End In Sight For Alpha

minektur writes "news.com has an article stating that DEC ... I mean Compaq .... Uh, I mean HP has decided to EOL the once mighty Alpha architecture. Let's all take a moment of silence." I was lucky enough to have access to a 533 MHz Alpha back when the fastest Pentiums were only around 200 MHz, and the Alpha architecture earned a special place in my heart. It will be missed.

Comments

A true shame...

[kakos]

The Alpha was always one of the better processors. It was fast and powerful and way ahead of its time. It is a shame that a truly great processor was killed by the economy and mergers galore. It will be missed.

Re:A true shame...

[jericho4.0]

I'm gonna tag along on yr post, since it seems to be to the point.

The Alpha rocked. Nuff said. I'm sorry it's EOL. And I'm sorry there are so many posts who think that little intel boxen are so much better.

Alpha gets added to the list of failed, technicaly better products. The Amiga, Beta video, the Newton, etc.

Re:A true shame...

[sql*kitten]

The Alpha was always one of the better processors. It was fast and powerful and way ahead of its time. It is a shame that a truly great processor was killed by the economy and mergers galore. It will be missed.

What killed it was DEC, whose management were naive enough to believe that great products sell themselves and there's very little need for marketing. Unfortunately for them, engineers don't make purchasing decisions. VMS on Alpha 5 years ago was 10 years ahead of where Solaris on UltraSPARC is now - seriously, in terms of reliability and scalability. VMScluster was a joy to use, and the Alpha gave superb performance for anything involving floating point. They should have owned the high-end workstation market (along with SGI) if technology was all that mattered, but Sun were smart enough to spend lavishly on their marketing, and it paid off massively for them.

If it hadn't been for that, Compaq would never have bought DEC, and would instead be back competing against Dell where they belong. The management of DEC have a lot to answer for - technology and engineering cannot exist in a vaccuum despite what Slashbots think, it goes hand in hand with marketing and sales.

Re:A true shame...

[evilviper]

Alpha gets added to the list of failed, technicaly better products.

No kidding. I really can't imagine why it is being dropped. I'd think HP would keep it around just so IBM doesn't take over the top spots for supercomputers.

Right now, the Alpha is firmly holding spots 2, 3, 6, 7, 18, 47, 57, 58, 59, 63, 79, 109, 110, 117, 118, 144, 179, 217, 245, 246, 337, 340, and 355 on the list of the 500 fastest supercomputers.

Sure, they can replace those slower systems with their other systems, but what about the 4 Alphas in the top 10 spots? What does HP have that can rival them in performance, while still keeping the prices down? I'd say if they kept the Alpha, rather than their own processors, they'd have a chance at finally gaining ground on the hi-end Unix server market where IBM and Sun dominate.

But, there's always hope for Alpha fans. Intel bought the technology, so if their new 64-bit processor (which shatters compatibility anyhow) doesn't perform well enough, they could just start making Alphas and call them their own.

AFAIK, there's nothing stopping Samsung (or anyone else involved) from continuing to build Alpha processors... Maybe API will try to keep the Alpha alive. It's been a good product for them for some time.

Or perhaps some other party might pick up the torch. Sun would be a good candidate, since they're in a tight competition with IBM, and the Alpha seems to be the only thing to top IBM's Power3 (and is doing so with half the number of processors!!!).

Come on HP. The Alpha has just as loyal a following as Apple... It's a big mistake not to start improving it and seeing what it can really do for you.

Re:A true shame...

[Xner]

Also spots 39 and 40, since the Cray T3E is basically a very fast toroidal interconnect and Alpha processors.

Re:A true shame...

[leandrod]

> I really can't imagine why it is being dropped.

Because HP is already committed to IPF (IA-64, Itanium) and thinks it will become better than what Alpha could be. That's the official story. The unofficial is that they gave up competing on products a long time ago, and now just want to do services. Presumably that's a mixture of dumb MBAs who repeat the "services are key" mantra without understanding you must have a product to service, and if you didn't manufactured it, its manufacturer is more likely to service it than you; and of realising that people want Wintel or Lintel "just because they are used to it", and because of volume driving down prices.

In other words, the RISC market was too fragmented, and instead of coordinating the RISCs by phasing out PA-RISC and going all the way Alpha they decided to instead try to sell Intel on their VLIW, and thus IPF was born. To be fair, the first blunder was Digital's when they failed to win Apple and Novell as Alpha users, then consequently to make Alpha a volume architecture with several licensees, OEMs, foundries, notebook versions and all that you need to go heads on against a monopoly.

> I'd think HP would keep it around just so IBM doesn't take over the top spots for supercomputers.

First, they do believe in IPF, or so it seems.

Second, is being in the top 500 supercomputers list important at all? I guess they'd rather be lucrative. I think the way they chose to be lucrative is mistaken, but that's probably their rationale.

> I'd say if they kept the Alpha, rather than their own processors, they'd have a chance at finally gaining ground on the hi-end Unix server market where IBM and Sun dominate.

Actually the PA-RISC has a nice position. Telcos tend to use predominantly the SuperDomes, due to HP's relationship with Amdocs. PA-RISCs are actually nice systems, and HP builds some nice systems around them. HP-UX isn't GNU/Linux or Solaris, but still it's Unix, so you can't throw it away. Too bad for them that Unisys will sell IPF machines that will be as nice as HP's, and so will other vendors, and some of them will have GNU/Linux or Unix to run on them.

> But, there's always hope for Alpha fans.

There isn't, see below.

> Intel bought the technology, so if their new 64-bit processor (which shatters compatibility anyhow) doesn't perform well enough, they could just start making Alphas and call them their own.

I doubt. Intel bought the patents and the documents, but most engineers left. Intel has lousy employee relationship, so they wouldn't be able to reproduce the in-house expertise Digital, Silicon Graphics, HP (before merge) had and that IBM, Sun now have. Also, they are already forcing customers to change the architecture. Would they risk it again, knowing each change in architecture is a chance of jumping ship to someone else with a better story to tell, like IBM or Sun?

> AFAIK, there's nothing stopping Samsung (or anyone else involved) from continuing to build Alpha processors...

First, there is no one else involved, only Samsung.

Second, Samsung can't compete. It does not have neither the focus, nor the ISVs, nor the customers, nor the applications, nor the systems, nor nothing needed to compete. Sun & IBM do, HP, Digital and Silicon Graphics had.

> Maybe API will try to keep the Alpha alive. It's been a good product for them for some time.

I doubt. Technically yes, but where are the volumes, the customers, the profits? Anyway they already jumped ship. They are now SiPackets, former API Networks, selling the HyperTransport stuff to AMD and the like.

> Or perhaps some other party might pick up the torch.

Forget it. Licenses are not available for the asking, even if you had loads of money. And you would have to get the engineers, and find the customers. Do you think anyone would, after the .com bubble?

> Sun would be a good candidate, since they're in a tight competition with IBM, and the Alpha seems to be the only thing to top IBM's Power3

Sun has already stated SPARC for them is binary compatibility and a viable future, not performance only. Their going Alpha would hurt more than help. They hope to get UltraSPARC to be competitive with POWER and IPF, and that's it.

> The Alpha has just as loyal a following as Apple...

There is a difference. There was never MS Office running on the Alpha, only MS Word and Excel, and these are gone now. There was never an Alpha notebook. Alphas and Macs were never in the same price bracket.

Re:A true shame...

[binaryDigit]

Intel bought the technology, so if their new 64-bit processor (which shatters compatibility anyhow) doesn't perform well enough, they could just start making Alphas and call them their own.

I thought that Intel bought the manufacturing arm and licensed some patented technologies (whatever DEC sued Intel over to begin with) and the right to manufacture Alphas, but not the whole intellectual rights to the Alpha architecture itself (which Compaq got)? So while Intel could certainly churn out Alphas, they could only churn out existing versions and not create new ones?

Or perhaps some other party might pick up the torch. Sun would be a good candidate, since they're in a tight competition with IBM, and the Alpha seems to be the only thing to top IBM's Power3 (and is doing so with half the number of processors!!!).

Egads, Sun would never abandon Sparc. They have spent billions on just simply developing the name in the marketplace, and to suddenly switch gears and drop Sparc to sell Alpha would be suicide. Most people who purchase Sun don't do it because their stuff is faster than anyone elses (because in general they are not), they buy it for the stability of the hardware and OS. Sun has been able to thrive even they've always been in the role of the lessor performer. Note that there aren't too many Sun's in the Top500, Sun just isn't that interested in that market.

Come on HP. The Alpha has just as loyal a following as Apple... It's a big mistake not to start improving it and seeing what it can really do for you.

No, it's a big mistake to try to sell computers using three different architectures (four if you count the overlap between PA-RISC and Itanium). It makes no sense at all to keep Alpha around (as much as I like Alpha). They've already bought into Itanium and PARISC still has legs while they wait for Itanium to mature. Now they can surely integrate more concepts from Alpha into future chips, but Alpha as an independent entity has no useful purpose in the HP landscape.

Maybe Transmeta will buy the rights and finally get a little oomph into those chips of theirs.

Re:A true shame...

[dildatron]

You can add to your list the HP calculators that were recently disconinued. I have a special part in my heart for the HP 48G. Now, the whole HP calc division has been dropped. To anyone ho has uesd one - they were much better than what TI had out at the same time. Unfortunately TI got all the schools hooked on TIs by giving them to teachers, giving large rebates, etc. and all you need to do is hook acedemia and you're in.

Re:A true shame...

[jbolden]

By the 48G HP hadn't done much interesting in years. Go back a bit further to the 28s and 15c (12c business line) and the difference becomes profound. Move to the late 70s 34c and you have a programable calculator at a time when most pocket calcs thought that doing square roots was a high end feature.

Re:A true shame...

[evilviper]

To be fair, the first blunder was Digital's when they failed to win Apple and Novell as Alpha users, then consequently to make Alpha a volume architecture with several licensees, OEMs, foundries, notebook versions and all that you need to go heads on against a monopoly.

I know Apple is currently looking for an alternative to PPC that will allow them to compete with PCs. I don't consider it Digital's mistake not getting Apple onboard, but Apple's mistake rather. As for Novell, they can barely keep themselves afloat, let alone push the new users to Alpha.

Second, is being in the top 500 supercomputers list important at all? I guess they'd rather be lucrative.

You don't believe the highest end computer market is lucrative? I'd say NEC, Cray, IBM, and Sun would disagree completely.

Re:A true shame...

[binaryDigit]

IBM dosen't actually make the x86 based machines themselves, they just have other manufacture them, but even still, the pc market is easy because it doesn't require the huge R&D spending that the workstation/server market does. Power4/3/RS3/PPC et al are architecturally similar and for the most part support the same ISA (I know they are not exactly the same, but they are very close). IBM uses them interchangably in their RS/6000 and AS/400 lines. The Z series they've had for a gazillion years (well not the Z series itself, but the whole mainframe line from back in the 360 days). So IBM itself only supports three major architectures x86, Power/PPC, and Z. HP doesn't have the benefit of having an architecture like the mainframes that are huge cash cows and have been since the dawn of (computing) time. If they did support Alpha, it would compete directly with 2 of their existing architectures (Itanium and PARISC). IBM, for the most part, doesn't have that problem, all three major architectures are geared towards very different markets (again, for the most part).

But anyway, my comment was about HP specifically. Keep in mind that even with the triple merger (HP/Compaq/DEC) HP is still a "smaller" company (revenue, market cap, etc), so comparing them with just about anyone else is not an apples/apples (no pun intended) comparison.

Re:A true shame...

[slashdot_commentator]

Oh come on, TI hooking academia was the least reason for the demise of the HP. The biggest reason was cost. I could get a crummy TI or Casio for 1/4 the cost of an HP calculator. And even ignoring the pocketbook, having to deal with reverse-infix would have been enough to chase away average consumers.

