The Ultimate GPU Benchmark (2009 - 2013)

Published: 2nd October 2022 (last update 2nd October 2022 )

The third part is ready. With new test platform and many interesting games. The range of video cards is quite wide this time, starting with Radeon HD 5000 / Geforce GTX 400 series and up to Radeon R9 290X and GeForce GTX 780 Ti.

Introduction

Here it is - the third part of the videocards benchmark project. The goal is to explore performance of the first and second generation of the DX11 GPUs. The multi-GPU technology in this era was greatly improved with frame packing - so of course there are several Crossfire and SLI setups.

The test system was upgraded to the X99 platform. Xeon E5 1650 v3 is equivalent of Core i7-5930k, including the unlocked multiplier. Performance should be slightly better than Sandy Bridge-E, despite the lower clock. Also Haswell-E doesn't suffer from any PCI-E 3.0 compatibility issues.

Test System - Hardware

Xeon E5-1650 v3 @ 4.5 GHz (6C/12T, fixed clock), NB @ 4 GHz
ASRock X99 Taichi
4 × 4 GB DDR4 3000 CL16-16-16-36-1T
Kingston HyperX Savage 120 GB
Toshiba 3 TB / 7200 rpm
Corsair RM1000i
Fractal Design Define S
Acer FX270HU (2560×1440)

Test System - OS and Drivers

Windows 7 x64
GeForce 377.83 (for Quadro 4000)
GeForce 391.35 (for other Nvidia Fermi GPUs)
GeForce 471.11 (for Nvidia Kepler GPUs)
Catalyst 15.11.1 (for Radeon HD 5000/6000 Series)
Adrenalin 19.12.1 (for Radeon HD 7000/R9 Series)
...
For all GPUs forced 16×AF, high quality filtering (if possible), vsync off

Test System - Games

Bioshock Infinite [2013, DX11]
Black Mesa [2020, DX9]
Control [2020, DX11]
Crysis [2007, DX9]
Crysis 3 [2013, DX11]
Dirt Rally [2015, DX11]
Doom [2016, OGL]
Far Cry 5 [2018, DX11]
GTA V [2015, DX11]
Kingdom Come: Deliverance [2018, DX11]
Mafia 2 [2010, DX9]
Metro: Last Light Redux [2014, DX11]
Project CARS [2015, DX11]
Rise of the Tomb Raider [2016, DX11]
Shadow Warrior [2013, DX11]
Tomb Raider [2013, DX11]
World of Tanks enCore RT [2019, DX11]

