[Tested] SAPPHIRE Radeon HD 6870 1024MB Review

SAPPHIRE Radeon HD 6870 Review Index

3 – SAPPHIRE Radeon HD 6870 OpenGL performances

Testbed:
– CPU: Core i7 960 @ 3.2GHz
– RAM: 4GB DDR3 Corsair Dominator
– Motherboard: GIGABYTE X58-A UD5
– Windows 7 64-bit
– Graphics drivers: Cat 10.10
– PSU: Corsair AX1200

PSU: Corsair AX1200

3.1 FurMark (OpenGL 2)

FurMark 1.8.2 has been used for the test. FurMark homepage is HERE.

Settings: 1920×1080 fullscreen, no AA, no postFX, 60sec, Xtreme mode UNCHECKED.

Rule: The higher the number of points, the faster the card is.

8259 points (Avg FPS: 138) – ASUS ENGTX580 ***UNLOCKED***, GPU Core: 871MHz, VDDC: 1.088V
6470 points – EVGA GeForce GTX 480
5420 points – ATI Radeon HD 5870
5161 points – MSI GeForce GTX 470
4693 points (Avg FPS: 78) – SAPPHIRE Radeon HD 6870, GPU core: 1020MHz
4641 points (Avg FPS: 78) – SAPPHIRE Radeon HD 6870, GPU core: 1000MHz
4583 points (Avg FPS: 76) – SAPPHIRE Radeon HD 6870, GPU core: 980MHz
4484 points (Avg FPS: 74) – ASUS EAH6870
4310 points (Avg FPS: 72) – SAPPHIRE Radeon HD 6870
3884 points – MSI N460GTX Cyclone 768D5 OC
2772 points – MSI R5770 Hawk

Remark: the score with a 1020MHz GPU is given only for the fun. Stressed by FurMark in Xtreme mode, the GPU clocked at 1020MHz is not stable (freeze).

3.2 TessMark (OpenGL 4)

Hardware tessellation is one of the big features of Direct3D 11 and OpenGL 4 capable graphics cards. TessMark is a new benchmark focused only on the tessellation engine of DX11 class cards. It’s a pure tessellation benchmark, it does not contain complex shader or other heavy texture fetches. TessMark shows an overview of the tessellation engine raw power, that’s all. DX11 specifies that the tessellation factor can vary from 1.0 up tp 64.0. Of course, for tessellation factors like 32 or 64, most of the tessellated triangles are smaller than… a pixel. In those cases, tessellation is useless and in a real world application such as a game, high tessellation factors won’t be used. But in the case of a synthetic benchmark, it’s always instructive to see how cards can handle the whole range of tessellation level.

TessMark 0.2.2 has been used for the test.

Settings: 1920×1080 fullscreen, no AA, 60sec, map set 1.

TessMark - OpenGL 4 tessellation

Rule: The higher the number of points, the faster the card is.

Tessellation factor 8.0: moderate

52188 (872FPS) – ASUS ENGTX580
48084 – EVGA GeForce GTX 480
38191 – MSI GeForce GTX 470
30512 – MSI N460GTX Cyclone 768D5 OC
27718 (462FPS) – Sapphire HD 6870, GPU core: 1000MHz
26223 – ASUS EAH6870
24161 (403FPS) – ATI Radeon HD 5870
23131 (386FPS) – Sapphire HD 6870
20745 – MSI R5770 Hawk

Tessellation factor 16.0: normal

36369 (545FPS) – ASUS ENGTX580
29196 – EVGA GeForce GTX 480
23316 – MSI GeForce GTX 470
17452 – MSI N460GTX Cyclone 768D5 OC
9255 (154FPS) – Sapphire HD 6870, GPU core: 1000MHz
8555 – ASUS EAH6870
8177 (136FPS) – Sapphire HD 6870
8018 (134FPS) – ATI Radeon HD 5870
7669 – MSI R5770 Hawk

Tessellation factor 32.0: extreme

15128 (252FPS) – ASUS ENGTX580
13008 – EVGA GeForce GTX 480
9997 – MSI GeForce GTX 470
6729 – MSI N460GTX Cyclone 768D5 OC
2299 – ASUS EAH6870
2246 (38FPS) – Sapphire HD 6870
2156 (36FPS) – ATI Radeon HD 5870
2159 – MSI R5770 Hawk

Tessellation factor 64.0: insane

4840 (81FPS) – ASUS ENGTX580
3963 – EVGA GeForce GTX 480
3169 – MSI GeForce GTX 470
1959 – MSI N460GTX Cyclone 768D5 OC
585 – ASUS EAH6870
574 (10FPS) – Sapphire HD 6870
565 – MSI R5770 Hawk
550 (10FPS) – ATI Radeon HD 5870

3.3 ShaderToyMark (OpenGL 2)

ShaderToyMark 0.1.0 is an OpenGL 2 benchmark, developed with GeeXLab,
and focused on pixel shaders only. The pixel shaders are heavily based on math (few texture fetches) and then ShaderToyMark can be seen as a kind of GPU computing benchmark.

