Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 


Forgot your password?
Close
typodupeerror
×
Technology

HP & IBM Unveil New Chips 69

Posted by justin++ from the let-me-guess-"beowulf!" dept.
In the CPU market, both IBM and HP have new offerings. The first chip is IBM's 500-MHz PPC 440 for embedded systems, etched at .18 microns, and the second is the HP PA-RISC 8600, which uses the same core as the 8500. The IBM chip is for embedded applications, while the second is for workstations. The HP debuts at 500 MHz, and will soon be followed by the 8800.
This discussion has been archived. No new comments can be posted.

HP & IBM Unveil New Chips More

HP & IBM Unveil New Chips

Comments Filter:
  • well, isn't that what intel does?
    furthermore, once a chip has already begun to get old, it will not stay at "consumer item" level for such a long time. some of those skip the "consumer" stage entirely.
    watch the consumer market behavior - they always buy the newest... pentiums and clones.
    why? because those are designated for the consumer market. they will not buy a comparable, 2-year-old chip. and since those would be out of production by that time, they'd probably still cost more, not come with support, etc etc.
  • I would love to see HP make a graphing calculator out of one of these babies! Imagine the games one could play if they improved the LCD as well!

    --KF6NUX
  • Actually, there is an excellent reason to not so much *move* MacOS to IA64 as to port it to that platform, the same way that Linux is ported to different platforms

    If Apple brought the MacOS to IA64, and did it before, say, Microsoft brought out Windows for that platform, then it would have a lot of people seriously considering whether Windows is worth the effort on IA64.

    It would be very interesting to see the OS battle that would be waged between the two sides if this happened. Unfortunately, given its success in the past with this type of thing, I fear that Microsoft would win. Oh Well.
  • It's interesting that you should bring up the question of porting Mac OS to the Intel platform. Many years ago, Apple wrote a version of System 6(I believe) that would run on the x86 architecture. Whoever was running the shop back then (Sculley, I believe), decided _not_ to ship it to compete with Windows. I don't know why Apple didn't ship this ported OS, but I am certain that it would have beaten out Windows 2.0 (the Windows state-of-the-art at that time). Now, however, Windows is approaching the usability and stability of Mac OS, and I doubt that the user base will be easily shaken (especially because Microsoft would pull tricks like not writing Office for the new OS). All that being said, if OS X is ready before NT 5 or Win2000, or whatever they're going to call it, for the IA-64, it would definitely stand a chance of taking over the server market. Time will tell...
  • SomeoneElse wrote:

    "For a 'Next Generation' system to include ISA slots is a joke. Why not build it on a MCA bus while you are at it."

    Are you somehow under the impression that MCA was/is an old-fashioned and inefficient bus architecture?

    If you are, then you are sorely mistaken -- it just got beaten in the market, even though it *was* technically better.

    And in one market it's still alive and well: It's the bus architecture IBM's R/S 6000 RISC workstations use -- quite successfully, AFAIK.

    Your statement is somewhat akin to saying "For a 'Next Generation' system to use Linux is a joke. Why not base it on NeXTstep while you are at it."



    Christian R. Conrad
    MY opinions, not my employer's - Hedengren, Finland.
  • No. I refuse to call things gay names like "itanium". I'm sick of product names like "GeForce 256" and "Athlon". From now on, I'm going to call it by its codename: NV10 and K7.

    -Warren
  • How many times:

    Linux doesn't run on 286's - it requires a multitasking processor which a 286 isn't capable of.
  • Not true - That was just something that someone said and then floated around /. Intel has said that Microsoft does have a working version of Win64 that will boot on the Merced prototypes. I thought we cleared this up awhile back.

    -----------

    "You can't shake the Devil's hand and say you're only kidding."

  • Actually workstation prices have been dropping to resonable levels.

    I'm not convinced that 21264 systems are cheaper than the HP's. HP-UX is a miserable Unix? I'm just a user but I am still getting work done on a very old 735 which has uptimes that have corresponded to power outages in my building for years.

    As far as performance HP's 8500's were comperable at lower clock speeds with the 21264. I'm not sure what the 8600 will do for numbers or price but....

