Slashdot Mirror
IBM To Produce Copper Alphas For Compaq
LinuxGeek writes: "IBM will be producing copper interconnect Alpha CPUs. The samples are already running 1.2GHz. Hopefully they can make them cheap and plentiful."
Similarly, sokoban
writes:
"News.com is reporting that IBM is planning to manufacture Copper Alpha
CPU's. Now that is a fast piece of equipment. Here is the
link." And nobody ever got fired for buying ... err, Alphas. Soon, 32-bit will seem so quaint, eh?
The problem is that most people will use gcc/g++ and not the DEC compilers on alpha. Or the intel proton compilers for x86, or Xlc from IBM for powerpc, or mcc for powerpc, or ... you get the point. At some point, even for performance computing, portability becomes a major factor. Vendor compilers are notoriously a pet optimization project and break in the real world..... if you're really interested in getting high-performance compilers on alpha, I suggest you:
1) fund development, or try to get optimization development funded. Intel has funded both strongarm and x86 opt work. Mips too. It's not a mystery why some of the gcc backends produce better code........
2) actually contribute work for the alpha port
3) get compaq/DEC to open their optimization work. IBM did this with the HAIFA scheduler, and it made a big difference. Imagine what a contribution DEC could make.....
Apple's MOtoroLa PowerPC is stuck at 500mhz...IBM has Passed that and are shipping 600 and 650mhz POWERers(PC)
Try building a Linux kernel (IE: 2.2.14) on a SparcStation LX, then you can talk about slow.
Huh? What does Intel technology have anything to do with this?
--
$x='S24;r)>63/* h@<5+oZ)32"5cz';$me='phroggy'x$];
$x=~y+ -xz+\0-Tx+;print$_^chop$me for split'',$x;
--
$x='S24;r)>63/* h@<5+oZ)32"5cz';$me='phroggy'x$];
$x=~y+ -xz+\0-Tx+;print$_^chop$me for split'',$x;
The PowerPC G4 uses 32-bit integer units, 64-bit float units, and a 128-bit vector processing unit; the technical details are here.
--
$x='S24;r)>63/* h@<5+oZ)32"5cz';$me='phroggy'x$];
$x=~y+ -xz+\0-Tx+;print$_^chop$me for split'',$x;
Actually I heard a rumor awhile back that Apple, IBM and Motorola were considering inviting AMD to join the PowerPC Alliance, so AMD could help out with fab; it makes sense, considering that AMD and Motorola already cross-license a lot of technologies.
--
$x='S24;r)>63/* h@<5+oZ)32"5cz';$me='phroggy'x$];
$x=~y+ -xz+\0-Tx+;print$_^chop$me for split'',$x;
--
$x='S24;r)>63/* h@<5+oZ)32"5cz';$me='phroggy'x$];
$x=~y+ -xz+\0-Tx+;print$_^chop$me for split'',$x;
In some numerical code (photometric redshift estimation), the Alpha 21264 consistently moves at least 3 times faster, but most often is closer to 5 times faster than similarly clocked P-IIIs.
Alphas really are damn fast, not just the chip but the whole architecture. If you're the kind of person that overbuilds everything you create (like me), then the Alpha will make you smile.
=-) =-) (we have a dual alpha)
-Paul
Altough some may think otherwise (or are just uninformed)... Microsoft Office (at least 4.3) did exist for the Alpha. As well as Microsoft Word 97 and Excel 97. For the endianess, the Alpha is little endian like x86 chips. If you don't know, just don't assume...
Who is "mcelrath", and why is he such a f*cking idiot? Who is Anonymous Coward and why are half of his posts worthless? If you have a point, make it. Someone trying to slime another post doesn't lower my opinion of the post unless there are some good facts stated about it.
Glad to hear it. For some reason I thought it had allready lost to P-III@1Hhz, or a K7@1GHz. The numbers I posted were SPECfp95 numbers (that's what AMD had). I do think someone else posted P-III SPEC2000's, but I've forgotten.
It would be nice to see the 1GHz K7 SPEC2000 numbers. As far as I know they meet SPECs def of shipping, so I wonder why AMD hasn't seen fit to publish them. Maybe they are waiting for the large integrated L2 cache version?
Try SPEC at http://www.spec.org . They have a very nice configurable search system for their CPU95 and CPU2000 benchmark results databases.
To take a few high points from the current data:
alpha 21264 667MHz, integer 424, FP 514
Pentium III 1GHz, integer 407, FP 273
HP PA-RISC 8600 552MHz integer 367, FP 338
Pentium III 800MHz integer 361, FP 260
Power-3 375MHz integer 248, FP 330
Sun Ultra 333MHz integer 133, FP 128
A Sun Ultra 300 is 100 by definition, twice as
big a number means that the test took half as long.
There are no Athlon figures up (anyone know them?).
Conclusion, the current generation of alphas really screams. A 1.2GHz version should be quite something! HP stays ahead in the astonishingly much performance per MHz ratings, pursued by IBM.
Not in SPECint2000, according to the figues above, although a 1GHz Athlon might five it a run for its money on your figures.
If you know where to look for it, you can find alpha based hardware at dump prices, just because they still bear the 'Digital' logo instead of Compaq.
They're not state of the art anymore, but if you want one for private use that is hardly going to matter. (At least the processor in my PC is rarely less than 90% idle for extended periods of time, and unless you're a dedicated gamer harddisk speed is almost certainly your bottleneck, not CPU speed)
What's this whole 'octane' thing anyway? It can't have much to do with the proportion of octane molecules.
-- Ed Avis ed@membled.com
One thing DEC has always been very good at is
emulation. Push them to port FX!32 (their 32-bit
intel emulation which ran NT/intel binaries on
NT/Alpha) to Linux, OpenVMS, and Tru64 instead of
letting it wither as it has since NT/Alpha died.
On top of that, stories abound about how antique the Alpha's chip fab technology was.
