The IDT7210 is a high-speed, low-power 16 x 16-bit parallel
multiplier-accumulator that is ideally suited for real-time digital
signal processing applications. Fabricated using CMOS
silicon gate technology, this device offers a very low-power
alternative to existing bipolar and NMOS counterparts, with
only 1/7 to 1/10 the power dissipation and exceptional speed
(25ns maximum) performance.
A pin and functional replacement for TRW’s TDC1010J the
IDT7210 operates from a single 5 volt supply and is compatible
with standard TTL logic levels. The architecture of the IDT7210
is fairly straightforward, featuring individual input and output
registers with clocked D-type flip-flop, a preload capability
which enables input data to be preloaded into the output
registers, individual three-state output ports for the Extended
Product (XTP) and Most Significant Product (MSP) and a
Least Significant Product output (LSP) which is multiplexed
with the Y input.
The X
IN and YIN data input registers may be specified
through the use of the Two’s Complement input (TC) as either
a two’s complement or an unsigned magnitude, yielding a fullprecision 32-bit result that may be accumulated to a full 35-bit
result. The three output registers – Extended Product (XTP),
Most Most Significant Product (MSP) and Least Significant
Product (LSP) – are controlled by the respective TSX, TSM
and TSL input lines. The LSP output can be routed through YIN
ports.
FUNCTIONAL BLOCK DIAGRAM
16
3
PREL
ACC, SUB,
416
CONTROL
REGISTER
MULTIPLIER ARRAY
ACCUMULATOR
MSP REGISTERLSP REGISTERXTP REGISTER
MSP
(P31-P16)
IDT7210
32
35
16
+
TSM
OUT
YREGISTER
IN
Y
(Y15-Y0/P15-P0)
TSL
16
2577 drw 01
PREL
1
IN
X
CLKXCLKYRND, TC
35
CLKP
TSX
3
XTPOUT
(P34-P32)
(X15-X0)
XREGISTER
+/–
MILITARY AND COMMERCIAL TEMPERATURE RANGESAUGUST 1995
IDT7210L
16 x 16 PARALLEL CMOS MULTIPLIER-ACCUMULATORMILITARY AND COMMERCIAL TEMPERATURE RANGES
DESCRIPTION (Continued)
The Accumulate input (ACC) enables the device to perform
either a multiply or a multiply-accumulate function. In the
multiply-accumulate mode, output data can be added to or
subtracted from previous results. When the Subtraction (SUB)
input is active simultaneously with an active ACC, a subtraction
can be performed. The double precision accumulated result is
rounded down to either a single precision or single precision
plus 3-bit extended result. In the multiply mode, the Extended
Product output (XTP) is sign extended in the two’s complement
mode or set to zero in the unsigned mode. The Round (RND)
control rounds up the Most Significant Product (MSP) and the
3-bit Extended Product (XTP) outputs. When Preload input
(PREL) is active, all the output buffers are forced into a highimpedance state (see Preload truth table) and external data
can be loaded into the output register by using the TSX, TSL
and TSM signals as input controls.
IDT7210L
16 x 16 PARALLEL CMOS MULTIPLIER-ACCUMULATORMILITARY AND COMMERCIAL TEMPERATURE RANGES
11
10
09
08
07
06
05
04
03
02
01
Pin 1
Designator
NCX15RNDCLKY TC PREL CLKP P33
X14CLKX VCC TSXP34NCX13
TSL SUBTSM
X11X12
X9X10
X7X8
X5X6
X3X4
X1X2
Y0,
X0
P0
Y1,P1Y3,P3Y5,P5Y7,
NC
Y2,
Y4,P4Y6,
P2
ABCDEFGHJKL
P6
ACC
G68-2
Y8,P8Y10,
P7
Y9,P9Y11,
GND
PGA
TOP VIEW
P10
P11
Y12,
P12
Y13,
P13
Y14,
P14
Y15,
P15
32
P
P30
P31
P28P29
P26P27
P24P25
P22P23
P20P21
P18P19
P16P17
NC
2577 drw 05
PIN DESCRIPTIONS
Pin NameI/ODescription
X0 - 15IData Inputs
Y0 - 15/ P0 - 15I/OMultiplexed I/O port. Y0 - 15 are data inputs and can be used to preload LSP register on PREL = 1. P0 - 15
are LSP register outputs - enabled by TSL.
P16 - 31I/OMSP register outputs - enabled by TSM. MSP register can be preloaded when PREL = 1.
P32 - 34I/OXTP register outputs - enabled by TSX. XTP register can be preloaded through these inputs when
PREL = 1.
