Slashdot Mirror
Russian E2k CPU at 135 SPECint95 / 350 SPECfp95 ???
"I get MPR. I've got about 7 minutes before I have to catch a bus, but, from the MPR issue itself:
The processor uses EPIC. The Elbrus team has been together for 40 years, originally designing supercomputers for the Soviet defense establishments. "They've developed computers based on superscalar, shared-memory multiprocessing and EPIC techniques long before papers on those subjects appeared in the West". MPR claims that the lack of a good semiconductor Fab has been what was holding them back. MPR says that the claims would be unbelievable except for the credibility of the team.
The X86 and IA64 compatibility rely on binary compilation assisted by emulation hooks, similar to what Transmeta is apparantly doing. Supposedly Dave Ditzel spent several years while at Sun working with the Elbrus team.
The processor exists only as an executable Verilog database. However, the E2K design is based on the Elbrus-3 processor that was fabricated in 1991. The Elbrus-3 was built in an "ancient process", used 15 million transistors in about 3000 LSI and MSI chips, and delivered twice the performance of a Cray Y-MP."
Some more he sent later:
" It is actually quite a long article... 6 pages plus the cover, I'm about two thirds through it. The architecture is in fact pretty stunning, and very similar to the Merced and the SPARC in several ways. It has a 64K, 4-way instruction cache: one i-cache only. It has two identical, synchronously-loaded 8K L1 data caches, and a 256K, 2-way, 4-bank L2 data cache. In addition, it has a 4K array pre-fetch buffer for use in loop overlapping. There are two regions, each with an L1 data cache, a 256-entry register file, and three ALUs. The regions are symmetric except that only one region has a divide function.
A great deal in this processor is left to the compiler, a fact that is demonstrated by the single, 64K i-cache; this will only work if the compiler does its job. Much also depends on the compiler's ability to identify instructions that can be executed in parallel. With an optimal instruction load, the multi-ported caches can provide a potential operand bandwidth of 288 Gbytes/sec at a processor clock of 1.2GHz. Much effort is expended to avoid branching; extensive branch prediction support is provided, and in some cases it will actually just go ahead and execute both sides of a branch to avoid doing the branch at all; with so many parallel execution paths, the cost of doing so is much lower than what would be the cost of branching.
When loops are identified, an effort is made to overlap the loop execution, taking advantage the same mechanism as used for the sliding register windows. The 4K FIFO Array prefetch buffer helps to feed data to the overlapped loop. In loop mode, for perfectly optimal code, the processor can rates as high as 23 operations per cycle.
Much of the processor is designed in standard static CMOS gates, but some of the critical paths through the processor use self-reset gates, which do not have a clock but rather are triggered by the completion of cycles in previous gates. According to MPR, these are estimated by Elbrus to run 10-15% faster than static CMOS gates.
Just a couple more facts about Elbrus: The Elbrus-1 computer was a "...superscaler, RISC, processor with out-of-order execution, speculative execution, and register renaming..." This machine was designed and built... between 1973 and 1979!! They dumped superscaler designs becuase they were too complex for the payoff. The Elbrus-3, built between 1985 and 1991, used "an EPIC-based VILW CPU", implemented as a "16-processor shared memory system"
They started working on the E2K in 1994, and it is now at Verilog RTL stage, with compilers and binary-compilation software written. MPR expresses great doubt that a home will ever be found to build the processor, what with the Russian economy as bad as it is, and most capable semiconductor houses already in the midst of implementing their own designs or just not wanting to compete with Intel."
Cool. Hope it's true, that ought to send some pricing shockwaves through Intel.
Well...350 specfp95 are pretty high specs :)
The A21264 architecture will reach 150 specfp95 in 1 year or something, so it doesn't sound *THAT* unlikely when there's no prototype or anything.
US5808926: Floating point addition methods and apparatus
Does this guy work for Transmeta yet?
First the supercomputer that fits in a toaster, whups up on a MIPS12000, does your taxes, sucks you off, and retails for $10...
Now a low power russian skunkworks chip that shames all current technology and works on pinwheel power... Woo!
Enjoy the life folks..
-MJEm
I prefer it that way.
hmmm...sounds like yet another conspiracy theory to me. I recently took a hardware design class from an ex-intel engineer who helped develop the 8080 and the 8085. After seeing how difficult it was to develop those chips, I have great respect for those guys. Pentium in '70s? Come now. You must be smoking something awfuly strong, or you are just stupid. I personally think the latter. :P
Well, this may be a very significant advance in computers, coming from Russia. But I think Russia contributed a lot already, with Vodka. When my NT Network crashes, I hit the bottle like a mofo.
...that can write upside down, zero gravity, anywhere.
The Russians bring a pencil to space.
256MB L2 cache... that would be great!
nasa puts windows 95 in space.
Cache that's on the CPU chip is usually referred to as L1 cache, and it's much faster than L2 cache that communicates with
the CPU via some sort of bus. I haven't read the article, but it seems that L2 is a typo and it should say
L1 instead. For comparison. the pentium pro has a 16K L1 cache and the p-II has
32k.
