Thank you for choosing the Hammerfall DSP MADI. This unique audio system is capable of
transferring digital audio data directly into a computer, from any device equipped with a MADI
interface. Installation is simple, even for the inexperienced user, thanks to the latest Plug and
Play technology. The numerous unique features and well thought-out configuration dialog puts
the Hammerfall DSP MADI at the very top of the range of digital audio interface cards.
The package contains drivers for Windows 2000 SP4, Windows XP, and Mac OS X.
Our high-performance philosophy guarantees maximum system performance by executing as
many functions as possible not in the driver (i.e. the CPU), but directly within the audio hardware.
2. Package Contents
Please check your Hammerfall DSP system's package contains each of the following:
Please check that your Hammerfall DSP MADI package contains each of the following:
• HDSP MADI PCI card
• HDSP MADI expansion board
• Quick Info guide
• RME Driver CD
• MIDI breakout cable
• Expansion Board ribbon cable (14-conductor)
3. System Requirements
• Windows 2000 SP4, Windows XP, Mac OS X (10.28 or higher)
• PCI Interface: a free PCI rev. 2.1 Busmaster slot
4. Brief Description and Characteristics
• Hammerfall design: 0% (zero!) CPU load, even using all 128 ASIO channels
Before installing the PCI card, please make sure the computer is switched off and the
power cable is disconnected from the mains supply. Inserting or removing a PCI card while
the computer is in operation can cause irreparable damage to both motherboard and card!
1. Disconnect the power cord and all other cables from the computer.
2. Remove the computer's housing. Further information on how to do this can be obtained
from your computer's instruction manual.
3. Important: Before removing the HDSP MADI from its protective bag, discharge any static in
your body by touching the metal chassis of the PC.
4. Prior to installation: Connect the HDSP MADI card to the Expansion Board using the supplied flat ribbon cable.
5. Insert the HDSP MADI firmly into a free PCI slot, press and fasten the screw.
6. Insert the Expansion Board and fasten the screw.
7. Replace the computer's housing.
8. Reconnect all cables including the power cord.
6. Hardware - Connectors
6.1 External Connectors
The bracket of the mainboard has two MADI interfaces, optical and coaxial
input and output each, a
MADI error LED and the
analog stereo output.
Identical signals are available at both the optical
and the coaxial output. An
obvious use for this would be to simply connect two devices, i.e. using the HDSP MADI as a
splitter (distribution 1 to 2).
The Expansion Board's
bracket has the word clock input and output.
Next to the input BNC
socket, a green LED dis-
plays the word clock input's LOCK state. Between the BNC sockets,
75 Ohm word clock termination can be activated
and verified by a yellow LED.
The included breakout cable is connected to the 9-pin Mini-DIN connector and provides two MIDI inputs and outputs via four 5-pin DIN connectors.
Note: If neither word clock I/O nor MIDI I/O are required, it is not necessary to install the Expansion Board at all.
Optional TCO
The optional Time Code Option is connected to the mainboard with a 10-pin flat ribbon cable.
The jack is located right beside the Expansion Board's one. Further details can be found in the
TCO's manual.
15-pin connector for the included HDSP MADI Expansion Board.
X7
10-pin connector for a connection of the optional Time Code Option (TCO).
X1
No function. Used to program the card in the factory.
Blue Jumper
The internal blue jumper X4 allows to change the voltage level at the coaxial MADI output. In
the lower position, the card generates 600 mVpp, according to the specification. Changing the
jumper to the upper position, the output voltage is increased to 1.2 Vpp. This setting is not intended to be used in normal operation. But in case of a very long or 'lossy' coaxial cable, this
setting may ensure an error-free operation of the transmission line.
7. Accessories
RME offers several optional components. Additionally parts of the HDSP MADI, like the special
breakout cables, are available separately.
ONK1 MADI Optical Network Cable, 3.3 ft (1 m)
ONK3 MADI Optical Network Cable, 10 ft (3 m)
ONK6 MADI Optical Network Cable, 20 ft (6 m)
ONK10 MADI Optical Network Cable, 33 ft (10 m)
ONK20 MADI Optical Network Cable, 66 ft (20 m)
8. Warranty
Each individual Hammerfall DSP undergoes comprehensive quality control and a complete test
at RME before shipping. The usage of high grade components allow us to offer a full two year
warranty. We accept a copy of the sales receipt as valid warranty legitimation.
If you suspect that your product is faulty, please contact your local retailer. The warranty does
not cover damage caused by improper installation or maltreatment - replacement or repair in
such cases can only be carried out at the owner’s expense.
RME does not accept claims for damages of any kind, especially consequential damage. Liability is limited to the value of the Hammerfall DSP. The general terms of business drawn up by
Synthax Audio AG apply at all times.
RME news, driver updates and further product information are available on our website:
http://www.rme-audio.com
If you prefer to read the information off-line, you can browse through a complete copy of the
RME website, found on the RME Driver CD (in the \rmeaudio.web directory).
Manufacturer:
IMM Elektronik GmbH, Leipziger Strasse 32, D-09648 Mittweida
Trademarks
All trademarks, registered or otherwise, are the property of their respective owners. RME,
DIGI96, SyncAlign, ZLM, SyncCheck, DIGICheck and Hammerfall are registered trademarks of
RME Intelligent Audio Solutions. HDSP MADI, TMS and TotalMix are trademarks of RME Intelligent Audio Solutions. Alesis and ADAT are registered trademarks of Alesis Corp. ADAT optical is a trademark of Alesis Corp. Microsoft, Windows 2000 and Windows XP are registered
trademarks or trademarks of Microsoft Corp. Steinberg, Cubase and VST are registered trademarks of Steinberg Media Technologies GmbH. ASIO is a trademark of Steinberg Media Technologies GmbH.
Copyright Matthias Carstens, 9/2005. Version 1.21
Current driver version: W2k/XP: 2.41, Mac OS X: 1.0a
Although the contents of this User’s Guide have been thoroughly checked for errors, RME can not guarantee that it is correct
throughout. RME does not accept responsibility for any misleading or incorrect information within this guide. Lending or
copying any part of the guide or the RME Driver CD, or any commercial exploitation of these media without express written
permission from RME Intelligent Audio Solutions is prohibited. RME reserves the right to change specifications at any time
without notice.
This device has been tested and found to comply with the EN55022 class B and EN50082-1
norms for digital devices, according to the European Council directive on counterpart laws in
the member states relating to electromagnetic compatibility (EMVG).
FCC
This device has been tested and found to comply with the requirements listed in FCC Regulations, part 15 for Class ‘B’ digital devices. Compliance with these requirements provides a reasonable level of assurance that your use of this product in a residential environment will not
result in harmful interference with other electronic devices.
This equipment generates radio frequencies and, if not installed and used according to the
instructions in the User’s Guide may cause interference harmful to the operation of other electronic devices.
Compliance with FCC regulations does not guarantee that interference will not occur in all installations. If this product is found to be the source of interference, which can be determined by
turning the unit off and on again, please try to eliminate the problem by using one of the following measures:
• Relocate either this product or the device that is being affected by the interference
• Use power outlets on different branch circuits, or install AC line filters
• Contact your local retailer or any qualified radio and television engineer
FCC compliance statement: Tested to comply with FCC standards for home or office use.
After the HDSP MADI has been installed correctly (see 5. Hardware Installation), and the computer has been switched on, Windows will recognize the new hardware component and start its
‘Hardware Wizard’. Insert the RME Driver CD into your CD-ROM drive, and follow further instructions which appear on your computer screen. The driver files are located in the directory
\HDSP_MADI_w2k on the RME Driver CD.
Windows will install the Hammerfall DSP System driver, and will register the card in the system
as a new audio device. After a reboot the HDSP MADI is ready for use.
In case the warning messages 'Digital signature not found', 'Do not install driver', 'not certified
driver' or similar come up: Don't listen to Microsoft, listen to us and continue with the installation.
In case the Hardware Wizard does not show up automatically after installation of the card,
do not attempt to install the drivers manually! An installation of drivers for non-recognized
hardware will cause a blue screen when booting Windows!
10.2 Driver Update
RME's driver updates often include a new madi.inf file. Also the revision number of the hardware might change (after a flash update). To prevent Windows 2000/XP from using an old
hdsp32.inf, or to copy some of the old driver files, be sure NOT to let Windows search for the
driver! Instead tell Windows what to do.
Under >Control Panel /System /Device Manager /Sound, Video and Game Controllers /RME
Hammerfall DSP MADI /Properties /Driver< you'll find the 'Update Driver' button. Select 'Install
from a list or specific location (advanced)', click 'Next', select 'Don't search I will choose
the driver to install', click 'Next', then 'Have Disk'. Now point to the driver update's directory.
10.3 Deinstalling the Drivers
A deinstallation of the HDSP's driver files is not necessary – and not supported by Windows
anyway. Thanks to full Plug & Play support, the driver files will not be loaded after the hardware
has been removed. If desired these files can then be deleted manually.
Unfortunately Windows Plug & Play methods do not cover the additional autorun entries of
TotalMix, the Settings dialog, and the registration of the ASIO driver. Those entries can be
removed from the registry through a software deinstallation request. This request can be found
(like all deinstallation entries) in Control Panel, Software. Click on the entry 'RME Hammerfall
DSP MADI'.
The Flash Update Tool updates the HDSP MADI to the latest firmware version. It requires an
already installed driver.
Start the program madi_fut.exe. The Flash Update Tool displays the current revision of the
HDSP MADI, and whether it needs an update or not. If so, then please press the 'Update' button. A progress bar will indicate when the flash process is finished. The bar moves slowly first
(program), then faster (verify).
If more than one interface card is installed, all cards can be flashed by changing to the next tab
and repeating the process.
After the update the PCI card need to be resettet. This is done by powering down and shutting
off the PC. A warm boot is not enough!
When the update fails (status: failure), the card's second BIOS will be used from the next cold
boot on (Secure BIOS Technology). Therefore the card stays fully functional. The flash process
should then be tried again on a different computer.
Note: Because of the changed hardware revision, Windows 2000/XP will start the hardware
assistant and wants to install new drivers. Do NOT let Windows search for new drivers, but
follow the instructions given in chapter 10.2.
11. Configuring the HDSP MADI
11.1 Settings Dialog
Configuration of the HDSP MADI is done via its own settings dialog. The panel 'Settings' can
be opened:
• by clicking on the hammer symbol in the Task Bar's system tray
The mixer of the HDSP MADI (TotalMix) can be opened:
• by clicking on the mixer icon in the Task Bar's system tray
The hardware of the HDSP system offers a number of helpful, well thought-of practical functions and options which affect how the card operates - it can be configured to suit many different requirements. The following is available in the 'Settings' dialog:
• Input selection
• Configuration of digital I/Os
• Synchronization behaviour
• State of input and output
• Current sample rate
• Latency
Any changes made in the Settings dialog are applied immediately - confirmation (e.g. by clicking on OK or exiting the dialog) is not required. However, settings should not be changed during playback or record if it can be avoided, as this can cause unwanted noises. Also, please
note that even in 'Stop' mode, several programs keep the recording and playback devices
open, which means that any new settings might not be applied immediately.
The status displays at the bottom of the dialog box give the user precise information about the
current status of the system, and the status of all digital signals.
All the card's settings described below are stored in a hardware memory, and are loaded immediately after a power-on of the computer. In clock mode Master even the last used sample
rate is set. Directly after switching on the computer, a stable and predictable clock state is
found at the HDSP MADI's outputs. This advanced technology completely eliminates disturbing
noises and clock network problems during power-up or re-boot.
Safe Mode
Frequency verifies the current digital input signal against the settings in the record program.
When de-activated a recording will always be allowed, even with non-valid input signals. Fre-
quency is valid for MME only.
SyncAlign guarantees synchronous channels when using MME multitrack software. This option
should only be switched off in case the used software does not work correctly with SyncAlign
activated.
Input activates redundancy operation. If the current input signal fails, the other input will be
used immediately, provided a valid signal is found there. Input also works as automatic input
selection, in case only optical or coaxial is present as input signal.
TMS activates the transmission of Channel Status data and Track Marker information from the
MADI input signal.
Buffer Size
The setting Buffer Size determines the latency between incoming and outgoing ASIO and GSIF
data, as well as affecting system stability (see chapter 13/14). Under Windows MME this setting
determines the DMA buffer size (see chapter 12.3).
SyncCheck
SyncCheck indicates whether there is
a valid signal (Lock) for the inputs
Word Clock, MADI and the optional
TCO (Video or LTC), or if there is a
valid and synchronous signal (Sync).
The AutoSync Reference display
shows the input and frequency of the
current sync source.
Time Code
Shows the timecode read out from
the LTC input of the optional TCO.
MADI In
Defines the input for the MADI signal. 'Optical' relates to the optical
input, 'Coaxial' to the BNC socket.
MADI Out
Defines the format of the MADI output signal. MADI can be a 56 or 64
channel signal.
96 kHz
Sample rates higher than 48 kHz can
be transmitted using the normal 48K
Frame, or using a native 96K Frame
at the card's output.
The unit can be configured to use its internal clock source (Master), or the clock source predefined via Pref. Sync Ref (AutoSync).
Pref. Sync Ref.
Used to pre-select the desired clock source. If the selected source isn't available, the unit will
change to the next available one. The current clock source and sample rate is displayed in the
AutoSync Ref display.
The automatic clock selection checks and changes between the clock sources Word Clock and
MADI, and Video and LTC when using the optional TCO.
System Clock
Shows the current clock state of the HDSP system. The system is either Master (using its own
clock) or Slave (see AutoSync Ref).
11.2 Settings dialog - DDS
Usually soundcards and audio interfaces generate their internal clock (master mode) by a
quartz. Therefore the internal clock can be set to 44.1 kHz or 48 kHz, but not to a value in between. SteadyClock, RME's sensational Low Jitter Clock System, is based on a Direct Digital Synthesizer (DDS). This superior circuitry can generate nearly any frequency with highest precision.
DDS has been implemented into the HDSP MADI with regard to the needs of professional
video applications, as well as to maximum flexibility. The dialog DDS includes both a list of
typical video frequencies (so called pull up/pull down at 0.1% and 4%) and two faders, which
allow to freely change the basic sample rate in steps of 1 Hz (!).
Application examples
DDS allows for a simultaneous change of speed and tune during record and playback. From
alignment to other sources up to creative effects – everything is possible..
DDS allows to intentionally de-tune the complete DAW. This way, the DAW can match instruments which have a wrong or unchangeable tuning.
DDS allows to define a specific sample rate. This feature can be is useful in case the system
randomly changes the sample rate – for unknown reasons. It also prevents a change from
Double Speed (96 kHz) to Single Speed (48 kHz), which would cause configuration and routing
problems by the changed amount of MADI channels.
The DDS dialog requires the HDSP MADI to be in clock mode Master! The frequency setting will only be applied to this one specific card!
Changing the sample rate in bigger steps during record/playback often results in a loss of
audio, or brings up warning messages of the audio software. Therefore the desired sample
rate should be set at least coarsely before starting the software.
Shows the sample rate as adjusted in
this dialog. The sample rate is defined
by the basic setting (Frequency), the
multiplier, and the position of the activated fader.
Frequency
Sets a fixed basic sample rate, which
can be modified by multiplier and
fader.
Freq. Multiplier
Changes the basic sample rate into
Single, Double oder Quad Speed
mode.
