Transistor count

Plot of transistor counts against dates of introduction. The curve shows counts doubling every two years.
A circuit board with a 4×4 array of SyNAPSE-developed chips. Built on 28 nm process technology, each chip using 5.4 billion transistors.

The transistor count is the number of transistors on an integrated circuit chip. Transistor count is the most common measure of semiconductor integrated circuit complexity. As of 2015, the highest transistor count in a commercially available CPU (in one chip) is over 5.5 billion transistors, in Intel's 18-core Xeon Haswell-EP.

According to Moore's law the transistor count of the integrated circuits doubles approximately every two years. On most modern microprocessors, the majority of transistors are contained in caches. In fact, IBM's Storage Controller at 7.1 billion transistors with 480 MB L4 cache in 2015, is bigger than any microprocessor before, on both counts; it is however not a CPU until combined with one (or more) of its companion IBM z13 microprocessor chip. That chip holds the rest of the cache, 64 MB L3 (and the other smaller levels) and processor logic, and while having the same dimensions has only 3.9 billion transistors (near record breaking for a microprocessor on its own). To date, combined, this mainframe is the biggest general purpose processor when only one Storage Controller is used, let alone when two are used.

On August 7, 2014, IBM announced their neuromorphic TrueNorth chip; their second generation chip backed by the SyNAPSE program, which, with 5.4 billion transistors (in 4096 cores), has more transistors than any chip IBM has ever made[1] (until their Storage Controller, mentioned above) and is the "second largest (CMOS) chip in the world"[2][3][4] (and therefore the largest "neuromorphic" one, as a GPU chip is bigger, but sixteen can be connected together).

Xilinx currently holds the "world-record" for a FPGA containing more than 20 billion transistors.

Microprocessors

A microprocessor incorporates the functions of a computer's central processing unit on a single integrated circuit. It is a multipurpose, programmable device that accepts digital data as input, processes it according to instructions stored in its memory, and provides results as output.

