The QuickRAM family of ESPs (Embedded Standard Products) offers FPGA logic in
combination with Dual-Port SRAM modules. The QL4016 is a 16,000 usable PLD gate
member of the QuickRAM family of ESPs. QuickRAM ESPs are fabricated on a 0.35 µm
four-layer metal process using QuickLogic's patented ViaLink
unique combination of high performance, high density, low cost, and extreme ease-of-use.
The QL4016 contains 320 logic cells and 10 Dual Port RAM modules (see Figure 1). Each
RAM module has 1,152 RAM bits, for a total of 11,520 bits. RAM Modules are Dual Port
(one read port, one write port) and can be configured into one of four modes:
64 (deep) × 18 (wide), 128 × 9, 256 × 4, or 512 × 2 (see
I/Os, the QL4016 is available in 84-pin PLCC, 100-pin TQFP, 100-pin CQFP and 144-pin
TQFP packages.
Designers can cascade multiple RAM modules to increase the depth or width allowed in
single modules by connecting corresponding address lines together and dividing the words
between modules (see
large as 16 bits wide in the smallest QuickRAM device and 44 bits wide in the largest device.
TM
technology to provide a
Figure 4). With a maximum of 82
Figure 2). This approach allows up to 512-deep configurations as
Software support for the complete QuickRAM family, including the QL4016, is available
through two basic packages. The turnkey QuickWorks
TM
package provides the most
complete ESP software solution from design entry to logic synthesis, to place and route, to
simulation. The QuickTools package provides a solution for designers who use Cadence,
Exemplar, Mentor, Synopsys, Synplicity, Viewlogic, Aldec, or other third-party tools for
design entry, synthesis, or simulation.
The QuickLogic
TM
variable grain logic cell features up to 16 simultaneous inputs and five
outputs within a cell that can be fragmented into five independent cells. Each cell has a fanin of 29 including register and control lines (see
• 110 bi-directional input/output pins, PCI-compliant for 5.0 V and 3.3 V buses for
-1/-2/-3/-4 speed grades
• 8 high-drive input/distributed network pins
Eight Low-Skew Distributed Networks
• Two array clock/control networks available to the logic cell flip-flop clock, set and reset
inputs—each driven by an input-only pin
• Six global clock/control networks available to the logic cell F1, clock, set and reset inputs
and the input and I/O register clock, reset and enable inputs as well as the output enable
contro—each driven by an input-only or I/O pin, or any logic cell output or I/O cell
feedback
High Performance Silicon
• Input + logic cell + output total delays = under 6 ns
AC Characteristic s at VCC = 3.3 V, TA = 25°C (K = 1.00)
To calculate delays, multiply the appropriate K factor from Table 10: Operating Range by the
following numbers in the tables provided.
QS
A1
A2
A3
A4
A5
A6
QS
OP
B1
B2
C1
C2
MP
MS
D1
D2
E1
E2
NP
NS
F1
F2
F3
F4
F5
F6
QC
QR
AZ
OZ
QZ
NZ
FZ
Figure 3: QuickRAM Logic Cell
Table 1: Logic Cell
SymbolParameter
t
PD
t
SU
t
H
t
CLK
t
CWHI
t
CWLO
t
SET
t
RESET
t
SW
t
RW
Combinatorial Delay
Setup Time
a
Hold Time0.00.00.00.00.0
Clock to Q Delay0.71.01.21.52.5
Clock High Time1.21.21.21.21.2
Clock Low Time1.21.21.21.21.2
Set Delay1.01.31.51.82.8
Reset Delay0.81.11.31.62.6
Set Width1.91.91.91.91.9
Reset Width1.81.81.81.81.8
a
a. These limits are derived from a representative selection of the slowest paths through the Quick-
RAM logic cell including typical net delays. Worst case delay values for specific paths should be
determined from timing analysis of your particular design.
Global Clock Buffer Delay0.8 0.8 0.9 0.9 1.1 1.2 1.3
a. The array dist ribu ted ne tw ork s co nsi st of 40 ha lf columns and the glo bal d is tributed networks con-
sist of 44 half columns, each driven by an independent buffer. The number of half columns used
does not affect clo ck buffer delay . The arra y clock has up to eight loads per ha lf column. The global
clock has up to 11 loads per half column.