Slashdot Mirror
Compaq May Nix Tru64 for Merced
Pivo writes "Sure I'd rather use Linux, and sure, "Tru64" is a cheezy name for a Unix variant, but for some reason I'm not happy to hear that Compaq may not release Tru64 for Merced after all. Nothing has been confirmed by Compaq however." But, according to the article, Compaq will keep supporting Unix on Alphas.
I love this guy. He must work for Microsoft. Since when was having "too many" choices a problem? We have linux, freebsd, openbsd, netbsd, solaris, tru64, sunOS, AIX, HP-UX, etc. All of them have niche markets. For example, netbsd is an excellent platform to build a firewall or intranet server on (good security), whereas linux makes an excellent server for a small-office setting (linux/samba - can't be beat). This analyst definately needs to get out more... he probably thinks NT and MSOffice are the only two products on the planet...
--
That would be like Apple porting the MACOS to x86 architecture
Another UNIX for Merced bites the dust. Seems like Merced will not be such a hot chip after all. Or is Compaq just shying away from the competition?
Perhaps after hearing the rumors, Compaq has decided that the best way to save themselves money is to delay porting their OS to another architecture, instead opting to wait until Transmeta comes out with that processor which can emulate other architectures. Certainly sounds cost-effective to me.
~ Kish
Or chip development.
Compaq has taken the OS provided them by Micro$oft and the CPU by Intel and made products.
From the DEC/Digital side, their Unix was not well accepted. The OEM mags were quoting that for every 10 users leaving VMS, only 2 stayed with Digital. (No one identified if it was DEC or its Unix as to why 80% left)
And, for Intel there are MANY Unixes, and a few non-Unix OSes. (PICK, THEOS, and some stuff from a company in Redmond) The Alpha choices are much less.
So, it does not suprise me to here that they are just going to keep working on what has already been developed...the Alpha product.
Because without a market for the Alpha processor, Compaq has alot of IP they can't get a return on.
If it was said on slashdot, it MUST be true!
Why would Compaq compete with itself?
Why would Compaq make a kick ass OS for harware it doesn't build
Why not keep Tru64 on the best 64bit platform?
It all makes perfect sense. I just want my slot a Alpha! (flash my bios too you know).
Let's see:
Irix will not run on Intel
Tru64 will not run on Intel
Win2000 will only run on Intel
So which OS will be able to run on more than one platform: Linux!
I have nothing against the other Unixes, even less against ths BSDs, but I think that after decades of fragmentation, the Unix world is coming together at last, in the form of Linux.
"..Terry Shannon, author of the Shannon Knows Compaq newsletter."..
"..believes it's likely Compaq will cut development of Tru64 on IA-64."
Wow, someone is sure going to look like a fool if they're wrong. :)
~ Kish
Why compete with themselves? According to the article, they already are planning to sell Monterey on their IA-64 systems. They probably figure that selling both TRU64 and Monterey for IA-64 would be redundant and a big waste of their money.
I think I see what they're trying to do:
Alpha-based systems: Digital branding (running OpenVMS or Tru64)
All others: Compaq branding (running Monterey or W2K)
although that might be a little too obvious to be correct.
Glückwünsche, haben Sie Slashdot ermordet, indem Sie zum korporativen Druck beugten und Subskriptionen einlei
can't stand when someone points out a strong point of *BSD
Actually everyday its becoming more & more obvious that Merced is really the Mckinley Beta. Even Intel has admitted there will be no performance gain with Merced & IA64 will only get into high gear when Mckinley comes out. Afterall, Merced partner HP even admitted it, saying that it's better to wait till Mckinley. Quite a few of the other vendors have also said they wont be ramping up their IA64 products till Mckinley hits the shelves.
You certainly can't be thinking I was serious..? That's pretty funny in and of itself. :) Oh well. It's always one extreme or the other.. when I'm direct, I'm the hammer slamming down into the nail, when I'm subtle.. "Hey, Joe, did you hear someone say something?"
~ Kish
"We have linux, freebsd, openbsd, netbsd, solaris, tru64, sunOS, AIX, HP-UX, etc. All of them have niche markets."
Well, all except those flavors of UNIX which are getting trashed because their developers are swinging their support over to Linux. :)
~ Kish
*BSD? Pick one, damn it. And define "everything". Will it run on my toaster? :)
~ Kish
To think I just got through bothering some guy who wrote an article for a BSD-related news site about his odd misrepresentations of GNU/Linux.. Care to define "light years"? For a "naughty upstart" it sure has done a good job is causing a few flavors of UNIX to fall already.