Fisher actually "donated" a lot of the R&D effort that went into the space pen. They thought they were gonna make themselves filthy rich, but space pens are really just novelty items.
That pencil sheds graphite in the compartment. Graphite the crew ends up breathing and that accumulates on the electronics.
Hmmm, good plan.
No wonder their saucers keep crashing
It's all fine and good to have high-end targets, but if your compilers don't expose the parallelism in the instruction stream, your hardware doesn't mean diddly. At the end of the day, when these guys say "oops, couldn't get it working", that'll be their excuse.
"A few small tweaks after the first x86 release to add IA-64 compatibility?" Puh-lease. New registers, predication bits, plus a dozen other low-level architectural things to imitate, and that's all on top of non-disclosed Pentium III security features, streaming SIMD, etc. all have to be engineered.
Probably Russian engineers trying to get funding.
The only odd thing is why MPR was so gullible? Maybe they weren't. I haven't seen my copy yet.
Wherever this L2 cache is housed, 256K is incredibly small.
That's the weirdest thing about this announcement. How on earth could they get such performance with such a small cache?
Heh. Maybe they've got a 1024k L1 cache. That would solve everybody's problems.
"Currently MCST is working on a new project, a powerful E2k microprocessor based on postRISC architecture with explicit parallelism and binary compilation system which allows for highly efficient and reliable implementation of Intelx86 binary code with approximate performance twice that of Intel's P7 (Merced). By now the project has passed the international technical evaluation in the major audit company Price Waterhouse, USA, and the search for strategic partner for joint implementation of the project is underway.
The key points of the project is development of the architecture and "superoptimizing" compiler that allows for efficient usage of parallel execution in the program. The work in these areas is the backbone of MCST system research. Another important direction of the company activities is creation of advanced binary compilation technology that provides 100% reliability and high efficiency of implementation of binary compatibility with Intel for the new architecture with x86 and SPARC architectures." (This last sentence is confusing in Russian too :))
And unmaned at that. The Space Shuttle cannot fly unmaned.
It never flew again because of money problems.
...the stupidest bunch of losers i have ever seen, I would not ever trust technology coming from a country which can not feed 75 % of its own people, Ruskies suck dick...
After reading Norr's article what impressed me most was not Russian vaporware (though that IS interesting) but how behind the curve Motorola is, chip-wise.
Norr's assesment of the G4 is pretty damning. This was meant to have been the line of defense against Merced, and now it looks to little, too late.
Why are you folks making such a fuss of this? It's VLIW, and performance numbers mean absolutely nothing when it comes to VLIW because you need a smart compiler that can take advantage of the hardware - I have yet to see one that can do this job with non-scientific code.
In case you kiddies don't work in the field, Boris Babayan (sp?) is a well-respected computer scientist and they are currently being "sponsored" by SUN.
Yes, so they have the latest, greatest Sun Workstations and design tools that we use over here.
_If_ the report is true and Boris hasn't been taken over by aliens, then his claims should be somewhat within reason.
The Russians ('nee Soviets) have always had brilliant scientists *and* engineers, but have somehow lacked the mechanism to get their high-end theoretical work into the hands of their draftsmen and steel-benders. They can't seem to put the '&' into R & D. The two always remained compartmentalized for them.
;-)
The 'R' is good: World-reknowned physicists, chemists, and mathemeticians.
The 'D' is good: They were able to create the Mig-25, a Mach 3 interceptor made of mostly steel (only used titanium for leading edges). We were totally freaked by it until we got our hands on one and found out it was just an amazing aeronautical engineering hack. The engines were only good for 300 hours of operation before requiring major overhaul, and they would pretty much fuse if you ran Mach 3 for more than a few minutes. Nonetheless, it was impressive what they could do with old technology -- kinda reminds me of Intel engineers
They just don't seem to be able to flow technology from the scientists to the engineers. Perhaps this is what the engine of capitalism does...
It is the other way around...
Not only Russians lunch American satelites,
but there is a LIMIT imposed by the US goverment
on the number of US satelates Russia is allowed
to lunch. Every time strings need to be pulled,
Clinton treatens to cut off that number.
As for the chip I hope it is true because it would
be so cool to have a cheaper alternative at lower
power. That is what is needed to keep Intel in
check...
We will now how true it is when the time comes.
To speculate now has no value.
They invented the 100 percent parallel computer.
They could before they stoped being commies
The Russians actually know what they are doing. they have don many things in the past that the US and other countries play catch up.
For instance they got the first person in space.
They have a missile that can fire backwards on a plane, the Us can't get our copy to work.
The Soviets have always had some smart engineers..the problem is with money..they have none. In many cases, they have gone low tech with superb results.
Buran did fly. It was considered to be quite a bit better than the shuttle.
Everyone laughed at the primitive electronics in their aircraft, until they realized that it not only worked, but was cheap and could not be destroyed by nuclear EMP.
As for satellites...who is launching whose now? The yanks just got government approval to have the Russians launch satellites for them. Why are the Soviets building space station pieces? Because they know it better than the US and have the lift vehicles with the highest payload capacity.