Tested Video Cards

	Radeon HD 5770	Radeon HD 5850	Radeon HD 5850 OC	Radeon HD 5870	Radeon HD 5970	3 × Radeon HD 5870	Radeon HD 6870	Radeon HD 6970 1 GB OC
GPU	Juniper	Cypress	Cypress	Cypress	2 × Cypress	3 × Cypress	Barts	Cayman
Architecture	Terascale 2	Terascale 2	Terascale 2	Terascale 2	Terascale 2	Terascale 2	Terascale 2	Terascale 3
Technology	40 nm	40 nm	40 nm	40 nm	40 nm	40 nm	40 nm	40 nm
Die Size	170 mm²	334 mm²	334 mm²	334 mm²	2 × 334 mm²	3 × 334 mm²	255 mm²	389 mm²
Transistor Count	1040 mil.	2154 mil.	2154 mil.	2154 mil.	2 × 2154 mil.	3 × 2154 mil.	1700 mil.	2640 mil.
Transistor Density	6.12 mil. / mm²	6.45 mil. / mm²	6.45 mil. / mm²	6.45 mil. / mm²	6.45 mil. / mm²	6.45 mil. / mm²	6.66 mil. / mm²	6.79 mil. / mm²
GPU Clock	860 MHz	725 MHz	980 MHz	850 MHz	725 MHz	850 MHz	920 MHz	925 MHz
Shader Clock	860 MHz	725 MHz	980 MHz	850 MHz	725 MHz	850 MHz	920 MHz	925 MHz
ROPs	16	32	32	32	2 × 32	3 × 32	32	32
TMUs	40	72	72	80	2 × 80	3 × 80	56	96
Compute Units	10	18	18	20	2 × 20	3 × 20	14	24
Shaders	800 Unified	1440 Unified	1440 Unified	1600 Unified	2 × 1600 Unified	3 × 1600 Unified	1120 Unified	1536 Unified
L1 Cache	10 × 8 kB	18 × 8 kB	18 × 8 kB	20 × 8 kB	2 × 20 × 8 kB	3 × 20 × 8 kB	14 × 8 kB	24 × 8 kB
L2 Cache	256 kB	512 kB	512 kB	512 kB	2 × 512 kB	3 × 512 kB	512 kB	512 kB
Memory	1024 MB GDDR5	1024 MB GDDR5	1024 MB GDDR5	1024 MB GDDR5	1024 MB GDDR5	1024 MB GDDR5	1024 MB GDDR5	1024 MB GDDR5
Memory Clock	4800 MHz	4000 MHz	5000 MHz	4800 MHz	4000 MHz	4800 MHz	4200 MHz	6000 MHz
Bus Width	128 bit	256 bit	256 bit	256 bit	2 × 256 bit	3 × 256 bit	256 bit	256 bit
Memory Bandwidth	76.8 GB/s	128 GB/s	160 GB/s	154 GB/s	256 GB/s	3 × 154 GB/s	134 GB/s	192 GB/s
Fillrate (Pixel)	13.8 GP/s	23.2 GP/s	31.4 GP/s	27.2 GP/s	2 × 23.2 GP/s	3 × 27.2 GP/s	29.4 GP/s	29.6 GP/s
Fillrate (Texel)	34.4 GT/s	52.2 GT/s	70.6 GT/s	68 GT/s	2 × 58 GT/s	3 × 68 GT/s	51.5 GT/s	88.8 GT/s
Compute Power (FP32)	1376 GFLOPS	2088 GFLOPS	2822 GFLOPS	2720 GFLOPS	2 × 2320 GFLOPS	3 × 2720 GFLOPS	2061 GFLOPS	2842 GFLOPS
Compute Power (FP64)	-	418 GFLOPS	564 GFLOPS	544 GFLOPS	2 × 464 GFLOPS	3 × 544 GFLOPS	-	711 GFLOPS
Bus Type	PCI-E 2.0	PCI-E 2.0	PCI-E 2.0	PCI-E 2.0	PCI-E 2.0	PCI-E 2.0	PCI-E 2.0	PCI-E 2.0
TDP	108 W	151 W	~250 W	188 W	294 W	~500 W	151 W	250 W
DirectX	11	11	11	11	11	11	11	11
OpenGL	4.5	4.5	4.5	4.5	4.5	4.5	4.5	4.5
Launch Year	2009	2009	2009	2009	2009	2009	2010	2010