CLKXIInput data X0 - 15 loaded in X input register on CLKX rising edge.
CLKYIInput data Y0 - 15 loaded in Y input register on CLKY rising edge.
CLKPIOutput data loaded into output register on rising edge of CLKP.
TSXITSX = 0 enables XTP outputs, TSX = 1 tristates P32 - 34 lines.
TSMITSM = 0 enables MSP outputs, TSM = 1 tristates P16 - 31 lines.
TSLITSL = 0 enables LSP outputs, TSL = 1 tristates P0 - 15 lines.
PRELIWhen PREL= 1 data is input on P0 - 15 lines. When PREL = 0, inputs on these lines are ignored.
ACCIThis input is loaded into the control register on the rising edge of (CLKX + CLKY).
When ACC = 1 and SUB = 0 an accumulate operation is performed. When ACC = 1 and SUB = 1, a
subtract operation is performed. When ACC = 0, the SUB input is a don't care and the device acts as a
simple multipler with no accumulation
SUBIThis input is loaded into the control register on the rising edge of (CLKX + CLKY).
This input is active only when ACC = 1. When SUB = 1 the contents of the output register are subtracted
from the result and stored back in the output register. When SUB = 0 the contents of the output register
are added to the result and stored back in the output register
TCIThis input is loaded into the control register on the rising edge of (CLKX + CLKY).
When TC = 1, the X and Y input are assumed to be in two's complement form. When TC = 0, X and Y
inputs are assumed to be in unsigned magnitude form
RNDIThis input is loaded into the control register on the rising edge of (CLKX + CLKY).
RND is inactive when low. RND = 1, adds a "1" to the most significant bit of the LSP, to round MSP and
XTP data
2577 tbl 01
11.23
Page 4
IDT7210L
16 x 16 PARALLEL CMOS MULTIPLIER-ACCUMULATORMILITARY AND COMMERCIAL TEMPERATURE RANGES
PRELOAD TRUTH TABLE
PRELTSXTSMTSLXTPMSPLSP
0000QQQ
0001QQHi Z
0010QHi ZQ
0011QHi ZHi Z
0100Hi ZQQ
0101Hi ZQHi Z
0110Hi ZHi ZQ
0111Hi ZHi ZHi Z
1000Hi ZHi ZHi Z
1001Hi ZHi ZPL
1010Hi ZPLHi Z
1011Hi ZPLPL
1100PLHi ZHi Z
1101PLHi ZPL
1110PLPLHi Z
1111PLPLPL
NOTES:2577 tbl 02
Hi Z = Output buffers at high impedance (output disabled)
Q = Output buffers at low impedance. Contents of output register will be
transferred to output pins.
PL = Output buffers at high impedance or output disabled. Preload data
supplied externally at output pins will be loaded into the output
register at the rising edge of CLKP.
NOTES ON TWO'S COMPLEMENT FORMATS
1. In two's complement notation, the location of the binary
point that signifies the separation of the fractional and
integer fileds is just after the sign, between the sign bit
(-2°) and the next significant bit for the multiplier inputs.
This same format is carried over to the output format,
except that the extended significance of the integer filed is
provided to extend the utility of the accumulator. In the
case of the output rotation, the output binary point is
located between the2° and 2
the binary point is arbitrary, as long as there is consistency
with both the input and output formats. The number filed
can be considered entirely integer with the binary point just
to the right of the least significant bit for the input, product
and the accumulated sum.
2. When in the non-accumulating mode, the first four bits (P
to P31) will all indicate the sign of the product. Additionally,
the P30 term will also indicate the sign with one exception,
when multiplying -1 x -1. With the additional bits that are
available in this multiplier, the –1 x –1 is a valid operation
that yields a +1 product.
3. In operations that require the accumulation of single products or sum of products, there is no change in format. To
allow for a valid summation beyond that available for a
single multiplication product, three additional significant
bits (guard bits) are provided. This is the same as if the
product was accumulated off-chip in a separate 35-bit wide
adder. Taking the sign at the most significant bit position
will guarantee that the largest number field will be used.
When the accumulated sum only occupies the right hand
portion of the accumulator, the sign will be extended into
the lesser significant bit positions.