Processor Transistor count Date of introduction Designer Process Area
TMS 1000 8,000 1971 Texas Instruments 8,000 nm
Intel 4004 2,300 1971 Intel 10,000 nm 12 mm²
Intel 8008 3,500 1972 Intel 10,000 nm 14 mm²
MOS Technology 6502 3,510[5] 1975 MOS Technology 8,000 nm 21 mm²
Motorola 6800 4,100 1974 Motorola 6,000 nm 16 mm²
Intel 8080 4,500 1974 Intel 6,000 nm 20 mm²
RCA 1802 5,000 1974 RCA 5,000 nm 27 mm²
Intel 8085 6,500 1976 Intel 3,000 nm 20 mm²
Zilog Z80 8,500 1976 Zilog 4,000 nm 18 mm²
Motorola 6809 9,000 1978 Motorola 5,000 nm 21 mm²
Intel 8086 29,000 1978 Intel 3,000 nm 33 mm²
Intel 8088 29,000 1979 Intel 3,000 nm 33 mm²
WDC 65C02 11,500[6] 1981 WDC 3,000 nm 6 mm²
Intel 80186 55,000 1982 Intel 3,000 nm 60 mm²
Motorola 68000 68,000 1979 Motorola 3,500 nm 44 mm²
Intel 80286 134,000 1982 Intel 1,500 nm 49 mm²
WDC 65C816 22,000[7] 1983 WDC 9 mm²
Motorola 68020 190,000[8] 1984 Motorola 2,000 nm 85 mm²
Intel 80386 275,000 1985 Intel 1,500 nm 104 mm²
ARM 1 25,000[8] 1985 Acorn 3,000 nm 50 mm²
Novix NC4016 16,000[9] 1985[10] Harris Corporation 3,000 nm[11]
ARM 2 30,000[8] 1986 Acorn 2,000 nm 30 mm²
TI Explorer's 32-bit Lisp machine chip 553,000[12] 1987 Texas Instruments
DEC WRL MultiTitan 180,000[13] 1988 DEC WRL 1,500 nm 61 mm²
Intel i960 250,000[14] 1988 Intel 600 nm
Intel 80486 1,180,235 1989 Intel 1000 nm 173 mm²
ARM 3 300,000 1989 Acorn
R4000 1,350,000 1991 MIPS 1,000 nm 213 mm²
ARM 6 35,000 1991 ARM
Pentium 3,100,000 1993 Intel 800 nm 294 mm²
ARM700 578,977[15] 1994 ARM 68.51 mm²
SA-110 2,500,000[8] 1995 Acorn/DEC/Apple 350 nm 50 mm²
ARM 9TDMI 111,000[8] 1999 Acorn 350 nm 4.8 mm²
Pentium Pro 5,500,000[16] 1995 Intel 500 nm 307 mm²
AMD K5 4,300,000 1996 AMD 500 nm 251 mm²
Pentium II Klamath 7,500,000 1997 Intel 350 nm 195 mm²
Pentium II Deschutes 7,500,000 1998 Intel 250 nm 113 mm²
AMD K6 8,800,000 1997 AMD 350 nm 162 mm²
Pentium III Katmai 9,500,000 1999 Intel 250 nm 128 mm²
Pentium III Coppermine 21,000,000 2000 Intel 180 nm 80 mm²
Pentium II Mobile Dixon 27,400,000 1999 Intel 180 nm 180 mm²
Pentium III Tualatin 45,000,000 2001 Intel 130 nm 81 mm²
AMD K6-III 21,300,000 1999 AMD 250 nm 118 mm²
AMD K7 22,000,000 1999 AMD 250 nm 184 mm²
Pentium 4 Willamette 42,000,000 2000 Intel 180 nm 217 mm²
Pentium 4 Northwood 55,000,000 2002 Intel 130 nm 145 mm²
Pentium 4 Prescott 112,000,000 2004 Intel 90 nm 110 mm²
Pentium 4 Prescott-2M 169,000,000 2005 Intel 90 nm 143 mm²
Pentium 4 Cedar Mill 184,000,000 2006 Intel 65 nm 90 mm²
Pentium D Smithfield 228,000,000 2005 Intel 90 nm 206 mm²
Pentium D Presler 362,000,000 2006 Intel 65 nm 162 mm²
Atom 47,000,000 2008 Intel 45 nm 24 mm²
Barton 54,300,000 2003 AMD 130 nm 101 mm²
AMD K8 105,900,000 2003 AMD 130 nm 193 mm²
Itanium 2 McKinley 220,000,000 2002 Intel 180 nm 421 mm²
Cell 241,000,000 2006 Sony/IBM/Toshiba 90 nm 221 mm²
Core 2 Duo Conroe 291,000,000 2006 Intel 65 nm 143 mm²