Coarse
Fader for coarse modification of the
basic sample rate. Click Active to activate it. Minimum step size 1 Hz.
Fine
Fader for fine modification of the basic
sample rate. Click Active to activate it.
Minimum step size 1 Hz.
Notes on the faders
A mouse click within the fader area, above or below the fader know, will move the fader with
the smallest step size up or down. Holding the Ctrl key while clicking will cause the fader to
jump to its center (0) position.
11.3 Clock Modes - Synchronisation
AutoSync
The HDSP MADI has been equipped with AutoSync, an automatic clock source selection,
which adopts the first available input with a valid digital signal as the clock reference input. The
input currently used as sync reference is shown in the AutoSync Ref status field, together with
its current sample frequency.
AutoSync guarantees that normal record and record-while-play will always work correctly. In
certain cases however, AutoSync may cause feedback in the digital carrier, so synchronization
breaks down. To remedy this, switch the HDSP’s clock mode over to 'Master'.
Via Pref. Sync Ref (preferred synchronization reference) a preferred input can be defined. As
long as the card sees a valid signal there, this input will be designated as the sync source, otherwise the other inputs will be scanned in turn. If none of the inputs are receiving a valid signal,
the card automatically switches clock mode to ‘Master’.
Thanks to its AutoSync technique and lightning fast PLLs, the HDSP is not only capable of
handling standard frequencies, but also any sample rate between 28 and 200 kHz. Even the
word clock input, which most users will use in varispeed operation, allows any frequency between 28 kHz and 200 kHz.
The HDSP MADI's outstanding clock control allows for a synchronization of the output signal to
the word clock's input signal not only at identical sample rates, but also at half, quarter, double
and quad sample rates. A playback of 192 kHz can easily be synchronized via a 48 kHz word
clock signal.
SyncCheck
If several digital devices are to be used simultaneously in a system, they not only have to operate with the same sample frequency but also be synchronous with each other. This is why
digital systems always need a single device defined as ‘master’, which sends the same clock
signal to all the other (‘slave’) devices.
Remember that a digital system can only have one master! If the HDSP MADI’s clock mode
is set to Master, all other devices must be set to Slave.
RME’s exclusive SyncCheck technology (first implemented in the Hammerfall) enables an
easy to use check and display of the current clock status. The SyncCheck field indicates
whether no signal (No Lock), a valid signal (Lock) or a valid and synchronous signal (Sync) is
present at each of the digital clock source inputs. The ‘AutoSync Ref’ display shows the current
sync source and the measured frequency.
In practice, SyncCheck provides the user with an easy way of checking whether all digital devices connected to the system are properly configured. With SyncCheck, finally anyone can
master this common source of error, previously one of the most complex issues in the digital
studio world.
The HDSP system can play back audio data only in supported modes (channels, PCM) and
formats (sample rate, bit resolution). Otherwise an error message appears (for example at 22
kHz and 8 bit).
In the audio application being used, HDSP must be selected as output device. This can often
be found in the Options, Preferences or Settings menus under Playback Device, Audio Devices, Audio etc.
We strongly recommend switching off all system sounds (via >Control Panel /Sounds<). Also
HDSP should not be the Preferred Device for playback, as this could cause loss of synchronization and unwanted noises. If you feel you cannot do without system sounds, you should consider buying a cheap Blaster clone and select this as Preferred Device in >Control Panel /Multimedia /Audio<.
The screenshot to the
right shows a typical
configuration dialog as
displayed by a (stereo) wave editor. After
selecting one of the
32 playback devices,
audio data is sent to
the according audio
channels.
Increasing the number
and/or size of audio
buffers may prevent
the audio signal from
breaking up, but also
increases latency i.e.
output is delayed. For
synchronized playback of audio and
MIDI (or similar), be
sure to activate the checkbox ‘Get position from audio driver’.
The HDSP MADI allows sample rates of up to 96 kHz via MADI. In this mode, only channels 1
to 32 are available.
When using popular DVD software player like WinDVD and PowerDVD, their audio data stream
can be sent to any AC-3/DTS capable receiver via the HDSP MADI. For this to work an output
wave device has to be selected in >Control Panel/ Sounds and Multimedia/ Audio<. Also check
'use preferred device only'.
You will notice that the DVD software's audio properties now allow to use 'SPDIF Out', 'Use
SPDIF' or to 'activate SPDIF output'. When selecting these, the software will transfer the nondecoded digital multichannel data stream to the HDSP. Naturally a successful decoding also
requires a MADI to AES converter like the RME ADI-642, converting the playback signals to
stereo AES3 or SPDIF.
Note: This 'SPDIF' signal sounds like chopped noise at highest level.
Multichannel
PowerDVD can also operate as software decoder, sending a DVD's multichannel data stream
directly to the analog outputs of the HDSP MADI. Supported are all modes, from 2 to 8 channels, at 16 bit resolution and 48 kHz sample rate.
For this to work an output wave device of the HDSP has to be selected in >Control Panel/ Sounds and Multimedia/ Audio<. Also check 'use preferred device only'. PowerDVD's audio
properties now lists several multichannel modes. If one of these is selected, PowerDVD sends
the decoded analog multichannel data to the HDSP MADI.
The device selected as Preferred Playback Device defines the first playback channel. Choosing
ADAT 3/4 and 6-channel mode, playback will happen on channels 3 to 8. Choosing MADI 3/4
and 6-channel mode, playback will happen on channels 3 to 8.
The channel assignment using PowerDVD is:
1 (first chosen playback channel) - Left
2 - Right
3 - Center
4 - LFE (Low Frequency Effects)
5 - SR (Surround Right)
6 - SL (Surround Left)
Note 1: Setting the card to be used as system playback device is against common sense, as
professional cards are not specialized to play back system sounds, and shouldn't be disturbed
by system events. To prevent this, be sure to re-assign this setting after usage, or to disable
any system sounds (tab Sounds, scheme 'No audio').
Note 2: The DVD player will be synced backwards from the HDSP card. So when using AutoSync and/or word clock, the playback speed and pitch follows the incoming clock signal.
Note 3: PowerDVD 5 no longer supports a starting channel other than channel 1. Use TotalMix
to send audio to any aother output port if necessary.
Note 4: In WinDVD 5 and 6, a channel-separated 5.1 playback using the HDSP requires a
change in the registry. Start regedit, go to >HKEY_CURRENT_USER / Software / InterVideo / DVD5 / AUDIOCHAN<, and set its value to 4 (hexadecimal). Additionally select 'Waveout' as
Audio Renderer in the Audio configuration dialog. Warning: Changes to the registry are
done at your own risk. Danger of complete data loss!
12.3 Low Latency under MME (Buffer Size Adjustment)
Using Windows 95 or 98 the MME buffer size was nothing to worry about. Latencies below 46
ms were not possible. Meanwhile both computers and operating system have become much
more powerful, and since Windows ME/2000/XP latencies far lower can be used. SAWStudio
and Sonar allowed to use such low settings from the start. Sequoia was updated in version
5.91, WaveLab in version 3.04.
In the HDSP's Settings dialog the MME buffersize (in fact the DMA buffer size) is set with the
same buttons as the ASIO buffer size. Our test computers allow to use settings down to 64
samples without clicks. Please note that this setting only defines the buffer size of the hardware. The true and effective latency is configured within the MME application!
Attention: the DMA buffers must not be larger than the application's buffers. Playback will
be stuttering and audio will be distorted.
This case can happen unnoticed when using ASIO and MME at the same time (multi-client)
and setting ASIO to 186 ms, while the buffers in the MME application are still set for a lower
latency.
Example: when you set the Hammerfall to 512 you can't use 128 in any program. But setting
the buffer size to 128 allows to use 128 and all higher values within the software.
Please also note that this is a you're-welcome-to-try feature. We can't guarantee that you will
be able to use 3 or 6 ms with MME. Simply check out by yourself which lowest setting your
system and software allows. Some motherboards with insufficient PCI bandwidth (like older VIA
boards) suffer from crackling at settings below 512. Be sure to set the buffer size to 512 or
higher in such a case (or get a new motherboard).
12.4 Multi-client Operation
RME audio interfaces support multi-client operation. This means several programs can be used
at the same time. Also all formats, like ASIO, MME and GSIF can be used simultaneously. The
use of multi-client operation requires to follow two simple rules:
I.e. it is not possible to use one software with 44.1 kHz and the other with 48 kHz.
• Different software can not use the same channels at the same time.
If for example Cubase uses channels 1/2, this playback pair can't be used in Gigasampler/Studio (GSIF) nor under MME (WaveLab etc.) anymore. This is no limitation at all, because TotalMix allows any output routing, and with this a playback of multiple software on the
same hardware outputs. Note that the inputs can be used at the same time, as the driver sends
the data to all applications simultaneously.
ASIO-Multiclient
RME audio interfaces support ASIO multi-client operation. It is possible to use more than one
ASIO software at the same time. Again the sample rate has to be identical, and each software
has to use its own playback channels. Once again the same inputs can be used simultaneously.
RME's sophisticated tool DIGICheck is an exception to this rule. It operates like an ASIO host,
using a special technique to access playback channels already occupied. Therefore DIGICheck
is able to analyse and display playback data from any software, no matter which format the
software uses.
Unlike analog soundcards which produce empty wave files (or noise) when no input signal is
present, digital I/O cards always need a valid input signal to start recording.
To take this into account, RME has included two unique features in the HDSP MADI: a comprehensive I/O signal status display (showing sample frequency, lock and sync status) in the
Settings dialog, and the protective Safe Mode / Frequency function.
If a 48 kHz signal is fed to the input and the application is set to 44.1 kHz, Safe Mode / Fre-quency stops the system from recording. This prevents faulty takes, which often go unnoticed
until later on in the production. Such tracks appear to have the wrong playback rate - the audio
quality as such is not affected.
The sample frequency shown in the Settings dialog (see chapter 11, screenshot Settings) is
useful as a quick display of the current configuration (the box itself and all connected external
equipment). If no sample frequency is recognized, it will read ‘No Lock’.
This way, configuring any suitable audio application for digital recording is simple. After selecting the required input, HDSP MADI displays the current sample frequency. This parameter can
then be changed in the application’s audio attributes (or similar) dialog.
The screenshot to the right shows a typical dialog
used for changing basic parameters such as sample frequency and resolution in an audio application.
Any bit resolution can be selected, providing it is
supported by both the audio hardware and the
software. Even if the input signal is 24 bit, the
application can still be set to record at 16-bit resolution. The lower 8 bits (and therefore any signals
about 96dB below maximum level) are lost entirely. On the other hand, there is nothing to gain
from recording a 16-bit signal at 24-bit resolution this would only waste precious space on the hard
disk.
It often makes sense to monitor the input signal or send it directly to the output. This can be
done at zero latency using TotalMix (see chapter 24).
Currently two solutions exist which enable an automated control of real-time monitoring. ZLM
allows monitoring in Punch I/O mode - with this the system behaves like a tape machine. This
method has been implemented in all versions of Samplitude (by Magix), and can be activated
using the global track option 'Hardware monitoring during Punch'. As ZLM is limited to MME,
this mode is no longer supported since TotalMix version 2.3.
The second solution is Steinberg’s ASIO protocol with our ASIO 2.0 drivers and all ASIO 2.0
compatible programs (even Samplitude…). When 'ASIO Direct Monitoring' has been switched
on, the input signal is routed in real-time to the output whenever a recording is started.
Start the ASIO software and select ASIO Hammerfall DSP as the audio I/O device. The 'ASIO
system control' button opens the HDSP's Settings dialog (see chapter 11, Configuration).
Hammerfall DSP supports
ASIO Direct Monitoring
(ADM). Please note that
currently Nuendo, Cubase
and Logic either do not
support ADM completely
or error-free.
Using emulated MIDI
drivers often causes a drift
and delay between audio
and MIDI. You should use
non-emulated (MME) MIDI
ports in such a case.
At a sample rate of 88.2 or
96 kHz, the number of
channels available at the
MADI input and output is
halfed.
13.2 Known Problems
If a computer does not provide sufficient CPU-power and/or sufficient PCI-bus transfer rates,
then drop outs, crackling and noise will appear. We recommend to deactivate all PlugIns to
verify that these are not the reason for such effects.
Additional hard disk controllers, both on-board and PCI based, aften violate the PCI specs. To
achieve the highest throughput they hog the PCI bus, even in their default setting. Thus when
working with low latencies heavy drop outs (clicks) are heard. Try to solve this problem by
changing the default setting of the controller (for example by reducing the 'PCI Bus Utilization').
When using more than one HDSP system, all units have to be in sync, see chapter 15. Else a
periodicly repeated noise will be heared.
Another common source of trouble is incorrect synchronization. ASIO does not support asynchronous operation, which means that the input and output signals not only have to use the
same sample frequency, but also have to be in sync. All devices connected to the Hammerfall
DSP must be properly configured for Full Duplex operation. As long as SyncCheck (in the Settings dialog) only displays Lock instead of Sync, the devices have not been set up properly!
The GSIF interface of the HDSP MADI allows direct operation with Gigastudio, with up to 32*
channels, 96 kHz and 24 bit. The new GSIF 2.0 is also supported with both audio and MIDI.
Gigastudio requires a lot of the computer’s calculation power. An optimum performance is
achieved with a stand-alone GSIF PC. However, when using the Hammerfall DSP, the latency
is always the same as the one selected for ASIO operation. This can cause performance problems on slower machines when using GSIF and ASIO at the same time.
Please note that the W2k/XP driver fully supports multi-client operation, including the combination MME/ASIO. So for example Cubase, Gigastudio and Sonar can be used simultaneously,
provided each of these programs uses its own audio channels exclusively. For example ASIO
could use channels 1/2 and Gigastudio (with GSIF) channels 3/4 simultaneously, and so on.
Simultaneous operation of GSIF and ASIO requires to use different channels. For example,
if Cubase uses tracks 1/2 these tracks can not be used by Gigastudio.
Common Problems
Please note that Gigastudio is running unexpectedly in the background (thus blocking its assigned audio channels), as soon as the Gigastudio MIDI ports are used – even when Gigastudio
itself hasn't been started. This causes a lot of confusion, as the driver seems to behave completely buggy, and the user does not recognize the simple reason for it – for example simultaneous operation of ASIO and GSIF on the same channels.
If Gigastudio starts up properly, loads gig files too, but won't play at all even when using the
virtual keyboard: Go to Hardware/Routing and select a valid MIDI input port. Note that blank is
not valid, but <none> is.
*The limitation of 32 channels is caused by Gigastudio 2.54. According to Tascam, Gigastudio
3 will support 64 channels.
The current drivers support operation of up to three HDSP MADI. HDSP MADI and AES-32 use
the same driver, therefore can be used at the same time. Please note that only one TCO of one
card can be used (of course). All units have to be in sync, i.e. have to receive valid sync information either via word clock or by using AutoSync and feeding synchronized signals.
• If one of the HDSP systems is set to clock mode Master, all others have to be set to clock
mode AutoSync, and have to be synced from the master, for example by feeding word
clock. The clock modes of all units have to be set up correctly in their Settings dialog.
• If all units are fed with a synchronous clock, i.e. all units show Sync in their Settings dialog,
all channels can be used at once. This is especially easy to handle under ASIO, as the ASIO
driver presents all units as one.
Note: TotalMix is part of the hardware of each HDSP system. Up to three mixers are available,
but these are separated and can't interchange data. Therefore a global mixer for all units is not
possible.
