• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Massive, overnight drop in performance...AMD295X2 - very weird!

AndyC_88

AndyC_88

AndyC_88

Associate
Joined
6 Jun 2016
Posts
3
Dear All,

Very curious problem here that I can't seem to figure out. I'd like to think I'm relatively IT-literate but this one has me absolutely stumped!

Background / System set up:

I'm currently running an i7 5630X, 16Gb DDR4, Corsair HX1000i and a Radeon 295X2 on an ASUS Rampage V Motherboard. Since I built it last year, it's been running consistently well, games are more than playable and I've been getting around 5.5k on FireStrike Ultra which seems about right for this set-up.

Problem:

Now, I moved my PC yesterday to a different location, and when I went to play FC4 (usual 80fps+) it was unplayable at about 15-20fps which I thought was very odd. I did a quick benchmark on 3dmark and I scored 1,700 - which is nearly 4,000 lower than my previous score...

Now, here's the problem...why would this be the case? I have tried the following things to no avail:

re-install latest drivers
re-seat the graphics card
check CF is enabled on MB
check CF is enabled in CCC
stress test to check temps (all ok)
turn off UPLS

The way it "feels" and the only way I can describe it is it never seems to get to full power...or that CF isn't truly enabled. If you take a look at the image below, GPU jumps to 100% then drops immediately...

AMD.jpg


So - any suggestions or comments at all would be greatly appreciated!

Kind regards,

Andy
 
Lokken86,

Thanks for replying. That was my thought - however the temps aren't getting above 50 degrees when using Furmark....which almost seems too low? I don't know how to explain it...it's almost like one of the GPU cores isn't working / functioning. Surely if this was the case the card just wouldn't work, though?

Anymore thoughts would be greatly appreciated!
 
I would reseat the GPU, while checking all connection due to moving it. Also check a different power socket. If the PSU has a voltage switch 230/110 ensure that it has not been moved.
 
Check the PCI-E speed and version it is running at, remember to stress it while checking to see where it sits at, under idle it can go to a slower PCI-E speed to save power, can be disabled in power options under the power profile you have set and the drivers for the card.
 
Sounds like a core might have died if the temps are at 50c and the frame rate has tanked. Load up afterburner and enter a game to see if both cores are going to full speeds.
 
AndyC_88 said:
Dear All,

Very curious problem here that I can't seem to figure out. I'd like to think I'm relatively IT-literate but this one has me absolutely stumped!

Background / System set up:

I'm currently running an i7 5630X, 16Gb DDR4, Corsair HX1000i and a Radeon 295X2 on an ASUS Rampage V Motherboard. Since I built it last year, it's been running consistently well, games are more than playable and I've been getting around 5.5k on FireStrike Ultra which seems about right for this set-up.

Problem:

Now, I moved my PC yesterday to a different location, and when I went to play FC4 (usual 80fps+) it was unplayable at about 15-20fps which I thought was very odd. I did a quick benchmark on 3dmark and I scored 1,700 - which is nearly 4,000 lower than my previous score...

Now, here's the problem...why would this be the case? I have tried the following things to no avail:

re-install latest drivers
re-seat the graphics card
check CF is enabled on MB
check CF is enabled in CCC
stress test to check temps (all ok)
turn off UPLS

The way it "feels" and the only way I can describe it is it never seems to get to full power...or that CF isn't truly enabled. If you take a look at the image below, GPU jumps to 100% then drops immediately...

So - any suggestions or comments at all would be greatly appreciated!

Kind regards,

Andy
Click to expand...
Uninstall MSI AB etc. Uninstall the drivers. Use DDU to wipe everything related to drivers. Use CCLeaner to clean what ever the MSI AB left behind.

Install the latest Crimson. DO NOT install MSI AB or any other 3rd party software.

Try to see if the issue is resolved.

Check the CPU temps and usage also. Because if you caught a virus it will cripple your CPU performance and consequently the GPU.

And STOP installing Evolve. It has issues.
 
Last edited:
Panos said:
:rolleyes:

Overclocked at 1100/1625 runs fine bellow 50C mark, if you replace the fan with 2 Apaches, and the case doesn't have 30C inside it, but room temp.
Click to expand...

How is replacing fans a 'stock' cooler?

I had a 295x2 as you well know, and mine was always throttling even when I replaced the fan with Corsair SP120's in push->pull
 
How about the low state power bug? There was a bug that had the gpu running at 300 mhz to save power and it kept it at that even if you was gaming at the time.

Go into the crimson software and go to Gaming/global settings and turn off power efficiency if it's on and see what happens
 
VincentHanna said:
How is replacing fans a 'stock' cooler?

I had a 295x2 as you well know, and mine was always throttling even when I replaced the fan with Corsair SP120's in push->pull
Click to expand...

Yeah I've tried mine with the stock fan, 2 x Apaches and 2 x SP120's and even at stock clocks, undervolted by 40mv some games can still cause throttling after a while. I'm running in a Phanteks Enthoo Pro case with 3x Akasa Viper 140mm intake fans so airflow really shouldn't be an issue either.
 
UPDATE: Thanks all very much for your suggestions - a complete wipe / re-install of drivers worked. Previously I had simply re-installed the drivers, the complete wipe seemed to do the trick. Seems odd though - I didn't change any settings or software, only the physical location of the PC...

After having a play with the settings, the only thing that comes close to replicating the problem is setting the desired temps to around 40 degrees so the card throttles performance immediately. (Even though this wasn't the case - before I re-installed the drivers they were set to the default 75)

So...seems a bit weird but I'll just chalk it down to one of lifes little mysteries...

Again, thanks all for your help.
 
You must log in or register to reply here.
Back
Top Bottom