More hopeful 1.0 experience...1.5 Ghz

[jmsandrsn]

More hopeful 1.0A experience...1.5 Ghz

I just got my Celeron 1.0A today. This is a Phillipine chip made in week 51 with a pack date of 12/18/2001. The default voltage is actually 1.475V despite it being said that they would be 1.5V (Oldfart’s was 1.475V as well).

For cooling I used the Millenium Glaciator and Artic Silver II. The mainboard is an ST6 with Mushkin REV3. I did the wire trick right from the start however the wire must have come off by the time it was mounted in the socket since default voltage showed as 1.475V and I could only adjust it up to 1.6V. All voltages and FSB that I mention are according to what was set in the BIOS rather than the true reading from CPUID and HW Doctor.

I first tried 133 FSB at default voltage and it booted into Windows (WIN98SE to be more specific) without problem. I ran Prime95 for about 5 minutes and everything seemed OK. I didn’t test it more thoroughly since I knew I would not be leaving it this low. I then tried 140 FSB at default and it would not load Windows. I didn’t want to play around with the voltage that much so I then went straight to 1.6V (the maximum I could go without the wire trick) and she booted right up at both 140 and 145 FSB. I ran Prime95 for a few minutes but didn’t do any thorough stability testing. I then went for 150 FSB at 1.6V and it would not load Windows.

I figured I could go higher with more voltage so I took the chip out and did the wire trick again and made sure that the wire was a little bit tighter this time. I then tried 150 FSB with 1.675V (the new default after wire trick) and it would run Prime95 OK but HW Doctor gave some type of illegal operation error and Sandra2001te crashed before even loading. I then upped the voltage to 1.7V (HW Doctor actually read this as 1.72 – 1.73V) and Prime95 runs OK and both Sandra and HW Doctor ran without those errors. I then ran Q3 demo for about 10 minutes and had no problems. Max temp under load was right around 35C.

Just for the heck of it, I tried 160 FSB and 1.8V but the system restarted right after posting. I guess the top speed is going to be somewhere between 1.5 – 1.6 Ghz. I will test some more and let you know how high I can go.

 

[oldfart]

Sounds like the same chip I got. 12/18/01. The pin hole method is easier to do. Just make a little U shaped wire and drop it in the holes.

Keep us up to date as to how it goes. My best stable setup was 140 FSB, 1.65 Vcore.

 

[DSTA]

Bravo! Bravo!

PS:

If you don't mind my obsessing with this: could you do me the favour of reading the core voltage with HW Doctor with the CPU under load?

 

[zoopa_man]

I'm curious of which Abit ST6 you have. Is it the Raid or just "regular" board ?
I've seen problems with the raid flavor of this board and wanted to know if you had any weird problems with it. I'm still shopping for a board to match my Celeron 1.1 I just got. Thanks for the input. I'll keep everyone posted on the 1.1's performance.

 

[jmsandrsn]

Bravo! Bravo!

PS:

If you don't mind my obsessing with this: could you do me the favour of reading the core voltage with HW Doctor with the CPU under load?

Well, I'm actually trying 155 FSB using 1.8V but it's going to be a no-go. It actually ran Prime95 for a few minutes withoug errors but Sandra gives illegal operations right after loading.

I watched the voltage fluctuate while running Prime95 at this speed and voltage(even though it's not really stable) and the voltage fluctuated between 1.81 - 1.84V. It fluctuates in the same range when sitting at idle.

 

[jmsandrsn]

I'm curious of which Abit ST6 you have. Is it the Raid or just "regular" board ?
I've seen problems with the raid flavor of this board and wanted to know if you had any weird problems with it. I'm still shopping for a board to match my Celeron 1.1 I just got. Thanks for the input. I'll keep everyone posted on the 1.1's performance.

I'm actually using the raid version but I'm not using the feature...it's disabled in the BIOS.

I guess I will probably just run it at 1.5 GHZ since 1.7V is a somewhat reasonable voltage. I could probably go a few Mhz higher on the FSB but it doesn't look like it's going to be worht all of the extra voltage.