16. DIGICheck
The DIGICheck software is a unique utility developed for testing, measuring and analysing
digital audio streams. Although this Windows software is fairly self-explanatory, it still includes
a comprehensive online help. DIGICheck 4.4 operates as multi-client ASIO host, therefore can
be used in parallel to any software, be it MME, ASIO or GSIF, with both inputs and outputs (!).
The following is a short summary of the currently available functions:
• Level Meter. High precision 24-bit resolution, 2/8/18 channels. Application examples: Peak
level measurement, RMS level measurement, over-detection, phase correlation measurement, dynamic range and signal-to-noise ratios, RMS to peak difference (loudness), long
term peak measurement, input check. Oversampling mode for levels higher than 0 dBFS.
Vertical and horizontal mode. Slow RMS and RLB weighting filter. Supports visualization
according to the K-system.
• Hardware Level Meter for Input, Playback and Output. As above, received pre-calculated
directly from the HDSP system hardware with near zero CPU load.
• Spectral Analyser. World wide unique 10-, 20- or 30-band display in analog bandpass-filter
technology. 192 kHz-capable!
• Vector Audio Scope. World wide unique Goniometer showing the typical afterglow of an
oscilloscope-tube. Includes Correlation meter and level meter.
• Totalyser. Spectral Analyser, Level Meter and Vector Audio Scope in a single window.
• Bit Statistics & Noise. Shows the true resolution of audio signals as well as errors and DC
offset. Includes Signal to Noise measurement in dB and dBA, plus DC measurement.
• Channel Status Display. Detailled analyzis and display of SPDIF and AES/EBU Channel
Status data.
• Completely multi-client. Open as many measurement windows as you like, on any chan-
nels and inputs or outputs!
To install DIGICheck, go to the \DIGICheck directory on the RME Driver CD and run setup.exe.
Follow the instructions prompted on the screen.
DIGICheck is conctantly improved. The latest version is always found on our website
www.rme-audio.de, section Downloads/Tools.
The newest information can always be found on our website www.rme-audio.com, section FAQ,
Latest Additions.
The input signal cannot be monitored in real-time
• ASIO Direct Monitoring has not been enabled, and/or monitoring has been disabled globally
(for example in TotalMix).
Playback works, but record doesn’t
• Check that there is a valid signal at the input. If so, the current sample frequency is dis-
played in the Settings dialog.
• Check whether the HDSP system has been selected as recording device in the audio appli-
cation.
• Check whether the sample frequency set in the audio application (‘Recording properties’ or
similar) matches the input signal.
• Check that cables/devices have not been connected in a closed loop. If so, set the systems’s clock mode to Master.
Crackle during record or playback
• Increase the number and size of buffers in the ‘Settings’ dialog or in the application.
• Try different cables (coaxial or optical) to rule out any defects here.
• Check that cables/devices have not been connected in a closed loop. If so, set the system’s
clock mode to ‘Master’.
• Increase the buffer size of the hard disk cache.
• Activate Busmaster mode for the hard disks.
• In case of a recently done BIOS update of the motherboard: Propably 'Load BIOS Defaults'
was loaded instead of 'Load Setup Defaults'. This sets the 'PCI Latency Timer' to 0 (default:
32).
Low Latency ASIO operation under Windows 2000/XP on single CPU systems:
• To use ASIO at lowest latencies under Windows 2000/XP even when only having one CPU,
the system performance has to be optimized for background tasks. Go to >Control Panel/ System/ Advanced/ Performance Options<. Change the default 'Applications' to 'Background
tasks'. The lowest usable latency will drop from 23 ms to around 3 ms.
Hammerfall DSP is found in the Device Manager (>Settings/ Control Panel/ System<), category 'Sound-, Video- and Gamecontroller'. A double click on 'Hammerfall DSP MADI' starts the
properties dialog. Choosing 'Resources' shows Interrupt and Memory Range.
The newest information on hardware problems can always be found on our website www.rme-
audio.com, section FAQ, Hardware Alert: about incompatible hardware.
The dialog 'New hardware component found’ does not appear:
• Check whether the PCI interface is correctly inserted in the PCI slot.
The card and drivers have been installed correctly, but playback does not work:
• Check whether the Hammerfall DSP appears in the Device Manager. If the ' Hammerfall
DSP’ device has a yellow exclamation mark, then there is an address or interrupt conflict.
• Even if there is no yellow exclamation mark, it is worth checking the ‘Resources’ tab anyway.
First fit the card (see 5. Hardware Installation), then switch on the computer and install the drivers from the RME Driver CD. The driver file is located in the folder HDSP MADI. Installation
works automatically by a double-click on the file hdspmadi.mpkg.
RME recommends to download the latest driver version from the RME website! If done, the
procedure is as follows:
Double-click onto hdspmadi_xx.gz to expand the archive file to hdspmadi_xx.tar and the
folder HDSP MADI xx, which includes the driver file hdspmadi.mpkg. Installation works automatically by a double-click on this file.
During driver installation the programs Settings and Mixer (TotalMix) will also be installed.
Both programs start automatically as soon as a HDSP system is detected. They stay in the
dock when exited, and remove themselves automatically from the dock when the HDSP system is removed.
Reboot the computer when installation is done.
18.2 Driver Update
In case of a driver update it's not necessary to remove the old driver first, it will be overwritten
during the installation.
Exception: driver updates from version <1.1. Remove the former Settings dialog and
TotalMix from the Login Items, and delete both files from your hard drive!
This driver version did not have the features AutoLoad, Dock Lock and AutoRemove. Therefore one has to make sure that both programs have been removed from the system, to prevent
the old Settings dialog and TotalMix from being loaded.
18.3 Flash Update
The Flash Update Tool updates the HDSP MADI card to the latest firmware version. It requires
an already installed driver.
Start the program HDSP MADI Flash. The Flash Update Tool displays the current revision of
the HDSP interface, and whether it needs an update or not. If so, then simply press the 'Update'
button. A progress bar will indicate when the flash process is finished. The bar moves slowly
first (program), then faster (verify).
If more than one interface card is installed, all cards can be flashed by changing to the next tab
and repeating the process.
After the update the PCI card needs to be resettet. This is done by powering down and shutting
off the PC. A warm boot is not enough!
When the update fails (status: failure), the card's second BIOS will be used from the next cold
boot on (Secure BIOS Technology). Therefore the card stays fully functional. The flash process
should then be tried again on a different computer.
Configuring the HDSP MADI is done via its own settings dialog. The panel 'Settings' can be
opened by clicking on the hammer icon in the dock. The mixer of the HDSP MADI, TotalMix,
can be opened by clicking on the mixer icon in the dock.
The Hammerfall DSP’s hardware offers a number of helpful, well thought-of practical functions
and options which affect how the card operates - it can be configured to suit many different
requirements. The following is available in the 'Settings' dialog:
• Input selection
• Configuration of digital I/Os
• Synchronization behaviour
• State of input and output
• Current sample rate
• Latency
Any changes performed in
the Settings dialog are applied immediately - confirmation (e.g. by exiting the
dialog) is not required.
However, settings should
not be changed during playback or record if it can be
avoided, as this can cause
unwanted noises.
The status displays at the
bottom of the dialog box
give the user precise information about the current
status of the system, and
the status of all digital signals.
Quick Boot
All the card's settings described below are stored in a hardware memory, and are loaded immediately after a power-on of the computer. In clock mode Master even the last used sample
rate is set. Directly after switching on the computer, a stable and predictable clock state is
found at the HDSP MADI's outputs. This advanced technology completely eliminates disturbing
noises and clock network problems during power-up or re-boot.
MADI In
Defines the input for the MADI signal. 'Optical' relates to the optical input, 'Coaxial' to the BNC
socket.
MADI Out
Defines the format of the MADI output signal. MADI can be a 56 or 64 channel signal.
96 kHz
Sample rates higher than 48 kHz can be transmitted using the normal 48K Frame, or using a
native 96K Frame at the card's output.
Input activates redundancy operation. If the current input signal fails, the other input will be
used immediately, provided a valid signal is found there. Input also works as automatic input
selection, in case only optical or coaxial is present as input signal.
Clock Mode
The unit can be configured to use its internal clock source (Master), or the clock source predefined via Pref. Sync Ref (AutoSync).
Pref. Sync Ref.
Used to pre-select the desired clock source. If the
selected source isn't available, the unit will change to
the next available one. The
current clock source and
sample rate is displayed in
the AutoSync Ref display.
The automatic clock selection checks and changes
between the clock sources
Word Clock and MADI, and
Video and LTC when using
the optional TCO.
Input Status
Displays the state of the current input signal:
• Channel format (64 or 56 channels)
• Frame format (48K or 96K)
• Sample rate (measured)
• Active input (optical or coaxial)
System Clock
Shows the current clock state of the HDSP system. The system is either Master (using its own
clock) or Slave (see AutoSync Ref).
SyncCheck
SyncCheck indicates whether there is a valid signal (Lock) for the inputs Word Clock, MADI
and the optional TCO (Video or LTC), or if there is a valid and synchronous signal (Sync). The
AutoSync Reference display shows the input and frequency of the current sync source.
Time Code
Shows the timecode read out from the LTC input of the optional TCO.
Usually soundcards and audio interfaces generate their internal clock (master mode) by a
quartz. Therefore the internal clock can be set to 44.1 kHz or 48 kHz, but not to a value in between. SteadyClock, RME's sensational Low Jitter Clock System, is based on a Direct Digital Synthesizer (DDS). This superior circuitry can generate nearly any frequency with highest precision.
DDS has been implemented into the HDSP MADI with regard to the needs of professional
video applications, as well as to maximum flexibility. The dialog DDS includes both a list of
typical video frequencies (so called pull up/pull down at 0.1% and 4%) and two faders, which
allow to freely change the basic sample rate in steps of 1 Hz (!).
The DDS dialog requires the HDSP MADI to be in clock mode Master! The frequency setting will only be applied to this one specific card!
Changing the sample rate in bigger steps during record/playback often results in a loss of
audio, or brings up warning messages of the audio software. Therefore the desired sample
rate should be set at least coarsely before starting the software.
DDS
Activates all settings of this dialog.
Value
Shows the sample rate as adjusted in
this dialog. The sample rate is defined by the basic setting (Frequency), the multiplier, and the position of the activated fader.
Frequency
Sets a fixed basic sample rate, which
can be modified by multiplier and
fader.
Freq. Multiplier
Changes the basic sample rate into
Single, Double oder Quad Speed
mode.
Coarse
Fader for coarse modification of the basic sample rate. Click Active to activate it. Minimum
step size 1 Hz.
Fine
Fader for fine modification of the basic sample rate. Click Active to activate it. Minimum step
size 1 Hz.
Notes on the faders
A mouse click within the fader area, above or below the fader know, will move the fader with
the smallest step size up or down. Holding the Ctrl key while clicking will cause the fader to
jump to its center (0).
DDS allows for a simultaneous change of speed and tune during record and playback. From
alignment to other sources up to creative effects – everything is possible..
DDS allows to intentionally de-tune the complete DAW. This way, the DAW can match instruments which have a wrong or unchangeable tuning.
DDS allows to define a specific sample rate. This feature can be is useful in case the system
randomly changes the sample rate – for unknown reasons. It also prevents a change from
Double Speed (96 kHz) to Single Speed (48 kHz), which would cause configuration and routing
problems by the changed amount of MADI channels.
19.3 Clock Modes - Synchronisation
AutoSync
The HDSP MADI has been equipped with AutoSync, an automatic clock source selection,
which adopts the first available input with a valid digital signal as the clock reference input. The
input currently used as sync reference is shown in the AutoSync Ref status field, together with
its current sample frequency.
AutoSync guarantees that normal record and record-while-play will always work correctly. In
certain cases however, AutoSync may cause feedback in the digital carrier, so synchronization
breaks down. To remedy this, switch the HDSP’s clock mode over to 'Master'.
Via Pref. Sync Ref (preferred synchronization reference) a preferred input can be defined. As
long as the card sees a valid signal there, this input will be designated as the sync source, otherwise the other inputs will be scanned in turn. If none of the inputs are receiving a valid signal,
the card automatically switches clock mode to ‘Master’.
Thanks to its AutoSync technique and lightning fast PLLs, the HDSP is not only capable of
handling standard frequencies, but also any sample rate between 28 and 200 kHz. Even the
word clock input, which most users will use in varispeed operation, allows any frequency between 28 kHz and 200 kHz.
The HDSP MADI's outstanding clock control allows for a synchronization of the output signal to
the word clock's input signal not only at identical sample rates, but also at half, quarter, double
and quad sample rates. A playback of 192 kHz can easily be synchronized via a 48 kHz word
clock signal.
SyncCheck
If several digital devices are to be used simultaneously in a system, they not only have to operate with the same sample frequency but also be synchronous with each other. This is why
digital systems always need a single device defined as ‘master’, which sends the same clock
signal to all the other (‘slave’) devices.
Remember that a digital system can only have one master! If the HDSP MADI’s clock
mode is set to Master, all other devices must be set to Slave.
RME’s exclusive SyncCheck technology (first implemented in the Hammerfall) enables an
easy to use check and display of the current clock status. The SyncCheck field indicates
whether no signal (No Lock), a valid signal (Lock) or a valid and synchronous signal (Sync) is
present at each of the digital clock source inputs. The ‘AutoSync Ref’ display shows the current
sync source and the measured frequency.
In practice, SyncCheck provides the user with an easy way of checking whether all digital devices connected to the system are properly configured. With SyncCheck, finally anyone can
master this common source of error, previously one of the most complex issues in the digital
studio world.
The driver with the file suffix gz provided by RME is a compressed TAR archive. TAR bundles
multiple files and folders into one file, but does not save memory space nor download time.
Both TAR and gz are supported natively by OS X, a double click on the file is all you need to
do.
Older browsers do not recognize gz as an archive, loading the file as a document. This results
in a cryptic looking text within the browser window. Downloading the file can be done via the
right mouse key, Save Target as. Despite this procedure, some older browsers like Netscape
4.78 will not save the file correctly - the archive will be corrupted.
The driver consists of a package file (pkg), which contains various folders and files, similar to
TAR. A double click will start the OS X installer. To save you the hassle of installing both audio
and MIDI drivers separately, the HDSP driver contains an additional meta package (mpkg),
that points to the single packages. Those single packages are not shown in the Finder, as they
reside within the invisible folder '.contained_packages'. Only the mpkg is visible. Important: an
installation can only be done with the complete folder. If only the mpkg is copied to a different
place, it will not find the single driver packages!
The actual audio driver appears as a kernel extension file. The installer copies it to >System/ Library/ Extensions<. Its name is HDSPMADI.kext. It is visible in the Finder, allowing you to
verify date and driver version. Yet, in fact this again is a folder containing subdirectories and
files.
Nonetheless, this 'driver file' can be removed by simply dragging it to the trash bin. This can be
helpful in case a driver installation fails. An incomplete installation can currently (10.3.2) only
be detected indirectly: The installation routine does not open a message window with a note
about a restart of the computer. This indicates that the driver file was not copied and the driver
was not installed!
Several users have observed that the installation routine occasionally stops and no longer
works correctly. This can be fixed by removing the corresponding extension file prior to installation. In some cases, also (or only) a repair of the disk permission will help.
