2.80 GHz – 3.40 GHz Frequencies Supporting Hyper-Threading
Technology1 for All Frequencies with 800 MHz Front Side Bus
February 2005
Document Number: 300561-003
Page 2
INFORMATION IN THIS DOCUMENT IS PROVIDED IN CONNECTION WITH INTEL® PRODUCTS. NO LICENSE, EXPRESS OR IMPLIED, BY
ESTOPPEL OR OTHERWISE, TO ANY INTELLECTUAL PRO PERTY RIGHTS IS GRANTED BY THIS DOCUMENT. EXCEPT AS PROVIDED IN
INTEL'S TERMS AND CONDITIONS OF SALE FOR SUCH PRODUCTS, INTEL ASSUMES NO LIABILITY WHATSOEV ER, AND INTEL DISCLAIMS
ANY EXPRESS OR IMPLIED WARRANTY, RELATING TO SALE AND/OR USE OF INTEL PRODUCTS INCLUDING LIABILITY OR WARRANTIES
RELATING T O FITNESS FOR A PARTICULAR PURPOSE, MERCHANTABILITY, OR INFRINGEMENT OF ANY PA TENT, COPYRIGHT OR OTHER
INTELLECTUAL PROPERTY RIGHT. Intel products are not intended for use in medical, life saving, or life sustaining applications.
Intel may make changes to specifications and product descriptions at any time, without notice.
Designers must not rely on the absence or characteristics of any features or instructions marked “reserved” or “un defined.” Intel reserves these for
future definition and shall have no responsibility whatsoever for conflicts or incompatibilities arising from future changes to them.
®
The Intel
from published specifications. Current characterized errata are available on request.
Pentium® 4 processor on 90 nm process may contain design defects or errors known as errata which may cause the product to deviate
Contact your local Intel sales office or your distributor to obtain the latest specifications and before placing your product order.
1
Hyper-Threading Technology requires a computer system with an Intel® Pentium® 4 processor supporting HT Technology and a Hyper-Threading
Technology enabled chipset, BIOS and operating system. Performance will vary depending on the specific hardware and software you use.
See http://www.intel.com/info/hyperthreading/
Intel, Pentium, Intel NetBurst, and the Intel logo are trademarks or registered trademarks of Intel Corporation or its subsidiaries in the United States
and other countries.
for more information including details on which processors support HT Technology.
31Fan Heatsink Power and Signal Specifications...................................................79
32Boxed Processor Fan Heatsink Set Points .........................................................81
.
6Datasheet
Page 7
Revision History
RevisionDescriptionDate
-001• Initial releaseFebruary 2004
-002• Added specifications for 3.20 GHz processors with PRB = 1
• Added ISGNT/ISLP specifications
• Updated thermal diode specifications
• Other changes marked with change bars
-003• Added specifications for 3.40 GHz processors with PRB = 0February 2005
April 2004
§
Datasheet7
Page 8
Intel® Pentium® 4 Processor on 90 nm
Process 2.80A/E GHz, 3E GHz, 3.20E
GHz, and 3.40E GHz
• Available at 2.80A/E GHz, 3E GHz,
3.20E GHz, and 3.40E GHz
• Supports Hyper-Threading Technology
(HT Technology) for all frequencies with
800 MHz front side bus (FSB)
1
• Binary compatible with applications
running on previous members of the Intel
microprocessor line
• Intel NetBurst
®
microarchitecture
• FSB frequencies at 533 MHz, and
800 MHz
• Hyper-Pipelined Technology
—Advance Dynamic Execution
—Very deep out-of-order execution
• Enhanced branch prediction
• Optimized for 32-bit applications running
on advanced 32-bit operating systems
• 478-Pin Package
§
• 16-KB Level 1 data cache
• 1-MB Advanced Transfer Cache (on-die,
full-speed Level 2 (L2) cache) with 8-way
associativity and Error Correcting Code
(ECC)
• 144 Streaming SIMD Extensions 2
(SSE2) instructions
• 13 Streaming SIMD Extensions 3 (SSE3)
instructions
• Enhanced floating point and multimedia
unit for enhanced video, audio,
encryption, and 3D performance
• Power Management capabilities
—System Management mode
—Multiple low-power states
• 8-way cache associativity provides
improved cache hit rate on load/store
operations
8Datasheet
Page 9
1Introduction
Introduction
The Intel® Pentium® 4 processor on 90 nm process is a follow on to the Intel® Pentium® 4
processor in the 478-pin package with enhancements to the Intel NetBurst
Pentium 4 processor on 90 nm process uses Flip-Chip Pin Grid Array (FC-mPGA4) package
technology, and plugs into a 478-pin surface mount, Zero Insertion Force (ZIF) socket, referred to
as the mPGA478B socket. The Pentium 4 processor on 90 nm process, like its predecessor, the
Pentium 4 processor in the 478-pin package, is based on the same Intel 32-bit microarchitecture
and maintains the tradition of compatibility with IA-32 software.
Note:In this document the Pentium 4 processor on 90 nm process is also referred to as the processor.
The Pentium 4 processor on 90 nm process supports Hyper-Threading Technology
Threading Technology allows a single, physical processor to function as two logical processors.
While some execution resources (such as caches, execution units, and buses) are shared, each
logical processor has its own architecture state with its own set of general-purpose registers,
control registers to provide increased system responsiveness in multitasking environments, and
headroom for next generation multithreaded applications. Intel recommends enabling HyperThreading Technology with Microsoft Wind ows* XP Profession al or Windows* XP Home, and
disabling Hyper-Threading Technology via the BIOS for all previous versions of Windows
operating systems. For more information on Hyper-Threading Technology, see www.intel.com/
info/hyperthreading. Refer to Section 6.1, for Hyper-Threading Technology configuration details.
In addition to supporting all the existing Streaming SIMD Extensions 2 (SSE2), there are 13 new
instructions that further extend the capabilities of Intel processor technology. These new
instructions are called Streaming SIMD Extensions 3 (SSE3).These new instructions enhance the
performance of optimized applications for the digital home such as video, image processing and
media compression technology. 3D graphics and other entertainment applications (such as gaming)
will have the opportunity to take advantage of these new instructions as platforms with the Pentium
4 processor on 90 nm process and SSE3 become available in the market place.
®
microarchitecture. The
1
. Hyper-
The processor’s Intel NetBurst microarchitecture front side bus (FSB) uses a split-transaction,
deferred reply protocol like the Pentium 4 processor. The Intel NetBurst microarchitecture FSB
uses Source-Synchronous Transfer (SST) of address and data to improve performance by
transferring data four times per bus clock (4X data transfer rate, as in AGP 4X). Along with the 4X
data bus, the address bus can deliver addresses two times per bus clock and is referred to as a
"double-clocked" or 2X address bus. Working together, the 4X data bus and 2X address bus
provide a data bus bandwidth of up to 6.4 GB/s.
Intel will enable support components for the processor including heatsink, heatsink retention
mechanism, and socket. Manufacturability is a high priority; hence, mechanical assembly may be
completed from the top of the baseboard and should not require any special tooling.
The processor includes an address bus powerdown capability that removes power from the address
and data pins when the FSB is not in use. This feature is always enabled on the processor.
Datasheet9
Page 10
Introduction
1.1Terminology
A ‘#’ symbol after a signal name refers to an active low signal, indicating a signal is in the active
state when driven to a low level. For example, when RESET# is low, a reset has been requested.
Conversely, when NMI is high, a nonmaskable interrupt has occurred. In the case of signals where
the name does not imply an active state but describes part of a binary sequence (such as address or
data), the ‘#’ symbol implies that the signal is inverted. For example, D[3:0] = ‘HLHL’ refers to a
hex ‘A’, and D[3:0]# = ‘LHLH’ also refers to a hex ‘A’ (H= High logic lev el, L= Low logic level).
“FSB” refers to the interface between the processor and system core logic (i.e., the chipset
components). The FSB is a multiprocessing interface to processors, memory, and I/O.
1.1.1Processor Packaging Terminology
Commonly used terms are explained here for clarification:
• Intel
• Processor — For this document, the term processor is the generic form of the Pentium 4
• Keep-out zone — The area on or near the processor that system design cannot use.
• Intel
• Intel
• Processor core — Processor core die with integrated L2 cache.
• FC-mPGA4package — The Pentium 4 processor on 90 nm process is available in a Flip-
®
Pentium® 4 processor on 90 nm process — Processor in the FC-mPGA4 package
with a 1-MB L2 cache.
processor on 90 nm process.
®
865G/865GV/865PE/865P chipset — Chipset that supports DDR memory technology
for the Pentium 4 processor on 90 nm process.
®
875P chipset — Chipset that supports DDR memory technology for the Pentium 4
processor on 90 nm process
Chip Micro Pin Grid Array 4 package, consisting of a processor core mounted on a pinned
substrate with an integrated heat spreader (IHS). This packaging technology employs a
1.27 mm [0.05 in] pitch for the substrate pins.
• mPGA478B socket — The Pentium 4 processor on 90 nm process mates with the system
board through a surface mount, 478-pin, zero insertion force (ZIF) socket.
• Integrated heat spreader (IHS) —A component of the processor package used to enhance
the thermal performance of the package. Component thermal solutions interface with the
processor at the IHS surface.
• Retention mechanism (RM)—Since the mPGA478B socket does not include any mechanical
features for heatsink attach, a retention mechanism is required. Component thermal solutions
should attach to the processor via a retention mechanism that is independent of the socket.
• Storage conditions — Refers to a non-operational state. The processor may be installed in a
platform, in a tray, or loose. Processors may be sealed in packaging or exposed to free air.
Under these conditions, processor pins should not be connected to any supply voltages, hav e
any I/Os biased, or receive any clocks. Upon exposure to “free air”(i.e. unsealed packaging or
a device removed from packaging material) the processor must handled in accordance with
moisture sensitivity labeling (MSL) as indicated on the packaging material.
• Functional operation — Refers to normal operating conditions in which all processor
specifications, including DC, AC, FSB, signal quality, mechanical and thermal, are satisfied.
10Datasheet
Page 11
1.2References
Material and concepts available in the following documents may be beneficial when reading this
document.
For clean on-chip power distribution, the processor has 85 VCC (power) and 179 VSS (ground)
pins. All power pins must be connected to V
ground plane.The processor VCC pins must be supplied by the voltage determined by the VID
(Voltage identification) pins.
2.2Decoupling Guidelines
Due to its large number of transistors and high internal clock speeds, the processor is capable of
generating large current swings between low and full power states. This may cause voltages on
power planes to sag below their minimum values if bulk decoupling is not adequate. Care must be
taken in the board design to ensure that the voltage provided to the processor remains within the
specifications listed in Table 9. Failure to do so can result in timing violations or reduced lifetime
of the component. For further information and design guidelin es, refer to the appropriate platform
design guide, and the Voltage Regulator-Down (VRD) 10.0 Design Guidelines for Desktop Socket
478.
, while all VSS pins must be connected to a system
CC
Electrical Specifications
2.2.1V
Decoupling
CC
Regulator solutions need to provide bulk capacitance with a low Effective Series Resistance (ESR)
and keep a low interconnect resistance from the regulator to the socket. Bulk decoupling for the
large current swings when the part is powering on, or entering/exiting low power states, must be
provided by the voltage regulator solution (VR). For more details on this topic, refer to the
appropriate platform design guide
for Desktop Socket 478.
