Jump to content
LaptopVideo2Go Forums

Global Uneven Framerates


titan818

Recommended Posts

Hey,

I've been getting uneven framerates in quite a few different games. Basically the framerate pulses up and down about 15-20fps about every second or two. This generally means the difference btwn a smooth 45 and a seizure-inducing 25. Upon start, (into 3d) the games run smooth, high fps for a minute or two, then they start the pulsing. The games I see it the most (all the time) in are COD-WaW and Left4Dead. I tried disabling multi-core rendering in both, but the problem persists. Up till today I was running DOXs 180.84s, but I've switched to 181.00. This problem occurs in both. Does anyone have any suggestions? Thanks

Toshiba X205-Sli3

Intel Core2Duo T7500 2.2GHz

GeForce 8600M GT Sli

181.00

Link to comment
Share on other sites

have you disabled sli? this is probably the cause. i get this sometimes. especially when testing drivers. umm try 179.28

Link to comment
Share on other sites

have you disabled sli? this is probably the cause. i get this sometimes. especially when testing drivers. umm try 179.28

Sorry for the delay, on 179.28 now, but problem is still occurring. Will try playing with SLi now.

Link to comment
Share on other sites

Sorry for the delay, on 179.28 now, but problem is still occurring. Will try playing with SLi now.

Disabled Sli and got basically what I expected- Lousy performance. (about half the previous framerate) However, the oscillation seemed to stop.

Link to comment
Share on other sites

Disabled Sli and got basically what I expected- Lousy performance. (about half the previous framerate) However, the oscillation seemed to stop.

Recently noticed that running Cod:Waw in windowed got rid of the uneven problem. (SLi switched on as well) Performance suffered severely though. Not a solution, but possibly a clue?

Edited by titan818
Link to comment
Share on other sites

Recently noticed that running Cod:Waw in windowed got rid of the uneven problem. (SLi switched on as well) Performance suffered severely though. Not a solution, but possibly a clue?

Bump?

Link to comment
Share on other sites

sorry for taking so long. xmass and all.

yeah sounds like SLi and driver performance issue. windowed doesnt take as much CPU, but take a lot more GPU power. I suspect trial and error using other driver releases. the new 180.** may fix this as they are made for SLi units.

Link to comment
Share on other sites

Will try working with the 180 series. I do remember some of them not working though. To rule out driver install/uninstall problems (and because I needed it anyway) I formatted and reinstalled Vista on my comp. Everythings so nice and clean, but the problem is still goin on. The one driver that did work was the last release from my OEM (176.05, like 4 months ago), but of course that gave pretty horrendous performance. In the meantime, I did make a pretty framerate plot using Fraps in Left4Dead. This is on the LV2Go 179.28. I tried NVidias as well. Anyway, in the plot you can see all goes well for awhile, then the spikes start.

post-12694-1230434972_thumb.jpg

Edited by Claw
Full quotation.
Link to comment
Share on other sites

What about running RivaTuner on the background with monitoring so you can see the clocks and temperatures during "normal" behaviour and spiking afterwards.

Link to comment
Share on other sites

Well now I've tried 180.88, Dox 180.84 and 181.00. Dox 180.84 and 180.88 got rid of the problem but gave pretty lousy performance, hovering around 30-35FPS. 181.00 had the problem but performance was awesome. (Up around 50-60FPS) I logged RivaTuner during these tests and didn't see any real fluctuation in any spec. However, during the 181.00 test, both of my cores got pretty near toasty. Core0 got to 83 which isnt too bad, but Core1 got up to around 96. Thoughts?

Link to comment
Share on other sites

That's within the specs - barely. C2D max temp is 100 C. With RMClock you could lower the voltages a little if you want to be safer.

Link to comment
Share on other sites

That's within the specs - barely. C2D max temp is 100 C. With RMClock you could lower the voltages a little if you want to be safer.

Fair enough, but could the high temp be a cause of the performance problem?

Link to comment
Share on other sites

Hey I thought I found the miracle driver at 180.70. But I tested again with the exact same conditions and got pretty different results. Core temps same as last post and clocks all normal. I can't figure this crap out.

post-12694-1230579217_thumb.jpg

Link to comment
Share on other sites

Fair enough, but could the high temp be a cause of the performance problem?

It might, you could do some math benchmarks and monitor the temps and check if the results get worse with raising temperatures. I could see my old lappie to clock the CPU down at about 78 C and the benchmark got 20 -30% worse - it was probably SiSoft Sandra benchmark AFAIR.

Link to comment
Share on other sites

It might, you could do some math benchmarks and monitor the temps and check if the results get worse with raising temperatures. I could see my old lappie to clock the CPU down at about 78 C and the benchmark got 20 -30% worse - it was probably SiSoft Sandra benchmark AFAIR.

Never seen this software before. Which specific benchmark should I look at and run?

Link to comment
Share on other sites

Processor Arithmetics, temperatures check with Speedfan graphs; it hogs almost 100% CPU so RivaTuner graphs didn't react.

Got about 70C with a brief test. There must be loads of test software available, do a search. PassMark is one that pops into my mind.

Link to comment
Share on other sites

Processor Arithmetics, temperatures check with Speedfan graphs; it hogs almost 100% CPU so RivaTuner graphs didn't react.

Got about 70C with a brief test. There must be loads of test software available, do a search. PassMark is one that pops into my mind.

Used Sandra- CPU Arithmetic, and the highest I see the cores get to is 74C and 73C. However I monitored CPU temp during left4dead and that got them up to about 81C and 78C.

Link to comment
Share on other sites

Probably ambient temperature around the chip is higher in the casing when all components are stressed.

Link to comment
Share on other sites

Probably ambient temperature around the chip is higher in the casing when all components are stressed.

Are these temperatures high enough to be a problem?

Link to comment
Share on other sites

If anyones interested 185.20 doesnt solve it either.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...