We have also received reports saying the driver update could not be installed on the system
disk - shown red crossed during the installation. Repairing permission may also help here. If
not, we're sorry, but have to recommend to contact Apple. Our driver has no knowledge of
folders, disks etc., the installation is handled completely by the OS X installer.
20.2 MIDI doesn't work
In some cases MIDI does not work after the installation of the HDSP driver. To be precise, applications do not show an installed MIDI port. The reason for this is usually visible within the
Audio MIDI Setup. It displays no RME MIDI device, or the device is greyed out and therefore
inactive. Mostly, removing the greyed out device and searching for MIDI devices again will
solve the problem. If this does not help, we recommend manual removal of the MIDI driver and
reinstallation of the complete driver. Otherwise repairing permissions may help.
The HDSP MIDI driver is a plugin. During installation it will be copied to >Library/ Audio/ MIDI Drivers<. It's name is Hammerfall DSP MIDI.plugin. The file can be displayed in the Finder
and also be removed by simply dragging it to the trash bin.
RME's Mac OS X driver supports all sampling frequencies provided by the hardware. Besides
192 kHz and 96 kHz this also includes 32 kHz and 64 kHz.
But not every software will support all the hardware's sample rates. For example Spark does
not display 32 kHz and 64 kHz. The hardware's capabilities can easily be verified in the Audio MIDI Setup. Select Audio devices under Properties of: and choose the Hammerfall DSP. A
click on Format will list the supported sample frequencies.
If the unit is in clock mode Master, selecting a sample rate will immediately set the device to
this frequency, which can be verified in the HDSP's settings dialog (System Clock). Format
thus allows you to activate any sampling frequency quickly and easily.
20.4 Repairing Disk Permissions
Repairing permission can solve problems with the installation process - plus many others. To
do this, launch Disk Utility located in Utilities. Select your system drive in the drive/volume list
to the left. The First Aid tab to the right now allows you to check and repair disk permissions.
20.5 PCI card and PCI slot compatibility
Unfortunately not every RME card will work in every PCI slot of an Apple computer. To our
knowledge, the current Hammerfall DSP systems can be used in any PCI slot of all G4 and G5
models. In case additional PCI cards of any manufacturer are used, it might happen that the
RME card is no longer found by the system. Swapping cards between slots can be helpful in
this case.
20.6 Various Information
The driver requires 10.2.8 or higher. Older versions of OS X are not and will not be supported.
Via >System Preferences/ Audio-MIDI Setup< the hardware can be configured for the system
wide usage. Programs that don't support card or channel selection will use the device selected
as Standard-Input and Standard-Output. (Soundstudio, Mplayer, Amplitube etc.).
In the lower part of the window, the audio hardware's capabilities are shown and can be
changed in some cases. On the record side no changes are possible. Programs that don't support channel selection will always use channels 1/2, the first stereo pair.
Since OS X 10.3 playback can be configured freely and to any of the available playback channels. This is done via Speaker Setup. Even multichannel playback (Surround, DVD Player)
can be set up easily.
OS X supports more than one audio device. Their simultaneous usage within one program had
been limited to Motu's Digital Performer until 10.3.9. Since 10.4 (Tiger) Core Audio offers the
function Aggregate Devices, which allows to combine several devices into one, so that a
multi-device operation is now possible with any software.
The Hammerfall DSP driver adds a number to each unit, so they are fully accessible in any
multicard-capable software.
Our experience so far is that using more than one HDSP MADI will work only on dedicated
server systems having multiple PCI busses. The PCI bus load of 128 channels plus the realtime behaviour necessary for audio are critical parameters. A usage of two cards at full track
count is therefore difficult to achieve on current Mac computers. See chapter 30.7, PCI – Performance.
HDSP MADI offers a hi-quality analog monitor output. The short circuit protected stereo line
output provides high output level, low impedance, and is available via a 6.3 mm (1/4") TRS
jack. Therefore it is also suitable for a direct use with headphones.
The analog output is directly driven from the channels 63/64, in Double Speed mode with
channels 33/34. Its output volume is controlled by the hardware output faders of channel 63/64
in TotalMix. Additionally the analog output can play back any input or playback signal (submix,
for example factory preset 5, 6 and 7).
RME's unique Speaker Protection reduces noise when switching the computer on and off, so there is no problem
even when using active monitors.
In case the output should operate as line
out, an adapter TRS plug to RCA phono
plugs, or TRS plug to TS plugs is required.
The pin assignment follows international
standards. The left channel is connected
to the tip, the right channel to the ring of
the TRS jack/plug.
22.2 MADI I/Os
The BNC input's ground-free design is built according to AES10-1991. The input impedance is
75 Ohm. It will operate error-free from about 180 mVpp on.
The optical input and output uses a FDDI (ISO/IEC 9413-3) compatible optical module, according to AES10-1991. More information can be found in chapter 30.1, MADI Basics.
HDSP MADI includes automatic input selection (Safe Mode Input). In case the current input
signal fails, the unit switches to the other input immediately. This mode, called redundancy
mode, offers improved safety against errors on the transmission line. Switching the inputs is
done in about one second. Redundancy operation is displayed in the Settings dialog.
The BNC output is built according to AES10-1991. The output's impedance is 75 Ohm. The
output voltage will be 600 mVpp when terminated with 75 Ohm. Changing the internal blue
jumper X4 to the upper position, the output voltage is increased to 1.2 Vpp. This setting is not
intended to be used in normal operation. But in case of a very long or 'lossy' coaxial cable, this
setting may ensure an error-free operation of the transmission line.
The HDSP MADI offers two MIDI I/O via 5-pin DIN connectors. The MIDI ports are added to the
system by the driver. Using MIDI capable software, these ports can be accessed under the
name MADI MIDI. Using more than one HDSP MADI, a consecutive number is added to the
port name, like MADI MIDI In 1 (2) etc.
The MIDI In port is available for both GSIF (GSIF-2 Low Latency) and standard MME MIDI
simultaneously.
The third MIDI port, MADI MIDI In 3 (1) and MADI MIDI Out 3 (1), receives and transmits MIDI
data via MADI. This allows for a direct communication between systems with HDSP MADI
cards. Additionally MIDI data can be transmitted from/to RME's ADI-648 and ADI-642. Also
ADI-648 and ADI-642 can be MIDI remote controlled without any additional line or cabling between computer (MADI card) and both units.
23. Word Clock
23.1 Word Clock Input and Output
SteadyClock guarantees an excellent performance in all clock modes. Based on the highly
efficient jitter suppression, the HDSP MADI refreshes and cleans up any clock signal, and provides it as reference clock at the BNC output (see chapter 30.6).
Input
The HDSP MADI's transformer isolated word clock input is active when Pref. Sync Ref in the
Settings dialog has been switched to Word Clock, the clock mode AutoSync has been activated, and a valid word clock signal is present. The signal at the BNC input can be Single,
Double or Quad Speed, the HDSP MADI automatically adapts to it. As soon as a valid signal is
detected, the green LED is lit, and the Settings dialog shows either Lock or Sync (see chapter
30.2).
Thanks to RME's Signal Adaptation Circuit, the word clock input still works correctly even with
heavily mis-shaped, dc-prone, too small or overshoot-prone signals. Thanks to automatic signal
centering, 300 mV (0.3V) input level are sufficient in principle. An additional hysteresis reduces
sensitivity to 1.0 V, so that over- and undershoots and high frequency disturbances don't cause
a wrong trigger.
The word clock input is shipped as high impedance type (not terminated). A push switch allows
to activate internal termination (75 Ohms). The switch is found beside the word clock input
socket. Use a small pencil or similar and carefully push the blue switch so that it snaps into its
lock position. The yellow LED will be lit when termination is active. Another push will release it
again and de-activate the termination.
Output
The word clock output of the HDSP MADI is constantly active, providing the current sample
frequency as word clock signal. As a result, in Master mode the provided word clock is defined
by the currently used software or the DDS dialog. In Slave mode the provided frequency is
identical to the one present at the currently chosen clock input. When the current clock signal
fails, the HDSP MADI switches to Master mode and adjusts itself to the next, best matching
frequency (44.1 kHz, 48 kHz etc.).
Selecting Single Speed in the Settings dialog (Word Clock Out) causes the output signal to
always stay within the range of 32 kHz to 48 kHz. So at 96 kHz and 192 kHz sample rate, the
output word clock is 48 kHz.
The received word clock signal can be distributed to other devices by using the word clock output. With this the usual T-adapter can be avoided, and the HDSP MADI operates as Signal Refresher. This kind of operation is highly recommended, because
• input and output are phase-locked and in phase (0°) to each other
• SteadyClock removes nearly all jitter from the input signal
• the exceptional input (1 Vpp sensitivity instead of the usual 2.5 Vpp, dc cut, Signal Adapta-
tion Circuit) plus SteadyClock guarantee a secure function even with highly critical word
clock signals
• the Expansion Board provides two word clock outputs with separated driver stages
Thanks to a low impedance, but short circuit proof output, the HDSP MADI delivers 4 Vpp to 75
Ohms. For wrong termination with 2 x 75 Ohms (37.5 Ohms), there are still 3.3 Vpp fed into the
network – per output!
23.2 Technical Description and Usage
In the analog domain one can connect any device to another device, a synchronization is not
necessary. Digital audio is different. It uses a clock, the sample frequency. The signal can only
be processed and transmitted when all participating devices share the same clock. If not, the
signal will suffer from wrong samples, distortion, crackle sounds and drop outs.
AES/EBU, SPDIF, ADAT and MADI are self-clocking, an additional word clock connection in
principle isn't necessary. But when using more than one device simultaneously problems are
likely to happen. For example any self-clocking will not work in a loop cabling, when there is no
'master' (main clock) inside the loop. Additionally the clock of all participating devices has to be
synchronous. This is often impossible with devices limited to playback, for example CD players,
as these have no SPDIF input, thus can't use the self clocking technique as clock reference.
In a digital studio synchronisation is maintained by connecting all devices to a central sync
source. For example the mixing desk works as master and sends a reference signal, the word
clock, to all other devices. Of course this will only work as long as all other devices are
equipped with a word clock or sync input, thus being able to work as slave (some professional
CD players indeed have a word clock input). Then all devices get the same clock and will work
in every possible combination with each other.
Remember that a digital system can only have one master!
But word clock is not only the 'great problem solver', it also has some disadvantages. The word
clock is based on a fraction of the really needed clock. For example SPDIF: 44.1 kHz word
clock (a simple square wave signal) has to be multiplied by 256 inside the device using a special PLL (to about 11.2 MHz). This signal then replaces the one from the quartz crystal. Big
disadvantage: because of the high multiplication factor the reconstructed clock will have great
deviations called jitter. The jitter of a word clock is multiple times higher than the one of a
quartz based clock.
The end of these problems should have been the so called Superclock, which uses 256 times
the word clock frequency. This equals the internal quartz frequency, so no PLL for multiplying is
needed and the clock can be used directly. But reality was different, the Superclock proved to
be much more critical than word clock. A square wave signal of 11 MHz distributed to several
devices - this simply means to fight with high frequency technology. Reflections, cable quality,
capacitive loads - at 44.1 kHz these factors may be ignored, at 11 MHz they are the end of the
clock network. Additionally it was found that a PLL not only generates jitter, but also also rejects
disturbances. The slow PLL works like a filter for induced and modulated frequencies above
several kHz. As the Superclock is used without any filtering such a kind of jitter and noise suppression is missing. No wonder Superclock did not become a commonly accepted standard.
The actual end of these problems is offered by the SteadyClock technology of the HDSP
MADI. Combining the advantages of modern and fastest digital technology with analog filter
techniques, re-gaining a low jitter clock signal of 22 MHz from a slow word clock of 44.1 kHz is
no problem anymore. Additionally, jitter on the input signal is highly rejected, so that even in
real world usage the re-gained clock signal is of highest quality.
This is especially true when extracting the word clock out of a MADI signal. Caused by the
MADI format itself, such a signal will have around 80 (!) ns of jitter, which is reduced to about 1
(!) ns by SteadyClock.
23.3 Cabling and Termination
Word clock signals are usually distributed in the form of a network, split with BNC T-adapters
and terminated with resistors. We recommend using off-the-shelf BNC cables to connect all
devices, as this type of cable is used for most computer networks. You will find all the necessary components (T-adapters, terminators, cables) in most electronics and/or computer stores.
Ideally, the word clock signal is a 5 Volt square wave with the frequency of the sample rate, of
which the harmonics go up to far above 500 kHz. To avoid voltage loss and reflections, both
the cable itself and the terminating resistor at the end of the chain should have an impedance
of 75 Ohm. If the voltage is too low, synchronization will fail. High frequency reflection effects
can cause both jitter and sync failure.
Unfortunately there are still many devices on the market, even newer digital mixing consoles,
which are supplied with a word clock output that can only be called unsatisfactory. If the output
breaks down to 3 Volts when terminating with 75 Ohms, you have to take into account that a
device, of which the input only works from 2.8 Volts and above, does not function correctly
already after 3 meter cable length. So it is not astonishing that because of the higher voltage,
word clock networks are in some cases more stable and reliable if cables are not terminated at
all.
Ideally all outputs of word clock delivering devices are designed with very low impedance, but
all word clock inputs with high impedance, in order to not weaken the signal on the chain. But
there are also negative examples, when the 75 Ohms are built into the device and cannot be
switched off. In this case the network load is often 2 x 75 Ohms, and the user is forced to buy a
special word clock distributor. Note that such a device is generally recommended for bigger
studios.
The HDSP MADI's word clock input can be high-impedance or terminated internally, ensuring
maximum flexibility. If termination is necessary (e.g. because the card is the last device in the
chain), activate the switch TERM between the BNC jacks on the Expansion Board so that the
yellow TERM LED lights up.
In case the HDSP MADI resides within a chain of devices receiving word clock, plug a Tadapter into its BNC input jack, and the cable supplying the word clock signal to one end of the
adapter. Connect the free end to the next device in the chain via a further BNC cable. The last
device in the chain should be terminated using another T-adapter and a 75 Ohm resistor
(available as short BNC plug). Of course devices with internal termination do not need Tadaptor and terminator plug.
Due to the outstanding SteadyClock technology of the HDSP MADI, we recommend not to
pass the input signal via T-adapter, but to use the card's word clock output instead. Thanks
to SteadyClock, the input signal will both be freed from jitter and - in case of loss or drop out
– be reset to a valid frequency.
The HDSP MADI's word clock input is active when Pref. Sync Ref in the Settings dialog has
been set to Word Clock, the clock mode AutoSync has been activated, and a valid word clock
signal is present. The signal at the BNC input can be Single, Double or Quad Speed, the HDSP
automatically adapts to it. As soon as a valid signal is detected, the green LED at the bracket is
lit, and the Settings dialog shows either Lock or Sync (see chapter 30.2). In the status display
AutoSync Ref the display changes to Word. This message has the same function as the green
Lock LED, but appears on the monitor, i.e. the user can check immediately whether a valid
word clock signal is present and is currently being used.
In the line Freq., the AutoSync Ref shows the frequency of the reference signal, measured by
the hardware. In this case the frequency of the current wordclock signal is measured and displayed.
The HDSP MADI includes a powerful digital real-time mixer, the Hammerfall DSP mixer, based
on RME’s unique, sample-rate independent TotalMix technology. It allows for practically unlimited mixing and routing operations, with all inputs and playback channels simultaneously, to any
hardware outputs.
