A Well-Chilled 750GHz Feasible Within 5 Years

drkhong writes: "...at least if you've got a good cooling system. IEEE Spectrum has an article about the next generation ICs. Using superconducting materials cooled down to 5K a peak of 750GHz has already been reached. Just think about how far light goes within one clock cycle, and then tell me you aren't impressed." These low-temperature devices are made of niobium (a superconducting metal), and use something called Josephson junction devices, resulting in chips for which the article states "there are no known physical barriers to decreasing size by a factor of 10 and thus increasing speed by a factor of 10, using lithography to move from today's 3-m linewidth to 0.3 m."

Comments

750 Ghz

[yamla]

A peak of 750 Ghz has already been reached? That is not at all what the article says. It notes that data rates of 750 Gb/sec have already been reached, an impressive but totally different thing.

This is, of course, very impressive but let us not forget that this requires cooling down to five degrees Kelvin. We are well past heatsinks and fans at this point. Unless the prices come down, it will cost around twenty THOUSAND dollars to cool the chip down this much.

It will be a long time before you see a system like this on your desktop. Unless we develop room-temperature superconductors, of course. But that would change everything...

Re:750 Ghz

[krlynch]

I wonder if there are enough particles in the universe to run a finite elements simulation for more than 4 hours in a 750 GHz CPU.

There are more than enough to keep such a CPU busy for nearly all eternity....A 750GHz CPU (even assuming 1 flop/cycle average throughput) would still have a hugely difficult time just doing QCD calculations of the interactions inside a SINGLE proton in anything approaching days! (I have a colleague doing lattice QCD who was just telling me about their new algorithms for hacking time off of certain types of lattice simulations, and they are talking about running for 16 CPUyears on a brand new 90Gflop machine! At 750 Gflops, you're still talking 2 CPU years! Don't ask me for details, though, as I don't know any....not my field).

For further consideration, there are about 10^80 particles in the universe (give or take a few orders of magnitude.....). Let's assume it only takes 10flops to update a single particle for one timestep (not even close, but let's run with it shall we?) That means we update 75 x 10^9 particles every second...let's round up and call it 10^11. That means it would take about 10^69 seconds to update one time step. Or 10^61 years. Which is roughly 10^45 times the age of the universe. Not to mention the amount of RAM you'd need to run this simulation on (which would take more particles to build than there are in the universe itself, but I digress.....)

Really monstrously fantastically mind-bogglingly large numbers are really really fun :-)

Re:open this box

[carleton]

Nope. That's the whole point of 128 bits. Assuming you're doing brute force, and have 1000 of these computers overclocked to run at 1000 GHz (to make the math easier), and of course assuming that they can do one trial per cycle,

2 ^ 128 trials * 1 cycle / trial * 1 second / (10^12 aggregate computer cycles) * 1 year / (3600 * 24 * 365) = 10790283070806014 years = 10 quadrillion years.

so what?

[crgrace]

Superconducting logic has been out for a VERY VERY long time. In fact, IBM burned tens of millions of dollars on the subject in the 1970s. The problems with superconductors are even WORSE than the problems with superfast III-V logic. UCSB has 70 GHz flip-flops made out of transferred-substrate heterojunction (III-V, Indium Phosphide and something else) transistors, but nobody thinks they will revolutionize computing, because they won't. So it is with superconducting logic.

There are two huge problems with superconducting logic that don't seem solvable in the near future. They are:

1. Cost : These things are enormously expensive to manufacture and operate, and it is the economy of scale of CMOS techology which has enabled, more than anything, the current computing revolution. Do you have any idea how expensive coolant and the dewar to use it in are to get something to 5K? Even the so-called "high-temperature" superconductors have to be pretty damn cold to function; they just don't need to go so close to absolute-zero.

2. Integration This is probably the killer. It will be extremely difficult to integrate many devices together. Even if myriad technical difficulties are overcome, the solution is not likely to be inexpensive, as CMOS technology is. For III-V semiconductors (which use much less exotic materials than superconductors), high defect rate, problems with lattic matching of the materials, and the lack of a high quality native oxide (like SiO2 in silicon) have made it impossible to achieve integration levels anywhere close to that achieved in silicon. Even GaAs, the most well-understood III-V semiconductor, can't be integrated to more than a few thousand devices. That's why we don't have 20 GHz GaAs microprosessors. And superconductors are even HARDER to deal with.

In summary, even if researchers are able to overcome almost insurrmountable odds to find away to reliably integrate meaningful numbers of these devices on a single die, I think it is very unlikely they will be able to do it cheaply, which is just as important as being able to do it at all. Otherwise, this technology will be of interest only to the military.

By the way, I know III-V semiconductors have a lot of very important uses, especially in optics and RF. It is a fact, however, that III-V logic is mainly of interest to the military and the space industry.

Re:so what?

[crgrace]

Disclaimer: I design analog-to-digital converters for a living.

The market the article talks about is the analog to digital converter market, not the desktop market.

True enough. I stand by my statements, however. For one thing, the article discusses an A/D fabricated in the technology. The die size was 1 cm^2. That is truly enormous and very would be extremely expensive as a product. Even if they can bring down the lithography, it is still very expensive. Second, while they say it runs at 12.8 GHz, because of the decimation filter it is obviously an oversampled converter but they don't give the oversampling ratio, so we have no idea of the actual conversion rate. I drew parallels between III-V materials (which have been around since the 1960s and must have achieved some kind of maturity) with superconducting electronics (which have been around since the late 1970s) and I think they still stand.

