if each using separate framebuffer than, what the advantage in cfx?if it got large vram advantage in crysis, then it would beat 580 in single ...in real, cfx scaling made here 69xx better..in metro sli, cfx both r giving 2x performance...and, pretty sure that any one buying 69xx cfx or 5xxgtx sli rig ,is not going to play below 1900x1200 res...so 69xx cfx is a clear winner.
AMD HD 6950 and 6970 released
if each using separate framebuffer than, what the advantage in cfx?if it got large vram advantage in crysis, then it would beat 580 in single ...in real, cfx scaling made here 69xx better..in metro sli, cfx both r giving 2x performance...and, pretty sure that any one buying 69xx cfx or 5xxgtx sli rig ,is not going to play below 1900x1200 res...so 69xx cfx is a clear winner.
vickybat
I am the night...I am...
The gtx series is also scaling well. In fact in newer games like metro, cfx in 69xx series is behind nvidia. Thats what i was talking about. Sli is not at all scaling bad and will improve on future driver release. Wait for newer games and see. The framebuffer is utilized better when two or more cards come into play. In this case 2gb for amd and 1.5gb and 1.25gb for nvidia.
And both 6970 and 6950 are behind nvidia 5 series in all benchmarks. So in performance nvidia is ahead. amd can only fight back in prices. Moreover gtx 560 is also on the way and has the potential to topple both 6870 and 6850 and maybe perform close to 6950.
J
Joker
Guest
choose a neutral game to compare scaling
anyways
*media.bestofmicro.com/R/7/273139/original/CrossFire%20Metro%202033%20Scaling.png
check out topgear's post citing benchmarks... hd 6970 = gtx 570 with both leading in respective benchmarks. but the latter is the better deal.
OP
^^ In single card config here's what the results out of 7 DX 11 aps :
Metro 2033 : DX11 PhysyX off
HD6970
DiRT 2 :
GTX 570
BFBC2 :
GTX 570* = HD6970 ( when OCed GTX 570 wins )
3D Mark 11 :
HD6970
Lost Planet 2 :
GTX 570
Just Cause 2 (DX11)
HD6970
Aliens Vs. Predator (DX11)
HD6970
GTX 570 winner of 3 titles and HD 6970 winner of 4 DX 11 title .
No doubt that can be possible - Read On Toms Hardware about EVGA GTX 460 FTW ! review.
In comparison it's neck to neck with GTX 470 and HD 6870 and in some games even outperformed them.
Metro 2033 : DX11 PhysyX off
HD6970
DiRT 2 :
GTX 570
BFBC2 :
GTX 570* = HD6970 ( when OCed GTX 570 wins )
3D Mark 11 :
HD6970
Lost Planet 2 :
GTX 570
Just Cause 2 (DX11)
HD6970
Aliens Vs. Predator (DX11)
HD6970
GTX 570 winner of 3 titles and HD 6970 winner of 4 DX 11 title .
No doubt that can be possible - Read On Toms Hardware about EVGA GTX 460 FTW ! review.
In comparison it's neck to neck with GTX 470 and HD 6870 and in some games even outperformed them.
Last edited:
@ topgear
from where do you get these benchmarks? please include the name of site as well.
yeah read about the evga ftw too.what i would have liked to see was each card OC'ed and then compared.as both 460 and 6850 OC like hell.
meanwhile, on guru3d there is review of msi 6850 cyclone. it reached >1GHZ on core. f***ing awesome!
from where do you get these benchmarks? please include the name of site as well.
yeah read about the evga ftw too.what i would have liked to see was each card OC'ed and then compared.as both 460 and 6850 OC like hell.
meanwhile, on guru3d there is review of msi 6850 cyclone. it reached >1GHZ on core. f***ing awesome!
performance wise evga 460gtx 1gb is equal or better than gtx470 stock..but it has 4~5%oc capability.460gtx is slight good overclocker than 6850 in referance cooler.but, custom cooled 6850 can reach 1.05ghz and overcome 460gtx oced.