Here are some typical applications for TotalMix:
• Setting up delay-free submixes (headphone mixes). The HDSP MADI allows for up to 32 (!)
fully independent stereo submixes. On an analog mixing desk, this would equal 64 (!) Aux
sends.
• Unlimited routing of inputs and outputs (free utilisation, patchbay functionality).
• Distributing signals to several outputs at a time. TotalMix offers state-of-the-art splitter and
distributor functions.
• Simultaneous playback of different programs using only one stereo output. The ASIO multi-
client driver allows to use several programs at the same time, but only on different playback
channels. TotalMix provides the means to mix and monitor these on a single stereo output.
• Mixing of the input signal to the playback signal (complete ASIO Direct Monitoring). RME
not only is the pioneer of ADM, but also offers the most complete implementation of the
ADM functions.
• Integration of external devices. Use TotalMix to insert external effects devices, be it in the
playback or in the record path. Depending on the current application, the functionality equals
insert or effects send and effects return, for example as used during real-time monitoring
when adding some reverb to the vocals.
Every single input channel, playback channel and hardware output features a Peak and RMS
level meter, calculated in hardware (hardware output is Peak only). These level displays are
very useful to determine the precense and routing destinations of the audio signals.
For a better understanding of the TotalMix mixer you should know the following:
• As shown in the block diagram (next page), the record signal usually stays un-altered. To-
talMix does not reside within the record path, and does not change the record level or the
audio data to be recorded (exception: loopback mode).
• The hardware input signal can be passed on as often as desired, even with different levels.
This is a big difference to conventional mixing desks, where the channel fader always controls the level for all routing destinations simultaneously.
• The level meter of inputs and playback channels are connected pre-fader, to be able to
visually monitor where a signal is currently present. The level meters of the hardware’s outputs are connected post-fader, thus displaying the actual output level.
The visual design of the TotalMix mixer is a result of its capability to route hardware inputs and
software playback channels to any hardware output. The HDSP MADI provides 64 input channels, 64 software playback channels, and 64 hardware output channels:
128 channels don't fit on the screen side by side, neither does such an arrangement provide a
useful overview. The input channel should be placed above the corresponding output channel.
Therefore, the channels have been arranged as known from an Inline desk, so that the row
Software Playback equals the Tape Return of a real mixing desk:
• Top row: Hardware inputs. The level shown is that of the input signal, i. e. fader independ-
ent. Via fader and routing field, any input channel can be routed and mixed to any hardware
output (bottom row).
• Middle row: Playback channels (playback tracks of the audio software). Via fader and rout-
ing field, any playback channel can be routed and mixed to any hardware output (third row).
• Bottom row (third row): Hardware outputs. Here, the total level of the output can be ad-
justed. This may be the level of connected loudspeakers, or the necessity to reduce the
level of an overloaded submix.
The following chapters explain step by step all functions of the user interface.
A single channel consists of various elements:
Input channels and playback channels each have a mute and solo button.
Below there is the panpot, realized as indicator bar (L/R) in order to save space.
In the field below, the present level is displayed in RMS or Peak, being updated about
every half a second. Overs (overload) are indicated here by an additional red dot.
Next is the fader with a level meter. The meter shows both peak values (zero attack, 1
sample is enough for a full scale display) by means of a yellow line, and mathematically correct RMS values by means of a green bar. The RMS display has a relatively
slow time constant, so that it shows the average loudness quite well.
Below the fader, the current gain and panorama values are shown.
The white area shows the channel name. Selecting one or more channels is done by
clicking on the white label which turns orange then. A right mouse click opens a dialog
to type in a new name.
The black area (routing field) shows the current routing target. A mouse click opens the routing
window to select a routing target. The list shows all currently activated routings by checkmarks
in front of the routing targets.
24.4 Tour de TotalMix
This chapter is a practical guide and introduction on how to use TotalMix, and on how TotalMix
works.
Starting up TotalMix the last settings are recalled automatically. When executing the application for the first time, a default file is loaded, sending all playback tracks 1:1 to the corresponding hardware outputs with 0 dB gain, and activating phones monitoring.
Hold down Ctrl and click on preset button 5 to make sure that factory preset 5 is loaded. The
faders in the top row are set to maximum attenuation (called m.a. in the following), so there is
no monitoring of the input channels. The Submix View is active, therefore for improved overview all outputs except 63/64 are greyed out. Additionally all faders are set to the current routing target 63/64. All faders of the middle row are set to 0 dB, so no matter on which channels a
playback happens, the audio will be audible via the Phones output. Just try it!
We will now create a submix on outputs 1+2. Please start a multitrack playback. In the third
row, click on the channels of hardware output 1 or 2. The Submix View changes to 1/2. Both
the fader settings and the output levels of all other channels are still visible, but greyed out for
improved orientation.
As soon as 1/2 became active, all faders of the second row jumped to their bottom position –
except those of playback channels 1/2. This is correct, because as mentioned above the factory preset includes a 1:1 routing. Click on 3/4 and the faders above are the only active ones,
same for 5/6 and so on.
Back to 1/2. Now you can change all the faders of all inputs and playback channels just as you
like, thus making any input and playback signals audible via the outputs 1/2. The panorama
can be changed too. Click into the area above the fader and drag the green bar in order to set
the panorama between left and right. The level meters of the third row display the level
changes in real-time.
As shown it is very easy to set up a specific submix for whatever output: select output channel,
set up fader and pans of inputs and playbacks – ready!
For advanced users sometimes it makes sense to work without Submix
View. Example: you want to see and set up some channels of different
submixes simultaneously, without the need to change between them all
the time. Switch off the Submix View by a click on the green button. Now
the black routing fields below the faders no longer show the same entry
(1+2), but completely different ones. The fader and pan position is the
one of the individually shown routing destination.
In playback channel 1 (middle row), labeled Out 1, click onto the routing
field below the label. A list pops up, showing a checkmark in front of
'1+2' and '63+64'. So currently playback channel 1 is sent to these two
routing destinations. Click onto '7+8'. The list disappears, the routing
field no longer shows '1+2', but '7+8'. Now move the fader with the
mouse. As soon as the fader value is unequal m.a., the present state is
being stored and routing is activated. Move the fader button to around 0
dB. The present gain value is displayed below the fader in green letters.
In the lower row, on channel 7, you can see the level of what you are
hearing from output 7. The level meter of the hardware output shows the
outgoing level. Click into the area above the fader and drag the mouse
in order to set the panorama, in this case the routing between channels 7
and 8. The present pan value is also being displayed below the fader.
Please carry out the same steps for Out 2 now, in order to route it to
output 8 as well.
In short: While editing the Submix 7+8 you have direct access to other
submixes on other channels, because their routing fields are set to different destinations. And you get a direct view of how their faders and
panoramas are set up.
This kind of visual presentation is a mighty one, but for many users it
is hard to understand, and it requires a deep understanding of complex routing visualizations. Therefore we usually re-commend to work
in Submix View.
Often signals are stereo, i. e. a pair of two channels. It is therefore helpful to be able to make the routing settings for two channels at once. Hold
down the Ctrl-key and click into the routing field of Out 3. The routing list
pops up with a checkmark at '3+4'. Select '7+8'. Now, Out 4 has already
been set to '7+8' as well.
When you want to set the fader to exactly 0 dB, this can be difficult,
depending on the mouse configuration. Move the fader close to the 0
position and now press the Shift-key. This activates the fine mode, which
stretches the mouse movements by a factor of 8. In this mode, a gain
setting accurate to 0.1 dB is no problem at all.
Please set Out 4 to a gain of around -20 dB and the pan close to center.
Now click onto the routing field. You'll now see three checkmarks, at
'3+4', '7+8' and '63+64'. Click onto '61+62'. The window disappears,
fader and pan jump to their initial values, the signal can now be routed to
the outputs 61 and 62. You can continue like this until all entries have
got a checkmark, i. e. you can send the signal to all outputs simultaneously.
You will certainly have noticed that the signal at the outputs 7/8 did not change while you were
routing channel 4 to other outputs and setting different gain values for those. With all analog
and most digital mixing desks, the fader setting would affect the level for every routed bus - not
so for TotalMix. TotalMix allows for setting all fader values individually. Therefore the faders
and the panpots jump to the appropriate setting as soon as another routing is chosen.
Sometimes you will want the routings not to be independent. Let's say you have sent a signal to
several submixes, and now want to change the signal's volume a bit on all these submixes.
Dragging the faders by use of the right mouse button activates Post Send mode and causes all
routings of the current input or playback channel to be changed in a relative way. Please note
that the fader settings of all routings are memorized. So when pulling the fader to the bottom
(maximum attenuation), the individual settings are back when you right click the mouse and
pull the fader up. The individual settings get lost in m.a. position as soon as the fader is clicked
with the left mouse button. As long as no single level is at m.a. position, the left mouse button
can be used to change the current routing's gain.
The checkmarks are un-checked by moving the fader to m.a. This setting deactivates the routing...why route if there is no level? Click onto '7+8' in the routing window, pull the fader down,
open the routing window again - the checkmark is gone.
The number of channels is reduced automatically when entering Double Speed mode (96 kHz).
The display is adjusted accordingly, and all fader settings remain stored.
24.5 Submix View
Such a wide range of possibilities make it difficult to maintain the overview. Because practically
all hardware outputs can be used for different submixes, as shown (up to 32 completely independent stereo submixes, 16 4-channel submixes etc.). And when opening the routing windows
you might see an army of checkmarks, but you don't get an overwiev, i.e., how the signals
come together and where. This problem is solved by Submix View mode. In this mode, all
routing fileds jump to the routing pair just being selected. You can then see immediately, which
channels, which fader and pan settings make a submix (for example '1+2'). At the same time
the Submix View simplifies setting up the mixer, as all channels can be set simultaneously to
the same routing destination with just one click.
Changing to a different destination (output channel) is done in any routing field, or by a click on
the desired output pair in the bottom row.
24.6 Mute und Solo
Mute operates pre-fader, thus mutes all currently active routings of the channel. As soon as any
Mute button is pressed, the Mute Master button lights up in the Quick Access area. It allows to
switch all selected mutes off and on again. You can comfortably make mute-groups or activate
and deactivate several Mutes simultaneously.
The same holds true for the Solo and the Solo Master buttons. As with conventional mixing
desks, Solo operates only for the output defined as Monitor Main, as a solo-in-place, post
fader. As soon as one Solo button is pressed, the Solo Master button lights up in the Quick
Access area. It allows to switch all selected Solos off and on again. You can comfortably make
solo-groups or activate and deactivate several Solos simultaneously.
This section includes additional options, further improving the handling of TotalMix. The Master
buttons for Mute and Solo have already been described, they allow for group-based working
with these functions.
In the View section the single mixer rows can be made visible or invisible. If the inputs are not
needed for a pristine playback mix, the whole upper row falls out of the picture after a click on
the Input button. If the hardware outputs don't interest you either, the window can thus be reduced to the playback channels to save space. All combinations are possible and allowed.
As described earlier, Submix sets all routing windows to the same selection. Deactivating
Submix automatically recalls the previous view. The mixer can be made smaller horizontally
and vertically. This way TotalMix can be made substantially smaller and space-saving on the
desktop/screen, if you have to monitor or set only a few channels or level meters.
The Presets are one of the mightiest and most useful features of TotalMix.
Behind the eight buttons, eight files are hidden (see next chapter). These contain the complete mixer state. All faders and other settings follow the changing
of preset(s) in real-time, just by a single mouse click. The Save button allows
for storing the present settings in any preset. You can change back and forth
between a signal distribution, complete input monitoring, a stereo and mono
mix, and various submixes without any problem.
Also here, RME's love for details can be seen. If any parameter is being altered
after loading a preset (e. g. moving a fader), the preset display flashes in order
to announce that something has been changed, still showing which state the
present mix is based on.
If no preset button is lit, another preset had been loaded via the File menu and
Open file. Mixer settings can of course be saved the usual way, and have long
file names.
Instead of single presets a complete bank of (8) presets can be loaded at once. Advantage:
The names defined for the preset buttons will be stored and loaded automatically.
Up to three HDSP MADI and AES-32 can be used simultaneously. The Unit buttons switch
between the cards. Holding down Ctrl while clicking on button Unit 2 or Unit 3 will open another
TotalMix window.
24.8 Presets
TotalMix includes eight factory presets, stored within the program. The user presets can be
changed at any time, because TotalMix stores and reads the changed presets from the files
preset11.mad to preset81.mad, located in Windows' hidden directory Documents and Set-
tings, <Username>, Local Settings, Application Data, RME TotalMix. On the Mac the location is
in the folder User, <Username>, Library / Preferences / Hammerfall DSP. The first number
indicates the current preset, the second number the current unit.
This method offers two major advantages:
• Presets modified by the user will not be overwritten when reinstalling or updating the driver
• The factory presets remain unchanged, and can be reloaded any time.
Mouse: The original factory presets can be reloaded by holding down the Ctrlkey and clicking on any preset button. Alternatively the files described above
can be renamed, moved to a different directory, or being deleted.
Keyboard: Using Ctrl and any number between 1 and 8 (not on the numeric
keypad!) will load the corresponding factory default preset. The key Alt will load
the user presets instead.
When loading a preset file, for example 'Main Monitor AN 1_2 plus headphone mix 3_4.mad',
the file name will be displayed in the title bar of the TotalMix window. Also when loading a preset by the preset buttons, the name of the preset is displayed in the title bar. This way it is always clear what the current TotalMix state is based on.
The eight factory presets offer a pretty good base to modify them to your personal needs. In all
factory presets Submix View is active by default.
Preset 1
Description: All channels routed 1:1, monitoring of all playback channels.
Details: All inputs maximum attenuation. All playback channels 0 dB, routed to the same out-
put. All outputs 0 dB. Level display set to RMS +3 dB. View Submix active.
Note: This preset is Default, offering the standard functionality of a I/O-card.
Preset 2
Description: All channels routed 1:1, input and playback monitoring. As Preset 1, plus 1:1 pass
through of all inputs.
Preset 3
Description: All channels routed 1:1, no input and no playback monitoring. All faders set to
maximum attenuation.
Preset 4
Description: All channels routed 1:1, input and playback monitoring. As Preset 2, but all inputs
muted.
Preset 5
Description: All channels routed 1:1, playback monitoring. Submix of all playback channels to
channels 63/64 (phones monitoring). Hardware output 63/64 selected and at –12 dB.
Preset 6
Description: As prest 5, but submix of all input channels to channels 63/64 (phones monitoring).
Preset 7
Description: As preset 5, but submix of all input and playback channels to channels 63/64
(phones monitoring).
Preset 8
Description: Panic. As Preset 4, but playback channels muted too (no output signal).
Preset Banks
Instead of a single preset, all eight presets can be stored and loaded at once. This is done via
Menu File, Save All Presets as and Open All Presets (file suffix .map). After the loading the
presets can be activated by the preset buttons. In case the presets have been renamed (see
chapter 24.11), these names will be stored and loaded too.
The Monitor panel provides several options usually found on analog mixing desks. It offers
quick access to monitoring functions which are needed all the time in typical studio work.
Monitor Main
Use the drop down menu to select the hardware outputs where your main monitors are connected to.
Dim
A click on this button will lower the volume of your main monitor output (see
above) by an amount set up in the Preferences dialog (see below). This is the
same as moving the third row faders down a bit, but much more convenient, as
the old setting is back by a simple mouse click.