1
bit positions. The location of
34
ABSOLUTE MAXIMUM RATINGS
SymbolRatingCommercialMilitaryUnit
VCCPower Supply
Voltage
VTERMTerminal Voltage
with Respect to
GND
TAOperating
Temperature
TBIASTemperature
Under Bias
TSTGStorage
Temperature
IOUTDC Output
Current
NOTE:2577 tbl 03
1. Stresses greater than those listed under ABSOLUTE MAXIMUM
RATINGS may cause permanent damage to the device. This is a stress
rating only and functional operation of the device at these or any other
conditions above those indicated in the operational sections of this
specification is not implied. Exposure to absolute maximum rating conditions for extended periods may affect reliability.
Operating Power Supply CurrentVCC = Max., Outputs Enabled
f= 10MHz
(2)
—4590—45110mA
CL = 50 pF
ICCQ1Quiescent Power Supply CurrentVIN≥ VIH, VIN≤ VIL—2030—2030mA
ICCQ2Quiescent Power Supply CurrentVIN≥ VCC –0.2V, V IN≤ 0.2V—410—412mA
(2,3)
ICC/f
Increase in Power Supply
VCC = Max., Outputs Disabled——6——8mA/
Current MHz
NOTES:
1. Typical implies VCC = 5V and TA = +25°C.
2. I
CC is measured at 10MHz and VIN = 0 to 3V. For frequencies greater than 10MHz, the following equation is used for the commercial range:
CC = 90+ 6(f –10)mA, where f = operating frequency in MHz. For the military range, ICC = 110 + 8(f –10). f = operating frequency in MHz, f = 1/tMA.
I
3. For frequencies greater than 10MHz, guaranteed by design, not production tested.
OL = 4mA for tMA > 55ns.
4. I
5. For conditions shown as Max. or Min., use appropriate value specified under electrical characteristics.
(1)
Max.Min. Typ.
(1)
Max.
Unit
MHz
2577 tbl 05
AC ELECTRICAL CHARACTERISTICS COMMERCIAL(VCC = 5V ± 10%, TA = 0° to +70°C)
7210L207210L257210L357210L457210L557210L65
SymbolParameterMin. Max. Min. Max. Min. Max. Min. Max. Min. Max. Min. Max. Unit
(1)
(2)
2.0202.0252.0352.0452.0552.065ns
2.0182.0202.0252.0252.0302.035ns
—18–20–25–25–30–30ns
t
MA
t
D
t
ENA
t
DIS
t
S
t
H
t
PW
t
HCL
NOTES: 2577 tbl 06
1. Transition is measured ±500mV from steady state voltage.
2. Minimum delays guaranteed but not tested
Multiply-Accumulate Time
Output Delay
(2)
3-State Enable Time—18–20–25–25–30–30ns
3-State Disable Time
Input Register Set-up Time10—12–12–15–20–25–ns
Input Register Hold Time3—3–3–3–3–3–ns
Clock Pulse Width9—10–10–15–20–25–ns
Relative Hold Time0—0–0–0–0–0–ns
AC ELECTRICAL CHARACTERISTICS MILITARY (VCC = 5V ± 10%, TA = –55° to +125°C)
7210L257210L307210L407210L557210L657210L75
SymbolParameterMin. Max. Min. Max. Min. Max. Min. Max. Min. Max. Min. Max. Unit
(1)
(2)
