Radeon HD 7950 Owner's Thread

[Eureka]

What was your original BIOS? Do you have a voltage locked card?

Mine was voltage locked, came with FX0 and wouldn't take any other BIOS.

 

[Belkov]

So you acknowledge my words - the new bioses would work flawlessly with revision 2 GB 7950 WF3.

 

[Deaks2]

Mine was voltage locked, came with FX0 and wouldn't take any other BIOS.

Interesting... My cards are voltage locked cards that came with FX0 too. I successfully ran FZ1, FY1 and F43. I am about to try FX1.

 

[Eureka]

Interesting... My cards are voltage locked cards that came with FX0 too. I successfully ran FZ1, FY1 and F43. I am about to try FX1.

What did you use to flash them?

 

[ArsenalGear]

I assume the card performs well in games? What is your Valley score on HD Extreme?

Wow...you know what I just tried? Changing the "Display Scaling Mode" in 3DMark from Centered to Stretched. Result? 1500pt jump.

8042 - http://www.3dmark.com/3dm11/6287180

I don't know if that's high enough, but I can't explain how that setting alters the score.

 

[WaTaGuMp]

Mine was voltage locked, came with FX0 and wouldn't take any other BIOS.

My card is better than yours, oh wait. :biggrin:

 

[Durvelle27]

Wow...you know what I just tried? Changing the "Display Scaling Mode" in 3DMark from Centered to Stretched. Result? 1500pt jump.

8042 - http://www.3dmark.com/3dm11/6287180

I don't know if that's high enough, but I can't explain how that setting alters the score.

weird my card scored higher

http://www.3dmark.com/3dm11/6217429

 

[Deaks2]

What did you use to flash them?

atiflash in a DOS environment.

FX1 locks up as soon as my PC puts a load on the cards. I am sticking with F43.

 

[ArsenalGear]

Wow...you know what I just tried? Changing the "Display Scaling Mode" in 3DMark from Centered to Stretched. Result? 1500pt jump.

8042 - http://www.3dmark.com/3dm11/6287180

I don't know if that's high enough, but I can't explain how that setting alters the score.

Well, I got another new score. This one jumped over 1000pts by simply raising the Catalyst "Power Control Settings" from 0 to 20%.

9157 - http://www.3dmark.com/3dm11/6288348

Makes me annoyed what other things I'm missing...

 

[3DVagabond]

Well, I got another new score. This one jumped over 1000pts by simply raising the Catalyst "Power Control Settings" from 0 to 20%.

9157 - http://www.3dmark.com/3dm11/6288348

Makes me annoyed what other things I'm missing...

Best I can figure is the power control setting is there simply to improve power consumption measurements. Everyone should just set it to +20%. It's not putting any additional power to the card, it's only allowing it to draw more if needed.

 

[Eureka]

atiflash in a DOS environment.

FX1 locks up as soon as my PC puts a load on the cards. I am sticking with F43.

I tried ATIFlash and it bricked my card whenever I tried F43 or F0. I could only reflash it back to FX0 using the other switch's BIOS.

 

[Zargon]

Best I can figure is the power control setting is there simply to improve power consumption measurements. Everyone should just set it to +20%. It's not putting any additional power to the card, it's only allowing it to draw more if needed.

It allows you to change the target TDP of the card

 

[OS]

I got one of those sapphire 7950 deals, mine reported a 49% ASIC quality, something stupid low.
It still overclocks ok, it does 1100/1550 with just catalyst driver, no voltage adjustments.
Still an awesome product. I had a 7870 also and the 7 series is great, esp since the last couple generations were more incremental upgrades.
I last had crossfire 6950s and I think a single heavily overclocked 7950 is almost as fast as stock dual 6950s.

 

[Durvelle27]

I got one of those sapphire 7950 deals, mine reported a 49% ASIC quality, something stupid low.
It still overclocks ok, it does 1100/1550 with just catalyst driver, no voltage adjustments.
Still an awesome product. I had a 7870 also and the 7 series is great, esp since the last couple generations were more incremental upgrades.
I last had crossfire 6950s and I think a single heavily overclocked 7950 is almost as fast as stock dual 6950s.