So...
...Wanna bet that IBM doesn't mind making the chip? Especially if they're looking at cross-licensing some stuff?
(Heck, how many of IBM's own RISC patents are part of the Alpha anyway?)
C'mon, IBM's got some cool silicon of their own. I would not be surprised if IBM itself decides to start ANOTHER line of machines based on the Alpha (RS/7000 anybody?)...
Naw... It's more likely that IBM is having some fun and keeping the chip foundries cooking and just irritating Intel as much as they're annoying Sun...
Is this a no-downside scenario or what?
-soup (GNUrd, Speaker to Machines) "Laugh at yourself- Why should everyone else have all the fun?" -Romanchek's 6th Ru
Oh yeah, well try installing Win2K on a pacemaker! Oh boy, did THAT take some time!
--Bob
1^2=1; (-1)^2=1; 1^2=(-1)^2; 1=-1; 1=0.
I HIGHLY recommend doing this if you're a little hardware savvy (like to tinker), and are fed up with Intel's crap (like me). But then again there's always the Athalon. But then again the alpha still beats the Athalon at floating point.
The DCG people have been very good to me as well. When I first got the thing, I accidentally nuked the Flash BIOS, because I'm an idiot. They were very agreeable, and sent me a new motherboard. All I paid for was shipping! Try and get that out of some no-name Taiwanese fly-by-night PC outfit!
--Bob
1^2=1; (-1)^2=1; 1^2=(-1)^2; 1=-1; 1=0.
Okay, now being in the lab at school, I can tell you more about these things. There are ten Ultra 5s in the lab. Each one has a Viewsonic PS790. And each has 3 dumb terminals attached to them in the lab. The dumbe terminals seem adequate for everything but using Netscape.
I attend PSU, and in one of our labs, we have about 10 Sparc 5s all with Ultra Sparc IIs, I believe. I may be wrong. Each Sparc has between 3-5 tektronix (spelling?) dumb terminals attached to them.
last post!
Actually, Netscape on the Alpha is only a rpm -i netscape.rpm away - Digital (please dont make me call it Compaq) has
There are no (repeat: ZERO) 900MHz Alphas available on the market at the moment. The best you can get is 750MHz, and you'll pay through the nose for that.
IBM stopped marketing home computers. Even though they were losing money hand over fist with them, they were still so crappy that they were giving IBM a black eye.
Now if we can just get Compaq to do the same - Somewhere on this thread there's probably some kid saying "Compaq Wildfire Alpha Servers suck!! After all, look at this Presario my dad bought me!!"
--
Business. Numbers. Money. People. Computer World.
Why, do build the Linux kernel, of course! Have you recently looked at the size of that beast? Even with a dual-Pentium mo-bo, 256 MB RAM, and kick-ass SCSI disk, I still have plenty of time to have a cup of coffee (sipping it VERY SLOWLY, of course), before the kernel build is complete.
And, of course, you never know when you'll get the urge to rebuild the kernel, so you better come prepared for that eventuality, that's what I say...
Actually, they just announced SOI on PowerPC today. ZDNet has the arti cle.
Basically, it'll debut in AS400 systems in Aug, RS/6000 later this year, and eventually in consumer PPC devices.
compiling mostly...while this may not be a home use, when something compiles at double the speed, it halfs the time needed to debug. Also, when you've got 10 emacs windows open, are compiling something, have mozilla open, etc etc, it bogs down the cpu...the only way of getting around this is A) speed, or B) SMP (i have 2 300 MHz processors that were 900 bucks each when i bought em).
Oh, and Unreal Tournament, HGII, Q3, etc, etc, etc.
One Microsoft Way
My plan is to pimp before they realize I'm a jackass. Hit 'em hard and fast.
Ok, yes, alphas are insanely fast. The problem is, they're not binarily compatible. They're sort of like Linux in that way...small market share, small base of users, but much better than the competition. Only reason people don't switch is because they won't be able to use everything they currently have with it, therefore making it less user friendly. I know they're not *meant* for home users, but it would be nice if they were ;-)
So, on another note, anyone know of anyplace (on the web or otherwise) that I can find out exactly what IS up with BIG_ENDIAN and LITTLE_ENDIAN? I'm not going to be using a BIG_ENDIAN machine anytime soon, but I've always been interested.
One Microsoft Way
My plan is to pimp before they realize I'm a jackass. Hit 'em hard and fast.
Wasnt Intel making Alpha chips for compaq because of FTC rulling?
Free Unix? Free Windows. http://www.reactos.com
Also, IBM does not hold a patent on copper interconnects, it's just that they happen to own the only fabs capable of making them. (AMD's copper-capable fab is co-owned by IBM, with the agreement that AMD will only use it for AMD-branded chips.)
Intel has stated (dang it, where's a link to that press release?) that at the present time, they don't need to move to copper to keep up. They may move to copper in a year or two, but not right away. Remember, the big thing with copper interconnects is that it lets you use smaller trace widths, and Intel seems to be doing just fine at .18 without moving to copper. They have even said that they could get down to .13 before needing to move to copper.
And, for the record, 'Coppermine' was never meant to imply copper interconnects. That codename (named after a river in Arizona) was in use internally to Intel before copper interconnects were a big deal.
Another non-functioning site was "uncertainty.microsoft.com."
The purpose of that site was not known.
In background, I am an alpha user, hell, I even have an old alphaserver as a playbox at home. Recently, I came into a position where I could bear to upgrade some servers, and was thinking of some little rackmount freenix powered boxes- maybe some SMP alphas.
I called my usual alpha dealer (it's been a while since I have had to talk alphas), to get prices for a mobo+cpu+ atx rackmount case, and was told that compaq had decreed an end to ATX form factor systems.
Instead, you have to get a proprietary form factor case and mobo. This puts the entry price for a bare mobo/case/cpu package at 1600+ukp, for a machine with a lot less power than a 900ukp althon box!