    From Microway's site Alpha Screamer 21264 @500MHz
    SpecInt95 28.6
    Specfp95 42.9

    From HP's site on a

    C3000 (8500 @400MHz)
    SpecInt95 30.3
    Specfp95 48.6

    B1000 (8500 @300MHz)
    SpecInt95 22.4
    Specfp95 38.8


    Prices?

    (Compaq's site was very slow)
    Microway 21264 -> 8,995
    HP C3000 -> 13,475
    HP B1000 -> 7,795

    The hp c3000 has more ram, hd, graphics than the b1000 or microway.

    I've also heard of extreme discounts on the HP's from my sys admin.
    I like the idea of an alpha heating my feet at better prices but the 21264 alpha's are not that compelling.
    And if you read that article, the 8X00 HP's are heading to clock speed in Alpha teritory.

  • Nice, are we really making progress? (Score:3)

    by SomeoneElse ( 90418 ) on Thursday October 07, 1999 @09:27PM (#1630330)
    Sure these chips are nice -- better performance is always great I suppose. But I think we've reached the point where before designing yet another CPU with yet another instruction set we should go and change the design of the PC a bit. If the Itanium includes any of the following, then it speaks for itself on just how crappy the PC is regardless of speed.

    1. IRQs. The IRQ controller in most PCs was designed by IBM if I remember correctly, circa *1982* or so. If we have to recompile and redesign all our apps and OSes anyway, can't we go ahead and redesign the other aspects of the circuitry to drive this thing?

    2. Don't tell me this thing will have a AT style keyboard port. Or a PS/2 port for that matter. Something that is supposed to represent bleeding edge technology built with a keyboard controller designed around 1980. Bravo Intel.

    3. ISA: don't even get me started. For a "Next Generation" system to include ISA slots is a joke. Why not build it on a MCA bus while you are at it.

    4. Floppy drive controller. Enough said. A 1.44 MB drive in a system attached to a 80 GB raid array. Great for backups, right?

    The sad thing is that all these wonderfully annoying and archaic antiques will likely be part of the "Next Generation" PCs built on Intel architecture. Says how far we've really come, doesn't it?

    Just my .02
  • or at least, it's really close to. If you want to help the Linux/PA-RISC port (you can get a 712 really cheap), see The Puffin Group's Linux/PA-RISC project [thepuffingroup.com].

    Though we'll need a lot of development time (the kernel is close to running, but glibc, binutils and the rest of userspace will need work), testers will be of help, too.

  • I am typing this on a dual celeron box running win2k rc2. with this motherboard, I could elect to use a USB keyboard, and IDE Superdisk (1.44/120mb floppy).

    This board has two unoccupied ISA slots. Intel and MS's pushed PC 99 spec already calls for the elimination of ISA slots. Over the last year, motherboards have moved from a typical 1/4/3 to a 1/5/2 format for the most part (AGP/PCI/ISA). Currently there is little demand for more than 5 PCI slots, which would necessitate an additional bridge controller (read cost), so they just are not being made. I am in the minority as someone who would like 6-7 PCI slots (right now I have NIC, SCSI, sound and video; I expect to get both an AGP card and a DVD decoder card which would leave me eventually with only 1 pci slot free).

    I love pc hardware because of its commodity nature. So much of my stuff trickles down from box to box. So, I am willing to accept the slow rate of change and the excrutiating time it has taken to kill things like ISA.

    matt
  • Why post this here? Is this on the topic? Who cares that you can't setup a computer in the correct fashion.

    We're getting faster and faster CPU's (almost 1GHz x86's), yet we're still basing it on 1980's technology. Let's put a nifty computer on a rear-wheel drive, carburetor car with all-drum brakes and no power steering!!!
  • I have no isa slots on my motherboard(its an intel i810 chipset) and I doubt that the intel 820 boards will have any isa at all.
  • nevermind, it worked after crashing 5 times..
  • Then again, he has a karma of 6.

  • PA-RISC and iTanium (Score:3)

    by Microlith ( 54737 ) on Thursday October 07, 1999 @06:50PM (#1630341)
    iTanium, the cpu with the cool translucent die. Comes in 5 fruity flavors.

    Anyways, since HP had a hand in with Intel in designing the Merced, will it also be able to emulate PA-RISC based software in addition to x86 software?
  • With Both IBM and Motorola moving into producing embedded PowerPC chips, What does this mean for Apple, who rely especially on Motorola for the CPUs in the current generation of PowerMacs?

