Slashdot Mirror
Looking Back at 1984 Report On "Radical Computing"
An anonymous reader writes "The Department of Defense has just released a long restricted report (PDF) by the JASON group entitled Radical Computing. This 1984 study outlines a number of alternate computing methods that could 'result in a radical improvement in computing.' The study attempts to explain the paradox of how the Russian lag in developing VLSI chips curiously did not critically hinder their accomplishments in space missions, ICBMs and chess computation. The authors speculate that the Russians might have achieved breakthroughs in alternative computing methods such as residue arithmetic and symbolic computing. (More cynical types assume the Russians bought or stole US chips from the French or other too-helpful go-betweens.)"
"The paper, published by the Government Attic website, also mentions how, eventually, highly parallel computers could make use of these alternative computational methods. Also discussed are such things as functional programming, interval arithmetic, recursive machines, multiple processor concurrency, fast recurrence evaluation, DDA machines, data-flow, and hyper-column cortex model. Which of these ideas ever came to fruition?"
Umm. Crap.
I've got nothing.
__ Someday, but not this morning, I'll finally learn to use the preview button.
The authors speculate that the Russians might have achieved breakthroughs in alternative computing methods such as residue arithmetic and symbolic computing.
Never propose a simple solution when exotic, impractical sounding one will do instead.
If it's for-profit but free, you're not the customer -- you're the product (e.g., the Slashdot Beta's "audience").
THIS is VLSI: http://www.mycpu.eu/ ...ahem...just not all in one chip.
We spent an awful lot of time and effort in the area of efficient function design as well. The crucial problem was how to derive a precise 'enough' result in a given number of CPU cycles. We did all kinds of functional partial solutions in order to break down complex problems into 'do-able' chunks. The simple fact is that computers aren't that good at Real Analysis, Solid Analytic Geometry and multidimensional trigonometry. You have to crush all that down into composite problems that computers ARE good at.
For further reading, see Byte Magazine from April 1991. Surely all good /. readers have a copy somewhere?
Never email donotemail@WeAreSpammers.com
most of there "insights" were over active imaginations.
Not quite. Sometimes, certainly, they just imagined the threat, but equally often they fell for some simple, yet clever, Soviet spoofs. Much was made in intelligence and in the popular press, for example, of those terrifying parade ground films showing division after division of Soviet infantry marching through Red Square, with air support flying over and armored divisions interspersed. It turned out at least once, however, that the hundreds of bombers flying overhead consisted of just a couple squadrons flying a continuous loop above the parade ground, circling behind the camera to pass by again and again. Very likely the same happened with the armor sometimes.
The Cold War was all about fear, and when analysts fell for something that seems stupid now it's not exclusively that they convinced themselves or became hysterical; the armed forces of both sides did a lot of work to keep up credible appearances of overwhelming force, usually without the actual hardware to back them up.
One of the main serious uses of computing, especially in the cold war, was solving partial differential equations. Whether these be for orbital calculations, stability analysis, EM simulation, etc..., solving partial differential equations is a critical part of any advanced engineering program.
The American approach really started in the 50s with the advent of programmable computers, and is very stereotypical: just find a decent approximation. Modern western engineering is all about using pretty advanced computers to find arbitrary numerical approximations to tricky PDEs. It's reached its culmination in modern engineering design, where most advanced products are designed and simulated in computers, and prototyping only occurs at the very end of the process.
The Soviets had computers.... some home built, some Western, but generally speaking they weren't very good. The Soviet approach was also very stereotypical: get an army of mathematicians and engineers to find exact analytic solutions to the problems you're trying to solve. You'd have armies of engineers and technicians designing things that in the west we'd give to a couple of engineers with some computer time.
The end result is that some Soviet engineering is stunningly brilliant. And a lot is absolute crap. One of the reasons the west won the cold war is that we were just much better at solving partial differential equations. This report is unsurprising... the Soviet approach just seems so stupid to any Western engineer unfamiliar with it, that you'd have to assume they had some magic trick up their sleeve. But nope, just a lot of brainpower misdirected into a lot of horribly inefficient pursuits.
I think it's safe to say they didn't have any exotic computer technology. Of course, hindsight is 20-20. ^.^
No? It's well known that the Soviets developed computers based on ternary logic (rather than binary) -- that seems pretty exotic to me. I thought it was equally well understood that it was more expedient to switch to clones of Western technology, so that's what happened.