~ Kish
Sun and Apple used to have Mac Application Enviroment for Solaris/SPARC. That product, however, is no longer available.
All we have here is a couple of analyists saying they believe Compaq is likely to cut development of Digital UNIX[1] on IA-64. Nothing official. Nothing from Compaq.
Now, granted, "Shannon Knows DEC" often gave us great insights into DEC. However, Mr. Shannon also blew it many a time. He's in the business of making predictions, and like weather predictions, they aren't always right. It is also worth pointing out that while Shannon knew DEC, he prolly doesn't know Compaq all that well.
In short: This is much ado about nothing.
I'm not saying it can't happen, just that this bit of information is mere speculation from the outside, and should be taken with a large amount of salt.
[1] I refuse to use the name "Tru64". That is the stupidest name for an OS I have ever heard.
dragonhawk@iname.microsoft.com
I do not like Microsoft. Remove them from my email address.
The fact that it's dry is what makes it funny (to weird people like me and the moderator with good taste who agreed).. I guess you'd have to recognize me from my previous posts to realize I was making a joke, unless you're better at catching onto subtle things like that than I usually am myself. :)
~ Kish
Compaq was never big into the server market before. They bought DEC to get their foot in the door. MS drops support, Tru64 is the only solution other than going with Linux, which I love but many people still won't admit that it is a real, robust, stable, OS. For them to drop Tru64 would just be shooting themselves in the foot. Why would a company buy into the server market, then essentially just walk out of it.
Either it is a rumor or they need some serious strategic help.
We have linux, freebsd, openbsd, netbsd, solaris, tru64, sunOS, AIX, HP-UX, etc. Hmm... how many of those are produced by the same vendor? This isn't an issue of someone boosting NT and MSOffice. In fact, NT was only mentioned in passing in the article. The problem for Compaq is that they're faced with marketing and R&D costs for half a dozen different OSes. That means a large amount of wasted effort and confusion. Why develop Tru64 for IA64 if it will be a multi-million dollar effort that only brings in peanuts? Worse, what if its only effect is to replace systems that would've run Tru64 on Alpha, which would have given Compaq profit on both the chip and the OS? In the long run, I can't see Tru64 as a contender. It has to face down Sun, MS, Intel, Linux, and HP, all of which have superior resources and mindshare. But in the short run at least, not porting to IA64 is probably the wiser choice. --JRZ
OK, for years we've been hearing about this wonderful IA64 architecture, and that it's going to be the be-all and end-all of CPUs. Naturally, you hear a lot about Microsoft promising 64-bit support, "shipping on the same day as Merced", and how Merced is a critical product for them to scale NT up and invade the datacenter. Of course, this is to be expected, because Microsoft and it's customers are pretty much stuck Intel platforms, so this would be a natural move (especially with the IA32 compatibility built into merced).
But at the same time, you have the big UNIX vendors (Sun, HP, IBM, DEC/Compaq) announcing that they too are also going to support Merced. Which is odd because these vendors make their own hardware and CPUs. I have to admit that I'm confused at the strategy, which on it's face seems to bolster WinTel.
Are the UNIX companies using IA64 to slowly get out of the CPU business? or the hardware business in general? That would be an odd strategy because right now they're making most of their money off hardware, and that's where the main differentiation is right now.
What happens when ZDNet benchmarks all of the commercial Unixes on some Dell PowerEdge? Does Vendor X really want their customers to see that they are 7% slower than Vendor Y on the same hardware? Or are they going to lock it down so that Vendor X Unix only runs on Vendor X Merced hardware. If so, what's the point?
Maybe Compaq/DEC is the first company that figured out that Unix-on-Merced is a loser strategy, and there's more money to be made with their own CPUs and hardware. (You have to figure Compaq would know - they are certainly going to be the premire IA64 hardware vendor for the Windows folks.)
Business. Numbers. Money. People. Computer World.
On the other hand, it could push more Merced users to Microsoft's NT (NT is being developed for Merced is it not?). As we all know, if there's one thing MS doesn't need, it's a bigger market share.
I suppose all we can do is hope that the first viewpoint is the correct one.
--
--
"Insert witty quote here."
So why bother porting for Merced when Linux is the biggest selling 64 bit OS already?
And Compaq firmly supports Linux. If you don't believe me go and download the Compaq/Digital Fortran and C compilers for Linux.
Known as the best optimized that you can get, they now are fully supported on Linux Alpha.
