Fastest PIII ever?

[sparkY1000]

Just wondering if anyone knows what the fastest PIII ever is?

I have checked ripping.org and hwbot and the fastest seems to be 2130MHz

http://valid.x86-secret.com/show_oc?id=195002

Has anyone seen higher?

[sparkY1000]

Anyone?

[Digital Dero]

everyone must be busy viewing the fastest 386 thread?

[sparkY1000]

lol I know its old tech but there is still a lot of fun to be had.

[EnJoY]

That would have been super-fast back in the day. Nice short stage pipe-line, good IPC.

[sparkY1000]

I managed this myself the other day...

http://valid.x86-secret.com/show_oc.php?id=340157

Just want to know how much more I need for PIII WR

[RunawayPrisoner]

120MHz, I'd guess. Good luck.

[sparkY1000]

Cool, I will post up some results when I manage it

[GraduS]

Good luck.

[sparkY1000]

Thanks GraduS

Have you got a Celeron 1400 to try?

I have got all the Tualatin Celerons now but my CUV266 wont go past 167MHz FSB at all tried maxing chipset voltage but it wont go any higher cant work it out anything worth trying?

[GraduS]

Thanks GraduS

Have you got a Celeron 1400 to try?

I have got all the Tualatin Celerons now but my CUV266 wont go past 167MHz FSB at all tried maxing chipset voltage but it wont go any higher cant work it out anything worth trying?

No. I wish to try 1300@1850(air) on LN2.
I pass 198fsb on cuv266, but only for cpu-z. http://valid.x86-secret.com/show_oc.php?id=330891
I made Caps mod, bios mod, Vmem, Vmch, Vcore, Vgtl but she wont go past over 185fsb stable.

[jcool]

Get the Abit ST6 with MCI PC150 SDRams. Will give you the most FSB, I am still running a P3-S 1,13 at 170 FSB with stock cooling.
Also the P3-S are way superior to the Celerons... just hard to geht the 1,4Ghz P3-S up high because of its default 133Mhz FSB and its power needs at a decent Vcore being too much for any unmodified mobo of that day.

[GraduS]

Get the Abit ST6 with MCI PC150 SDRams. Will give you the most FSB, I am still running a P3-S 1,13 at 170 FSB with stock cooling.

1:1 fsb:ram?
I test tusl on i815 with SDR, she can run 210+fsb. But this system more slowly then cuv266 with 1:1 DDR.

[jcool]

Yeah 1:1. I thought you were trying to achieve the highest clockspeed possible?
If you are looking for the most in absolute speed, I'd think there is no way around a tB1 1,4Ghz P3-S. It's way faster than a Tualatin Celeron even at lower clockspeeds, due to the 512K L2, lower L2 latency as well (0 namely), the HW data prefetch unit also seen in the Athlon XPs and many more things.

I frequent a German HW forum called tualatin.de, it evolved around the Tualatin core CPUs in 2002.
Nowhere on the web is there a more comprehensive collection of articles, comparisons and reviews about Tualatin- and P3-related hardware.
I remember reading that a Celeron would need like 300Mhz more clockspeed to match the P3-S.

[sparkY1000]

No. I wish to try 1300@1850(air) on LN2.
I pass 198fsb on cuv266, but only for cpu-z. http://valid.x86-secret.com/show_oc.php?id=330891
I made Caps mod, bios mod, Vmem, Vmch, Vcore, Vgtl but she wont go past over 185fsb stable.

I was amazed by that score with your PIII-S it was fantastic! I would love to get sub60s pi1m with a PIII but it would take stability at 190MHz+.

I might have to look into mods for Vmch and Vgtl got plenty of voltage for RAM as it is and can increase Vcore to 3.5V lol! For caps mod do you mean you have replaced all the important caps? BIOS mod? Wow! You know x86 assembly code?

Without a doubt the L2 cache of the PIII-S 1400 is far superior performance wise but all out frequency for a PIII you need a tB1 Celeron most likely a 1300 or 1400.
Got a 1300 the other day to try and hoping for some nice results with my phase and plenty of volts.

[sparkY1000]

Was trying my 1300 today and everything was going well got to just shy of 2GHz without any problems but I think I had tightened my phase mount upto much so the lug I use to attach my phase change mount onto snapped almost breaking my suction line! :O

Might have to get a new adapter if I cant fix it

[RAW-Raptor22]

That would have been super-fast back in the day. Nice short stage pipe-line, good IPC.

Probably like a 3.2GHz P4. My 1200 at 1550MHz beats 2400MHz P4Bs at benchmarks...

When I get my 1300 rig up I will bench that badboy too.

[GraduS]

I remember reading that a Celeron would need like 300Mhz more clockspeed to match the P3-S.

Yes. Its truth.

I was amazed by that score with your PIII-S it was fantastic! I would love to get sub60s pi1m with a PIII but it would take stability at 190MHz+.

thx. I think 200+ fsb

For caps mod do you mean you have replaced all the important caps?

I replace +5v, +12v, cpu, mem, mch caps.

BIOS mod? Wow! You know x86 assembly code?

No) I know other method

[GraduS]

I start playing with TUSL2-C:
p3-s 1400 SL6BY 2104 : http://valid.x86-secret.com/show_oc.php?id=377716 WR
celeron 1200 SL6C8 2119 : http://valid.x86-secret.com/show_oc.php?id=377714 WR
celeron 1400 QKE0 QS 2227 : http://valid.x86-secret.com/show_oc.php?id=377363 WR

[RAW-Raptor22]

I start playing with TUSL2-C:
p3-s 1400 SL6BY 2104 : http://valid.x86-secret.com/show_oc.php?id=377716 WR
celeron 1200 SL6C8 2119 : http://valid.x86-secret.com/show_oc.php?id=377714 WR
celeron 1400 QKE0 QS 2227 : http://valid.x86-secret.com/show_oc.php?id=377363 WR

Wow, that is really awesome dude, props!

[jcool]

Amazing. What's the max you can run Pi 1M at? With the P3S that is, the Celerons won't come close there.

[GraduS]

[cTx] Raptor22, jcool thx)

I can run pi1M at 2080Mhz 198fsb and only 148 2-2-2 mem, result 1m16s. I don' have 200mhz ram....((((

[jcool]

Hm, well. You'd need MCI 150 mems, they can run ~190 CL3 or even 200, I don't know. Runs 170 with 2-2-2. They were made in Germany back then, best SDRams ever.

[celemine1Gig]

Hm, well. You'd need MCI 150 mems, they can run ~190 CL3 or even 200, I don't know. Runs 170 with 2-2-2. They were made in Germany back then, best SDRams ever.

To be honest, I don't think that it has to be MCI PC150 RAM. I think that MCI never produced their own chips. All they did was buying chips from other manufacturers and binning them to get the highest performing ones. At least that's what they do today (roughly). So, in conclusion there should be chips from other manufacturers out there that should perform identical. I don't know which chips you should be looking for, because at that time back then I had no clue. But I'm pretty sure that there are alternatives.

[STEvil]

I've got 2x256 6ns SEITEC ram.. pushed it over 175mhz 2-2-2-x on Abit KT7A-RAID many many years ago.