How to diagnose T60p freezing?

[cacophony]

I have a T60p that I recently upgraded to a 128GB Samsung 830 SSD and Windows 7 Ultimate (fresh install). Computer ran flawlessly for the first month but in the last couple days I've gotten two freezes. Both occurred about 5-10 minutes into watching a youtube videos within firefox on full screen mode. The two times it occurred the video would freeze and the system would be completely unresponsive to all keys (ESC, CTR-ALT-DEL, etc). The only keys that did anything were the keyboard light and volume/mute buttons. I know the volume buttons worked because a tone/buzz continued to play when the video froze, and the volume of this unpleasant tone was controllable. After waiting a few minutes, I just held down the power until it shut down.

After rebooting, the Event Viewer logs only show the forced shutdown. The prior event was almost 24 hours earlier.

I ram memtest86 for a few hours and it didn't produce any errors.

I had a theory that maybe the CPU got to TJ Max and did a forced shutdown, but the last time it occurred I had RealTemp monitoring running, and although I couldn't see the temp while watching the video I did have the alarm set to go off at 90 degrees (10 prior to TJ Max), and didn't hear any alarm. I'm also fairly certain that it was running at about 70 degrees around that time and I didn't hear the fan speed increase.

There don't seem to be an tools in the installed Samsung SSD Magician software check to make sure the drive is functioning properly. I did run the performance and optimization tool and it didn't report any issues. After the first crash I did run chkdsk on startup and it didn't appear to find any issues.

I have Windows entirely up to date, and so is Firefox and Flash. Very little is installed on the machine: Firefox, Thunderbird, Lightroom, WinSCP, putty, VNC Viewer, Samsung SSD Magician, some printer tools. I have 2 GB of RAM, and I'm typically running at about 50% utilization.

What can I do here? This is really frustrating as stability has always been very important to me, and I ran Windows XP on multiple earlier Thinkpads with not a single freeze over the last 10 years.

[cacophony]

Just had the thought that I haven't been monitoring GPU temps. I just installed Thinkpad Fan Control.
Perhaps the ATI V5250 reached it's max temp and shut down? Not sure if I have to configure TPFC for my T60p, or even if it's a good idea to run for my case.

[RealBlackStuff]

GPU-heat has always been the enemy in a T60p.
Also, with 2GB you are a bit low on memory.
Replace 1 stick with a 2GB for a total of 3GB (which is the T60/T60p max. anyway, due to a chipset limitation).
Also, try the same Youtube film in another browser, like IE9.
I like FF, but it is not the most efficient browser.

[hhmcsv]

Sounds to me as a SSD problem, I had similar with a different drive, see this thread:

http://forum.thinkpads.com/viewtopic.php?f=29&t=105987

.......but my present SSD has not showed any freezing for more than 3 months now

[cacophony]

I haven't seen the issue since installing TP Fan Control, so my current theory is that GPU shutdown occurred because of poor fan control. I had never installed any of the Thinkpad Power Management utilities, so perhaps the default Windows 7 fan control wasn't adequate for my setup. I am noticing that the TP Fan Control utility does seem to keep the temps more under control.

IMO, the fact that the two instances have both occurred under heavy CPU and GPU load is a pretty strong indication that the problem is CPU/GPU related, not SSD related.

[Peter Reed]

I have a T60p (T7200 w 4 GB RAM) with ATI Mobility Fire GL graphics chip (GPU) and had the same shutdown problem after I replaced the heat sink and fan due to a noisy fan. I downloaded TPFC.63 and realized the GPU temps were the problem. Prior to my fix (see below) I was seeing idle temps of 55-60C for the CPU and 79-80 for the GPU. These would rise to 95+C for the CPU and 100+C for the GPU under load which would cause occasional shutdowns.

There are several threads on this forum addressing how to solve the GPU heat problems. The solution I chose and am very happy with is as follows:

[*]Purchased an .8mm thick copper shim on Amazon for $4 (UXcell). The shim is 14mm x 14mm.
[*]Removed the heat sink (HS) and fan and cleaned the CPU, ATI, and Northbridge chips with Artic Clean and Thermal Surface Pufirier solutions which came with my Artic Silver 5 thermal compound (also purchased via Amazon).
[*]Removed the thermal pad for the GPU only, leaving the thermal pad for the Northbridge which does not appear to be a heat issue.
[*]Cleaned the HS with the above two solutions where the CPU and GPU seat.
[*]Put a thin film of Artic Silver 5 on the CPU, HS CPU seat area, and both sides of the copper shim (using an old credit card to smooth out the compund).
[*]Placed the copper shim on the GPU chip (using needle nose pliers so as not to touch the comound) with approx 1 mm overhang from the edges of the GPU.
[*]Using the needle nose pliers I bend the tabs up slighty where the screws attach the GPU Northbridge holddown piece. I used this method to give more downward pressure on the HS over the GPU and Northbridge versus tape on a quarter and dime as suggested by other forum members. I tried the quarter and dime approach but it appeared to be too high and I was worried about over stressing the hold down screws.
[*]Put everything back together.

Results have been excellent as follows: At idle the CPU temps are still in the 55-57C range but the GPU temps are now in the 66-69C range which is 10C less than prior to my mod. Under load the GPU does not exceed 87C. Please note it takes a few days for the Artic Sliver 5 to settle in. By the next day I achieved the above temps.

Hope this helps any member considering the GPU heat mod.