AOA Forums

AOA Forums (http://www.aoaforums.com/forum/)
-   Intel Motherboards & CPUs (http://www.aoaforums.com/forum/5-intel-motherboards-and-cpus/)
-   -   Celery 1.2 @1.7 or 1.8Ghz? (http://www.aoaforums.com/forum/intel-motherboards-and-cpus/1314-celery-1-2-1-7-1-a.html)

Kris 17th November, 2001 05:27 PM

Celery 1.2 @1.6-1.8Ghz?
 
Who's getting 1.6-1.8Ghz? (or hoping to get there :eek: )...

Seems MANY are getting outrageous overclocks out of these new Celerons with the SK6 HSF... looks like the BE6-2 & ST6 are good boards.. I have a TUSL2C headed my way...

Anyone done the pin mod yet?

Pinky 18th November, 2001 01:18 AM

I've done the PIN mod to get 1.675V (think they were vid1 and vid2).

1.50ghz is a reasonable overclock without taking off the IHS (integrated heat spreader), but you'd need to remove it to go higher, and most of the time that's under extreme conditions.

I will be able to better answer your question in about a week ;).

Kris 18th November, 2001 01:54 AM

Saw your comments Pinky in the same thread with some other folks claiming to be seeing 1.6 to 1.7 pretty regular...

I'm not to sure about taking off the IHS myself... although I did lap the die of my P3 SECC chip to reduce temps...

I understand the die is very small... can't imagine trying to hang 2+ lbs of copper off of a small die like that... *crunch* not without shims or rubber feet or some kind of protection.... :D

dimmreaper 18th November, 2001 04:18 AM

"Integrated Heat Spreader", what a Joke! They should have called it a "Clumsy Technician Protection Mechanism" . . . .

<rant>

I often here folks, saying "AMD should put a heat spreader on there CPU's". People need to realize the the shiny thing on the new Intel cores makes them run hotter. Just as the plate that protected the core of the K6-X flip chips did. Yes the K6-X were flip chips, I have removed the caps from many of them in the quest for clock speed. The reason why AMD did not carry the cap from the K6-Xs over to the K7-2s(that's thunderbird) was that it would hinder clock speed considerably. If one thermal junction is bad, two is horrible . . .

</rant>

Kris 19th November, 2001 05:46 AM

From what I understand there is little to no variation on temps full load/no load with the Taulatin... :cool:

Pinky, that about right with you too?

dimmreaper 19th November, 2001 06:01 AM

Pinky, just out of curiosity. What temps do you see with your Tulip, and what did you see with your Crappermine?(I have nothing against the P3, I own two of 'em) I assume both CPUs were in the same board, yes?

Pinky 19th November, 2001 08:02 PM

Quote:

Originally posted by dimmreaper
Pinky, just out of curiosity. What temps do you see with your Tulip, and what did you see with your Crappermine?(I have nothing against the P3, I own two of 'em) I assume both CPUs were in the same board, yes?
I see less difference between idle and load temps with the tullie (only 3-4C spread) as opposed to the regular P3s I've had, which could go as high as 10C from idle to load.

I am working on developing some 'feet' for my glaciator II and delta before I start experimenting with what could be a fragile core.

if you don't hear from me for a few days you know why :).

Kris 19th November, 2001 08:35 PM

Pinky is there any similarity between the Millenium Glacier & the Galciator II ?

I can post a pic if that wil help...

dimmreaper 20th November, 2001 04:33 AM

I wasn't so much concerned about the temperate "spread", or range, as the actual temperatures. How hot does the tupil get as opposed to the cD0?

dimmreaper 20th November, 2001 04:35 AM

Quote:

Originally posted by Kris
Pinky is there any similarity between the Millenium Glacier & the Galciator II ?
The Glaciator II has a couple of pieces of the base machined out. This serves two purposes.

1) It promotes more air flow past the fins.
2) It allows air to blow directly on the PGA package, which inturn cools that package, which inturn tricks in socket thermistors in to thinking the Glaciator II is a lot better than the Glaciator I than what it is.

Kris 20th November, 2001 05:23 AM

So what's the verdict on the Glacier (you have one, yes?)?