Oh, almost forgot: They are FREE!
Enjoy.
Maurice W. Hilarius Voice: (778) 347-9907
>Um what has fallen so far? Linux killed what?
>Irix? Solaris? *BSD ?
Well, it could be argued that SCO has lost quite a few customers to Linux...
As for Solaris/x86, it never was a strong contender, so..
The way I parse the recent SGI statements about their future direction, they seem to basically leave the low-to-midrange server market to Linux (consider their donations to Linux: xfs, OpenGL stuff etc) and concentrate on the high end. After all, the money is mostly in hardware these days.
Anyway, Linux does not need to kill any other OS to be sucessful. It sure has brought back the spotlights on Unix, a turn of events beneficial to all Unices (well perhaps with the exception of SCO..) Not to speak of the whole Open Source / Free Software bandwagon.
good riddance to Tru64 on Merced. Was anyone really waiting for this anyway? Any IT guy at a company who would have forgone the blazing power of the alpha chip for the half-baked 64 bit product of Intel (who has brought you half-baked CISC x86 technology for many years now) is out of their mind. anyone willing to pay the liscencing fees of Tru64 should be willing to pay for the alpha platform as well. almost everyone knows that merced will only be the precursor to the more powerful mckinley and whatever comes beyond it, so if i were in charge of purchasing anything having to do with IA-64, i would either buy something else or wait for mckinley. and one more thing: besides the standard unix features, the main advantage to Tru64 Unix is the alpha platform and the alpha compiler, which builds extremely efficient code for the alpha. if compaq sells tru64 on unix, any compiler they ship with it will seem wimpy next to intel's compiler. even if compaq strikes a deal with intel and compaq ships the intel compiler with Merced Tru64, people will still be able to get the intel compiler for their platform without compaqs help.
porting Tru64 or any other proprietary UNIX (i.e., IRIX, HP-UX, AIX, etc.) to Merced just doesn't make sense.
p.s. I've noticed that Sun ships an intel version of Solaris right now and i'm assuming that they'll upgrade it to merced. does anyone know how this helps/hurts them (Sun)? does sun sell intel machines in addition to their sparc products?
Admittedly, the Compaq compilers are excellent (nothing short of mind-blowing, really) - however, the Compaq beta license is free in the Free Beer sense only. It would hardly qualify as Free Software or even OSD compliant.
It would be refreshing to see Compaq embrace the GPL (or an OSD-compliant) license a la SGI though I'm sure many software patents preclude doing this with their compilers...
Having the ability to roll this optimized code into gcc/egcs could have the ability to sell a lot more Alphas ;-)
only free as in free beer, i saw no indication
that it would ever really be free. (compaq
seems to have a problem with this IP thing)
a much better solution would be for compaq to make
the back end to gcc or have cygnus do it (with
full spec) that could solve the little compatibility issues the web site complains about.
Do you have any proof for the 80% figure or did you just make it up?
Compaq has designed a lot of "chips" or ASICs over the years. Until about 2 years ago they designed all of their own chipsets and various other support chips.
Um reliability, cost are two areas where linux is superior to many os's especially windows. Funny trolls never used to log in.
Speaking strictly for myself and not for my employer, I'd like to find the idiot within the Company who came up with the brilliant idea to take a perfectly good OS and burden it with such a stupid name as "Tru64".
Upon finding said idiot, I will thrash him soundly about the head and shoulders with an old VT100 keyboard.
:-)
Hey, I work there and I still call it "Digital Unix."
--
I gave my boss a reality check. It bounced.
The project was called "Star Trek" and ran on standard PCs. Apple actually demoed it to Dell.
. html
See Bob Cringely's article on it:
http://www.pbs.org/cringely/archive/nov697_text
"Compaq engineers quietly admit that Linux on Alpha runs better than Compaq's own Unix. So don't be surprised if Compaq kills its True64 Unix product in favor of Linux. Compared to free, it is hard to justify the True64 R & D investment, especially given the Titanic mindset inside Compaq these days."
0 2.html
http://www.pbs.org/cringely/pulpit/pulpit199909
Tru64's future may be limited on Alpha as well.
Compaq can't compete with e-Machines, yet I see Compaq PCs in Best Buy. Compaq can't compete with Dell, yet they desperately want to emulate them. Compaq, with Alpha and Tru64, can compete with HP, IBM, and Sun, but the "Not Invented Here" syndrome will prevent it.