Re:A true shame...

[leandrod]

> I know Apple is currently looking for an alternative to PPC that will allow them to compete with PCs.

First, CPU performance is important only for gamers and big servers, and neither is an important market for Apple. Second, most of performance these days is in building a balanced system, so Apple has to focus more on storage, memory, bus, video and the such. Third, the alternative for Apple may be IBM POWER4 or AMD Opteron or even UltraSPARC, but I cannot see Apple having enough clout to revive neither Alpha, nor MIPS, nor anything else.

> I don't consider it Digital's mistake not getting Apple onboard, but Apple's mistake rather.

And why not? Legend has that Apple wanted the Alpha badly, and Digital went so far as to plan a notebook version of Alpha around 21264 time, but much before that the deal was dead on water because DEC insisted on Apple building the next Mac OS on OpenVMS, instead of insisting on Taligent, Pink and Copeland. Which probably would have been a good thing as timing goes as each of these projects got cancelled and Mac OS X came too late, but the result would have been a more closed Mac OS then the current one. It would probably have saved the Alpha.

So whom to blame, Apple for refusing to see that OpenVMS was far better then what they had at the time, or Digital for being too wise for its own good?

> As for Novell, they can barely keep themselves afloat, let alone push the new users to Alpha.

You seem to be under the illusion that I am talking current events, but I am talking ancient History, around 1.99[0-2].

Actually this was the time when Novell was king of the (corporate office automation) PC networking, and wanted to get a scalable Netware. The stumbling block was also OpenVMS, which in retrospect would have been a good thing. Too bad Novell never got Caldera to fly, because that could have been a road to scalability, freedom and performance.

> You don't believe the highest end computer market is lucrative? I'd say NEC, Cray, IBM, and Sun would disagree completely.

I do not say I do not believe, but rather it ain't necessarily so. Cray has to do that, because it can do little else. The others can afford to do some stuff for PR reasons, but HP is in dire straits to justify this stupid blunder. It is putting in risk the cash cows by choosing to go along Intel instead of keeping PA-RISC, Alpha, Unix and its midrange systems as a focus. Itanium will give it no edge, but it may be forced to try to build some Itanium supercomputers to justify its decisions. Keeping the Alpha just for supercomputers is barely a proposition, specially when it is already killing lots of OpenVMS stuff.

Re:A true shame...

[leandrod]

> In my experience (primarily in Europe and Asia), Alphas dominate telcos

I should have qualified. AFAIK Amdocs dominates wireless all over the world as volumes go (biggest customers) and wireline in the US. Wireline is mainframe, wireless HP-UX. It might be that Alpha dominates wireline out of the US, and smaller wireless, or that I am wrong. My experience is with the billing system Amdocs, but I know that operations do use Alphas a lot.

Re:A true shame...

[Ben Hutchings]

I've heard from someone who worked with it that most of the management software for GSM networks runs on VMS, presumably on Alphas now.

Re:A true shame...

[leandrod]

Might well be. I once worked at a CDMA operator who had at least to Alphas, but they were running Digital Unix.

This might just be a good thing

[Raul654]

I think this goes to show that it's not just about building a better mousetrap. You have to build a better mousetrap and then show everyone that it's SO much better than what is out there that it is worth the transition costs. It's something they teach in engineering 101, and it's the same problem microsoft has been bumping into for years now (and basically arm-twisting everyone to upgrade)

Re:This might just be a good thing

[darkov]

I think this goes to show that it's not just about building a better mousetrap.

No, this is a case of money and influence over technology. Good technology. Bad politics. You could build a processor that executes instructions before they're fetched from memory and the Pentium would still be a best seller.

They're really nothing good about the death of the Alpha.

Re:This might just be a good thing

[Fesh]

I think that was his/her point... *chuckle*

Re:This might just be a good thing

[uncleFester]

No, this is a case of money and influence over technology.

No, I think this is a case of poor marketing. It's a superior product, so much so the competetion even bought into it.. but the product was saddled with two companies who couldn't market the product. Digital was notoriously poor in marketing, and when Compaq bought them it was merely a product-rounding move. Compaq, after all, made their money in Intel-based crap and Capellas never really pushed the Alpha as the strong superior product it was.

The Alpha was a decent hardware/OS setup: I ran a number of them at my last job, supporting boxen using Oracle. The boxes were solid computers (even for older 4100-series machines!), Tru64 was fairly solid (only a tricky NFS glitch on one machine spotted a perfect record with them) and the 1 1/2 years I spent with Dec/Compaq/Tru64 was suprisingly excellent. It's a shame the companies involved pretty much killed them due to stupidity.

Re:This might just be a good thing

[darkov]

I think that was his/her point... *chuckle*

Just becuase I'm a hermaphrodite it doesn't mean you can laugh at me.

Re:This might just be a good thing

[scaprallion]

I recall a conversation with a Digital rep a few years back where he claimed his company would've marketed fried chicken as "warm dead bird". I have to agree. The Alpha was a *great* architecture that never went anywhere due to lack of marketing.

Re:This might just be a good thing

[Fesh]

Fine, fine... Sheesh.

Its point.

Still lives within the EV6 AMD Athlon...

[SirDaShadow]

AMD sure had it right with its decision of making the Athlon Architecture based on the Alpha...to the point that it outlived the Alpha itself...

Re:Still lives within the EV6 AMD Athlon...

[io333]

Were they? Both the Alpha and the Athlon seem to have run into a speed wall whereas the apparent long term bumpability of the P4 is mind boggling.

Re:Still lives within the EV6 AMD Athlon...

[kc8apf]

The speed increases on the p4 is due to the use of a 22 stage pipeline. The Athlon and Alpha do not have nearly that long a pipeline and as such do not scale in Mhz as easily, but they get more work done per clock, hence why a slower Athlon is on par with a p4.

Re:Still lives within the EV6 AMD Athlon...

[Anonymous Coward]

AMD sure had it right with its decision of making the Athlon Architecture based on the Alpha...to the point that it outlived the Alpha itself...

They only licensed the bus architecture. Nothing else...

Re:Still lives within the EV6 AMD Athlon...

[Chasing Amy]

This is true, but hyperthreading seems to have great potential for fixing the weaknesses inherent in having such a long pipeline. Few apps have specifically been optimized for it yet, but even so it provides a small to large increase in productivity depending on how many threads you have going at once and how much each app is optimized for hyperthreading or dual processors. The benchmarks posted at places like Anandtech and Tom's Hardware demonstrate this, even at this early stage.

Add to that the fact that Intel is pushing for developers to compile using optimizations for hyperthreading and dual processors, and to make apps more multithreaded, and you get an even greater likelihood of performance increases in the future. The cost of that long pipeline is clearly being lowered, and P4 with hyperthreading can get more done per clock cycle than the P4 without.

I was one of the people who laughed at Intel when the P4 was released in its original incarnation, believing the Athlon's Alpha-like brute force would continue to trounce the comparatively puny NetBurst architecture at every turn. But in the end, the larger cache, faster FSB, and now Hyperthreading ability of the newer P4, seem to be adding up to be just as valuable as the P4's GHz scalability.

All I can say is, brute force doesn't seem to cut it any more. Intel is finally improving the little things, and not just clockspeed. The fact that next year Intel is planning to move to an 800MHz effective FSB with matching dual-channel 400MHz DDR memory just goes to show that. Who ever would have thought? :-)

Re:Still lives within the EV6 AMD Athlon...

[darkov]

Few apps have specifically been optimized for it yet...make apps more multithreaded

This is a reccurring story in the development of parrallelism. It would be great in any form if people just developed for it, but even multithreading is quite tricky to implement compared to a single stream execution environment. And in most apps you just can't get the fine-grained parrallelism that would yield really good speed improvements.

This is a software problem, and no amount of hardware will make a significant difference.

Re:Still lives within the EV6 AMD Athlon...

[jericho4.0]

I hate to agree with you, but the P4 _is_ looking very good right now...HT is just waiting to be expoited by software.

IBM's Power4's look good too, though. Long term trends will shift away from Intel solutions (IMHO).It remains to be seen what the better choice is. The market decides in the end, and that isn't always the best choice.

Re:Still lives within the EV6 AMD Athlon...

[Znork]

...except of course, the other execution units data is probably not in the L1 cache either. And that you now effectively have half the cache size since you're sharing it between two execution units.

There is some performance to be gained with hyperthreading but mainly when you have some math intensive code running on one unit and other stuff running on the other. The main gain will be in running things like seti@home in the background...

Re:Still lives within the EV6 AMD Athlon...

[Isle]

They only licensed the bus architecture. Nothing else...

And hired most of the alpha-architects that fled from when Compaq bought DEC, and let the plans for the 21364 inspire good parts of the Athlon internals.

Re:Still lives within the EV6 AMD Athlon...

[Lussarn]

He is talking about the EV6 bus. The CPU is different but uses similiar chipset. There are Alpha motherboards using AMD760 chipset that of course where made for Athlons.

Re:Still lives within the EV6 AMD Athlon...

[operagost]

Well, the EV4/5 was far faster than a Pentium at the same clock speed in FP and just as fast in integer performance, so I can't say they took the speed demon route. It was simply a better CPU. In fact, the Xeon now runs at a higher clock, but it's still spanked by slower Alphas (rightly so, they're still expensive).

Re:Still lives within the EV6 AMD Athlon...

[Glock27]

This is true, but hyperthreading seems to have great potential for fixing the weaknesses inherent in having such a long pipeline.

If I want SMP, I'll buy a multiway Hammer, thank you. ;-) Most likely I'll be able to get two Hammers for the cost of a high-end P4.

Few apps have specifically been optimized for it yet, but even so it provides a small to large increase in productivity depending on how many threads you have going at once and how much each app is optimized for hyperthreading or dual processors. The benchmarks posted at places like Anandtech and Tom's Hardware demonstrate this, even at this early stage.

The actual benefits of HT have yet to be realized in any actual application as far as I'm aware. We'll see how things sort out. One thing that isn't so great about HT is that applications that charge for multiple CPUs see multiple CPUs when it's enabled. Once again, I'd prefer to simply have a second "real" CPU.

I think AMD has a better approach. Keep the individual processors simpler (Hammer die size is smaller than P4 at 130 nm.), and enable up to 8-way SMP at a nominal cost. 64-bit memory addressing is a killer addition as well.

On a side note, is Intel inventing a new dialect of C++ on it's own? I've seen some pretty bizarre code samples in Intel ads lately...

Re:Still lives within the EV6 AMD Athlon...

[Glock27]

Or am I not seeing the point and you aren't talking about instruction set?

The way fast x86 chips have worked for awhile is to have a front end decode the x86 instructions into "micro-ops", then feed those micro-ops to many functional units on the back end. The back end is RISC-like, and amenable to RISC techniques. Many CPU engineers working on AMD products have come from the Alpha team, and AMD's CPUs have definitely benefited. Look at the floating point performance.

BTW, on the subject of floating point performance, I'd like more information on the Itanium SPEC results. Specifically, did Intel use the same optimization that boosted Sun's UltraSparc 3 scores? That was pretty close to cheating, if it didn't actually cross the line...

Re:Still lives within the EV6 AMD Athlon...

[drinkypoo]

Unfortunately hyperthreading seems to be giving only a small benefit today. If it were more than a gimmick then it would produce more of a performance bump on heavily loaded systems. Having multiple CPUs provides a benefit when you are running multiple processes, not just when running multiple threads.

Athlon will be replaced for all purposes by Hammer, under the Athlon 64 brand name, so the fact that it's seemingly running out of steam matters not at all. It will continue to get cheaper and cheaper until it is phased out while still providing comparable performance to a P4. Maybe it will end up being half the speed, but 1/8 the price; By that time Hammer will be starting to roll out.