My card reports 81.5% ASIC Quality

 

[Skurge]

Has anyone seen reviews of an overclocked 7950 vs a stock 6990 at 1440/1600p. I'm thinking side grading. I don't want a small downgrade, but not too much.

 

[Nomercy1]

Need some help with this one guys. Have a 7950 sapphire dual-X. Loved the card when i got it overclocked nicely 1150 - 1400 mem . Now its been running smooth at these speeds for 4 - 5 months now and i did a lot of benching during the overclock process. And i have never changed off stock voltage

Now i decided to see today what my benches are after not looking for a while and.... WTF

Old score heaven 3.0 maxxed 1080p 45 fps average
New score heaven 3.0 maxxed 1080p 38.2 FPS


Ok little weird got Heaven 4.0 and new Futuremark.

heaven 4.0 Ultra 1080p maxxed. 24.6 FPS 620 score

3dmark Firestrike 3348 Graphics score.

Now this is with the new 13.3 beta 3 drivers. Obviously something is wrong, i went to stock speeds and get almost the same results. The GPU is not throttling, power is maxxed its just scoring terribly. No hardware Changes have been made to the PC. The temps max out at 68c while benching. mem temps 62c while benching according to gpu-z.


Tried rolling back the drivers and saw some improvment but i dont understand whats happening.

 

[3DVagabond]

Has anyone seen reviews of an overclocked 7950 vs a stock 6990 at 1440/1600p. I'm thinking side grading. I don't want a small downgrade, but not too much.

Overall, the 7970GHz runs pretty close to neck and neck with the 6990. A decent O/C'ing 7950 can pass stock 7970GHz. Of course crossfire performance isn't as consistent as single gpu. That means it's going to go back and forth with games that scale well and those that don't. I would say that overall the gaming experience is going to be better with an O/C'd 7950, but some games are going to be better on the 6990. The best thing to do is compare performance in reviews of the 7970GHz with what you get with your 6990 in the games you play. A bit of work, but that's how I would do it.

 

[WaTaGuMp]

Looks like I got a bad 7950. Took back the Gigabyte for another one. This one wont overclock at all. So ordered a Sapphire. Gonna end up being cheaper and hopefully better.

 

[czglory]

just got a new system with his ICEQ 7950 and got artifacts immediately upon booting to the bios. The card was working before it was shipped out... this is my first AMD card and my experience has not been satisfactory so far

 

[Nomercy1]

Little update from my previous post. After many many hours of troubleshooting i have come to the following results with my sapphire dual-x

1) anything over 1075mhz Core and the GPU will begin to throttle while benchmarking. This is not a result of the temp of the card ( which now maxes at 60-61c ) It never use to do this

2) the memory cannot be overclocked at all. it will run @ 1300mhz but dropping it back down to 1250 nets me a 2fps gain in benchmarks. I have tried increases the voltage on the memory with no results. The mem temps are low as well. never gone over 63c. I also tried to run the memory at 1200mhz underclock and it just crashed everytime.


Overall im disappointed with this slight overclock but it is what it is. I am a bit puzzled that the gpu throttles after 1075 now when i would run it at 1125 before w/o any issues.

I always knew the memory didnt overclock well as it would always crash at anything over 1300mhz.


So overall im not sure if its just the new benchmarking software ( heaven 4.0/valley/3dmark fire ) That is exposing a weak overclock or what but definelty bummed out that this is all i can get my card to do now.

 

[Belkov]

Did you change your bios? It is reaaly weard. May be the problem is in your PSU.

 

[psolord]

I made two new videos of the scene in Crysis 3 Only Human level, where Prophet rides the Vtol. I chose this scene because it's automated up to the point I recorded, so it would be the same for both recordings. Also there's a lot of camera panning, which is the worse motion for the viewer, if there's stuttering.


Here's the video of the 7950 crossfire. This is how I play and this is what I see. 60fps vsynced. Actually it's better in real life because Crysis 3 is very cpu heavy and Fraps needs a lot of cpu as well so its difficult to keep a steady 60fps, but still it's good enough.