, and the Voltage Regulator-Down (VRD) 10.0 Design Guidelines
2.2.2FSB GTL+ Decoupling
The processor integrates signal termination on the die as well as incorporating high frequency
decoupling capacitance on the processor package. Decoupling must also be provided by the system
baseboard for proper GTL+ bus operation. For more information, refer to the appropriate platform
design guide.
Datasheet13
Page 14
Electrical Specifications
2.2.3FSB Clock (BCLK[1:0]) and Processor Clocking
BCLK[1:0] directly controls the FSB interface speed as well as the core frequency of the processor .
As in previous generation processors, the processor core frequency is a multiple of the BCLK[1:0]
frequency. No user intervention is necessary, and the processor will automatically run at the speed
indicated on the package. The processor uses a differential clocking implementation.
Table 2. Core Frequency to FSB Multiplier Configuration
1.Individual processors operate only at or below the rated frequency.
Core Frequency
(133 MHz BCLK/533 MHz FSB)
2.3Voltage Identification
The VID specification for the processor is supported by the Voltage Regulator-Down (VRD) 10.0
Design Guidelines for Desktop Socket 478. The voltage set by the VID pins is the maximum
voltage allowed by the processor. A minimum voltage is provided in Table 9 and changes with
frequency. This allows processors running at a higher frequency to have a relaxed minimum
voltage specification. The specifications have been set such that one voltage regulator can work
with all supported frequencies.
Core Frequency
(200 MHz BCLK/800 MHz FSB)
Notes
1
1
1
1
1
1
Individual processor VID values may be calibrated during manufacturing such that two devices at
the same speed may have different VID settings.
The processor uses six voltage identification pins, VID[5:0], to support automatic selection of
power supply voltages. Table 3 specifies the voltage level corresponding to the state of VID[5:0].
A ‘1’ in this table refers to a high voltage level and a ‘0’ refers to low voltage level. If the processor
socket is empty (VID[5:0] = x11111), or the voltage regulation circuit cannot supply the voltage
that is requested, it must disable itself. See the Voltage Regulator-Down (VRD) 10.0 Design Guidelines for Desktop Socket 478 for more details.
Power source characteristics must be guaranteed to be stable when the supply to the voltage
regulator is stable.
The processor’s Voltage Identification circuit requires an independent 1.2 V supply and some other
power sequencing considerations.
silicon. Since these PLLs are analog, they require low noise power supplies for minimum jitter.
Jitter is detrimental to the system: it degrades external I/O timings as well as internal core timings
(i.e., maximum frequency). To prevent this degradation, these supplies must be low pass filtered
from V
CC
.
are power sources required by the PLL clock generators on the processor
The AC low-pass requirements, with input at V
• < 0.2 dB gain in pass band
• < 0.5 dB attenuation in pass band < 1 Hz
• > 34 dB attenuation from 1 MHz to 66 MHz
• > 28 dB attenuation from 66 MHz to core frequency
The filter requirements are illustrated in Figure 1. For recommendations on implementing the filter,
2. No specification exists for frequencies beyond fcore (core frequency).
3. fpeak, if existent, should be less than 0.05 MHz.
16Datasheet
High
Frequency
Band
Page 17
2.4Reserved, Unused, and TESTHI Pins
All RESERVED pins must remain unconnected. Connection of these pins to VCC, VSS, or to any
other signal (including each other) can result in component malfunction or incompatibility with
future processors. See Chapter 4 for a pin listing of the processor and the location of all
RESERVED pins.
For reliable operation, always connect unused inputs or bidirectional sign als to an appropriate
signal level. In a system level design, on-die termination has been included on the processor to
allow signals to be terminated within the processor silicon. Most unused GTL+ inputs should be
left as no connects since GTL+ termination is provided on the processor silicon. However, see
Table 5 for details on GTL+ signals that do not include on-die termination. Unused active high
inputs should be connected through a resistor to ground (V
unconnected; however, this may interfere with some test access port (TAP) functions, complicate
debug probing, and prevent boundary scan testing. A resistor must be used when tying
bidirectional signals to power or ground. When tying any signal to power or ground, a resistor will
also allow for system testability. For unused GTL+ input or I/O signals, use pull-up resistors of the
same value as the on-die termination resistors (R
TAP, Asynchronous GTL+ inputs, and Asynchronous GTL+ outputs do not include on-die
termination. Inputs and used outputs must be terminated on the system board. Unused outputs may
be terminated on the system board or left unconnected. Note that leaving unused outputs
unterminated may interfere with some TAP functions, complicate debug probing, and prevent
boundary scan testing. Signal termination for these signal types is discussed in the appropriate
platform design guide.
).
TT
). Unused outputs can be left
SS
Electrical Specifications
The TESTHI pins must be tied to the processor V
resistor has a resistance value within ±20% of the impedance of the board transmission line traces.
For example, if the trace impedance is 60 Ω, then a value between 48 Ω and 72 Ω is required.
The TESTHI pins may use individual pull-up resistors or be grouped together as detailed below . A
matched resistor must be used for each group:
using a matched resistor, where a matched
CC
• TESTHI[1:0]
• TESTHI[7:2]
• TESTHI8 – cannot be grouped with other TESTHI signals
• TESTHI9 – cannot be grouped with other TESTHI signals
• TESTHI10 – cannot be grouped with other TESTHI signals
• TESTHI11 – cannot be grouped with other TESTHI signals
• TESTHI12 – cannot be grouped with other TESTHI signals
Datasheet17
Page 18
Electrical Specifications
2.5FSB Signal Groups
The FSB signals have been combined into groups by buffer type. GTL+ input signals have
differential input buffers that use GTLREF as a reference level. In this document, the term "GTL+
Input" refers to the GTL+ input group as well as the GTL+ I/O group when receiving. Similarly,
"GTL+ Output" refers to the GTL+ output group as well as the GTL+ I/O group when driving.
With the implementation of a source synchronous data bus comes the need to specify two sets of
timing parameters. One set is for common clock signals that are dependent on the rising edge of
BCLK0 (ADS#, HIT#, HITM#, etc.) and the second set is for the source synchronous signals that
are relative to their respective strobe lines (data and address) as well as the rising edge of BCLK0.
Asychronous signals are still present (A20M#, IGNNE#, etc.) and can become active at any time
during the clock cycle. Table 4 identifies which signals are common clock, source synchronous,
and asynchronous.
Table 4. FSB Pin Groups
Signal GroupTypeSignals
GTL+ Common Clock
Input
GTL+ Common Clock I/O
GTL+ Source
Synchronous I/O
GTL+ Strobes
Asynchronous GTL+
Input
Asynchronous GTL+
Output
Asynchronous GTL+
Input/Output
TAP InputSynchronous to TCK TCK, TDI, TMS, TRST#
TAP OutputSynchronous to TCK TDO
FSB ClockClockBCLK[1:0], ITP_CLK[1:0]
Power/Other
NOTES:
1.Refer to Section 4.2 for signal descriptions.
2.The value of these pins during the active-to-inactive edge of RESET# defines the processor configuration
options. See Section 6.1 for details.
3.In processor systems where there is no debug port implemented on the system board, these signals are used
to support a debug port interposer. In systems with the debug port implemented on the system board, these
signals are no connects.
Legacy input signals (such as A20M#, IGNNE#, INIT#, SMI#, SLP#, and STPCLK#) use CMOS
input buffers. All of these signals follow the same DC requirements as G TL+ signals; however, the
outputs are not actively driven high (during a logical 0-to-1 transition) by the processor. These
signals do not have setup or hold time specifications in relation to BCLK[1:0].
2.7Test Access Port (TAP) Connection
Due to the voltage levels supported by other components in the Test Access Port (TAP) logic, it is
recommended that the processor be first in the TAP chain and followed by any other components
within the system. A translation buffer should be used to connect to the rest of the chain unless one
of the other components is capable of accepting an input of the appropriate voltage level. Similar
considerations must be made for TCK, TMS, TRST#, TDI, and TDO. Tw o copies of each signal
may be required, with each driving a different voltage level.
Datasheet19
Page 20
Electrical Specifications
2.8FSB Frequency Select Signals (BSEL[1:0])
The BSEL[1:0] signals are used to select the frequency of the processor input clock (BCLK[1:0]).
Table 7 defines the possible combinations of the signals and the frequency associated with each
combination. The required frequency is determined by the processor, chipset, and clock
synthesizer. All agents must operate at the same frequency.
The processor operates at a 533 MHz or 80 0 MHz FSB frequency (selected by a 133 MHz or
200 MHz BCLK[1:0] frequency). Individual processors will only operate at their specified FSB
frequency.
For more information about these pins, refer to Section 4.2 and the appropriate platform design
guide.
Table 7. BSEL[1:0] Frequency Table for BCLK[1:0]
BSEL1BSEL0Function
LLRESERVED
LH133 MHz
HL200 MHz
HHRESERVED
20Datasheet
Page 21
Electrical Specifications
2.9Absolute Maximum and Minimum Ratings
Table 8 specifies absolut e m aximum and minimum ratings. Within functional operation limits,
functionality and long-term reliability can be expected.
At conditions outside functional operation condition limits, but within absolute maximum and
minimum ratings, neither functionality nor long-term reliability can be expected. If a device is
returned to conditions within functional operation limits after having been subjected to conditions
outside these limits, but within the absolute max im u m and minimum ratings, the device may be
functional, but with its lifetime degraded depending on exposure to conditions exceeding the
functional operation condition limits.
At conditions exceeding absolute maximum and minimum ratings, neither functionality nor longterm reliability can be expected. Moreover, if a device is subjected to these conditions for any
length of time then, when returned to conditions within the functional operating condition limits, it
will either not function, or its reliability will be severely degraded.
Although the processor contains protective circuitry to resist damage from static electric discharge,
precautions should always be taken to avoid high static voltages or electric fields.
Table 8. Processo r DC Absolute Maximum Ratings
SymbolParameterMinMax UnitNotes
V
CC
T
C
TSTORAGEProcessor storage temperature –40 +85°C
NOTES:
1. For functional operation, all processor electrical, signal quality, mechanical, and thermal specifications must
be satisfied.
2. Storage temperature is applicable to storage conditions only. In this scenario, the processor must not receive
a clock, and no lands can be connected to a voltage bias. Storage within these limits will not affect the longterm reliability of the device. For functional operation, refer to the processor case temperature specifications.
3. This rating applies to the processor and does not include any tray or packaging.
Any processor supply voltage with
respect to V
Processor case temperatureSee Section 5See Section 5°C
SS
2.10Processor DC Specifications
The processor DC speci fications in this sectio n ar e de fined at th e pr oc essor co r e sili con and not
at the package pins unless noted otherwise. See Chapter 4 for the pin signal definitions and signal
pin assignments. Most of the signals on the processor FSB are in the GTL+ signal group. The DC
specifications for these signals are listed in Table 12.
Previously, legacy signals and Test Access Port (TAP) signals to the processor used low-voltage
CMOS buffer types. However, these interfaces now follow DC specifications similar to GTL+. The
DC specifications for these signal groups are listed in Table 13 and Table 14.
Table 9 through Table 17 list the DC specifications for the processor and are valid only while
meeting specifications for case temperature, clock frequency, and input voltages. Care should be
taken to read all notes associated with each parameter.
- 0.31.55V
1
2, 3
2, 3
MSR_PLATFORM_BRV bit 18 is a Platform Requirement Bit (PRB) that indicates that the
processor has specific platform requirements.