In the past, you could build a nice powerful system with your own parts, and not pay the awful penalty of having to put up with hugely inflated prices chaged by compaq for that awful plastic proprietary stuff they peddle (can you say "lock in"? I knew you could).
It was something of a different ball game when it was Digital kit.. my old alphaserver is built like a fscking tank, and I can see where the money went. However, when I look at my compaq floor standing raid towers, all I see is overpriced, tacky and plastic.
It's a real shame that Compaq hoovered up DEC in the first place (ex-VAX/VMS users tend to get a little overly emotional at this point :). It's rather like Albert Einstein having to work for Laurel and Hardy, to my mind. Yes, I know that L+H had a much bigger target market, but Uncle Albert commanded the respect, and had a certain dignity :)
Anyway, by taking a massive step backwards, and making the Alpha suffer proprietary form factors once more, compaq has priced itself out of yet another market, by charging surreal amounts for products of dubious value.
My heart doesn't bleed for them, I hope they rot.
-A-
My usual dealers are UK based (I'm reesident in the UK), and I was going by what they told me. However, I will be investigating this when I am a little less tired... Before the pound drops too much against the dollar, mebbe.
Thanks for the link- I prefer to build my own boxes, not only from a price POV, but because I love to know what's going into them, much like rolling your own linux kernel...
MSB LSB
0 31
1 32
31 0
32 1
Obviously I meant the standard GNU libm compiled with GCC...
On a 386sx20 the 0.9 series used to take about 6 hours. I always used to leave it to compile over night. It's scary to think I actually managed to get X & netscape running on that box!
I ported my company's product to Linux on Intel and of course I used the GNU toolset. The next step was Linux on Alpha and I kept getting suggestions to use the Compaq compilers/libs.
Forget it! If I stick with the tools all I have to do is type "make" and fix maybe 3 32/64 bit issues. If I switch tools I have to fix orders of magnitude more problems with the makefiles, libraries, etc. And who gives a rip about performance in an email server? It spends most of its time waiting for user input or talking over the network--the CPU is rarely if ever the bottleneck.
--
Have Exchange users? Want to run Linux? Can't afford OpenMail?
Linux MAPI Server!
http://www.openone.com/software/MailOne/
(Exchange Migration HOWTO coming soon)
You'd have more power with 80-octane gas, but you'd probably need to tune the engine differently, perhaps using valves/pistons with less compression, otherwise it would ping to death.
Remember, 140 octane == water.
Well, everyone is cheering for 64 bit, and with good reason. IA-64, Alphas, Sparcs, all run at enormous speeds and utterly annihilate IA-32 machines. :)
Of course, there are always the problems of general market penetration. I am still waiting for the day when I can use a 64 bit machine for something other than just a server. (I wants Unreal!
But here is my main point. First there was 16 bit. Then 32. Then 64. And, of course, there are a few 128 bit machines floating around. My question is, why not just blow it all away with a grand 256 bits? By this I don't mean 256 bit instructions, but a 256 bit wide bus, 256 bit integer processing capabilities, 256 bit vector-style bulk integer computations, etc... I don't see any reason for this not to happen. It doesn't require us to break any quantum limits or develop any new fab processes. It's just a bigger design.
Oh, and on a side note. Does the Alpha support the predicate processing type that IA-64 has? Or is that a unique feature?
Actually Alphas aren't usually targeted to home users. They are primarily popular in the corporate realm. Also, keep in mind that alphas have been at 900MHz (with DOUBLE speed cache for quite some time. And why would a home user want a 1.2 GHz AMD Athlon? Quake3. And that's about it.
Just a quite note.. I remember reading somewhere a quote from one of the Linux leaders (it was David Young in a magazine interview, I believe) that Compaq hardware isn't the best choice when running Linux, because they are very proprietary [sp] with their extensions (it went on to say that's why Dell has been successful; they are always ready to use open-standard hardware.)
Or am I misinterpreting this, and all it means is that Compaq will make the CPUs and not the boxes (I can't quite tell.)
Eitherway, Alpha's are cool, and I hope more people relize their potential.
Pete
// Damn I wish these comment boxes were wider, I hate HTML
...
>>Microsoft never ported MS Office to Alpha.
Not entirely true. They ported Word and Excel ( i have the CD's). Corel ported all of wordperfect suite 8 to Alpha/NT (minus netscape and that mail center thingge)
>>On Linux, there is a severe paucity of >>"commercial" applications for Alpha, and some of >>the critical applications for "desktop" use are not available.
Well we dont have staroffice but we DO have applixware. Anyways I do most of my work with opensource tools.
>>Notably, Netscape is not readily available. (I know that the Digital UNIX version can be hacked into submission,but it is not an easy rpm -i netscape.rpm away...)
actually yeah it is as easy as RPM -ivh . Look on the Redhat-6.2 CD.
Tru64-compat-1.0-3.alpha.rpm
netscape-common-4.72-1.alpha.rpm
netscape-navigator-4.72-1.alpha.rpm
netscape-communicator-4.72-1.alpha.rpm
and Java works too.
>>On the server side, Oracle for Linux is not available. Ditto for Informix, Sybase, DB/2,
So does that mean I can't use Postgresql or MySQL? Oracle and company will port when the time is right.
>More critically, Linux/Alpha is a considerably more "brittle" platform than Linux/IA-32.
Do you mean unstable? hardly the case.
>>Linux may not be the rats-nest of spectacularly nonportable software that Windows NT is, but getting things like KDE, GNOME, GIMP, and such has been a fairly slow process. Hardware compatibility lists are similarly "brittle."
KDE runs excellently on Alphalinux and so has the GIMP since the RH-5.x days. GNOME works reasonably well. What hardware support is so lacking? There is support for a few RAID cards, nearly every SCSI and network card out there.