Breakfast served all day!
Knowing the parameters they have to meet now, amateurs have managed suborbital rockets with minimal computation. With the recent change in the upper bounds of amateur spaceflight (ie. when FAA says NASA takes over permissions) and the knowledge in hand, amateur orbital flight is a matter of time. NASA helped develop and made use of VLSI not because it made what they were doing possible, but because it made what they were already doing easier.
As for doing without, the Russians provided us with proof positive during the Apollo-Soyuz Test Project. They flew first and we went up to meet them because we had better aim. Our guys used an HP 48 handheld for calculations and their clock was fed by the signals from the atomic clock at National Bureau of Standards. When we got there we saw they were using, respectively, slide rules, pencil and paper, and a stop watch. But our having the better technology did not prevent them from getting there. And their having lesser technology did not prevent them successfully participating in the several cat-and-mouse rendezvous practices that followed the first.
"I may be synthetic, but I'm not stupid." -- Bishop 341-B
(More cynical types assume the Russians bought or stole US chips from the French or other too-helpful go-betweens.)
Back in the early '90s, one of my professors had come over from the USSR to teach Comp Sci. The local ACM chapter, at least a couple of times if not more, had him give a talk on the state of computing in Russia. This was exactly what he laid out. That shell companies were setup in France to lease IBM equipment (all you could do in those days for this very reason). The shell would fly-by-night the IBM to Russia where they would part it out. Notably, iirc, Romania was where they reverse engineered the machine code of the OS back into a somewhat usable assembly language. This, he would explain, was why all the really nasty viruses for PCs came from Romania - because the writers could eyeball instruction code and tell you what it was going to do. They also knew every crevice of the system, which became the advent of viruses hanging out in BIOS's and system clock memory.
He eventually became uncomfortable giving the talks and stopped, to my knowledge.
TFS reflects what US didn't know back then, not the current state of knowledge...
Apart from Setun mentioned by other posters (which, although interesting, didn't really influence much the race in technology; about which the pdf is all about) there's also, most importantly, this gem:
http://en.wikipedia.org/wiki/Elbrus_(computer)
Soviet domestically developed supercomputers. Multiprocessor superscalar RISC machine few years before the report from TFS was written; later VLIW long before the Itanium. Used specifically in "how the hell Soviets are keeping up" areas
(the man apparently responsible for them works for Intel for some time now...)
One that hath name thou can not otter
I worked in a college computer lab with a Russian expat
He was extremely familiar with DCL (Digital Command Language) and VAX architecture. Apparently, he had spent years working on DEC VAX clones in the old Soviet Union.
I also remember reading that DEC would etch stuff like "check six" in Russian onto integrated circuits to let the Russians know that they knew it was being reverse engineered
Wherever You Go, There You Are
Thank you for your comment, Sznupi, but I have a few remarks, that I may make because I am myself a computer designer. The Setun example was served to me several times in the last decade and the same arguments apply :
1) Just look around you : where does ternary logic live ? in some Russians' fond memory. OK.
Show me where ternary logic can replace things : AFAIK, it is used in *some* multiply hardware, under the name of Booth recoding.
http://en.wikipedia.org/wiki/Booth%27s_multiplication_algorithm That's all, and it is not always practical : booth computations are a bit faster but recoding is a pain.
http://www.fpga-guru.com/multipli.htm
2) do the maths : Ternary logic values on binary wires :
- either you use 2 wires to encode 3 values and you lose 1/4 of the coding space (as in any base conversion)
- either you use the 3-wire 1-hot encoding and... well, you win nothing.
Now imagine you have binary memory : you lose 1/4 of the capacity. You can recode data so you lose less, but the less you lose in space, the slower it runs because it adds complex base-conversion circuits, with all the carry chains and the likes.
Memory in the first Russian ternary computer was certainly magnetic core memory : with the epoch's electronics, it was not difficult to encore magnetic 3 fluxes. But it does not work well in today's very high speed logic, where noise resilience and process variations can kill electric margins.
Conclusion : we live in a binary world, it's not by mistake.
Now if some electronic circuit worked WELL in ternary, it would not be enough : it would have to work WAY BETTER than today's binary circuits to even consider acceptation.
Don't get me wrong : I respect Russians a lot. But we all make mistakes and invent our little prides... All engineers have their failures... It's part of our learning. It is a greater failure to not learn from our mistakes.
