Terahertz processors? How is this possible?

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

f4phantom2500

Platinum Member
Dec 3, 2006
2,284
1
0
Unless some major scientific breakthrough is made, I don't see how a THz CPU is possible. Right now, we have enough trouble trying to cool our CPU's @ 4 GHz using water cooling. I doubt we'll be seeing such computer technology until 2050-2060, if at all. We'll probably have CPU's with Hundreds or Thousands of cores running at a few GHz each.

I think it's far more likely that we'll be using quantum computers by that timeframe.

http://en.wikipedia.org/wiki/Quantum_computing

I don't think it's too far off, some dudes at Yale built a quantum processor last year

http://www.sciencedaily.com/releases/2009/06/090628171949.htm

Very exciting stuff
 

evident

Lifer
Apr 5, 2005
12,116
733
126

f4phantom2500

Platinum Member
Dec 3, 2006
2,284
1
0
haven't they been trying to produce optical computing as well since the 80s? as in a laser would replace a transistor in representing a 1 / 0?

quantum computing is a fundamentally different approach; it operates on the principles of quantum mechanics. things like superposition and entanglement come into play. its many orders of magnitude faster than classical computing.

http://www.sciencedaily.com/releases/2001/09/010913074828.htm

that article's old but you get the idea. the wiki in my previous post goes into good detail.
 

jvroig

Platinum Member
Nov 4, 2009
2,394
1
81
Would such a processor emit any radiation?
I assume you mean hazardous / life-threatening radiation such as those by radioactive isotopes found in nuclear power plants or research labs (and not more generic radiation such as electromagnetic radation from your cellphone). Not at all. Don't be afraid of the "quantum" in the name, quantum mechanics is present in sub atomic levels of everything, not just in "nuclear"-like things.
 

f4phantom2500

Platinum Member
Dec 3, 2006
2,284
1
0
I assume you mean hazardous / life-threatening radiation such as those by radioactive isotopes found in nuclear power plants or research labs (and not more generic radiation such as electromagnetic radation from your cellphone). Not at all. Don't be afraid of the "quantum" in the name, quantum mechanics is present in sub atomic levels of everything, not just in "nuclear"-like things.

i think he may be referring to the optical processors...either way there would be no hazardous radiation.
 

Swivelguy2

Member
Sep 9, 2009
116
0
0
Okay, here's the straight dope about photonic computers:

What the University of Utah article from the OP seems to be talking about is yet another optical waveguide technology. There are lots of things like this already in the academic literature, and at shorter wavelength too.

But waveguides are the photonic equivalent of copper wire. I don't see how they can claim anything's coming in 10 years when we're still working on getting copper wire down. Did the first computer come just 10 years after copper wire was reliably manufactured? Hell no.

Once we get copper wire down, we're going to have to develop the photonic equivalent of circuit elements, primarily transistors. That'll take a while.
 

Matthiasa

Diamond Member
May 4, 2009
5,755
23
81
I assume you mean hazardous / life-threatening radiation such as those by radioactive isotopes found in nuclear power plants or research labs (and not more generic radiation such as electromagnetic radation from your cellphone). Not at all. Don't be afraid of the "quantum" in the name, quantum mechanics is present in sub atomic levels of everything, not just in "nuclear"-like things.

Quantum mechanics is also present at the macroscopic level as well. :p

But yeah, it would be nice to have optical and quantum computes someday.
 

Candymancan21

Senior member
Jun 8, 2009
278
3
81
http://news.softpedia.com/news/Terahertz-Computers-Coming-Up-Soon-83487.shtml

Right now, we have enough trouble trying to cool our CPU's @ 4 GHz using water cooling.



Speak for yourself... Im running my E8400 CO yea CO @ 4.2ghz currently 1.296v and its watercooled using a Maze 3 block... Yea Maze 3 and a old black ice extreme radiatior with a 1048 pump.. All the original parts i had back in the Athlon xp 2400+ days... Im not even use 2 fans on my radiator anymore i just 1 90mm fan..

It stay idle around 37-39c and load doesnt reach anyhigher then around 60c. I can run this cpu up to 4.4ghz if i wanted. So i dunno i have no issues watercooling this thing. Maybe the newer watercooling blocks just arent as good as the old one im using. I had to modifiy the block and make brackets to work on 775 but hey it works right ?


Im suprised you never heared of optical computer's.. Its actually called Quantom Computing.
 
Last edited:

taltamir

Lifer
Mar 21, 2004
13,576
6
76
read the actual article people...
the THz is the frequency of the LIGHT they used not the frequency of the CPU. (HZ = Cycles per Second).
They are trying to build 30Ghz CPUs using THz light interconnect... (hopefully they mean 30ghz retail CPUs... because netburst is theoretically 40ghz)

Electricity actually propagates at the speed of light. the advantages of optical INTERCONNECTS (that means the fsb, not the actual CPU itself) is that it produces much less heat, the signal doesn't get weaker the longer you travel, and you don't have to worry about interference and crosstalk... basically, it takes a lot less power (due to multiple reasons) and produces less heat. Cooler + Lower power requirement = can push clockspeed higher