[geeks] New Itanium machines from SGI

[Chris Byrne]

No under current process and materials technology we have reached a
limit. We can get any faster than we can disipate the heat from. The
bigger the wires the more heat there is. The smaller the wire the less
heat. One we hit the smallest wire that silicon can support and the
higest heat level it will tolerate (actually at that point we are also
worrying about electromigration and random interference as well. Never
mind the yeilds issue) that's it.

Its been a well known topic in the semiconductor industry for a few
years now, We have some ideas as to what we can do but none of them are
within 5 years of production or probably even ten years unless someone
really kicks the research into overdrive. They havent done that because
its so expensive.

Chris Byrne

> -----Original Message-----
> From: geeks-bounces at sunhelp.org 
> [mailto:geeks-bounces at sunhelp.org] On Behalf Of Gavin Hubbard
> Sent: Wednesday, January 08, 2003 22:47
> To: The Geeks List
> Subject: RE: [geeks] New Itanium machines from SGI
> 
> 
> >So what exactly do we do. We've gone as far as we can go 
> vertically, now
> >its time to go horizontally. 
> >
> >Chris Byrne
> 
> I'm not sure I'd agree that we've pushed silicon to the limit (I can
> remember when HDDs were about to reach 'the limit' before GMR 
> heads were
> developed). Haven't solid-state research labs turned out working logic
> gates at 100s of GHz?
> 
> I know the simple gates are a long_long_long way from a modern
> microprocessor, but Intel sure spends a lot of money on R&D.
> 
> Regards,
> 
> Gavin
> _______________________________________________
> GEEKS:  http://www.sunhelp.org/mailman/listinfo/geeks