Typical American arrogance.
cheers mate, I wouldn't say better.
>The processor uses EPIC.
I think you mean VLIW (Very Long Instruction Word), not EPIC (Explicitly Parallel Instruction Computing). VLIW is a common usage term like RISC or CISC which describes a style (or, more like, philosophy) of CPU design. Intel wanted to sound cool when they designed IA-64, so instead of describing it with the general term (VLIW), they created the acronym EPIC (surely a marketing decision, along the lines of "pentium"). I believe EPIC (like "pentium") is trademarked. Thus, AMD can make a 586, but not a "pentium". Likewise, another company can make a VLIW chip, but not an "EPIC" chip.
little marketing games we like to play...
--Lenny (not really anonymous)
Didn't you know that Russia has the biggest microchips in the world?
The L(evel) number only says how close the cache is to the CPU, not if it's on or off chip, on another board or wherever. L1 is the cache closest to the CPU, L2 next etc. You can have as many levels as you like.
The K6-3, and the alpha 21164 both use 3 levels of cache for instance. 2 levels on-chip and 1 level on the mobo.
I got the same impression re: G4 from Norrs's article; pretty unimpressive.
It said the SHIPPING G4 would be 450 mhz- thats shipping, which won't see volume for awhile (i.e. 9 months) It also said the integer performance would be basicly the same as todays G3. No big woop. By then the Intel Model-T will be spinning 800 mhz, not to mention all the other groovvy silicone that will surface in what is basically a light year, technology wise.
Meanwhile, IBM has their G4 (Ativec-less) running 650 NOW- and that the beginning for those chip wizards. Unfortuately for Apple, IBM doesn't want to use Activec, because it doesn't do much for a server, which is the main use for IBM.
IBM and Motrola aren't co-developing the PPC chip anymore, though there is cooperation. IBM was tired of giving Motorola the free lunch, technology wise. Motorola is WAY behind IBM chip fabs, silicone on silicone, advanced copper et al. Motorola likes that el cheapo low power embedded market much better than chase that high performance market.
Motorola IS dragging their behind on the G4 Activec version. Which really sucks for Apple, since Motorola is Apple's one and only supplier for the Activec G4.
Is doesn't help that Motorola is very Anti-Apple after Jobs pulled the plug on the clones. After that Motorola cleaned Macs out of their offices- even brand new Macs - and install INTEL windows machines. How's that for a little clue about how they feel about their good customer, Apple.
The Power PC is GREAT architecture. Its a pity Motorola isn't doing much with it, except for making really smart toasters.
256Kb may be small as a direct mapped L2 cache, but as a 2 or 4 way set assioated cache it would kick some serious ass!!
(Thats why the 300a doesn't outperform the 21264 even though they both have 128K of on chip cache, the alpha's big cache is 4(perhaps 2) way set assoative..
The only question is, who are they going to sell the design to? There is no way in hell they'll ever be able to manufacture it in a competitive timeframe.
With regard to your emulator notion, I could except metrics, like MFLOPS or MIPS in that respect, but it is too much of a stretch to extrapolate Specmarks. Those test suites are very complicated...the SpecFP is actually a suite of real scientific applications...Poisson Solvers, CFD, particle simulations. Those measures rely heavily on the memory and cache architecture. Now, they may have a 0.5 micron version which ran at 1/10 the MHz and reached SpecFP of 35 and they are performing a trivial scaling. I just find it very difficult to believe that they have a memory subsystem that can keep such a CPU fed. For example, the current 21164 Alpha systems use SDRAM DIMMs in pairs, forming a 128-bit data path. That Alpha system "only" manages 20 to 30 SpecFP. To keep this Russian CPU fed, I suspect that you will need at least a 256-bit data bus, maybe 512!
In short, Specmarks can only be quoted for an entire system, not for a CPU. They have clearly made gross assumptions in reaching those numbers.
The G4 is in silicon *now*. Are we to assume it will not increase in clock speed until Merced shows up? (And when exactly _will_ Merced show up?) These early G4s are also not multicore.
;-)
Further, Norr's assessment only said that integer performance would not be much higher than in the current G3. Floating point will increase significantly. Most importantly, AltiVec was not really addressed, and AltiVec is much more than a bolt-on like MMX or even KNI.
AltiVec is a full-fledged processing unit like the integer or FP units, with its own set of registers. It will totally take over some functions currently performed by the integer unit, particularly the most processor-intensive.
The G4 should ship later this year. When Merced finally shows, then we'll see how they stack up. Beyond that, I want to see what a Merced laptop looks like
...mmm...
:-)
olga! i love you!
hope to see you soon,... !
For the 3 years I worked on probation in MCST (Moscow Centre of Sparc Technology - it where Elbrus team located) while studing in Moscow Institute of Physics & Technology. Many of the kernel's group developers (including Babayan and Sokhin) lecturing to us (MIPT's students). Believe me - they know what they do, they know what to do and they know how to do. And I'm sure - in a 3 years their E2k will be done not in the
model, but in the "metall"!
mudak!
nah - Slashdot comment boards have the largest number of vaccum tubes in active service...