Datasheet21
Page 22
Electrical Specifications
Table 9. Voltage and Current Specifications
SymbolParameterMinTypMaxUnitNotes
VID rangeVID1.2501.400V
VCC Loadline A processors
V
CC
3.20E GHz (PRB = 1)
3.40E GHz (PRB = 1)
See Table 10 and
Figure 2
VID – I
(max) * 1.45 mΩV
CC
VCC Loadline B processors
2.80A/E GHz (PRB = 0)
V
CC
3E GHz (PRB = 0)
3.20E GHz (PRB = 0)
See Table 11 and
Figure 3
VID – I
(max) * 1.45 mΩV
CC
3.40E GHz (PRB = 0)
I
for processor with multiple VID:
CC
I
CC
I
SGNT
I
SLP
I
TCC
I
CC_VCCA
I
CC_VCCIOPLL
I
CC_GTLREF
I
CC_VCCVID/
VCCVIDLB
2.80A/E GHz (PRB = 0)
3E GHz (PRB = 0)
3.20E GHz (PRB = 0)
3.40E GHz (PRB = 0)
3.20E GHz (PRB = 1)
3.40E GHz (PRB = 1)
I
Stop-Grant:
CC
2.80A/E GHz (PRB = 0)
3E GHz (PRB = 0)
3.20E GHz (PRB = 0)
3.40E GHz (PRB = 0)
3.20E GHz (PRB = 1)
3.40E GHz (PRB = 1)
ICC TCC activeI
ICC for PLL pins60mA
ICC for I/O PLL pin60mA
ICC for GTLREF pins (all pins)200µA
ICC for V
CCVID/VCCVIDLB
78
78
78
A
78
91
91
40
40
40
A
40
50
50
CC
A
150mA
NOTES:
1.Unless otherwise noted, all specifications in this table are based on estimates and simulations or empirical data.
2.Individual processor VID values may be calibrated during manufacturing such that two devices at the same speed may ha ve
different VID settings.
3.These voltages are targets only. A variable voltage source shoul d exist on systems in the event that a different voltage is required. See Section 2.3 and Table3 for more information.
4.The voltage specification requirements are measured across VCC_SENSE and VSS_SENSE pins at the socket with a
100 MHz bandwidth oscilloscope, 1.5 pF maximum probe capacitance, and 1 MΩ minimum impedance. The maximum length
of ground wire on the probe should be less than 5 mm. Ensure external noise from the system is not coupled into the oscilloscope probe.
5.Refer to Table 10/Figure 2 or Table 11/Figure 3 for the minimum, typical, and maximum V
processor should not be subjected to any V
V
should never exceed the VID voltage. Failure to adhere to this specification can shorten the processor lifetime.
CC
6.I
7.The current specified is also for the AutoHALT State.
8.I
9.The maximum instantaneous current the processor will draw while the thermal control circuit is active as indicated by the as-
is specified at V
CC_MAX
Stop-Grant and ICC Sleep are specified at V
CC
CC_MAX
sertion of PROCHOT# is the same as the maximum I
and ICC combination wherein VCC exceeds V
CC
.
CC_MAX
for the processor.
CC
allowed for a given current. The
CC
for a given current. Moreover,
CC_MAX
1
2
3,4,5
3,4,5
6
7,8,10
9
10
10
10
22Datasheet
Page 23
10. These parameters are based on design characterization and are not tested.
Table 10. VCC Static and Transient Tolerance for Loadline A
1.The loadline specification includes both static and transient limits except for overshoot allowed as shown in
Section 2.11.
2.This table is intended to aid in reading discrete points on Figure 2.
3.The loadlines specify voltage limits at the die measured at the VCC_SENSE and VSS_SENSE pins. V oltage
regulation feedback for voltage regulator circuits must be taken from processor VCC and VSS pins. Refer
to the Voltage Regulator-Down (VRD) 10.0 Design Guide for socket loadline guidelines and VR implementation details for 478_VR_CONFIG_A.
Electrical Specifications
1,2,3
Datasheet23
Page 24
Electrical Specifications
0
Figure 2. VCC Static and Transient Tolerance for Loadline A
05101520253035404550556065707580859
VID - 0.000
Vcc
Icc [A]
VID - 0.038
VID - 0.076
VID - 0.114
Vcc [V]
Maximum
Vcc
VID - 0.152
Typical
Vcc
Minimum
VID - 0.190
VID - 0.228
NOTES:
1. The loadline specification includes both static and transient limits except for overshoot allowed as shown in
Section 2.11.
2. This loadline specification shows the deviation from the VID set point.
3. The loadlines specify voltage limits at the die measured at the VCC_SENSE and VSS_SENSE pins. V ol t age
regulation feedback for voltage regulator circuits must be taken from processor VCC and VSS pins. Refer to
the Voltage Regulator-Down (VRD) 10.0 Design Guide for socket loadline guidelines and VR implementation
details for 478_VR_CONFIG_A.
24Datasheet
Page 25
Table 11. VCC Static and Transient Tolerance for Loadline B
1.The loadline specification includes both static and transient limits except for overshoot allowed
as shown in Section 2.11.
2.This table is intended to aid in reading discrete points on Figure 3.
3.The loadlines specify voltage limits at the die measured at the VCC_SENSE and VSS_SENSE
pins. Voltage regulation feedback for voltage regulator circuits must be taken from processor
VCC and VSS pins. Refer to the Voltage Regulator-Down (VRD) 10.0 Design Guide for socket
loadline guidelines and VR implementation details for 478_VR_CONFIG_B.
Electrical Specifications
1,2,3
Datasheet25
Page 26
Electrical Specifications
Figure 3. VCC Static and Transient Tolerance for Loadline B
051015202530354045505560657075
VID - 0.000
VID - 0.025
VID - 0.050
VID - 0.075
VID - 0.100
Vcc
Maximum
Vcc [V]
Icc [A]
VID - 0.125
VID - 0.150
Vcc
Typical
Vcc
Minimum
VID - 0.175
VID - 0.200
NOTES:
1. The loadline specification includes both static and transient limits except for overshoot allowed as shown in
Section 2.11.
2. This loadline specification shows the deviation from the VID set point.
3. The loadlines specify voltage limits at the die measured at the VCC_SENSE and VSS_SENSE pins. V ol t age
regulation feedback for voltage regulator circuits must be taken from processor VCC and VSS pins. Refer to
the Voltage Regulator-Down (VRD) 10.0 Design Guide for socket loadline guidelines and VR implementation
details for 478_VR_CONFIG_B.
26Datasheet
Page 27
Electrical Specifications
Table 12. GTL+ Signal Group DC Specifications
SymbolParameterMinMaxUnitNotes
V
IL
V
IH
V
OH
I
OL
I
LI
I
LO
ONBuffer On Resistance812Ω
R
Input Low Voltage0.0GTLREF – (0.10 * VCC)V
Input High VoltageGTLREF + (0.10 * VCC)VCCV
Output High Voltage0.90*V
represents the amount of hysteresis, nominally centered about 0.5 * V
CC
for all TAP inputs.
CC
1, 2
3
4
4
V
4
5
6
6
Table 15. VCCVID DC Specifications
SymbolParameterMinTypMaxUnitNotes
V
CCVID
V
CCVIDLB
Voltage1.141.21.26V
Voltage1.141.21.26V
Table 16. VIDPWRGD DC Specifications
SymbolParameterMin TypMaxUnitNotes
V
V
IH
Input Low Voltage0.3V
IL
Input High Voltage0.9V
28Datasheet
Page 29
.
Table 17. BSEL [1:0] and VID[5:0] DC Specifications
SymbolParameterMaxUnitNotes
R
(BSEL) Buffer On Resistance60Ω
ON
(VID)Buffer On Resistance60Ω
R
ON
Electrical Specifications
1
2
2
I
OL
I
LO
V
TOL
Max Pin Current8mA
Output Leakage Current200µA
Voltage Tolerance3.3 + 5%V
NOTES:
1.Unless otherwise noted, all specifications in this table apply to all processor frequencies.
2.These parameters are not tested and are based on design simulations.
3.Leakage to VSS with pin held at 2.5 V.
Table 18. BOOTSELECT DC Specifications
SymbolParameterMinTypMaxUnitNotes
V
V
Input Low Voltage0.2 * V
IL
Input High Voltage0.8 * V
IH
CCVID
NOTES:
1.These parameters are not tested and are based on design simulations.
2.11VCC Overshoot Specification
The processor can tolerate short transient overshoot events where VCC exceeds the VID voltage
when transitioning from a high-to-low current load condition. This overshoot cannot exceed
VID + V
overshoot event must not exceed T
VID). These specifications apply to the processor die voltage as measured across the VCC_SENSE
and VSS_SENSE pins.
OS_MAX
(V
is the maximum allowable overshoot voltage). The time duration of the
OS_MAX
OS_MAX
(T
is the maximum allowable time duration above
OS_MAX
CCVID
3
V
V
1
1
Table 19. VCC Overshoot Specifications
SymbolParameterMinTypMaxUnitFigureNotes
V
OS_MAX
T
OS_MAX
Datasheet29
Magnitude of VCC overshoot above
VID
Time duration of VCC overshoot
above VID
0.050V4
25µs4
Page 30
Electrical Specifications
Figure 4. V
Overshoot Example Wavef orm
CC
VID + 0.050
VID
Voltage (V)
TOS: Overshoot time above VID
V
: Overshoot above VID
OS
NOTES:
1. V
is measured overshoot voltage.
OS
2. T
is measured time duration above VID.
OS
2.11 .1Die Voltage Validation
Example Overshoot Waveform
V
T
OS
Time
OS
Overshoot events from application testing on real processors must meet the specifications in
Table 19 when measured across the VCC_SENSE and VSS_SENSE pins. Overshoot events that
are < 10 ns in duration may be ignored. These measurements of processor die level overshoot
should be taken with a 100 MHz bandwidth limited oscilloscope.
§
30Datasheet
Page 31
Package Mechanical Specifications
3Package Mechanical
Specifications
3.1Package Mechanical Specifications
The Pentium 4 processor on 90 nm process is in a Flip-Chip Pin Grid Array (FC-mPGA4) package
that interfaces with the motherboard via a mPGA478B socket. The package consists of a processor
core mounted on a substrate pin-carrier. An integrated heat spreader (IHS) is attached to the
package substrate and core and serves as the mating surface for processor component thermal
solutions (such as a heatsink). Figure 5 shows a sketch of the processor package components and
how they are assembled together. Refer to the mPGA479, mPGA478A, mPGA478B, mPGA478C, and mPGA476 Socket Design Guidelines for complete details on the mPGA478B socket.
The package components shown in Figure 5 include the following:
• Integrated Heat Spreader (IHS)
• Thermal Interface Material (TIM)
• Processor core (die)
• Package substrate
• Capacitors
Figure 5. Processor Package Assembly
CORE (D IE)
CORE (D IE)
CAPACIT ORS
CAPACIT ORS
NOTE:
1. Socket and motherboard are included for reference and are not part of processor package.
TIM
TIM
IH S
IH S
SUBSTRATE
SUBSTRATE
MOTHERBOARD
MOTHERBOARD
SOCKET
SOCKET
Datasheet31
Page 32
Package Mechanical Specifications
3.1.1Package Mechanical Drawing
The package mechanical drawings are shown in Figure 6 and Figure 7. The drawings include
dimensions necessary to design a thermal solution for the processor. These dimensions include:
• Package reference with tolerances (total height, length, width, etc.)
• IHS parallelism and tilt
• Pin dimensions
• Top-side and back-side component keep-out dimensions
• Reference datums
All drawing dimensions are in mm [in].
