iP3/T+Tualatin NOT for overclocking!

Crashman

Polypheme
Former Staff
I am no longer recommending the iP3/T for overclocking with the Cleron Tualatin. I have tried several pin methods for raising the core voltage with no affect, leaving me unable to raise the bus speed.
Furthermore, with regards to all the praise the thing has gotten, it is DEFINATELY NOT faster than the PIII 1000EB, in fact the performance is SLIGHTLY WORSE than my old PIII 700@933 (but still superior to any other Celeron and possibly better than the ultra-rare PIII 1000E)

What's the frequency, Kenneth?
 
I thought the tualatin celeron was the new PIII. Same cache size. It should work. Are you sure that old gfx card isn't to blame?

j/k

Strange though.

<b><font color=blue>~scribble~</font color=blue></b> :wink: <A HREF="http://www.ud.com/home.htm" target="_new">Help cure cancer.</A>
 

Crashman

Polypheme
Former Staff
I am almost 100% sure the bus speed is to blame!
ATI Radeon DDR 3DMark-2001:
w/Celeron 850 2444 marks
w/PIII 933 3200 marks
w/Celeron 1200 Tualatin 3059 marks.

What's the frequency, Kenneth?
 

phsstpok

Splendid
Dec 31, 2007
5,600
1
25,780
Hey, Crash. You just need to be more persistent.

Click the link below for a nice review of the iP3/T with Celeron 1.2, complete with overclocking. The author describes how he got core voltage up to 1.8 volts and how he squeaked out an overclock of 1.5ghz.

It seems the iP3/T - Tualatin-core Celeron combination is seriously hindered by low memory bandwidth.

I hope you can read German (and I believe you once mentioned you can). Babel Fish didn't do too badly with this one, though, but you need the original link to expand the pictures.

<A HREF="http://www.hardoverclock.net/html/reviews/pl-ip3t/pl-ip3t_1.htm" target="_new">http://www.hardoverclock.net/html/reviews/pl-ip3t/pl-ip3t_1.htm</A>

<b>We are all beta testers!</b><P ID="edit"><FONT SIZE=-1><EM>Edited by phsstpok on 01/19/02 01:42 AM.</EM></FONT></P>
 

Crashman

Polypheme
Former Staff
Während die Caches von P-III Prozessoren längst nicht mehr von Latenzzyklen gebremst werden, wurde der Cache des Celerons mit einem Wartezyklus ausgestattet, was die CPU verlangsamt!
. OK, so that explains a little bit of the performance difference, is this wait state hard wired into the CPU or simply an instruction sent to BIOS (hackable)?

What's the frequency, Kenneth?
 

phsstpok

Splendid
Dec 31, 2007
5,600
1
25,780
You had to make me dig up the translator again. I had one year of German and remember almost none it.

is this wait state hard wired into the CPU or simply an instruction sent to BIOS (hackable)?
Are you asking me? You're the resident expert.

The translation wasn't good enough for me to scrutinize what was said. Is that Cache Wait State imposed somehow by the iP3/T - Celeron combo or is it just a function on the Celeron itself? Did Intel hamstring another Celeron internally?

The author of the article mentioned doing something creative to get 1.8 volts. Is the method different than those you have discussed in the past?

<b>We are all beta testers!</b><P ID="edit"><FONT SIZE=-1><EM>Edited by phsstpok on 01/19/02 02:54 AM.</EM></FONT></P>
 

Crashman

Polypheme
Former Staff
It's the same method, yes. The difference here is that the Vid25mv pin used to be another VSS pin, so you have to use a different pin to get to VSS. Maybe I just got a suckier processor then them?

What's the frequency, Kenneth?
 
Sorry, I was forgetting about the Celeron FSB. That explains it then. How much FSB did you get from it?

<b><font color=blue>~scribble~</font color=blue></b> :wink: <A HREF="http://www.ud.com/home.htm" target="_new">Help cure cancer.</A>
 
Instability = either insufficient voltages or high temps. So I'm out of ideas unless you want me to send you a set of stablisers.

<b><font color=blue>~scribble~</font color=blue></b> :wink: <A HREF="http://www.ud.com/home.htm" target="_new">Help cure cancer.</A>
 

Crashman

Polypheme
Former Staff
hehe, yeh, stabilizers for a PC. Anywho, can't seem to get any stable overclock whatsoever, regardless of voltage modification attempts, so I'm stuck at stock!

What's the frequency, Kenneth?
 

phsstpok

Splendid
Dec 31, 2007
5,600
1
25,780
Don't give up! I think I have some help for you!

Check this thread at

<A HREF="http://forums.overclockers.ws/vb/showthread.php?s=&threadid=56519" target="_new">http://forums.overclockers.ws/vb/showthread.php?s=&threadid=56519</A>

The claim is that mod'ing the CPU does no good. The mod must be done on the iP3/T.

Looks like it's not for the feint of heart. The mod involves surface soldering and cutting traces on the iP3/T.

Here are the money shots.

<A HREF="http://koti.mbnet.fi/~alderan/mod/isot/vid.jpg" target="_new">http://koti.mbnet.fi/~alderan/mod/isot/vid.jpg</A>
<A HREF="http://koti.mbnet.fi/~alderan/mod/isot/katkaisu.jpg" target="_new">http://koti.mbnet.fi/~alderan/mod/isot/katkaisu.jpg</A>

(not my photos and I don't know who to credit)

<b>We are all beta testers!</b>
 
Curious as to which mobo you're using?

Is it <A HREF="http://fae.abit.com.tw/eng/faq/qa/2001/2001100401.htm" target="_new">Abit?</A>

I know that's not the type you're trying to overclock. Just wondered if you knew that one.

:cool: <b><font color=blue>The Cisco Kid</font color=blue></b> :cool:
 

Crashman

Polypheme
Former Staff
There seems to be some question as to whether it is necessary to perform such mods or whether the pin trick is adequate. Hoping someone with more money than I will work it out!

What's the frequency, Kenneth?