Ternary computations were a "local minimum" for a given time and technology. And I don't regret the time when the US's supercomputers (http://en.wikipedia.org/wiki/CDC_6600) used 60-bit words and 6-bit bytes. The next generations of Cray designs went to 64-bits wide registers and 8-bit bytes, and they even adopted (reluctantly) IEEE Floating point numbers. This proves that even when technical merit is stellar, it is useless (and even laughable) if it can't interface to the other computers. Adapt or die bragging.
Not to mention that the Shuttle, and Earth-vicinity spacecraft generally, don't really need much computing power. You have ground-side machines to do the heavy lifting (which isn't all that heavy) and you transmit the plan to the orbiting craft. All it has to do is execute.
The report really does sound like a bit of sophisticated propaganda to convince Congressmen to fund nifty research mathematics. It is very strangely focused like a review article on niche mathematics and computer science.
The solution the paradox is simple: the USSR really was behind, but in the particular military areas mentioned (ICBMs, spaceflight), it is clear that advanced VLSI is not necessary. The USSR was not so far behind (or at all) in hard engineering like metallurgy, thermodynamics, rocketry etc, all the areas which are absolutely necessary for spaceflight.
Remember that the difference between the West and USSR was in economic efficiency. VLSI was just way too expensive---so Soviets had to make do when the West would use economical, high performing chips. The necessary computers embedded inside weaponry and rockets through 1984 simply didn't need to be that complicated. They usually had to run a simple control loop & switching system, which was designed and simulated off-line by large stationary computers in the lab. And more often in the USSR's case, analytical pencil & paper computations. The USSR had a much stronger applied mathematical understanding of nonlinear dynamics and chaos---in the USSR fluid mechanics wasn't shunted off as a boring part of civil engineering, but stayed with the high-level physics community the whole time. The West started recognizing the importance right about in the mid 1980's.
The deficiency in high performance semiconductors DID, in truth, hurt their military capacity in some areas: those areas where advanced semiconductor technology is essential, and not just an economically effective choice.
Primary examples are anything which involves combined analog/digital operations, for instance CCD imagers, and modern wireless digital communication devices. A critical example: high resolution spy satellites which transmitted the results by radio and not film canister.
For instance: despite great space flight experience, the USSR didn't come remotely close to having a capability in the 1980's like the Global Positioning System, or relatively cheap spread-spectrum communications (almost everything we have now is from original military developments), or fancy infrared imagers and image analysis software embedded in a warhead's targeting system. All those require advanced, embedded, launchable, semiconductor technology---a cloned VAX in a building won't cut it.
After seeing the results of the Gulf War in 1990 a Soviet general was very relieved that they never went to war with the West. The USSR was astonished at the capability of precision bombing from the F-117 et al and the necessary logistics & ground & airborne communication systems supporting such a campaign. Iraq didn't have the capability and certainly training of the USSR but 1990 Iraq had some decent Soviet hardware, which was nearly totally ineffective in combat.
It meant that in a war in Europe NATO could have smashed a Soviet armored assault without nuclear weaponry (, and the USSR strongly underestimated this conventional capability driven by technology.
One lesson is that the technological capabilities of Chinese weaponry today shouldn't be underestimated.
Most practical problems can be solved with marginally acceptable accuracy without computers. In "the old days," modeling efforts were utterly crippled by the lack of computers so we had to give everything a good margin for safety and hope it was enough.
Try to design an engine that meets modern emissions requirements without a computer.
Try to make detailed predictions about the behavior of any circuit containing multiple transistors without a computer.
Try to design a modern-scale bridge without a computer.
Your post Sir, is a fine example of brainwash. Anybody with a bit of background in remote sensing can tell you that in the beginning of the 90s satellite pics with a resolution of about 0.5 metres could easily be bought on the market, coming from decommissioned Soviet-Union satellites. And them pics surely weren't sent to earth by canister either (that practice ended in the 60s, thank you very much).
I won't make any remark about your admiration of the precision bombing from the F-117 and the associated cost/effectiveness comparison (at least not until I've visited Iraq and Afghanistan), but really, using the Iraqi Army, after 12 years of embargo, as an example force in comparing the effectiveness of the US military versus the Warschau Pact?
What I do wonder about, really, is that after your headlong demonstration of the inferiority of Soviet material, you come to the next conclusion: "One lesson is that the technological capabilities of Chinese weaponry today shouldn't be underestimated." Underestimated?