is that safe?

[BSH]

I revisited the settings, and it seems unstable with 1.7v, but 1.725v is okay. is that still safe? almost 10% more, than the standard voltage!
and why does my abit kt7 board measere every voltage 10-20% less, then the standard? for example: core voltage 1.59v (instead of 1.7v), 3.2v (instead of 3.3v), 11.5v (instead of 12v), etc.
does it mean, my psu is so low quality, or the measure of the bios is so inaccurate?

[Aman99]

Well what kind of power supply do you have? 1.725 is very safe as long as you have decent cooling, I'm at 1.85 volts (bios reads 1.91) myself.

Aaron

[Rob Cork]

1.725 is fine - default voltage for high-end Tbirds is 1.75V! 1.85V is the max normally allowed by most boards and is perfectly safe as long as you can stay below 50C (IMHO), which shouldn't be hard. I've been running this duron at about 2.27V, give or take 0.05V, for over 5 months, and it's still going. I remember being scared to go over 1.8V when I started, but you quickly get over that :D

[-inox-]

yeah its nothing to do with your psu my vltages are not exact either but it think that its to do with the board and i would be pretty difficult to keep it exactly wright any way i have

duron 700@933
and according to mbm5 it says vcore 1.74 12+ @ 11.77
3.3 @3.45
so i think its likr that on loads of boards

[David]

i thought going over 1.85volts was dangerous?

[Justy303]

nelly (Jun 04, 2001 06:37 a.m.):
i thought going over 1.85volts was dangerous?

I have myself been at 2.25v, and that didn't kill my duron. Of course, it's necessary to have a good cooling system at these voltages. I guess it will shorten the life of the cpu, but WTH, if it burns up, that's at least a good reason for an upgrade.

[BSH]

one more time... i thought, the system is stable, coz it ran 2-3 hours without problems (after that, i shot it down).
but now i tested all the games i have, and everything went ok, until the F***ing colin mcrae rally 2: it went good, til the first frame of the real game (i mean, over the menus), then sudden death, dropped me back to windows.
i increased to 1.75v, and now everything is fine.
here pops up a new question: how can i decide, if my system is stable (and i mean, REALLY stable!)?
and how can i decide, if the crash is caused by a software bug or by hardware instability?
give me tips, please!

[TT120]

Please watch the language BSH. We have a lot of young people here and we want to provide a good place for them and everyone to come and learn. Language like that is inappropriate. I'm not getting on your case about it, just letting you know. :):):)

[Rob Cork]

BSH (Jun 04, 2001 08:51 a.m.):
one more time... i thought, the system is stable, coz it ran 2-3 hours without problems (after that, i shot it down).
but now i tested all the games i have, and everything went ok, until the F***ing colin mcrae rally 2: it went good, til the first frame of the real game (i mean, over the menus), then sudden death, dropped me back to windows.
i increased to 1.75v, and now everything is fine.
here pops up a new question: how can i decide, if my system is stable (and i mean, REALLY stable!)?
and how can i decide, if the crash is caused by a software bug or by hardware instability?
give me tips, please!
There's different levels of stability I guess - if all you want to do is surf the net and listen to mp3s (me in a nutshell most the time) then you can probably get away with a higher overclock. Instability shows up the more the cpu is stressed so if you're playing games all the time unstability is likely to show up at slightly lower speeds. Typical 'tests' for stability tend to be running Seti@home 24/7, or Prime95, or the Q3 demo loop, for a considerable period of time - days, not hours. In your case, maybe trying different video drivers / Via 4in1 / bioses might help - try all the options available.

Seeing as you want to play games on yours, I'd try running 3dMark (2000 or 2001) repeatedly, as well as maybe leaving it overnight running the Q3 demo loop (not sure how to do that though).

[Justy303]

I would agree to that. The 3dmark2001 seems to be very oc-sensitive, just like CMR2.

[BSH]

and what, if i leave the machine up, and it goes too hot and burns out, while i'm not there? is that the only way?
now i'm testing the stability with motherboard probe set up to shut down, when it get's too hot, and i'm using background processes (totaltest, hot potato, etc.) to keep 100% usage. it seems stable.
what about software cooling? (rain, for example) i use it with good results: 10 celsius down, when idle.
now i have 45C with a few apps running (mid cpu usage), but normally goes down to 41C, and up to 50C with heavy load. is that safe?:)
my case temperature goes up to 33C. i have a very heat producing tv card (with passive cooler), a vanta (with active cooler), a sb awe64 gold (slightly warm, without cooling), sb128 (not warm), 7200rpm hdd (with cooler, nice warm), 5400hdd (in rack, with cooling, hot!), cdrom (warm) and a cd writer (not used, not warm).

[asmodean]

Adding some case case fans should alleviate the problem.
Add a fan to that tv tuner too, if you cannot touch the HS for more than couple of seconds. That 5,4k rpm HDD could also possibly use cooling, if the warm air gets to spread around.

My 2 pennies.