Meanwhile as we all know the P4 has an extremely long pipeline for its generation which raises the cost of branch misprediction. Those high clock rates come at a high price. The greater parallelism of the Athlon's core when compared to a single P4 core (it remains to be seen how much real-world performance we will get from hyperthreading once more applications are optimized) and its shorter pipeline provide obvious benefits today. And let us not forget the overwhelming cost difference. If Intel keeps increasing clock rate as they have been then they may regain control of the high-end 32 bit desktop market but AMD will continue to own the low-end and I strongly suspect that unless Itanium 2 is dramatically cheaper than even half of what Itanium 1 costs by the time Hammer rolls out, that AMD will start a 64 bit desktop trend and Intel may never recover that market.

Re:Alphas

[CaptainCarrot]

Or VMS, perhaps?

RIP Alpha...

[pVoid]

The processor on which I discovered for the first time what super-scalar execution meant ...

Also the processor which for some odd reason doesn't support rotate (or was it shift?) operations =)... <sniff still>.

Maybe it's time I actually went out and bought one now, they'll soon be like vintage Cadillacs.

On another note, HP has got some major huevos man, making such a drastic shift in technology requires it.
I sure hope Itaniums happen. 256 integer registers makes me drool.... DROOOOOL

Re:RIP Alpha...

[Keith_Beef]

HP has got some major huevos man, making such a drastic shift in technology requires it

And HP is putting all those huevos in one basket...

Alpha is the Omega

[BJH]

Screw Itanium. An architecture that requires a highly-tuned compiler to optimize software well enough that it doesn't stutter from huge branch prediction misses is not a worthy successor to the Alpha. Excellent firmware, elegant architecture, good speed/MHz ratio...

I own three Alphas (a Personal Workstation 600, an Alphastation 255, and a homebuilt machine using a PC64+ motherboard), and they're great machines to use. I'm currently on the lookout for an ES40 - when I see one for below a couple of thousand dollars, I'm going to snap it up.

Re:Alpha is the Omega

[g4dget]

I agree with what you are saying. I think the Itanium will be very damaging to the evolution of software: it will be much harder to create new compilers for that architecture. C/C++, Java, and .NET will become even more dominant as fewer and fewer smaller compiler efforts can compete.

Re:Alpha is the Omega

[jbolden]

I don't think so. Intel already has a good compiler for Itanium and there are lots of language X -> C scripts which don't increase code execution times. Further I think there is strong evidence that Intel is going to make sure that GCC does a good job (which gets you about 20 languages). So I don't think the static languages will have much trouble.

At the same time Parrot is going excellently. In about a week a developer knocked off a quickbasic -> parrot environment. So parrot supports: Perl, Python, Java, Ruby, GWBasic, quickbasic, TCL and other already. Writting for the Parrot environment is easy and I could definitely see someone using a good compiler for the C -> Parrot work. So I don't think they dynamic languages will have much trouble.

Re:Alpha is the Omega

[g4dget]

I don't think so. Intel already has a good compiler for Itanium and there are lots of language X -> C scripts which don't increase code execution times.

X-to-C compilation is an efficient solution only for a very limited set of languages: C just doesn't make a good backend, and, furthermore, many languages require frequent runtime code generations, which is simply too expensive with an external C compiler.

Further I think there is strong evidence that Intel is going to make sure that GCC does a good job (which gets you about 20 languages).

Yes, and they all roughly have the same world view as C.

At the same time Parrot is going excellently.

Parrot is a byte code interpreter. Of course, you can write byte code interpreters in C and then compile them with a C compiler. That is not a very efficient use of processor resources: you'd be better off compiling natively to a much simpler processor than an Itanium, without the VLIW features. And that's my point: Itanium further and unnecessarily enshrines the C view of the world, putting any alternatives at a grave disadvantage. That's not the way to make progress in software.

Re:Alpha is the Omega

[jbolden]

X-to-C compilation is an efficient solution only for a very limited set of languages: C just doesn't make a good backend, and, furthermore, many languages require frequent runtime code generations, which is simply too expensive with an external C compiler.

Parrot is a byte code interpreter. Of course, you can write byte code interpreters in C and then compile them with a C compiler. That is not a very efficient use of processor resources: you'd be better off compiling natively to a much simpler processor than an Itanium, without the VLIW features. And that's my point: Itanium further and unnecessarily enshrines the C view of the world, putting any alternatives at a grave disadvantage. That's not the way to make progress in software.



I'm having some trouble picturing the kind of language you are talking about. Could you give an example? Obviously GCC is not for direct use on languages which don't require run time complication. OTOH Lex&Yacc seem to work those and GCC compiles the YACC. Or again Parrot.

Anyway could you give me some sort of example where you can't bootstrap from good C compilation?

haha Are we talking about Alphas or osx here?

[TheCouchPotatoFamine]

sure sounds the same

quality = more money, people. you are all free to drive Yugos if you please.

That aside, my father had an alpha for FreeBSD development, i believe. I used to play Larn on it. before i ever heard the word "final fantasy" Sorry to hear it got sucker punched. (Remember CEO's: If you're not number one, then buy number one and kill it. It's cheaper then investing in R&D.)

shucks.

Re:haha Are we talking about Alphas or osx here?

[jericho4.0]

You've never written a line of code in your life.

Maybe VIA will buy it off of HP!

[Newer Guy]

Then you'll see it for 29.95 on Pricewatch..and not need a fan. I can see it now.. the VIA/Cyrix Alpha DLC!

The day of a single very powerful CPU is over...

[Wolfier]

Distributed computing is going to be the trend...if I can stack together a few cheap chips to rival a single high performance chip, what would I do?

Given the exponential relationship of price to performance (i.e. a marginal performance increase will cost you a LOT more) associated with processors, I'll take the cheaper approach.

Granted, many apps don't fully use distributed processing power, but the ones that need most CPU probably do.

Alpha gone, but the best is yet to come.

[Anonymous Coward]

The Alpha chipset was fortold in the aincient scriptures, "I am the ALPHA" God mentioned more than once. This is where it gets interesting -- "And The OMEGA", God added. All of the core prophets prosthesized of Gods eventual metamorpheses to "OMEGA". It is widely believed now by computer industry analysts that they were referring to the coming release of the first Microsoft created CPU and chipset, named "OMEGA".
Bill Gates said at NorCON '02 that Microsofts products in the next five years would become the cornerstone of peoples mental and physical existences, again, he is referring to "OMEGA".

Or Linux...

[anonymous cupboard]

They ran Linux quite nicely. Many Digital engineers had input into Linux

Re:Or Linux...

[anonymous cupboard]

You are lucky, my biggest client uses OpenVMS, and although HPaQ wants to move them to Itanium, it will be a while before Itanium works reliably enough under VMS (or any other system at the moment).

Linux is fine and great but we need something that works like a VMS cluster to run things. In particular, we need the distributed lock manager and a cluster wide file system. HPaQ promises OpenVMS on Itanium, but that will take time. On the other hand, they are committed to supporting VMS for another 18 years or so because of DoD contracts that they inherited.

Alpha licensees

[velco]

IIRC, Samsung develops Alpha processors, I guess the rumours for Alpha's death are greatly exaggerrated.

Re:Alpha licensees

[Nexx]

IIRC, they quit doing so a while ago, as their Alpha stuff didn't sell.

I hate the 20s reply-send delay. It's highly annoying :p

Re:It's Technological Evolution

[pVoid]

Indeed.

In fact, this is much more than 'the circle of life' as in "lion eats elk, and then dies, and fertilizes grass of elk"...
No, HP is phasing out Alpha to brandish Itanium. So it's like HP's grandfather just died so now they can take care of their own children.

(do I make sense at 4 in the AM?)

Broadcast Flag

[denisonbigred]

Perhaps this is part of the "inter-industry conspiracy." And the alpha doesn't even have analog outputs... what will they kill next?

the most foolhard gamble ever?

[merc_sa]

the high end server market is notorious for being slow to adopt change. Now HP is trying to get rid
of not one but TWO legacy architecture in favor of the unproven itanium.. Though both PA-RISC and
Alpha were niche players, they were highly regarded in their market. Maybe I'm just being a cynic, but
somehow I got a feeling Carly is pulling a SGI and migrating to a platform because everybody ELSE
thought it was a good idea..

Though I'm a big Sun box fan, I still have to give the proper respect for those two well regarded
chips. RIP PA-RISC, Alpha..

Re:the most foolhard gamble ever?

[joib]

Looks like HP is changing from an engineering company to a bunch of vacuum cleaner salesmen, like say, Dell. They spun off their measurement systems stuff (Agilent), killing PA-RISC and alpha, what do they have left? Reselling Intel boxes and perhaps some consulting.

On the other hand, chip development costs seem to grow exponentially, so keeping on developing two high-end architectures for a very small market doesn't really make sense economically.

Re:the most foolhard gamble ever?

[am 2k]

Don't forget that they killed the calculator department. I really like my HP48GX, and I think it's far superior compared to the TI-series.

Re:the most foolhard gamble ever?

[uncleFester]

HP has been 'getting rid' of PA-RISC for ages. I remember attending an seminar on the virtues of Merced in the 1996-1997 timeperiod. At that time they planned phased out PA-RISC CPUs and were going to do PA emulation on Itanium by 2000 at the latest, to allow older HPUX installations to make a smooth transition.

In other words.. they'be been singing this song for at least 5 years and the Merced/cKinley delays have royally screwed their plans. They did have other plans on the horizon (though at the time I believe the roadmap only went to the 8500 or 8600 chips.. the 8700+ processors were probably a mad scramble when they realized it was going to be even longer.

Re:the most foolhard gamble ever?

[nutznboltz]

1. jilt customers
2. do it again
3 ???
4. (don't say it) profit!

(sorry)

Re:the most foolhard gamble ever?

[evilviper]

chip development costs seem to grow exponentially, so keeping on developing two high-end architectures for a very small market doesn't really make sense economically.

It sure does when that marke happens to be the highest paying market around. The Alpha holds the spots as the #2 & #3 fastest computers on earth, beating even IBM, while doing so with half the number of processors.

You might as well say Rolls Royce should start reselling Fords since their own cars are only a Niche market.

Re:the most foolhard gamble ever?

[sql*kitten]

HP has been 'getting rid' of PA-RISC for ages. I remember attending an seminar on the virtues of Merced in the 1996-1997 timeperiod. At that time they planned phased out PA-RISC CPUs and were going to do PA emulation on Itanium by 2000 at the latest, to allow older HPUX installations to make a smooth transition.

They're not alone there. The same indecision and delay cost SGI dearly because it diverted resources from the R14000 team. In the consumer space there's only one real rival to Itanium and that's the PowerPC, and even that is only a blip on Intel's radar. Ultimately it's about software; superior processors aren't enough on their own, there must also be a critical mass of software to run on them. Until there is, MIPS, SPARC, Alpha, PowerPC (even in the Mac) et al will only sell into niche markets.

Re:the most foolhard gamble ever?

[drinkypoo]

I don't understand how Sledgehammer isn't a real rival to Itanium; An efficient 64 bit processor which can be treated like a 32 bit processor if you so desire, which can run in a 32-bit equivalent of "real" mode in which it looks EXACTLY like a 32 bit chip, can switch back and forth between 32 and 64, or run purely 64 -- just like the 32 bit chips doing both 16 and 32 bit modes. Incidentally for all those who think that this is a bad idea, it's the OS that screws up flipping back and forth between real and protected mode (Which is to say, all windows 9x derivatives except windows ME) and not the CPU itself. It's a sound idea at the CPU level but the OS might fuck up, of course.

Re:the most foolhard gamble ever?

[MyHair]

... what do they have left? Reselling Intel boxes and perhaps some consulting.

P R I N T E R S .

And scanners, digital cameras, PDAs, etc..

Dropping two proven workstation/server architectures in favor of an unproven processor from a consumer-grade processor manufacturer doesn't sound too wise to me, but HP has advantages in the PC market.