Core 2 Duo Allendale 169,000,000 2007 Intel 65 nm 111 mm²
Itanium 2 Madison 6M 410,000,000 2003 Intel 130 nm 374 mm²
AMD K10 quad-core 2M L3 463,000,000[17] 2007 AMD 65 nm 283 mm²
ARM Cortex-A9 26,000,000[18] 2007 ARM 45 nm 31 mm²
Core 2 Duo Wolfdale 3M 230,000,000 2008 Intel 45 nm 83 mm²
Itanium 2 with 9 MB cache 592,000,000 2004 Intel 130 nm 432 mm²
Core 2 Duo Wolfdale 411,000,000 2007 Intel 45 nm 107 mm²
Core i7 (Quad) 731,000,000 2008 Intel 45 nm 263 mm²
AMD K10 quad-core 6M L3 758,000,000[17] 2008 AMD 45 nm 258 mm²
POWER6 789,000,000 2007 IBM 65 nm 341 mm²
Six-core Opteron 2400 904,000,000 2009 AMD 45 nm 346 mm²
16-core SPARC T3 1,000,000,000[19] 2010 Sun/Oracle 40 nm 377 mm²
Apple A7 (dual-core ARM64 "mobile SoC") 1,000,000,000 2013 Apple 28 nm 102 mm²
Quad-core + GPU Core i7 1,160,000,000 2011 Intel 32 nm 216 mm²
Six-core Core i7 (Gulftown) 1,170,000,000 2010 Intel 32 nm 240 mm²
8-core POWER7 32M L3 1,200,000,000 2010 IBM 45 nm 567 mm²
8-core AMD Bulldozer 1,200,000,000[20] 2012 AMD 32 nm 315 mm²
Quad-core + GPU AMD Trinity 1,303,000,000 2012 AMD 32 nm 246 mm²
Quad-core z196[21] 1,400,000,000 2010 IBM 45 nm 512 mm²
Quad-core + GPU Core i7 Ivy Bridge 1,400,000,000 2012 Intel 22 nm 160 mm²
Quad-core + GPU Core i7 Haswell 1,400,000,000[22] 2014 Intel 22 nm 177 mm²
Dual-core Itanium 2 1,700,000,000[23] 2006 Intel 90 nm 596 mm²
Quad-core + GPU GT2 Core i7 Skylake K cca 1,750,000,000 2015 Intel 14 nm 122 mm²
Six-core Core i7 Ivy Bridge E 1,860,000,000 2013 Intel 22 nm 256 mm²
Duo-core + GPU Iris Core i7 Broadwell-U 1,900,000,000[24] 2015 Intel 14 nm 133 mm²
Six-core Xeon 7400 1,900,000,000 2008 Intel 45 nm 503 mm²
Quad-core Itanium Tukwila 2,000,000,000[25] 2010 Intel 65 nm 699 mm²
Apple A8 (dual-core ARM64 "mobile SoC") 2,000,000,000 2014 Apple 20 nm 89 mm²
8-core POWER7+ 80 MB L3 cache 2,100,000,000 2012 IBM 32 nm 567 mm²
Six-core Core i7/8-core Xeon E5
(Sandy Bridge-E/EP) 		2,270,000,000[26] 2011 Intel 32 nm 434 mm²
8-core Xeon Nehalem-EX 2,300,000,000[27] 2010 Intel 45 nm 684 mm²
8-core Core i7 Haswell-E 2,600,000,000[28] 2014 Intel 22 nm 355 mm²
10-core Xeon Westmere-EX 2,600,000,000 2011 Intel 32 nm 512 mm²
Six-core zEC12 2,750,000,000 2012 IBM 32 nm 597 mm²
Apple A8X (tri-core ARM64 "mobile SoC") 3,000,000,000[29] 2014 Apple 20 nm 128 mm²
8-core Itanium Poulson 3,100,000,000 2012 Intel 32 nm 544 mm²
IBM z13 3,990,000,000 2015 IBM 22 nm 678 mm²
12-core POWER8 4,200,000,000 2013 IBM 22 nm 650 mm²
15-core Xeon Ivy Bridge-EX 4,310,000,000[30] 2014 Intel 22 nm 541 mm²
61-core Xeon Phi 5,000,000,000[31] 2012 Intel 22 nm 350 mm²
Xbox One main SoC 5,000,000,000 2013 Microsoft/AMD 28 nm 363 mm²
18-core Xeon Haswell-E5 5,560,000,000[32] 2014 Intel 22 nm 661 mm²
IBM z13 Storage Controller 7,100,000,000 2015 IBM 22 nm 678 mm²
22-core Xeon Broadwell-E5 ~7,200,000,000[33] 2016 Intel 14 nm 456 mm²
SPARC M7 10,000,000,000[34] 2015 Oracle 20 nm