>>I've tried running the Diamond Rio package, rio on my Alpha box; it gets quite confused, probably due to some 32-bit-ism in the "driver."
Packages can't be fixed if you don't tell anyone that they are broken.
>> More critical is the paucity of graphics cards supported by XFree86 on Alpha. It is probably similar for other 64 bit platforms; you're restricted to whatever there are "open" drivers for, and there are some cards (Cirrus comes to mind) that have architectures that are distinctly unfriendly to 64 bit operations.
That's strange... I don't know how my Voodoo 3000 works then with XFree-4.0. Nevermind the 3dfx framebuffer and ggi. Which is very fast BTW. The only thing Alpha is missing out on right now is hardware accelerated graphics drivers on the newest cards and thats being worked on now.
>>All of this adds up to Alpha not being a particularly consumer-friendly platform. People that "just want it to work" will find it somewhat challenging in this regard.
If thats true then Sparc and SGI are just as "consumer-unfriendly platform". 64 bit computing at the time when windows NT was on Alpha was never meant to be a home PC but a high performance workstation. What people are you referring to?
>>This is presently the case for Linux/Alpha; it was the case for Windows NT/Alpha.
Absolutley not. More hardware and software is supported by Alphalinux than NT ever had. Also the community and commercial support greater than NT ever had...
>>The same will likely be true for IA-64, for a goodly while, by the way. The pains that Alpha has had may help IA-64 porting efforts to be less painful, but what "cramps" persist are likely to be irritants to adopters of IA-64.
at first yes but with the influx of 64 bit developers and commercial interest will make those growing pains short and sweet.
>>In short, I don't think this was a likely niche, except amongst "daft hobbyists," which doesn't represent all that "nice" a niche...
Thats rude. I know alot of UNIX admins and MCSE's that say the same thing about Linux in general. I apologize that most people who use Alphas have no choice but to use the fastest and most mature 64 bit platform out there to get their WORK done.
Regards,
Peter
--
www.alphalinux.org
www.alphalinux.org
Yes you can get a new Alpha for under $3500US.
Go to http://www.alpha-processor.com and sign up for the developer program. You can get a new 21264-based system with AGP graphics for $2999.
mike (I work for API)
What's my Karma Mr. Burns? "Excellent"
That's not a Compaq Alpha, it's an API Alpha motherboard. http://www.alpha-processor.com I work at API.
What's my Karma Mr. Burns? "Excellent"
The reason why Microsoft didn't port the Office suite to the Alpha platform is that most of the Office applications will run under FX!32. I was able to get portions of Office 95 and 97 running, but the machine I had was fairly old and only had 48MB of RAM, not enough to get some programs running.
:(
Microsoft did make a version of Word and Excel for Windows NT (a long time ago... around Office 4.2 for Windows) that was supported on the Alpha platform, but never made any newer versions
I agree that the Alpha platform [at this very day] isn't very ``consumer-friendly'' since it was never pushed as a consumer-level platform. Compaq nor Digital really thought that the Alpha could be produced cheap enough (due to large on-die cache, larger fabrication process, number of pins, etc) to be something other than an enterprise/high-end RISC workstation solution.
Digital introduced and sold the 21164PC processor to be the ``Intel killer'' of processors, but lack of support at both Microsoft and Digital killed the killer. Even though the 21164PC processor was a severely castrated version of the 21164 (sounds like the Celeron 266/300 v. Pentium II and the Celeron II v. Pentium III) that still had problems with low yields and huge power consumption.
I'd love to see the Alpha and other RISC processors get a bigger piece of the consumer slice, but it will take a lot of changes at Compaq to do so.
Heck, at one point in time, Dell would advertise on their website that one or two series of their storage area network solution used the Alpha processor. Now they just say that they use ``RISC'' processors (probably because Intel didn't like Dell advertising competitive technologies).
I take it you are referring to Linux only - as our compilers for Tru64 UNIX and OpenVMS are VERY highly optimized for the latest generation of Alpha processors. With the introduction of Compaq Fortran, C and C++ for Linux Alpha, that has changed too - many users find applications speed up a lot using our compilers (especially Fortran). gcc for Alpha is really quite good, as it is, though.
Steve Lionel
But they do have many other alpha motherboards (164pc and 164sx) as well as other models I don't know. Plus, they have many used SPARC's and SGI's around, if you want to expand your hackerdom away from the x86.
If you're in the boston area, they're definitely worth a visit. Their used/salvage area is totally kickass!
make world, not war
Actually, I've found that rebuilding KDE(2) takes MUCH longer than the kernel... I used to think the kernel took a while, but it doesn't compare to a comparable sized C++ build.
[Posted from deep within the northen forests of North America, via TCP/IP tunneled over carrier pidgeon]
After I went to Ground Zero -- Compaq -- I came away with a discreet impression that they really don't want to talk to a mere peon like me. Their web site, apparently, is really put together more like a business-to-business (or government) type of a deal.
You're defintly right about that... my experience trying to spec a nice Alpha box at Compaq's site was about the same as yours. However Microway seems quite willing to sell single machines to private customers. They have a 533 Mhz 21164 (with a reasonably good setup) for $2000. For some reason the price for a NT box and a Linux box is the same, though - I guess the best move would be to order the NT machine and then install Linux (keep the NT CDs for posterity or something).
But now with the Alpha unable to sustain a big clock rate advantage, does the Alpha get any sort of real performance advantage when compared to x86, or particularly other less "reduced" architectures like HP's, Sun's or IBM's?
Alpha will still kick the crap out of x86, especially in floating point. HPPA's are supposed to be as fast or faster than Alpha's at floating point, though I don't know much about HPs stuff. Ultra's are slow. They're nice when you put 4 or 8 into a server. I don't have benchmarks, but an Ultra5 running Solaris 'feels' slower than my P-II 350 running Linux. And Sun's pricing will make you feel like an Alpha is cheap. PowerPCs are pretty cheap (and very nice chips), but not as fast as an Alpha, AFAIK.
