P4 1.8A @ 3.0 GHz!!

Toggle sidebar Toggle sidebar
T

Technonut

Diamond Member
Mar 19, 2000
4,041
0
0
P4 1.8A SL68Q Packdate: 09/04/02
ABIT IT7 (Flashed to Overclockers BIOS)
512MB Corsair XMS 3200 Rev. 2.1
Antec 550W TruePower PSU
80GB Maxtor D740X
Retail 128MB Radeon 8500 (Will install my 9700 next)
Acoustic Edge

FSB @ 167 / AGP 82 / PCI 41 MHz ;) (668 MHz Data Rate)

I started out with the P4 retail HSF just to see what it would do, and was way too hot at 2.75 GHz / 1.65V even after removing the TIM and applying AS3. I moved on to my original plan, which is watercooled with This custom made waterblock.

I removed the Northbridge HS from the IT7, applied AS3, and installed a nice Sunon 40mm fan. I then volt modded the P4 by joining the base of 4 pins with conductive paint to open higher BIOS CPU voltage options. (5 minute job)

After carefully applying a thin layer of AS3 to the P4 and installing the Black Ice original radiator and 500 GPH Danner pump, it was time to boot up and tweak. ;)

After all was done, I settled for the above specs. I have been looping 3DMark2001 SE for hours now with Prime95 running in the background. Just running Prime 95 for an hour, my temps did not exceed 55.5 C. This is very good considering that the IT7 is known to read temps higher than shown. :) Core voltage is reading at 1.79V. On to Sandra..........

CPU: 5760/3672
MM: 11913/14886

Memory: 3407/3406 :Q

PC Mark 2002

CPU: 7419
Memory: 7330
HDD: 708

I will post links to screen shots later. It feels good to set a goal of 3.0 GHz and reach it. :) I plan on letting the 3DMark2001 SE bench & Prime95 loop all night just to be sure, but it has been going for hours so far.

 
O

oldfart

Lifer
Dec 2, 1999
10,207
0
0
Nice overclock!
FSB @ 167 / AGP 82 / PCI 41 MHz
Click to expand...
Thats a bit over the edge of what I consider a good PCI speed. I dont like to go above 38 MHz. I thought the IT7 could lock AGP/PCI @ 66/33??
 
T

Technonut

Diamond Member
Mar 19, 2000
4,041
0
0
Thanks...
Thats a bit over the edge of what I consider a good PCI speed. I dont like to go above 38 MHz. I thought the IT7 could lock AGP/PCI @ 66/33??
Click to expand...
Yes, it does. I probably will have to set it at that when I throw my 9700 in. For the time being though, my installed components have always been able to handle those speeds. ;)
 
C

CrazySaint

Platinum Member
May 3, 2002
2,441
0
0
Sweeet OC, man! Especially since my chip has the same specs are yours :D

Core voltage is reading at 1.79V.
Click to expand...

Voltage is a bit high there, be careful you don't ruin your chip. Duvie and THUGSROOK have both ruined chips by running them at 1.8v. I know you're an experienced OC'er and probably already know that, but better safe than sorry ;)
 
T

Technonut

Diamond Member
Mar 19, 2000
4,041
0
0
Voltage is a bit high there, be careful you don't ruin your chip. Duvie and THUGSROOK have both ruined chips by running them at 1.8v. I know you're an experienced OC'er and probably already know that, but better safe than sorry
Click to expand...
Appreciate the concern. :) My watercooler is working very well though. The radiator and block is not even really warm. This will be a good test on the current Packdate. If it smokes..... RIP...It will have been a fun ride. :p
 
E

EdipisReks

Platinum Member
Sep 30, 2000
2,722
0
0
it's not the heat that kills them at 1.8v, it's the electromigration. the 13 nanometer copper process is very resistant to electromigration, but 1.8v is pretty high. nice overclock, regardless.
 
F

fluxquantum

Platinum Member
Oct 27, 2000
2,398
1
71
wow nice overclock. i overclocked my p4 1.6a to 2.3 but started running into stability problems. i upped the voltage and things were cool but still unstable. i bet you it's my ram. i am using rambus and i don't think rambus overclocks well.
 
T

Technonut

Diamond Member
Mar 19, 2000
4,041
0
0
any info on how well it did/does on a more conservative 1.7v?
Click to expand...
THUGSROOK, Not really... I just kind of went for it. I already knew that it would pull 2.75 GHz with stock cooling at 1.65V. I wanted to grab 3.0 GHz. :) BTW, were you air cooling at 1.80V when your CPU went out? I know of the process concerning electromigration, but I also know that it could take quite some time to occur.
 