Transistorized computers

The "second generation" of computers (transistor computers) featured boards filled with discrete transistors and magnetic memory cores.

Processor Transistor count Date of introduction Manufacturer Process Area
"Transistor Computer" (full size) 200 discrete point-contact transistors 1955 University of Manchester ? ?
"Metrovick 950" 200 discrete junction transistors 1956 Metropolitan-Vickers ? ?
PDP-1 2,700 discrete transistors 1959 Digital Equipment Corporation
PDP-8/s ? discrete transistors (519 logic gates) 1971 ? Digital Equipment Corporation ? ?
M18 FADAC 1,600 discrete transistors 1960 Autonetics ? ?
D-17B 1,521 discrete transistors 1962 Autonetics ? ?
Apollo Guidance Computer 12,300 (4,100 ICs, each containing a single 3-transistor 3-input NOR gate) 1966 Raytheon / MIT Instrumentation Laboratory ? ?

GPUs

A graphics processing unit (GPU) is a specialized electronic circuit designed to rapidly manipulate and alter memory to accelerate the building of images in a frame buffer intended for output to a display.

Processor Transistor count Date of introduction Manufacturer Process Area
NV3 3,500,000 1997 NVIDIA 350 nm 90 mm²
Rage 128 8,000,000 1999 AMD 250 nm 70 mm²
NV5 15,000,000 1999 Nvidia 250 nm
NV10 23,000,000 1999 Nvidia 220 nm 111 mm²
NV11 20,000,000 2000 Nvidia 180 nm 65 mm²
NV15 25,000,000 2000 Nvidia 180 nm 81 mm²
R100 30,000,000 2000 AMD 180 nm 97 mm²
NV20 57,000,000 2001 Nvidia 150 nm 128 mm²
R200 60,000,000 2001 AMD 150 nm 68 mm²
NV25 63,000,000 2002 Nvidia 150 nm 142 mm²
R300 107,000,000 2002 AMD 150 nm 218 mm²
R360 117,000,000 2003 AMD 150 nm 218 mm²
NV38 135,000,000 2003 Nvidia 130 nm 207 mm²
R480 160,000,000 2004 AMD 130 nm 297 mm²
NV40 222,000,000 2004 Nvidia 130 nm 305 mm²
G70 303,000,000 2005 Nvidia 110 nm 333 mm²
R520 321,000,000 2005 AMD 90 nm 288 mm²
R580 384,000,000 2006 AMD 90 nm 352 mm²
G80 681,000,000 2006 Nvidia 90 nm 480 mm²
R600 700,000,000 2007 AMD 80 nm 420 mm²
G92 754,000,000 2007 Nvidia 65 nm 324 mm²
RV790 959,000,000[35] 2008 AMD 55 nm 282 mm²
GT200 Tesla 1,400,000,000[36] 2008 Nvidia 65 nm 576 mm²
Cypress RV870 2,154,000,000[37] 2009 AMD 40 nm 334 mm²
Cayman RV970 2,640,000,000 2010 AMD 40 nm 389 mm²
GF100 Fermi 3,200,000,000[38] Mar 2010 Nvidia 40 nm 526 mm²
GF110 Fermi 3,000,000,000[38] Nov 2010 Nvidia 40 nm 520 mm²
GK104 Kepler 3,540,000,000[39] 2012 Nvidia 28 nm 294 mm²
Tahiti 4,312,711,873[40] 2011 AMD 28 nm 365 mm²
GK110 Kepler 7,080,000,000[41] 2012[42] Nvidia 28 nm 561 mm²
Hawaii 6,300,000,000 2013 AMD 28 nm 438 mm²
GM204 Maxwell 5,200,000,000 2014 Nvidia 28 nm 398 mm²
GM200 Maxwell 8,100,000,000 2015 Nvidia 28 nm 601 mm²
Fiji 8,900,000,000 2015 AMD 28 nm 596 mm²
GP100 Pascal 15,300,000,000[43] 2016 Nvidia 16 nm 610 mm²

FPGA

A field-programmable gate array (FPGA) is an integrated circuit designed to be configured by a customer or a designer after manufacturing.

FPGA Transistor count Date of introduction Manufacturer Process Area
Virtex ~70,000,000 1997 Xilinx
Virtex-E ~200,000,000 1998 Xilinx
Virtex-II ~350,000,000 2000 Xilinx 130 nm
Virtex-II PRO ~430,000,000 2002 Xilinx
Virtex-4 1,000,000,000 2004 Xilinx 90 nm
Virtex-5 1,100,000,000[44] 2006 Xilinx 65 nm
Stratix IV 2,500,000,000[45] 2008 Altera 40 nm
Stratix V 3,800,000,000[46] 2011 Altera 28 nm
Virtex-7 6,800,000,000[47] 2011 Xilinx 28 nm
Virtex-Ultrascale XCVU440 20,000,000,000+[48] 2014 Xilinx 20 nm

Logic functions

Transistor count for generic logic functions is based on static CMOS implementation.[49]

Function Transistor count
NOT 2
Buffer 4
NAND 2-input 4
NOR 2-input 4
AND 2-input 6
OR 2-input 6
NAND 3-input 6
NOR 3-input 6
XOR 2-input 6
XNOR 2-input 8
MUX 2-input with TG 6
MUX 4-input with TG 18
NOT MUX 2-input 8
MUX 4-input 24
1-bit Adder full 28
1-bit Adder–subtractor 48
AND-OR-INVERT [50] 6
Latch, D gated 8
Flip-flop, edge triggered dynamic D with reset 12
8-bit multiplier[51] 3,000
16-bit multiplier[51] 9,000
32-bit multiplier[52] 21,000
small-scale integration 2–100[53]
medium-scale integration 100–500[53]
large-scale integration 500–20,000[53]
very-large-scale integration 20,000–1,000,000[53]
ultra-large scale integration >1,000,000

Memory

Semiconductor memory is an electronic data storage device, often used as computer memory, implemented on integrated circuit.