This is the video of the single 7950 at the same settings. I only disabled vsync for this one, otherwise I would be getting a steady 30fps, thus artificially making my point even stronger on a difference that is already huge.


What I see here, is two cards running the game at perfect smooth 60fps and still have some spare cycles, while the single card is struggling to hit 40fps at full load. (<-this hints at greater fan speed, thus more noise, another benefit of dual gpu)


Call me crazy, but this is what dual gpu is all about. Effectively almost doubling your processing power pool, so you can use it for a smoother framerate and overall better gaming experience. I'd suggest to people that are quick to dismiss dual gpu systems, to actually sit in front of one and do some real life testing, because I do, for more than 5 years now (voodoo2 sli doesn't count I guess) and I have mostly faced situations like the one described above.

I don't defend AMD here, but dual gpu solutions. Both my secondary system with 570 SLI and my tertiary with 5850 CFX, still exhibit the same behavior, alas with lower settings.

Actually my older 5850 CFX system would be far more suitable for comparisons, since it's much more gpu power starved and the second card can do wonders, but it lacks the cpu power to provide good fraps recordings.

This post is aimed at people that are preaching single gpu powa at any cost and as a counter measure to recent crossfire frametime findings. I know that me using vsync alleviates most or all of these issues, but this is how I believe proper gaming must be done.

It is also a good example of what 60fps gaming vs less than 60fps gaming looks like.

 

[Nomercy1]

Did you change your bios? It is reaaly weard. May be the problem is in your PSU.

If your reffering to me i have been thinking it may have something to do with it. i have more than enough juice unless its dieing which would be very odd. And no, no bios changes have been made

 

[Belkov]

If your reffering to me i have been thinking it may have something to do with it. i have more than enough juice unless its dieing which would be very odd. And no, no bios changes have been made

You can check whether the voltage shown with the software using by you is real - try to connect a multicet - you can google how it works, because i can't explain it very well in English. Basicaly you must put the pin of the multicet in one ot psu connectors, but with some settings on the multicet - this i can't explain.

 

[Belkov]

I made two new videos of the scene in Crysis 3 Only Human level, where Prophet rides the Vtol. I chose this scene because it's automated up to the point I recorded, so it would be the same for both recordings. Also there's a lot of camera panning, which is the worse motion for the viewer, if there's stuttering.


Here's the video of the 7950 crossfire. This is how I play and this is what I see. 60fps vsynced. Actually it's better in real life because Crysis 3 is very cpu heavy and Fraps needs a lot of cpu as well so its difficult to keep a steady 60fps, but still it's good enough.

This is the video of the single 7950 at the same settings. I only disabled vsync for this one, otherwise I would be getting a steady 30fps, thus artificially making my point even stronger on a difference that is already huge.


What I see here, is two cards running the game at perfect smooth 60fps and still have some spare cycles, while the single card is struggling to hit 40fps at full load. (<-this hints at greater fan speed, thus more noise, another benefit of dual gpu)


Call me crazy, but this is what dual gpu is all about. Effectively almost doubling your processing power pool, so you can use it for a smoother framerate and overall better gaming experience. I'd suggest to people that are quick to dismiss dual gpu systems, to actually sit in front of one and do some real life testing, because I do, for more than 5 years now (voodoo2 sli doesn't count I guess) and I have mostly faced situations like the one described above.

I don't defend AMD here, but dual gpu solutions. Both my secondary system with 570 SLI and my tertiary with 5850 CFX, still exhibit the same behavior, alas with lower settings.

Actually my older 5850 CFX system would be far more suitable for comparisons, since it's much more gpu power starved and the second card can do wonders, but it lacks the cpu power to provide good fraps recordings.

This post is aimed at people that are preaching single gpu powa at any cost and as a counter measure to recent crossfire frametime findings. I know that me using vsync alleviates most or all of these issues, but this is how I believe proper gaming must be done.

It is also a good example of what 60fps gaming vs less than 60fps gaming looks like.

Well, for now, with only 2-3 games so demanding, i don't think it's a worthy deal to grab 2 or more 7950s. May be in the future, but not now - for example i don't like crysis, far cry and tomb rider - i don't intend playing them, so what is the point for me to grab two cards - all depending of the customer preferences...