HD 6870 & HD 6850 vs. GTX 460 1GB: An Overclocking Study
i was talking about multigpu solution.not single.in metro both sli(5xx),cfx(6xxx) scaling is equal to 2x.see guru3d 6950cfx review.in case of single gpu, 6970 is a clear winner against 570,and a 6950 @870mhz/5.8ghz equal to 570 in 8xmsaa and 2500x1600 resolution.may be coz of 2gb vram,but mind it, we r getting that 2gb 6950 in a lot better price.
Last edited:
vickybat
I am the night...I am...
Can you justify the statement in bold? I don't see the 6970 to be a clear winner at all. If 6950 can be overclocked, so can the 570 which will blow the pants out of 6950.
The gtx 570 is priced perfectly for the perfomance it delivers and if price will be reduced, then we can get an even sweeter deal.
@ asingh
Buddy i meant that mutiple gpu's can better utilize a large framebuffer when rendering in real time (read procedural rendering) than a single one (though from the same family or generation of gpu's).
For example 2 6950's in cf will take the advantage og 2gb vram than a single 6950. Same goes for a single gtx 570 with 1.25gb and two 570's in sli which utilize the same amount of vram as in a single board. But i think 570 would have scaled or performed much better with more framebuffer , lets say somethin equivalent to the caymans.
asingh
Aspiring Novelist
@VickyBat:
I think in multi-GPU subsystems the theoretical framebuffer does not go into 2x. Each frame is rendered twice -- once on each card. Not sure how you can correlate 2 GPUs will take better advantage of VRAM vs. a single accelerator.
Though the nVidia boards have less VRAM but their data path is wider compared to ATIs.
I think in multi-GPU subsystems the theoretical framebuffer does not go into 2x. Each frame is rendered twice -- once on each card. Not sure how you can correlate 2 GPUs will take better advantage of VRAM vs. a single accelerator.
Though the nVidia boards have less VRAM but their data path is wider compared to ATIs.
tkin
Back to school!!
There are different techniques, one other approach is to split the frame into two or more parts depending on gpus, the load balancing makes sure that the gpus get frames proportional to their processing power, this is called scissor, ati used it before in mixed crossfire setup(just like lucid hydra).
OP
Here's what I've compiled about SLI and CF form two websites to know about these two a little bit better - I know it's not perfect but it will give us some clear idea for sure :
First Cross Fire :
Crossfire system comes with 3 ways of splitting the workload :
AFR - Alternate Frame Rendering.
Alternate frame rendering is also a technique that Nvidia's SLI can use. AFR is as simple as it sounds. One frame is fully rendered by one card while the second card is already working on the next frame before switching back again. This can almost double performance upto the point where the work on one frame becomes two much and then other modes become more useful. Also some games cannot use AFR as each frame can be dependent on the frame before it, therefore frame 2 cannot be rendered until frame 1 has been completed, defeating the object of AFR entirely. I a crossfire system I would recommend using Radeon cards capable of running the frame rate on there own adequately or this system becomes less efficient.
Be aware that you require one of each type to make this work and they must be from the same range i.e. both X800's. However unlike Nvidia's solution you can mix and match varieties of these cards. For example you can pair a X800 XT Crossfire edition with a standard X800 pro. You wont see as much benefit as having two X800 XT's but it is possible and could help some people with lower speed X800 / X850 cards.
Supertiling
Supertiling is exclusive to ATI, Nvidia doesn't have a solution similar to this in their SLI technology. Supertiling shares the workload of a frame between the two graphics cards by splitting it up in the form of a chess board. Tile one is sent to the first graphics card for processing, tile two is sent to the second graphics card, then tile 3 is sent to the first card again and so on until the frame is fully rendered.