I feel Compaq will never be the company it once was. Perhaps Compaq will port the clustering ability that they worked so hard to put into Tru64 into Linux, and become a premier Linux on Merced vendor. But I doubt it.
Don't forget that Compaq has been the leader in PC x86 servers for sometime, and that still is the real profit center for the company. Especially with Dell kicking their ass on the desktop. (Maybe some of you don't an 8-way Xeon to be a *real* servers, oh well,)
After spending a ton of money on Intel IA32 and IA64 stuf, buying Tandem and DEC, it would be very bizarre to drop out of the 'server' market. Without it, they're just another Gateway 2000.
Business. Numbers. Money. People. Computer World.
Seems you're right :-(
Even OpenVMS, i guess, won't be Digital OpenVMS anymore. Compaq OpenVMS just doesn't sound right.
In fact, Compaq doesn't sound right for anything that used to be put out by Digital. I would have thought that they would at least keep the name, if only for marketing reasons.
We'll miss you, Digital.
Glückwünsche, haben Sie Slashdot ermordet, indem Sie zum korporativen Druck beugten und Subskriptionen einlei
On Solaris GNU stuff mostly compiles without tweaking, on Irix it mostly doesn't. Heck, Solaris even provides a free GNU compiler that you can use for compiling your favorite compiler.
--
"L'IT c'est moi!"
From my limited experience with Digital Unix (supporting a professor's workstation) I can't see any real advantage to Digital Unix over, say Solaris, other than the fact that it runs on Alphas. It would seem foolish to port the OS to a radically different architecture that doesn't improve on the Alpha's strengths. Does anyone else have any insight in the matter?
--
"L'IT c'est moi!"
Maybe true64,which is OSF based, might be more difficult to port to merced. After all isn't OSF based on the mach kernel and therefore an entirely different beast when compared to svr4 unices? Wouldn't that mean unlike ibm and sco they would have to go it alone in the porting? Wouldn't that mean it would be up to OSF to do the work to get their kernel ready for merced(or have I lost track of something here?)
Joe
jlrice@columbus.crosswinds.net
The licensing fees aren't what's so costly. It's the support contract that can cost you an arm and a leg.
CUR ALLOC 20195.....5804M
Compaq already has a 64-bit chip in the Alpha. Intel apparently has been having loads of problems with the Merced chip. It's the first 64-bit chip out from Intel. If you know their track record, you know why I won't have a Merced.
Compaq is streamlining. Rather than make the large initial investment in IA-64 and EPIC (which will require a large investment to just port over everything DEC wrote and get it to actually compile and run in an optimized manner, yet alone work right), they are making what they have work.
IA-64 will not be in most data centers for a few years until McKinley is out. Intel has a nasty track record with first release chips, especially the Pentium, 486, and 386.
Why build on a foundation of sand, which I consider IA-64 to be until it is tested and working, and not a Merced chip, when you have a 64-bit chip that is running in very high-end data centers already?
Compaq is doing the right thing in this case. This is a case much like AmigaOS, where the hardware and software are very tightly integrated. You just can't recompile Tru64 for another chip. Tru64 takes advantage of a lot of hardware specific to the Alpha chip and the servers itself. It's reliable, and it works extremely well for HA environments.
Having Tru64 on Merced would undercut Compaq and could possibly force them to not make Alpha chips. While this is something intel wants, it's not a good thing. Having Tru64 on Merced could cause Intel to become the Microsoft of consumer and server chips, because it would eliminate a great competitor.
I personally enjoy running Linux on those chips too, since they are just so well-designed and built to scale, unlike the x86 family.
I applaud Compaq (if this is true) for avoiding the train wreck that's going to occur when IA-64 comes out. You're going to see so many issues dealing with how IA-32 apps run on it in emulation mode (which is slow), and how compilers will have to be reworked to use EPIC to actually get the speed increase. x86 is a nasty chip design compared to the Alpha or MIPS chips.
If IA-64 catches, I would not be surprised to see Compaq offer it at all. However, with anything new from Intel of this magnitude, it is better to avoid it. They are sticking with their 64-bit design and making it better, and I applaud them for bringing competition to IA-64 in.
Yes, Digital Unix, aka Tru64 Unix is based on the Mach 2.5 (or thereabouts kernel).
But this doesn't mean it's hard to port, quite the opposite. When I worked at CMU, I had accounts on Mach machines running on many types of hardware: VAX, Sun3 (680x0), Sun4(Sparc), PMAX, Omeron, Alpha, and probaly others I forget. Mach is pretty darn easy to port. There's very little that is machine dependant. It may be the easiest OS to port in common use today (probably only NT comes close).