Also, you can't get 32 or 64 proccessor setups for x86, which you can for some of the better Unix-vendor RISC machines.
Anything I missed? Is Intel going to go Copper, or are they just going to keep on naming their processors so that they sound like it?
Willamette (the P-IV) is supposed to have copper. That's not coming out until (IRRC) Q3 2001.
Any system running Solaris will probably be slower than a system running GNU/Linux. Have you tried Linux for Sparc?
Trust me, I'd much prefer to use Linux over Solaris, but those are my work machines. I don't think my boss would go for me installing Linux on any of _them. Or the ones in the public CS lab, for that matter. Most of my university has 'standardized' on Solaris (so we can use the same OS for servers and workstations, better support, whatever).
would be Compaq's new chip, the Beta. Hey, they made CPUs called the ARM and the Thumb, right? I think I heard about one called the Hand too, but hopefully that was just the voices in my head.
I know I am!
I'm also thinking it might make slashdot load a little faster.
However, if you want to go cheap, then don't buy branded machines from the Q. Pick up an Alpha motherboard from a third party.
http://www.alphalinux.org/ would be a good place to start looking.
Well, i know I'm excited....
----------------------------------
Now don't get me wrong, I'm not saying that faster compiling isn't good and wouldn't speed up the debugging process.
--
I got a sparcstation 10 on my desk at work. Being an NT shop (and I'm the only UNIX guy) I have to use NT all day on a P200, which isn't all that much faster (thanx to NT) than the sparc.
I've never used any alpha other than a multia, but that was enough to make me want a real alpha. After getting used to my sparc 10, alphas kinda seem to me like a middle ground between intel and sparc. Good starting point for those who've never used anything but intel systems... just not having a standard BIOS was a bit of a shock the first time (a few years ago).
I want to get another alpha eventually, but right now I'm saving up for a used Ultra II. Something in the 300-400mHz range. And then triple-boot linux, solaris, and BSD...
Those who can't do, teach. Those who can't teach either, do tech support.
http://www.whatis.com/bigendia.htm
that was the first hit I got on google searching for "big endian little endian".
At least you didnt submit it to "ask slashdot".
--
http://gammatron.weblogger.com
Any system running Solaris will probably be slower than a system running GNU/Linux. Have you tried Linux for Sparc?
Will I retire or break 10K?
I hope this deal will help out compaq in their server line. Over the past couple years they have lost the desktop market to other more "e" OEMs. As for the Alpha processors themselves, I hope that this PR will somehow influence upper-management into stepping away from x86 technology.
Isn't this what power computing is all about? I don't know much about Alpha machines in particular, but I myself have found that most alternatives to x86 provide a more robust system that can take a real server's load. My experience with x86 has been that hardware itself can often be flaky leading to hard lockups and erratic behaviours. Then again, maybe that's just my personal experience/opinion.
I would figure that by paying these extra prices, you *should* be getting a superior machine with quality assurances. As far as I can tell, you aren't considering these additions in your pricing. Then again, someone correct me if I'm wrong in saying that you get more for your money with Alpha boxen than with x86.
Try it on a Sparc IPC (25Mhz) with the 2.0.36 kernel before they finally fixed the mmu slowdown bug. It took somewhere around 5 hours :)
Only the State obtains its revenue by coercion. - Murray Rothbard
Uhhh...I don't know what you're talking about. Two things: AFAIK (I worked on the VC compiler for Alpha) FX32 was never "outlawed". In fact, it was integrated into the Alpha version of Win2000. Second, FX32 is a moot point since Compaq dumped NT as an Alpha OS. Even if FX32 was "outlawed", it wouldn't be why Alpha will never be a consumer platform. It would be because Windows is a dead OS, as far as the Alpha is concerned (boy that's fun to say).
And the FX!32 technology is still available for Linux, where it's known as em86, allowing you to run Intel binaries on your Alpha Linux system for the odd application that doesn't include source :-(
Information wants to be free -- but informants want to be paid.
Try this link: http://www.alpha-processor.com/resellers/system_ve ndors.shtml
I don't know if the proces will be to your liking, but at least you'll have 31 vendors to evaluate.
Information wants to be free -- but informants want to be paid.
They did port Word 97 and Excel 97 to native Alpha NT 4.0 (I even have copies), but the ports were not very good, and in general it worked better to run the x86 version under FX!32 (which does what Transmeta's code-morphing software does, except that the Alpha doesn't have the advantage of being hardware-optimized for the purpose).
Information wants to be free -- but informants want to be paid.
Not all that surprising once you learn that a lot of DEC's Alpha hardware engineers now work at AMD (and are the ones responsible for the Sledgehammer design).
Information wants to be free -- but informants want to be paid.
I found a page comparing the 500 mhz Alpha 21264, the 400 Mhz Apple G4 (using optimized code), and a 600 Mhz P3 "Coppermine". Here is the link. Basically, the G4 with altivec smokes the P3, but the Alpha absoloutely demolishes the G4 and P3. The alpha would probably even beat the G4 and P3 if it gave them a clock advantage of 300 Mhz. With a 1.2 Ghz Alpha you will be hard pressed to find something faster
09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0 is the magic number.
Hey man, it's all about the marketing, and the image ... Intel's been at this game for years.
JAVA
with faster chips it will become practical to have platform independant software that can actually run at a decent speed. Along the same lines faster chips will enable os emulations to run at a much nicer clip.
the other is
"The Future"
It may sound stupid but really, the argument that chips are fast enough because "everything I need to run, I have the power to run" has been put forth since the dawn of the computer age. What you really need the power for are those things you need to run that don't exist yet because there's not been powerful enough chips about for people to seriously consider writing them.
me.
