P-####FX oveheating problems explained and solved

Okay, I have figured out why P series computers overheat so damned often.... there is no 'lock' on the MAXIMUM overclocking of the GPU that the driver does AUTOMATICALLY unless you are running RivaTuner and have it set to only allow a maximum overclock on the thing.

My GPU was running at 600MHZ Core ROP, 1500 Core shader, and 602MHZ memory clock... it's no damned wonder why the thing was overheating so damned bad, to the point of a hard lockup!

To solve this problem, the only thing that you can do is run RivaTuner and have 'constant performance' turned on OR have it set for a maximum driver overclock of the DEFAULT memory, clock, etc. settings in RivaTuner.

I have reported this to NVidia, who have said that it is a known issue with ALL laptop graphics cards, which they are trying to fix at the moment.

They have no 'special' drivers for laptops themselves, so they are trying at the moment to detect that it's a M-class GPU (mobile class) and dial down the maximum overclock to compensate and keep the thing in a small thermal envelope of, at most, 70 Celsius.

 

what are you smoking?

 

What are you trying to say?
I have done insane clocks as in my sig and also, no overheating issues have surfaced under 24/7 use on 635/950/1625. It boils down to HOW you maintain your notebook.

 

I second this, he probably didn't get to the chopper...

He probably needs to clean his vents.

 

MY P7811FX has the Hard lockup issue and I have replaced the thermal paste and cleaned the vents & heatsinks.
So this problem does not only happen to people who don't clean their computers.

 

This is not a personal attack at you, but I am disturbed by many things you just posted.

#1) Why have you made this thread? I don't know if you've noticed, but the ''overheating'' theory has been debunked over a year ago. Why the machine locks up, no one still knows. It has been singled out that the GPU is most likely the reason though. I've had my P-7811FX lock up at 62*C, which is not even considered "hot"!

#2) Graphics cards don't overclock by themselves. NEVER NEVER NEVER. They will underclock to conserve battery life, lessen heat output when not playing games, but they will never overclock themselves, unless you find a way to modify the vBIOS and that's not likely. Those clocks you posted, ARE the m9800GTS' clocks. They were manufactured to be that way, it is not a coincidence that they are that "high". It's a gaming notebook with a high-performance graphics card.

#3) We had a very helpful fellow by the name of nomios, who discovered the Rivatuner, "Force constant 3D performance" well over 2 months ago. His post was here. He even helped assist you with your issue in this thread. Please give credit, where credit is due.

#4) I take most of you say with a grain of salt. Nvidia never even answered to the public how they ended up manufacturing faulty GPU's, which one's were defective and how many GPU's were affected by this. It's always helpful to email the company you have an issue with, but honestly, they just don't care; it's really that sad. If you had any links to prove that they are actually attempting to limit the GPU from exceeding 70*C, I'd LOVE to see it.

Honestly, the first paragraph is just ugh. . . In other words, your explanation is just horrid, why not just link the post that nomios linked to you? Also, the solution is vague and completely ripped off of nomios' post.

 

I will start looking into this issue in detail after work today. I'm rather sick of the computer locking up 3 or 4 times a day, usually in conjunction with playing WOW. It's good to hear that it's a common problem, and not just my hardware dying. No, my vents are not clogged, the computer is only a month old, and the problem started almost immediately. The only thing I have done that could be affecting the issue is disable powermizer, but I don't have much of a choice there because PM causes dpc latency to go through the roof when running multimedia. I'll start by monitoring my temps, but I don't think it is overheating, because the laptop is not particularly warm to the touch, and the randomness of when it crashes does not imply that it is more apt to crash once it has warmed up. I'll also see if running it with PM on makes a difference. WOW is about the only time I have the computer under full load, and there seems to be a direct correlation between it and the crashing. I have yet to have it crash without having run wow first, but it is not necessarily when the game is at peak load that it crashes, in fact, occasionally it does not lock up until the game has been closed for a few minutes. So, after watching the temps, and playing with powermizer for a bit, and upgrading any upgradable drivers, I'll probably give the fixes listed above a try. I will be back to update my progress. Nvidia is starting to seem more buggy than ATI nowadays.


*EDIT*


I am running a GTX 260M in the p-7915u

 

No, it isn't 'how you maintain your notebook'. I take my notebook TOTALLY APART and clean it very well every 3 months or so (used to be every year).... so there is no dust or anything in it to cause these problems.

The problem is that there is not the same amount of 'cooling potential' in a laptop as in a regular desktop PC, therefore some 'overclocks' are just too much in the small thermal window that these things have that they are able to cool the chips inside of them to do.

That extreme overclock that the drivers do automatically? Way too much, even NVidia themselves when I sent the e-mail to them about this said that it was WAY TOO MUCH of an overclock for a laptop PC...... unless you have gone the extra distance and WATERCOOLED THE LAPTOP or done something that the normal user does not have the experience to do.

Heck, they said even in a DESKTOP..... that overclock would be pushing the thermal envelope of the 9800 series processors without watercooling.

