Celeron 700@893 voltage?

[Guest]

For the ones that read my previous message: i finally bought a Celeron II 700 (the price was about the same with 600 and 633).
What made me happy is the fact that my new processor is the new C II with cC0 stepping.
I've overclocked it and seems to be stable at 893MHz@85MHzBus, although i expected more...

I have the following question:
I'm working at 893Mhz now, with 1.85 or 1.90 volts.
With bus speed set at 86,87 or 88 some software will create errors and Windows will stop responding (Win2000Pro). When set at 89 or above Windows won't start at all. At 100Mhz i can't even boot (monitor doesn't turn on), even with 1.95volts.
Do you think that a higher voltage would help me to run @90MHz FSB speed or higher?
And if yes would it be safe enough for the CPU? (My motherboard supports max 1.95, but i can trick it through the SlotKet)

My motherboard is an Abit BE6-II (newest bios) and i'm using the MSI Master Slotket (that can be set to 1.50-2.00v).
Note that i've set the L2 Chache Latency to the slowest setting available (i will try faster speeds later on, when i'm sure about the max stable frequency) and i've tried many PCI/FSB and PCI/AGP division settings.

Thank you!

Something

 

[Guest]

If they really had the same price you should have bought one with a lower multiplier, in order to be able to use a higher FSB!

My cel600 of the cB0 stepping reaches 945 @ 1.90V @ 55C.

In answer to your questions; Yes upping the voltage always help, but when you do that, you should also improve the cooling. But NOWHERE in your mail do you mention the CPU temperature. Thats a shame, because thats also a good indicator of how high you can go with your current cooling-solution.

Its OK to play with the timing of the cache, but did you try also with the timings of your main memory? -If you're using old PC66 that might be one of the things holding you back.

Henrik

My first post! 🙂

-The Dane In Spain-
cel600@945

 

[Guest]

My CPU is stable @893MHz @ 1.85v.
Temperatures are about 43C idle and 50C @ full usage.
I've got 3x128MB PC100 memory.
Thanx for your post!


Something

 

[Guest]

Well, the temperature sounds OK then...
Just for the hell of it, you could try LOWERING the CPU-voltage. I know it sounds crazy, but just today I had this experience with a friends Athlon 900 with an inferior cooler. We had to lower the voltage 0.05V before it would run "stable" (Ha!) at 1000.
(Remember that the cB0 stepping celerons were marked with 1.5V?)

Henrik

-The Dane In Spain-
cel600@945

 

[slvr_phoenix]

You could always try underclocking the multiplier so that you can raise the bus higher. Even just to get it back to 800MHz with a 100MHz bus the system would probably be faster because of the memory bandwidth. But I'd bet that your threshold is 900MHz and by lowering the multiplier and raising the bus you could reach it.

Just a guess though.

- Sanity is purely based on point-of-view.

 

[Guest]

Yes, it really sounds crazy, but i'll try it.
The cB0 works @1.5, the cC0 @1.7 (mine is a cC0).
Anyway, i noticed that it would run Windows @91FSB (with very high voltage 2.00v), but i prefer keeping it a bit slower than frying it!

Now I'm working 100% stable @914 @87 @1.85v




Something

 

[Guest]

I think it's impossible to underclock the multiplier of a Celeron, or PIII (even PII i think).
Is there really a way of doing this??

Something

 

[Guest]

Nope..

But phoenix is very experienced assembler who thinks in terms of physics, and I suppose if you expose the core long enough to atomic radiation the multiplier will change and so will a lot of other characteristics of the processor, but hey.

(he he)



***Hey I run Intel... but let's get real***

 

[Guest]

hahahahaha!!!

😉

Something