Hmm.. I guess maybe its time I got rid of this old solbourne workstation with the 25Mhz sparc chip.. I wonder if I could just pop that alpha onto this MB.. Ha Ha.. Yeah, its fast- but can you heat your house with it? -
air and light and time and space
Chips are getting real fsat these days. Could someone tell me some justifiable reasons for home users to have this much speed? (I know its great to have speed, but I just want to know what other people's reasons are)starcraflinux!!
I do not believe that current IBM copper technology uses SOI wafers. Those processes are still in development, and if I remember correctly were supposed to be rolled out in a year or two in their ASIC fabs, and later for powerpc etc.
Just so you know, IBM didn't actually have much to do with the aptiva. The hardware is made by Acer, it just has IBM branding. IBM has kind of had a love/hate relationship with the desktop, as AFAIK they're not even selling the aptiva's through brick and mortar any more.
This Sig Intentionally left blank
Eagles may fly, but weasles don't get sucked into jet engines!
:-)
:-)
Hmmm... I always thought it was "Eagles may soar, but weasles don't get sucked into jet engines!"
Viva la buchni!!!!!
Oh, and for the moderators out there... I'm quite happy with this Copper thing... IBM really needs to put something decent on the market for a change . Yes, they are a great company, but you have to admit they've been slipping lately. Anyone who has ever owned an "Aptiva" shares my grief.
Oh well, at least they still have great mainframes.
I'm looking good replacement for my noname alpha ... God let it be this.. ok... ?? ;)
Amiga - so the world may KNOW!!!!
Not a thing would be my guess. Intel hasn't been anywhere near the top of the technology curve for many years. Nevertheless, they are immensely profitable. Ask yourself: what would you do? I would keep on hype, hype, hyping it up selling to gullible, brain-dead consumers until the consumers either kill themselves off in fits of idiocy or wise up. And I'd laugh all the way to the bank.
Careful with the POWER versus PowerPC, as they're not the same thing. The POWER architecture is really fast, even at low clocks. I remember that SpecFP used to have a 200MHz POWER3 chip in second place after the 500MHz Alpha 21264, ahead of the PA-RISC and everyone else. I think they've at least doubled the clocks by now, and I think they're in the POWER4 generation as well. They had a 64-bit version, the RS-II 64, and I believe they've also moved that ahead to the RS-III 64.
But I've never been able to really keep track of IBM's processors, it just seems tough to get info and keep it straight. Especially with the POWER versus PowerPC mixup.
-Paul Komarek
Otoh, they support linux but not Tru64 (no Tru64 PALcode)...
"My opinions are my own, and I've got *lots* of them!"
The (current) 21264's are aggressively out-of-order speculative-execution CPU's, and on typical "random" codes will execute more than twice as many instructions per clock tick as the '164 does.
"My opinions are my own, and I've got *lots* of them!"
"My opinions are my own, and I've got *lots* of them!"
If you're not part of the solution, you're part of the precipitate.
Consider:
Notably, Netscape is not readily available. (I know that the Digital UNIX version can be hacked into submission, but it is not an easy rpm -i netscape.rpm away...)
Linux may not be the rats-nest of spectacularly nonportable software that Windows NT is, but getting things like KDE, GNOME, GIMP, and such has been a fairly slow process.
I've tried running the Diamond Rio package, rio on my Alpha box; it gets quite confused, probably due to some 32-bit-ism in the "driver."
More critical is the paucity of graphics cards supported by XFree86 on Alpha. It is probably similar for other 64 bit platforms; you're restricted to whatever there are "open" drivers for, and there are some cards (Cirrus comes to mind) that have architectures that are distinctly unfriendly to 64 bit operations.
All of this adds up to Alpha not being a particularly consumer-friendly platform. People that "just want it to work" will find it somewhat challenging in this regard.
This is presently the case for Linux/Alpha; it was the case for Windows NT/Alpha.
The same will likely be true for IA-64, for a goodly while, by the way. The pains that Alpha has had may help IA-64 porting efforts to be less painful, but what "cramps" persist are likely to be irritants to adopters of IA-64.
In short, I don't think this was a likely niche, except amongst "daft hobbyists," which doesn't represent all that "nice" a niche...
If you're not part of the solution, you're part of the precipitate.
Heh, consider this:
:-(
[marcus@earth marcus]$ uptime
10:44pm up 331 days, 9:55, 6 users, load average: 0.14, 0.05, 0.01
[marcus@earth marcus]$ uname -a
Linux earth 2.2.5 #3 Sun May 30 13:10:30 CDT 1999 alpha unknown
That's an OLD alpha axppci33(266MHz, 256K cache, 64MB RAM, 6GB of disks) that does qmail, 2 apache servers(one being a devel/demo of java servlets), leafnode, ftpd, local DNS, PPP on three dial-in lines plus firewalling and masquerade for 6 boxes on the local LAN and the dial-ins 24/7 and obviously working on that 365 part. Its plenty fast as there is no problem saturating the 10Mb ethernet with apache(raw files) or ftpd.
Total cost of about $550.00.
Since the ADSL line is about to arrive, I'm going to have to take it down to add another NIC.
No, it does not run X, Netscape, nor any "office" software. Consumer platform? No, it's flawless just as it is.
Good judgement comes from experience, and experience comes from bad judgement.
- W. Wriston, former Citibank CEO
AMD has some SPECfp-95 numbers here. They show a 800Mhz athlon at 26.6 (P-III shown at 19.8), 900Mhz at 27.5, and 1000Mhz 29.4 They use to have some SPECint numbers as well, showing about a 15% lead over the P-III, but that was at a clock speed where the Athlon had the half speed L2 cache, and it was when the P-III also had a half speed cache (P-IIIs with the smaller, faster, more associtave cache L2 are a bit faster on many benchmarks then the older P-IIIs). So I don't know how the current P-III stands in int to the current K7. Probbaly not as good or AMD would publish those numbers.