32Datasheet
Page 33
Figure 6. Processor Package Drawing (Sheet 1 of 2)
Package Mechanical Specifications
Datasheet33
Page 34
Package Mechanical Specifications
Figure 7. Processor Package Drawing (Sheet 2 of 2)
34Datasheet
Page 35
Package Mechanical Specifications
3.1.2Processor Component Keep-out Zones
The processor may contain components on the substrate that define component keep-out zone
requirements. A thermal and mechanical solution design must not intrude into the required keepout zones. Decoupling capacitors are typically mounted to either the topside or pin-side of the
package substrate. See Figure 6 and Figure 7 for keep-out zones.
The location and quantity of package capacitors may change due to manufacturing efficiencies but
will remain within the component keep-in.
3.1.3Package Loading Specifications
Table 20 provides dynamic and static load specifications for the processor package. These
mechanical maximum load limits should not be exceeded during heatsink assembly, shipping
conditions, or standard use condition. Also, any mechanical system or component testing should
not exceed the maximum limits. The processor package substrate should not be used as a
mechanical reference or load-bearing surface for thermal and mechanical solution. The minimum
.
Table 20. Processor Loading Specifications
loading specification must be maintained by any thermal and mechanical solutions.
ParameterMinimumMaximumNotes
Static44 N [10 lbf]445 N [100 lbf]
Dynamic890 N [200 lbf]
Transient667 N [150 lbf]
NOTES:
1.These specifications apply to uniform compressive loading in a direction normal to the processor IHS.
2.This is the maximum force that can be applied by a heatsink retention clip. The clip must also provide the
minimum specified load on the processor package.
3.These specifications are based on limited testing for design characterization. Loading limits are for the package only and does not include the limits of the processor socket.
4.Dynamic loading is defined as an 11 ms duration average load superimposed on the static load requirement.
5.Transient loading is defined as a 2 second duration peak load superimposed on the static load requirement,
representative of loads experienced by the package during heatsink installation.
1, 2, 3
1, 3, 4
1, 3, 5
3.1.4Package Handling Guidelines
Table 21 includes a list of guidelines on package handling in terms of recommended maximum
loading on the processor IHS relative to a fixed substrate. These package handling loads may be
experienced during heatsink removal.
Table 21. Package Handling Guidelines
ParameterMaximum RecommendedNotes
Shear356 N [80 lbf]
Tensile156 N [35 lbf]
Torque8 N-m [70 lbf-in]
NOTES:
1.A shear load is defined as a load applied to the IHS in a direction parallel to the IHS top surface.
2.These guidelines are based on limited testing for design characterization.
3.A tensile load is defined as a pulling load applied to the IHS in a direction normal to the IHS surface.
4.A torque load is defined as a twisting load applied to the IHS in an axis of rotation normal to the IHS top surface.
Datasheet35
1, 2
2, 3
2, 4
Page 36
Package Mechanical Specifications
3.1.5Package Insertion Specifications
The processor can be inserted into and removed from a mPGA478B socket 15 times. The socket
should meet the mPGA478B requirements detailed in the mPGA479, mPGA478A, mPGA478B, mPGA478C, and mPGA476 Socket Design Guidelines.
3.1.6Processor Mass Specification
The typical mass of the processor is 19 g [0.67 oz]. This mass [weight] includes all the components
that are included in the package.
3.1.7Processor Materials
Table 22 lists some of the package components and associated materials.
This chapter provides the processor pinout and signal description.
4.1Processor Pin Assignments
The pinout footprint is shown in Figure 10 and Figure 11. These figures represent the pinout
arranged by pin number. Table 23 provides the pinout arranged alphabetically by signal name and
Table 24 provides the pinout arranged numerically by pin number.
Datasheet39
Page 40
Pin List and Signal Description
Figure 10. Pinout Diagram (Top View—Left Side)
26252423222120191817161514
AF SKTOCC# Reserved Reserved BC LK1BCLK0VCCVSSVCCVSSVCCVSSVCCVSSAF
OPTIMIZED/
AE
COMPAT#
AD ITP_CLK1 TESTHI12 TESTHI0VSSVSSAVSSVCCIOPLLVCCVSSVCCVSSVCCVSSAD
AC ITP_CLK0VSSTESTHI4 TESTHI5VSSTESTHI2 TESTHI3VSSVCCVSSVCCVSSVCCAC
A[35:3]# (Address) define a 2
subphase 1 of the address phase, these pins transmit the address of a
transaction. In subphase 2, these pins transmit transaction type information.
A[35:3]#
A20M#Input
ADS#
Input/
Output
Input/
Output
These signals must connect the appropriate pins of all agents on the processor
FSB. A[35:3]# are protected by parity signals AP[1:0]#. A[35:3]# are source
synchronous signals and are latched into the receiving buffers by ADSTB[1:0]#.
On the active-to-inactive transition of RESET#, the processor samples a subset
of the A[35:3]# pins to determine power-on configuration. See Section 6.1 for
more details.
If A20M# (Address-20 Mask) is asserted, the processor masks physical address
bit 20 (A20#) before looking up a line in any internal cache and before driving a
read/write transaction on the bus. Asserting A20M# emulates the 8086
processor's address wrap-around at the 1-MB boundary. Assertion of A20M# is
only supported in real mode.
A20M# is an asynchronous signal. However, to ensure recognition of this signal
following an input/output write instruction, it must be valid along with the TRDY#
assertion of the corresponding input/output write bus transaction.
ADS# (Address Strobe) is asserted to indicate the validity of the transaction
address on the A[35:3]# and REQ[4:0]# pins. All bus agents observe the ADS#
activation to begin parity checking, protocol checking, address decode, internal
snoop, or deferred reply ID match operations associated with the new
transaction.
Address strobes are used to latch A[35:3]# and REQ[4:0]# on their rising and
falling edges. Strobes are associated with signals as shown below.
36
-byte physical memory address space. In
ADSTB[1:0]#
AP[1:0]#
BCLK[1:0]Input
Input/
Output
Input/
Output
SignalsAssociated Strobe
REQ[4:0]#, A[16:3]# ADSTB0#
A[35:17]#ADSTB1#
AP[1:0]# (Address Parity) are driven by the request initiator along with ADS#,
A[35:3]#, and the transaction type on the REQ[4:0]#. A correct parity signal is
high if an even number of covered signals are low and low if an odd number of
covered signals are low. This allows parity to be high when all the covered
signals are high. AP[1:0]# should connect the appropriate pins of all processor
FSB agents. The following table defines the coverage model of these signals.
Request SignalsSubphase 1Subphase 2
A[35:24]#AP0#AP1#
A[23:3]#AP1#AP0#
REQ[4:0]#AP1#AP0#
The differential pair BCLK (Bus Clock) determines the FSB frequency. All
processor FSB agents must receive these signals to drive their outputs and latch
their inputs.
All external timing parameters are specified with respect to the rising edge of
BCLK0 crossing V
CROSS
.
54Datasheet
Page 55
Table 25. Signal Description (Page 2 of 8)
NameTypeDescription
BINIT# (Bus Initialization) may be observed and driven by all processor FSB
agents and, if used, must connect the appropriate pins of all such agents. If the
BINIT# driver is enabled during power-on configuration, BINIT# is asserted to
signal any bus condition that prevents reliable future operation.
If BINIT# observation is enabled during power-on configuration, and BINIT# is
BINIT#
BNR#
BOOTSELECTInput
BPM[5:0]#
BPRI#Input
BR0#
BSEL[1:0]Output
COMP[1:0]Analog
Input/
Output
Input/
Output
Input/
Output
Input/
Output
sampled asserted, symmetric agents reset their bus LOCK# activity and bus
request arbitration state machines. The bus agents do not reset their IOQ and
transaction tracking state machines upon observation of BINIT# activation. Once
the BINIT# assertion has been observed, the bus agents will re-arbitrate for the
FSB and attempt completion of their bus queue and IOQ entries.
If BINIT# observation is disabled during power-on configuration, a central agent
may handle an assertion of BINIT# as appropriate to the error handling
architecture of the system.
BNR# (Block Next Request) is used to assert a bus stall by any bus agent who is
unable to accept new bus transactions. During a bus stall, the current bus owner
cannot issue any new transactions.
This input is required to determine whether the processor is installed in a
platform that supports the processor. The processor will not operate if this pin is
low. This input has a weak internal pullup.
BPM[5:0]# (Breakpoint Monitor) are breakpoint and performance monitor
signals. They are outputs from the processor that indicate the status of
breakpoints and programmable counters used for monitoring processor
performance. BPM[5:0]# should connect the appropriate pins of all processor
FSB agents.
BPM4# provides PRDY# (Probe Ready) functionality for the TAP port. PRDY# is
a processor output used by debug tools to determine processor debug
readiness.
BPM5# provides PREQ# (Probe Request) functionality for the TAP port. PREQ#
is used by debug tools to request debug operation of the processor.
Refer to the Intel
more detailed information.
These signals do not have on-die termination. Refer to Section 2.4, and the
®
865G/865GV/865PE/865P Chipset Platform Design Guide for termination
Intel
requirements.
BPRI# (Bus Priority Request) is used to arbitrate for ownership of the processor
FSB. It must connect the appropriate pins of all processor FSB agents.
Observing BPRI# active (as asserted by the priority agent) causes all other
agents to stop issuing new requests, unless such requests are part of an
ongoing locked operation. The priority agent keeps BPRI# asserted until all of its
requests are completed, then releases the bus by de-asserting BPRI#.
BR0# (Bus Request) drives the BREQ0# signal in the system and is used by the
processor to request the bus. During power-on configuration, this pin is sampled
to determine the agent ID = 0.
This signal does not have on-die termination and must be terminated.
The BCLK[1:0] frequency select signals BSEL[1:0] (Bus Select) are used to
select the processor input clock frequency. Table 7 defines the possible
combinations of the signals and the frequency associated with each
combination. The required frequency is determined by the processor, chipset,
and clock synthesizer. All agents must operate at the same frequency . For more
information about these pins, including termination recommendations, refer to
Section 2.8 and the appropriate platform design guidelines.
COMP[1:0] must be terminated on the system board using precision resistors.
Refer to the Intel
details on implementation.
Pin List and Signal Description
®
865G/865GV/865PE/865P Chipset Platform Design Guide for
®
865G/865GV/865PE/865P Chipset Platform Design Guide for
Datasheet55
Page 56
Pin List and Signal Description
Table 25. Signal Description (Page 3 of 8)
NameTypeDescription
D[63:0]# (Data) are the data signals. These signals provide a 64-bit data path
between the processor FSB agents, and must connect the appropriate pins on
all such agents. The data driver asserts DRDY# to indicate a valid data transfer.
D[63:0]# are quad-pumped signals and will thus be driven four times in a
common clock period. D[63:0]# are latched from the falling edge of both
DSTBP[3:0]# and DSTBN[3:0]#. Each group of 16 data signals correspond to a
pair of one DSTBP# and one DSTBN#. The following table shows the grouping
of data signals to data strobes and DBI#.
Quad-Pumped Signal Groups
D[63:0]#
DBI[3:0]#
Input/
Output
Furthermore, the DBI# pins determine the polarity of the data signals. Each
group of 16 data signals corresponds to one DBI# signal. When the DBI# signal
is active, the corresponding data group is inverted and therefore sampled active
high.
DBI[3:0]# (Data Bus Inversion) are source synchronous and indicate the polarity
of the D[63:0]# signals.The DBI[3:0]# signals are activated when the data on the
data bus is inverted. If more than half the data bits, within a 16-bit group, would
have been asserted electrically low, the bus agent may invert the data bus
signals for that particular sub-phase for that 16-bit group.
