It seems to me that not everyone agrees about the idea of a "burn in period" when overclocking things, what does everyone here think?
And how long is a burn-in time supposed to be?
It seems to me that not everyone agrees about the idea of a "burn in period" when overclocking things, what does everyone here think?
And how long is a burn-in time supposed to be?
More than a week, it seems.
I am not getting much temperature difference, and its been a week.
After my few months burn in I am stuck 100mhz BELOW what it would reach when I originally got the chip....
NS
lies!
chips will burn in, but its more about voltage than time tbh. samsung ram chips burn in very nicely
That's good to hear you think so, Agent. I hope you're correct because I intend to avoid burning in. Is anyone else in serious agreement? Disagreement? The last of my stuff has been shipped and it should be assembled soon.
...Nightshade737 that's really a drag about your memory. Thanks for sharing the results.
"Electromigration" :S this guy knows his stuff ;-)
3D Mark 2k1 - 20661
If you get a customer, or an employee, who thinks he's Charles Bronson, take the butt of your gun and smash their nose in.
I had a really excellent paper on electromigration and it talked about "burning in". The amount of facts backing it up was amazing. I'm just so pee'd because i cant find it
Remember all the 1.6 P4's about a year and a half back, clocking to 2.4+ gig ? - within a month the eletromigration started to kick in, chips became unstable due to the extra voltage, and the tracks degraded much quicker. Some of the people that had been running at extreme voltages couldn't get their chip to run at default any more.
Granted, AMD chips seem to be less susceptible to this fate at the moment - although when we see the AMD 64's become more mainstream, and people start putting 1.85+ volts, I'm sure we'll see a similar effect somewhere along the line.
Thermal paste however, does gain a small advantage due to process of changing its form under heat.
Er... I was talking about my Processor, not my memory (used to do 2060, now only does roughly 1960mhz).
NS
i got a few sticks of pc2700, each one wouldnt boot at first under 2.7v, after a whlie clocked at 200fsb at 2.6 they managed 2.7, is this not burning in?
/me goes andreads about electomigration
http://www.csl.mete.metu.edu.tr/Elec...ation/emig.htm
what would electomigration have to do with burning in? i woulda thought it would be the silicon that the burning in happens, not the conductors to it?
I have never found a difference in OCing a chip from the day i buy till i sell it , only my findings of course , so i agree with agent Mostly BS
Agents quite right about the Dubbed Northwood Sudden Death Syndrome. No one ever quite managed to explain that one.
Im still a sceptic to it all, i understand the premise of electromigration, its much like shoes. You try them they fit but sometimes there just not right. Wear then for a little while and it feels like the things where made for your feet. Keep wearing them and they simple age and fall apart.
Pherhaps its one of the relics in the electronic world, since designs have shrunk vastly over the years, parts have become more sensitive and susceptible to "wear" on the conductive paths leading to leakage and break down of these path ways.
Personally, I've found that after having my set-up for around 3 months, I don't need as much volts (for my CPU) as I did at the start.
ie. At the start, 2.2Ghz, 1.7v. Now, 1.65v. Start, 2.4Ghz, 1.85v. Now, 1.825v. Take note though, I have not been able to increase my max oc speed. It is still at 2.45Ghz as it was from the beginning.
Previous System
||3DMark01 - 18,533|3DMark03 - 6,501|Aqua Mark 3 - 43,912|[AthonXP 2500+@ 2.4Ghz - 12*200, Radeon 9800SE (unlocked) 440/369]||
Current System
||3DMark01 - 30,949|3DMark03 - 18,122|3DMark05 - 8,483|3DMark06 - 4,222|||
There are currently 1 users browsing this thread. (0 members and 1 guests)