XFX 7970 GPU core osculating?

[BogusJoe]

Got two of the FX-797G-TDFC and they are defective when the cards are ran at 100% GPU usage. Stock settings GPU @ 1050mhz, Memory @ 1500mhz, and Power control setting @ 0%. It does not matter whether the cards are in CrossFireX enabled mode or Disabled mode. Have also tried this with (both cards) just a single card in the system at a time. Here is what happens. The card (s) runs great until in hits 100%. At the moment of 100% the GPU core clock starts osculating from 537mhz to 1000mhz. Here is a list of programs I have used and it happens every time it hits 100% GPU usage:

1. Heaven 3.0
2. 3DMark 11
3. 3D Vantage
4. Windows Experience
5. Crysis 1
6. Crysis 2
7. Battle Field 3
8. PC Mark 7
9. FurMark 1.10.4

FurMark (Benchmark preset: 1080) is the quickest program to run to produce the same results. I first noticed this when running GPUShark v0.6.7 to monitor the card (s). Also used GPUZ v0.6.7 and logged to file to verify the results. Then loaded MSI AfterBurner to monitor and log the results and this is where the numbers from 537mhz to 1000mhz showed up. The core clock is very erratic once the GPU hit 100% and the VDDC was a constant 1.213v.

At first I thought it might be a driver issue. Started with Catalyst 13.1, then moved down to Cat 12.10, then down to 12.8 Everytime I uninstalled a driver, I ran the AMD Cleanup util 1.2.1.0 and Driver Sweeper and then ran CCleaner to remove any leftover registry entries.

System information:
Windows 7 64bit Ultimate with latest patches and updates
Mother Board: Asus Rampage IV Extreme Bios v3301
Processor: Intel I7 3930K
Ram: 32 gig DDR3 2134 Gskill
Power Supply: SilverStone Olympus 1500
VGA Card 1: in slot 1 or BusID 1 FX-79G-TDFC
VGA Card 2: in Slot 3 or BusID 3 FX-79G-TDFC

Has anyone else ran into this situation? Once a GPU hits 100% usage does it clock itself down?

 

[bigshootr8]

This is a very odd issue to be experiencing I'm curious what others may say. I read this and I don't have a answer but I'm curious what the people with the 7970 think.

 

[nitrium]

Thermal throttling maybe? What sort of temps you recording when it hits the 100% load?

 

[BogusJoe]

Thermal throttling maybe? What sort of temps you recording when it hits the 100% load?

Thanks for asking. Forgot to mention that. The temps are 55c and the MaxTemp is 63c. I have yet to see anything over 70c.

 

[Sakkura]

Well if it's just happening in Furmark and the like, it could just be the driver reacting to the stress test and throttling down. But if it happens in games, I don't know...

 

[Supernova1138]

I had this issue with my 7870 on the Catalyst 12.7 Beta driver. Are you running any web browsers in the background while you are playing/stress testing the card? Running a web browser with GPU acceleration enabled would cause my card to not stay at maximum clocks. Disabling GPU acceleration, or not running my web browser in the background helped with this issue.

Another possible fix for you would be to increase your core clock speed by 2MHz. For some strange reason, that would also fix this problem while I was using the 12.7 Beta driver. The issue ultimately disappeared entirely for me after moving up to Catalyst 12.8.

My guess is something in 13.1 has caused this issue to return for some users, and it is somehow remaining, even after trying to purge the drivers. If you want a more permanent solution aside from the two workarounds I posted above, you may have to wait for a newer driver revision to fix this, or possibly consider reinstalling Windows.

 

[BogusJoe]

Yes, I have tried to move the core up. Tried 1100mhz and 1130mhz and it did the same thing. I did not try just a 2mhz increase. I did not have a web browser open. Did not try any other drivers except 12.8, 12.10, and 13.1 I thought maybe it was a driver issue too. Getting ready to do a reload of Windows tomorrow. Will let you know of the outcome.

 

[helper800]

The GPU manufacturers have made it so that when you try to run Furmark it detects it and throttles your GPU's core clock regardless of what setting you run Furmark at. Furmark is extremely stressful on GPU's and that's why the manufactures have done this. I had similar problems with my XFX DD 7950 and have already made a thread on the Furmark issue though it was a while ago.

 

[nitrium]

I know that on AMD cards with PowerTune if in any way the GPU's TDP is exceeded, throttling automatically kicks in. It is built in at the global level. I now strongly suspect this is what's happening with your card. It is not a defect, but a feature! ;-).
http://www.amd.com/uk/Documents/PowerTune_Technology_Whitepaper.pdf
http://www.anandtech.com/show/6025/radeon-hd-7970-ghz-edition-review-catching-up-to-gtx-680/2

 

[bigshootr8]

The GPU manufacturers have made it so that when you try to run Furmark it detects it and throttles your GPU's core clock regardless of what setting you run Furmark at. Furmark is extremely stressful on GPU's and that's why the manufactures have done this. I had similar problems with my XFX DD 7950 and have already made a thread on the Furmark issue though it was a while ago.

If that is the case then that is the issue right there. I can understand a card doing that with furmark to prevent it from being damaged.

 

[Imago2]

The hd 7000-series seem to adjust the core clock automatically when there is a lack of providing voltage to keep it at a current stable level. I noticed it when I was experiementing on my voltage controls. Maybe your chip (unlike others) need some more voltage? It wouldn´t be the first time manufactures compile OC-profiles without paying attention on a stable voltage level

 

[BogusJoe]

I have been in contact with XFX since Jan 27th. Here is the last bit of information from their support.

[KEVIN_C 1/31/2013 7:32:12 PM] Hello we have actually seen that normally with powerful cards, usually with powerful benchmark settings, think of it as a vehicle rev limiter when the card gets to 100% it will fluctuate the clock speeds so as not to burn itself out. My question to you is does this ever happen in a normal game? It is not going to want to get to 100% and stay at 100% power so it tries to alleviate itself but fluctuating constantly when it gets that high.

My reply:
[ 2/2/2013 9:51:30 PM] Yes, it does this in game too as indicated in the first part of this ticket. It also does this at 900mhz. I will have to check into your "so-called" Rev Limiter. So, you are saying if I run these cards at 900mhz @ 100% during gaming, it will burn these cards out, even with the GPU temp at 70c. I paid $900.00 for a set of High Performance cards from XFX only to be limited from running at 100% and clock itself down below the advertised 1050 with Power Boost? I know this is a new card and new design, but I have cards from XFX from the 4870X2 (had problems and returned with 5870 replacements), to the 6970 and was able to run @ 100% for hours at a time without problems or the core clock osculating like it has on these cards.


Guys, I don't think I am wrong here. 900mhz, below the factory stated Gigahertz Edition 1050 and the core is osculating wildly as soon as it hits 100% gpu usage and the temps are well within limits! Anyway, returning the cards and got a pair of EVGA 680 Classified coming in. Thank you for replying to this post. I would like others to be aware of the (as I call it) False Advertisement.