BSOD - Infrequent, browsing mostly. IRQL_NOT_LESS_OR_EQUAL

Do the stress tests next. Monitor temperatures as you do so to make sure your hardware does not overheat. Stop the tests at any indication of overheating.

Okay, little update.

I've installed the temp monitoring software and been trying to get a feel for them. I've also done a furmark burn-in test "Burn-in benchmark 1920x1080 15 min), I'm not sure if this was the correct one to do. But, I didn't have any crashes during it (but it was only 15min).

Furmark tells me my benchmark was 2171 points. I had 21.4k frames, an avrg frame rate of 23, and my max temp was 81C. The temperature is according to furmark (I had the HwInfo thingie open, but I couldn't find the temperature readings until after the benchmark had finished (and during it, I was too nervous to try minimise furmark).

I don't know if 81C tells you much, I understand GPU's generally tolerate (or are at least subjected to) higher temps than CPU's? I think the max temp for the HD6850 is 90C, but whether that is being generous or not, I don't know. It seemed to level off quite a lot after 5 mins of furmark, and settled into 79-81 after about 7.5 mins.

I think I ought to do the test again however, leaving it on for longer, and this time making sure to note the temperatures in Hwinfo, and possibly others.


Also, I kind of want to apologise, I feel lately, I've been a bit lax at replying, and acting on your advice, so I'm sorry about that. I mentioned I'm at university, I'm in my third year, and though I can honestly say I've worked very hard the whole semester, the last two weeks are a busy and stressful time. (I feel like I'm undergoing my own stress test at times). So I hope you'll be patient with me.

Having said that, I'm hoping to say, get all the stress tests completed by the end of the weekend, that seems like a fair and realistic goal.

All the best,
Dan.

I am a PhD student myself, and I remember the stress you are going through. I completely understand where you are coming from, so do things at whatever pace works for you. :)

81 C is fine for the card and nothing to be concerned about. I would recommend running longer as you said and checking for any video artifacts. Monitor the temperatures in whatever program seems to give the values you would expect. They should all be about the same as it gets hotter as most of the error in such software is in the idle temperature sensing. Most are reliable under stress unless you see something that just makes absolutely no sense (like temperatures lower under stress than at idle or temperatures below room temperature).

Probably quite naive to think I could have all the tests done by the weekend! But, I've ran the Furmark test for 90 minutes at the weekend, max temp was 79C (down from 81C, although it was a bit of a colder a day, if that would affect it), no crashes either, so that's looking a bit promising, tho I could probably still run it for a lot longer.

As for the rest, I finish friday, so I should be able to get stuck into them after the weekend and finish them quite quickly.

Heres hoping. Again, thanks for your patience. :)

You're welcome. No need to run FurMark longer if it balanced out at 79 C over a fairly long period of time. No artifacts, I assume?

Let us know how the other steps go as you are able to get to them.

I don't know as I've seen an artefact before, but I'm pretty much certain I hadn't had any (I guess it's the sort of thing you'd spot a mile off).

I wouldn't say that artifacts are obvious, but most of the time, they are hard to miss. Weird lines show up on the screen, the graphics are not as smooth and become pixelated, etc. I imagine if you did not see artifacts, it was clean.

I'm pretty certain it was clean in all fairness. I'm a graphic designer too, so I'd imagine I'd spot a visual anomaly fairly quickly.

Okay, I'm pretty much done with uni now, although I'm still a little bit busy in some ways, but I'll be able to start testing things much more quickly. I tried a pass of the Video Card memory test program, and it gave no errors (this was just one pass ofc).

So, I'm going to put it on a continual test either tonight, or tomorrow, and report back when its done seven or so passes. Then I'll move on to passmark, and then the BurnTest, assuming all is well.


All the best,
Dan.

Okay, I ran the Graphics Card Memory test, and it did 9 passes, and told me it had no errors.
There was thing that appeared in the dialogue box:

Trying 32bpp BGR:888 mode...NOT SUPPORTED (Code: 80004001)
Trying 16bpp BGR:565 mode...NOT SUPPORTED (Code: 80004001)

Those two popped up each pass, but I don't know if it constitutes an error exactly, as I said, VMT didn't tell me there were any.

I'll head onto the next two stress tests now.
Dan,