When I was out of work and desperate for money I took a job selling computers retail at a large office store. I was surprised at how some people wanted their PC, scanner, camera, PDA, printer and other accessories to have the same brand name on them. They really think it helps them all work together, and they may be right: one company to call if you're having trouble printing your digital photo through your PC and printer. That means a lot to a non-geek.

And I presume the PC add-ons have higher profit margins that the PCs themselves.

Re:the most foolhard gamble ever?

[slashdot_commentator]

Ultimately it's about software; superior processors aren't enough on their own, there must also be a critical mass of software to run on them.

Which could have been mitigated by the use of Linux and open source software. Alphas were meant for the higher-end server market, not to run Windoze for the enduser consumer. Server software at that time was either proprietary or public domain software (a tiny percentage). The point being that DEC/Alpha was running in the same commercial software environment as everyone else.

Until there is, MIPS, SPARC, Alpha, PowerPC (even in the Mac) et al will only sell into niche markets.

The 64bit CPU arrived 5 years before the marketeers were hyping 64bit computing. What should have happened was that DEC should have pushed Alpha for the server market, and forced Intel to be penetrating Alpha's marketshare today. This could have even been accomplished by Compaq, who was looking to establish a backoffice presence with servers running NT. Alpha had years of production while Itanium was still bumbling in development.

The lesson here is Carpe Diem

Stil going strong.

[global_diffusion]

I was lucky enough to have access to a 533 MHz Alpha

That's funny. I'm still using two of those (dual procs) to run calculations. They really do rock.

Alpha?

[JanusFury]

Alpha's dead? I thought BSD was dead.

Re:In Soviet Russia...

[Zork the Almighty]

I see. So in Soviet Russia your joke would be funny also, yes ?

Re:Alphas

[kc8apf]

Which is now known as Tru64. Of course, depending on what manual page you look at it, the name varies from OSF/1 to Digital UNIX to Tru64.

Unbelievable

[jsse]

Imagine telling this to a geek just back from a 15-year coma:

"HP finally decides to retire Alpha..."
"HP bought Alpha?!"
"Yeah, after they merged with Compaq, which bought DEC..."
"Compaq bought DEC????!!!!!"

Re:Unbelievable

[SecretAsianMan]

Good idea. I'll use the diversion to steal his VAXen. I've been looking for a good heating system for this winter.

Re:Unbelievable

[RajivSLK]

A geek who who awoke from a 15 year comma would probably be put in a lab and studied.

"Dr. you say this geek has no knowledge of 'Slashdot'? Truely amazing..."

They would probably put throw him into a basement in the depths of IBM to write legacy code for 15 year old applications. Unaware of recent advents such as 'the web', 'slashdot' and 'massive internet pr0n archives' he would be the most productive geek ever.

Re:Unbelievable

[jsse]

The first thing he'd ask is to get a telephone line to dial up to the BBS he used to hang around to see if his ranking is still 'VIP - 1 hour time limit, 1 MB download limit' even he hasn't uploaded anything for the past 15 years.

Hopefully he could find his oldmates in the chatroom.

Re:Unbelievable

[TummyX]

Great. You'd put him into another 15-year coma.

Re:Unbelievable

[_Spirit]

I know spelling can be bad around here but a 15 year COMMA ? Hell I would have trouble making a 15 year point.

Re:Unbelievable

[operagost]

What's sad is all this happened relatively recently. That geek would only need to be in a five year coma. I'm waiting for the REALLY big fish to gobble up HP. Maybe Microsoft will stop dabbling and finally make its bold move for world dominance.

Re:Unbelievable

[tcc]

> Imagine telling this to a geek just back from a 15-year coma:

This scenario is total science fiction.

In the real world, you'd ask him how much MSFT shares he has, offer him double that price and .... profit! :)

Re:The day of a single very powerful CPU is over..

[MalleusEBHC]

Granted, many apps don't fully use distributed processing power, but the ones that need most CPU probably do.

Even after considering the extra money spent to develop an app to scale well with parallel processing, the savings from using multiple "cheap" processors compared to one expensive high performance processor will still make it worth it. Not only that, but then you have an app that you can scale up as needed (assuming it was designed well) without having to purchase a whole new set of hardware, but rather just by adding a couple more processors to your current cluster.

OpenVMS for x86?

[SecretAsianMan]

Does this mean they're going to port OpenVMS to x86 or something else, or are they going to EOL that, too? I think I had heard some rumors somewhere of a VMS/x86 port.

Re:OpenVMS for x86?

[Anonymous Coward]

Does this mean they're going to port OpenVMS to x86 or something else, or are they going to EOL that, too?

VMS Engineers are well along in the port to Itaniac. The port will be relatively easy because of the portability built into VMS during the VAX->Alpha port and the fact that the memory management in Itaniac looks a lot like the VAX. This according to a VMS Engineer who spoke at a conference I attended in Nashua NH (where the VMS engineers are).

I think I had heard some rumors somewhere of a VMS/x86 port.

Won't this silly rumor ever die????

It doesn't seem that long ago...

[DCowern]

Wow... I remember the GIGANTIC PC Computing headline circa mid-1997 proclaiming "Windows at 500Mhz". It seemed so earth-shattering back then... half a gigahertz. :-P

Maybe now that DEC/Compaq/HP is EOLing them, we'll see some really cheap ones start popping up on eBay once PHBs decide they don't want unsupported boxen. I wouldn't mind adding one of these to my collection. Would make a pretty nice linux workstation.

Re:And then there was one

[tqft]

what about transmeta? But rather than running in (intel) emulation mode, how about some of you (or maybe me when I can afford the time and money) try and use the VLIW concept to the full. The idea of getting smallish code blocks to execute as single instructions must have some appeal to some fo you speed freaks.

you know -- the current generation still rule

[cballowe]

I know they had the megahertz win in the early days against intel, but man I love my alphas. The architecture of the entire system that DEC built around them is just so nice. Maybe it's because I haven't played with any E10K or better hardware from Sun or whatever HP has to compete on that level, but I wouldn't give up my Compaq GS series hardware for anything. Not to mention Tru64 and TruCluster -- I swear Tru64 has the best man pages of any Unix - free or commercial. I often find myself going to the Tru64 pages for info on various standard syscalls.

I don't think it's a wise move for HP -- I wish Compaq had known what they were buying when they bought Digital.

Re:you know -- the current generation still rule

[supersnail]

If alphas were so nice how come nobody bought them!

Reason one. Loosey IO bandwith on alpha hardware. This made them useless for standard commercial processing or database work and confined them in the scientific number crunching niche at a time when research money was tight.

Reason two. Loosey operating systems. ULTRIX was immature and buggy, VMS was VMS (some people love it, not sure why), and NT ......

Reason three. It was a memory hog. With a "int" set to 64 bits as standard and each machine instruction taking up a lot of room (256 bits I think) you needed tons of real memory to run "Hello World", remember memory was expensive then.

Re:you know -- the current generation still rule

[pesc]

Loosey operating systems. ULTRIX was immature and buggy

AFAIK, Alphas never ran Ultrix. They ran OSF/1 (later renamed Tru64). The first true 64-bit OS released 10 years ago which still beat newer attempts in good, clean 64-bit design (including, for example, HP-UX).

VMS was VMS (some people love it, not sure why)
They run it because of the reliability and clustering capabilities. Which VMS had 15 years ago and no UNIX yet has emulated...

It was a memory hog. With a "int" set to 64 bits as standard
No, an int was 32 bits. A long (and void*) were 64 bits, just as it SHOULD be.

See LP64

and each machine instruction taking up a lot of room (256 bits I think)
Have you ever used an Alpha? Each instruction is 32 bits long.

Re:you know -- the current generation still rule

[Anonymous Coward]

Memory bandwidth varied, depending on the model. There were some with decent bandwidth, if you paid for it. Just like if you pay for it with, say, Sun. The cheaper Alphas were pretty badly crippled with both a slow bus and small L2, though...

Ultrix was also MIPS. And only MIPS. It ran on the DEC RISCStations, etc. but was never ported to the Alpha. Unix on the Alpha was OSF/1, Digital Unix or Tru64, depending on week, direction of wind, etc. And it's very different from the MACH-based Tru64/whatever-its-called-this-week.

Ultrix itself was an odd mish-mash of SysV and BSD. It wasn't actually all that bad.

As for VMS, people love it because it's rock-solid, very secure and has excellent documentation both online and in dead-tree format. It makes unix look like a toy OS in many respects, quite frankly. And I don't particularly like it - too bloody verbose - but I can certainly see why it's popular in certain circles.

As for the instruction size, it was fixed at 32bits. A far cry from 256 bits. Int being 64bits makes sense on a 64bit processor. And it's not like you can't define your own types if it really bothers you...

As for nobody buying them... um... there's a hell of a lot of them out there. I own one. I've used many. They were just another victim of the desktop PC getting powerful enough to handle work that until then had been the realm of very expensive workstations. So were SGI. So were Sun, to a lesser degree.

Digital bet their shirt on 64bit computing being their big selling point, but only those who actually needed it were willing to pay the premiums. And very few need it.

Re:you know -- the current generation still rule

[johnalex]

Sorry, but people didn't buy Alphas for those reasons. They didn't buy Alphas because DEC couldn't sell the devil a glass of ice water.

We still have a DEC Alpha 2100 here. Bought it in 1994. Aside from hardware maintenance, the thing never goes down. We've upgraded the hard drives and the memory, but otherwise, we've left it alone. It accomplishes the job we want it to do.

OpenVMS is stable. You can't break it. When DEC built something, they built it to run pretty near forever. That may have been their problem: no built-in obsolescence.

I'll say about the Alpha what I said a while back about Unix to a bunch of pseudo-geeks on a credit union list: If you've never used it, you wouldn't understand, and we can't explain it to you. If you want reliability, you use something that can do the job. Alphas and OpenVMS did the job.

Of course, now maybe our data processing company will get off the stick and at least do some research about offering a Unix-based platform.

Re:you know -- the current generation still rule

[Scott Wood]

Actually, instructions on Alpha take only 32 bits each, not 64 or 256.

Tell me this

[Raul654]

Ok, I've been wanting to ask this question on slashdot for sometime, and this is as good a time as any. Just how much of the Joe sixpack user's stuff is parallelize-able? Can you effeciently farm out quake/UT/Morrowind/Maple/Matlab/etc to multiple processors? (ok, maple and matlab may not be what joe user wants, but you get the idea)

Re:Tell me this

[Natalie's Hot Grits]

"Can you effeciently farm out quake/UT/Morrowind/Maple/Matlab/etc to multiple processors?"

The answer is yes... sortof...

Quake/UT are 3d First Person Shooter games. This means that as far as the client side goes (Joe sixpack's computer) there is no immediate advantage. Most processing power is done on rendering in real time, which can't be done on more than one host (however, its still possible to do it on more than one video card, or multiple CPU systems in SMP mode).

However, on the Quake/UT server side, everything changes. Most of the multiplayer games are limited by 1) network bandwidth and 2) cpu power when it comes to scaling (limits usually around ~32 users). Battlefield 1942 (kindof like RTCW) is an excellent example of a game that could dramatically imporve by distributed computing. For one thing, that game's server is dramatically CPU limited. You can't get more than ~32 clients(out of a 64 person max) out of a server with acceptable results, even on a 100Mbit switched LAN. However, if you could distribute out the server, have maybe 2, 3, 20 servers, each in charge of 1/2, 1/3, 1/20th of the users in the game (sortof like IRC chat) then you could successfully scale to any # of users without having any CPU scaling issues. I think everquest might have a hint of this type of technology.

