Sentential03's burn in's really do work for me.

Tat · Nov 6, 2004

Man.. I really wanna try this out but my prime wont freaking load 100% in blend test with piority 10 since I reinstall the window. Dam it I have no idea why it is doing this

. So I cant really know if this gonna work. But I really wanna push my vcore down a little :bang head

. Oh well, if anyone know how to fix this problem please let me know

CandymanCan · Nov 6, 2004

Tat said:
Man.. I really wanna try this out but my prime wont freaking load 100% in blend test with piority 10 since I reinstall the window. Dam it I have no idea why it is doing this . So I cant really know if this gonna work. But I really wanna push my vcore down a little . Oh well, if anyone know how to fix this problem please let me know

You need to run Cpu Burn-In to burn the cpu in not prime95. http://users.bigpond.net.au/cpuburn/

Prime95 is only to test the stability afterwords.

I have an Update now. I ran the burn it for a few hours @ 2550mhz 1.65v a nd before which alot of people already know that anything beyond 2500mhz for me was unstable and always failed prime95 after a few minutes. Well i just stopped running prime95 after my burn in @ 2550mhz 1.65 and i got ZERO error's after 2 hours and 37 min, when prime95 before would fail after like 5 minutes.

So 2550mhz is now stable for me. :clap:

, Im going for 2600mhz and if i cant get that stable then im ganna try and see if can lower my vcore on the 2550mhz setting 1.6or 1.55v is my goal @ that speed

felinusz · Nov 6, 2004

Do you think that this might work for a GPU as well, applied slighty differently?

I.E. keeping the GPU Vcore at stock, but overclocking it to the very brink of stability (minor artifacting), and then stressing it. Then overclocking it even farther with the same stock Vcore after stressing it, repeating the stress - after awhile upping the Vcore a little bit, and taking a small speed step upwards?

But would stressing a GPU at an unstable (minor artifacting) speed do damage to the card?

This sounds like an interesting concept, I'm going to try it out on my Mobile Barton 2600+, possibly my 9800 Pro R350 as well, if I can get the guts together to try it

Man.. I really wanna try this out but my prime wont freaking load 100% in blend test with piority 10 since I reinstall the window.

I'm having the same problem right now - inexplicably, after I reinstalled Prime95 on my WinXP partition, the program will no longer bring my CPU up to 100% useage. It'll go up to 15-30%, and then fall to ~7%, never making any progress past the first test - although not erroring or freezing either.

Prime95 still works fine on my Win2K partition though... very strange.

Aphex_Tom_9 · Nov 7, 2004

okay, im not quite sure if it helped, but i ran it for 3+ hours, and now im priming at 16.5x172 (2842mhz) at 1.98 vcore, which i believe is a sold improvement...not sure though, this chip's been acting pretty funny. Stepping closer and closer to 3ghz...my goal is to run 16.5x182 (3003mhz) or maybe 17x182 (3094, call it 3.1ghz) stable.

Avg · Nov 7, 2004

How long should I let it burn in?

Adaman · Nov 7, 2004

Sounds cool...i'll have to try it. It's hard to find the time though, guess i'll have to do it while i'm sleeping, heh. My CPU is a Vcore hog.

Thanks for the info.

Zebbo · Nov 7, 2004

Avg said:
How long should I let it burn in?

from eight to ten hours

BeerHunter · Nov 7, 2004

subtotal said:
theres a thread (somewhere) that details that, the leading theory at the moment is that it improves some of the transistors in the cpu

like to see that.

Tat · Nov 7, 2004

CandymanCan said:
You need to run Cpu Burn-In to burn the cpu in not prime95. http://users.bigpond.net.au/cpuburn/

Prime95 is only to test the stability afterwords.

I have an Update now. I ran the burn it for a few hours @ 2550mhz 1.65v a nd before which alot of people already know that anything beyond 2500mhz for me was unstable and always failed prime95 after a few minutes. Well i just stopped running prime95 after my burn in @ 2550mhz 1.65 and i got ZERO error's after 2 hours and 37 min, when prime95 before would fail after like 5 minutes.

So 2550mhz is now stable for me. , Im going for 2600mhz and if i cant get that stable then im ganna try and see if can lower my vcore on the 2550mhz setting 1.6or 1.55v is my goal @ that speed

I'm solid stable with my setting rite now. Just that I have to use 1.875 bio set (1.824 -1.81 reading) to get to this speed. I've 24 hrs primed and solid 3D and everything then I reisntalled the window. Now, I just cant prime anymore so there is noway, I cant do stablize test

if I wanna burn it and lower my Vcore. I've tried to reinstall window twice but samething happened.

cV · Nov 7, 2004

Burning-in: The witch-doctor tale of modernity.

amd_luvarboi · Nov 7, 2004

cV said:
Burning-in: The witch-doctor tale of modernity.

