Same here. Multicore off then on and no crashes now for a while.

Seems turning it off and on again works with everything.

Sadly despite hopes, the crashes are returning more and more often again. Multi-core is off now and it is still crashing, as it was when turning it back on again.

The previous engine error:
Engine Error
Can't load lump 53, allocation of 17878673
is back now and again.

Also I have just got booted during a map change on HH1 and got the following error:
Engine Error
CMDLCache: Out of memory

First time I have ever seen this one.


Edit: Just googled the new error and it is mentioning heapsize as a possible solution. Anyone ever altered this much? Any thoughts?

The fact that your system is unstable as shown by the memory test crashing would be a prime culprit.

Download orthos (google it, it is easy to find) and see how long you can run that.

Happy to give Orthos a run later, but a little worried about the time you are meant to run this for. Seems an awful lot of stress to put the computer through. Can it be paused in any way or when you stop and start again does it pick up from where it stopped using the infomation stored?

Also, I still would have thought other games would have been crashing if it was instability as the issue.

Try putting "-heapsize X" in TF2 launch options where X is equal to half of your total ram.
e.g. If you have 4gb, then "-heapsize 2048000"

A short run should be sufficient to prove stability for gaming.. 4 or 5 loops

Get a temperature monitor working so that you can see if your cpu is overheating.. coretemp or similar.

Thanks Sir , I have added that in the launch settings (3GB for me -heapsize 1536000), and shall try that later today.

Thanks also Mr K3D for that, I did a very short run of about 20 mins earlier, before I had to use the computer and got the following result:

Type: Blend - stress CPU and RAM Min: 8 Max: 4096 InPlace: No Mem: 2814 Time: 15
CPU: 2400MHz FSB: 266MHz [266MHz x 9.0 est.]
21/06/2010 11:50
Launching 2 threads...
2:Using CPU #1
1:Using CPU #0
1:Beginning a continuous self-test to check your computer.
1:Press Stop to end this test.
2:Beginning a continuous self-test to check your computer.
2:Press Stop to end this test.
1:Test 1, 4000 Lucas-Lehmer iterations of M19922945 using 1024K FFT length.
2:Test 1, 4000 Lucas-Lehmer iterations of M19922945 using 1024K FFT length.
2:Test 2, 4000 Lucas-Lehmer iterations of M19922943 using 1024K FFT length.
1:Test 2, 4000 Lucas-Lehmer iterations of M19922943 using 1024K FFT length.
2:Test 3, 4000 Lucas-Lehmer iterations of M19374367 using 1024K FFT length.
1:Test 3, 4000 Lucas-Lehmer iterations of M19374367 using 1024K FFT length.
2:Test 4, 4000 Lucas-Lehmer iterations of M19174369 using 1024K FFT length.
1:Test 4, 4000 Lucas-Lehmer iterations of M19174369 using 1024K FFT length.
2:Test 5, 4000 Lucas-Lehmer iterations of M18874369 using 1024K FFT length.
1:Test 5, 4000 Lucas-Lehmer iterations of M18874369 using 1024K FFT length.
2:Self-test 1024K passed!
2:Test 1, 800000 Lucas-Lehmer iterations of M172031 using 8K FFT length.
1:Self-test 1024K passed!
1:Test 1, 800000 Lucas-Lehmer iterations of M172031 using 8K FFT length.
2:Torture Test ran 17 minutes 50 seconds - 0 errors, 0 warnings.
1:Torture Test ran 17 minutes 50 seconds - 0 errors, 0 warnings.
2:Execution halted.

I presume running the default Blend is the way to go with this.

Sadly the heapsize doesn't seem to have worked. Just crashed at the first map change with this error:

Engine Error
Can't load lump 53, allocation of 12663721 bytes failed!!!

I shall run the Orthos program as long as I can in a bit and see what happens.