View Full Version : Best burn in procedure ????
Hi guys and gals. I'm new here. I have a question for you all. What is the best burn in procedure for O'C your CPU. I have an ABIT VP6 with dual PIII 1ghz which I have at 1.13 (FSB 152) stable. They are 1 week old and have been at that speed since I got them. Now should I up the voltage so they get nice and hot and then leave them be for awhile???
Cpu 1 and 2 at idle are 23 - 24 c
Cpu 1 at full load is 33 - 34
Cpu 2 at full load is 30 - 32
MB is at 25 - 26 all the time.
I have run this at high speeds as high as 1.2 but anything after 1.13 is not stable in warm weather. I have 6 - 80mm case fans and thermoengine hsf with delta fans on my cpu's. They are cDO stepping and a matched set as well as my ram which are Cas2 (256mb * 4). Now if this is as high as I can realisticly go then I am more then happy with it since it is Quuuuuiiiiccckkkk !!!!! But it would be nice to know a good way to burn these chips in and know if I can go higher. Thanks all..
Oh and by the way VP6 ROCKS!!!!!!!!!!!!!
:)
outhouse
09-01-01, 09:13 PM
Duel p3 1g's dont overclock that well as the MB limits the amount that it can be done, you are at about where most duel set ups get. If you want to try and burn in the 1g do not get it hot these CPU's cannot take the heat other CPU's can, when they say burn in they mean running them hard not hot, to burn in raise your CV to 1.9 as long as your temps are under 45C full load try and leave your FSB at default speeds and then run Seti or another burn in program for 2 days to 2 weeks, you can still use your PC fot whatever you like but when your not put the burn in program back on. It may cost you some bandwidth but you can turn your memory down some it may help try setting it to minus clock speed and see if that will help.
Yeah I thought as much, since when I take one CPU out the sky is damn near the limit :) Much more stable at higher FSB but it is still interesting to know how high I can make this puppy go. Sure is quick though. Oh yeah I run Seti 24/7 on here as we speak.
Thanks for you advice
outhouse
09-01-01, 10:10 PM
since you have duels i'm sure you could run two work units at the same time, :)
How do you run 2 at a time?
cyberey66
09-01-01, 11:04 PM
I'm still confused about burning-in. Does anyone know any links that explain what goes on.
Well I'm off to searching about burning-in.:cool:
outhouse
09-01-01, 11:29 PM
Originally posted by Lainar
How do you run 2 at a time?
Check Octoman's post rite above yours it has the exact dirrections to your question.
outhouse
09-01-01, 11:39 PM
Originally posted by satan
I'm still confused about burning-in. Does anyone know any links that explain what goes on.
Well I'm off to searching about burning-in.:cool:
There should be something in the tips section but i'll try to help some.
When you burn in your just working the CPU hard with a higher CV which gets the CPU used to high CV, There are gates that open and close at a rate you cannot see info passes through these gates when you burn in your just getting these gates used to opening and closing at a rate they where never intended, I think and i could be wrong that the CV gives them extra power which makes them move a little more efficiently even though the MHZ output is the same then when you raise the FSB they move more quickly thus raising the MHZ.
To be honest I wouldn't waste time doing a burn in procedure, I've never seen any convincing science (some interesting theories but nothing concrete) and you risk wasting a lot of time for possibally no benefits. Have you tryed actually increasing the voltage to overclock? try both the vcore and vio. Also some good cooling for the northbridge as the standar vp6 heatsink may not be enough when running to fast cpus at out of spec bus speeds.
Burn-in in the electronics industry refers to the process of operating an electronic device at full load and temperature for an extended period of time to see if there are any defects in it that will lead to premature failure. If there are any defects the odds are it will fail during the burn-in period. If not it will likely last quite a while.
So, Phil is right. This notion that "burning in" a cpu will somehow make it clock higher is really unfounded and is mostly just a myth. Any benefits you do see in overclockability resulting from "burning in" a cpu are probably attributed to things like the thermal paste heating up and conforming better to the CPU and heatsink or perhaps the chip taking a better seat in the socket.
outhouse
09-02-01, 11:36 AM
Well i have to disagree with both of you, as it stands the mojority of people here do burn in and most have had succes and i know for a fact its not the thermal paste conforming because 1 temps have not changed for better or worse and 2 I have had my current CPU burn in twice once at 1.9 and then again at 2.1. Your rite there is no evidence and no one knows exactly why but the fact remains true it does work and has helped many to gain stability where they was none before. To this day i have never heard anyone say they had bad results from burning in only good ones. As far as the industry standard for electronic components that may be true but as far as overclocking goes you dont want to get your CPU close to its limits with heat as it can cause damage, once I tried running my CPU at 10C below its rated temp and actually lost ground luckily I only lost about 15 or 20MHZ and did not fry my chip. As far as waisting time if you use Seti and its running in the background your not waisting anytime in fact your helping this website if you have joined the team.
I didn't say I don't beleive in burning in, more that I haven't seen any real evidence of it, I would not waste my time now and possibally reduce the life of my chip on something that may only be superstition. Any way when you hear all these people saying that burning in worked for them they only achieved dropping the voltage back by like 0.1v or a few extra mhz, never massive gains.
cyberey66
09-02-01, 05:44 PM
Well I don't care if it works or not, I'm really looking for how it works, if it does. I printed a 32 page read on CPUs. Maybe it'll have some stuff that'll connect with what people say. To me it just doesn't make sence. Well my CPU is always at full load now because of F@H, so I'll see what happens.
theres all kinds of theories but like I said no proof, some people think its a form of electromigration, others say it's like running a new car in (stop laughing at the back)
but the complete lack of scientific evidence makes me sceptical.
vBulletin® v3.8.7, Copyright ©2000-2012, vBulletin Solutions, Inc.