Supertiling sounds like a very complex method of splitting up a screen but take this as an example, the traditional method is to split the screen in half horizontally. The top half of the screen is sent to the first graphics card and the bottom half is sent to the second graphics card. Supertiling does not support OpenGL and it doesn't give optimal performance all the time.
Scissor Frame Rendering (SFR)
Scissor frame rendering or SFR is very similar to the original method of SLI and Nvidia's Split Frame Rendering. SFR is the basic cut the screen in half and send half to the one graphics card and half to the other. There are some differences however between the original voodoo 2 SLI, Nvidia's Split Frame Rendering and ATI's Scissor Frame Rendering.
The original SLI simple cut the frame in half, no mathematics was required no calculations just cut in half and sent to two graphics cards. Nvidia's solution is the same but has a dynamic load calculation built in. It does not necessarily split the screen 50/50. It will calculate the load at the top of the frame and the bottom and split the frame accordingly.
ATI's solution is a little bit of both of the above. The screen is not always split 50/50 but its not calculated on the fly. This can save some calculations free up more clock cycles but can cause a slight drop from optimal efficiency. The load is set by the application / game and that value is fixed throughout the session. So if the screen is split 60 / 40 then that is how it will stay. The main advantage of the ATI solution here is that it can split the screen horizontally like the others, but also vertically. If splitting the screen vertically would give better results
Source
Now SLi :
The graphics memory is NOT doubled in SLI mode. If you have two 128MB graphics cards, you do NOT have an effective 256MB. Most of the games operate in AFR (alternate frame rendering). The first card renders one full frame and then the next card renders the next frame and so on. If you picture SLI working in this way it is easy to see that each frame only has 128MB of memory to work with.
If you feel you need 256MB or even 512MB of graphics memory for certain games, then you need to make sure that each card has that much memory. Also keep in mind that if you mix a 128MB GeForce 6600GT with a 256MB GeForce 6600GT that BOTH cards will operate with 128MB of graphics memory (the lowest common memory size of the two cards).
Source
First Cross Fire :
Crossfire system comes with 3 ways of splitting the workload :
AFR - Alternate Frame Rendering.
Alternate frame rendering is also a technique that Nvidia's SLI can use. AFR is as simple as it sounds. One frame is fully rendered by one card while the second card is already working on the next frame before switching back again. This can almost double performance upto the point where the work on one frame becomes two much and then other modes become more useful. Also some games cannot use AFR as each frame can be dependent on the frame before it, therefore frame 2 cannot be rendered until frame 1 has been completed, defeating the object of AFR entirely. I a crossfire system I would recommend using Radeon cards capable of running the frame rate on there own adequately or this system becomes less efficient.
Be aware that you require one of each type to make this work and they must be from the same range i.e. both X800's. However unlike Nvidia's solution you can mix and match varieties of these cards. For example you can pair a X800 XT Crossfire edition with a standard X800 pro. You wont see as much benefit as having two X800 XT's but it is possible and could help some people with lower speed X800 / X850 cards.
Supertiling
Supertiling is exclusive to ATI, Nvidia doesn't have a solution similar to this in their SLI technology. Supertiling shares the workload of a frame between the two graphics cards by splitting it up in the form of a chess board. Tile one is sent to the first graphics card for processing, tile two is sent to the second graphics card, then tile 3 is sent to the first card again and so on until the frame is fully rendered.
Supertiling sounds like a very complex method of splitting up a screen but take this as an example, the traditional method is to split the screen in half horizontally. The top half of the screen is sent to the first graphics card and the bottom half is sent to the second graphics card. Supertiling does not support OpenGL and it doesn't give optimal performance all the time.
Scissor Frame Rendering (SFR)
Scissor frame rendering or SFR is very similar to the original method of SLI and Nvidia's Split Frame Rendering. SFR is the basic cut the screen in half and send half to the one graphics card and half to the other. There are some differences however between the original voodoo 2 SLI, Nvidia's Split Frame Rendering and ATI's Scissor Frame Rendering.
