View Full Version : Arghhh...What's causing the problems!!!???
After running my barton 2500+ @ 3200+ for 6 months, i got better case cooling and deciding to push it further. my cpu is locked, and so far im at 215x11 @ 1.8v. my 2x512mb corsair xms pc3200 is at 2-2-3-11 timings @ stock 2.6v. the bios is D23 Sub-Zer0. the problem I'm having is random reboots, bluescreen errors, and errors while booting. i dont think these started until i flashed from my stock bios to d23, but im not sure. im having trouble determining my optimal overclock. whats causing the problems? is it not enough cpu voltage? not enough ram voltage? the bios? also, i cant complete 3dmark 03 w/o getting a reboot or bluescreen. also, sometimes my pc reboots or goes to a blue screen as it is booting up. i ran prime95 torture test for 4 hours without any errors. is there an easy way to find out what the problem is? what do u suggest i do?
Have you tried out memtest or made sure it wasn't the video card that was unstable or too OCed? I had these problems too when I had too much FSB and not enough voltage. It could also be the actual mobo chipset overheating though. Too many variables at this point to make a conclusion. Check everything and make sure all fans work, make sure your video card isn't overheating, make sure you test the RAM with memtest, make sure the computer is clean, etc.
Xymurgy
06-30-04, 09:15 PM
Try setting the RAM voltage to 2.8v
i idle at 44-45c, and full load at 48-49c
my video card is not overclocked...well, i have overdrive on, buts thats it.
I'd definetly crank up that voltage to 2.9v and see if you get better timings as IIRC, you have CH-5 modules. If everything is the same as always the summer heat is what is messing with you probably i.e. chipset possibly (I think). If you have a spare fan, plug it in and make it blow on the chipset and see if that works. It could always be the OS crapping out on you.
I'd definetly crank up that voltage to 2.9v and see if you get better timings as IIRC, you have CH-5 modules. If everything is the same as always the summer heat is what is messing with you i.e. chipset possibly (I think). If you have a spare fan, plug it in and make it blow on the chipset and see if that works. It could always be the OS crapping out on you.
well, i tried 2.9v earlier, and the same thing happened.
Why not flash your BIOS back to the original or reflash it? Could laways have been a bad flash that messed up in a tiny way which results in this.
JTanczos
07-01-04, 12:07 PM
I would definitely raise the voltage to at least 2.8 and keep it there. Ramguy said that alot of their "stock" voltages for memory are reading too low and have to be raised manually.
Ide make sure the ram itself is still good by running memtest for a few hours. First thing I always do when it reboots randomly a few times. When it does reboot you should touch your finger to your NB cooler and see if its noticeably hot. If it is you definitely need a better cooler. 2 of my asus boards actually suffered heat damage cause my northbridge cooler wasn't enough. Stupid things had those active coolers with those tiny wussy fans on em. Should of replaced em but I thought it would be enough.
Also read what the BSOD says. What kind of error its bringing up. Might give you a clue as to what the problem is. When my 1100 was crashing it told me the ACPI had an error of some kind then rebooted so I knew it was a physical problem with with the board.
Also..... Check your hard drives. On mine the write heads on the one platter started to fail and it couldn't write the swap file and crashed windows. Luckily I was able to back up the drive. Great fun.
Good luck.
JT
mrspec3
07-01-04, 01:01 PM
I would definitely raise the voltage to at least 2.8 and keep it there. Ramguy said that alot of their "stock" voltages for memory are reading too low and have to be raised manually.
Your saying corsairs admits that it rates the stock voltages of its ram to low so you have to crank it up to get stock speed? :-/ Where did he say this?
Anyways I would crank out memtest and also try the sticks in single chanel or one at a time. It sonds like it could be your video card as well. Make sure AGP lock is on or set to 66Mhz and if you can crank up the vagp voltage to like 1.6-1.7V. Then also try to disable the overdrive on the card because an overclocked card can make the system unstable.
Next what kind of PSU do you have and what are the 12V, 5V and 3.3V lines reading under load (DMM OR MBM5)>? If your psu is crappy that would be the first place I would start
Steve
i can complete 3dmark 01 and 03 @ 210x11 @ 1.8v and 2-2-3-11 dual channal timings with 2.9v (havent tried 2.8v). does this seem ok?
EDIT: just failed prime after 2 minutes "hardware failure". this is the 1st time ive ever failed it, even when testing at faster speeds. so now im clueless what to do...
my psu is an antec 380 watt true power.
JTanczos
07-01-04, 04:05 PM
Your saying corsairs admits that it rates the stock voltages of its ram to low so you have to crank it up to get stock speed? :-/ Where did he say this?
When I posted over there I asked what kind of ram to get. I questioned timings and voltages also. He told me a LL one and told me to use 2.8 instead of the default. I did searching and saw in many posts he recommends 2.8 for added stability on XMS modules. I may be wrong since im going off memory(no pun intended :p) and it might just be LL modules.
JT
mrspec3
07-02-04, 09:18 PM
bump
Did you try anything I posted before?
Blondie
07-03-04, 07:15 AM
You say your getting blue screens, but your not giving the error- that usually helps. Is the blue screen appearing too quick for you to get the error? If so go to device manager, advanced tab, startup & recovery, settings button. Under system failure, uncheck automatically restart. This will give you the error.
dkdgbroyles
07-03-04, 05:27 PM
I get blue screens also. About every 3 to 6 days playing games. The message is usually "no page file_ fault". Also it says "windows is shutting down to protect components". Also it says try turning off or on cache, I can't remember which. In this hot room, my CPU can get to 52C but in the winter I stay at 44-48C at load. I use xp by the way.
flapperhead
07-03-04, 05:35 PM
ur probably hitting a combination of stresses in ur gaming that either the cpu/memory or mb cant handle. try backing the memory down 3/5 mhz and see how it responds..or adding voltage to the cpu/memory..
ive been busy lately so ive left it at 210 fsb. will raising the chipset voltage do anything?
timings are 2-2-3-11 2.9v
i guess its stable. i get about 45c idle, 50c load. if i get another blue screen ill post the error. any more suggestions on how to get the best performance i can out this machine?
Blondie
07-07-04, 06:32 AM
I've got a question for you- I see your running the Abit NF7-S. We have this MB in my friends son machine. I've never updated the BIOS on it. How stable has the machine been since you updated that if you did? I hate to update his and then he has problems. I had it here for a week while fixing it after he was infected with trojans.
I've got a question for you- I see your running the Abit NF7-S. We have this MB in my friends son machine. I've never updated the BIOS on it. How stable has the machine been since you updated that if you did? I hate to update his and then he has problems. I had it here for a week while fixing it after he was infected with trojans.
i never had a bsod until i updated the bios. then again, i never OCed beyond a 3200+ with the stock bios.
Blondie
07-07-04, 09:27 AM
:beer: Thanks, I think I'll leave his alone.
vBulletin® v3.8.7, Copyright ©2000-2012, vBulletin Solutions, Inc.