	Radeon HD 6970 2 GB OC	Radeon HD 6990	Radeon HD 7850 1 GB	Radeon R9 270X OC	Radeon HD 7950 Boost	Radeon HD 7970	Radeon R9 280X	2 × Radeon R9 280X
GPU	Cayman	Antiles	Pitcairn	Pitcairn	Tahiti	Tahiti	Tahiti	2 × Tahiti
Architecture	Terascale 3	Terascale 3	CGN 1	CGN 1	CGN 1	CGN 1	CGN 1	CGN 1
Technology	40 nm	40 nm	28 nm	28 nm	28 nm	28 nm	28 nm	28 nm
Die Size	389 mm²	2 × 389 mm²	212 mm²	212 mm²	352 mm²	352 mm²	352 mm²	2 × 352 mm²
Transistor Count	2640 mil.	2 × 2640 mil.	2800 mil.	2800 mil.	4313 mil.	4313 mil.	4313 mil.	2 × 4313 mil.
Transistor Density	6.79 mil. / mm²	6.79 mil. / mm²	13.2 mil. / mm²	13.2 mil. / mm²	12.3 mil. / mm²	12.3 mil. / mm²	12.3 mil. / mm²	12.3 mil. / mm²
GPU Clock	900 MHz	880 MHz	900 MHz	1120 MHz	860 MHz	925 MHz	1100 MHz	1100 MHz
Shader Clock	900 MHz	880 MHz	900 MHz	1120 MHz	860 MHz	925 MHz	1100 MHz	1100 MHz
ROPs	32	2 × 32	32	32	32	32	32	2 × 32
TMUs	96	2 × 96	64	80	112	128	128	2 × 128
Compute Units	24	2 × 24	16	20	28	32	32	2 × 32
Shaders	1536 Unified	2 × 1536 Unified	1024 Unified	1280 Unified	1792 Unified	2048 Unified	2048 Unified	2 × 2048 Unified
L1 Cache	24 × 8 kB	2 × 24 × 8 kB	16 × 8 kB	20 × 8 kB	28 × 16 kB	32 × 16 kB	32 × 16 kB	2 × 32 × 16 kB
L2 Cache	512 kB	2 × 512 kB	512 kB	512 kB	768 kB	768 kB	768 kB	2 ×768 kB
Memory	2048 MB GDDR5	2048 MB GDDR5	1024 MB GDDR5	2048 MB GDDR5	3072 MB GDDR5	3072 MB GDDR5	3072 MB GDDR5	3072 MB GDDR5
Memory Clock	5800 MHz	5000 MHz	4800 MHz	5600 MHz	5000 MHz	5500 MHz	6000 MHz	6000 MHz
Bus Width	256 bit	2 × 256 bit	256 bit	256 bit	384 bit	384 bit	384 bit	2 × 384 bit
Memory Bandwidth	186 GB/s	2 × 160 GB/s	154 GB/s	179 GB/s	240 GB/s	264 GB/s	288 GB/s	2 × 288 GB/s
Fillrate (Pixel)	28.8 GP/s	2 × 28.2 GP/s	28.8 GP/s	35.8 GP/s	27.5 GP/s	29.6 GP/s	35.2 GP/s	2 × 35.2 GP/s
Fillrate (Texel)	86.4 GT/s	2 × 84.5 GT/s	57.6 GT/s	89.6 GT/s	96.3 GT/s	118.4 GT/s	140.8 GT/s	2 × 140.8 GT/s
Compute Power (FP32)	2765 GFLOPS	2 × 2703 GFLOPS	1843 GFLOPS	2867 GFLOPS	3082 GFLOPS	3789 GFLOPS	4506 GFLOPS	2 × 4506 GFLOPS
Compute Power (FP64)	691 GFLOPS	2 × 676 GFLOPS	115 GFLOPS	179 GFLOPS	771 GFLOPS	947 GFLOPS	1127 GFLOPS	2 × 1127 GFLOPS
Bus Type	PCI-E 2.0	PCI-E 2.0	PCI-E 2.0	PCI-E 2.0	PCI-E 3.0	PCI-E 3.0	PCI-E 3.0	PCI-E 3.0
TDP	250 W	375 W	126 W	180 W	200 W	260 W	295 W	2 × 295 W
DirectX	11	11	12	12	12	12	12	12
OpenGL	4.5	4.5	4.6	4.6	4.6	4.6	4.6	4.6
Launch Year	2010	2011	2012	2013	2012	2012	2013	2013

