Official HWBOT Partner
Official HWBOT Partner

Submission Details

29 sec 703 ms with
Elite Submitted

Images

Hardware Details

Processor

Cooling
Liquid Nitrogen
Cores
3,581.00 MHz  (91%)

Memory

Cooling
Air (Custom)
Type
512 MB  DDR SD-RAM Speed @ 256 MHz
Timings
tCAS: 2.0  tRCD: 2  tRP: 2  tRAS: 5
Batch
BH-5

Videocard

Model
GeForce4 MX 4000 64-bit (NV18) (GeForce 4 series)
Cooling
Air (Stock)
Speed
250.00 MHz  /  200.00 MHz

Motherboard

Cooling
Air (Custom)

Comments

ruggero
ruggero

...It's a incredible frequency :D

 

Do you have tested on air this cpu?

NoMS
NoMS

This is the first ever sub-30s PI 1M on Athlon XP or I miss something? :D

pcwerman
pcwerman

You're right ! This is world first sub 30 sec.I think he is wizard... :D

Mr.Scott
Mr.Scott

Way to go Ivan. :celebration:

I.nfraR.ed
I.nfraR.ed

Ruggero, I've tested it on water only. 2900+ 1M with vcore < 2V. Valid 3000+.

ruggero
ruggero
Ruggero, I've tested it on water only. 2900+ 1M with vcore < 2V. Valid 3000+.

 

Ok,thank's and again congratulation for the score.

Massman
Massman

Whoah ... just whoah!!

Moose83
Moose83

Mother of AMD, you guys totally nailed it :D

Maggehagge
Maggehagge

Did you forgot that nForce 2 chipset always works best with x-x-x-11 timings? both stability and performance IIRC. or maybe superPI is faster with 2-2-2-5 instead of 2-2-2-11?

I.nfraR.ed
I.nfraR.ed

No, I haven't forgot, but 2-2-2-5 is faster than 2-2-2-11 on my board/bios.

Mr.Scott
Mr.Scott
No, I haven't forgot, but 2-2-2-5 is faster than 2-2-2-11 on my board/bios.

 

Mine generally are also. ;)

I.nfraR.ed
I.nfraR.ed

Nothing special - 1.85V or so. Don't remember exactly, but this particular board does not scale past 2V on the chipset.

TerraRaptor
TerraRaptor

Okay, thanks. I need >2.3v to go 240fsb 2-2-2 on A7n8x-e deluxe (2.0v - no go, only 230mhz 2-2-2) - it works fine in 1M but cannot complete 32m (dunno if this is overheating, missing 5V caps (desoldered them to mod different mobo) or memory - should check this today). My old result was commented by myself that I had used 2.48v vdd to go 250 2-2-2 and I'm not sure if this is clever/safe thing (I use stock radiator on NB).

I.nfraR.ed
I.nfraR.ed

Try with higher volts on the 3.3V rail. Modded bios is also very important, but I guess you're using one, since you're able to do 250MHz. You can also try L12 hardware mod.

 

I haven't yet killed NF2 chipset with voltage, but if you're not cooling it fine, these volts won't help you.

 

Generally speaking, my boards does not scale good with NB voltage. I have only one board that likes high volts and another one that can't complete 32M at 250MHz with more than 1.6V on the NB :)

 

I've also had NF7 that can't do more than 250MHz for benches no matter volts/bios/cpu.

TerraRaptor
TerraRaptor

Thank you Ivan. I have resoldered caps, changed nb TIM and went to 3.7v - now everything is stable @252 2-2-2 but one of my winbond sticks suck - now i need to get my 2*256mb and 2*512mb winbonds sets back from friend and I'll be able to enter socket A competition.

d3mox
d3mox

Can you tell me how to mod 3.3v rail?

I.nfraR.ed
I.nfraR.ed

There are plenty of PSU modding guides over the net, you can check x-bit's guide or get a PSU with adjustable rails.

TerraRaptor, I've never went up to 3.7V.

TerraRaptor
TerraRaptor

D3mox I had a standalone psu with 3.3v mod - just because overvolted 3.3 rail is not that good for other setups. It is also important to have a psu that is able to provide as much power as possible @3.3+5.0v rails (i.e. my current coolermaster 800w has 150w which leads to shutdowns of a7n8x @2ghz barton aircooled, my old huntkey was rated 180w and so was better).

GraduS
GraduS

Very result! When I finish with s370 I try socketA too.