To xtyfyb, yes, what I posted WAS somewhat taken verbatim from that other person's post.... and I am sorry that I didn't give them credit.... but up until NOW, NVidia had not admitted that there was a known issue that was traced back to the drivers for the P-####FX series and 9800 series (maybe even the latest in Gateway and other's offerings) in general..... we now have that with the e-mails (I will see if I still have them and post them if I do, I have a bad habit of deleting e-mails right after they are no longer useful to my current needs) from NVidia.

Sad thing is that ATI/AMD graphics processors don't have this issue, because they have a SOFTWARE LOCK on the maximum overclock without overriding that lock and a 'maximum temperature' that the graphics processor is allowed to go to before automatically dialing back the overclock, even if you have set it higher. NVidia didn't include that in their drivers, which was an oversight of the HIGHEST CALIBER!

The last issue is that with that 'constant performance' turned on.... you lose a HECK OF A LOT of the graphics potential of your graphics processor. According to Futuremark and PCPitstop's testers, it is CUT MORE THAN IN HALF with that constant performance turned on! Which is not acceptable in the slighest.

 

I bow before your clocks.

I use a more comfortable 600/900/1475 for everyday-purpose gaming. My P-6860FX has never faltered on me. Love this piece of hardware.

 

My 7805u is set to 700/900/1700 for everyday use and havent experienced anything wrong at all.

you can bow down to me instead

 

I bow to him because he is a fellow P6860FX owner.

We cannot compete against the 7XXX series with the 9800.

 

Well, enabling powermizer just seemed to make things worse, and is a moot point anyway, since it causes entirely unacceptable dpcl issues when it is running. The gpu is running @ 550mhz. I have done no over clocking. Temps for the cpu and gpu seem to stay under 60 degrees C, so it really does not look like the computer is overheating, and I do not see any reason to under clock it. I reupdated all my Nvidia software last night, and it did not have any subsequent hardware locks, but did have a blue screen. I'll probably test and monitor for a little while and then try the Nomios fix if it is still locking up on me.

 

Really? Proper Maintenance of your laptop does not impact your temperatures? I should make note of this. /sarcasm

Perhaps you didn't put it back together very well. A lack of pressure/contact between your thermal pads and heatsinks could very well be your problem.

True, yet many p-78XX can hit 700mhz+ core/ 900mhz+ memory without a hiccup and still less than 70C temps.

It's not an "extreme overclock" if it is the standard clock for that gpu.

600 MHz core clock/1500 MHz shader clock/800 MHz memory clock.
600 MHz core clock/1500 MHz shader clock/800 MHz memory clock.

NVIDIA GeForce 9800 GT GPU Engine Specs:
CUDA Cores 112
Graphics Clock (MHz) 600 MHz
Processor Clock (MHz) 1500 MHz

Isn't that convenient for you?

Not if you use RivaTuner for ATI overclocking.

RivaTuner is not the best overclocking software for mobile nVidia cards. Nvidia Performance tools is designed for these cards. NVIDIA System Tools with ESA Support

 

Well I guess this thread is heading nowhere.

Provide evidence to substantiate your claims pls.

Karmin pls lock.

 

I love my P 6860FX too.

 

just ignore him you guys, oblivious just about all of you know he's off his rocker. Posting things like the gpu is auto overclocking is just plain crazy talk. He even claims in one of his old post that he's trained to repair computers....if he was, it was only to answer the phone and to say nothing more than I don't know.

 

Well, I hope I’m not jinxing myself, but after two solid raid nights, my computer has not locked up since applying Nomios' fix. I'm not ready to declare victory yet but this looks very promising. Who would have thought such a random setting would fix this issue.

 

Nomios and I were just at the end of our collective ropes I think. I did a lot of research into trying to figure out why my laptop was overheating and had dealt with the issue for over a year. I finally found a way to underclock the GPU which also fixed the problem but I felt it was unnecessary to do so if the performance level was pushed as Nomios had indicated in his directions. My power supply was generating a lot of heat and thus my laptop was overheating while charging. When playing without it, however, my temps dropped dramtically. when I used HWmonitor to check my temps after using Nomios' fix they dropped to around 120F from 165F as if I wasn't using the power supply. I also found that each laptop should be considered unique because they all have different temperature thresholds even if they're the same model.

I think it's safe to say that if you used Nomios' fix and haven't experience any lockups since then you're good to go. I've only had two or three lockups since October because of the overall room temperature in my house.

 

Um how your power adapter makes your laptop overheat makes no sense at all. Allow me to spell it out for you. Downclocking your gpu from STOCK clocks to prevent overheating is not a fix. That means you have a defective gpu. If your laptop is stuttering while on the adapter odds are its the adapter causing issues. You guys aren't fixing anything your just delaying a potential problem that could require a complete replacement of your system and by delaying it the issue may not resurface till your out of warranty. Use the warranty now if you are having issues thats what its there for.