	Radeon R9 290	Radeon R9 290X OC	Radeon R9 380X	GeForce GTS 450	GeForce GTX 460 OEM	GeForce GTX 460 OC	GeForce GTX 470	GeForce GTX 480
GPU	Hawaii	Hawaii	Tonga	GF106	GF104	GF104	GF100	GF100
Architecture	CGN 2	CGN 2	CGN 3	Fermi	Fermi	Fermi	Fermi	Fermi
Technology	28 nm	28 nm	28 nm	40 nm	40 nm	40 nm	40 nm	40 nm
Die Size	438 mm²	438 mm²	366 mm²	238 mm²	332 mm²	332 mm²	526 mm²	526 mm²
Transistor Count	6200 mil.	6200 mil.	5000 mil.	1170 mil.	1950 mil.	1950 mil.	3200 mil.	3200 mil.
Transistor Density	14.2 mil. / mm²	14.2 mil. / mm²	13.7 mil. / mm²	4.92 mil. / mm²	5.87 mil. / mm²	5.87 mil. / mm²	6.08 mil. / mm²	6.08 mil. / mm²
GPU Clock	950 MHz	1100 MHz	980 MHz	810 MHz	650 MHz	925 MHz	608 MHz	700 MHz
Shader Clock	950 MHz	1100 MHz	980 MHz	1620 MHz	1300 MHz	1850 MHz	1216 MHz	1400 MHz
ROPs	64	64	32	16	32	32	40	48
TMUs	160	176	128	32	56	56	56	60
Compute Units	40	44	32	4	7	7	14	15
Shaders	2560 Unified	2816 Unified	2048 Unified	192 Unified	336 Unified	336 Unified	448 Unified	480 Unified
L1 Cache	40 × 16 kB	44 × 16 kB	32 × 16 kB	4 × 64 kB	7 × 64 kB	7 × 64 kB	14 × 64 kB	15 × 64 kB
L2 Cache	1024 kB	1024 kB	512 kB	256 kB	512 kB	512 kB	768 kB	768 kB
Memory	4096 MB GDDR5	4096 MB GDDR5	4096 MB GDDR5	1024 MB GDDR5	2048 MB GDDR5	1024 MB GDDR5	1280 MB GDDR5	1536 MB GDDR5
Memory Clock	5000 MHz	6200 MHz	5700 MHz	3600 MHz	3400 MHz	4200 MHz	3350 MHz	3700 MHz
Bus Width	512 bit	512 bit	256 bit	128 bit	256 bit	256 bit	320 bit	384 bit
Memory Bandwidth	320 GB/s	397 GB/s	182 GB/s	57.6 GB/s	109 GB/s	134 GB/s	134 GB/s	178 GB/s
Fillrate (Pixel)	60.8 GP/s	70.4 GP/s	31.4 GP/s	13 GP/s	20.8 GP/s	29.6 GP/s	24.3 GP/s	33.6 GP/s
Fillrate (Texel)	152 GT/s	193.6 GT/s	125.4 GT/s	25.9 GT/s	36.4 GT/s	51.8 GT/s	34 GT/s	42 GT/s
Compute Power (FP32)	4864 GFLOPS	6195 GFLOPS	4014 GFLOPS	622 GFLOPS	874 GFLOPS	1243 GFLOPS	1089 GFLOPS	1344 GFLOPS
Compute Power (FP64)	608 GFLOPS	774 GFLOPS	251 GFLOPS	52 GFLOPS	73 GFLOPS	104 GFLOPS	136 GFLOPS	168 GFLOPS
Bus Type	PCI-E 3.0	PCI-E 3.0	PCI-E 3.0	PCI-E 2.0	PCI-E 2.0	PCI-E 2.0	PCI-E 2.0	PCI-E 2.0
TDP	275 W	~350 W	190 W	106 W	150 W	~225 W	215 W	250 W
DirectX	12	12	12	11	11	11	11	11
OpenGL	4.6	4.6	4.6	4.6	4.6	4.6	4.6	4.6
Launch Year	2013	2013	2015	2010	2010	2010	2010	2010