My belief is that this is a laboratory curiosity with little commerical potential. I'm sure the military is very interested in using it with radar, however.

By the way, the IEEE is well known for pumping up "cutting edge" technologies that never reach their potential. Remember "fuzzy logic"?

Whatever happened to GaAs CPUs

[pjrc]

In the early 90s, it was widely believed that GaAs transistors would replace good 'ole Si for microprocessors. They're a lot faster, after all.

Well, several things happened:

• Nobody figured out how to make reseasonalbe P-channel devices.
• Small geometries were much harder, because III/V and II/VI type (more than one element) semiconductors suffer from a whole bunch of problems where, which in my limited understanding (I'm a circuits guy) are due to the wrong atom at a place in the crystal lattice, such as a Ga where and As should have been.
• It's easy to take Silicon Dioxide (glass) for granted, until you try to figure out a good way to make insulators on other materials.
• All the while, good ole Si kept getting better and better... not only faster, but higher densities. Today's CPU speed is as much a function of using lots of transistors as it is their speed. As more transistors were available, everyone invested a lot of research and thought into ways to use them to run code faster (superscaler architecture, branch prediction, out-of-order and speculative execution, etc)

Now I've been watching the J-junction for several years now, though I know much less about how it really works that I ought to. I do know there's a big difference between a test device and processes that produce only thousands of them to being viable for a modern microprocessor. GaAs transistors are hugely popular for RF applications, where you only need a small number of them. Today nobody believes the world will eventually be overtaken by GaAs based microprocessors.

It seems unlike the world will really be overtaken by J-Junction microprocessors, at least in our lifetimes. Maybe that's just wishful thinking, since I've got a lot of energy invested in transistors, and with a bit of luck that'll remain valuable for another 25 years... but then again, look what happened to all those guys how only knew about tubes!

Anyways, the point is that there's a big difference between a small number of insanely fast test devices to a high density processor with all the other requisites to make a reasonable microprocessor.

Next from AMD....

[supabeast!]

The Freon!

Cost is Going to Preclude It from Being Bought

[nachoworld]

This is absurd. The cost of keeping such a superconductor at 5 K is going to keep the general public, and even most corporations, from buying this technology. It's expensive to keep a box at that temperature in the lab (I should know, I'm a chemist). Only the US government would be willing to shell out the money for these low-maintenence devices (maybe). Corporations would rather just use the money to buy the computing speed in multiple CPUs rather than as one - it'd be a hellofa lot cheaper.

Perhaps the people working on the project will eventually be able to use a superconducting material that works at liquid nitrogen temp instead of niobium (perhaps a yttrium complex like we use now? - I don't know the specifics of this 700GHz IC or whether it would be able to use Yttrium complexes). In that case, the cost will go down and perhaps we'll see more corporations buying this tech. In order for personal consumers to buy a 700GHz computer, we'd have to have room-temp or near-room-temp superconductors.

But then we run into one of the hugest physics problems of the late twentieth century. The scientific community no longer has the enthusiasm it once had for searching for that "perfect" superconductor.

---

Cheaper Hardware?

[dmatos]

With the advent of the 300MHz processor, the 233 I purchased became dirt cheap. Now that there are 1.5 GHz chips out, you can get an 800 MHz chip dirt cheap. When the 750GHz chips are produced, I will be lined up to buy an obsolete 500GHz chip that will be fast enough to start windows from boot in less than three minutes! Yay bleeding edge subsidizing second-stringers!

Only 10 Quadrillion Years?

[mr_gerbik]

Thats precisely why Strom Thurman should use 256bit encryption!

2 ^ 256 trials * 1 cycle / trial * 1 second / (10^12 aggregate computer cycles) * 1 year / (3600 * 24 * 365) = 31600000000000000000000000000000000000000000000000 00000 years... he should be retired by then at least.

Yeah, and a 2 million stage pipeline

[Mr Z]

One problem with these high clock rates is that you end up having to pipeline things rather excessively all over the place. I'd imagine at 750GHz that even a single 64-bit ADD would be pipelined over multiple cycles, due to transport delay!

Think about it: Light travels about 1 foot per nanosecond (30cm). At 1GHz speeds, a signal could travel well across a die if it were unimpeded (eg. could travel at the speed of light). In fact, it could theoretically travel most of the way across the motherboard in one clock period. At 750GHz, light travels 0.4mm per clock tick -- about 1/20th the way across a typical CPU die (assuming a die in the range 8mm x 8mm to 10mm x 10mm die -- not too far off what we build today). We're talking 20 pipeline stages just to get from one edge of the die to the other, if we can travel at the full speed of light in a vacuum. And the bad news is that we probably can't -- just look at todays CPUs!

What'll happen is that highly parallelizable problems will speed up, and inherently serial problems will end up staying the same. All of your number crunching for playing video games will rocket along since the calculations can be pipelined and parallelized, but the twisty, turny, five-instructions-and-a-branch control code won't speed up much.

--Joe
--
Program Intellivision!

WOW!

[clinko]

Joe Consumer -

"750GHZ! WOW! NOW I CAN RUN AOL EVEN FASTER!
AND WITH 56k AOL IS FASTER THAN EVER!"

Finally!

[Reality Master 101]

I can get 100,000 frames / second on Q3. Dammit, I can see the difference!!

--