We know that in order to store a single bit (which may be 1 or 0), one flip-flop is required, made of around eight transistors. Typical CMOS Static random-access memory (SRAM) consists of 6 transistors. For Dynamic random-access memory (DRAM), 1T1C, which means one transistor and one capacitor structure is common. Capacitor charged or not is used to store 1 or 0. For flash memory, the data is stored in floating gate, and the resistance of the transistor is sensed to interpret the data stored. Depending on how fine scale the resistance could be separated, one transistor could store up to 3-bits, meaning eight distinctive level of resistance possible per transistor. However, the fine the scale comes with cost of repeatability therefore reliability. Typically, low grade 2-bits MLC flash is used for flash drive, so a 16 GB flash drive contains roughly 64 billion transistors.

Chip Capacity & type Transistor count Date of introduction Manufacturer Process Area
? 256-bit ROM bipolar TTL ? 1965 Sylvania ? ?
? 1024-bit ROM MOS ? 1965 General Microelectronics ? ?
SP95 16-bit SRAM bipolar ? 1965 IBM ? ?
? 128-bit RAM ? 1969 IBM ? ?
512-bit PROM bipolar TTL ? 1970 Radiation Inc. ? ?
93400 256-bit RAM ? 1970 Fairchild ? ?
1103[54][55] 1 kb DRAM ?1,024 1970 Intel ? ?
1702 Erasable PROM 2 kb EPROM ? 1971 Intel ? ?
? 8 Mb DRAM ?8,388,608 January 6, 1984 (1986) Hitachi ? ?
?[56][57] 64 Mb DRAM ?67,108,864 1994 NEC, Samsung 320 nm ?
?[58][59] 256 Mb DRAM ?268,435,456 June 12, 1995 IBM, SIEMENS AG, Toshiba Corp. 250 nm 286 mm²
?[60] 1 Gb DRAM ?1,073,741,824 January 9, 1995 (2001) Hitachi ? ?
? 64 Gb DRAM ?68,719,476,736 2007(9)? ? ? ?
?[61] 128 Gb DRAM ?137,438,953,472 July 5, 2012 Samsung 30 nm ?

Parallel systems

Historically, each processing element in earlier parallel systems—like all CPUs of that time—was a serial computer built out of multiple chips. As transistor counts per chip increases, each processing element could be built out of fewer chips, and then later each multi-core processor chip could contain more processing elements.[62]

Goodyear MPP: (1983?) 8 pixel processors per chip, 3,000 to 8,000 transistors per chip.[62]

Brunel University Scape (single-chip array-processing element): (1983) 256 pixel processors per chip, 120,000 to 140,000 transistors per chip.[62]

Cell Broadband Engine: (2006) 9 cores per chip, 234 million transistors per chip.[63]