Nick May (who can't be bothered to login - nick@fukkers.com)
Well - they have had the equivalent
of RTL simulators and such for quite awhile.
I work with the Russian fellow that wrote
several of these tools.
They are VERY clever folks and such
claims would appear quite likely to
be true - I suspect I'll be having an
interesting talk with my Russian
friends next Monday!
Oh - just to give this some credibility -
I design IC's for a living now ( and have
in the past worked for one of the mid-80's
VLIW companies..that didn't make it.)
The russians have always kept hidden their technology, so it's impossible to say what they have in store. Their biggest problem is marketing. They don't know how to sell the stuff. But aparently these guys have. Letting SUN into the game was a good idea too. SUN's not a small company people, they have access to many things. Including high-tech designs that other people dream of. And the SPARC processor kicks ass, as an Intel based server can never reach the level of reliability and speed (toughness)of a SPARC machine. Nuff said, I belive them. I admit, I want to believe them, because it's time that someone made a thing like thins and marketed it right. I'm getting tired of having to upgrade my machine every few months so that the software runs properly.
Okay...it's a quote from "Dr. Strangelove". Don't flame me too badly.
Hey, no offense, but Stalin died in '53. :)
So who did this Russian company steal the technology from? Transmeta? IIRC the Russians.. err.. Soviets.. never made anything original and I doubt they're going to start anytime soon.
Yeah and nearly blew up the spacecraft in the process in your effort to get there.
Grow up. The cold war is over (or should be anyway)
What's that have to do with the price of tea in China? Win95 serves its purpose when its purpose is to just give computer illiterate astronauts access to email.
da kakoi zhe on pedik, on zhe pidar
You are just a little brainwashed looser. I can enlighten you on a number of technologies that west could only dream to put their hands on, but, once again, you are just a looser and won't be able to comprehend.
Before you dis MIR too much, keep in mind that it is WAY WAY past its design life. Consider how long it's been in space, AND the length of the average 'stay' on MIR. Then compare that to our long since defunct 'spacelab'....
The Russians have proven over and over that they are better at building rockets that work than we are. They launch their stuff in nearly any weather including below freezing, high crosswinds, etc. We on the other hand require nearly perfectly clear days, low winds, specific cloud ceilings, etc etc etc. Consider how often our launches get scrubbed due to weather. That's nearly unheard of for them.
Their track record (launches vs failures) is much better than ours. There are a LOT of american companies that would have gladly had their stuff launched on a Russian vehicle, (rather than wait years for their turn on a shuttle or Atlas) HOWEVER there were laws that specifically forbade that. I'm not sure how many are still in effect, but it's my understanding that if you want to put a package up on anything other than a NASA vehicle, it requires extensive paperwork, and darned near permission from congress and the president or you get thrown in jail for treason or some such.
If you don't make the instruction set public, nobody can write code to run on your processor.
Without code that runs on it. a processor is useless.
If you want to have a market for your processor, and have it adopted quickly, you have to make the instruction set public very early on, or at least make it available to a large number of software developers (unless you also want to write all the compilers yourself)
well put. Now if we could just do something about the signal to noise ratio.
Or come up with better filtering so I'm not wasting brain bandwidth reading such intellegent comments as "rooskies suck disk" and similar dross
http://www.osp.ru/cw/1998/03/business/ 05.htm
http://www.infoart.ru/chat/arc/babayan.h tm
http://www1.infoart.ru:8000/ it/press/cwm/29_98/sun.htm
Acording to that they are developed several compilers for sun (including Java). That's where the money comming from. Boris Babayan is the father of Elbrus-1, 2, 3 and now 2000.
back when i had a security clearance, and programmed crays, reagan was president, and we all feared the evil empire. there was a rumor that the ruskies had built special-purpose computers that were far more powerful than the conventional estimates of their technology.
the story went: "suppose one were to build a computer that might work one day a week and sit around for six days getting fixed. but on that one day, it'd kick so much tail it'd make up for the downtime."
sounds like that's what they were working on that was "discarded" because it was too complicated.
NJET!
m l)
OK, maybe I can suppose the Elbrus group was not a hoax and did some work
for Sun, and the project existed. But the claims made in the article
supposedly quoting MPR are just too far off.
(see: http://slashdot.org/articles/99/02/19/1543222.sht
1. "The processor uses EPIC".
Right. How? Did they have a NDA with Intel 5 years years ago? Ah non,
it is explained later that it was "EPIC-like". So it doesn't use EPIC.
The sentence in the article is nothing but a misleading eye-catcher.
2. "The Elbrus team has been together for 40 years".
The only 4 (sic!) employees given on the home page don't look like they
are 65.
3. "Elbrus-3 delivered twice the performance of a Cray Y-MP".
It did? Where is it? Where are the publications about it? Where are
the test programs, the operating system, the architecture specification?
