Wirth broke three records on stage and another four before the keynote started with an overclocked quad-core Penryn desktop processor.
Otellini's keynote this morning focuses on bringing extreme to the mainstream and during his talk about extreme overclocking, he invited Charles Wirth, owner of the renowned extreme overclocker site XtremeSystems.org onto the stage.
Wirth arrived on stage and was greeted by Otellini, who asked him what he'd been doing this morning. Wirth explained that he'd been playing with a quad-core Yorkfield processor in an unmodified motherboard using an X38 chipset and some Corsair Dominator DDR3 memory.
He was also running a GeForce 8800 Ultra running at its default clock speeds in the system too.
Before the keynote, he explained that he'd already broken four world records and then broke another three benchmarks live on stage in just over two minutes.
According to a slide displayed during the live demo, he broke the highest quad-core frequency, the highest 3DMark 06 and 3DMark 05 CPU scores, as well as PCMark 2005 too.

On stage, he broke the record for SuperPI 1M, which he achieved in eight seconds, followed by Aquamark 3 and Cinebench 10. He set scores of 273,000 for Aquamark 3 and over 20,000 for Cinebench 10.
We spoke to Charles after the keynote and he said that he's actually hit well over 23,000 in Cinebench 10 in the last few days. He also said that he hasn't hit his chip's maximum core speed yet.
"Like Paul said earlier today with a gate leakage - we can directly see that by pushing voltage into the core. The more voltage we push into this core, the faster we can push this processor. I have not reached the top speed of this processor yet, and I did reach 5.56GHz already on an unmodded board. This board has no voltage modifications of any kind."
In order to break these records, Wirth used a custom-made three stage cascade cooler that cooled the CPU to -160'C.
As Wirth was leaving the stage, Otellini joked "call me when you hit 6GHz." Knowing how hard these guys push processors, I wouldn't be surprised if that happens long before the Penryn family launch on November 12th.
Discuss in the forums
21 Comments
Discuss in the forums ReplyAny more numbers? I want to know what the current 3Dmark06 number is to compare against my paltry rig.
Though he was saying the board was unmodded and i was like cool, so we all get this kind of speed....then he goes to say the chip was cooled to -160C !
maybe not then? haha....
Sam
He should have gone for LN2 instead of the cascade system though. -198c love should do some interesting things. Or better yet LHe, which is EVEN colder. Or just work in the near-absolute-zero of outer space :D
and there is absolutely no questioning Fugger's reputation for those that don't know
"Liquid nitrogen is a compact and readily transported source of nitrogen gas without pressurization."
It's always worth questioning the results. If someone has proven themselves (as I assume he has), then you give them the benefit of the doubt. But you should always be ready to question results, especially when they're as exteme as this. It's not calling into doubt anyone's reputation, it's just a matter of verifying the results.
@admins
By the way, your Register link has an extra "http://" in it, which leads you to this page:
http://www.http.com//forums.bit-tech.net/register.php?do=signup
You might want to fix that.