9 replies to this topic

[mfenech]

http://drudgereport.com/flash3.htm

[Eldmannen]

250 times faster on silicon?
I doubt it, silicon is getting near is end...
250 times faster, sure possible with nanotechnology, optical computing, etc but I dont think you can get 250 times faster on silicon...

Also, it is highly unlikely that it is the x86 architecture they're talking about...

[lokoike]

My friend emailed me that exact same article, but that is the only place I have heard of this so far, so I'm inclined to believe it is a fake, or highly overexaggerated. If these guys really knew of a new chip technology that was 250 times faster than current chips, you'd think they would know a little more about it....

[Glenn]

There's quite a few articles but it's not too practical yet:

June 20, 2006
IBM chip breaks speed records
Filed under: Processors
By "freezing" chips to 451 degrees below zero, researchers have produced a semiconductor that operates 250 times faster than chips commonly used today.

According to a report in today's New York Times, researchers at IBM and Georgia Institute of Technology achieved the speed milestone using a cryogenic test station to freeze the chip with liquid helium to temperatures normally found only in outer space.

[zaphirer]

Glenn, on Jun 20 2006, 02:42 PM, said:

By "freezing" chips to 451 degrees below zero, researchers have produced a semiconductor that operates 250 times faster than chips commonly used today.

I guess I would have to splurge on a new watercooling system then. My fans probably couldn't take it

[lokoike]

Thanks for the info, Glenn.

I am a bit confused though. Judging by the info you listed, it doesn't sound like a new chip at all; it is simply a new way of cooling a processor. The article mfenech linked to stated that not only was this a new chip, but it could potentially be on the market in 1-2 years! How on Earth could a company manage to design a cooling system or supercool running chip that operates at near absolute zero in just a couple years?! Something doesn't add up...

I'm wondering if perhaps this new chip (assuming it isn't a hoax) is actually a totally new processor design, as opposed to just some super low-temperature cooling system, which I don't see showing up commercially any time soon!

Quote

The achievement is a major step in the evolution of computer semiconductor technology

Notice that cooling isn't mentioned at all. It makes me wonder...

[Glenn]

The article is ambiguous. It could mean the chip has to operate at -451 deg or it could mean that supercooling it during the production process causes some permanent change in the chip material.

[Eldmannen]

Well the IBM could operate at 350 GHz at normal room temperature.

But this is "germanium-silicon".

Also this is not an CPU. It only has few transistors, it dont have several million transistors and a instruction set such as x86 or something.
They were just testing to see how far they theoretically could go on germanium-silicon. So the chip they have today is not a CPU and it is pretty useless.

[Humpty]

More info as Eldmannen stated.

Quote:

At room temperature, the chip operated at 350 GHz, slightly slower than the record but still over 87 times faster than the fastest commercially available chips. According to IBM, SiGe chips operate faster at very cold temperatures.

Article

[zaphirer]

The question is, when can I get one? Cheap?