I don't know the answer about things like Maple or Matlab, but I'm pretty sure they could at least take advantage of distributed computing a little. (again, depends on how you use it)

Re:Tell me this

[joib]

I'd say a lot of joe sixpack's software can be parallelized to some extent. But as very few consumers have multi-cpu computers, and win9x and xp home don't support SMP, the extra hassle of writing multithreaded programs is just not worth it. Perhaps when hyperthreading becomes more common, developers will start to use multiple threads. Perhaps xp home support it, I don't know.

Re:Tell me this

[vofka]

win9x and xp home don't support SMP

Windows XP Home does not support more than one Physical CPU, however it does support Hyperthreading (ie. more than one logical CPU). Though this is not true SMP, using SMP techniques for coding and compiling applications can yield performance increases on HT capable CPU's - and since the latest desktop P4 has HT enabled, Joe Public will soon be able to take advantage of Multiprocessor aware apps, even on XP Home.

Re:Tell me this

[fuzzybunny]

Oh ok this is a great idea--

Parallelize Joe Sixpack's mundane everyday computing tasks.

That way IE can format every hard drive in the company when it catches a malicious activeX control!

It will make your desktop support monkey's job more efficient and enjoyable, because every PC in a given environment will have the same crash at the same time every time!

:-)

Re:Tell me this

[sql*kitten]

I'd say a lot of joe sixpack's software can be parallelized to some extent. But as very few consumers have multi-cpu computers, and win9x and xp home don't support SMP, the extra hassle of writing multithreaded programs is just not worth it. Perhaps when hyperthreading becomes more common, developers will start to use multiple threads. Perhaps xp home support it, I don't know.


Not quite true, there are tangible benefits to multithreading even on single-CPU machines. For example, your web browser can be downloading the next page while you scroll around the page that you're currently on. Word can spellcheck your document asynchonously in the background while you are typing. Most Java applications actually delegate redrawing the screen to a thread while their main processing continues in the foreground.

With these types of applications, the work of implementing threads has to be done anyway, for user-experience reasons. Therefore, you can take advantage of multiple CPUs at no extra development cost.

Re:Alphas

[SecretAsianMan]

the name varies from OSF/1 to Digital UNIX to Tru64

And I still wonder why they dropped ULTRIX way back when. That's gotta be the coolest OS name ever.

Tru64 or HP-UX on Itanium?

[ragnarok]

Are they going to produce one of their Unixes for Itanium or try to convice people to move to Windows? Some of us have to produce software that actually works (ie industrial and civil control software) and NT just ain't gonna cut it.

Well, I guess we can all move to Linux or Solaris.

Re:Tru64 or HP-UX on Itanium?

[sys$manager]

HP-UX already runs on Itanium, they sell IA-64 HPUX boxes.

Re:Tru64 or HP-UX on Itanium?

[jaavaaguru]

Both Alpha and PA-RISC were fast and powerful - I thought that PA-RISC was quicker than a similarly spec'd Sun.

Unfortunately it had a poor excuse for an mmap implementation which made it useless for a lot of multi-user server applications (Apache being an example that springs to mind).

Hopefully HP will port HPUX to Itanium now, and we might have a Unix distro from HP that actually works.

My XL 300

[IcarusMoth]

I'm going to go and have a nice cry with my XLT 300, after finally getting linux to run on it. I love my Alpha I got mine used from a company called Great lakes computing, I know they still sell them. I got mine for $750 like 5 years ago, and with Debian on it, It Rocks Hard!! Now If you must excuse me, I need to get some kleenex "OH Hal, I have some bad news"

Re:The day of a single very powerful CPU is over..

[anonymous cupboard]

Um, the Alpha was ideal for that. Dick Sites, the main architect used to work for Cray and he used a lot of ideas to make Alphas work well together. You want 16 processors, it will do it and do it well.

Re:Alphas

[MavEtJu]

if they could run something other than NT/AIX

*cough* freebsd *cough*

Re:does it matter any more ?

[LinuxGeek]

Wow, you have P4 Xeons that are 64-bit processors and can support 20gigs+ of memory? No? Then how can you call them more powerful than an EV6 Alpha?

I will miss the Alphas, they were doomed the moment that Intel got control of them.

Alpha and Linux

[Koos Baster]

So, why do they favour Itanium over proven hardware like PA-RISC and Alpha? Simple: software. Alpha's epitaph was written the day Microsoft decided to stop NT development for Alpha's. Without commercial software, what's good hardware worth?

My answer would be: A LOT! A few years ago I bought a (once expensive) 266MHz Alpha for about $300, without any software. It took a while to get Red Hat 6 running, but the machine really rocks! As most of us know, Linux per se does *not* require x86 hardware. I guess you could even go through the trouble of getting Wine to run Win32 binaries under Bochs, if performance is not your primary issue. However, in my daily usage I hardly ever need anything outside Linux. In those cases - when someone sends me a Word document - I use and old Toshiba laptop, running Mandrake.

So why is x86 hardware the de facto standard Linux hardware? Good: price/performance ratio. Why is x86 relatively cheap? Large sales volumes. Why so? Windows won't run on anything else. Why do people buy Windows? Because everyone does.

It's just the everlasting circle that won't be broken anytime soon. Not by better hardware (Alpha, PowerPC, MIPS, StrongARM) and not by better software (Linux, BSDs other Unices). It's so depressing...

--
Programming is like sex... make one mistake, and support it the rest of your life

Re:Alpha and Linux

[jericho4.0]

I have Mips, Alpha, and Sparc hardware being extemely useful. These platforms are so much better than intel hardware it's not even funny.

Get 'em while the're cheap, kids!

Re:Alpha and Linux

[jaavaaguru]

It's just the everlasting circle that won't be broken anytime soon

One word: Palladium.

Once MS start forcing people to use only MS-approved software, prevent you ripping CDs and copying MP3s, force you to use even more and more MS proprietary formats and extortionate licensing models, how many people do you think will still like to use MS Software?

I rate software on cost, reliability and useability. MS's current stuff doesn't rate too highly on any of the above.

Re:Alpha and Linux

[captaineo]

Don't forget Digital's groundbreaking FX32 emulation software. FX32 allowed you to run x86 Windows software on Alpha NT. The cool part was that after running for a while in slow software emulation, FX32 went back through the x86 code and translated the hotspots into native Alpha code. It was not unusual for translated x86 applications to run faster under FX32 than they did on contemporary x86 machines!

(FX32 was a major help for the visual effects industry, where Alphas were really popular at one point - 3D apps like Lightwave were available in native Alpha versions, but all the support tools, like Photoshop, were not...)

Re:Alpha and Linux

[Alomex]

Most *nix systems have used for enterprise class systems for well over 20 years

Not true. I worked for large manufacturer which sold unix boxen and big iron and in the late 80s and we wouldn't even dream of selling a unix machine to anybody save the engineering department of a corporation.

Re:Alpha and Linux

[Alomex]

Still, technically speaking, VHS was inferior to the other two.

This is a myth. VHS was able to record a movie in a single tape. In at lest this aspect VHS was superior to betamax.

Re:Alpha and Linux

[operagost]

WTF... Alpha was for minis. Desktop and workstation use was important (see the killer FP unit) but still secondary. Since the enterprise shiznit like Oracle was ported and available as soon as Alpha appeared, software wasn't an object. FX32 was available to run x86 stuff on NT Alpha, but it was discontinued because NO ONE GAVE A CRAP. People just didn't want (relatively) cheapass, weak Windows NT on their $10,000+ Alpha.

It's really all about marketing. The line "PHBs want Windows because they run it on their eMachine at home" is just a fable. There's certainly dunderheads like that about, but they're not a significant market force. Most respond to marketing targeted at the enterprise, not the Family PC reader. You can't run Windows XP or Office XP on an IBM iSeries, but they still sell because IBM's marketing can kick ass.

Re:Alpha and Linux

[Chris Mattern]

> Once MS start forcing people to use only MS-approved software, prevent you ripping CDs and
> copying MP3s, force you to use even more and more MS proprietary formats and extortionate licensing
> models, how many people do you think will still like to use MS Software?

If Palladium does start alienating users (I suspect it will, but the jury's still out on that), MS will drop it like a hot rock. Gates' amazing ability to turn MS on a dime when his monopoly is threatened is the main reason he still has a monopoly (See: Internet).

Chris Mattern

Re:Alpha and Linux

[TheAncientHacker]

They also didn't want anything but Intel on their $5,000 Windows NT desktops, either.

Remember that Windows NT was available on:

• x86
• Alpha
• PowerPC
• MIPS
• Clipper

and even with Microsoft investing large amounts of money and effort(Windows NT was written on MIPS and ported to x86 to make sure no x86isms crept in) in making the multiple platforms available NOBODY bought anything but the x86 versions in any numbers.

Re:Alpha and Linux

[jbolden]

It's just the everlasting circle that won't be broken anytime soon.

Wordperfect, Lotus 1-2-3, Wordstar, dB4, Havard Graphics, Borland Turbo Pascal, Netscape.

Computer monopolies can fall fast, the cycle can be broken. Microsoft and Intel are extremely competitive putting market share above profits and offering an excellent value proposition. Most of their compitors died because they got greedy or they got lazy.

Re:Alpha and Linux

[Jason Earl]

If Palladium doesn't fly (and personally I don't think it will), then Microsoft is screwed. Microsoft has 99% of a market that is done growing, it has an 85% profit margin on its most popular products, and it's stock price is still configured for a very hefty growth rate.

Quite frankly, it ain't gonna happen without Palladium.

Even if Microsoft maintains its current revenues eventually Wall Street is going to want to lower their price to earnings ratio. Microsoft stock will drop to one fourth its current price and billions of dollars of value will go up in smoke. If you think that the $40 billion Microsoft has in the bank is a big deal, you should take a look at the cash value of Bill Gate's share of Microsoft stock. Microsoft has to grow, and selling an operating system and an office suite simply isn't going to get them there. Being the middle man in every entertainment transaction just might.

And that's not even taking into consideration the fact that OpenOffice and Linux are undercutting the value of Microsoft's current cash cows. Microsoft isn't going to back down on Palladium, they can't afford to.

Re:Alpha and Linux

[nathanm]

Still, technically speaking, VHS was inferior to the other two.

This is a myth. VHS was able to record a movie in a single tape. In at lest this aspect VHS was superior to betamax.

Notice the other parent poster said technically speaking? Ability to record on one tape is irrelevent, and you have to reduce the quality to record more on the tape!

Re:Alpha and Linux

[Alomex]

Notice the other parent poster said technically speaking?

I would consider capacity a technical characteristic, as was the decision to sacrifice recording quality in exchange for capacity in minutes.

They'll Live to Regret This

[turgid]

HP will live to regret this. They're retiring a mature, stable, established and best-of-breed architecture (Alpha) for an unproven, late, incompatible, expensive, clumsy one (itanium). Their competitors must be laughing all the way to the bank. Just what is HP doing? Why do large corporations make such crazy decisions?

Re:They'll Live to Regret This

[Large Green Mallard]

HP live to regret this?

Maybe. But then again... maybe not :)

Re:They'll Live to Regret This

[Alomex]

They're retiring a mature, stable, established and best-of-breed architecture (Alpha)

That nobody uses or cares about...

for an unproven, late, incompatible, expensive, clumsy one (itanium).

Late? yes.

Incompatible? With what? IBM 360s?

Expensive? Not any more than the alpha when it came out.

Clumsy? On the contrary, the Itanium design is top notch.

Re:They'll Live to Regret This

[kin_korn_karn]

you said best-of-breed. That invalidates your opinion completely.

Re:They'll Live to Regret This

[turgid]

You're right. With a bit of luck they'll wither away to become just another clone maker :-)

Re:future

[Alan Partridge]

most likely the Clawhammer / Sledgehammer family - the world moves on, and so do the people who designed the Alpha. It would be good to see govt. earning their keep by actually trying to stop monopolies forming in important industries - diversity = choice, and choice is good. I never used a product that contained an Alpha CPU, but I've never bought an x86 based machine either...

