Problems with my M17x R3.....

Ever since I bought my Alienware M17x R3 I have had nothing but issues with the game play.

I get 120+ FPS for a minute or two and then the FPS drops down to 15-20. This happens with every game I play (TF2, CS:GO, Crysis 2).

I am getting very desperate here. I've had this laptop for a little longer than a year and have tried what seems like everything. I have opened it up and cleaned it, re pasted the video card 3 times now, tried a multitude of drivers and I always have the same effect. My GPU temperature never goes above 71 but the clock speed drops on its own for some reason. Screenshot attached.

Is there anyone out there that can help?

 

sounds like the 580m powerthrottle maybe? Have you tried the fix for that?

 

Looks like he already has tried the bios fix from the screenshot with P1 showing 620?

Have you installing msiafterburner and setting up the on-screen-display to see exactly what is going on during gaming? The clock rate should identify what 'P' state the card is in. You can drop P1 down a few cycles to see when that cuts in. Also temps can drop very quickly when you switch out - it does sound a little high, my 580 at standard clocks never went above 68C and run most games around 62c-64c.
The P12 state in the screenshot is normal for power save in windows, you can stop that (but no real need to) by setting the card to max performance in the driver - reboot required.
HTH

 

ive heard of similar issues and it was what mickyD said...

 

OP, can you list your storage subsystem stats. I've had a similar issue and tracked it down to HDD settings and RAID controller inability to adjust the drives default settings. See this thread for more details:
http://forum.notebookreview.com/ali...are-m17x-r3-fps-issues-not-overheating-3.html

 

Check out this article, it might help a little

 

I'm interested to hear...What temps do you get IN furmark?

That should have been updated with the information that DELL acknowledged there was a flaw in the original 580M heatsink design and actually redesigned it with later shipments and that ChrisM specifically says in another thread that if you run a GPU stress-test application and it hits 78c before 2 minutes that the GPU and/or Heatsink may be faulty and should be replaced if under warranty.

 

First time I tried furmark. Got 68C max at 1920x1080. All other settings at default. It does cause the P1 throttle at points during the run so I upped it to 620 to match the base clock speed. Although the voltage still drops it seems stable.