T61 NVS 140M GPU temperatures

I have a T61 that I've had since March 2008. I recently decided to check the internal temperatures and what struck my attention was the GPU. According to GPU-Z, it's been ranging around 65C during normal use (internet browsing) while plugged in. When I ran the stress test in the ThinkVantage Toolbox, it reached 76C.

However, according to TPfancontrol, the GPU temperature has been around 55C to 59C, during normal use, so I'm not sure which to believe.

Are these temperatures normal? The laptop runs great with no problems whatsoever and has been running regularly for at least 12 hours per day since I got it. I'm concerned though that these temperatures are excessive and could eventually lead to a failure. It is out of warranty, so if these temperatures are not normal, there is nothing I can do really do.

Any feedback would be greatly appreciated.

 

these temps are perfectly normal. for a fact, i have lots of youtube vids being streamed and lots of tabs open in my firefox and lots of other softwares running and i'm on the upper 60's on my gpu. gpu z is accurate. what you could do is try to clean your vents.

 

Thanks a lot! This is reassuring.

I think what concerned me is that the GPU was running hotter than any other component, but I guess for a graphics component, that's normal.

 

Yeah those number are in the right ball park. However, keep in mind that your card is on the list of cards affected by the faulty chips so it may eventually fail despite the temperatures seeming normal.

 

I was just reading about this faulty Nvidia issue. VERY frustrating and I am definitely not pleased.

Any idea if there is a planned recall of some sort?

 

Planned recall? lol.... they allocated 200 million (US i think) to "fixing" the problem in all affected units, there was no recall and won't be since nVidia would probably be out of business as we speak if they came out with the truth in full. Supposedly the failure rates are only "slightly" higher than unaffected GPU's, but they've never actually released any estimates or actual figures, which i really would want them to.

 

Nvidia earned around 2 billion USD for the GPU that was affected by the recall, yet they only paid out 200 million USD, including the associated legal fee.... considering that the GPU are physically soldered to the motherboards, the computer companies basically have to foot the bill for the motherboard (unfortunately there is more than just a GPU on a motherboard), shipping, labour, lost earning, lost credibility, etc..... no wonder companies don't want to admit it, especially in the face of tighter earning margin for laptops.

 

MXM (any revision) would be useful here if it were in EVERY LAPTOP... alot cheaper and it would solve the problem for good.

 

this is probably why Nvidia wanted the laptop companies to use MXM... but laptop company refuses for two reasons, one is the greater cost of using this solution (nvidia MXM graphics card cost a lot money compared to the GPU chip by itself) and also many people are forced buy new laptops because their GPU is outdated. If Lenovo does MXM solution they will whitelist it too, like their other replaceable mPCIExpress solution, to control what people use in their laptop....

Also, laptop graphics card is bit different to computer graphics card upgrade... given the fact that there is only one fan that is shared across the CPU and GPU for heat disposal. One would hate to think, some body cook their CPU and GPU, because they decided to upgrade to some high octane GPU without considering whether the thermal management can deal with all that heat coming off the GPU + CPU. Warranty would also be something of a great spot, what if the user fried the CPU because, he decided that he needed a GPU that should only be used in W700 type laptop....

 

TPFan always gives a lower reading relative to HWMonitor and GPUZ. I don't know if its better to rely on TPFan. With the same specs as you, mine has never hit 65C using HWM nor GPUZ. Always runs about mid 50's to highest about 60C but average seems to be 56 to 58 like that. My T61 is also on more than 12 hours per day. Ive had no problems related to the card that I am aware of but its not used to play games. Months back I did try NFSPD and the game did shut down.

 

i had my t61p gpu hit the 80's one time when I was playing blu-ray over a usb drive. that's the only time i've ever seen it get that high. most of the time it hovers in the 50's...60's maybe if i'm playing a dvd.

i since sold all my blu-rays, though. crap technology.

 

I periodically game on my T61, mostly Call of Duty and Call of Duty 2. I'm fortunate in that I've never had my computer shut down due to overheating, so I guess I've been lucky so far. When I was addicted to COD, I used to spend several hours on the weekend playing and never had a single problem.

However, I'm still not happy with these high temperatures though. I bought this computer in mind that it would last well into 2012. It's an amazing machine, almost two years on and it is still running as if I just got it, but now that I know that my video card is running at above normal temperatures, I'm now concerned that it will prematurely die, something which is not supposed to happen with a ThinkPad...