Mono
Sets the stereo output defined above to monaural playback. Useful to check for
mono compatibility and phase problems.
Talkback
A click on this button will dim the Main Monitor output (see above) by an amount
set up in the Preferences dialog. At the same time the control room's microphone signal (source defined in Preferences) is sent to the three destinations
described below as Monitor Phones. The mic level is adjusted with the channel's
input fader.
Monitor Phones 1/2/3
Use the drop down menu to select the hardware outputs where the submixes are sent to. These
submixes are usually phones mixdowns for the musicians. A click on the button allows to hear
the specific submix via the Main Monitor outputs. So when setting up or modifying the submix
for the musician this process can be monitored easily and any time. Or in other words: you can
easily check other hardware outputs/submixes by using the Monitor Phones function, without
the need to copy/paste routings back and forth, or to reconfigure the cabling at the hardware.
24.10 Preferences
The dialog box Preferences is available via the menu
Options or directly via F3.
Talkback
Input: Select the input channel of the Talkback signal
(microphone in control room).
Dim: Amount of attenuation of the Main Monitor output in dB.
Listenback
Input: Select the input channel of the Listenback signal (microphone in recording room).
Dim: Amount of attenuation of the Monitor Phones
outputs in dB.
Note: The Mute button of the Talkback and Listenback
channel is still active. Therefore it is not necessary to
select <NONE>, in case one of both shall be deactivated.
Dim: Amount of attenuation of the Main Monitor output in dB. Activated by the Dim button in
the Monitor panel.
Stereo Pan Law
The Pan Law can be set to -6 dB, -4.5 dB, -3 dB and 0 dB. The value chosen defines the level
attenuation in pan center position. This setting is useful because the ASIO host often supports
different pan laws too. Selecting the same value here and in the ASIO host, ASIO Direct Monitoring works perfectly, as both ASIO host and TotalMix use the same pan law. Of course, when
not using ADM it can be changed to a setting different from the factory preset of –6 dB as well.
You will most probably find that -3 dB gives a much more stable loudness when moving an
object between left and right.
24.11 Editing the Names
The channel names shown in the white label area can be
edited. A right mouse click on the white name field brings up
the dialog box Enter Name. Any name can be entered in this
dialog. Enter/Return closes the dialog box, the white label
now shows the first letters of the new name. ESC cancels the
process and closes the dialog box.
Moving the mouse above the label
brings up a tool tip with the complete
name.
The hardware outputs (third row) can be edited in the same
way. In this case, the names in the routing drop down
menus will change automatically. Additionally the names in
the drop down menus of the Monitor section will change as
well.
The preset buttons can get meaningful
names in the same way. Move the mouse
above a preset button, a right mouse
click will bring up the dialog box. Note
that the name shows up as tool tip only,
as soon as the mouse stays above the
preset button.
The preset button names are not stored in the preset files, but globally in the registry, so won't
change when loading any file or saving any state as preset. But loading a preset bank (see
chapter 24.8) the names will be updated.
In many situations TotalMix can be controlled quickly and comfortably by the keyboard, making
the mixer setup considerably easier and faster. The Shift-key for the fine mode for faders and
panpots has already been mentioned. The Ctrl-key can do far more than changing the routing
pairwise:
• Clicking anywhere into the fader area with the Ctrl-key pressed, sets the fader to 0 dB.
• Clicking anywhere into the pan area with the Ctrl-key pressed, sets the panorama to <C>
meaning Center.
• Clicking a preset button while holding down Ctrl, the original factory preset will be loaded.
• Using Ctrl and any number between 1 and 8 (not on the numeric keypad!) will load the cor-
responding factory default preset. Alt plus number loads the user preset.
• Using multiple HDSP MADIs, clicking the button Unit 2 while holding down Ctrl opens a
second TotalMix window for the second HDSP system, instead of replacing the window contents.
The faders can also be moved pairwise, corresponding to the stereo-routing settings. This is
achieved by pressing the Alt-key and is especially comfortable when setting the SPDIF and
Phones output level. Even the panoramas can be operated with Alt, from stereo through mono
to inversed channels, and also the Mute and Solo buttons (ganged or inversed switching!).
At the same time, TotalMix also supports combinations of these keys. If you press Ctrl and Alt
at the same time, clicking with the mouse makes the faders jump to 0 dB pairwise, and they
can be set pairwise by Shift-Alt in fine mode.
Also very useful: the faders have two mouse areas. The first area is the fader button, which can
be grabbed at any place without changing the current position. This avoids unwanted changes
when clicking onto it. The second area is the whole fader setting area. Clicking into this area
makes the fader jump to the mouse at once. If for instance you want to set several faders to
m.a., it is sufficient to click onto the lower end of the fader path. Which happens pairwise with
the Alt-key pressed.
Using the hotkeys I, O and P the complete row of Input, Playback and Output channels each
can be toggled between visible and invisible. Hotkey S switches Submix view on/off. Those
four hotkeys have the same functionality as the buttons in the View section of the Quick Access Panel. The Level Meter Setup dialog can be opened via F2 (as in DIGICheck). The dialog
box Preferences is opened via F3.
Hotkey M toggles Mute Master on/off (and with this performs a global mute on/off). Hotkey X
toggles the Matrix view on/off (see chapter 25), hotkey T the mixer view. Hotkey L links all
faders as stereo pairs.
Further hotkeys are available to control the configuration of the Level Meter (see chapter
24.14):
Key 4 or 6: Display range 40 or 60 dB
Key E or R: Numerical display showing Peak or RMS
Key 0 or 3: RMS display absolute or relative to 0 dBFS
Always on Top: When active (checked) the TotalMix window will always be on top of the Win-
dows desktop.
Note: This function may result in problems with windows containing help text, as the TotalMix
window will even be on top of those windows, so the help text isn't readable.
Deactivate Screensaver: When active (checked) any activated Windows screensaver will be
disabled temporarily.
Ignore Position: When active, the windows size and position stored in a file or preset will not
be used. The routing will be activated, but the window will not change.
ASIO Direct Monitoring (Windows only): When de-activated any ADM commands will be
ignored by TotalMix. In other words, ASIO Direct Monitoring is globally de-activated.
Link Faders: Selecting this option all faders will be treated as stereo pairs and moved pairwise. Hotkey L.
Level Meter Setup: Configuration of the Level Meters. Hotkey F2. See chapter 24.14.
Preferences: Opens a dialog box to configure several functions, like Pan Law, Dim, Talkback
Dim, Listenback Dim. See chapter 24.10.
Enable MIDI Control: Turns MIDI control on.The channels which are currently under MIDI
control are indicated by a colour change of the info field below the faders, black turns to yellow.
Deactivate MIDI in Background: Disables the MIDI control as soon as another application is
in the focus, or in case TotalMix has been minimized.
The HDSP MADI calculates all the display values Peak, Over and RMS in hardware, in order to
be capable of using them independent of the software in use, and to significantly reduce the
CPU load.
Tip: This feature, the Hardware Level Meter, is used by DIGICheck (Windows only, see chapter 16) to display Peak/RMS level meters off all channels, nearly without any CPU load.
The level meters integrated in TotalMix - considering their size - cannot be compared with
DIGICheck. Nevertheless they already include many useful functions.
Peak and RMS is displayed for every channel. 'Level Meter Setup' (menu Options or F2) and
direct keyboard entry (hotkeys) make various options available:
• Display range 40 or 60 dB (hotkey 4 or 6)
• Release time of the Peak display (Fast/Medium/Slow)
• Numerical display selectable either Peak or RMS (Hotkey E or R)
• Number of consecutive samples for Overload display (1 to 15)
• RMS display absolute or relative to 0 dBFS (Hotkey 3 or 0)
The latter is a point often overlooked, but nonetheless
important. A RMS measurement shows 3 dB less for
sine signals. While this is mathematically correct, it is
not very reasonable for a level meter. Therefore the
RMS readout is usually corrected by 3 dB, so that a full
scale sine signal shows 0 dBFS on both Peak and
RMS meters. This setting also yields directly readable
signal-to-noise values. Otherwise the value shown with
noise is 3 dB better than it actually is (because the
reference is not 0 dB, but -3 dB). For example in
WaveLab.
The value displayed in the text field is independent of
the setting 40/60 dB, it represents the full 24 bit range
of the RMS measurement, thus making possible a SNR
measurement 'RMS unweighted', which you would
otherwise need extremely expensive measurement
devices for. An example: An RME ADI-8 DS connected
to the HDSP MADI's ADAT port will show around -113
dBFS on all eight channel's input level meters.
This level display will constantly bring the reduced dynamic range of your equipment, maybe of
the whole studio, in front of your eyes. Nice to have everything 24 bit - but still noise and hum
everywhere in the range around -90 dB or worse... sorry, but this is the disappointing reality.
The up-side about it is that TotalMix allows for constantly monitoring the signal quality without
effort. Thus it can be a valuable tool for sound optimization and error removal in the studio.
Measuring SNR (Signal to Noise) requires to press R (for RMS) and 0 (for referring to 0
dBFS, a full scale signal). The text display will then show the same value as an expensive
measurement system, when measuring ‘RMS unweighted’.
The mixer window of TotalMix looks and operates similar to mixing desks, as it is based on a
conventional stereo design. The matrix display presents a different method of assigning and
routing channels, based on a single channel or monaural design. The matrix view of the HDSP
looks and works like a conventional patchbay, adding functionality way beyond comparable
hardware and software soutions. While most patchbays will allow you to connect inputs to outputs with just the original level (1:1, or 0 dB, as known from mechanical patchbays), TotalMix
allows you to use a freely definable gain value per crosspoint.
Matrix and TotalMix are different ways of displaying the same processes. Because of this both
views are always fully synchronized. Each change in one view is immediately reflected in the
other view as well.
25.2 Elements of the Matrix View
The visual design of the TotalMix Matrix is mainly determined by the architecture of the HDSP
system:
• Horizontal labels: All hardware outputs
• Vertical labels: All hardware inputs. Below are all play-
back channels (software playback channels)
• Green 0.0 dB field: Standard 1:1 routing
• Black gain field: Shows the current gain value as dB
• Orange gain field: This routing is muted.
To maintain overview when the window size has been reduced, the left and upper labels are
floating. They won't left the visible area when scrolling.
25.3 Operation
Using the Matrix is a breeze. It is very easy to indentify the current crosspoint, because the
outer labels light up in orange according to the mouse position.
If input 1 is to be routed to output 1, use the mouse and click one time on crosspoint In 1 / 1.
The green 0.0 dB field pops in, another click removes it. To change the gain (equals the use of
a different fader position, see simultaneous display of the mixer view), hold Ctrl down and drag
the mouse up or down, starting from the gain field. The value within the field changes accordingly. The corresponding fader in the mixer view is moving simultaneously, in case the currently modified routing is visible.
Note the difference between the left side, representing the inputs and software playback channels, and the upper side, representing the hardware outputs. Moving a fader in row 1 or 2 in
TotalMix view, only the specific levels (max. 2) of this routing will change within the Matrix. But
moving a fader in row 3 will make all vertically activated levels move at once (for example
63/64, analog output).
A gain field marked orange indicates activated mute status. Mute can only be changed in the
mixer view.
The Matrix not always replaces the mixer view, but it significantly enhances the routing capabilities and - more important - is a brilliant way to get a fast overview of all active routings. It
shows you in a glance what's going on. And since the Matrix operates monaural, it is very easy
to set up specific routings with specific gains.
Example 1: You want TotalMix to route all software outputs to all corresponding hardware outputs, and have a submix of all inputs and software outputs on the Phones output (equals factory preset 2). Setting up such a submix is easy. But how to check at a later time, that all settings are still exactly the way you wanted them to be, not sending audio to a different output?
The most effective method to check a routing in mixer view is the Submix View, stepping
through all existing software outputs, and having a very concentrated look at the faders and
displayed levels of each routing. That doesn't sound comfortably nor error-free, right? Here is
where the Matrix shines. In the Matrix view, you simply see a line from upper left to lower right,
all fields marked as unity gain. Plus two rows vertically all at the same level setting. You just
need 2 seconds to be sure no unwanted routing is active anywhere, and that all levels match
precisely!
Example 2: The Matrix allows you to set up routings which would be nearly impossible to
achieve by fiddling around with level and pan. Let's say you want to send input 1 to output 1 at
0 dB, to output 2 at -3 dB, to output 3 at -6 dB and to output 4 at -9 dB. Each time you set up
the right channel (2/4), the change in pan destroys the gain setting of the left channel (1/2). A
real hassle! In Matrix view, you simply click on the corresponding routing point, set the level via
Ctrl-mouse, and move on. You can see in TotalMix view how pan changes to achieve this special gain and routing when performing the second (fourth...) setting.
26. TotalMix Super-Features
26.1 ASIO Direct Monitoring (Windows only)
Start Samplitude, Sequoia, Cubase or Nuendo and TotalMix. Activate ADM (ASIO Direct Monitoring), and move a fader in the ASIO host. Now watch the corresponding fader in TotalMix
magically move too. TotalMix reflects all ADM gain and pan changes in realtime. Please note
that faders only move when the currently activated routing (currently visible routing) corresponds to the one in the ASIO host. Also note that the Matrix will show any change, as it shows
all possible routings in one view.
With this TotalMix has become a wonderful debugging tool for ADM. Just move the host's fader
and pan, and see what kind of ADM commands TotalMix receives.
The hardware output row faders are included in all gain calculations, in every possible way.
Example: you have lowered the output level of a submix, or just a specific channel, by some
dB. The audio signal passed through via ADM will be attenuated by the value set in the third
row.
Click on the white name label of channel 1 and 2 in TotalMix. Be sure to have channel 3's fader
set to a different position and click on its label too. All three labels have changed to the colour
orange, which means they are selected. Now moving any of these faders will make the other
faders move too. This is called 'building a group of faders', or ganging faders, maintaining their
relative position.
Building groups or ganging can be done in any row, but is limited to operate horizontally within
one row. If you usually don't need this, you can at least gang the analog outputs. The advantage over holding the Alt-key is that Alt sets both channels to the same level (can be handy
too), while grouping via selection will retain any offset (if you need one channel to be louder all
the time etc.).
Note: The relative positions are memorized until the faders are pulled down so that they reach
upper or lower maximum position and the group is changed (select another channel or deselect
one of the group).
Tip: Gang some submixes and watch all routing levels change in the Matrix view.
26.3 Copy Routings to other Channels
TotalMix allows to copy complete routing schemes of inputs and outputs.
Example 1: You have input 5 (guitar) routed within several submixes/hardware outputs (=
headphones). Now you'll get another input with keyboards that should appear in the same way
on all headphones. Select input 5, open the menu Edit. It shows 'Copy In 5'. Now select the
desired new input, for example In 8. The menu now shows 'Paste In 5 to In 8'. Click on it done. If you are familiar with this functionality just use Ctrl-C and Ctrl-V. Else the self updating
menu will always let you know what actually will happen.
Tip: Have the Matrix window open as second window when doing this. It will show the new
routings immediately, so copying is easier to understand and to follow.
Example 2: You have built a comprehensive submix on outputs 4/5, but now need the exact
same signal also on the outputs 6/7. Click on Out 4, Ctrl-C, click on Out 6, Ctrl-V, same with
5/7 - you're done!
The Matrix shows you the difference between both examples. Example 1 means copying lines
(horizontally), while example 2 means copying rows (vertically).