When did it beat a Cray Y-MP? Which Y-MP? On what benchmark? I'm taking
bets for bottles of champagne that it never beat a Y-MP, not on
memory bandwidth, not on MFLOPS, not on anything except bogon throughput!
4. "It should also be both x86 and IA64 compatible"
Bogus. No ancient processor can be IA64 compatible because IA64 didn't
exist then. Even today, there's veryvery probably no non-HP/Intel
IA64 compatible design.
If they think that "IA64 compatible" means something like "in principle
it would be possible to write an IA64 emulator and execute it on the
other architecture" then *anything* is IA64 compatible, even my Z80.
That is a worthless claim.
5. "has a CPU design that achieves 135 SPECint95 and 350 SPECfp at 1.2GHz"
Aha? They claim they have, but where is it? No known microprocessor
CPU design anywhere would deliver that amount of performance even if
we'd assume linear scaling with clock frequency. Particularly bogus is
the fp claim, because that is very dependent on the memory system.
6. "the design is very similar to the Merced and the SPARC"
It cannot be very similar to both of them, because they are very
different! If the author means the similarity is because the L1/2
cache sizes are the same, then my VW and a Porsche are the same
because their fuel tanks take the same amount of fuel...
7. "the processor can rates as high as 23 operations per cycle."
Well, then it better have at least 23 functional units on the chip
that run in parallel! Does it?
8. etc. pp., but I'm getting tired.
Last not least, the article claims Merced SPECint 45 and fp 70 as if
they were matter of fact, i.e. not qualified as "estimates". This
serves to illustrate the quality of the guesses made.
I'm not saying there aren't good designers in Russia. But hey, they're
living in the same physical world as we are.
Cheers,
HK
--
Henrik Klagges - IT Analyst
henrik@strategypartners.com
PGPKey available on request
1.) VLIW and "binary compilation" aren't
:)).
Intel's creation. I doubt Dr. Bobaian
would call the design EPIC although
he might say it competes with it.
2.) It's the enterprise, not the team that
has been around since 1957.
3.) Twice the performance of a Y-MP is
really no big deal especially in 1991.
The important thing is the VLIW design
experience of the company which allowed
them to get this level of performance
from LSI components!
4.) Given enough design abstraction this
is not a bogus claim. In fact it could
emulate any instruction set given the
right "compiler". The whole point of
EPIC and other similar design philosophies
is to extend the degrees of freedom in
the architecture by moving much of what
was previously hardwired in silicon into
the domain of the compiler.
5.) What is your baseline for the linear scaling?
You are comparing apples and oranges. EPIC
and whatsuch is NOT in the "known processor"
domain to start with. I don't find it so
bizarre to obtain 2.5X fp performance of the
Alpha 21264 if the instruction stream can
be super-optimized through the compiler in a
way that no "fit-all-possibilities" hardware
compromise can.
6.) The reference is to the cache layout, which
is taken as indicating the input of Sun
engineers in the project.
7.) Just like in the Merced there are multiple
pipelines for floating and integer
instructions. The Elbrus design likely
has 23 pipelines (must include some I/O
ones). Of course without the "magic"
compiler the thing is useless.
I agree about estimates for non-existing hardware
(or "mushware"
Narod, da chto vy obraschaete vnimanie na utomlennogo neprivychnoi mozgovoy deyatelnostju gringo, mesto kotoromu u parashi? On zhe nichego vokrug sebia ne vidit i znat' ne khochet -- tipichnaya nevezhestvennaya serost'...
P.S. "Offtopic, Inflammatory, Inappropriate, Illegal, or Offensive comments might be moderated." What about this?
Stalin died in 1953, not in 1963.
Posted by BikE_PUnX:
kid, how old are you?
russia has one of the most advanced aerospace programs in the world. remember, they managed to go from nothing to putting a man in orbit in less that 50 years. just because they have problems with their currencey (ans yes, that is what their problems mainly are -- everyone has enough food and clothes and heat to survive), do not assume everything else has gone to crap also.
---
"If you can't fix it with duct tape, it's fucked."
Yes, the west laughed about those vacume tubes in military hardware for years. Then we thought about EMP (electro-magnetic pulse). IIRC everyone got really quiet for a few years after that.
The belief that Russians cannot out innovate an American company is no different that the U.S. government's apparent position that only Americans can create strong crypto technology.
The USSR got a man in space first. The USA put a man on the moon first, USSR landed a probe on Venus first, US made it to Mars first, USSR holds the record for space station service time, etc, etc... The space race is really a tie in the long run.
http://www.ipmce.su/
http://www.el2000.ru/
K, I found them, now someone translate for me? :-)
Basically, I think it's BS. It's not real, it's vaporware, not hardware. There is no existing stuff, it's "planned" for thier "project." They are partners with Sun, BUT.... The project is not even lead by a Russian, it's lead by an Armenian(sp?) guy, and Russians have a trust issue over that, because it might not even benifit thier country (which could use a technological boost).
Anyway, that's the spin I get from http://www.el2000.ru/news/presentation.html
That raises a red flag for me-- when did Intel make the IA64 instruction set available to other processor designers?