References

  1. ↑ Simonite, Tom (August 7, 2014). "IBM Chip Processes Data Similar to the Way Your Brain Does".
  2. ↑ Anthony, Sebastian (August 7, 2014). "IBM cracks open a new era of computing with brain-like chip: 4096 cores, 1 million neurons, 5.4 billion transistors". Retrieved January 12, 2014.
  3. ↑ http://www.extremetech.com/extreme/187612-ibm-cracks-open-a-new-era-of-computing-with-brain-like-chip-4096-cores-1-million-neurons-5-4-billion-transistors
  4. ↑ "SyNAPSE program develops advanced brain-inspired chip". August 7, 2014.
  5. ↑ "The MOS 6502 and the Best Layout Guy in the World". swtch.com. Retrieved 2014-08-09.
  6. ↑ Microprocessors: 1971 to 1976 Christiansen
  7. ↑ "Microprocessors 1976 to 1981". weber.edu. Retrieved 2014-08-09.
  8. 1 2 3 4 5 Demone, Paul (2000-11-09). "ARM's Race to World Domination". real world technologies. Retrieved 2015-07-20.
  9. ↑ Hand, Tom. "The Harris RTX 2000 Microcontroller" (PDF). mpeforth.com. Retrieved 2014-08-09.
  10. ↑ "Forth chips list". UltraTechnology. 2001-03-15. Retrieved 2014-08-09.
  11. ↑ Koopman, Philip J. (1989). "4.4 Architecture of the Novix NC4016". Stack Computers: the new wave. Ellis Horwood Series in Computers and Their Applications. Carnegie Mellon University. ISBN 0745804187. Retrieved 2014-08-09.
  12. ↑ Bosshart, P.; Hewes, C.; Mi-Chang Chang; Kwok-Kit Chau; Hoac, C.; Houston, T.; Kalyan, V.; Lusky, S.; Mahant-Shetti, S.; Matzke, D.; Ruparel, K.; Ching-Hao Shaw; Sridhar, T.; Stark, D. (October 1987). "A 553K-Transistor LISP Processor Chip". IEEE Journal of Solid-State Circuits. sc-22 (5): 202–3. doi:10.1109/ISSCC.1987.1157084.
  13. ↑ Jouppi, Norman P.; Tang, Jeffrey Y. F. (July 1989). "A 20-MIPS Sustained 32-bit CMOS Microprocessor with High Ratio of Sustained to Peak Performance" (PDF). Western Research Laboratory, Digital Equipment Corporation. p. i. WRL Research Report 89/11.
  14. ↑ "The CPU shack museum". CPUshack.com. 2005-05-15. Retrieved 2014-08-09.
  15. ↑ "ARM7 Statistics". Poppyfields.net. 1994-05-27. Retrieved 2014-08-09.
  16. ↑ "PC Guide Intel Pentium Pro ("P6")". PCGuide.com. 2001-04-17. Retrieved 2014-08-09.
  17. 1 2 Toepelt, Bert (2009-01-08). "AMD Phenom II X4: 45nm Benchmarked — The Phenom II And AMD's Dragon Platform". TomsHardware.com. Retrieved 2014-08-09.
  18. ↑ "ARM (Advanced RISC Machines) Processors". EngineersGarage.com. Retrieved 2014-08-09.
  19. ↑ Stokes, Jon (2010-02-10). "Sun's 1 billion-transistor, 16-core Niagara 3 processor". ArsTechnica.com. Retrieved 2014-08-09.
  20. ↑ "Intel's Atom Architecture: The Journey Begins". AnandTech. Retrieved April 4, 2010.
  21. ↑ "IBM to Ship World's Fastest Microprocessor". IBM. 2010-09-01. Retrieved 2014-08-09.
  22. ↑ Shimpi, Lal. "The Haswell Review: Intel Core i7-4770K & i5-4670K Tested". anandtech. Retrieved 20 November 2014.
  23. ↑ "PRESS KIT — Dual-core Intel Itanium Processor". Intel. Retrieved 2014-08-09.
  24. ↑ "Intel's Broadwell-U arrives aboard 15W, 28W mobile processors". TechReport. Retrieved 5 January 2015.
  25. ↑ "Itanium Tukwila." AFP. February 5, 2008. Retrieved on February 5, 2008.
  26. ↑ Angelini, Chris (2011-11-14). "Intel Core i7-3960X Review: Sandy Bridge-E And X79 Express". TomsHardware.com. Retrieved 2014-08-09.
  27. ↑ "Intel Previews Intel Xeon 'Nehalem-EX' Processor." May 26, 2009. Retrieved on May 28, 2009.
  28. ↑ "." August 29, 2014. Retrieved on August 29, 2014.
  29. ↑ "Apple A8X". NotebookCheck. Retrieved 2015-07-20.
  30. ↑ "Intel Readying 15-core Xeon E7 v2". AnandTech. Retrieved 2014-08-09.
  31. ↑ "Intel Xeon Phi SE10X". TechPowerUp. Retrieved 2015-07-20.
  32. ↑ "Intel Xeon E5-2600 v3 Processor Overview: Haswell-EP Up to 18 Cores". pcper. Retrieved 29 January 2015.