I have never used mine, but thought I might give it a shot... now, you've got me wondering.. :(

dimmreaper 20th November, 2001 05:36 AM

I use my Millenium Glaciator all the time. It kicks arse . . .

dimmreaper 20th November, 2001 05:59 AM

Quote:

Originally posted by Pinky
I am working on developing some 'feet' for my glaciator II and delta before I start experimenting with what could be a fragile core.
Perhaps this Idea I've been toying around with will be of use to you Pinky.

Squeeze a bead of silicone rubber around the outside edge of the top of the CPU package. Then turn the CPU upside down and set it on a piece of wax-paper(on a flat table or similar surface) until the silicone hardens. The theory is that this will give the core some protection against crooked HSFs.

I got an old 500E that I could try this on if your interested.

Kris 20th November, 2001 06:11 AM

yes... Jeff... this works well... I have a 486 hetasink & fan on my little ICS clock chip on my P3V4X mobo... used the same basic tactic...

Any ideas for removing the heat spreader?

dimmreaper 20th November, 2001 06:25 AM

Quote:

Originally posted by Kris
Any ideas for removing the heat spreader?
Xacto knife? Bench grinder? Heck I have no Idea what is holding it on the Tulips, so can't offer much advise except experiment to find out, and perhaps damage a CPU or two in the process.

Kris 20th November, 2001 06:33 AM

Quote:

Originally posted by dimmreaper
I wasn't so much concerned about the temperate "spread", or range, as the actual temperatures. How hot does the tupil get as opposed to the cD0?
Thought you might like this on the Celery 3:

Quote:

the first thing I noticed is that it it does not take much to cool it - at 1200 MHz and 1.475v, it radiates only 29.9 watts. At 1680 and 1.55 v, it radiates 49.3 watts, about half what you would expect at about the same speed with an overclocked T-Bird.

The second thing I noticed was the minimal temp rise when stressed - about 2C over idle temps. According to Motherboard Monitor, idle temps were 37C and under Prime 95, 39C (these are on-die temps). Under stress, watercooling the CPU resulted in a rise of not more than 4.5C over ambient temps (measured at the waterblock's base).

The overclocked temp rise over spec was minimal - I ran Prime 95 at 1200 MHz with a reported voltage was 1.42v (BIOS set at Default 1.475) for an hour and got a delta of 2.7C, with on-die temps of 33C, at an ambient temp of 20.6C. Overclocked to 12 x 138, the temp rise was only an additional 2C at the waterblock and 6C on die compared to temps at spec.

The third major item I noticed is that hiking the voltage beyond 1.65v did not make any difference in performance. The most stable speed I attained was 12 x 138; this required only 1.55 volts.
Not so bad.... :D

dimmreaper 20th November, 2001 06:37 AM

Thanks Kris! I can't wait to see how hot AMD's .13 CPUs are going to get.

Kris 20th November, 2001 06:51 AM

Agreed... this lastest die shrink is really going to put the bite on HSF manufacturers I think :)

Besides... if the P3 did so well @ .13... then shouldn't the K7 be phenomenol?

dimmreaper 20th November, 2001 07:10 AM

Quote:

Originally posted by Kris
Besides... if the P3 did so well @ .13... then shouldn't the K7 be phenomenol?
It is reasonable to assume the the K7 will do quite well. Especially when you consider that the high-voltage low-current(low being a relative term here) design eliments of the Palomino core will be carried over. Such designs are typically quite scalable. I won't be suprised if the AMD .13 parts debut with core voltages of 1.5 or even 1.6V!

Some of the three-phase voltage regulators used currently on SocketA motherboards can be implemented in such a way as to provide as little as 1.075 volts(in .025V incroments), this means that there is a good chance that future die-shrinks(.09mcron) will be supported by such boards.

dimmreaper 20th November, 2001 06:55 PM

Just though of something Kris. If you really want to push the a Tulip, why not go for some water/peltier goodness? A small 118W peltier would be sufficient.


All times are GMT +1. The time now is 08:41 PM.


Copyright ©2001 - 2010, AOA Forums


Search Engine Friendly URLs by vBSEO 3.3.0