The original SLI simple cut the frame in half, no mathematics was required no calculations just cut in half and sent to two graphics cards. Nvidia's solution is the same but has a dynamic load calculation built in. It does not necessarily split the screen 50/50. It will calculate the load at the top of the frame and the bottom and split the frame accordingly.
ATI's solution is a little bit of both of the above. The screen is not always split 50/50 but its not calculated on the fly. This can save some calculations free up more clock cycles but can cause a slight drop from optimal efficiency. The load is set by the application / game and that value is fixed throughout the session. So if the screen is split 60 / 40 then that is how it will stay. The main advantage of the ATI solution here is that it can split the screen horizontally like the others, but also vertically. If splitting the screen vertically would give better results
Source
Now SLi :
The graphics memory is NOT doubled in SLI mode. If you have two 128MB graphics cards, you do NOT have an effective 256MB. Most of the games operate in AFR (alternate frame rendering). The first card renders one full frame and then the next card renders the next frame and so on. If you picture SLI working in this way it is easy to see that each frame only has 128MB of memory to work with.
If you feel you need 256MB or even 512MB of graphics memory for certain games, then you need to make sure that each card has that much memory. Also keep in mind that if you mix a 128MB GeForce 6600GT with a 256MB GeForce 6600GT that BOTH cards will operate with 128MB of graphics memory (the lowest common memory size of the two cards).
Source
skeletor
Chosen of the Omnissiah
PowerColor Radeon HD6950 2GB PCI-E 2.1 Graphics Card @ 17.5k.
I suggest you to move out, look around in shops and get real quotes like my friend did. The site you quoted has everything overpriced by around 2-4k.
pauldmps
Banned
Radeon HD 6950 Unlocks to HD 6970
Source: Techtree.com India > News > Hardware > Radeon HD 6950 Unlocks to HD 6970
Tutorial: AMD Radeon HD 6950 to HD 6970 Mod | techPowerUp
Those who settled for the Cayman Pro (HD 6950) over its more expensive cousin will be delighted to hear that it can be easily, and so far reliably, unlocked into the more powerful Cayman XT (HD 6970). Coincidentally, the Cayman Pro and XT are essentially the same GPU die with 128 shaders disabled. Fortunately for us, AMD hasn't locked the shaders on the silicon level, but simply by reconfiguring the VGA BIOS to disable the same. All it takes is a simple BIOS flash to unlock the shaders and transform your cheaper HD 6950 into a full fledged HD 6970.
Techpowerup.com has a detailed guide replete with all the dope you need to unlock the Cayman XT. Recovery is easy with the Caymans, thanks to a recovery BIOS feature. Just make sure you back up your existing BIOS before you meddle with it. The unlock process is reported to have been successful across AMD engineering samples, as well as on the HIS and Asus retail versions of the card, which again are one and the same.
Source: Techtree.com India > News > Hardware > Radeon HD 6950 Unlocks to HD 6970
Tutorial: AMD Radeon HD 6950 to HD 6970 Mod | techPowerUp
Those who settled for the Cayman Pro (HD 6950) over its more expensive cousin will be delighted to hear that it can be easily, and so far reliably, unlocked into the more powerful Cayman XT (HD 6970). Coincidentally, the Cayman Pro and XT are essentially the same GPU die with 128 shaders disabled. Fortunately for us, AMD hasn't locked the shaders on the silicon level, but simply by reconfiguring the VGA BIOS to disable the same. All it takes is a simple BIOS flash to unlock the shaders and transform your cheaper HD 6950 into a full fledged HD 6970.
Techpowerup.com has a detailed guide replete with all the dope you need to unlock the Cayman XT. Recovery is easy with the Caymans, thanks to a recovery BIOS feature. Just make sure you back up your existing BIOS before you meddle with it. The unlock process is reported to have been successful across AMD engineering samples, as well as on the HIS and Asus retail versions of the card, which again are one and the same.