Despite the advent of lower-power designs from Intel and AMD recently, today’s processors are still power-hungry beasties. One company says it has a solution to the problem, which may lead to the kind of clock speed increases we used to see back in the heyday of Intel’s ‘it’s all about CPU speed’ fixation.
Early stages of development eh? I’ve got my doubts based on some other theoretical-on-paper products that didn’t turn out so well in reality including but not limited to netburst, glaze 3d, rambus, and the K6III.
Ok, maybe the Glaze 3D doesn’t count, that’s vaporware.
The idea of square CPU cycles is an innovative approach though.
But Rambus is succesfull recently … by sueing DDR RAM manufacturers over IP infringement.
Multigig says the pulses remain synchronized…
Multigig’s solution is still in the early phases of testing.
This idea is such that it needs a pun to summarize it. You have doubts. I have doubts too.
..but remember that 75% of 50 Watt is still 13x as much as ARM chips, or a number of MIPS and PPC chips consume
But still: getting more juice from x86 using less juice can be a good thing. For one: saves on the CO2.
Yes it’s more then ARM, but is that really Apples to Apples performance wise? Especially when you’re talking about the more complex instructions x86 can do compared to some ARM chips? Anyone have any benchmarks comparing performance/power ratio of X86 and ARM?
If you’re looking for a low-power system today (not some years in the future), check out http://www.mini-itx.com (and no, I don’t work for them or own VIA stock).
I’ve noticed that DamnSmallLinux sells mini-itx systems on their web site. Plenty of other sources, if you look around.
Edited 2006-05-10 01:35
VIA has onoe of their processors running at like 1.5w
Right. And you conveniently forgot to mention its piss-poor performance…
No worse than any of there other ones, which aren’t THAT bad.
I mean, they don’t compare to Intel or AMD but are great if you need a useable performance w/ modern software and low-power reqs.
I could almost hear the *yoink* as you copy/pasted not only the Ars article title, but also the first paragraph, and didn’t even bother to attribute it to them.
Not this old debate again. They are attributing it to ArsTechnica. This is done by the link, and by the “Linked by Thom Holwerda”.
It clearly states it’s a link and it shows the first part of the article. Nothing wrong in that. You are just not reading.
I wonder why we are told about products to come, instead about products, which are here 🙂
You can go to http://www.epiacenter.com , which seems to be more active than http://www.mini-itx.com and you will find there link to reviews to C7-M based notebooks already ….
cheers,
Petr
Anyone seen the new VIA ULV’s (ultra-low-voltage)? The 1.5GHz ULV consumes 7.5W. That’s great. Go VIA. I just bought a 533MHz fanless VIA C3, which uses about 2.6W, for my server.
http://www.via.com.tw/en/products/processors/eden_ulv/