Example 3: Let's say the guitarist finished his recording, and you now need the same signal
again on all headphones, but this time it comes from the recording software (playback row). No
problem, you can even copy between rows 1 and 2 (copying between row 3 and 1/2 isn't possible).
But how to select while a group is active? De-selecting the group first? Not necessary! TotalMix
always updates the copy and paste process with the last selection. This way you don't have to
de-activate any group-selections when desiring to perform a copy and paste action.
26.4 Delete Routings
The fastest way to delete complex routings: select a channel in the mixer view, click on the
menu entry Edit and select Delete. Or simply hit the Del-key. Attention: there is no undo in TotalMix, so be careful with this function!
The HDSP series uses TotalMix also for a routings of the subgroup outputs (=hardware outputs,
bottom row) to the recording software. Unfortunately this feature is not available with the HDSP
MADI, as the FPGA of the card has no resources left. Therefore this chapter describes the
loopback mode when used with an external cable loop. As the HDSP MADI has only one input,
an external cable loop will only make sense for the following examples in case the signal had
been split up into several wires. This can be done easily when using the ADI-648, splitting the
MADI line into eight 8-channel ADAT lines. Then a loopback of specific channels is possible.
A loopback is used to record the playback signal. This way, complete submixes can be recorded, the playback of a software can be recorded by another software, and several input
signals can be mixed into one record channel. Please note these important issues:
The connection of digital output and input generates a clock loop, and with this a malfunction, in case the card has not been switched into clock mode Master, or an external clock
signal of highest priority is used.
The latter is the case when the card is in AutoSync mode, and is synchronized by an external
clock signal from the input selected in Pref Sync Ref.
Connecting digital output and input can cause a digital feedback, which is more severe
than any analog one. Caution!
This is a problem for both TotalMix (monitoring an input signal to the same output channel) and
the DAW software (which usually activates monitoring in the same way).
Recording a Software's playback
In real world application, recording a software's output with another software will show the following problem: The record software tries to open the same playback channel as the playback
software (already active), or the playback one has already opened the input channel which
should be used by the record software.
This problem can easily be solved. First make sure that all rules for proper multi-client operation are met (not using the same record/playback channels in both programs). Then route the
playback signal via TotalMix to a hardware output in the range of the record software, and send
it to the record software via the loopback cable.
Mixing several input signals into one record channel
In some cases it is useful to record several sources in only one track. For example when using
two microphones when recording instruments and loudspeakers. TotalMix loopback saves an
external mixing desk. Simply route/mix the input signals all to the same output (third row), then
send this output to a record channel via the loopback cable. This way any number of input
channels from different sources can be recorded into one single track.
Note: The data recorded via loopback is delayed by about 3 samples, when using an additional
ADI-648 by about 6 samples. This value is extremely low, because the HDSP 9652 provides
digital interfacing only. And therefore the additional delay can simply be ignored.
With TotalMix a usage of external hardware - like effects devices - is easy and flexible.
Example 1: The singer (microphone input channel 1) shall have some reverb on his head-
phones (outputs 11/12). A direct routing In 1 to Out 11/12 for monitoring had been set up already. The external reverb is connected to a free output, for example channel 8. In active
mode Submix View click on channel 8 in the bottom row. Drag the fader of input 1 to about 0
dB and the panorama fully to the right. Adjust the input level at the reverb unit to an optimal
setting. Next the output of the reverb unit is connected to a free stereo input, for example 5/6.
Use the TotalMix level meters to adjust a matching output level at the reverb unit. Now click on
channels 11/12 in the bottom row, and move the fader of inputs 5/6 until the reverb effect gets
a bit too loud in the headphones. Now click on channel 8 in the bottom row again and drag
fader 1 down a bit until the mix of original signal and reverb is perfect for the singer.
The described procedure is completely identical to the one when using an analog mixing desk.
There the signal of the singer is sent to an output (usually labeled Aux), from there to a reverb
unit, sent back from the reverb unit as stereo wet signal (no original sound), back in through a
stereo input (e.g. Effect return) and mixed to the monitoring signal. The only difference: The
Aux sends on mixing desks are post-fader. Changing the level of the original signal causes a
change of the effects level (here the reverb) too, so that both always have the same ratio.
Tip: Such a functionality is available in TotalMix via the right mouse button! Dragging the faders by use of the right mouse button causes all routings of the current input or playback channel
to be changed in a relative way. This completely equals the function Aux post fader.
Example 2: Inserting an effects device can be done as above, even within the record path.
Other than in the example above the reverb unit also sends the original signal, and there is no
routing of input 1 directly to outputs 11/12. To insert an effects device like a Compressor/Limiter directly into the record path, the input signal of channel 1 is sent by TotalMix to any
output, to the Compressor, back from the Compressor to any input. This input is now selected
within the record software.
TotalMix can be remote controlled via MIDI. It is compatible to the widely spread Mackie Control protocol, so TotalMix can be controlled with all hardware controllers supporting this standard. Examples are the Mackie Control, Tascam US-2400 or Behringer BCF 2000.
Additionally, the stereo output faders (lowest row) which are set up as MonitorMain outputs in
the Monitor panel can also be controlled by the standard Control Change Volume via MIDI channel 1. With this, the main volume of the HDSP MADI is controlable from nearly any MIDI
equipped hardware device.
27.2 Setup
• Open the Preferences dialog (menu Options or F3). Select the MIDI Input and MIDI Output
port where your controller is connected to.
• When no feedback is needed (when using only standard MIDI commands instead of Mackie
Control protocol) select NONE as MIDI Output.
• Check Enable MIDI Control in the Options menu.
27.3 Operation
The channels being under MIDI control are indicated by a colour change of the info field below
the faders, black turns to yellow.
The 8-fader block can be moved horizontally and vertically, in steps of one or eight channels.
Faders can be selected to gang them.
In Submix View mode, the current routing destination (output bus) can be selected via REC Ch.
1 – 8. This equals the selection of a different output channel in the lowest row by a mouse click
when in Submix View. In MIDI operation it is not necessary to jump to the lowest row to perform
this selection. This way even the routing can be easily changed via MIDI.
Full LC Display Support: This option in Preferences (F3) activates complete Mackie Control
LCD support with eight channel names and eight volume/pan values.
Attention: this feature causes heavy overload of the MIDI port when ganging more than 2
faders! In such a case, or when using the Behringer BCF2000, turn off this option.
When Full LC Display Support is turned off, only a brief information about the first fader of the
block (channel and row) is sent. This brief information is also available on the LED display of
the Behringer BCF2000.
Tip for Mac OS X users: LC Xview (www.opuslocus.com) provides an on-screen display
emulating the hardware displays of a Logic/Mackie Control, for use with controllers that can
emulate a Logic/Mackie Control but do not have a display. Examples include the Behringer
BCF2000 and Edirol PCR series.
Deactivate MIDI in Background (menu Options) disables the MIDI control as soon as another
application is in the focus, or in case TotalMix has been minimized. This way the hardware
controller will control the main DAW application only, except when TotalMix is in the foreground. Often the DAW application can be set to become inactive in background too, so that
MIDI control is switched between TotalMix and the application automatically when switching
between both applications.
TotalMix also supports the 9th fader of the Mackie Control. This fader (labeled Master) will
control the stereo output faders (lowest row) which are set up as Main Monitor outputs in the
Monitor panel. Always and only.
27.4 Mapping
TotalMix supports the following Mackie Control surface elements*:
Element:Meaning in TotalMix:
Channel faders 1 – 8 volume
Master fader Main Monitor channel's faders
SEL(1-8) + DYNAMICS reset fader to Unity Gain
V-Pots 1 – 8 pan
pressing V-Pot knobs pan = center
CHANNEL LEFT or REWIND move one channel left
CHANNEL RIGHT or FAST FORWARD move one channel right
BANK LEFT or ARROW LEFT move eight channels left
BANK RIGHT or ARROW RIGHT move eight channels right
ARROW UP or Assignable1/PAGE+ move one row up
ARROW DOWN or Assignable2/PAGE- move one row down
EQ Master Mute
PLUGINS/INSERT Master Solo
STOP Dim Main Monitor
PLAY Talkback
PAN Mono Main Monitor
MUTE Ch. 1 – 8 Mute
SOLO Ch. 1 – 8 Solo
SELECT Ch. 1 – 8 Select
REC Ch. 1 – 8 in Submix mode only: select output bus
F1 - F8 load preset 1 - 8
F9 select Main Monitor
F10 - F12 Monitor Phones 1 - 3
*Tested with Behringer BCF2000 Firmware v1.07 in Mackie Control emulation for Steinberg mode and with Mackie Control
under Mac OS X.
The stereo output faders (lowest row) which are set up as MonitorMain outputs in the Monitor
panel can also be controlled by the standard Control Change Volume via MIDI channel 1.
With this, the main volume of the HDSP MADI is controlable from nearly any MIDI equipped
hardware device.
Even if you don't want to control all faders and pans, some buttons are highly desired to be
available in 'hardware'. These are mainly the Talkback and the Dim button, and the new monitoring options (listen to Phones submixes). Fortunately a Mackie Control compatible controller
is not required to control these buttons, as they are steered by simple Note On/Off commands
on MIDI channel 1.
The notes are (hex / decimal / keys):
Monitor Main: 3E / 62 / D 3
Dim: 5D / 93 / A 5
Mono: 2A / 42 / #F 1
Talkback: 5E / 94 / #A 5
An example of a small MIDI controller covering such MIDI functionality (and even some more)
is the Behringer BCN44. This little box has 4 pots and 8 buttons for all the above functions –
for less than 60 Euros.
27.6 Loopback Detection
The Mackie Control protocol requires feedback of the received commands, back to the hardware controller. So usually TotalMix will be set up with both a MIDI input and MIDI output. Unfortunately any small error in wiring and setup will cause a MIDI feedback loop here, which then
completely blocks the computer (the CPU).
To prevent the computer from freezing, TotalMix sends a special MIDI note every 0.5 seconds
to its MIDI output. As soon as it detects this special note at the input, the MIDI functionality is
disabled. After fixing the loopback, check Enable MIDI Control under Options to reactivate the
TotalMix MIDI.
Not all information to and around our products fit in a manual. Therefore RME offers a lot more
and detailed information in the Tech Infos. The very latest Tech Infos can be found on our
website, section News & Infos, or the directory \rmeaudio.web\techinfo on the RME Driver
CD. These are some of the currently available Tech Infos:
Synchronization II (DIGI96 series)
Digital audio synchronization - technical background and pitfalls.
Installation problems - Problem descriptions and solutions.
Driver updates Hammerfall DSP – Lists all changes of the driver updates.
DIGICheck: Analysis, tests and measurements with RME audio hardware
A description of DIGICheck, including technical background information.
ADI-8 Inside
Technical information about the RME ADI-8 (24-bit AD/DA converter).
Many background information on laptops and tests of notebooks:
HDSP System: Notebook Basics - Notebook Hardware
HDSP System: Notebook Basics - The Audio Notebook in Practice
HDSP System: Notebook Basics - Background Knowledge and Tuning
HDSP System: Notebook Tests - Compatibility and Performance
The digital mixer of the Hammerfall DSP in theory and practise
HDSP System: TotalMix - Hardware and Technology
HDSP System: TotalMix - Software, features, operation
MADI, the serial Multichannel Audio Digital Interface, has been defined already in 1989 as an
extension of the existing AES3 standard following several manufacturers' wish. The format also
known as AES/EBU, a balanced bi-phase signal, is limited to two channels. Simply put, MADI
contains 28 of those AES/EBU signals in serial, i. e. after one another, and the sample rate can
still even vary by +/-12.5%. The limit which cannot be exceeded is a data rate of 100Mbit/s.
Because an exact sampling frequency is used in most cases, the 64 channel mode was introduced officially in 2001. It allows for a maximum sample rate of 48 kHz + ca. 1%, corresponding to 32 channels at 96 kHz, without exceeding the maximum data rate of 100 Mbit/s. The
effective data rate of the port is 125 Mbit/s due to additional coding.
Older devices understand and generate only the 56 channel format. Newer devices often work
in the 64 channel format, but offer still no more than 56 audio channels. The rest is being eaten
up by control commands for mixer settings etc.. The ADI-648 and the HDSP MADI show that
this can be done in a much better way, with an invisible transmission of 16 MIDI channels and
the MADI signal still being 100% compatible.
For the transmission of the MADI signal, proved methods known from network technology were
applied. Most people know unbalanced (coaxial) cables with 75 Ohms BNC plugs, they are not
expensive and easy to get. The optical interface is much more interesting due to its complete
galvanic separation, but for many users it is a mystery, because very few have ever dealt with
huge cabinets full of professional network technology. Therefore here are some explanations
regarding 'MADI optical'.
• The cables used are standard in computer network technology. They are thus not at all
expensive, but unfortunately not available in every computer store.
• The cables have an internal fibre of only 50 or 62.5 µm diameter and a coating of 125 µm.
They are called network cables 62.5/125 or 50/125, the former mostly being blue and the
latter mostly being orange. Although in many cases not clearly labeled, these are always (!)
glass fibre cables. Plastic fibre cables (POF, plastic optical fibre) can not be manufactured
in such small diameters.
• The plugs used are also an industry standard and called SC. Please don't mix them up with
ST connectors, which look similar to BNC connectors and are being screwed. Plugs used in
the past (MIC/R) were unnecessarily big and are not being used any longer.
• The cables are available as a duplex variant (2 cables being glued together) or as a sim-
plex variant (1 cable). The ADI-648's opto module supports both variants.
• The transmission uses the multimode technique which supports cable lengths of up to al-
most 2 km. Single mode allows for much longer distances, but it uses a completely different
fibre (8 µm). By the way, due to the wave-length of the light being used (1300 nm), the optical signal is invisible to the human eye.
Digital signals consist of a carrier and the data. If a digital signal is applied to an input, the receiver has to synchronize to the carrier clock in order to read the data correctly. To achieve
this, the receiver uses a PLL (Phase Locked Loop). As soon as the receiver meets the exact
frequency of the incoming signal, it is locked. This Lock state remains even with small changes
of the frequency, because the PLL tracks the receiver's frequency.
If a MADI signal is applied to the HDSP MADI, the unit indicates LOCK, i. e. a valid input signal. This information is presented in the HDSP MADI's Settings dialog. In the status display
SyncCheck, the state of all clocks is decoded and shown as simple text (No Lock, Lock, Sync).
Unfortunately, LOCK does not necessarily mean that the received signal is correct with respect
to the clock which processes the read out of the embedded data. Example: The HDSP MADI is
set to 44.1 kHz internally (clock mode Master), and a mixing desk with MADI output is connected to the card's MADI input. The status display will show LOCK immediately, but usually
the mixing desk's sample rate is generated internally (it is Master too), and thus slightly higher
or lower than the HDSP MADI's internal sample rate. Result: When reading out the data, there
will frequently be read errors that cause clicks and drop outs.
Also when using multiple clock signals, a simple LOCK is not sufficient. The above described
problem can be solved elegantly by setting the HDSP MADI from Master to AutoSync (its internal clock will then be the clock delivered by the mixing desk). But in case the card is clocked to
word clock, this signal can also be un-synchronous, and there will again be a slight difference in
the sample rate, and therefore clicks and drop outs.
