I've jsut build a completely new PC and when i tried to install windows, one the final stages it BSOD with the IRQ etc error. I tried taking out 2 RAM sticks and it installed fine and now windows is booted. However, to check if this RAM is fine i downloaded a windows memory testing program (i know its not very thorough for many reasons but still....) and immediately (give or take a few seconds) it came up with the exact same BSOD. I've noticed when i check my RAM its extremely hot, almost too hot to touch. Is this normal? It all doesnt make sense.....4 sticks didnt work....now 2 sticks work for browsing but as soon as RAM is stressed/used...BSOD......perhaps it happened earlier due to faster heat increase due to more RAM sticks closer packed together? Anybody have any ideas please? Just spend over £1000 on this new PC, really stressing me out Thanks
Ok this is interesting, check out CPU-Z....wtf? http://i214.photobucket.com/albums/cc207/Igniseus/cpu-z.jpg EDIT: Okay im new to this feature but it turns out the CPU will automatically under-lock depending on load? Temps that can be monitored are all fine..... http://i214.photobucket.com/albums/cc207/Igniseus/temps.jpg
Ok.....so i managed to somehow get it running with all 4GB of ram stable until when i installed Stranglehold it died about half way in. The BSOD went on and off so fast i couldnt see it, but i guess itwas the usual. Anyway, i felt the RAM immediately after it was literally too hot to touch, 2 seconds on it and it burnt my finger abit. Also of note, my PWM temp goes from about 38C idle to 50C under load. Heres my spec's btw: IP35-E Q6600 Geil 4x1GB PC-6400 Corsair HX620W Coolermaster Ammo 533 with 1 front 90mm fan and 1 rear 120mm fan both at 1000rpm (max). 8800GTX SB X-Fi 2x 750GB HDD Also this is all whilst the case is open, and flat so heat can rise out and i have a ceiling fan also, and my window is wide open, so it will be much cooler than when sealed up and in my desk. Thanks
How much voltage are you runing through the ram? Can you check the bios settings an find out? and yes the multi on the cpu will drop to 6 when its not being used. you can turn that off in the bios.
Right just to add, i've ran 4 passes of mem test 86 on test #5 with 0 errors. I had the case open to let cool air get in. This was done at 1.90v on the RAM. I shall test a game now and see if it holds for 5 mins.
I'm not being too hopeful but it seems the voltage increase and MCH increase might have fixed it, but thats going off 2 mins on a couple of games, but things didnt tend to last longer than that before anyway. I need to go to work now, so i will test further later, so lets hope for the best
Why did you have it on such a low voltage anyway, the range is 1.9v to 2.2v for that particular model if I'm not mistaken?
I cant agree with that one. Auto for me ment an unstable system I had to set the voltage and timings for my system to work without getting bluescreens.
Then you are not lucky because the 'auto' setting is meant for those who don't know what voltage and timings to use. Auto should work for everyone, theorically
Sure if you buy generic ram it "should" work in "theory". but this is real life and whats supposed to happen is rarely even close to what actually happens. There is no harm in manually setting the voltage and timings anyway. Especially if you are setting them to what the stock values are.
Site spec says 1.9+ but the SPD and generally everywhere else says 1.80. I'm going to do some more testing soon so i'm not certain whether the voltages fixed it yet or not.
dont put too much faith in the spd. The max for ddr2 as per the spd is 6400 (or something like that) and my ram is at 8500 running with 2.2v. stock. With the higher voltage is the ram still too hot to touch? I dont know about stability but that will shorten the lifespan of the ram. Might want to consider putting some direct airflow on it.