With TPfancontrol, if I max out the fan speed (setting 7), it cools the GPU down to 51/52 degrees, but it's not practical to have the fan running at full speed all the time. I'm going to open it and blow a can of compressed air across the fan and heat sink so see if that helps.

 

stressing your Nvidia GPU will lead to electromigration on your gpu bump material, which is going to cause it to fail...

 

UPDATE:

I opened my T61 removing the palm rest and keyboard. I used a can of compressed air to clean out the area, especially around the fan and heat sink.

I am going to keep testing it, but it appears to really have helped! So far it is hovering between 57C/58C/59C which is an excellent improvement.

I am surprised about this improvement as it didn't appear to be very dusty, but this is a welcome improvement nonetheless!

I will keep monitoring it, but I am pretty sure the dust blowout definitely helped.

 

I'll blame electromigration when mine fails...

 

I thought to first order its heat/cool cycles, not heat itself?!

 

that was what some people (including me) initially thought, but closer inspections by some experts and more information released by Nvidia on this information, have shown that electromigration on the bump material is the root cause of the failure.

 

UPDATE 2:

The temperature is staying stable at 59C, sometimes going up to 60C, but most of the time is at 59C. It briefly peaked up to 62C initially, but quickly cooled down to 59C/60C.

One thing that is striking though is how long it's taking to reach these temperatures. Before I cleaned the inside, I would find it at 65C immediately when it finished booting up.

However, since I cleaned it, it takes about 15-20 minutes for it to reach 59C/60C.

Either way, I'm happy with the 5C improvement. I guess this is what I'll have to live with.

 

And where are the temps before the 15 to 20 minute spike up to 59 to 60C? Please use GPU-Z and keep a data log.

I rarely dream. Last night I had a nightmare that as I was using my TP the red pointer button started disintegrating and the whole keyboard went up in smoke due to the Nvidia GPU.

 

The temps started around 45C-50C and gradually increased. At first when I saw that the temperatures stuck around 53C/54C, I got very excited, but then they gradually kept rising.

Unfortunately after an extended period of basic use (internet browsing, I have several tabs open in Firefox), it does go above 59C/60C and hovers around 64C/65C, occasionally dipping down to 62C/63C. It did briefly reach 66C/67C. Overall, it is cooler than before the cleaning, but I guess not as much as I had initially thought. It's just taking much longer to reach these high temperatures than before I did the dust blowout.

I've been using both HWMonitor and GPU-Z to check the temperature readings and they're the same.

With TPfancontrol, if I max out the fan speed (setting 7), it drops temperatures to around 52C/53C. I just don't think it's practical to run the fan at full speed all the time, although it's a good idea during gaming.

My T61 is out of the Lenovo warranty, but is covered by Visa for two more months. I'm thinking about calling them tomorrow.

I might be having that dream soon...

 

TP fairy is probably angry....

 

Lol, mine must be pretty far away from failure since my GPU is at 38C according to TPfancontrol and 48C in HWmonitor, i don't know which one to trust but i would think the one with higher readings would be more direct of a reading using simple logic, these are still great temps.

 

I'm amazed you have such low temps.

 

Those are great temperatures, especially since your video card (FX 570M) is much more powerful than the NVS 140M.

 

Yeah, basically it isn't gimped, but the cards potential really comes to fruition with a game that is WELL OPTIMIZED... so, NOT GTA IV. A game like Dirt 2 however, really makes this card shine because Dirt 2 is so well optimized i can max everything out pretty much, other than a 1360x900 resolution, which still looks fine due to the fact that it was console centered and scaled for the consoles.

 

I know this is an older thread, but I'm having the same problem with my T61 which I bought around March 08, too. At first I thought it was a problem with my CPU, too, due to a test with SpeedFan which warranted temps of 70-80C for the GPU and ~60C for each core on the CPU when it is idle. I don't know if those temps are normal or not for the CPU, but the GPU's heightened temp really bothered me.

My warranty actually has NOT expired. But I haven't dealt with Lenovo Tech Support very much in the past. I had to have a fan replaced & that was no problem, but I'm not really sure how to broach the subject of "hey, does my graphics card have a faulty chip?"

Would they replace the mobo over that? I have a warranty that covers *everything,* so I'd hope so, but just wondering if you guys have some advice?


The laptop is running XP SP3, with a t9300 cpu and the nVidia quadro 140m.