Sun microsystems lists them as a partner. They used to be the Moscow Center of SPARC Technology.
What the hell you mean "where it's not needed"?
It's needed everywhere.
"Be nice, veer left, and never stop thinking" Iain Banks - Walking On Glass
I've always knew that military didn't use the crap stolen by KGB from West (i.e. OS/360, VAX clones)...
:-)
Now I at least know the name
If the soviet government wasn't so paranoid about their military technologies we could really have another Silicon Valley now..
I was unaware of any computer hardware coming from Russia.
--
"Reactionaries must be deprived of the right to voice their opinions; only the people have that right." - Mao
"Pentium" is just a marketing name.
--
"Reactionaries must be deprived of the right to voice their opinions; only the people have that right." - Mao
RISC chips have traditionally larger caches to be able to cope with the larger number of instructions. Newer Alphas have 4+ meg of L2 cache.
--
"Reactionaries must be deprived of the right to voice their opinions; only the people have that right." - Mao
I've known for some time that the Russians had the same level of scientific and engineering that the rest of the world does (I wouldn't be surprised if China is right on up there with us- Nature reveals her secrets to anyone who asks the right questions...) but that the USSR happened to be hampered in a few critical resource areas and as such, some things suffered (like semiconductor tech)- while other things flourished like optics (they make some of the finest in the world!) and tube technology (don't laugh- tubes are intrinsically immune to EMP attacks that'd take out solid state devices and they are the ONLY route to high power levels.).
I'm surprised that the people behind this design haven't contacted the people that got behind Svetlana Electron Devices and got something going with them. (Svetlana is one of the largest suppliers of vacuum tubes in the world)
I am not merely a "consumer" or a "taxpayer". I am a Citizen of the State of Texas
How can they say what the SPECint95 and SPECfp95 ratings will be when they don't even have a prototype? And what secret knowledge have they scryed from Intel and HP that they know enough of the inner-workings of the Merced chips to tell what those specs will be?
We might as well read the Farmer's Almanac to determine this "supposed" chip's ratings. Their
design may be innovative or it may be non-existant, but we won't know until something
is done with it. Blech.
--
Aaron Gaudio
"The fool finds ignorance all around him.
"Every man is a mob, a chain gang of idiots." - Jonathan Nolan, Memento Mori
See this page on the BESM-6 computer.
Wasn't that just a question of economics rather than technology? Given that they couldn't afford to feed themselves, how do you suppose they were going to afford all that kerosene?
I'm out of my tree just now but please feel free to leave a banana.
Well, this is not a hoax. The chip is based on a
post-VLIW design. Apparently it is
5 seconds close to be baked in the silicon.
Some details are available at the Elbrus
web site: www.el2000.ru (in russian).
This company was building computers since 1959.
There have been too much self-indignation on how Asia having to rely on Silicon Valley on everything....
Now that the Russians are proving that they can produce something this amazing even without a good semiconductor Fab.
Those of you in Japan, Taiwan, Korea, China and India better take heed --- you've got the Fabs, why don't you work with the Russians and produce this amazing chip?
Are you listening, Asia?
This is the BIG CHANCE for you now !!!
JUST DO IT !!
Muchas Gracias, Señor Edward Snowden !
There have been too much self-indignation on how Asia having to rely on Silicon Valley on everything....
Now that the Russians are proving that they can produce something this amazing even without a good semiconductor Fab.
Those of you in Japan, Taiwan, Korea, China and India better take heed --- you've got the Fabs, why don't you work with the Russians and produce this amazing chip?
Are you listening, Asia?
This is the BIG CHANCE for you now !!!
JUST DO IT !!
Muchas Gracias, Señor Edward Snowden !
From the "about page":
They are working with Transmeta on "commercial projects, related to creating apparatus and programming environments for most modern calculating architechtures" (or something to this effect)
Interesting...
Actually it said it would be about 5 times faster than intel's merced, which is at least 2 years off.
Entirely possible they've been doing work for Western companies under contract, and have all the software etc. Isn't that hard to envisage.
Moore's Law actually only indirectly affects CPU speed.
Moore's Law states that transistor density doubles every 18 months. Finding how to use all those transistors usually results in far less than a 2x speeded.
Strikes me that it's very probable that all us North American folk have been fed a line of bullshit by the media and governments all these years. Little wonder it's shocking news to hear that 'the Russkies' have advanced technology -- we've been damn near programmed to think they're a bunch of ignorant rat-bastard Commies who deserve to starve for not kow-towing our brand of government. After hearing this bit of CPU news, I'm now curious to hear about what other bits of advanced technology they've invented/been working on/kicked our ass on...
--
Don't like it? Respond with words, not karma.
Fountain pen.
Capillary action. No need for gravity. Ne pas?
--
Don't like it? Respond with words, not karma.
wouldn't it be nice to get that E2000 design under GPL and got a Free CPU that kickes ass in a same way as Linux does? I would feel much more like a free person then. But... it's not likely having Elbrus key people military industrial complex heritage. So, as long as it's not Free it's of not much interest really. Since I would agree that chances of Elbrus team pulling it in Russia is quite slim. If they come to the West for that they would be used and abused. Not by our community of course :-) ... And we need to put more effort into Freedom CPU project.