The orignal poster wanted to know if the Alpha was competitave with lower clock speeds. The answers shown here say the 21264 is. I want to point out that the picture was somewhat diffrent with the 21164 and 21064. They are strictly in-order machines, and the 21064 has very strict instruction grouping rules.
The 21064 (frequently) does less work per clock then a P-III, P-II, and even frequently a PPro. Of corse the 21064 was new a very long time ago (8 years?). The 21064 did somewhat less work per clock then most of it's contempary RISCs. But it was clocked much much faster. (i.e. 133Mhz 21064 vs the 50Mhz TI SuperSPARC, definitly much faster then the 386s and 486s of it's day as well)
The 21164 did more work per clock then the 21064. It's contemparary RISCs were doing much more work per clock. Far more out of order dispatch and specultave execution in the other RISCs, and in the PPro which Intel brought out during the 21164's lifetime (I think). However the 21264 once again had a dramitaic clock speed advantage (2x or more in some cases - 3x to 4x over the Intels for a while).
The 21264 is a diffrent kind of Alpha. The first one to do a lot of out of order and specultave execution. Clock for clock it gets far more work done then the 21164. Which is good because it no longer outclocks many of it's foes. It has even lost the top slot in SPECint which the Alpha has held more or less since it's introduction (not at all times, but never lost it for more then a few months that I recall -- then again I didn't check every month!). Even the SPECfp it shows some weeknesses it hasn't in a long time (I recall it having a firmer grip on #1 here, losing when HP announced new HPPA machines, and to some of the RS/6000s, but allways regaining it fairly soon after).
Seeing a 21264 with a really fast clock will definitly be an impressave thing. It will strain it's 8M L2 cache at 1.2Mhz for sure. Probbably blow right past it. I wonder if they will have a much larget L2, or if they will use a big L3?
I also look forward to the 21364. I don't think Compaq has said a lot about it, other then SMT. Simultanius Multi-Threading, which looks similar to SMP on a chip, but it lets the CPU assign functional units on a per-cycle basis to acheve maximum throughput. I wonder what else the 21364 will do. I wonder if it will go back to a simpler in-order less-specultave execution model (using SMT to tolarate the memory latency).
I heard the 21364 taped out in December as planned. I don't know how the test samples were, or what the rest of the schedule is.
P.S. dispite never having owned an Alpha, I'm definitly a member of the "Alpha is fast cult". A pity it has allways been cheeper to ray trace using more CPUs of something else (the K7 last I did my estimates).
I'm a musician, and I would dearly love to have one of these things to set up a realtime synthesis/audio processing engine for jamming through MIDI. I consider that to be a "home use."
I used to do quite a bit with POVRAY and 3D modelling. It would be neat to have a machine that could render a frame in a few seconds (instead of 20-30 minutes). That would really cut down on the time required to develop a scene.
You could compile the kernel really fast! ;-)
It turns out that arithmetic operators cost half as many transistors/operator if the bit order goes from lowest-order bit to highest-order bit, as in an Intel chip, than they do if the bit order is highest-order bit first.
Having just finished my final year in Computer Engineering learning chip design, I can say with confidence that this isn't true. The implementation is the same no matter which order the bits are in (just use a mirror image of the layout).
You might also want to visit the EGCS site, which mentions Alpha updates to GCC as recently as March. GCC does support the newer EV6 architecture. The problem with Linux is more that large portions of it are hand-coded for the GCC assembler -- and optimizing kernel code doesn't do much since so little time is spent in the kernel anyway. ;-)
So IBM is fabbing alphas at 1.2 GHz, meanwhile the POWER line, which is their baby, seems to be stuck at 500 MHz or so (last time I checked). What gives?
BTW, an Alpha I purchased a few years back is STILL faster at floating point than the latest from Intel or AMD (OK, the Athlon is close).
And since FX32! (Intel x86 emulator) was outlawed (because dynamic recompilation violated software EULA's becuase it automatically reversed engineered the object code), I don't expect the Alpha to ever the viable in the end-consumer marketplace.
Uhhh...I don't know what you're talking about. Two things:
AFAIK (I worked on the VC compiler for Alpha) FX32 was never "outlawed". In fact, it was integrated into the Alpha version of Win2000.
Second, FX32 is a moot point since Compaq dumped NT as an Alpha OS. Even if FX32 was "outlawed", it wouldn't be why Alpha will never be a consumer platform. It would be because Windows is a dead OS, as far as the Alpha is concerned (boy that's fun to say).
Having said that, I don't think Compaq has any inclination or incentive to price/market the Alpha toward end users.
--GnrcMan--
Well, regardless of that, I worked with the team that makes the proprietary Compaq compiler and it truely is amazing.
But you are right, EGCS is making inroads as far as Alpha optimization goes(working on the Alpha compiler is my pet project(haven't submitted anything yet) now that I'm no longer working on VC for Alpha, which used the same optimizer, "GEM"), but GEM is the only way to go right now.
--GnrcMan--
www.alphalinux.org
The link is a vendor list, alphalinux.org is a nice place to look for info. Also check comp.os.linux.alpha .
You won't be able to get a new alpha for less than $3500US, but used ones are available for cheap, sometimes free.
Ace's has some benchmarks (under linux, no less) of the different compilers. They prove your point quite well.
-
__
Comment submitted. There will be a delay before you understand what you posted.
Are you using your Alpha box to make money? (Click here then scroll down to the tree.) Or are you just running the distributed.net client?
Will I retire or break 10K?
Man.. this is great! First off, hat's off to the Alpha. I've used several of these systems, and all kidding aside, a single processor 533 with 512Meg will happily run 6 quake servers! That's a serious beating.
I'm glad to see some of Intel's technology rolling over into the other processors. Hardware companies should take cues from the open source movement. If you see something you like, use it in your product and come out with something even better. I've gotten worn out from the Gigahertz wars, but some seriously beefy Alphas are always welcome in my place.