2.0252.0302.0402.0552.0652.075ns
2.0202.0202.0252.0302.0352.035ns
—20–20–25–25–30–30ns
t
MA
t
D
t
ENA
t
DIS
t
S
t
H
t
PW
t
HCL
NOTES:2577 tbl 07
1. Transition is measured ±500mV from steady state voltage.
2. Minimum delays guaranteed but not tested
Multiply-Accumulate Time
Output Delay
(2)
3-State Enable Time—20–20–25–30–30–35ns
3-State Disable Time
Input Register Set-up Time12—12–15–20–25–25–ns
Input Register Hold Time3—3–3–3–3–3–ns
Clock Pulse Width10—10–15–20–25–25–ns
Relative Hold Time0—0–0–0–0–0–ns
11.25
Page 6
IDT7210L
16 x 16 PARALLEL CMOS MULTIPLIER-ACCUMULATORMILITARY AND COMMERCIAL TEMPERATURE RANGES
INPUT
INPUT
CLOCK
OUTPUT
CLOCK
PRELOAD
THREE-STATE
CONTROL
OUTPUT
CONTROL AND
DATA
IN
tStH
tPW
tDIStENA
HIGH IMPEDANCE
tHCL
tMA
tPW
tDIStStHtENA
DATAOUTPRELOAD IN DATADATAOUT
tD
Figure 1. Timing Diagram
11.26
Page 7
IDT7210L
16 x 16 PARALLEL CMOS MULTIPLIER-ACCUMULATORMILITARY AND COMMERCIAL TEMPERATURE RANGES
DIGIT
VALUE
DIGIT
VALUE
–16
2
–15
2
–14
2
–13
2
–12
2
–11
2
–10
2
–9
2–82
–7
2
–6
2
–5
2
–4
2
–3
2
–2
2
–1
2
SIGNAL
0
–32
P
1
–31
P
2
–30
P
3
–29
P
4
–28
P
5
–27
P
6
–26
P
7
–25
P
8
–24
P
9
–23
P
10
–22
P
11
–21
P
12
–20
P
13
–19
P
14
–18
P
15
–17
P
16
–16
P
17
–15
P
18
–14
P
19
–13
P
20
–12
P
21
–11
P
22
–10
P
23
–9
P
24
–8
P
25
–7
P
26
–6
P
27
–5
P
28
–4
P
29
–3
P
30
–2
P
31
–1
P
32
0
P
33
1
P
34
2
P
=
2
2
2577 drw 11
2
2
2
2
2
2
LSP
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
MSP
Figure 3. Fractional Unsigned Mgnitude Notation
2
2
2
2
2
2
2
2
2
2
XTP
2
SIGNAL
0
X
1
X
2
X
3
X
4
X
5
X
6
X
7
X
8
X
9
X
10
X
11
X
12
X
13
X
14
X
15
X
BINARY POINT
DIGIT
VALUE
–15
2
–14
2
–13
2
–12
2
–11
2
–10
2
–9
2
–8
2
–7
2
–6
2
–5
2
–4
2
–3
2
–2
2
–1
2
0
–2
DIGIT
SIGNAL
0
–15
Y
1
–14
Y
2
–13
Y
3
–12
Y
4
–11
Y
5
–10
Y
6
–9
Y
7
–8
Y
8
–7
Y
9
–6
Y
10
–5
Y
11
–4
Y
12
–3
Y
13
–2
Y
14
–1
Y
0
15
Y
X
VALUE
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
–2
DIGIT
SIGNAL
0
–30
P
1
–29
P
2
–28
P
3
–27
P
4
–26
P
5
–25
P
6
–24
P
7
–23
P
8
–22
P
9
–21
P
10
–20
P
11
–19
P
12
–18
P
13
–17
P
14
–16
P
15
–15
P
16
–14
P
17
–13
P
18
–12
P
19
–11
P
20
–10
P
21
–9
P
22
–8
P
23
–7
P
24
–6
P
25
–5
P
26
–4
P
27
–3
P
28
–2
P
29
–1
P
30
0
P
31
1
P
32
2
P
33
3
P
4
34
P
=
VALUE
2
2
2
2
2
2
2
2
LSP
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
XTPMSP
–2
2577 drw 10
SIGNAL
1 X0
X
3 X2
X
5 X4
X
Figure 2. Fractional Two’s Complement Notation.
7 X6
X
9 X8
X
11 X 10
X
13 X 12
X
15 X14
X
BINARY POINT
DIGIT
VALUE
–16
2
–15
2
–14
2
–13
2
–12
2
–11
2
–10
2
–9
2–82
–7
2
–6
2
–5
2
–4
2
–3
2
–2
2
–1
2
SIGNAL
0
Y
1
Y
2
Y
3
Y
4
Y
5
Y
6
Y
7
Y
8
Y
9
Y
10
Y
11
Y
12
Y
13
Y
14
Y
15
Y
X
11.27
Page 8
IDT7210L
16 x 16 PARALLEL CMOS MULTIPLIER-ACCUMULATORMILITARY AND COMMERCIAL TEMPERATURE RANGES
SIGNAL
0
X
BINARY POINT
1
X
2
X
3
X
4
X
5
X
6
X
7
X
8
X
9
X
10
X
11
X
12
X
13
X
14
X
15
X
DIGIT
VALUE
0
2
1
2
2
2
3
2
4
2
5
2
6
2
7
2