	2 × GeForce GTX 480	GeForce GTX 560 Ti OC	2 × GeForce GTX 560 Ti OC	GeForce GTX 570 OC	GeForce GTX 580 OC	2 × GeForce GTX 580 OC	GeForce GTX 660 OEM	GeForce GTX 670
GPU	2 × GF100	GF114	2 × GF114	GF110	GF110	2 × GF110	GK104	GK104
Architecture	Fermi	Fermi	Fermi	Fermi	Fermi	Fermi	Kepler	Kepler
Technology	40 nm	40 nm	40 nm	40 nm	40 nm	40 nm	28 nm	28 nm
Die Size	2 × 526 mm²	332 mm²	2 × 332 mm²	520 mm²	520 mm²	520 mm²	294 mm²	294 mm²
Transistor Count	2 × 3200 mil.	1950 mil.	2 × 1950 mil.	3000 mil.	3000 mil.	3000 mil.	3540 mil.	3540 mil.
Transistor Density	6.08 mil. / mm²	5.87 mil. / mm²	5.87 mil. / mm²	5.77 mil. / mm²	5.77 mil. / mm²	5.77 mil. / mm²	12 mil. / mm²	12 mil. / mm²
GPU Clock	700 MHz	1000 MHz	950 MHz	845 MHz	900 MHz	855 MHz	950 MHz	1070 MHz
Shader Clock	1400 MHz	2000 MHz	1900 MHz	1690 MHz	1800 MHz	1710 MHz	950 MHz	1070 MHz
ROPs	2 × 48	32	2 × 32	40	48	2 × 48	24	32
TMUs	2 × 60	64	2 × 64	60	64	2 × 64	96	112
Compute Units	2 × 15	8	2 × 8	15	16	2 × 16	6	7
Shaders	2 × 480 Unified	384 Unified	2 × 384 Unified	480 Unified	512 Unified	2 × 512 Unified	1152 Unified	1344 Unified
L1 Cache	2 × 15 × 64 kB	8 × 64 kB	2 × 8 × 64 kB	15 × 64 kB	16 × 64 kB	2 × 16 × 64 kB	6 × 16 kB + 48 kB Tex	7 × 16 kB + 48 kB Tex
L2 Cache	2 × 768 kB	512 kB	2 × 512 kB	768 kB	768 kB	2 × 768 kB	384 kB	512 kB
Memory	1536 MB GDDR5	1024 MB GDDR5	1024 MB GDDR5	1280 MB GDDR5	1536 MB GDDR5	1536 MB GDDR5	3072 MB GDDR5	2048 MB GDDR5
Memory Clock	3700 MHz	4800 MHz	4580 MHz	3800 MHz	4800 MHz	4600 MHz	5600 MHz	6000 MHz
Bus Width	2 × 384 bit	256 bit	2 × 256 bit	320 bit	384 bit	2 × 384 bit	192 bit	256 bit
Memory Bandwidth	2 × 178 GB/s	154 GB/s	2 × 147 GB/s	152 GB/s	230 GB/s	2 × 221 GB/s	134 GB/s	192 GB/s
Fillrate (Pixel)	2 × 33.6 GP/s	32 GP/s	2 × 30.4 GP/s	33.8 GP/s	43.2 GP/s	2 × 41 GP/s	22.8 GP/s	34.2 GP/s
Fillrate (Texel)	2 × 42 GT/s	64 GT/s	2 × 60.8 GT/s	50.7 GT/s	57.6 GT/s	2 × 54.7 GT/s	91.2 GT/s	119.8 GT/s
Compute Power (FP32)	2 × 1344 GFLOPS	1536 GFLOPS	2 × 1459 GFLOPS	1622 GFLOPS	1843 GFLOPS	2 × 1751 GFLOPS	2189 GFLOPS	2876 GFLOPS
Compute Power (FP64)	2 × 168 GFLOPS	128 GFLOPS	2 × 122 GFLOPS	203 GFLOPS	154 GFLOPS	2 × 146 GFLOPS	91 GFLOPS	120 GFLOPS
Bus Type	PCI-E 2.0	PCI-E 2.0	PCI-E 2.0	PCI-E 2.0	PCI-E 2.0	PCI-E 2.0	PCI-E 3.0	PCI-E 3.0
TDP	2 × 250 W	~250 W	~450 W	220 W	~320 W	~600 W	105 W	200 W
DirectX	11	11	11	11	11	11	12	12
OpenGL	4.6	4.6	4.6	4.6	4.6	4.6	4.6	4.6
Launch Year	2010	2011	2011	2010	2010	2010	2012	2012