Vassili Leonov
I think they are more than a Sun partner. I remember Sun buying them in perestroika days. I also remember HP being quite upset since they were going to buy it, but Sun somehow beat them to it. That is what I remember anyway.
Sinan
Should ask Babayan's (Elbrus inventor) daughter, Osana, about that. Last what I heard from her, her dad was not very happy with how his team was working. Luck of funds etc. :) :)
BTW - she leaves in Palo Alto... Rather natural.. Called brain drain.. (Say hello to myself)
:)
Hope she does not read this...
In case she does - hi, how about Kirkwood this weekend..
<^>_<(ô ô)>_<^>
Ok, It's fine, it's dandy, it's not even seen the silicon yet.... this thing is 100% vapor, nothing more than a set of design goals really.
I'd like to see some hardware manufacturers quit talking about the far and wide and start talking about the here and now. Give us new high performance hardware that will be on the market soon, it is nice to see what's coming up, but c'mon, nice concept, lets see if it works when we build it... That just doesn't work for me.
Zanthor
Is this anything like their Buran, which was supposed to kick our Space Shuttle's butt, but never got beyond the testing phase - all the while ours were launching THEIR satelites??
I'll believe Russian cutting egde technology when I see it. Until then, Yuri Gagarin was their greatest scientific achievement of the century, and that's old news. They certainly DO NOT lack the brain power, their scientists are brilliant. But with little resources, you can't bring ideas to fruition. This is an attempt at investment funding.
Sorry to the Russian readers, but if your neighbors are burning furniture for heat, you're not 'cutting edge'. Yes, we have homeless too, but not for lack of resources.
Theory is nice, publish it. Practice is better.
-- What you do today will cost you a day of your life.
But at least I learned something in the precess.
Cheers!
-- What you do today will cost you a day of your life.
Come on. I'd like to think that Americans have grown up a little bit since the cold war. BTW, communism is actually a VERY good concept. Support everyone, regardless of their abilities to work. Welfare, Social Security, Unemployment are all basically communist programs. Communism is great in theory, it's just fallen apart in every real world trial. It's really a shame. I personally would like nothing more than a world in which money doesn't exist.
-matt
is that an eastern company will pick up and make this chip if western companies don't. Someone like Hitachi. I think they'd love to get the jump on something so high-end. Well, if the chip is actually producable, and the results are real.
He said, "You'll be able to tell your grandchildren that you helped assemble the first NT supercomputer," and I cringed.
not in Russia. :)
Intersting considering russian silicon technology wasn't very competitive. I guess necessity is the mother of invention. Put it togeather with a modern fab and you have something truly world class. Aint hybrid vigor wonderful.
The russians also once claimed to have control over cold fusion technology.
As any true skeptic would say, I'll believe it when I see it.
It was something like this:
Rus. politicians were freaked because of an imaginary technological gap between Shuttle and soviet rockets, so they ordered to immediately make 'our own'. Many design solutions were copied from Shuttle and...
good 'ol rockets are just as good and cost twice less. Also, Mir payed off while Shuttles haven't.
I never heard of russians doing anything good in consumer electronics. I remember when I was in Russia, 99% of computers were made from cheapo asian parts while 1% was compaqs, ibms and such. When you talk about defence technology, Russia was severely lacking development until '63, when Stalin died. Before that, Cybernetics was not considered 'real science'. Still, this chip is possible, if you consider that defence technology is being transferred to consumer market ever since the USSR breakup. This definitely looks more plausible than recent hypercomputer sensation.
-- ATTENTION: do not read this sig. It doesn't say much.
I agree..
The article that started this thread was saying that: Buran was a failure, even though it was copied from our shuttle but they couldn't pull it off, hence they can't design the chip on their own.
My point was that Buran's failure is due to the fact it was mindlessly copied from shuttle.
I don't know who's space program is superior.. I think even experts would have trouble proving that one side is better, due to the fact that space programs are very complex and multi-faceted and an expert who is sufficiently aware of one field in space program wouldn't know much about other fields.
I definitely wouldn't even try to say which program is superior.. its impossible to say just by following news or reading a few books targeted at general population.
-- ATTENTION: do not read this sig. It doesn't say much.
This seems like a classic case of vaporware. They have no Western sponsors and are admittedly several years from production. It never ceases to amaze me how often people will fall for this trick: "Our CPU will be twice as good as Intel's best current offerings! Look for it in less than two years!" Of course, given Intel's track record and Moore's law, Intel will have something twice as good as their own current best in 18 months. Meanwhile, though, their competitor gets lots of positive press.
Anyway, the Pentium was actually declassified military technology that Intel obtained in the early 70s; they've been releasing it piecemeal ever since.
Pardon my ignorance, but what Intel chip has ever had more? The Pentium Pro had >256K L2 cache, but it was on a separate chip (though it was in the same package). The latest Celery^Hon has 128K. Are there non-Intel processors that have more than 256K L2 cache included on the chip?
