XP1700B DUTC Overclocking/Undervolting Results

[ral]

CPU: AXDA1700DUT3C 9231796250192 JIUHB0308RPKW
Motherboard: Epox 8K9A
Cooling Volcano 7+ (medium speed)
OS: Win98SE

I got the following results overclocking my cpu. To trest stability I use multiple sessions of Raven Shield and Rowan's Battle of Britain. My system is stable at:

1500MHz (9*166) - 1.4v (mobo min vcore)
1584MHz (9.5*166) - 1.4v
1667MHz (10*166) - 1.4v
1750Mhz (10.5*166) - 1.4v
1834MHz (11*166) - 1.475v
1916MHz (11.5*166) - 1.55v
2000MHz (12*166) - 1.625v
2083MHz (12.5*166) - 1.70v

Did not trst beyong 2083MHZ

From the post I have read on these boards, I gather my cpu is an average overclocker... but seems to be a pretty awesome undervolter

Posted for info purposes.

[SteveI]

Thanks for the info... very useful indeed.

[ral]

SteveI

Your estimate of th max potential of my CPU may be dead on...

Looks like the CPU is very predictable... after 1750MHz I need an additional 0.75 to gain aother 83MHz. I might actualy reach 2.25 GHz if I used a 1.85 vcore...and maybe even 2.5 with 2.0 vcore.

[SteveI]

Unfortunately, with most CPUs, you start hitting diminishing returns. At some point, it will take a lot more voltage to get a little more mhz. Hopefully, you're not there yet...

[ral]

Might try 1.775v deep in the cold of night (well it nevere really gets that cold here) and see if I can reach 2170MHz... and if that morks might try 1.85v next...

I dont think my HSF is a good choise for trying 2.0v

But for everyday use I have kind of settle with 1834MHz at 1.475v.

ral runs away so his cpu can fight another day.

[Liquid3D]

Originally posted by SteveI
Unfortunately, with most CPUs, you start hitting diminishing returns. At some point, it will take a lot more voltage to get a little more mhz. Hopefully, you're not there yet...

Isn't that the truth! I have a Prometeia on the way, but my Vantec/Thermalright combo has yeilded unbelievable results. I see my temps as lower then most wantercooled setup's as this recent pic below will show, and currently my system is folding at 30%. But I've hit that ceiling where I just couldn't get past 2400MHz at 2.1V. Perhaps with the Prometeia it'll change, but considering at 2.1V the temps were still under 25C I don't see how, yet pray regardless.

But it's funny you posted this, because I just tried an experiment where I lowered my 2400 AIUCB to 1.50V from 1.675V, which I upped it to from 1.65V because I was/am running the CPU at 12x166. So I'm now going to begin re-clocking at the lower voltages, and I wonder what the true bottleneck was when I attempted to reach 2400MHz +? Each time I got an corrupt sysfile error which is usually indicative of HDD problems, even so when overclocking I put my HDD in PIO Mode-4, in BIOS and WindowsXP. So it's back to the drawing board to establish an entire set of benchmarks prior to the arrival of my Prometeia. I just hope my replacement Tyan Tachyon get's back soon. Anyway thankyou for providing this valuable undervolting info. What could be more fundamental to CPU potentiality then thermal dissipation, hence current.

[ral]

SteveI...

I think I am reaching the point of dimishing returns. I was able to get into Windows at 2167MHz (13x166) @ 1.75v. But the system hngs shortly after. Went up to 1.775 and got into Windows but programs have errors when I launch or use them. Reven Shield stable at 1.8v.

Looks like I would need at least 1.9v to get to 2250MHz. Looks like I am going to really really stop here. Temps are 7-8 degrees above default.