DBR# (Debug Reset) is used only in processor systems where no debug port is
DBR#Output
DBSY#
DEFER#Input
DP[3:0]#
Input/
Output
Input/
Output
implemented on the system board. DBR# is used by a debug port interposer so
that an in-target probe can drive system reset. If a debug port is implemented in
the system, DBR# is a no connect in the system. DBR# is not a processor signal.
DBSY# (Data Bus Busy) is asserted by the agent responsible for driving data on
the processor FSB to indicate that the data bus is in use. The data bus is
released after DBSY# is de-asserted. This signal must connect the appropriate
pins on all processor FSB agents.
DEFER# is asserted by an agent to indicate that a transaction cannot be
guaranteed in-order completion. Assertion of DEFER# is normally the
responsibility of the addressed memory or input/output agent. This signal must
connect the appropriate pins of all processor FSB agents.
DP[3:0]# (Data Parity) provide parity protection for the D[63:0]# signals. They
are driven by the agent responsible for driving D[63:0]#, and must connect the
appropriate pins of all processor FSB agents.
56Datasheet
Page 57
Table 25. Signal Description (Page 4 of 8)
NameTypeDescription
Pin List and Signal Description
DRDY#
DSTBN[3:0]#
DSTBP[3:0]#
FERR#/PBE#Output
GTLREFInput
HIT#
HITM#
IERR#Output
Input/
Output
Input/
Output
Input/
Output
Input/
Output
Input/
Output
DRDY# (Data Ready) is asserted by the data driver on each data transfer,
indicating valid data on the data bus. In a multi-common clock data transfer,
DRDY# may be de-asserted to insert idle clocks. This signal must connect the
appropriate pins of all processor FSB agents.
FERR#/PBE# (Floating Point Error/Pending Break Event) is a multiplexed signal
and its meaning is qualified by STPCLK#. When STPCLK# is not asserted,
FERR#/PBE# indicates a floating-point error and will be asserted when the
processor detects an unmasked floating-point error. When STPCLK# is not
asserted, FERR#/PBE# is similar to the ERROR# signal on the Intel 387
coprocessor, and is included for compatibility with systems using MS-DOS*-type
floating-point error reporting.
When STPCLK# is asserted, an assertion of FERR#/PBE# indicates that the
processor has a pending break event waiting for service. The assertion of
FERR#/PBE# indicates that the processor should be returned to the Normal
state. For additional information on the pending break event functionality,
including the identification of support of the feature and enable/disable
information, refer to volume 3 of the Intel Architecture Software Developer's Manual and the Intel Processor Identification and the CPUID Instruction
application note.
GTLREF determines the signal reference level for GTL+ input pins. GTLREF is
used by the GTL+ receivers to determine if a signal is a logical 0 or logical 1.
Refer to the Intel
more information.
HIT# (Snoop Hit) and HITM# (Hit Modified) convey transaction snoop operation
results. Any FSB agent may assert both HIT# and HITM# together to indicate
that it requires a snoop stall, which can be continued by reasserting HIT# and
HITM# together.
IERR# (Internal Error) is asserted by a processor as the result of an internal
error. Assertion of IERR# is usually accompanied by a SHUTDOWN transaction
on the processor FSB. This transaction may optionally be converted to an
external error signal (e.g., NMI) by system core logic. The processor will keep
IERR# asserted until the assertion of RESET#.
This signal does not have on-die termination. Refer to Section 2.4 for
termination requirements.
®
865G/865GV/865PE/865P Chipset Platform Design Guide for
Datasheet57
Page 58
Pin List and Signal Description
Table 25. Signal Description (Page 5 of 8)
NameTypeDescription
IGNNE# (Ignore Numeric Error) is asserted to force the processor to ignore a
numeric error and continue to execute noncontrol floating-point instructions. If
IGNNE# is de-asserted, the processor generates an exception on a noncontrol
IGNNE#Input
INIT#Input
ITP_CLK[1:0]Input
LINT[1:0]Input
LOCK#
MCERR#
OPTIMIZED/
COMPAT#
Input/
Output
Input/
Output
Input
floating-point instruction if a previous floating-point instruction caused an error.
IGNNE# has no effect when the NE bit in Control Register 0 (CR0) is set.
IGNNE# is an asynchronous signal. However, to ensure recognition of this signal
following an input/output write instruction, it must be valid along with the TRDY#
assertion of the corresponding input/output write bus transaction.
INIT# (Initialization), when asserted, resets integer registers inside the processor
without affecting its internal caches or floating-point registers. The processor
then begins execution at the power-on Reset vector configured during power-on
configuration. The processor continues to handle snoop requests during INIT#
assertion. INIT# is an asynchronous signal and must connect the appropriate
pins of all processor FSB agents.
If INIT# is sampled active on the active-to-inactive transition of RESET#, then
the processor executes its Built-in Self-Test (BIST).
ITP_CLK[1:0] are copies of BCLK that are used only in processor systems
where no debug port is implemented on the system board. ITP_CLK[1:0] are
used as BCLK[1:0] references for a debug port implemented on an interposer. If
a debug port is implemented in the system, ITP_CLK[1:0] are no connects in the
system. These are not processor signals.
LINT[1:0] (Local APIC Interrupt) must connect the appropriate pins of all APIC
Bus agents. When the APIC is disabled, the LINT0 signal becomes INTR (a
maskable interrupt request signal) and LINT1 becomes NMI (a nonmaskable
interrupt). INTR and NMI are backward compatible with the signals of those
names on the Pentium processor. Both signals are asynchronous.
Both of these signals must be software configured via BIOS programming of the
APIC register space to be used either as NMI/INTR or LINT[1:0]. Because the
APIC is enabled by default after Reset, operation of these pins as LINT[1:0] is
the default configuration.
LOCK# indicates to the system that a transaction must occur atomically. This
signal must connect the appropriate pins of all processor FSB agents. For a
locked sequence of transactions, LOCK# is asserted from the beginning of the
first transaction to the end of the last transaction.
When the priority agent asserts BPRI# to arbitrate for ownership of the processor
FSB, it will wait until it observes LOCK# de-asserted. This enables symmetric
agents to retain ownership of the processor FSB throughout the bus locked
operation and ensures the atomicity of lock.
MCERR# (Machine Check Error) is asserted to indicate an unrecoverable error
without a bus protocol violation. It may be driven by all processor FSB agents.
MCERR# assertion conditions are configurable at a system level. Assertion
options are defined by the following options:
For more details regarding machine check architecture, refer to the IA-32 Software Developer’s Manual, Volume 3: System Programming Guide.
This signal should be left as a no connect on the baseboard to indicate that the
baseboard supports the Intel® Pentium® 4 processor on 90 nm process. This
input has a weak internal pull-up.
• Enabled or disabled.
• Asserted, if configured, for internal errors along with IERR#.
• Asserted, if configured, by the request initiator of a bus transaction after it
observes an error.
• Asserted by any bus agent when it observes an error in a bus transaction.
58Datasheet
Page 59
Table 25. Signal Description (Page 6 of 8)
NameTypeDescription
As an output, PROCHOT# (Processor Hot) will go active when the processor
temperature monitoring sensor detects that the processor has reached its
PROCHOT#
PWRGOODInput
REQ[4:0]#
RESET#Input
RS[2:0]#Input
RSP#Input
SKTOCC#Output
SLP#Input
Input/
Output
Input/
Output
maximum safe operating temperature. This indicates that the processor Thermal
Control Circuit (TCC) has been activated, if enabled. As an input, assertion of
PROCHOT# by the system activates the TCC, if enabled. The TCC remains
active until the system de-asserts PROCHOT#.
PWRGOOD (Power Good) is a processor input. The processor requires this
signal to be a clean indication that the clocks and power supplies are stable and
within their specifications. The term ‘Clean’ implies that the signal will remain low
(capable of sinking leakage current), without glitches, from the time that the
power supplies are turned on until they come within specification. The signal
must then transition monotonically to a high state. PWRGOOD can be driven
inactive at any time, but clocks and power must again be stable before a
subsequent rising edge of PWRGOOD.
The PWRGOOD signal must be supplied to the processor; it is used to protect
internal circuits against voltage sequencing issues. It should be driven high
throughout boundary scan operation.
REQ[4:0]# (Request Command) must connect the appropriate pins of all
processor FSB agents. They are asserted by the current bus owner to define the
currently active transaction type. These signals are source synchronous to
ADSTB0#. Refer to the AP[1:0]# signal description for a details on parity
checking of these signals.
Asserting the RESET# signal resets the processor to a known state and
invalidates its internal caches without writing back any of their contents. For a
power-on Reset, RESET# must stay active for at least 1 ms after V
have reached their proper specifications. On observing active RESET#, all FSB
agents will de-assert their outputs within two clocks. RESET# must not be kept
asserted for more than 10 ms while PWRGOOD is asserted.
A number of bus signals are sampled at the active-to-inactive transition of
RESET# for power-on configuration. These configuration options are described
in the Section 6.1.
This signal does not have on-die termination and must be terminated on
the system board.
RS[2:0]# (Response Status) are driven by the response agent (the agent
responsible for completion of the current transaction), and must connect the
appropriate pins of all processor FSB agents.
RSP# (Response Parity) is driven by the response agent (the agent responsible
for completion of the current transaction) during assertion of RS[2:0]#, the
signals for which RSP# provides parity protection. It must connect to the
appropriate pins of all processor FSB agents.
A correct parity signal is high if an even number of covered signals are low and
low if an odd number of covered signals are low. While RS[2:0]# = 000, RSP# is
also high, since this indicates it is not being driven by any agent guaranteeing
correct parity.
SKTOCC# (Socket Occupied) will be pulled to ground by the processor. System
board designers may use this pin to determine if the processor is present.
SLP# (Sleep), when asserted in Stop-Grant state, causes the processor to enter
the Sleep state. During Sleep state, the processor stops providing internal clock
signals to all units, leaving only the Phase-Locked Loop (PLL) still operating.
Processors in this state will not recognize snoops or interrupts. The processor
will recognize only assertion of the RESET# signal, and de-assertion of SLP#. If
SLP# is de-asserted, the processor exits the Sleep state and returns to StopGrant state, restarting its internal clock signals to the bus and processor core
units.
Pin List and Signal Description
and BCLK
CC
Datasheet59
Page 60
Pin List and Signal Description
Table 25. Signal Description (Page 7 of 8)
NameTypeDescription
SMI# (System Management Interrupt) is asserted asynchronously by system
logic. On accepting a System Management Interrupt, the processor saves the
current state and enters System Management Mode (SMM). An SMI
SMI#Input
STPCLK#Input
TCKInput
TDIInput
TDOOutput
TESTHI[12:0]Input
THERMDAOtherThermal Diode Anode. See Section 5.2.6.
THERMDCOtherThermal Diode Cathode. See Section 5.2.6.
THERMTRIP#Output
TMSInput
TRDY#Input
TRST#Input
VCCInput
VCCAInput
Acknowledge transaction is issued, and the processor begins program execution
from the SMM handler.
If SMI# is asserted during the de-assertion of RESET#, the processor will tristate
its outputs.
STPCLK# (Stop Clock), when asserted, causes the processor to enter a low
power Stop-Grant state. The processor issues a Stop-Grant Acknowledge
transaction, and stops providing internal clock signals to all processor core units
except the FSB and APIC units. The processor continues to snoop bus
transactions and service interrupts while in Stop-Grant state. When STPCLK# is
de-asserted, the processor restarts its internal clock to all units and resumes
execution. The assertion of STPCLK# has no effect on the bus clock; STPCLK#
is an asynchronous input.