Re:Alphas

[jericho4.0]

I'm just sorry that someone modded you funny when you weren't joking.....

Epidemics

[Ektanoor]

It seems that HP caught compaqague. A terrible and deadly virus that destroys all the brain zones related to innovation, risk, calculation and self-estimation. Earlier we saw several companies being caught by this epidemics, the most notable, DEC, where the virus spread with such furor that in a question of months a once well-known company turned into another department in the corner of the company.

The fact that HP dropped a lot of its support for open source, closed the production of the Alpha architecture and seems to scale down other important sectors are a clear show that the desease got deeply into the corporation ranks. Soon we probably will see turning from blue to red and naming itself Compaq.

Depressing is right. . .

[Chaset]

Depressing is right. Can you imagine what kind of chips we'd be seeing if Intel's ungodly amount of financial and engineering resources were being poured into something like the Alpha rather than kludging and hacking the x86 generation after generation?

It is one of my peeves that CPU architectural superiority means little in a world where x86 is the "default", and the negative feedback loop (Intel is cheap -> people buy it -> Intel is cheaper) seems to have no end in sight.

That and the fact that Intel can use its x86 cash cow to keep funding the Itanium whether or not it has any real merit. Not saying that it doesn't (EPIC IS a cool idea), but in a level playing field, do you think they can get away with just throwing so many transistors at the problem?

As various promising architectures die off (Alpha, PA-RISC, who's next? POWER?), in the end, was the computing community better served by the dominance of one architecture designed for the lowest common denominator? It's all speculation, sure, but I think not.

A shame

[bryan1945]

the Beta never got released...

Re:And then there was one

[Hoser McMoose]

All well and good in concept, except for that fact that there isn't a single operating system, or even a compiler in existance for Transmeta's native VLIW mode. Actually, for that matter, there isn't even any documentation available for it's VLIW instructions, so even assembly or raw byte code is out.

Face it, the Transmeta Crusoe IS an x86 chip, it just uses a really odd-ball way of getting around any Intel patents. Also, with Intel's Banias processor on the near horizon, Transmeta's days are severely numbered IMO.

Re:does it matter any more ?

[master_p]

You are over reacting. 64-bit computing is useful to a very small percentage of the computing world, and those who need it are already have one. Furthermore, the applications that need more power than current top-of-the-line processors are well served from multiprocessor systems, clusters, etc. For example, scientific applications like nuclear reaction simulation, fluid dynamics computation and other similar things run in super computers. The Alpha CPU is a capable chip, do doubt, but it does not make a big difference in those applications that need such a big horsepower.

You are referring to advantages of the Alpha like SMT and 64-bit addressing. Xeons and Pentiums 4 HT already have SMT (hyper threading), and the Hammer from AMD is in the right track for a commercially viable 64-bit x86 chip.

There are also other 64-bit chips, like the Ultrasparc. Personally, as a user, small office/home office and developer of defense applications certainly don't need 64-bits.

I don't really see any advantage of the Alpha on the market. Sure, it is a good chip, and a technological marvel, but a dead one. In other words, it is not marketable any more(i.e., no one really needs it).

Open the specifications and all information

[xirtam_work]

Now is the time for HP to show that it's not just going to waste all the time and effort everyone put into the Alpha.

The now have the opporunitity to publish all information regarding the Alpha to the community so that anyone who wants to can continue to provide support. Or, if anyone wanted to, produce their own Alpha based chips.

By allowing the continued use of the Alpha they could extend the life of these systems instead of killing them off in favour of newer systems. I know that they probably will not want to, but hey, it's a nice guesture to make.

I seem to recall something about "open" processors before, such as an open sparc or something, so this wouldn't be the first time it was done, just the first time that a big corporation allowed it to be done with their 'redundant' interlectual property. I also think that this would be good for preservation purposes and to have more information about micro-processors of our era for future generations. Just look at the mess that NASA have been in before when older components obsolete.

CPU clock speed...

[Anonymous Coward]

What does a 533 MHz Alpha have to do with a 200 MHz Pentium? You are comparing clock speed of totally different architectures. What one can do in one clock step can be much more (or much less) than what the other can. I'm not saying that Pentium are, or where, faster than Alphas. I've heard that they were really good but, you can't compare CPU performance just by comparing the clock speed. Actually I have a couple of questions. The Alphas are RISC CPUs so, what clock speed do they have to run at to top the Pentium at 3GHz? Will it "fry" the computer?

Re: CPU clock speed...

[brokeninside]

It's ironic that you mention this, but if you look at SPEC scores, the 500 and 533 Mhz Alpha chips have the highest SPEC/Mhz ration of any CPU for which results have been submitted to SPEC.

Look at the archived results on the SPEC website. (You'll have to do the arithmetic yourself, they only provide scores, not scores/Mhz.)

Re:CPU clock speed...

[nathanm]

The Alphas are RISC CPUs so, what clock speed
do they have to run at to top the Pentium at 3GHz?

According to the latest SPEC benchmarks, to beat a 3060 MHz P4 in floating point, an Alpha would need to run at 1340 MHz, not much higher than the current 1250 MHz Alpha. For integer performance, it would need to run at about 1600 MHz.

However, Intel's own Itanium gets almost 1.5x the floating point performance at 1/3 the clock speed. Also, the PIII would outperform the P4 in integer ops at equal speed and come close in floating point.

Here's the data:

SPECfp 2000

CPU MHz base peak
AthlonXP(2800) 2000 782 843 ASUS A7N8X
P4 3060 1092 1103 Dell Precision Workstation 350
PIII 1000 329 340 Dell Precision Workstation 420
Xeon 2800 878 887 Dell Precision Workstation 530
Sparc64 1350 1004 1241 Fujitsu PrimePower 900
Alpha 1250 1019 1365 hp Alphaserver ES45 68/1250
Itanium2 1000 1431 1431 hp server rx5670
Power4 1450 1221 1295 IBM eServer pSeries 650 Model 6M2

SPECint 2000

CPU MHz base peak
AthlonXP(2800) 2000 878 913 ASUS A7N8X
P4 3060 1085 1130 Dell Precision Workstation 350
PIII 1000 454 462 Dell Precision Workstation 420
Xeon 2800 921 957 Dell Precision Workstation 530
Sparc64 1350 747 847 Fujitsu PrimePower 900
Alpha 1250 845 928 hp Alphaserver ES45 68/1250
Itanium2 1000 807 --- hp server rx5670
Power4 1450 909 935 IBM eServer pSeries 650 Model 6M2

they had a better mousetrap

[g4dget]

The Alpha was one of the best 64bit processors out there, years before the Itanium. It should have been highly successful. It failed for two reasons: one was that Alpha-based systems were priced out of the market, and the other was that it is hard (though not impossible) to compete against Intel.

A better strategy for Alpha might have been to do whatever was necessary to price it not much higher than a corresponding Pentium-based system at the time and get lots of market share and software support quickly. But this would have required deep pockets over several years, and pretty much only Intel can afford to do that.

Now, of course, we are getting a worse mouse trap: Itanium is just a horrendous architecture. It should never have seen the light of day. But Intel will manage to push it on us, whether we want it or not, because pretty much all the alternatives are effectively gone. Only AMD's 64bit chip holds out some promise because you can switch to it without changing over your entire hardware and software infrastructure.

Re:they had a better mousetrap

[g4dget]

they are definitely competetive with the other RISC architectures.

Those other RISC architectures haven't been doing that well either.

They are aimed at the high end, not the low-end.

All that matters to most people is price/performance ratios. When the Pentium got better floating point performance per dollar, it took over.

Re:The day of a single very powerful CPU is over..

[sql*kitten]

Distributed computing is going to be the trend...if I can stack together a few cheap chips to rival a single high performance chip, what would I do?

Well, Be thought the same thing, and look what happened to them. Turned out one processor per person was enough after all, for the vast majority of users. Or should I say one general-purpose processor per person, a modern graphics card is more powerful than the CPU for its specialized task. And don't forget you won't just have to buy more processors, but the motherboard to support them - compare the prices of single, dual and quad hardware.

Granted, many apps don't fully use distributed processing power, but the ones that need most CPU probably do.

I think you are confused between distributed computing and SMP. They are different design approaches to different problems. A task that executes well (quickly + cheaply) on one won't necessarily execute well on another, even if the CPUs on both are identical.

What kiled the Alpha ?

[Networkpro]

You've got to have a target audience first. Digital had been slipping for quite a while some due to its outragious support costs. When I was living in Norhtern Virgina in the 80's I had to get a DEC engineer on site for some hardware work and it cost me $5,000.00 just to have him dispatched! I'm a DEC user from the 780 cluster days who moved on to Burroughs, then Unisys, then HP/UX, then Sun back to HP andcurrently am on Sun. Software is always the determining factor on servers.

Re:Alphas

[mark_lybarger]

what about the sparc III? or maybe that was only FreeBSD?

My Alpha

[superid]

I still have two Alpha 266's sitting silent in my basement. Yes they were ok, but I will remember them for being difficult to boot and load.

I know there are normal learning curve issues, it was my first non-intel platform and there was a lot for me to learn.

However, what I discovered was that I couldn't install the RH 7.1 ISO for Alpha, nor the RH 7.0, I had to go back to RH 6.2. This gave me the impression that people (or at least RedHat) had already given up supporting the platform.

Once it was running, it ran PostgreSQL just fine.

Don't cry over split milk

[ites]

The Alpha was never a Windows machine, obvious from the start. We did several projects on NT/Alpha and the sheer difficulty of getting software for the CPU meant it could never compete with Intel. In the Unix market, the combination of Alpha/Digital Unix was very reliable, and we still support some customers who use this, but frankly we can run the same applications on Linux/Intel and it's unclear what advantages the Alpha boxes give. Lastly, the Alpha/OpenVMS combination gave the best results, because OpenVMS is a really solid OS that makes excellent use of the Alpha. We also support a customer (a large tour operator) who uses this configuration: Alpha/OpenVMS/RDB/ACMS.

DEC's strength was always engineering, not marketing, but they were killed by the commoditization of IT due to the twin forces of IT marketing giants (Compaq, Microsoft, Oracle) and open software (mainly Linux). It's clear today that any advantages the Alpha and/or OpenVMS give are completely wiped out by the cheapness of mass produced solutions.

HP is not taking a big risk betting on Itanium because the CPU is almost entirely irrelevant in today's market. My notebook runs 2-3 times faster than the front-end Alpha's used by our tour operator client, and it's only the lack of decent software such as the multithreading ACMS clients we wrote (able to handle 500+ terminals on a modest Alpha) that prevents us using Linux instead, on whatever box happens to be lying around. (And yes, we'll do a port of ACMS and the multithreaded clients so that our client can switch away from his Alpha/OpenVMS clusters).

Anyhow, the demise of Digital and all their technology was clear from the day Dave Custer and his team went to work for Microsoft on NT.

Re:Don't cry over split milk

[drinkypoo]

Yes, even if you're an intel type, you can build a dual xeon for less than it costs you to buy a new uniprocessor alpha. Of course soon you won't be able to buy a new one so this comparison will be impossible but a dual xeon can be had for about the price of a dual G4 powermac. (Kind of sad for Apple, that.)

Re:Don't cry over split milk

[nathanm]

Of course soon you won't be able to buy a new one

The article said HP would continue developing Alpha thru 2004, and selling Alpha boxes as long as there was demand, which they estimated as 2006.

Alpha+NT

[ViXX0r]

A while back we had a couple of MCSE's on our site for "job training". One of the tasks we set before them was getting a copy of MS Proxy Server running on an NT machine. Apparently they had no luck getting the "i386" install to work (the machine in question was indeed of the i386 variety), so they tried the install in the "Alpha" directory.

