• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

OCing an oldie - Celly 366

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

gofra

Member
Joined
Jun 8, 2005
Greetings Intel bunch!

I just had a funny and joyfull experience OCing my GFs Celeron 366 and just out of fun, I decided to post my results here.

L_Intel-FV524RX366%20(bottom).jpg

The heart...

bigfan.jpg

Preparation - do you see the AXP Barton box cooler on? Heh, it fits just great! I was worried a bit how the socket card will handle the extra weight but then I realised that it sits on on of the DIM slots nicely. I took off small alluminium heatsink and its 50mm stock fan and mounted on a very quiet low RPM 80 mm fan that was only bolted on one side (since it was to big). So, a little inovation was neccessary:

inovative.jpg

Weee...perfect so far!

Soon I realised, that the NB under this giant cooler will have warm airflow over itself, so I decided to mount on a small ALU heatsink that came on my Epox board as a default. Again (almost unbeleviable), it fitted so nicely that the CPU heatsink above leaned on totaly horizontaly. Now, the problem of warm air was still here. I decided to "reverse mount" the stock CPU fan above, on the big heatsink...like this:

faaans2.jpg


Again, the bolts only came on one side. And since this one was quite a screamer (I opened and cleaned it but no good) I had to put two rubbers in between. Here's how it came out:

lilfan.jpg


Ok! Off to the joy of OCing! I discovered very soon, that my multis are locked at 5,5. After some reading, I found out no one so far was able to unlock them. So what now? Well, raising the FSB seemed like a great idea but I was affraid to do it. Namely, the stock 66 FSB makes the PCI use 33 and AGP 66 which is normal. I only had the chance to raise the FSB to somewhere around 70 or 85. That would probably make PCI and AGP work at funny speeds and I just didnt want that. Then I took a closer look at the slot-card and read through the manual of a simmilar board. There were two jumpers, one on the card and the other one on the board - both with the sign TURBO66 on. Since it was allready working at this speed I took the jumpers off. My oh my!

voila.jpg


POST! With 50% OC! Now, Windows wasnt installed yet, so there was no fear of corrupt data yet. Again, I went Googling for some info and found out that either 66 or 100 MHz FSB will set the PCI and AGP speeds correct. Wow, nice!
But prior to that, I set all back to default and installed Windows 98 on. After I was done with it, I took the jumpers off again and set the voltage (again, with the jumpers on the slot-card) a bit higher (from 2.0V default to 2.1V).
The moment of truth...Reboot...POST...OS loading...entered WINDOWS! Yeeha! But soon, my system froze. So I raised the voltage even higher, up to 2.2 this time and repeated. Worked. Tested some programs...worked. I even ran Super Pi and it did allright up until the 512K test. Bummer! At 2.3 I couldnt start Windows...at 2.4 neither. The cooler was warm but no big deal so I raised it up to 2.5. Still no go. The board had no sensors on so I couldnt read the fan speeds or the temperatures and decided to give it a rest. So much for my first Intel OCing experience.

Just out of curiosity, I set my jumpers back to default 2.0 V and gave it another try (I found it strange that higher voltage didnt work). Windows worked. Super Pi did too. 256K...512K...even 1M!! I was surprised. 2M passed allright and so did 4M! Now, since this is a Celly 366 and my GF only uses it for Word, Firefox and Excel I stopped testing it. Im not realy sure about the stability but I got a strong feeling it will do just fine.

Man, 50% OC at stock voltage...Call me sentimental, but those were the days.

My first Intel OCing experience - pure fun!



Cheers!
 

WillysNut

Senior Jeep Nut
Joined
Apr 18, 2001
Location
NY, USA
Ahhh the days of old...You're bringing back some memories for me. Congrats on your first successful OC. Great writeup. Enjoy the fruits of your labor!
 

Ssetre

Member
Joined
Aug 14, 2004
Location
Saskatchewan, Canada
HeH looks like you could cool a Thunderbird athlon with all that.

I got a 366 that ran stock for about 5 minutes the rest of the time it was bangin away at 550.

I just need to find the vid-pin guide for doing up a Coppermine.
 

ropey

Member
Joined
Oct 6, 2001
Location
Canada
Mr_Fuchs said:
366mhz and 550mz.. at this point in the game its the same thing,
Not so. Using windows 98 with 366MHz on a 66MHz bus vs 550MHz on a 100MHz bus WILL show quite a difference in system use.

Maybe no difference in your game but on the proper software, benchmarks and real life WILL show a marked difference.

R
 
OP
gofra

gofra

Member
Joined
Jun 8, 2005
..and what a difference it is. I was lucky there was a PC 100 and not 66 memory stick on, otherwise I might had some problems. Seriously, its way quicker. OS loads almost 100% quicker, all the test times I first tried were cut to 2/3 at least. Surfing? Same...So yes, its a big difference.

But true, its an oldie and it realy makes no difference if you compare it to the systems of the present day (which, Im sure is what Mr_Fuchs wanted to say ;).

/offtopic:

btw, Ssetre, this one is a Mendocino - Coppermines were faster. But Im sure you'll find plenty of material on the net. Today for example, I found out folks running 550 with stock cooling and voltage. Makes you think huh?
 

ropey

Member
Joined
Oct 6, 2001
Location
Canada
That this is the Mendicino is the proof of your excellent overclocking work as very few .25 micron 2v Mendicino's were able to make the jump from 66MHz to 100MHz. Most made the 75MHz bus jump, less the 83MHz jump and even less the 100MHz jump even though the 100MHz jump brought the AGP/PCI bus back into spec.

Personally, I have never found one that did allow me this however, I have never completed the modifications as you did either.

So, as I said, "Well Done"!

As an aside:

Almost all .18 micron 1.6v Coppermine Celeron II's made the jump easily and with no modifications necessary at all. The die shrink allowed for a fair voltage drop and the Coppermines ran extremely cool thus were able not only to make the move from 66/100 but some also made the move from 66/133. I have personally taken Coppermine Celeron's such as 766's @ 66MHz to ~1.GHz @ 133MHz. Intel moved to the 100MHz bus after the 766MHz Coppermine Celeron II's and thus the 800MHz Coppermine Celeron II was 8 X 100. These also made 1GHz fairly easily.

R
 

Ssetre

Member
Joined
Aug 14, 2004
Location
Saskatchewan, Canada
Mine is a 366 Mendocino 2.0vcore and it does 550 all day with nothing more than a simple flick of a jumper.

My Celeron-II 633 does 950 all day as well with a flick of a jumper. But I have to pinmod to get more voltage. And all the information has been yanked from the net :(.

Ah back in the days when a Celeron didn't suck compared to its Pentium brother.
 

batboy

Senior Moment
Joined
Jan 12, 2001
Location
Kansas, USA
Sort of reminds me of back in the good ole days when I had my trusty 300A Celeron. I ran it at 463 MHz (103 FSB) for well over a year (65% O/C). Sold it to a buddy who made some cooling improvements similar to what you did and with 2.2v he was able to hit a stable 504 MHz. Those old overclocked Cellies with full speed on-die cache smoked the more expensive P-II's running at the same speed.
 
OP
gofra

gofra

Member
Joined
Jun 8, 2005
Uhm...closing on 600 now - still doing exceptionally well!

I almost caught up with my ole man's "newer" Duron 650 :). But hey, I bet I can take that one up into the "heights" too...