TCK (T est Clock) provides the clock input for the processor Test Bus (also known
as the Test Access Port).
TDI (Test Data In) transfers serial test data into the processor. TDI provides the
serial input needed for JTAG specification support.
TDO (Test Data Out) transfers serial test data out of the processor. TDO
provides the serial output needed for JTAG specification support.
TESTHI[12:0] must be connected to a V
proper processor operation. See Section 2.4 for more details.
In the event of a catastrophic cooling failure, the processor will automatically
shut down when the silicon has reached a temperature approximately 20 °C
above the maximum T
processor junction temperature has reached a level beyond which permanent
silicon damage may occur. Upon assertion of THERMTRIP#, the processor will
shut off its internal clocks (thus, halting program execution) in an attempt to
reduce the processor junction temperature. To protect the processor, its core
voltage (V
of the THERMTRIP# signal is enabled within 10 µs of the assertion of
PWRGOOD and is disabled on de-assertion of PWRGOOD. Once activated,
THERMTRIP# remains latched until PWRGOOD is de-asserted. While the deassertion of the PWRGOOD signal will de-assert THERMTRIP#, if the
processor’s junction temperature remains at or above the trip level,
THERMTRIP# will again be asserted within 10 µs of the assertion of
PWRGOOD.
TMS (Test Mode Select) is a JTAG specification support signal used by debug
tools.
TRDY# (Target Ready) is asserted by the target to indicate that it is ready to
receive a write or implicit writeback data transfer. TRDY# must connect the
appropriate pins of all FSB agents.
TRST# (Test Reset) resets the Test Access Port (TAP) logic. TRST# must be
driven low during power on Reset.
VCC are the power pins for the processor. The voltage supplied to these pins is
determined by the VID[5:0] pins.
VCCA provides isolated power for the internal processor core PLLs. Refer to the
Intel
implementation details.
power source through a resistor for
CC
. Assertion of THERMTRIP# (Thermal Trip) indicates the
C
) must be removed following the assertion of THERMTRIP#. Driving
CC
®
865G/865GV/865PE/865P Chipset Platform Design Guide for complete
60Datasheet
Page 61
Table 25. Signal Description (Page 8 of 8)
NameTypeDescription
Pin List and Signal Description
VCCIOPLLInput
VCCIOPLL
guidelines for VCCA, and refer to the Intel
provides isolated power for internal processor FSB PLLs. Follow the
®
865G/865GV/865PE/865P Chipset
Platform Design Guide for complete implementation details.
VCC_SENSE is an isolated low impedance connection to processor core power
VCC_SENSEOutput
VCCVIDInput
(V
). It can be used to sense or measure voltage near the silicon with little
CC
noise.
1.2 V is required to be supplied to the VCCVID pin if the platform is going to
support the processor. Refer to the Intel
®
865G/865GV/865PE/865P Chipset
Platform Design Guide for more information.
VCCVIDLBInput
1.2 V is required to be supplied to the VCCVIDLB pin if the platform is going to
support the processor. Refer to the Intel
®
865G/865GV/865PE/865P Chipset
Platform Design Guide for more information.
VID[5:0] (Voltage ID) pins are used to support automatic selection of power
supply voltages (V
processor and must be pulled up to 3.3 V with 1 kΩ 5% resistors. The voltage
). These are open drain signals that are driven by the
CC
supply for these pins must be valid before the voltage regulator (VR) can supply
V
VID[5:0]Output
to the processor. Conversely, the VR output must be disabled until the
CC
voltage supply for the VID pins becomes valid. The VID pins are needed to
support the processor voltage specification variations. See Table 3 for definitions
of these pins. The VR must supply the voltage that is requested by the pins, or
disable itself.
VIDPWRGDInput
VSSInput
The processor requires this input to determine that the V
voltages are stable and within specification.
VSS are the ground pins for the processor and should be connected to the
system ground plane.
VSSAInputVSSA is the isolated ground for internal PLLs.
VSS_SENSEOutput
VSS_SENSE is an isolated low impedance connection to processor core V
can be used to sense or measure ground near the silicon with little noise.
CCVID
and V
CCVIDLB
SS
. It
§
Datasheet61
Page 62
Pin List and Signal Description
62Datasheet
Page 63
Thermal Specifications and Design Considerations
5Thermal Specifications and Design
Considerations
5.1Processor Thermal Specifications
The processor requires a thermal solution to maintain temperatures within operating limits as set
forth in Section 5.1.1. Any attempt to operate the processor outside these operating limits may
result in permanent damage to the processor and potentially other components within the system.
As processor technology changes, thermal management becomes increasingly crucial when
building computer systems. Maintaining the proper thermal environment is key to reliable, longterm system operation.
A complete thermal solution includes both component and system level thermal management
features. Component level thermal solutions can include active or passive heatsinks attached to the
processor Integrated Heat Spreader (IHS). Typical system level thermal soluti ons may consist of
system fans combined with ducting and venting.
For more information on designing a component level thermal solution, refer to the Intel
Pentium
Note:The boxed processor will ship with a component thermal solution. Refer to Chapter 7 for details on
the boxed processor.
®
4 Processor on 90 nm Process Thermal Design Guidelines.
5.1.1Thermal Specifications
T o allow for the optimal operation and long-term reliability of Intel processor-based systems, the
system/processor thermal solution should be designed such that the processor remains within the
minimum and maximum case temperature (T
Thermal Design Power (TDP) value listed per frequency in Table 26. Thermal solutions not
designed to provide this level of thermal capability may affect the long-term reliability of the
processor and system. For more details on thermal solution design, refer to the IntelProcessor on 90 nm Process Thermal Design Guidelines.
The Pentium 4 processor on 90 nm process introduces a new methodology for managing processor
temperatures that is intended to support acoustic noise reduction through fan speed control.
Selection of the appropriate fan speed is based on the temperature reported by the processor’s
Thermal Diode. If the diode temperature is greater than or equal to T
temperature must remain at or below the temperature as specified by the thermal profile. If the
diode temperature is less than T
profile, but the diode temperature must remain at or below T
speed control must be designed to take these conditions into account. Systems that do not alter the
fan speed only need to guarantee the case temperature meets the thermal profile specifications.
control
®
) specifications when operating at or below the
C
®
Pentium® 4
, then the processor case
control
, then the case temperature is permitted to exceed the thermal
. Systems that implement fan
control
Datasheet63
Page 64
Thermal Specifications and Design Considerations
To determine a processor's case temperature specification based on the thermal profile, it is
necessary to accurately measure processor power dissipation. Intel has developed a methodology
for accurate power measurement that correlates to Intel test temperature and voltage conditions.
Refer to the Intel
®
Pentium® 4 Processor on 90 nm Process Thermal Design Guidelines for the
details of this methodology.
The case temperature is defined at the geometric top center of the processor IHS. Analysis
indicates that real applications are unlikely to cause the processor to consume maximum power
dissipation for sustained periods of time. Intel recommends that complete thermal solution designs
target the Thermal Design Power (TDP) indicated in Table 26 instead of the maximum processor
power consumption. The Thermal Monitor feature is intended to help protect the processor in the
unlikely event that an application exceeds the TDP recommendation for a sustained period of time.
For more details on the usage of this feature, refer to Section 5.2. In all cases, the Thermal
Monitor feature must be enabled for the processor to remain within specification.
Table 26. Processor Thermal Specifications
Core Frequency
(GHz)
2.80A/E (PRB = 0)895See Table 27 and Figure 12
3E (PRB = 0)895See Table 27 and Figure 12
3.20E (PRB = 0)895See Table 27 and Figure 12
3.40E (PRB = 0)895See Table 27 and Figure 12
3.20E (PRB = 1)1035See Table 27 and Figure 12
3.40E (PRB = 1)1035See Table 27 and Figure 12
NOTES:
1.Thermal Design Power (TDP) should be used for processor thermal solution design targets. The TDP is
not the maximum power that the processor can dissipate.
2.This table shows the maximum TDP for a given frequency range. Individual processors may have a lower
TDP. Therefore, the maximum T
Figure 12 and Table 27 for the allowed combinations of power and T
Thermal Design
Power (W)
C
Minimum T
will vary depending on the TDP of the individual processor. Refer to
The maximum and minimum case temperatures (TC) are specified in Table 27. These temperature
specifications are meant to help ensure proper operation of the processor. Figure 13 illustrates
where Intel recommends T
temperature measurement methodology, refer to the IntelProcess Thermal Design Guidelines.
Figure 13. Case Temperature (TC) Measurement Location
M easure from edge of processor IHS
35 m m x 35 mm sub strate [1.378 in x 1.378 in]
thermal measurements should be made. For detailed guidelines on
C
15.5 mm
[0 .61 in]
31 mm x 31 m m IHS [1.22 x 1.22 in]
®
Pentium® 4 Processor on 90 nm
M easure T
(geometric center of IHS)
15.5 mm
[0 .61 in]
at t his point
C
66Datasheet
Page 67
Thermal Specifications and Design Considerations
5.2Processor Thermal Features
5.2.1Thermal Monitor
The Thermal Monitor feature helps control the processor temperature by activating the TCC when
the processor silicon reaches its maximum operating temperature. The TCC reduces processor
power consumption as needed by modulating (starting and stopping) the internal processor core
clocks. The Thermal Monitor feature must be enabled for the processor to be operating within specifications. The temperature at which Thermal Monitor activates the thermal control
circuit is not user configurable and is not software visible. Bus traffic is snooped in the normal
manner, and interrupt requests are latched (and serviced during the time that the clocks are on)
while the TCC is active.
When the Thermal Monitor feature is enabled and a high temperature situation exists (i.e., TCC is
active), the clocks are modulated by alternately turning the clocks off and on at a duty cycle
specific to the processor (typically 30–50%). Clocks often will not be off for more than
3.0 microseconds when the TCC is active. Cycle times are processor speed dependent and decrease
as processor core frequencies increase. A small amount of hysteresis has been included to prevent
rapid active/inactive transitions of the TCC when the processor temperature is near its maximum
operating temperature. Once the temperature has dropped below the maximum operating
temperature and the hysteresis timer has expired, the TCC goes inactive and clock modulation
ceases.
With a properly designed and characterized thermal solution, it is anticipated that the TCC would
only be activated for very short periods of time when running the most power intensive
applications. The processor performance impact due to these brief periods of TCC activation is
expected to be so minor that it would be immeasurable. An under-designed thermal solution that is
not able to prevent excessive activation of the TCC in the anticipated ambient environment may
cause a noticeable performance loss, and in some cases may result in a T
specified maximum temperature and may affect the long-term reliability of the processor. In
addition, a thermal solution that is significantly under-designed may not be capable of cooling the
processor, even when the TCC is active continuously. Refer to the Intel90 nm Process Thermal Design Guidelines for information on designing a thermal solution.
The duty cycle for the TCC, when activated by the Thermal Monitor, is factory configured and
cannot be modified. The Thermal Monitor does not require any additional hardware, software
drivers, or interrupt handling routines.
that exceeds the
C
®
Pentium® 4 Processor on
Datasheet67
Page 68
Thermal Specifications and Design Considerations
5.2.2On-Demand Mode
The processor provides an auxiliary mechanism that allows system software to force the processor
to reduce its power consumption. This mechanism is referred to as "On-Demand" mode and is
distinct from the Thermal Monitor feature. On-Demand mode is intended as a means to reduce
system level power consumption. Systems using the processor must not rely on software usage of
this mechanism to limit the processor temperature.