Err, the PPro is and was available with 512K and 1MB L2 caches, while the P-II has 512K, and the Xeon is availabe with multiple MBs of cache.
Err, the PPro had ZERO L2 cache on the IC inside the pin package. If you open up the pin package you found one square of silicon for the CPU and another square of silicon for the cache. P-II has L2 cache in a separate pin package, though it's inside the cartride. Same is true for Xeon.
Meantime, some PA-RISC, SGI, and Alpha systems have more L2 than some of us have system RAM.
So are those on the same chip as the processor, or a separate chip?
My original question stands unanswered.
Not quite correct. In the beginning, CPUs communicated directly with the RAM. When CPUs got faster, L1 cache was implemented. It was NOT on the CPU chip. It was on separate chips, elsewhere on the motherboard. With the 486, 8K of L1 cache was on the CPU chip. Many motherboards also added L2 cache. With the P-II, the L2 cache was taken off the motherboard and put on a separate board that also held the CPU. The Celeron has both L1 and L2 cache on the CPU chip. This is the situation that would appear to be the case with the Russian chip - both L1 and L2 cache on the CPU chip...
...or it could be just a typo.
Yeah, but you've got to reduce the chip size by a factor of four to reduce your linear measurements by a factor of two. Theoretically, that cuts the time the signals spend moving around by a factor of two. But you've also got to worry about the latency within each transistor, so transistor density alone isn't the only factor.
Yup. That's why my original statement was "given Intel's track record and Moore's law"
Of course, one might argue that Intel's track record is no predictor of future performance, but that's true of all empirical evidence, including Moore's law.
I found their website: http://www.el2000.ru
It's all in Russian. I attempted to figure it out using my very basic school russian. They seem to be a Sun distributor for Russia.
I think the new chip is described here. At least I think it is.
I can understand individual words but not much more than that. So if some one knows Russian. Please translate.
-Pelle
Very abbreviated translation of http://www.el2000.ru/company/comp.html
*
Elbrus is a spin-off company (actually tree companies) of Institute for Precise Mechanics and Computing Equipment (IPMCE), who designed supercomputers used for Soviet missile control, missile defence, nuclear research and other applications. R&D people are mostly from IPMCE.
...
From the history of IPMCE
1957-59 : M-40 - a computer on vacuum tubes, used in first successful missile-defence test.
1964-69 : 5E92B - 2-processor system on ICs, main component in first missile-defense system
of Moscow.
1973-79 : ELBRUS-1 - first SMP system on medium-integration ISc, 10-processor system.
1977-84 : ELBRUS-2 - SMP supercomputer, based on ECL (?) high-integration ICs, 10 processors,
used in Space Flight Control Center, in nuclear research centers, for military applications.
1985-94 : ELBRUS-3 - LSI, ECL high-integration ICs, 16 processors, 2 times performance of CRAY-YM. Wasn't put in production for financial reasons.
Some of IPMCE design achievements
1955 - high-speed arithmetics
1964 - fault-tolerant non-stop architecture with full hardware control (?)
1979 - SMP system with shared memory, 10 CPU
1979 - protected programming technology, hardware support for data types
1986 - an architecture with explicit parallelism
1986 - binary compilation technology
...
Novosibirsk and StPetersburg offices were formed in 1992, and work mostly on compilers and Java technology.
...
MCST (Moscow Center for SPARC Tecnologies) - a part of ELBRUS group, 400+ employees
...
Recent designs :
- a SPARC-compatible CPU for military applications
- universal high-performance system ELBRUS-90micro, works with 100+ different network and I/O-adapters.
Since 1992 MCST works with Sun Miscrosystem: design and support of compilers, operating systems, perspective workstation design, Java technologies.
Long-time partnership with Compass Design Automation B.V. (a VLSI Technology, Inc. company), a VLSI CAD specialist.
...
Currently MCST works on new E2K CPU design, using postRISC architecture with explicit parallelism and binary compilation technology, providing effective and reliable execution of Intelx86 binaries, with estimated performance of up to 2 times higher than Intel's P7 (Merced).
...
Key elements of the project are CPU architecture and a complementary "superoptimizing" compiler, which permits fullest usage of hardware parallelism.
Another important direction is improvements of binary compilation technology, which provides 100% reliability and high efficiency of binary compatibility modes for x86 and SPARC architectures.
...
MCST also works on number of other joint projects in hardware and software areas, with Transmeta and other companies.
***********************************************
So those guys certainly have know-how, experience and gray matter to design kick-ass computers.
Manufacturing is another story...
Sorry, but I've never worked for Transmeta.
Oppositely, some guys who are now at Transmeta worked closely with us some time ago.
Very clever guys they were and are! Luck to them!
Yes on necesity but it shouldn't be too surprising given the number of "minds" that create most of the profound technologies aren't that many and they tend to be a bit scattered across the globe.
I look forward to watching this one unfold.
Carl Forhan
Songbird Productions
http://songbird.atari.net