 

[oldfart]

Hmm. So it seems the C1.0A is not as good an overclocker as the C1.2. Like my exp, it requires quite a bit of Vcore to get it up to 1.4 - 1.5. I never tried 1.7 Vcore. I went to 1.675, 145 FSB. I could not load Win2K @ that setting.

 

[DSTA]

Well, I'm actually trying 155 FSB using 1.8V but it's going to be a no-go. It actually ran Prime95 for a few minutes withoug errors but Sandra gives illegal operations right after loading.

I watched the voltage fluctuate while running Prime95 at this speed and voltage(even though it's not really stable) and the voltage fluctuated between 1.81 - 1.84V. It fluctuates in the same range when sitting at idle.

Thanks very much for the voltage read out. Looking pretty normal, then - perhaps it's time to find some other worthy obsession

Anyway 150 FSB @ 1.7 does not sound bad at all! A bit more difficult than I had hoped for, but well worth it.

Northwood at 2.53 GHz might be a bit faster, but no 50% overclock .

 

[DSTA]

Hmm. So it seems the C1.0A is not as good an overclocker as the C1.2. Like my exp, it requires quite a bit of Vcore to get it up to 1.4 - 1.5. I never tried 1.7 Vcore. I went to 1.675, 145 FSB. I could not load Win2K @ that setting.

Yup. As said, I would have thought it to be a little easier. Also means that the 1.1A might be bit on the edge as a recommendation for people looking for simple "BIOS overclocking".

 

[oldfart]

If the 1.1A is the same, it could be a little dificult. To hit the 133 FSB mark, you are @ 1.463. Thats up there if its like the 1.0A. Who is going to blaze that trail?

 

[zoopa_man]

I'll be the trail blazer for that one.

Hope to have decided on a board by sunday this weekend after the computer show. If they let me bring my PC into work to play with it I should have it built, set up and ready to goof with by two weekends form now. Depends on what I find at the puter show and how much time I have to play with these things.

 

[Yodums]

Thats nuts dude.

150fsb is crazy good job.

 

[jmsandrsn]

Hope to have decided on a board by sunday this weekend after the computer show.

I've been very satisfied with the ST6-R. I've been using this board since it first came out around July or so. Only allows you to increase the voltage by 0.125V but wire trick isn't that hard to get higher voltages. I've been able to run memory at CAS 2 all the way up to 166/167 FSB.

I've also got the TUSL2-C but prefer the ST6 in this case.

Good luck with whatever you chose.

 

[Yodums]

So what exactly is your fsb now?

 

[jmsandrsn]

So what exactly is your fsb now?

I'm going to leave it at 150 FSB for now. Might be able to go a few Mhz higher but I not going to worry about it for now.

 

[mikester]

very well done, congratchulashuns to you.

 

[Flu!d]

So you got the chip to run at 150mhz FSB @ 1.7 volts? Well I guess it wasn't the Gigabyte mobo that was making the chip a hard overclocker after all....I guess you could say that just because it is a Tualatin based chip dosen't mean that it will nessecarily overclock on the smell of an oily rag!

 

[Rooski]

jmsandrsn,

Have you actually ever tested your system to be sure your fsb can work stable at those higher speeds, before installing the 1.0a?

 

[jmsandrsn]

jmsandrsn,

Have you actually ever tested your system to be sure your fsb can work stable at those higher speeds, before installing the 1.0a?

Same components are stable up to 167 FSB using a PIII 1.2 @ 1.5 Ghz.

I wasn't able to loop 3DMark2000 overnight...must have crashed back to the desktop at some point. Will try upping the voltage up to 1.725V and see if this helps. This was at 150 FSB at 1.7V.

 

[ol' man]

Well I know of a few people that have to run a 1.8 vcore to get the 1.2GHz cel stable @ even 1.5GHz. I had to run a 1.725v to be stable at 1.5GHz. I would not say every chip is going to behave like this one jmsandrsn has. I still think it is a very good OC. I know if I had that chip I could get it to go abit higher with my IHS mod Sounds though like most should not attempt that though.

Hey jmsandrsn, you should do some benchies for us anyway at 150FSB.