<A HREF="http://biz.yahoo.com/bw/011203/22033_1.html" target="_new"> Here </A> is AMD's answer to Intel, apparently. Granted, by their own admssion, the technology won't be in use until <i>2009</i>, it's nice to see that they are quite willing to go their own way.
It seems to me that AMD always underestimates their abilities, while Intel always overestimates theirs. I’ll bet we see these nano-gate processors from both companies around 2007 (somewhere between Intel 2005 and AMD 2009).
Just in case you missed my message under raystonns post titled:"Intel's Breakthrough in Chip Transistor Design"
Here it is again...
Looks like Intel are not the only ones spending money on research and development.
<A HREF="http://www.theinquirer.net/03120109.htm" target="_new">http://www.theinquirer.net/03120109.htm</A>
As soon as Intel publicly announce 'a new breakthrough' every body jumps on the bandwagon. Its all part of the rat race I guess, don't know if all these 'breakthroughs' will be practically useful or is just hype to keep investors happy.
Personally I think IBM are the pioneers in new technology (science and computing), although I don't like the company itself. I'm sure there are many other companies we haven't even heard of that have made bigger breakthroughs but don't bother annoucing them. My point being the people who make the 'real' breakthroughs often don't get mentioned, the large corporates who sponser the smaller companies usually receive all the attention. This is called 'outsourcing' which is what usually happens in large corporations. Don't get me wrong I'm not implying Intel, IBM or AMD are doing this or saying large multinationals are bad (in fact I think they are very important), I'm just telling you the way it is.
<font color=purple>~* K6-2 @ 333MHz *~
I don't need a 'Gigahertz' chip to surf the web just yet ;-)</font color=purple>
anyway here's a link to AMD's new "3.3 THz" transistor.
that about 3 times faster than the Intel trannie that Raystonn announced recently.
and before anyone says anything.. Yes I did notice, and thought it was weird, that they said this..
"The 15-nm transistor, devised in AMD's Submicron Development Center, is a CMOS-based, 0.8-Volt device, designed to handle switching speeds of 0.3-ps, or 3.33 trillion switches per second. "
to me, that sounds like they haven't actually stretched the transistor yet. they're just estimating.. or worse hoping... but the fact is the design in small enough that it's feasible to see it going faster than all other current transistors.
Cool
to bad about the ol' waiting til 2009 thing..
"I spilled coffee all over my wife's nighty... ...serves me right for wearing it?!?"
AMD's work was presented at 2001 IEDM as a lat-news paper, which means it missed the end-of-June paper submission deadline but was accepted as "breaking news" by the conference. However, it does contain same kind of electrical data as the Intel's paper. What these two papers have are single transistor characteristics and results of ring oscillators (RO). A ring oscillator (RO) consists of typically hundreds of transistors that in a configuration one can measure the average switching time of each transistor. This is far cry from the operating clock speed of a real microprocessor. As a frame of reference, nowadays transistors in mass production may have an RO switching time of a few picoseconds (ps) to tens of ps. Therefore, although the 0.3 ps of AMD (or sub-1 ps of Intel) sounds impressive, it will not produce "THz" microprocessors. In fact, there are still MILES to go before any of these technologies comes close to production.
The only thing about all those brouhahas is that Intel marketing is very aggressive, making a big deal out of this "terahertz," while AMD is actually kind of layback here.
**Spin all you want, but we the paying consumers will have the final word**
Thank you for telling it better than i could
I am fully aware that THz transistors is not even close to THz processors.
But in the same vein, if AMD can produce faster transistors in the future much like Intel will have faster switching transistors.. and that IS the fundemental number when calculating a CPU speed... then it should silence all these people that say Intel/Northwood and so on, is gonna cane AMD/Hammer and so on...
no, change that... It won't silence them, but maybe they'll have sensible arguments about which processor can best benefit from this stuff.
Fact is, both AMD and Intel have tech. roadmaps that are advancing..
we don't know who will win anymore than we did 2 years ago when P4 speculations may have started.
and when somebody trolls and says yada blah blah is better, blah blah..
REMEMBER THIS -> Knowing wh's processor will be better and *thinking* you know who's processor will be better, Are 2 very very different things.
"I spilled coffee all over my wife's nighty... ...serves me right for wearing it?!?"