  33. ↑ "Broadwell-EP Architecture - Intel Xeon E5-2600 v4 Broadwell-EP Review". Tom's Hardware. 2016-03-31. Retrieved 2016-04-04.
  34. ↑ http://www.enterprisetech.com/2014/08/13/oracle-cranks-cores-32-sparc-m7-chip/
  35. ↑ "The Radeon HD 4850 & 4870: AMD Wins at $199 and $299". AnandTech.com. Retrieved 2014-08-09.
  36. ↑ "NVIDIA's 1.4 Billion Transistor GPU: GT200 Arrives as the GeForce GTX 280 & 260". AnandTech.com. Retrieved 2014-08-09.
  37. ↑ "Radeon 5870 specifications". AMD. Retrieved 2014-08-09.
  38. 1 2 Glaskowsky, Peter. "ATI and Nvidia face off-obliquely". CNET. Retrieved 2014-08-09.
  39. ↑ "Whitepaper: NVIDIA GeForce GTX 680" (PDF). NVIDIA. 2012.
  40. ↑ Woligroski, Don (2011-12-22). "AMD Radeon HD 7970". TomsHardware.com. Retrieved 2014-08-09.
  41. ↑ http://www.nvidia.com/content/PDF/kepler/NVIDIA-Kepler-GK110-Architecture-Whitepaper.pdf
  42. ↑ http://www.anandtech.com/show/6446/nvidia-launches-tesla-k20-k20x-gk110-arrives-at-last
  43. ↑ https://devblogs.nvidia.com/parallelforall/inside-pascal/
  44. ↑ "Taiwan Company UMC Delivers 65nm FPGAs to Xilinx." SDA-ASIA Thursday, November 9, 2006.
  45. ↑ ""Altera's new 40nm FPGAs — 2.5 billion transistors!". pldesignline.com.
  46. ↑ "Altera unveils 28-nm Stratix V FPGA family". April 20, 2010. Retrieved April 20, 2010.
  47. ↑ "Xilinx Announces world’s highest capacity FPGA." October 25, 2011. Retrieved on October 25, 2011.
  48. ↑ "http://www.xilinx.com/publications/archives/xcell/Xcell86.pdf" May 2014. Retrieved on June 3, 2014.
  49. ↑ Jan M. Rabaey, Digital Integrated Circuits, Fall 2001: Course Notes, Chapter 6: Designing Combinatorial Logic Gates in CMOS, retrieved October 27, 2012.
  50. ↑ Richard F. Tinder (January 2000). Engineering Digital Design. Academic Press. ISBN 978-0-12-691295-1.
  51. 1 2 Ghasemizadeh H., Fathi A., Ghasemizadeh A. (2012). "High Speed 16×16-bit Low-Latency Pipelined Booth Multiplier". Electrical and Electronic Engineering 2 (3): 121–7. doi:10.5923/j.eee.20120203.03. ISSN 2162-9455.
  52. ↑ Asadi, P. (2007). "Energy-efficient 32 × 32-bit multiplier in tunable near-zero threshold CMOS" (PDF). World Applied Sciences Journal 2 (4): 341–7. ISSN 1818-4952.
  53. 1 2 3 4 "IEEE Standard 100: The Authoritative Dictionary of IEEE Standards Terms" (7th ed.). 2000. doi:10.1109/IEEESTD.2000.322230. IEEE Std 100-2000.
  54. ↑ Intel: 35 Years of Innovation (1968–2003) Intel, 2003
  55. ↑ The DRAM memory of Robert Dennard history-computer.com
  56. ↑ NEC to build 64-Mbit DRAM line in U.S. (NEC Corp.; dynamic random access memory) Highbeam Business, October 24, 1994
  57. ↑ NEC, Samsung sampling 64-Mbit DRAMs Highbeam Business, April 17, 1995
  58. ↑ Alliance unwraps 256-Mbit DRAM. (IBM Corp, Toshiba Corp, Siemens AG) Highbeam Business, June 12, 1995
  59. ↑ International chip trio delivers memory jump. (Siemens AG, IBM Corp. and Toshiba Corp. plan to develop a 256-Mbit dynamic random access memory chip)(Tech Trends)(Brief Article) Highbeam Business, July 3, 1995
  60. ↑ Breaking the gigabit barrier, DRAMs at ISSCC portend major system-design impact. (dynamic random access memory; International Solid-State Circuits Conference; Hitachi Ltd. and NEC Corp. research and development) Highbeam Business, January 9, 1995
  61. ↑ Samsung announces 16GB DDR4 DIMM to be released in 2014 TweakTown, July 5, 2012
  62. 1 2 3 Smith, Kevin (August 11, 1983). "Image processor handles 256 pixels simultaneously". Electronics.
  63. ↑ "Cell chip: Hit or hype?" by Michael Kanellos 2005

External links

This article is issued from Wikipedia - version of the Sunday, May 01, 2016. The text is available under the Creative Commons Attribution/Share Alike but additional terms may apply for the media files.