T

THUGSROOK

Elite Member
Feb 3, 2001
11,847
0
0
yes aftermarket air cooling.

at the time me and Duvie were running identicle setups....
we both decided to see what we can get on 1.8v and we both fried them w/in a week. (temps were below 50*C)

:)
 
T

Technonut

Diamond Member
Mar 19, 2000
4,041
0
0
at the time me and Duvie were running identicle setups.... we both decided to see what we can get on 1.8v and we both fried them w/in a week. (temps were below 50*C)
Click to expand...
Sounds kind of like stuff Insane3D and I used to do "back in the day." We managed to pull off some impressive OCs with Slot A Athlons. :)

Don't get me wrong, it's not like I would enjoy pissing the cost of the processor away. It is just a hobby with certain risks that one has to be willing to accept. Better a processor than a high end racing engine huh? ;)
 
D

Duvie

Elite Member
Feb 5, 2001
16,215
0
71
Thise guys are behind the times and incorrect on some of their asumptions....Thugs and I had a lot of this stuff figured out back in may...


1) I lost my chip when I pushed it to 1.83v actual and it stayed there fine for one week and though it prime95 for 24 hours one week later while divx encoding it BSOD on me and the rest was torment....It didn't lose total oc capability...It still ran 1.8@2.25ghz with 1.58v fine for 2-1/2 more months until I replaced it anyways...

2) My chip was not using the wire trick as asus had no start issues since they overvikted and had an overvolt jumper...So that assumptions was clearly wrong...

3) Most system guying ou back then were water cooling system where guys pushed past 1.8v and thought they were fine cause they had insanely low temps....As mentioned it was reported those low temps actually may have caused damaged as well...The shifting from exrteme hot and extremem cold may have fatigued the components...Most water systems seem to go harder and faster then like mine and thugs that just started showing diminishing oc returns...

4) I have yet to see anyone lose a chip under 1.7v actual...yes actual not what is set in bios sinve they can often be overvolted severly by Asus mobos and give some owners wrong idea of there true voltage.

5) Temp is a factor but in most cases at failure ppls temps were very good with loading under 50c...well below 66-70 danger zone area...then calculate the factor that many of the boards are differing in temp monitoring severly as well.
 
T

Technonut

Diamond Member
Mar 19, 2000
4,041
0
0
Duvie, you bring up some interesting points concerning the thread at Overclockers.com. The
"extreme cold to extreme heat" theory regarding watercooling... I do not agree with. Unless using pelts, the waterblock temp on startup is not less than room temperature. It is no different than cold booting a system with an air cooled metal HSF.

Anyway.. I decided to really stress test my 3.0 GHz overclock. I have been running UT DM set as spectator with 16 bots (999 frag limit per map) automatically looping through many different maps. Prime95 is running in the background. It has been going for over 24 hours now. I paused the game to check the CPU temp, and was pleased to see it at 50.5 C. (The AS3 must have set in. :) )

I will continue to run this 24 hours a day all week. (Maybe 2 weeks if it does not crash) I will just continue running this rig (8K5A2+ / MP 2100+ @ 2.0GHz 421.36 MHz DDR System Bus) until my P4@3.0 GHz testing is complete. ;)
 
T

Technonut

Diamond Member
Mar 19, 2000
4,041
0
0
The BIOS is a modded it7_av bios from a Overclockers.com forum member with the following:

-increased options for ddr voltage.
-increased pci dividers
-overclockers.com screen mesage
-overclockers epa logo
-changed boot screen font color
-increased agp options

changed defaults:
-agp apature is at 128 instead of 64
-enhanced dram is enabled
-third boot divice is cdrom
-ddr voltage is 2.6 instead of 2.5
-ps/2 mouse sopport is disabled

The ABIT IT7 MAX was purchased from Newegg. :)
 
S

sean2002

Golden Member
Apr 9, 2001
1,538
0
0
Data corruption waiting to happen. Can you show a screen shot of Prime95 running for any long period of time? 8+ hours
 
T

Technonut

Diamond Member
Mar 19, 2000
4,041
0
0
Data corruption waiting to happen. Can you show a screen shot of Prime95 running for any long period of time? 8+ hours
Click to expand...
Prime95 has been running in the background with UT running for over 24 hours now. Like I said above, I only paused UT once to see the temp. I will be more than happy to post a screenshot of Prime95 running for 24 hours after my stess testing is complete. :)

As far as data corruption is concerned, I have been pushing very high PCI bus speeds for years with Maxtor hard drives and the Highpoint controller on various mobos. From what I gather (and experienced) , the HP controller is more forgiving at a high bus speed than the onboard IDE controllers.
 
D

Duvie

Elite Member
Feb 5, 2001
16,215
0
71
heat to cold extreme and no new topic and has been supported by some of the guys in the field around here as a danger to chips as well...Are you extremely cold and at what temp I am not sure if it applies to you....
 
T

Technonut

Diamond Member
Mar 19, 2000
4,041
0
0
Duvie, to quote myself:
Unless using pelts, the waterblock temp on startup is not less than room temperature. It is no different than cold booting a system with an air cooled metal HSF.
Click to expand...
The water and waterblock temperature is not going to be below the room temperature at start up unless you have a peltier installed. (Hence, no extreme cold) I just have a waterblock, radiator, pump, and reservoir.
 
You must log in or register to reply here.

TRENDING THREADS

COMPANY
RESOURCES
FOLLOW
Top Bottom