I do hope that some companies will take notice and start porting their games (like HalfLife) to the Alpha platform. They'd make totally superior gaming machines since the FPU is actually worth something as compared to Intel.
1.2GHz Alpha... 4MB cache... Drool!!! When can I get one?
-What have you contributed lately?
You'll probably be able to find a EV56 machine for around $2000.
I recently purchased a DEC Alpha DPW 433au (EV56 21164a/Miata mobo) from egghead.com for US$1,199.   They recently had them avail. for US$1,049 (I just checked and I don't see it anymore... they only had 4 left and they may have been snapped up).
What is impossible to find is a place that will sell you just a motherboard or just an Alpha.
Amen!   Although I've seen more places that will sell the mobo and parts like E.L.I or other places that will sell you "inexpensive" (in quotes - although considering, the prices aren't too bad) systems, like DCG, but other than those, it has been a pain trying to find an upgrade processor for less than I paid for the whole system!
I still like my little alpha though...   heh heh.
-- Win2k: "It's not so much that it's only 65,000 bugs, it's just that they stopped at 65,535 to prevent an overflow."
Did you check Compaq and Samsung's marketing company's website? Alpha Processor Inc. They have a list of system vendor's. You'll probably be able to find a EV56 machine for around $2000.
What is impossible to find is a place that will sell you just a motherboard or just an Alpha. If I could go out and buy my own EV6 and mb and build the rest of the machine around it, I would have done so a long time ago.
Bill
Does anyone have any comparisons of Alpha vs other processors at equivalent clocks? I remember comparing processors 3 years ago, when the Alpha was getting comparatively high clock rates. When you really looked at a ratio of Work/MHz (of course, the definition of "work" can vary tremendously), the Alpha wasn't very "efficient" in terms of how much it did in a clock. Of course, the was the entire intention behind RISC. Lots of clocks by reducing how much was done each clock.
But now with the Alpha unable to sustain a big clock rate advantage, does the Alpha get any sort of real performance advantage when compared to x86, or particularly other less "reduced" architectures like HP's, Sun's or IBM's?
--
Sometimes it's best to just let stupid people be stupid.
Right. Just the other day I was mulling over perhaps getting a medium to a low-range Alpha. And I reached a conclusion that the high price is the real problem with Alphas.
I think it's mostly has to do with the fact that Compaq, it seems, doesn't really want to market this hardware to consumers. Here I was, seriously considering one, but...
For starters, there doesn't appear to be a lot of places selling those things out there. I've tried searching the 'Net, I didn't find anything. I couldn't believe it when I searched shop.yahoo.com, which basically aggregates hundreds (if not thousands) of merchants' e-commerce shops under one roof, and came up dry.
After I went to Ground Zero -- Compaq -- I came away with a discreet impression that they really don't want to talk to a mere peon like me. Their web site, apparently, is really put together more like a business-to-business (or government) type of a deal.
That's really a shame. I think that I'm not the only one who might be interested in looking into alternatives to the x86 platforms. If Compaq only invested a small portion of that R&D that went into this newfangled chip into, instead, investing in marketing the Alpha to consumers, they might find themselves a nice niche amongst consumers too.
According to this EETimes article, high-frequency induction is a growing concern. It cites an academic paper stating "copper is much more problematic than aluminum when it comes to inductance...."
SAN DIEGO, Calif. -- The perils of chip design at 0.18 micron and below mandate new research in a number of areas, according to presenters at the recent International Symposium on Physical Design (ISPD) 2000 here. Calls were made for breakthroughs in areas such as power analysis and estimation, inductance modeling, analog layout, incremental CAD and process variability.
Beyond academic papers, this year's ISPD included a keynote speech by Aart de Geus, chairman and chief executive officer of Synopsys Inc. (Mountain View, Calif.) In his keynote speech, de Geus called for better estimation and analysis tools, and for attention to new problems such as inductance.
"Finally, after 25 years, inductance is coming back," de Geus said. "With high current, inductance does matter." But it won't be a big issue for most designers, he added, until feature sizes drop below 0.18 micron.
A paper given by Li-Fu Chang of Frequency Technology Inc. (Santa Clara, Calif.) outlined that company's research in inductance modeling of on-chip copper interconnects. Copper is much more problematic than aluminum when it comes to inductance, the paper notes.
Chang said the paper presents a "full chip" inductance modeling architecture. "Inductance modeling is much harder work than Rs and Cs [resistance and capacitance]," he noted.
How much do you want to bet that when I try to specify copper Alphas, that dimwit teenager in Academic Purchasing (who thinks he's 3l337 because he can remember the HardOCP URL 2 times out of three) will get me a stack'o'coppermines instead?
Oh, and think of all the lovely conversations you'll overhear waiting in line with that last Y power connector for your Beowulf as the wannabe Mr. Know-it-alls ("I order all my systems on-line!") tie up the Saturday salesclerks with impossibly self-contradictory orders.
_____________
If you can go to bed, knowing you did a valuable thing today, you're very lucky. If you can't... it's not bedtime
Well, it would seem that the future is in Copper interconects. I'm just wondering what is Intel going to do about it. Last time I read up on it they refuzed to license the technology from IBM.
Anything I missed? Is Intel going to go Copper, or are they just going to keep on naming their processors so that they sound like it?
Yes, these things will totally kick ass, but if you use gcc and glibm it would be like running 80 octane gasoline in a Corvette. This is a brand new architecture that the gnu development is years behind on. Compaq has compilers, libraries and prebuilt apps that run _much_ faster. I don't think anything has been done to gcc since the EV5 architecture.
The Compaq tools are mostly available for free, but you can't compile kernels with them. We're trying to catalog what can actually be compiled with them but most Linux system source assumes that you are using gcc. Sad, really. I'm sure that things would run much better when using a compiler that actually knows how many execution units the processor has...