If bit 4 of the ACPI P_CNT Control Register (located in the processor IA32_THERM_CONTROL
MSR) is written to a 1, the processor will immediately reduce its power consumption via
modulation (starting and stopping) of the internal core clock, independent of the processor
temperature. When using On-Demand mode, the duty cycle of the clock modulation is
programmable via bits 3:1 of the same ACPI P_CNT Control Register. In On-Demand mode, the
duty cycle can be programmed from 12.5% on/ 87.5% off, to 87.5% on/12.5% off in 12.5%
increments. On-Demand mode may be used in conjunction with the Thermal Monitor . If the system
tries to enable On-Demand mode at the same time the TCC is engaged, the factory configured duty
cycle of the TCC will override the duty cycle selected by the On-Demand mode.
5.2.3PROCHOT# Signal Pin
An external signal, PROCHOT# (processor hot), is asserted when the processor die temperature
has reached its maximum operating temperature. If the Thermal Monitor is enabled (note that the
Thermal Monitor must be enabled for the processor to be operating within specification), the TCC
will be active when PROCHOT# is asserted. The processor can be configured to generate an
interrupt upon the assertion or de-assertion of PROCHOT#. Refer to the Intel Architecture Software Developer's Manuals for specific register and programming details.
The processor implements a bi-directional PROCHOT# capability to allow system designs to
protect various components from over-temperature situations. The PROCHOT# signal is bidirectional in that it can either signal when the processor has reached its maximum operating
temperature or be driven from an external source to activate the TCC. The ability to activate the
TCC via PROCHOT# can provide a means for thermal protection of system components.
One application is the thermal protection of voltage regulators (VR). System designers can create a
circuit to monitor the VR temperature and activate the TCC when the temperature limit of the VR
is reached. By asserting PROCHOT# (pulled-low) and activating the TCC, the VR can cool down
as a result of reduced processor power consumption. Bi-directional PROCHOT# can allow VR
thermal designs to target maximum s ustain ed curr ent instead of maximum current. Systems should
still provide proper cooling for the VR, and rely on bi-directional PROCHOT# only as a backup in
case of system cooling failure. The system thermal design should allow the power delivery
circuitry to operate within its temperature specification even while the processor is operating at its
Thermal Design Power. With a properly designed and characterized thermal solution, it is
anticipated that bi-directional PROCHOT# would only be asserted for very short periods of time
when running the most power intensive applications. An under-designed thermal solution that is
not able to prevent excessive assertion of PROCHOT# in the anticipated ambient environment may
cause a noticeable performance loss. Refer to the Intel
Platform Design Guide and the Voltage Regulator-Down (VRD) 10.0 Design Guid elines for
Desktop Socket 478 for details on implementing the bi-directional PROCHOT# feature.
®
865G/865GV/865PE/865P Chipset
68Datasheet
Page 69
5.2.4THERMTRIP# Signal Pin
Regardless of whether or not the Thermal Monitor feature is enabled, in the event of a catastrophic
cooling failure, the processor will automatically shut down when the silicon has reached an
elevated temperature (refer to the THERMTRIP# definition in Table 25). At this point, the FSB
signal THERMTRIP# will go active and stay active as described in Table 25. THERMTRIP#
activation is independent of processor activity and does not generate any bus cycles. If
THERMTRIP# is asserted, processor core voltage (V
Thermal Specifications and Design Considerations
) must be removed.
CC
5.2.5T
T
value for T
T
control
control
control
and Fan Speed Reduction
is a temperature specification based on a temperature reading from the thermal diode. The
will be calibrated in manufacturing and configured for each processor. The
control
temperature for a given processor can be obtained by reading the
IA32_TEMPERATURE_TARGET MSR in the processor. The T
IA32_TEMPERATURE_TARGET MSR needs to be converted from Hexadecimal to Decimal and
added to a base value of 50 °C.
The value of T
When T
diode
is above T
may vary from 00h to 1Eh (0 to 30 °C).
control
in Table 27 and Figure 12; otherwise, the processor temperature can be maintained at T
lower) as measured by the thermal diode.
The purpose of this feature is to support acoustic optimization through fan speed control.
5.2.6Thermal Diode
The processor incorporates an on-die thermal diode. A thermal sensor located on the system board
may monitor the die temperature of the processor fo r th ermal management/long term die
temperature change purposes. Table 28 and Table 29 provide the diode parameter and interface
specifications. This thermal diode is separate from the Thermal Monitor’s thermal sensor and
cannot be used to predict the behavior of the Thermal Monitor.
, then Tc must be at or below T
control
value that is read from the
control
as defined by the thermal profile
c(max)
control
(or
Datasheet69
Page 70
Thermal Specifications and Design Considerations
Table 28. Thermal Diode Parameters
SymbolParameterMinTypMaxUnitNotes
I
Forward Bias Current11187uA
FW
nDiode Ideality Factor1.00831.0111.023
R
Series Resistance3.2423.333.594Ω
T
NOTES:
1.Intel does not support or recommend operation of the thermal diode under reverse bias.
2.Characterized at 75 °C.
3.Not 100% tested. Specified by design characterization.
4.The ideality factor, n, represents the deviation from ideal diode behavior as exemplified by the diode equation:
I
= IS * (e
where I
Constant, and T = absolute temperature (Kelvin).
= saturation current, q = electronic charge, VD = voltage across the diode, k = Boltzmann
S
FW
5.Devices found to have an ideality factor of 1.0183 to 1.023 will create a temperature error of approximately
2 C higher than the actual temperature. In order to minimize any potential acoustic impact of this temperature
error, Tcontrol will be increased by 2 C on these parts.
6.The series resistance, RT, is provided to allow for a more accurate measurement of the thermal diode temperature. R
, as defined, includes the pins of the processor but does not include any socket resistance or
T
board trace resistance between the socket and the external remote diode thermal sensor. RT can be used
by remote diode thermal sensors with automatic series resistance cancellation to calibrate out this error term.
Another application is that a temperature offset can be manually calculated and programmed into an offset
register in the remote diode thermal sensors as exemplified by the equation:
T
error
where T
charge.
= sensor temperature error, N = sensor current ratio, k = Boltzmann Constant, q = electronic
error
qVD/nkT
= [RT * (N-1) * I
–1)
FWmin
1
2, 3, 4, 5
2, 3, 6
] / [nk/q * ln N]
Table 29. Thermal Diode Interface
Pin NamePin NumberPin Description
THERMDAB3diode anode
THERMDCC4diode cathode
§
70Datasheet
Page 71
6Features
This chapter contains power-on configuration options and clock control/low power state
descriptions.
6.1Power-On Configuration Options
Several configuration options can be configured by hardware. The processor samples the hardware
configuration at reset, on the active-to-inactive transition of RESET#. For specifications on these
options, refer to Table 30.
The sampled information configures the processor for subsequent operation. These configuration
options cannot be changed except by another reset. All resets reconfigure the processor; for reset
purposes, the processor does not distinguish between a "warm" reset and a "power-on" reset.
Table 30. Power-On Configuration Option Pins
Configuration OptionPin
Output tristateSMI#
Execute BISTINIT#
In Order Queue pipelining (set IOQ depth to 1)A7#
Disable MCERR# observationA9#
Disable BINIT# observationA10#
APIC Cluster ID (0-3)A[12:11]#
Disable bus parkingA15#
Disable Hyper-Threading TechnologyA31#
Symmetric agent arbitration IDBR0#
RESERVEDA[6:3]#, A8#, A[14:13]#, A[16:30]#, A[32:35]#
NOTES:
1.Asserting this signal during RESET# will select the corresponding option.
2.Address pins not identified in this table as configuration options should not be asserted during RESET#.
Features
1,2
Datasheet71
Page 72
Features
6.2Clock Control and Low Power States
The processor allows the use of AutoHALT, Stop-Grant, and Sleep states to reduce power
consumption by stopping the clock to internal sections of the processor, depending on each
particular state. See Figure 14 for a visual representation of the processor low power states.
6.2.1Normal State—State 1
This is the normal operating state for the processor.
6.2.2AutoHALT Powerdown State—State 2
AutoHALT is a low power state entered when the processor executes the HALT instruction. The
processor transitions to the Normal state upon the occurrence of SMI#, BINIT#, INIT#, or
LINT[1:0] (NMI, INTR). RESET# causes the processor to immediately initialize itself.
The return from a System Management Interrupt (SMI) handler can be to either Normal Mode or
the AutoHALT powerdown state. See the Intel Architecture Software Developer's Manual, Volume III: System Programmer's Guide for more information.
The system can generate a STPCLK# while the processor is in the AutoHALT powerdown state.
When the system de-asserts the STPCLK# interrupt, the processor will return execution to the
HALT state.
While in AutoHALT powerdown state, the processor will process FSB snoops and interrupts.
Figure 14. Stop Clock State Machine
2. Auto HALT Power Down State
BCLK running.
Snoops and interrupts allowed.
Snoop
Event
Occurs
4. HALT/Grant Snoop State
BCLK running.
Service snoops to caches.
Snoop
Event
Serviced
HALT Instruction and
HALT Bus Cycle Generated
INIT#, BINIT#, INTR, NMI,
SMI#, RESET#
S
T
P
C
L
K
#
A
s
s
e
r
t
e
d
S
T
P
C
L
K
#
D
e
-
a
s
s
e
r
t
e
d
Snoop Event Occurs
Snoop Event Serviced
1. Normal State
Normal execution.
STPCLK#
Asserted
3. Stop Grant State
BCLK running.
Snoops and interrupts allowed.
SLP#
Asserted
5. Sleep State
BCLK running.
No snoops or interrupts allowed.
STPCLK#
De-asserted
SLP#
De-asserted
72Datasheet
Page 73
6.2.3Stop-Grant State—State 3
When the STPCLK# pin is asserted, the Stop-Grant state of the processor is entered 20 bus clocks
after the response phase of the processor-issued Stop Grant Acknowledge special bus cycle.
Since the GTL+ signal pins receive power from the FSB, these pins should not be driven (allowing
the level to return to V
addition, all other input pins on the FSB should be driven to the inactive state.
BINIT# will not be serviced while the processor is in Stop-Grant state. The event will be latched
and can be serviced by software upon exit from the Stop Grant state.
RESET# will cause the processor to immediately initialize itself, but the processor will stay in
Stop-Grant state. A transition back to the Normal state occurs with the de-assertion of the
STPCLK# signal. When re-entering the Stop Grant state from the Sleep state, STPCLK# should
only be de-asserted one or more bus clocks after the de-assertion of SLP#.
A transition to the HALT/Grant Snoop state occurs when the processor detects a snoop on the FSB
(see Section 6.2.4). A transition to the Sleep state (see Section 6.2.5) occurs with the assertion of
the SLP# signal.
While in the Stop-Grant State, SMI#, INIT#, BINIT#, and LINT[1:0] are latched by the processor,
and only serviced when the processor returns to the Normal State. Only one occurrence of each
event will be recognized upon return to the Normal state.
) for minimum power drawn by the termination resistors in this state. In
CC
Features
While in Stop-Grant state, the processor will process snoops on the FSB and it will latch interrupts
delivered on the FSB.
The PBE# signal can be driven when the processor is in Stop-Grant state. PBE# will be asserted if
there is any pending interrupt latched within the processor. Pending interrupts that are blocked by
the EFLAGS.IF bit being clear will still cause assertion of PBE#. Assertion of PBE# indicates to
system logic that it should return the processor to the Normal state.