8
2
9
2
10
2
11
2
12
2
13
2
14
2
15
–2
SIGNAL
DIGIT
0
0
Y
1
1
Y
2
2
Y
3
3
Y
4
4
Y
5
5
Y
6
6
Y
7
7
Y
8
8
Y
9
9
Y
10
10
Y
11
11
Y
12
12
Y
13
13
Y
14
14
Y
15
15
Y
X
VALUE
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
–2
SIGNAL
DIGIT
0
0
P
1
1
P
2
2
P
3
3
P
4
4
P
5
5
P
6
6
P
7
7
P
8
8
P
9
9
P
10
10
P
11
11
P
12
12
P
13
13
P
14
14
P
15
15
P
16
16
P
17
17
P
18
18
P
19
19
P
20
20
P
21
21
P
22
22
P
23
23
P
24
24
P
25
25
P
26
26
P
27
27
P
28
28
P
29
29
P
30
30
P
31
31
P
32
32
P
33
33
P
34
34
P
=
VALUE
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
XTPMSPLSP
–2
2577 drw 12
BINARY POINT
Figure 4. Integer Two's Complement Notation
SIGNAL
DIGIT
0
0
X
1
1
X
2
2
X
3
3
X
4
4
X
5
5
X
6
6
X
7
7
X
8
8
X
9
9
X
10
10
X
11
11
X
12
12
X
13
13
X
14
14
X
15
15
X
VALUE
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
SIGNAL
DIGIT
0
0
Y
1
1
Y
2
2
Y
3
3
Y
4
4
Y
5
5
Y
6
6
Y
7
7
Y
8
8
Y
9
9
Y
10
10
Y
11
11
Y
12
12
Y
13
13
Y
14
14
Y
15
15
Y
X
VALUE
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
DIGIT
SIGNAL
0
0
P
1
1
P
2
2
P
3
3
P
4
4
P
5
5
P
6
6
P
7
7
P
8
8
P
9
9
P
10
10
P
11
11
P
12
12
P
13
13
P
14
14
P
15
15
P
16
16
P
17
17
P
18
18
P
19
19
P
20
20
P
21
21
P
22
22
P
23
23
P
24
24
P
25
25
P
26
26
P
27
27
P
28
28
P
29
29
P
30
30
P
31
31
P
32
32
P
33
33
P
34
34
P
=
VALUE
2
2
2
2
2
2
2
2
LSP
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
XTPMSP
2
2577 drw 13
Figure 5. Integer Unsigned Magnitude Notation
11.28
Page 9
IDT7210L
16 x 16 PARALLEL CMOS MULTIPLIER-ACCUMULATORMILITARY AND COMMERCIAL TEMPERATURE RANGES
TEST CIRCUITS AND WAVEFORMS
TEST CIRCUITS FOR ALL OUTPUTS
V
CC
V
OUT
50pF
C L
Pulse
Generator
V
IN
D.U.T.
T
R
7.0V
500Ω
500Ω
2577 drw 06
SWITCH POSITION
Test
Open Drain
Disable Low
Enable Low
All Other Tests
DEFINITIONS:
L= Load capacitance: includes jig and probe capacitance.
C
R
T = Termination resistance: should be equal to ZOUT of the Pulse
Generator.
Switch
Closed
Open
2577 lnk 09
SET-UP, HOLD AND RELEASE TIMES
DATA
INPUT
TIMING
tSU
tH
INPUT
ASYNCHRONOUS CONTROL
PRESET
tREM
CLEAR
ETC.
SYNCHRONOUS CONTROL
PRESET
CLOCK ENABLE
CLEAR
tSU
tH
ETC.
PROPAGATION DELAY
SAME PHASE
INPUT TRANSITION
tPLH
OUTPUT
tPLHtPHL
OPPOSITE PHASE
INPUT TRANSITION
tPHL
3V
1.5V
0V
3V
1.5V
0V
3V
1.5V
0V
3V
1.5V
0V
2577 drw 07
3V
1.5V
0V
V
OH
1.5V
VOL
3V
1.5V
0V
2577 drw 09
PULSE WIDTH
LOW-HIGH-LOW
PULSE
tW
HIGH-LOW-HIGH
PULSE
1.5V
1.5V
2577 drw 08
ENABLE AND DISABLE TIMES
ENABLEDISABLE
3V
CONTROL
INPUT
PLZtPZL
t
OUTPUT
NORMALLY
LOW
SWITCH
CLOSED
3.5V
1.5V
tPZHtPHZ
OUTPUT
NORMALLY
HIGH
NOTES:
1. Diagram shown for input Control Enable-LOW and input Control DisableHIGH
2. Pulse Generator for All Pulses: Rate ≤ 1.0MHz; t
SWITCH
OPEN
1.5V
0V
F≤ 2.5ns; tR≤ 2.5ns
0.3V
0.3V
1.5V
0V
3.5V
VOL
VOH
0V
2577 drw 10
11.29
Page 10
IDT7210L
16 x 16 PARALLEL CMOS MULTIPLIER-ACCUMULATORMILITARY AND COMMERCIAL TEMPERATURE RANGES