Join Us On Facebook

Compare any two graphics cards:
VS

Radeon HD 5870 vs Radeon HD 6950

Intro

The Radeon HD 5870 comes with a clock speed of 850 MHz and a GDDR5 memory frequency of 1200 MHz. It also features a 256-bit bus, and makes use of a 40 nm design. It features 1600(320x5) SPUs, 80 Texture Address Units, and 32 Raster Operation Units.

Compare those specifications to the Radeon HD 6950, which comes with core speeds of 800 MHz on the GPU, and 1250 MHz on the 1024 MB of GDDR5 RAM. It features 1408 SPUs along with 88 TAUs and 32 ROPs.

(No game benchmarks for this combination yet.)

Power Usage and Theoretical Benchmarks

Power Consumption (Max TDP)

Radeon HD 5870 188 Watts
Radeon HD 6950 200 Watts
Difference: 12 Watts (6%)

Memory Bandwidth

Theoretically speaking, the Radeon HD 6950 should be 4% faster than the Radeon HD 5870 overall, because of its greater data rate. (explain)

Radeon HD 6950 160000 MB/sec
Radeon HD 5870 153600 MB/sec
Difference: 6400 (4%)

Texel Rate

The Radeon HD 6950 is a bit (about 4%) better at anisotropic filtering than the Radeon HD 5870. (explain)

Radeon HD 6950 70400 Mtexels/sec
Radeon HD 5870 68000 Mtexels/sec
Difference: 2400 (4%)

Pixel Rate

If running with high levels of AA is important to you, then the Radeon HD 5870 is the winner, though not by far. (explain)

Radeon HD 5870 27200 Mpixels/sec
Radeon HD 6950 25600 Mpixels/sec
Difference: 1600 (6%)

Please note that the above 'benchmarks' are all just theoretical - the results were calculated based on the card's specifications, and real-world performance may (and probably will) vary at least a bit.

Price Comparison

Radeon HD 5870

Amazon.com

Radeon HD 6950

Amazon.com

Please note that the price comparisons are based on search keywords - sometimes it might show cards with very similar names that are not exactly the same as the one chosen in the comparison. We do try to filter out the wrong results as best we can, though.

Specifications

Model Radeon HD 5870 Radeon HD 6950
Manufacturer AMD AMD
Year September 23, 2009 December 2010
Code Name Cypress XT Cayman Pro
Fab Process 40 nm 40 nm
Bus PCIe 2.1 x16 PCIe x16
Memory 1024 MB 1024 MB
Core Speed 850 MHz 800 MHz
Shader Speed N/A MHz (N/A) MHz
Memory Speed 1200 MHz (4800 MHz effective) 1250 MHz (5000 MHz effective)
Unified Shaders 1600(320x5) 1408
Texture Mapping Units 80 88
Render Output Units 32 32
Bus Type GDDR5 GDDR5
Bus Width 256-bit 256-bit
DirectX Version DirectX 11 DirectX 11
OpenGL Version OpenGL 3.2 OpenGL 4.1
Power (Max TDP) 188 watts 200 watts
Shader Model 5.0 5.0
Bandwidth 153600 MB/sec 160000 MB/sec
Texel Rate 68000 Mtexels/sec 70400 Mtexels/sec
Pixel Rate 27200 Mpixels/sec 25600 Mpixels/sec

Memory Bandwidth: Memory bandwidth is the maximum amount of information (counted in MB per second) that can be moved past the external memory interface within a second. The number is calculated by multiplying the interface width by its memory speed. If the card has DDR type RAM, it must be multiplied by 2 again. If it uses DDR5, multiply by ANOTHER 2x. The better the card's memory bandwidth, the better the card will be in general. It especially helps with AA, HDR and high resolutions.

Texel Rate: Texel rate is the maximum number of texture map elements (texels) that can be applied per second. This number is calculated by multiplying the total amount of texture units of the card by the core clock speed of the chip. The higher the texel rate, the better the video card will be at texture filtering (anisotropic filtering - AF). It is measured in millions of texels in a second.

Pixel Rate: Pixel rate is the maximum number of pixels the video card can possibly write to the local memory per second - measured in millions of pixels per second. The figure is worked out by multiplying the number of Render Output Units by the clock speed of the card. ROPs (Raster Operations Pipelines - aka Render Output Units) are responsible for drawing the pixels (image) on the screen. The actual pixel fill rate is also dependant on many other factors, especially the memory bandwidth - the lower the bandwidth is, the lower the ability to reach the maximum fill rate.

Comments

8 Responses to “Radeon HD 5870 vs Radeon HD 6950”
mauro says:
muy buena la pagina es exactamente lo que estaba buscando...muy pero muy buena, mas aya de que los resultados no sean exactos, la idea esta...y es exelente, felicitaciones!!
[...] habe dann ist der Unterschied nicht wirklich merkbar. Hier sieht du ganz grob den Unterschied Radeon HD 5870 vs Radeon HD 6950 – Performance Comparison Benchmarks @ Hardware Compare Zitieren + [...]
[...] to be enjoyed? How big difference is there really between HD5870 and HD6950? Please enlight me.. http://www.hwcompare.com/6132/radeon...adeon-hd-6950/ An answer would be appreciated.... Attached Images [...]
Viking2121 says:
No difference at all, Overclock a tad and it faster than the 6950, only thing thats better is more vram.
larsV12 says:
@mauro .. like Viking says - only, i'd add that the 6950 offers OpenGL 4.1 support. What that means, in rel terms, is it might be slightly better optimized for newer, 3D gaming titles that utilise OpenGL. Other than that? They're very evenly matched cards .. so put it like this: flip a coin - heads for the 5870, tails for the 6950 - to pick one ... i guarantee you won't be disappointed with the outcome ;P
George Kohler says:
The 2GB 6950 is the way to go! You can flash the 6970 firmware and unlock the locked shader cores and if you use the ASUS firmware you can even add a 20% overclock to blow the 5870 out of the water. Crisis 3 with everything maxed out hits 136 FPS with the ASUS 6970 firmware and 20% boost and runs at 89 degrees C well below the 95C design limit since it is after all the same exact reference hardware only hobbled by a greedy company to dole out performance. I love finding out these cheesy ploys and exploiting them.
larsV12 says:
@George K ... the 6950 1GB has the same unlock capability ( ..i should know, the rig i'm typing this on has a 6950-1G @ 1536 SPU's ;P ). The 6950-2G's obviously will rule at high res' over the 1GB versions though. I'm getting high 60's, temperature wise (on ref' cooler), with Crysis & idling @ 30-35. Wanna know how to keep these babies running cool - without after-market cooling (... and short of putting the rig in a freezer)?? ... ask me how ;P
LarsV12 says:
Actually, after testing the 5870 vs 6950, on a modest 5.5% bump on the core (to 900), for the 5870, and a 4% bump on the memory clock (to 1250), the 5870 returns, pixel fill: 28.8GP/s, texture fill: 72GT/s and mem' b/w @ 160GB/s. That's slightly better than parity with the 6950. Also consider that the 5870 is known to be able to clock up to core: 1000 & mem:1300 memory: http://www.tweaktown.com/articles/3286/gigabyte_radeon_hd_5870_1gb_super_overclock_video_card_overclocked_higher/index11.html --> Add to the fact 5870's have 1600 shader cores vs just 1408 for 6950's, and all the sudden the 6950, by all rights, looks more like a 'side-grade' rather than an upgrade.

Your email address will not be published.


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Spam Protection by WP-SpamFree