6.2.4HALT/Grant Snoop State—State 4
The processor responds to snoop or interrupt transactions on the FSB while in Stop-Grant state or
in AutoHALT powerdown state. During a snoop or interrupt transaction, the processor enters the
HALT/Grant Snoop state. The processor will stay in this state until the snoop on the FSB has been
serviced (whether by the processor or another agent on the FSB) or the interrupt has been latched.
After the snoop is serviced or the interrupt is latched, the processor will return to the Stop-Grant
state or AutoHALT powerdow n state, as appropriate.
Datasheet73
Page 74
Features
6.2.5Sleep State—State 5
The Sleep state is a very low power state in which the processor maintains its context, maintains
the phase-locked loop (PLL), and has stopped all internal clocks. The Sleep state can only be
entered from Stop-Grant state. Once in the Stop-Grant state, the processor will enter the Sleep state
upon the assertion of the SLP# signal. The SLP# pin should only be asserted when the processor is
in the Stop Grant state. SLP# assertions while the processor is not in the Stop Grant state is out of
specification and may result in erroneous processor operation.
Snoop events that occur while in Sleep State or during a transition into or out of Sleep state will
cause unpredictable behavior.
In the Sleep state, the processor is incapable of responding to snoop transactions or latching
interrupt signals. No transitions or assertions of signals (with the exception of SLP# or RESET#)
are allowed on the FSB while the processor is in Sleep state. Any transition on an input signal
before the processor has returned to Stop-Grant state will result in unpredictable behavior.
If RESET# is driven active while the processor is in the Sleep state, and held active as specified in
the RESET# pin specification, then the processor will reset itself, ignoring the transition through
Stop-Grant State. If RESET# is driven active while the processor is in the Sleep State, the SLP#
and STPCLK# signals should be de-asserted immediately after RESET# is asserted to ensure the
processor correctly executes the reset sequence.
Once in the Sleep state, the SLP# pin must be de-asserted if another asynchronous FSB event needs
to occur. The SLP# pin has a minimum assertion of one BCLK period.
When the processor is in the Sleep state, it will not respond to interrupts or snoop transactions.
§
74Datasheet
Page 75
Boxed Processor Specifications
7Boxed Processor Specifications
The processor will also be offered as an Intel boxed processor. Intel boxed processors are intended
for system integrators who build systems from baseboards and standard components. The boxed
processor will be supplied with a cooling solution. This chapter documents baseboard and system
requirements for the cooling solution that will be supplied with the boxed processor. This chapter is
particularly important for OEMs that manufacture baseboards for system integrators. Unless
otherwise noted, all figures in this chapter are dimensioned in millimeters and inches [in brackets].
Figure 15 shows a mechanical representation of a boxed processor.
Note:
• Drawings in this section reflect only the specifications on the Intel boxed processor product.
These dimensions should not be used as a generic keep-out zone for all cooling solutions. It is
the system designer's responsibility to consider their proprietary cooling solution when
designing to the required keep-out zone on their system platform and chassis. Refer to the
®
Intel
Pentium® 4 Processor on 90 nm Process Thermal Design Guidelines for further
guidance. Contact your local Intel Sales Representative for this document.
Figure 15. Mechanical Representation of the Boxed Processor
NOTE: The airflow of the fan heatsink is into the center and out of the sides of the fan heatsink.
Datasheet75
Page 76
Boxed Processor Specifications
7.1Mechanical Specifications
7.1.1Boxed Processor Cooling Solution Dimensions
This section documents the mechanical specifications of the boxed processor. The boxed processor
will be shipped with an unattached fan heatsink. Figure 15 shows a mechanical representation of
the boxed processor.
Clearance is required around the fan heatsink to ensure unimpeded airflow for proper cooling. The
physical space requirements and dimensions for the boxed processor with assembled fan heatsink
are shown in Figure 16 (side views), and Figure 17 (top view). The airspace requirements for the
boxed processor fan heatsink must also be incorporated into new baseboard and system designs.
Airspace requirements are shown in Figure 20 and Figure 21. Note that some figures have
centerlines shown (marked with alphabetic designations) to clarify relative dimensioning.
Figure 16. Space Requiremen ts for the Boxed Processor (Side View)
76Datasheet
Page 77
Boxed Processor Specifications
Figure 17. Space Requirements for the Boxed Processor (Top View)
7.1.2Boxed Processor Fan Heatsink Weight
The boxed processor fan heatsink will not weigh more than 450 grams. See Chapter 5 and the
®
Intel
Pentium® 4 Processor on 90 nm Process Thermal Design Guidelines for details on the
processor weight and heatsink requirements.
Note:The processor retention mechanism, based on the Intel reference design, should be used to ensure
compatibility with the heatsink attach clip assembly and the boxed processor thermal solution. The
heatsink attach clip assembly is latched to the retention tab features at each corner of the retention
mechanism.
The target load applied by the clips to the processor heat spreader for Intel's reference design is
75 ±15 lbf (maximum load is constrained by the package load capability). It is normal to observe a
bow or bend in the board due to this compressive load on the processor package and the socket.
The level of bow or bend depends on the motherboard material properties and component layout.
Any additional board stiffening devices (such as plates) are not necessary and should not be used
along with the reference mechanical components and boxed processor. Using such devices
increases the compressive load on the processor package and socket, likely beyond the maximum
load that is specified for those components. See the IntelThermal Design Guidelines for details on the Intel reference design.
Chassis that have adequate clearance between the motherboard and chassis wall (minimum
0.250 in ch) should be selected to ensure the board's underside bend does not contact the chassis.
Datasheet77
®
Pentium® 4 Processor on 90 nm Process
Page 78
Boxed Processor Specifications
7.1.3Boxed Processor Retention Mechanism and Heatsink
Attach Clip Assembly
The boxed processor thermal solution requires a processor retention mechanism and a heatsink
attach clip assembly to secure the processor and fan heatsink in the baseboard socket. The boxed
processor will not ship with retention mechanisms but will ship with the heatsink attach clip
assembly. Baseboards designed for use by system integrators should include the retention
mechanism that supports the boxed processor. Baseboard documentation should include
appropriate retention mechanism installation instructions.
7.2Electrical Requirements
7.2.1Fan Heatsink Power Supply
The boxed processor's fan heatsink requires a +12 V power supply. A fan power cable will be
shipped with the boxed processor to draw power from a power header on the baseboard. The power
cable connector and pinout are shown in Figure 18. Baseboards must provide a matched power
header to support the boxed processor. Table 31 contains specifications for the input and output
signals at the fan heatsink connector. The fan heatsink outputs a SENSE signal that is an opencollector output that pulses at a rate of 2 pulses per fan revolution. A baseboard pull-up resistor
provides VOH to match the system board-mounted fan speed monitor requirements, if applicable.
Use of the SENSE signal is optional. If the SENSE signal is not used, pin 3 of the connector should
be tied to GND.
Note: The motherboard must supply a constant +12 V to the processor's power header to ensure proper
operation of the variable speed fan for the boxed processor.
The power header on the baseboard must be positioned to allow the fan heatsink power cable to
reach it. The power header identification and location should be documented in the platform
documentation, or on the system board itself. Figure 19 shows the location of the fan power
connector relative to the processor socket. The baseboard power header should be positioned
within 4.33 inches from the center of the processor socket.
Figure 18. Boxed Processor Fan Heatsink Power Cable Connector Description
PinSignal
1
12
GND
2
+12V
3
SENSE
3
Straight square pin, 3-pin terminal housing w ith
polarizing ribs and friction locking ramp .
0.100" pin pitch, 0.025" square pin w idth.
Waldom*/Molex* P/N 22-01-3037 or equivalent.
Match with straight pin, friction lock header on m otherboard
Waldom/Molex P/N 22-23-2031, AMP* P/N 640456-3,
or equivalent.
78Datasheet
Page 79
Boxed Processor Specifications
Table 31. Fan Heatsink Power and Signal Specifications
DescriptionMinTypMaxUnitNotes
+12 V: 12 volt fan power supply10.21213.8V
IC: Fan current draw740mA
SENSE: SENSE frequency2
NOTES:
1.Baseboard should pull this pin up to 5 V with a resistor.
pulses per fan
revolution
Figure 19. Baseboard Power Header Placement Relative to Processor Socket
1
7.3Thermal Specifications
This section describes the cooling requirements of the fan heatsink solution used by the boxed
processor.
7.3.1Boxed Processor Cooling Requirements
The boxed processor may be di rectly coo led wi th a fan heatsink. However, meeting the processor's
temperature specification is also a function of the thermal design of the entire system, and
ultimately the responsibility of the system integrator. The processor temperature specification is
presented in Chapter 5. The boxed processor fan heatsink is able to keep the processor temperature
within the specifications (see Table 26) in chassis that provide good thermal m ana gem ent. For the
Datasheet79
Page 80
Boxed Processor Specifications
boxed processor fan heatsink to operate properly, it is critical that the airflow provided to the fan
heatsink is unimpeded. Airflow of the fan heatsink is into the center and out of the sides of the fan
heatsink. Airspace is required around the fan to ensure that the airflow through the fan heatsink is
not blocked. Blocking the airflow to the fan heatsink reduces the cooling efficiency and decreases
fan life. Figure 20 and Figure 21 illustrate an acceptable airspace clearance for the fan heatsink.
The air temperature entering the fan is required to be at or below 38 °C. Again, meeting the
processor's temperature specification is the responsibility of the system integrator.
The boxed processor fan operates at different speeds over a short range of internal chassis
temperatures. This allows the processor fan to operate at a lower speed and noise level, while
internal chassis temperatures are low. If the internal chassis temperature increases beyond a lower
set point, the fan speed will rise linearly with the internal temperature until the higher set point is
reached. At that point, the fan speed is at its maximum. As fan speed increases, so does fan noise
levels. Systems should be designed to provide adequate air around the boxed processor fan
heatsink that remains below the lower set point. These set points, represented in Figure 22 and
Table 32, can vary by a few degrees from fan heatsink to fan heatsink. The internal chassis
temperature should be kept below 38 ºC. Meeting the processor's temperature specification
(see Chapter 5) is the responsibilit y of the system integ rato r.
Note:The motherboard must supply a constant +12 V to the processor's power header to ensure proper
operation of the variable speed fan for the boxed processor (refer to Table 31 for the specific
requirements).
Figure 22. Boxed Processor Fan Heatsink Set Points
Boxed Processor Specifications
Higher Set Point
Highest Noise Level
Increasing Fan
Speed & Noise
Lower Set Point
Lowest Noise Level
X
Internal Chassis Temperature (D e grees C)
Table 32. Boxed Processor Fan Heatsink Set Points
Boxed Processor Fan
Heatsink Set Point (ºC)
X ≤ 30 °C
Y = 34 °C
Z ≥ 38 °C
NOTES:
1.Set point variance is approximately ±1 °C from fan heatsink to fan heatsink.
When the internal chassis temperature is below or equal to this set point,
the fan operates at its lowest speed. Recommended maximum internal
chassis temperature for nominal operating environment.
When the internal chassis temperature is at this point, the fan operates
between its lowest and highest speeds. Recommended maximum
internal chassis temperature for worst-case operating environment.
When the internal chassis temperature is above or equal to this set point,
the fan operates at its highest speed.
Boxed Processor Fan SpeedNotes
YZ
1
1
§
Datasheet81
Loading...
+ hidden pages
You need points to download manuals.
1 point = 1 manual.
You can buy points or you can get point for every manual you upload.