	GeForce GTX 690	GeForce GTX 760 OC	GeForce GTX 770 OC	GeForce GTX 780 @ 155 W	GeForce GTX 780 OC	GeForce GTX 780 Ti OC	Quadro 4000
GPU	2 × GK104	GK104	GK104	GK110	GK110	GK110	GF100
Architecture	Kepler	Kepler	Kepler	Kepler	Kepler	Kepler	Fermi
Technology	28 nm	28 nm	28 nm	28 nm	28 nm	28 nm	40 nm
Die Size	2 × 294 mm²	294 mm²	294 mm²	561 mm²	561 mm²	561 mm²	526 mm²
Transistor Count	2 × 3540 mil.	3540 mil.	3540 mil.	7100 mil.	7100 mil.	7100 mil.	3200 mil.
Transistor Density	12 mil. / mm²	12 mil. / mm²	12 mil. / mm²	12.7 mil. / mm²	12.7 mil. / mm²	12.7 mil. / mm²	6.08 mil. / mm²
GPU Clock	1050 MHz	1175 MHz	1290 MHz	900 MHz	1100 MHz	1200 MHz	475 MHz
Shader Clock	1050 MHz	1175 MHz	1290 MHz	900 MHz	1100 MHz	1200 MHz	950 MHz
ROPs	2 × 32	32	32	48	48	48	32
TMUs	2 × 128	96	128	192	192	240	32
Compute Units	2 × 8	6	8	12	12	15	8
Shaders	2 × 1536 Unified	1152 Unified	1536 Unified	2304 Unified	2304 Unified	2880 Unified	256 Unified
L1 Cache	2 × 8 × 16 kB + 48 kB Tex	6 × 16 kB + 48 kB Tex	8 × 16 kB + 48 kB Tex	12 × 16 kB + 48 kB Tex	12 × 16 kB + 48 kB Tex	15 × 16 kB + 48 kB Tex	8 × 64 kB
L2 Cache	2 × 512 kB	512 kB	512 kB	1536 kB	1536 kB	1536 kB	512 kB
Memory	2048 MB GDDR5	2048 MB GDDR5	4096 MB GDDR5	3072 MB GDDR5	3072 MB GDDR5	3072 MB GDDR5	2048 MB GDDR5
Memory Clock	6000 MHz	6000 MHz	8000 MHz	6000 MHz	6000 MHz	7000 MHz	2800 MHz
Bus Width	2 × 256 bit	256 bit	256 bit	384 bit	384 bit	384 bit	256 bit
Memory Bandwidth	2 × 192 GB/s	192 GB/s	256 GB/s	288 GB/s	288 GB/s	336 GB/s	89.6 GB/s
Fillrate (Pixel)	2 × 33.6 GP/s	37.6 GP/s	41.3 GP/s	43.2 GP/s	52.8 GP/s	57.6 GP/s	15200 MP/s
Fillrate (Texel)	2 × 134.4 GT/s	112.8 GT/s	165.1 GT/s	172.2 GT/s	211.2 GT/s	288 GT/s	15200 MT/s
Compute Power (FP32)	2 × 3076 GFLOPS	2707 GFLOPS	3963 GFLOPS	4147 GFLOPS	5069 GFLOPS	6912 GFLOPS	486 GFLOPS
Compute Power (FP64)	2 × 128 GFLOPS	113 GFLOPS	165 GFLOPS	173 GFLOPS	211.8 GFLOPS	288 GFLOPS	243 GFLOPS
Bus Type	PCI-E 3.0	PCI-E 3.0	PCI-E 3.0	PCI-E 3.0	PCI-E 3.0	PCI-E 3.0	PCI-E 2.0
TDP	256 W	250 W	~ 320 W	~ 155 W	275 W	300 W	142 W
DirectX	12	12	12	12	12	12	11
OpenGL	4.6	4.6	4.6	4.6	4.6	4.6	4.6
Launch Year	2012	2013	2013	2013	2013	2013	2010