In order to display those problems, the HDSP MADI includes SyncCheck®. It checks all clocks
used for synchronicity. If they are not synchronous to each other, the status display will show
LOCK. If they are synchronous to each other (i. e. absolutely identical), the status display will
change to SYNC. In the example above it would have been obvious immediately that the entry
LOCK is shown in SyncCheck instead of SYNC, right after connecting the mixing desk. With
external synchronisation via word clock, both entries Word Clock and MADI must display
SYNC.
In practice, SyncCheck allows for a quick overview of the correct configuration of all digital
devices. So one of the most difficult and error-prone topics of the digital studio world finally
becomes easy to handle.
The term Zero Latency Monitoring has been introduced by RME in 1998 for the DIGI96 series
of audio cards. It stands for the ability to pass-through the computer's input signal at the interface directly to the output. Since then, the idea behind has become one of the most important
features of modern hard disk recording. In the year 2000, RME published two ground-breaking
Tech Infos on the topics Low Latency Background, which are still up-to-date: Monitoring, ZLM and ASIO, and Buffer and Latency Jitter, both found on the RME Driver CD and the RME website.
How much Zero is Zero?
From a technical view there is no zero. Even the analog pass-through is subject to phase errors, equalling a delay between input and output. However, delays below certain values can
subjectively be claimed to be a zero-latency. This applies to analog routing and mixing, and in
our opinion also to RME's Zero Latency Monitoring. The term describes the digital path of the
audio data from the input of the interface to its output. The digital receiver of the HDSP MADI
can't operate un-buffered, and together with TotalMix and the output via the transmitter, it
causes a typical delay of 3 samples. At 44.1 kHz this equals about 68 µs (0.000068 s). In Double Speed mode, the delay doubles to 6 samples, for both ADAT and SPDIF.
Oversampling
While the delays of digital interfaces can be disregarded altogether, the analog inputs and outputs do cause a significant delay. Modern converter chips operate with 64 or 128 times oversampling plus digital filtering, in order to move the error-prone analog filters away from the
audible frequency range as far as possible. This typically generates a delay of one millisecond.
A playback and re-record of the same signal via DA and AD (loopback) then causes an offset
of the newly recorded track of about 2 ms. The following table lists the delays of the HDSP
MADI's DA-converter for the headphones output:
Sample frequency kHz 44.1 48 88.2 96 176.4 192
DA (43.4 x 1/fs) ms 0.98 0.9
DA (87.5 x 1/fs) ms 0.99 0.91
DA (176.8 x 1/fs) ms 1 0.92
Buffer Size (Latency)Windows: This option found in the Settings dialog defines the size of the buffers for the audio
data used in ASIO and GSIF (see chapter 13 and 14).
Mac OS X: The buffer size is defined within the application. Only some do not offer any setting.
For example iTunes is fixed to 512 samples.
General: A setting of 64 samples at 44.1 kHz causes a latency of 1.5 ms, for record and playback each. But when performing a digital loopback test no latency/offset can be detected. The
reason is that the software naturally knows the size of the buffers, therefore is able to position
the newly recorded data at a place equalling a latency-free system.
AD/DA Offset under ASIO and OS X: ASIO (Windows) and Core Audio (Mac OS X) allow for
the signalling of an offset value to correct buffer independent delays, like AD- and DAconversion or the Safety Buffer described below. An analog loopback test will then show no
offset, because the application shifts the recorded data accordingly. Because in real world operation analog record and playback is unavoidable, the drivers include an offset value matching
the HDSP MADI's converter delays.
Because the HDSP MADI is a completely digital interface, and the delays introduced by external AD/DA-converters or other digital interfaces are unknown to unit and driver, the drivers
include the digital offset values (3 / 6 samples). Therefore the delays caused by external converters have to be taken care off in the record software, which usually means that the user has
to enter specific offset values manually.
Note: Cubase and Nuendo display the latency values signalled from the driver separately for
record and playback. While with our digital cards these values equal exactly the buffer size (for
example 3 ms at 128 samples), the HDSP MADI displays an additional millisecond – the time
needed for the AD/DA-conversion.
Core Audios Safety Offset
Under OS X, every audio interface has to use a so called satety offset, otherwise Core Audio
won't operate click-free. The HDSP MADI uses a safety offset of 32 samples. This offset is
signalled to the system, and the software can calculate and display the total latency of buffer
size plus AD/DA offset plus safety offset for the current sample rate.
30.4 DS - Double Speed
When activating the Double Speed mode the HDSP MADI operates at double sample rate. The
internal clock 44.1 kHz turns to 88.2 kHz, 48 kHz to 96 kHz. The internal resolution is still 24
bit.
Sample rates above 48 kHz were not always taken for granted, and are still not widely used
because of the CD format (44.1 kHz) dominating everything. Before 1998 there were no receiver/transmitter circuits available that could receive or transmit more than 48 kHz. Therefore
a work-around was used: instead of two channels, one AES line only carries one channel,
whose odd and even samples are being distributed to the former left and right channels. By
this, you get the double amount of data, i. e. also double sample rate. Of course in order to
transmit a stereo signal two AES/EBU ports are necessary then.
This transmission mode is called Double Wire in the professional studio world, and is also
known as S/MUX (abbreviation for Sample Multiplexing) in connection with the multichannel
ADAT format. The AES3 specification uses the uncommon term Single channel double sam-pling frequency mode.
Not before February 1998, Crystal shipped the first 'single wire' receiver/transmitters that could
also work with double sample rate. It was then possible to transmit two channels of 96 kHz data
via one AES/EBU port.
But Double Wire is still far from being dead. On one hand, there are still many devices which
can't handle more than 48 kHz, e. g. digital tape recorders. But also other common interfaces
like ADAT or TDIF are still using this technique.
With MADI, sample multiplexing is often used as well to offer sample rates higher than 48 kHz.
The HDSP MADI supports all formats. 96 kHz can be received and transmitted both as 48K
Frame (using S/MUX) and as native 96K Frame. In 48K Frame Double Speed mode, the HDSP
MADI distributes the data of one channel to two consecutive MADI channels. This reduces the
available channel count from 64 to 32.
As the transmission of double rate signals with 48K Frame is done at standard sample rate
(Single Speed), the MADI ports still operate at 44.1 kHz or 48 kHz.
Due to the small number of available devices that use sample rates up to 192 kHz, but even
more due to a missing real world application (CD...), Quad Speed has had no broad success so
far. An implementation of the ADAT format as double S/MUX (S/MUX4) results in only two
channels per optical output. Devices using this method are few.
In earlier times the transmission of 192 kHz had not been possible via Single Wire, so once
again sample multiplexing was used: instead of two channels, one AES line transmits only one
half of a channel. A transmission of one channel requires two AES/EBU lines, stereo requires
even four. This transmission mode is being called Quad Wire in the professional studio world.
The AES3 specification does not mention Quad Wire.
With MADI, sample multiplexing is used as well to offer sample rates higher than 96 kHz. In
fact, technical reasons require to use this method beyond 96 kHz. A 192K or 384K Frame format would not be fully compatible to the MADI standard. Therefore 192 kHz is supported as
S/MUX4 only.
In 48K Frame Quad Speed mode, the HDSP MADI distributes the data of one channel to four
consecutive MADI channels. This reduces the available channel count from 64 to 16.
As the transmission of quad rate signals with 48K Frame is done at standard sample rate (Single Speed), the MADI ports still operate at 44.1 kHz or 48 kHz.
The SteadyClock technology of the HDSP MADI guarantees an excellent performance in all
clock modes. Its highly efficient jitter suppression refreshes and cleans up any clock signal, and
provides it as reference clock at the word clock output.
Usually a clock section consists of an analog PLL for external synchronization and several
quartz oscillators for internal synchronisation. SteadyClock requires only one quartz, using a
frequency not equalling digital audio. Latest circuit designs like hi-speed digital synthesizer,
digital PLL, 100 MHz sample rate and analog filtering allow RME to realize a completely newly
developed clock technology, right within the FPGA at lowest costs. The clock's performance
exceeds even professional expectations. Despite its remarkable features, SteadyClock reacts
quite fast compared to other techniques. It locks in fractions of a second to the input signal,
follows even extreme varipitch changes with phase accuracy, and locks directly within a range
of 25 kHz up to 200 kHz.
SteadyClock has originally been developed to gain a stable and clean
clock from the heavily jittery MADI
data signal. The embedded MADI
clock suffers from about 80 ns jitter,
caused by the time resolution of 125
MHz within the format. Common jitter
values for other devices are 5 ns,
while a very good clock will have less
than 2 ns.
The picture to the right shows the
MADI input signal with 80 ns of jitter
(top graph, yellow). Thanks to
SteadyClock this signal turns into a
clock with less than 2 ns jitter (lower graph, blue).
The other input sources of the HDSP
MADI, word clock, Video and LTC,
gain a lot from SteadyClock as well. In
fact, extracting a low jitter clock from
LTC is not possible without a SteadyClock similar technique at all!
The screnshot to the right shows an
extremely jittery word clock signal of
about 50 ns jitter (top graph, yellow).
Again SteadyClock provides an extreme clean-up. The filtered clock
shows less than 2 ns jitter (lower
graph, blue).
The cleaned and jitter-freed signal can be used as reference clock for any application, without
any problem. The signal processed by SteadyClock is of course not only used internally, but
also available at the HDSP MADI 's word clock outputs. It is also used to clock the digital MADI
output.
The HDSP MADI card's sheer number of audio channels makes it more demanding for a computer's PCI bus performance than any other audio card. Furthermore, measurements of pure
data throughput are not sufficient for measuring realtime audio performance or compatibility.
Large amounts of data may be transferred in short and fast bursts with small interruptions,
which will result in a relatively high data rate when measured averaged, but audio signals will
suffer clicks and dropouts because of the interruptions.
Theoretically, PCI can transfer up to 133 MByte/s. A single HDSP MADI card will cause about
24.6 MByte/s of traffic. So three cards should work in any modern computer without problems -
but in fact they won't. That is, they can be installed and accessed, but audio will crackle like
hell...
According to our research, even using two cards with all 128 channels of playback and recording is not possible on usual single-PCI-bus systems. The limit is around 80 channels in
each direction, and varies depending on how many channels of inputs and outputs are active.
Here's an example of Intel's popular 875 chipset, which uses the ICH5 as southbridge. The
crackle-free maximum is:
• 128 channels record with up to 96 channels playback
• 64 channels record with up to 128 channels playback
If a PATA harddisk is used instead of SATA in this specific system, all 128 channels can be
used in both directions without clicks. But neither type of disk will allow simultaneous recording
and playback of so many tracks here.
With 80 channels transferring data each way, and a seemingly low PCI load of about 30.7
MByte/s, every additional system activity will cause a short interruption of the PCI bus activity.
Be it disk or network access, depending on chipset and board architecture, this will cause a
disturbance of the audio signal. These disturbances begin to appear on channel 64, and the
stronger they get, the more lower channels will be affected also. Therefore, they may not be
noticed immediately.
Although a single MADI card will usually work trouble-free on modern computers, some points
should be noted:
• The PCI-bus should be kept free of other devices. This also applies to on-board components
such as modems, USB devices or network adapters.
• If disturbances occur during hard drive activity: Modern SATA hard drives often feature
extremely high peak data rates, which are unnecessary for audio playback and recording,
but tend to disturb the PCI bus. In this case, taking the step back to PATA controllers and
drives can turn out to be a big step forward. On Macs, keeping audio files on an external
FireWire drive may be worthwile. Even FireWire 800 efficiently limits peak performance, but
the average data rate will be sufficient for many audio tracks.
Using two HDSP MADI cards to full capacity requires boards with a very high transfer rate between north bridge and south bridge, in order to also allow the processing of other data (hard
drive...) in real-time. The 266 MByte/s of an Intel 875 chipset are simply not sufficient. The
solution is found in motherboards with server chipsets that usually incorporate several separated PCI busses. One good example is the Tyan 8KW (S2885), which not only operates flawlessly with two MADI cards, but also provides ample CPU power, thanks to Dual Opteron
CPUs.
Sample rate range originally used in Digital Audio. Typical applications are 32 kHz (digital radio
broadcast), 44.1 kHz (CD), and 48 kHz (DAT).
Double Speed
Doubles the original sample rate range, in order to achieve higher audio quality and improved
audio processing. 64 kHz is practically never used, 88.2 kHz is quite rare in spite of certain
advantages. 96 kHz is a common format. Sometimes called Double Fast.
Quad Speed
Controversially discussed way of ensuring hi-end audio quality and processing by quadrupling
the sample frequency. 128 kHz is non-existant, 176.4 kHz is rare, if at all then 192 kHz is used,
e.g. for DVD Audio.
Single Wire
Standard audio data transfer, where the audio signal's sample rate is equal to the rate of the
digital signal. Used from 32 to 192 kHz. Sometimes called Single Wide.
Double Wire
Before 1998 there were no receiver/transmitter circuits available that could receive or transmit
more than 48 kHz. Higher sample rates were transferred by splitting odd and even bits across
the L/R channels of a single AES connection. This provides for twice the data rate, and hence
twice the sample rate. A stereo signal subsequently requires two AES/EBU ports.
The Double Wire method is an industry standard today, however it has a number of different
names, like Dual AES, Double Wide, Dual Line and Wide Wire. The AES3 specification uses
the uncommon term Single channel double sampling frequency mode. When used with the
ADAT format, the term S/MUX is commonly used.
Double Wire not only works with Single Speed signals, but also with Double Speed. As an example, Pro Tools HD, whose AES receiver/transmitter only work up to 96 kHz, uses Double
Wire to transmit 192 kHz. Four channels of 96 kHz turn into two channels of 192 kHz.
Quad Wire
Similar to Double Wire, with samples of one channel spread across four channels. This way
single speed devices can transmit up to 192 kHz, but need two AES/EBU ports to transmit one
channel. Also called Quad AES.
S/MUX
Since the ADAT hardware interface is limited to Single Speed, the Double Wire method is used
for sample rates up to 96 kHz, but usually referred to as S/MUX (Sample Multiplexing). An
ADAT port supports four channels this way. With MADI S/MUX is used as well, to transmit up to
96kHz although the 48K Frame format is used.
S/MUX4
The Quad Wire method allows to transmit two channels at up to 192 kHz via ADAT. The
method is referred to as S/MUX4. With MADI S/MUX4 is used as well, to transmit up to 192
kHz although the 48K Frame format is used.
Note: All conversions of the described methods are lossless. The existing samples are just
spread or re-united between the channels.
48K Frame
Most often used MADI format. Supports up to 64 channels at up to 48 kHz.
Frame format for up to 32 channels at up to 96 kHz. The advantage of this format against 48K
Frame using S/MUX: the receiver can detect the real (double) sample rate on its own and immediately. With 48K Frame and S/MUX, the user has to set up the correct sample rate in all
involved devices manually.
Note about the preliminary state of this manual
This manual describes several features not yet found and supported in the current MADI driver.
These are:
Windows
Sample rates higher than 96 kHz
Mac OS X
Sample rates higher than 96 kHz
Settingsdialog - DDS
TotalMix Monitor Panel
TotalMix MIDI Remote Control
Because the earlier manual included no longer valid information in several cases, and all of the
extended features of TotalMix are at least already present under Windows, we decided to
release this manual anyway. Both Mac and Windows driver of the HDSP MADI will be updated
within 2005 to the state described in this manual, and will have the same functionality under
Mac OS X and Windows.