Of course, they got the "unsupported architecture" (or "incompatable binary" or whatever) error and called me in from the next office to see this bizarre message. I had to explain to them that the files in that directory were for the Alpha CPU and they needed to use the ones in i386.

Neither of these MCSEs had even heard of the Alpha. It truly saddened me. I think it was then that I lost what little good impressions of the MCSE certification track that I had left (which was few, trust me).

Re:Alpha+NT

[TheAncientHacker]

Then they got lucky in passing their exams since "what processor architectures does Windows NT run on" was an objective on the Windows NT exams as were topics on binary versus source compatibility.

Future? How about now?

[Spoing]

Add to that the fact that Intel is pushing for developers to compile using optimizations for hyperthreading and dual processors, and to make apps more multithreaded, and you get an even greater likelihood of performance increases in the future.

Has anyone put this to the test -- recompiled Linux or one of the BSDs with P4 optimizations (or just SMP) and noticed any difference?

Already done

[DABANSHEE]

Ever heard of Windows NT

Mind you that's more 3 parts VMS + 1 part OS/2 + 1 part MS tedium

What if they had...

[sig]

It kind of makes you wonder what would have happened if DEC had accepted Apple's offer to base PowerMacs on the alpha. Back when Apple was getting ready to leave the venerable M68040 series of chips, it approached DEC and said it wanted to make a deal with them to produce alphas for Macs. The CEO of DEC said no, because he wanted to focus the companies efforts on extending the life of VAX/VMS for one more generation, and getting involved with Apple would be a distraction. Of course, he was fired shortly there after for being a knucklehead, but by then it was too late. Apple had teamed up with Motorola and IBM to develop the PowerPC architecture. Still, you gotta wonder what would have happened if he had had a clue and played with Apple.

My Alpha

[LighthouseJ]

I'm glad I got my Alpha when I had the chance, a 21066 chip on a Digital AXPpci 33 motherboard. Sure, it's not one of the fastest ones out there, but I paid $150 for it and it works fine with RH 6.something.

One of these days, I want to snatch up one of several Alpha's on eBay and they have some really nice ones for not much money at all.

For example at this auction, with 5 hours left (at the time I'm writing this) you can get a dual 533MHz Alpha with everything you need for $520, install an OS and you're ready to go. I'd only want to exchange the 6 4.3GB drives for bigger capacities.

Death of Alpha Predicted -- News at 11

[Greyfox]

Didn't Compaq say they were going to stop Alpha production something like 5 years ago? I seem to recall some other company (Not AMD -- Samsung maybe?) licensing the archetecture. Maybe HP will have more luck killing it off...

It's a pity DEC was even worse at marketing than IBM is. My assembly language book from college talks about how DEC introduced their 16 bit machine in 1976 and how a 32 bit machine may be created one day in the near future but that they'd be prohibitively expensive and never gain widespread popularity due to the price. DEC was comfortably ahead of everyone else for high performance mini-and-desktop computing, and they blew it. Their software always stuck me as very well thought out and easy to work with. There was only one problem -- all the managers were buying IBM. Oh well...

HP

[Quill_28]

Quote from former HP employee I know:

"HP will be a printer company in two years"

Re:HP

[Zwack]

I believe that you friend is wrong. HP do more (a lot more) than make printers.
They make storage devices (tape drives, SAN disk arrays...)
They make computers (PCs, HP3000, HP9000(HP-UX machines)...)
They make software (OpenView, HP-UX, OpenVMS...)
They have four research labs worldwide (2 US, 1 UK, 1 JP)

Sure they make printers, but I doubt that that is even the majority of their business. Some companies I know of have hundreds of HP-UX machines, ranging from A class (2U 1-2 proc server) all the way up to superdome (stand alone larger than a 19" rack, 32 or 64 proc servers). Considering that HP, Sun and IBM all sell Unix Servers that Linux is not yet ready to compete with (One application we use is AIX or VMS only and I know a lot of other companies in the same sector worldwide use the same application) HP is unlikely ever to become a "printer company". It may drop it's PC lines... but that is far from being the same thing.

Oh, and I don't mean to offend anyone with the "Unix Servers that Linux is not yet ready to compete with" bit, but it's purely the hardware. When did you last see a Linux box with partitioning (multiple real servers running on the same physical hardware (dedicated resources, not shared resources and emulation)) massive amounts of memory, more than 8 CPUs? I believe that Linux is capable of being ported to this hardware (IBM have already done the same thing with their Z series) but then who would you buy the hardware from... Sun, HP or IBM... they still make the sale, I don't think they are so bothered about the OS as selling you the hardware and the support agreement. When you can go down to your local computer store and pick up OEM partitioning Unix server parts to build your own equivalent Linux box THEN I will retract my statement.

Z.

Re:The day of a single very powerful CPU is over..

[Wolfier]

Well, Be thought the same thing, and look what happened to them. Turned out one processor per person was enough after all, for the vast majority of users.
Well, I think Be failed for some really different reasons. Just like you can't say, "hey, Mozart is a music genius and see what happened to him - he's dead!"

For PC, yes, one CPU per person is enough - I'd extend that further - one person doesn't even need one CPU. If you think this way, then it's pretty clear why we stack processors together - use 10 CPU configuration to serve 30 users on dumb terminals! Isn't that cheap? I think so...that's what a lot of people do when they're short of $$$.

I think you are confused between distributed computing and SMP. They are different design approaches to different problems. A task that executes well (quickly + cheaply) on one won't necessarily execute well on another, even if the CPUs on both are identical.

No, I'm not confused - just putting them together in order to avoid confusion for the people who read it ;) - I've written programs for both. I can say, they're different designs for a similar problem - to use parallel processing power somehow. Writing apps using SMP is easy nowadays if you use a good OS - actually if you use threads a good OS would do the jobs for you although not as ideally as you would do on your own.

On the other hand, multiple CPU on a cluster let's say, is more difficult. There is, I think, 1 or 2 good OS that would help you do the job, but it's not trivial and require participation on your part. For example, LAM/MPI, a very common and 'easy' approach, is still requires pretty explicit communication code in your programs.

I run the first customer code on it

[Hiroto. S]

I think it was back in 1992, I was working for a company called Applicon (or Schlumberger CAD/CAM division) and we were major reseller of Digital hardware with our CAD/CAM software. Due to that relashonship, we had early access to Alpha platform and I did trial porting of our 3D graphics library to the platform.

I mean it was really early. The environment was that we cross compiled/linked the image on a VAX, copy the image to dual mounted disk, dismount the disk, boot Alpha and mount the disk and run the image. No debugger. When the image crashed, I got register dump. Not even stack trace. Network to the box was Digital's LAT (Local Area Transport), so I used Xlib over that transport.

I think I spend a couple weeks there. In 2nd week, we got debugger, version X.0001 or something. When I finally got our library to start running some simple rendering test, the picture didn't look right. A squre cube looked very distorted. Run a quick test of trigonometry functions. Hmmm, sin() returning value bigger than 1.0 didn't look right. Was told that I was a first one to excersize floating point on their chip. It was fixed shortly and we got nice pictures drawn.

I was told that we were the first external customer to run code on Alpha. And of course, we were doing all that work on the only ture operating systerm on Alpha: VMS :-)

Another interesting but far less practical project I got involved later was to try out Digital's binary translator which translated DECstation (MIPS) Ultrix's binary image into Digital Unix (or was it still OSF/1) Alpha binary image. It was pretty impressive. It took our image, which was more than 40M on Ultrix (about 6million lines of PL/I, C and Fortran), and created image more than 80M of size. It was still maintaining whole MIPS image inside it because it has to interpret the MIPS binary in some complicated situation. I think it was for something like exception handling which our PL/I code heavily used. After they fixed the last problem regarding this exception handling, the translated image actually passed through our basic regression test suite. I was not involved but there was also VAX/VMS to Alpha/VMS binary translator, which we played with too. If I remember correctly, some VMS softwares on early Alpha/VMS were actually binary translated images. We never shipped anything using those translators (it is pretty much impossible to debug the translated image), but it was a interesting excersize.

Hiroto

Re:The day of a single very powerful CPU is over..

[Wolfier]

Architecture-wise.

Making a distributed framework were easier on the Alphas than the Intels....but most the time you only do it once - you can say it is a one time overhead. ...and once somebody builds a cluster for the Intel, the "ideal" thing advantage is pretty much gone - if you choose to use a cheap Linux cluster framework (e.g. beowulf!) and don't insist on reinventing the wheel.

Alpha processors were more expensive, and suffers from most of the disadvantages a less popular product generally suffers (less support, less programmers, etc) compared to Intel and AMD.

That's why when the x86's finally catches up with parallelization, Alpha is destined to go away...

Re:The day of a single very powerful CPU is over..

[Doomdark]

You may think so, but somehow how distributed computing has "been coming" for decades now. Just like neural networking, intelligent agents and quantum computing (ok, last one not as long as others). Every few years new modification of DC tags along; right now it's "grid computing" (throw in whatever CPUs you got an manage them transparently), tomorrow it'll be something else (at least name changes for marketing).

There are a few niches where distributed computing is really useful. So far best examples have been, let's see, cracking encryption and searching for ET.

There are many reasons why distributed computing is not "the" general solution for need-for-speed... here are some of the obvious ones:

• In many cases, algorithms need enough messaging that overhead just kills any benefits from scaling.
• It's much more difficult to parallelize algorithms to scale properly, outside the obviously scalable cases (like cracking encryption)
• Maintenance for N individual systems is much more expensive than maintaining lesser number of more powerful systems.

And finally, like somebody else mentioned, perhaps you are mixing SMP and DC here a bit; SMP has fewer problems and is where I see the future, more than with distribution (actually, more than SMP, processor-level multi-threading).

OT: your sig

[ShavenYak]

Arithmetic according to C: float x = 3.14159; int y = 1/2 * x; Value of y? zero.

Why would you say 'int y = 1/2 * x' anyway? Using 'int y = x / 2' is more efficient, and you get the answer you expect (1).

RE: Alpha's

[fshalor]

I use an AS 500 running GNU/Debian every day. It's twin sits outside for general use. Another Alpha, an AS250 is running headless in the corner, printserving and doing some network stuff. (Also Debian.) These three computers have NEVER been shut off since I installed Linux on them, except to move them physically.

I can't say I miss them, cuz they're still here! DEC (Sorry, they perfer to be called Digital Equipment Corperation or Digital I think.) really did they're job too well in developing these guys.

However, HP's being smart. Since the research is being done with other architechtures, it's best to follow suite.

I fell in love with these boxes upon the first boot...Having "more" and "cat" in the bios just rocked my world. I wish others would have taken the hint!

Another cause...

[r_j_prahad]

... was DEC's adoption of exclusionary tactics. Smack dab in the middle of the interoperability wars that bestowed fleeting fame on AT&T UNIX, DEC decided to build walls around every product they made. OSF/1 may have been a superior O/S but it wasn't SVR4 and it wasn't FIPS compliant, and for a while we couldn't purchase it. And a just a little while later, it didn't matter.

Ken Olsen was spot on with his snake oil pronouncement, but it helped kill the company.

Welcome to the new world of HP

[swordgeek]

This is only to be expected, and I was actually expecting it to be announced about six months ago.

Carly Fiorina has made it clear that HP is no longer a technology company, but a sales company. They are no longer willing to take risks, they are no longer willing to develop new ideas and different architectures, and very ironcally, they are no longer willing to invent. If you need proof, just look at their nearly-dead calculator division.

The Alpha is dead. RPN is nearly dead. The spirits of Hewlett and Packard are dead, and Carly is going to make a very successful printer sales company by killing them. Unfortunately.

How we lost the beloved Alpha.

[hateddamntruth]

"I'd think HP would keep it around just so IBM doesn't take over the top spots for supercomputers."