Note he said supporting the merced, so OSes running on other processors are not included. Of course, he's still wrong since he left out Windows 2000.
Given limited resources (the usual case, certainly Compaq's case), it is always in the best interest of business to support fewer, more profitable options. If it were otherwise you'd be running a charity, and not a profitable business.
Seems like a sensible business decision on Compaq's part to me.
- They have their own 64-bit architecture that is well established
- When you're offering a proprietary OS it makes sense to keep down the number of platforms you support
The more interesting decisions for them are:
1. Should they continue to push NT on Alpha?
2. Should they sell and support NT on IA64?
To a large extent these depend on which way Microsoft and the PC software industry jumps - you don't buy software to fit the hardware, you do it the other way. If Microsoft pushes IA64 as the main platform, then that will force Compaq's hand as far as NT goes. However, the widespread use of NT on non-x86 platforms which was touted when NT was new has failed to materialise, with the excpetion of a small Alpha following, and I think x86 will be with us for a while yet.
What I'd be interested to know is how the memory bus and cache coherency work on multi-cpu IA64 machines intended for the NT server market. If they are not significantly better than the current abysmal Pentium architecture offerings then in conjunction with the porting/emulation issue and general teething troubles there is a big incentive just to stay with x86 based hardware, which is what many shops will do.
AMD is a small but significant wildcard here - I guess they'll be forced to bet the farm in the short to medium term on the x86 architecture, which Intel will leave open to them only when the pickings become slim. If Athlon's bus lives up to its server performance promises, there is even the possibility of Microsoft and Intel both wearing big albumen face packs, with a lot of NT shops opting for NT4 on Athlon and bypassing IA64 and Win2K for the time being.
Just because it's POSIX compliant doesn't make it UNIX. UNIX flavors are flavors of UNIX because they are derived from UNIX. GNU/Linux is not derived from UNIX, although the BSD derivatives are. *shrug*
~ Kish
Well, until you finish trolling and answer my query, which I posed first, I'll feel free to leave you in the dark, kid.
~ Kish
>Unix world is coming together at last, in the form of Linux
It would be nice to think that. And the people who TRIED to bring at least the x86 op-code Unix together over at www.86open.org drink the same kool-aid
But, the 'one Linux binary' concept is a market FAILURE as of this time.
Why?
Because **YOU** the consumer of shrink-wrapped binaries are not asking for binaries that run ANY Linux implementation, but instead ask for a linux binary and then tell the company what distro they run. So the company makes the binary for RedHat. So, instead of being seen as part of one big happy world, "if you arn't running the same distro as I am running you are my enemy" is the attitude.
Each of the distro-vendors want thier product to make them money. And the users of distro X don't want their product of choice to go away. Hence the radical 'My distro is better than your distro' or, heaven forbid, you run Linux binaries on SCO, Solaris, or BSD.
As long as there are so many different distribution vendors, all wanting to have a makret difference, the LSB effort will continue to be a failure. For a working LSB makes the 100+ distros "the same" wrt 'linux binaries'.
If you are wanting 'world domination' for Linux, you had better START asking for shink-wrapped binaries that will run on SCO, Solaris, BSD AND ANY LINUX distro. **YOU** the binary consumer are in the driver seat. Until the concept of exclusion is changed to INCLUSION, this world domination shtick is a bust. BSD/SCO/SUN have met you 1/2 way. They have Linux binary modes. Why won't *YOU* as the consumer of shrink-wrapped binaries work to be sure *EVERYONE* is included. The world is everyone....enclude them!
If it was said on slashdot, it MUST be true!
This may be good for Linux. If they do what SGI is doing and start moving towards Linux there could be some more drivers for Linux, and they could add in the stability and security of the kernel. Who knows maybe they work on the port of Linux to the Merced instead.
As I have said before, and Richard Stallman also has said: it looks like Linux is uniting the various *NIX flavors. lxrun under Solaris, the ports under FreeBSD, SGI dropping IRIX, and moving to LInux, (Compaq moving towards Linux it seems).
If the *NIX flavors unite administratino of the various *NIX system will be easier. IE rather than have to learn the AIX way, the Solaris way, the Linux way, the BSD way, there will be one unified administration method. Yes I realize that the systems are nto that different, but they are different enought that there is a learnign curve for administrators moving from one to the next. By a unified administration method there would be no learning curve, or a very minimal one.
Only 'flamers' flame!