Just might be worth a try though.

enduro · Nov 7, 2004

then I reisntalled the window. Now, I just cant prime anymore so there is noway, I cant do stablize test

You reinstalled Windows, and now prime can't pass? Hmmmm. Interesting. I'll have to trade with my little brother for a minute and see if I can pass Prime with his ram. That would be an interesting test.

felinusz · Nov 7, 2004

Burning-in: The witch-doctor tale of modernity.

There's no hamr in trying, especially since many people have seen some signifigant MHz:Vcore gains after trying out various burn-in techniques. I personally have never seen any signifigant gain on my stable maximum MHz/Vcore, although I still burn-in every chip that I get (and I've been through quite a few). Why not?

Wouldn't this burn-in technique fly in the face of the "other" widely-accepted burn-in procedure, whereby one lowers the MHz and temperature to a minimum, whilst raising the Vcore voltage as high as possible to condition the chip?

I did this to my Mobile Barton for 8 hours last night, at 2500 MHz, with 1.6V, which was Prime95 stable for about 5 seconds before the procedure. After 8 hours of this, the overclock was still stable for less than ten seconds with a 1.6V Vcore voltage - so no MHz:Vcore gains for me just yet.

I'm currently repeating the process at 2700 MHz, with 1.8V of Vcore, which is stable for less than a second in Prime95.

Do you think that this might work for a GPU as well, applied slighty differently?

Does anyone have any thoughts on this?

Sentential · Nov 7, 2004

felinusz said:
I did this to my Mobile Barton for 8 hours last night, at 2500 MHz, with 1.6V, which was Prime95 stable for about 5 seconds before the procedure. After 8 hours of this, the overclock was still stable for less than ten seconds with a 1.6V Vcore voltage - so no MHz:Vcore gains for me just yet.

I'm currently repeating the process at 2700 MHz, with 1.8V of Vcore, which is stable for less than a second in Prime95.

Does anyone have any thoughts on this?

A few seconds in Prime95 stable, isnt stable enough. It must be stable enough to run for a few minutes.. no more.

The purpose of this is to take the voltage to its critical point, just before the CPU begins to fail.

You dont want to choke the CPU off, because that will do nothing. If you give it the absolutle minimal volts, riiiight at the point of failure. If you burn it in there you will see significant jumps as I just have.

I burned in 2570 last night at the critical vcore of 1.56v. With 1.56v it would prime for 2-10 mins, but no more. I burned it in for a good 12 hours last night. Now 2570 has become perfectally stable.

I will try again later tonight in my final push for 2.6ghz

Sentential · Nov 7, 2004

Finally making solid progress on 2.6ghz. Instead of failing instantally, it now goes for 4 mins. :clap:

Now burning in 2.6

felinusz · Nov 7, 2004

Thanks Sent, I'll try it out again, with some more voltage (enough for ~5 minutes of Prime95) later on tonight, and see what happens.

For the record, this "CPU Burn-in" program doesn't display anything when it's running properly, with error checking off, right?

Npetune3000 · Nov 7, 2004

It looks like whatever gains that i had made have gone away. Dammit :/ ....I dont think i will be purchasing that XP-90 afterall.

SilverJag · Nov 7, 2004

Npetune3000 said:
It looks like whatever gains that i had made have gone away. Dammit :/ ....I dont think i will be purchasing that XP-90 afterall.

What do you mean? Can you please explain more?

Npetune3000 · Nov 7, 2004

AFter the burn in I was stable w/ 2.5ghz @1.55volts down from 1.6volts but that is no longer the the case. In order to reach 2.5ghz now i have to use 1.6volts.

If I was alble to reach 2.6ghz under 1.65 volts I was planning to purchase a XP-90 to keep it cool. However, since the best I can get is 2.5ghz, it would be a waste to purchase the XP-90 when i can get 2.4ghz w/ deafult Vcore. Not to mention, I can run my chip @ stock speeds with 1.075 Vcore stable so buying the XP-90 would be a waste of funds i do not have.,

ïß®äHïM · Nov 8, 2004

Sorry for bumpin a relatively old thread, but do u guys run the CPU Burn-in program with enabled or disabled error-checking.

Sentential03's burn in's really do work for me.

Member

Disabled

Senior Overclocking Magus

Member

Senior Member

Member

Mushkin Tech Rep

Member

Member

Member

Member

Member

Senior Overclocking Magus

Contributing Member

Contributing Member

Senior Overclocking Magus

Member

Member

Member

Registered

Similar threads