After studying a number of processors indepthly and doing a research project with the Alpha, I came to the realization that the Alpha engineers were truly the best in the world. (My next favorites were the highly-scalable SPARC by Sun, and the PowerPC, another very smart RISC processor by IBM, Motorola, and Apple.) I had also read a book about the pains the Alpha engineers went through to design a processor so far ahead of its time. For some reason, however, DEC couldn't stay afloat and the company exchanged hands a number of times, and, presummably, lost a lot of their engineers. Compaq inherits the company, and merges with HP. Funny thing is HP was already in bed with Intel and helped design the 64-bit Itanium. Now, even though Alpha had gone 64-bit since around 1995, it doesn't make sense for HP to compete with Intel using the Alpha, after all its efforts to help Intel create the Itanium and monopolize the 64-bit market.

When something like this happens, there is always guaranteed to be a fall guy. Before you know it, Alpha, the most outstanding processor in the world, is history.

Re:Alphas

[glenmark]

You mean "set def sys$login"

That's why most VMS users have symbolic and logical definitions in their login.com files to create shortcuts, such as

$ HOME == "device:[homedir]"

I type HOME and there I am.

For that matter, I could drop into the long discontinued posix shell, or install bash from the GNV package (Gnu for VMS), and just use *nix commands. I don't, though. I prefer to use commands that aren't quite so cryptic.

For the record, VAX != VMS. VAX and Alphas are hardware platforms, each of which run various operating systems, including VMS and various flavors of Unix (and Unix-like systems).

You've got it backwards

[TheAncientHacker]

Dave Cutler and his team from DECWest moved up the hill to Microsoft as a result of DEC killing their Prism project. (which is one of the biggest reasons DEC is now a department of the Compaq division of HP)

Also, Alpha was a great Windows machine. It ran Windows NT beautifully and what little software there was ran very well. It's interesting to note that you wouldn't have had trouble finding software for Alpha Windows NT if you'd been able to go with all Microsoft stuff. MS software was, for years, available in Alpha versions.

Re:You've got it backwards

[ites]

If Digital or MS had given away MSVC for Alpha, we'd have seen a decent range of NT/Alpha software. The problem is Windows was popular mainly thanks to the wide range of available software. NT/Alpha could never catch up.
'No software' is always an exageration. The question is 'enough software to compete?' and for NT/Alpha, the answer was always 'not yet'.
Finally, although NT ran well enough on the Alpha, as did many other OSes, this was never enough reason for businesses buying Alphas. NT is not an enterprise OS! Digital Unix and OpenVMS are.

Although I agree, it's way too late

[Featureless]

The EOL may be today, but the massive R&D effort necessary to keep a high-end CPU architecture in the game ended a long time ago. Even if HP reversed course, they'd be starting from scratch, in a sense. The real end of the Alpha happened some time ago, probably during or not long after the DEC/Compaq merger.

HP is taking a lot of heat over this.

[emil]

Remember, HP is killing off PA-RISC as well as Alpha. There have been a number of comments by HP executives recently that they don't hope to turn a profit in the enterprise business for some time because of Itanium costs.

However, the combined Alpha and PA-RISC business is now larger than Sun.

For HP to have executed this merger in a way that retained customers (which they have not), they would have had to deliver EV8. That they have not done so seals their fate. Carly has partnered so much that no one at HP runs the company any longer (to say nothing of what they are doing to their resellers).

Re:Alphas

[CoolVibe]

That would be something I'd love to try. If only for the superior clustering capabilities VMS has. Can you say ueber fast POVray renders?

Of course, KDE on VMS would be hella cool too.

Most Alpha engineers are still with Intel

[AlphaMaker]

>> Intel bought the technology, so if their new 64-bit processor (which shatters compatibility anyhow) doesn't perform well enough, they could just start making Alphas and call them their own.

"I doubt. Intel bought the patents and the documents, but most engineers left. Intel has lousy employee relationship, so they wouldn't be able to reproduce the in-house expertise Digital, Silicon Graphics, HP (before merge) had and that IBM, Sun now have."

That is a widely spread fiction. Only a small handful out of several hundred engineers have left since the Compaq deal with Intel. In fact, there are 3 or four who are Intel Fellows. That is essentially a VP-level engineering position, the highest possible. True engineers can't resist working on something new and sexy and Intel gave it to them. They get to design the next generation Itanium from the ground up. As a result, most of the engineers elected to stay.

We'll have to see how it all turns out in a few years.

Re:Most Alpha engineers are still with Intel

[leandrod]

> That is a widely spread fiction. Only a small handful out of several hundred engineers have left since the Compaq deal with Intel.

Maybe. I would like to see (1) numbers & (2) a qualitative analysis about this.

But yet, there were some high-profile defections, not only after but even before the Compaq deal. There were defections when Digital deteriorated, there were defections when Compaq bought Digital, there were defections when Compaq continued the deterioration trend Digital had started, then again when HP continued even lower, and finally when HP ditched the Alpha in a very high-mannered and questionable, both from technical and business perspectives, decision. Or so was reported at each point.

So I wouldn't discount this as a myth without further evidence.

Re:Most Alpha engineers are still with Intel

[AlphaMaker]

Since I work in the group, I know firsthand. Since I still work for the group, I won't give specific numbers.

Every company out there has some level of attrition each year. In the cpu design business, it tends to happen at the end of a project. The industry average when not in a recession is about 10% annually. Every major Alpha implementation (EV4, EV5, EV6, and EV7) has had several engineers leave after more or less completing the project. Previous to the Compaq merger with DEC, the Alpha attrition rate was well below industry average. Afterwards during the hot dot-com economy, we approached the 10% industry average. (silicon valley is actually worse on average) One thing to keep in mind that while several very experienced engineers have left, it is still a very experienced design team.

Absolute numbers don't actually mean anything by themselves. You could lose the best 10 engineers or the worst 10 engineers. Where will you ever truly get "evidence" about something like this? Surely not from the average internet columnist. I wanted to set the story straight, since it seems to be a popular myth to propagate in tech columns. None of these columnists has had sufficient inside knowledge on which to base their claims. I couldn't prove it either unless you came to work here. %^)

Re:Most Alpha engineers are still with Intel

[leandrod]

Thanks for the info, even if I hoped for something a little bit more /juicy/! ;)

OK, so the picture you paint is that the Alpha team was more stable than average, and now it is just the average. So, absent other considerations, it went from top to average, and that is not good in itself.

Moreover, now that Alpha is officially ended, after being all but dead for several years now as far as an architecture with a future goes, what will happen? I do not hope for any of these apple-in-the-sky scenarios of Intel having a change of heart after a Itanium fiasco, but I do think Itanium is already a technical fiasco and always will be, even if the market gobbles it up as it has been doing with everything Intel.

So even if the team is still good enough, while not top notch, I doubt it can make a difference. All this stuff from columnists about Alpha tech making its way into Itanium is but incremental improvements to a doomed enterprise in my view, since their basic philosophies are so opposed to each other. I may be wrong, but I still cherish that Alpha-vs-IA64 paper.

Re:Alphas

[glenmark]

Typo and poor proofreading on my part. Make that symbolic def
$ HOME == "set def device:[homedir]"

alpha-athlon?

[Dave_bsr]

Actually, the Athlon is a RISC CPU that translate x86 instructions into it's own instructions. I don't know if that is because it's based on the Alpha or not, but I had heard the above from some people who should know whether it was true or not.

Re:alpha-athlon?

[Ben Hutchings]

It is not really a RISC CPU, but its core certainly is quite RISC-like. Decoding x86 instructions into variable numbers of simple micro-ops makes it easier to schedule and process them efficiently. The same approach was used in the K6 and has been used in all Intel chips since the Pentium Pro (excluding the Pentium with MMX).

...huevos...

[Dave_bsr]

that's a good wordplay. Mod's, even if you don't get it, I do...so mod the above up please. "Huevos in one basket..." hahahaha.

It's not OVER

[drinkypoo]

It's just on pause for a while. The next technology in CPUs will probably bring it back. "Tired of needing 1,000 node clusters to do your simulations? Try new QuantumCpu3000Pro! And do the work of 1,000 old machines with one computer!"

Re:The day of a single very powerful CPU is over..

[poot_rootbeer]

if I can stack together a few cheap chips to
rival a single high performance chip, what would I do?

You'd probably fly to work on a unicorn, and eat sunshine and moonbeams for lunch, because you'd be in Fantasy Land.

(Given today's existing products and sufficiently meaningful values for 'a few', 'cheap', 'rival', and 'high performance', that is)

Why do some chips succeed and some fail?

[00_NOP]

I noticed this week that development on the venerable z80, described as a microprocessor back in the day, but now relegated to the world of the "microcontroller" continues - with chips being etched on/placed on glass and certainly I still have a working one in my Gameboy Color.

But the z80 is a technology that is over a quarter of a century old - the gap between us and it is nearly as big as between it and "Colossus". So why is it still in use, but the Alpha is to die?

Sure the z80 is cheaper than running water but remember all you get is an addressable 65k and, what, 2 Mhz?

New HP vs Old HP motto's

[Indy1]

Old HP - "Invent."

New HP - "Merge, layoff, go out of business."

Re:Alphas

[CaptainCarrot]

DECWindows is the VMS port of X, so replacing it with XFree86 would be pointless, and probably make it even less robust than it is. Perhaps you meant to talk about the old Session Manager, and if you are you're right; it's archaic. It's so archaic that no one I know of uses it any more; newer DECWindows installations use the CDE. I'd prefer KDE or Gnome anyway, mind you...

Re:Alphas

[CaptainCarrot]

So type "cd" if you want; that's what I do. If you didn't understand how to define symbols in your LOGIN.COM to make commonly used commands a bit easier to type in, you didn't understand VMS well enough to take full advantage of it.

Really it's no more complicated than Unix, but there have been a lot more people spending a lot more time learning Unix than VMS, especially since DEC went down the toilet. Of course *nix is going to seem easier. It's more familiar. To someone like myself who uses VMS almost exclusively at work, Unix seems complicated and arcane. It's all a matter of what you're accustomed to.

Re:Alphas

[CaptainCarrot]

VMS + 1 = WNT, in the same way that IBM - 1 = HAL.

I think that was the intent all along. The programmer in my group who wrote all our device drivers looked into WinNT internals once, and comparing it to VMS called it "deja vu all over again."

Re:In Soviet Russia...

[Zork the Almighty]

Quick ! Someone mod down this troll !

Re:The day of a single very powerful CPU is over..

[anonymous cupboard]

First there may be some confusion between multiprocessing and clusters. You can cluster anything, you just need some good software. Clustering started with the VAX architecture, a high-end CISC machine. It could run on a lot of different machines and I'm interested in what IBM are planning with their emulation of the distributed lock manager. Cross system resource locking is fundemental to all kinds of projects (such as cluster wide file systems).In any case, if HPaq stand by their commitments (to the DOD amongst others), they will get VMS up on Itanium and this means VMS clustering. As most of the hard work was done during the Alpha port, getting it up should be easy. Only performance will be a question as much of VMS is still written in VAX assembler and then 'compiled' for the Alpha. VAX assembler is pretty high level so this is feasible.

Multprocessing is different. You just need some good locking primitives and ways to keep caches coherent with the necessary connections brought out of the chip. If the architecture doesn't support multiprocessing, it is very, very hard to graft it on later.

Don't sell the Power4 short.

[PeterM from Berkeley]

I think the Power4 has a real shot at taking over the customers that Alpha used to serve. It has a Real Company behind it, not a succession of laughing-stocks. I think AMD won't be able to match Power4's performance, because of all the x86 baggage.

On the other hand, IBM doesn't seem to want to market Power4 agressively, nor price it competitively in price/perf with x86. I think they could sell a lot of them if they could offer 1.5x perf at less than 2.5x the price.

If they come out with a version of the Power4 which uses significantly less power as well, it'd be very interesting to those who set up clusters of computers, as it'll reduce demand on infrastructure.

PeterM