Safe voltage for CC0 (1.7V) P3???

[krusty the klown]

Hi peeps http://www.sysopt.com/forum/smile.gif

Anyone recommend a safe maximum voltage for CC0 chips? Does the old 10% figure hold (I know some folk go higher, but I personally am not too keen), or does the actual value of the voltage matter more??

I'm running shiny new P3 700E CC0s in a VP6 (wanting to get to 2*933 http://www.sysopt.com/forum/smile.gif) and when I set 1.85V in BIOS, both the BIOS 'health monitor' and Sandra report 1.89V. This is slightly over 10% increase (which would be 1.87V). Now, if the chip was a 1.65V flavour, I'd be a bit nervous of 1.89, but for the CC0, it's only 11.2% - which doesn't seem that much to me.

Is 1.89 OK??

TIA http://www.sysopt.com/forum/smile.gif

[richamies]

Hey Krusty,

Personally I think it all depends on your cooling. I'm just using a Golden Orb, but I know this CPU(Celeron cC0) runs very cool. I've had it up to 2.1v for long periods(when I was burning in - which didn't help!). I've also had it up to 2.5v for a few minutes(trying to benchmark at 1190mhz - 140FSB!).

If you're happy with your cooling, then I'd personally say 2.1v TOPS. Realistic maximum for a long period would be around 1.95v.

Mine runs at over a gig with 1.7v, so that's where I keep it. Memory gets weak after 124FSB, and the CPU needs a touch more voltage so I keep it at default. My thoughts are that if you are needing over 1.85(which is fine for these babies) then heat will probably be more of an issue than not enough voltage. Mine runs at 850 with 1.3v.

Forgot to mention mine is a 566 Celeron http://www.sysopt.com/forum/smile.gif.

Rich
http://www.sysopt.com/forum/smile.gif

[krusty the klown]

Well, the cooling is and isn't an issue...

I'm only running retail heatsinks (but with no thermal pad + Arctic Silver) and if I run @ 2*933 /1.89V with the case off, they're OK (mid to low 40's ºC), but with the case on, they get up to ~50ºC under 100% sustained CPU load (too hot for my liking). When I remove the case and put my hand near the CPUs, I can feel a 'core' of hot air. I need to fit a blowhole to get some cool air in the system and in the summer, I doubt my retail sinks will hack 933, but I'm trying to source a QUIET pair of heatsinks (and having no joy - UK).

Hmmm... WTF, 11.2% over voltage shouldn't be a problem... http://www.sysopt.com/forum/smile.gif

[Roy]

I consider +15% to be the threshold of "aggressive". I've run a cB0 at 2.0V, but it didn't help the OC speed and I backed off to 1.85V. The cC0's revised substrate and increased V spec suggests better heat capacity. I think the 1133 is (was? http://www.sysopt.com/forum/wink.gif) spec'd at 1.8V. I'm going to buy a cC0 850 today and see what it'll do. It's the 1133 in disguise and I'm cautiously hopeful.

[Tom Pico]

Isn't safe a relative term? Intel says its absolutley safe at default voltage/FSB. At anything else, you takes your chances. Obviously, heat is a killer so adequate cooling increases your margin of safety. Check CPU temps at default and o/c'd - how much hotter is it running? Keep in mind that you'll no doubt want to replace that chip after 2 - 3 years anyway so shortening its lifespan may not mean anything. But also balance that against how much of an increase you're getting. With my 700E CC0 I can run at 980MHz (140 FSB) at default voltage. To run at the next higher step allowed by my motherboard - 1015MHz (145 FSB) I have to increase my core voltage to 1.80. Worth it? Not to me. There is no discernable difference between 980 and 1015 MHz. Sure when I run at 1015 I get a higher score in 3D Mark 2000 (3D Mark 2001 crashes) but I can see no difference in actual game play. So for me (I'm old and so I play it safe) 980MHz will do for now http://www.sysopt.com/forum/smile.gif. Especially since I remember how happy I was to move up to a 486/33MHz pc!