Overclocking: Tom's Sets World Record

Page 3 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Status
Not open for further replies.
5.112 Ghz....wow. Well...I got 3.7ghz on my e6750...sounds pretty slow now. I wonder if It was stable enough to run a 1M Super PI. I got 14.1, I wonder what this oc'd machine could do.
 
[citation][nom]physx7[/nom]Read my above post...[/citation]
Yes I read that part, but that doesn't necessarily mean the record is highest clock speed although that is implied. I'm going to have to assume that's what they mean regardless.
 


I wonder what the record overclock for a Q6600 really is....
 
this goes to show that the limt to how high the frequency can go is heat, not the properties of the silicon. back in the 80s when CMOS chips were first being introduced, they used a 5V standard, now it is around 1.2V, a decrease of around a factor of 5. the size of the transistor, however, has decreased by a factor of about 100, meaning the electric field across each transistor that is conducting is 20 times greater, letting the heat monster out.
 
Exactly Kevinmcg.. only 2 cores i,m sure Quads have 4 cores!!..

5 months late .. not true results .. and they didnt break the record ?.. well they may have 5 months ago .. lol ..

Toms why sooo late in publishing the story .. maybe you thought you would be busted?? even if it was only 2 cores .. i,m sure CPUZ shows 4 for a Quad?

LIES all LIES!!..
 
hmm next time cool the GPU and the RAM and try and play crysis or something next time. but well done guys, excellent acheivement. how long till you can have my set up running at those speeds?
 
[citation][nom]fudgeboy[/nom]hmm next time cool the GPU and the RAM and try and play crysis or something next time. but well done guys, excellent acheivement. how long till you can have my set up running at those speeds?[/citation]
At those speeds they were probably on the edge of stability, I doubt they'd be playing crysis any time soon.
 
A fun experiment, but to what practical purpose? The system isn't stable, and quite frankly, you're duplicating similar experiments that have already been done. Perhaps next time you may wish to consider the highest overall stable system speed by your choice benchmark, then relate this to computations per time increment per watt power usage. Just a thought.
 
I did a somewhat similar experiment with my own CPU. I was mostly interested in how cool i could get the cpu. I was running a AMD 939 3700 with the stock fan. I was able to get it down below 0 Celsius. At which point the mobo could no longer give me an accurate reading, because it did the opposite and gave me a reading of 999 Celsius. And since it was my main system figured i should stop the experiement so that i had a system at the end of the day.

So i understand way they did not use the mobo for the temperature readings
 
Wait, a "record" oc on a Q6600 with two cores disabled--that's not truly a record OC on a Q6600 since it's not a fully-functional Q6600--besides, other dual-core 65nm processors have been taken above 5GHz--since this was essentially just an E6600, it's not impressive. I wanna see benchmarks--superpi or something...
 
Why is Liquid Nitrogen used to cool the CPU?
liquid helium is like -450F or something like that
they use helium for superconductors, because it has a much lower temp than nitrogen
 
And how do you propose they pour it in without it boiling off too fast? It boils at 4.2 Kelvin, I doubt your average air conditioner will be able to cool the ambient temperature enough to make it last longer.
 
[citation][nom]Luscious[/nom]All have already been beaten. This link was from April:http://www.xtremesystems.org/forum [...] p?t=184718Kiddies, get back in the sand pit with your buckets and spades![/citation]

To be fair, this was done in February, so maybe it was a record then.
 
Status
Not open for further replies.