ShaderToyMark - OpenGL 2 pixel shader

Settings: 960×540 windowed, no AA, 60sec

306 points (51 FPS) – ASUS ENGTX580
263 points (43 FPS) – GeForce GTX 480
189 points (31 FPS) – ATI Radeon HD 5870
184 points (30 FPS) – ASUS EAH6870
179 (29FPS) – Sapphire HD 6870
156 points (26 FPS) – MSI N460GTX Cyclone
104 points (17 FPS) – MSI R5770 Hawk
46 points (7 FPS) – GeForce 9800 GTX
36 points (6 FPS) – EVGA GTX 280
33 points (5 FPS) – GeForce GTX 260

3.4 Maxon CINEBENCH R11.5 (OpenGL 2)

CINEBENCH is an OpenGL benchmark based on Cinema4D.

Maxon CINEBENCH R11.5

Settings: windowed mode 1920×1080.

61.79 FPS – Sapphire Radeon HD 6870
61.22 FPS – Radeon HD 5870
42.84 FPS – GeForce GTX 480
42.53 FPS – MSI N460GTX Cyclone

3.5 OpenGL 4 Mountains demo

Mountains demo is an OpenGL 4 demo that shows hierarchical-Z map based occlusion culling in action.

OpenGL 4 Mountains demo

Settings: default window size: 1024×768, ICR enabled (Instance Cloud Reduction), Hi-Z enabled and dynamic LOD enabled.

674 FPS – ASUS ENGTX580
568 FPS – EVGA GTX 480
350 FPS – MSI N460GTX Cyclone 768D5
255 FPS – ASUS EAH6870
235 FPS – Sapphire Radeon HD 6870
231 FPS – Radeon HD 5870
220 FPS – MSI R5770 Hawk

3.6 GluxMark2 (OpenGL 2)

GluxMark2 is a purely syntethic OpenGL benchmark
and tries to measure performance from every point of view by using programmable graphics pipeline (vertex, geometry and fragment/pixel shaders).

OpenGL 2 - GluxMark2

Preset: high-end (1920×1080, MSAA: 8X)

9216 points (OpenCL: +2776 = 11992 points) – GeForce GTX 480
8099 points – Radeon HD 5870
6615 points – Sapphire Radeon HD 6870
5367 points (OpenCL: +2789 = 8156 points) – MSI N460GTX Cyclone

3.7 Unigine Heaven (OpenGL 4)

For this last OpenGL test, I used Ungine Heaven 2.1, one of the standard Direct3D / OpenGL synthetic benchmark.

OpenGL 4 - Unigine Heaven 2.1

Settings: 1920×1080 fullscreen, OpenGL rendering, tessellation: normal, shaders: high, AA: 4X, 16X anisotropic filtering.

46.4 FPS, Scores: 1168 – ASUS ENGTX580
38.7 FPS, Scores: 974 – EVGA GeForce GTX 480
24.5 FPS, Scores: 617 – MSI N460GTX Cyclone 768D5 OC
15.9 FPS, Scores: 400 – ATI Radeon HD 5870
13.6 FPS, Scores: 342 – ASUS EAH6870
13.5 FPS, Scores: 339 – SAPPHIRE HD6870
9 FPS, Scores: 227 – MSI R5770 Hawk

SAPPHIRE Radeon HD 6870 Review Index

13 thoughts on “[Tested] SAPPHIRE Radeon HD 6870 1024MB Review”

  1. JeGX Post Author

    GTX 580 Crysis score is for the next review. I’m testing a GTX 580 from EVGA…

  2. Lokavidu

    guys seriously consider putting the index on the bottom of the site too.. using home key is a bit annoying and unfamiliar

  3. iPristy

    OK, will you then compare 6870 vs 580 ?

    Radeons are fast enough in games, artificial bench demos it’s about the same as Sunspider bench who cares about that.

  4. JeGX Post Author

    @iPristy: sure, all previous scores (HD 5870, HD 6870, GTX 4xx) will be in EVGA GTX 580 review.

  5. Pingback: [Tested and Burned] EVGA GeForce GTX 580 SC Review - 3D Tech News, Pixel Hacking, Data Visualization and 3D Programming - Geeks3D.com

  6. Pingback: [GPU Tool] ShaderToyMark 0.2.0 Demomakers’ Benchmark Available with Better SLI and CrossFire Support - 3D Tech News, Pixel Hacking, Data Visualization and 3D Programming - Geeks3D.com

  7. Pingback: AMD Catalyst 10.12 WHQL (OpenGL 4.1, OpenCL 1.1) - 3D Tech News, Pixel Hacking, Data Visualization and 3D Programming - Geeks3D.com

  8. Pingback: [Tested and Burned] SAPPHIRE Radeon HD 6970 2048MB Review - 3D Tech News, Pixel Hacking, Data Visualization and 3D Programming - Geeks3D.com

  9. Pingback: [Tested] ASUS Radeon HD 6950 2GB GDDR5 Review - 3D Tech News, Pixel Hacking, Data Visualization and 3D Programming - Geeks3D.com

  10. Pingback: [Tested and Burned] ASUS GeForce GTX 560 Ti DirectCU II TOP Review - 3D Tech News, Pixel Hacking, Data Visualization and 3D Programming - Geeks3D.com

  11. kromar

    these opengl tests do not at all reflect the performance of real opengl applications.
    when using the latest fermi architecutre in opengl heavy 3d apps like blender, maya and so on the true weakness of these cards emerge.

    it would be really great if you could add a test done in blender to the ones yu already did to see how the “real” opengl performance is.

Comments are closed.