News HP designs new chip layout

http://www.bit-tech.net/news/2007/01/17/HP_designs_new_chip_layout/

 

Uhm, that will be a big benefit, but that's way too far away to talk about yet.. And will computing not be a whole different ballgame by then??

 

Now if i only owned HP, my bank acount would looks something similar like this:
100000000000000000000000000000000000000

Sam

 

closer to 116500000000 actually

I suppose HP have lots of disused equipment from making their 9000 series server CPUs now that they've gone to Itaniums. I guess they carried on playing with them instead of flogging them off

 

royalties..... crap..... more expensive chips here we come.
i will be updating my computer this year.

 

If they're royalties are as overpriced as their current products we'll see CPUs hit the 250€ mark... for entry level CPUs.

Other than that: Awesome! 1W TDP anyone?

 

Well, the article totally missed the point. If you read the /. article, it says they're talking about FPGAs - not standard ICs. All this new tech achieves is reducing the die size to gate count ratio on programmable logic.

8 times the current power in 2010 follows Moore's law pretty much exactly.

ch424

 

your right about it being for FPGAs (Field programmable gate array) and not standard IC processors, but it would be 8 times the 2010 capacity and not 8 times the 2007 capacity so it would actually be beating moores law (at least for FPGAs).

plus

so that would make them even more compact, basicaly its making FPGAs performance closer to that of standard ICs.

so although its not very significant for us pc folk, it is significant for the industry.

 

ch424,

I see your point about the FPGAs, I'll look into this and possibly amend the article if there's need. But as for the 8 times the power in 2010, I think that's outside of Moore's law?

2007 = 1
2009 = 2
2011 = 4
2013 = 8

Double the power every 2 years. so to get 8x response out of anything in 3 is a bit outside of it. Maybe I'm not understanding something about your math?

 

Sorry, I was wrong about Moore's Law, I don't know what I was thinking.

The Ars article and Slashdot comments are interesting though.

ch424