    Or will Apple just move to the IA-64 architecture?
  • Be serious and name a high end chip that DOESN'T run Linux. Good Luck.
  • LinuxPPC should run on the IBM chip with no problem. However, the port to HP PA-RISC (FYI PA=Precision Architecture) is still in the works. You can find more info at the Puffin Group page here [thepuffingroup.com]
  • VA linux will see to it that by the time The Public gets silicon, linux will be running on it. The day you can order a merced system from dell with win2000, you'll be able to get a redhat-based merced system from VA.

    -Chris
  • Merced == iTanium...




    Anyways, they said that Linux was booting and running on Merced, and Win64 was not, coincidentally.




    If I'm wrong please post additional info here so I may learn.
  • How do these processors compare to the compaq alpha 21264's. This is the real 64 bit competition. Hope their price structure is reasonable as well.


    -- Moondog
  • suprisingly accurate too. it's not offtopic, it should be moderated as funny
  • Yeah, but who would use linux over irix, most people in the sgi community wouldn't(at least the ones i've talked to). Linux is not always best for everything.

  • Re:PA-RISC and iTanium (Score:3)

    by um... Lucas ( 13147 ) on Thursday October 07, 1999 @07:10PM (#1630355) Journal
    I believe either Merced or McKinnely will have NATIVE PA-RISC support, rather than emulated support as is the case of x86... Either way, it's cool that they're progressing...

    A while back, I did some calculating of chips based on their SPEC performance, and MHz for MHz, the PA-RISC series is the fastest line of chips. Alpha wins in the end because it uses much higher clocks, and now with the 21264, it's actually accomplishing more per clock, but still, HP's chips clobber all others if they're all at the same clock speed.
  • I doubt it has an FPU... the die is 4mm^2 compared to the PPC750 which is 40mm^2. That's 100 times smaller than a G3. Not much room for anything but a basic PPC core. Still would make a great processor for a small handheld or portable game system.
    --
  • 1 watt? So, will this thing not need on of those power sucking fans on a heatsink?
    --
  • The IRQ controller was originally designed by Intel for the 8080A CPU. IBM just used the commodity support chips that were available at the time the original IBM PC was designed.

    ISA slots are still useful for simple I/O cards that don't need the features, expense and complications of PCI.

  • I guess my sysadmin is just a god then. Nice going matt!
  • 500MHz at 1 Watt
    The strongarm only uses 1/4 watt, but its only 110 MHz or so.

    It says its based on the powerPC core, so i wonder if linux would run on it ?
  • HP entered into some form of agreement with either Apple or Carnegie-Mellon in order to gain access to, or share private advances in, the Mach microkernel and/or MkLinux. Since Linux is GPLed, I assume it was for the microkernell, but regardless, if HP's had engineers on Mach for over a year, then at least a variant of MkLinux should run.
  • Computers (and thus, computer chips) are no longer only a luxury product.
    and again, a new chip is released, with high-end designation.
    it's interesting to see how many chips are low-end designated, and how much of them really are low-end.
    when a company designates a "low-end", 2999$ chip, it doesn't constitute as low-end for me.
    i'll most probably never going to get use that kind of chip.
    that's a real pity, isn't it?
    my current tally of lowend chips counts only a handful of vendors, while there's a lot more.
    why can't we get better computing at more affordable prices, than seeing companies release yet another high-performance, high-end, highly-overpriced chip??
    also have a look at the old post about the ultra-5 "lowend" machine.
  • Umm... Motorola has been doing embedded PowerPC from nearly day 1. They have not been exceedingly popular due to cost & complexity. And, of course, Motorola makes a *lot* of money off of other embedded stuff, like HC1x, HC705, and Coldfire. IBM won't give two shits, IMHO, since they seem to shy away from embedded products.

    So, why would this affect what lives in Macs? 'splain!

  • Re:PA-RISC and iTanium (Score:4)

    by Guy Harris ( 3803 ) <guy@alum.mit.edu> on Thursday October 07, 1999 @07:15PM (#1630368)
    Anyways, since HP had a hand in with Intel in designing the Merced, will it also be able to emulate PA-RISC based software in addition to x86 software?

    Not by itself; the code would be translated to IA-64 code by software. This part of HP's versionof the IA-64 Application Instruction Set Architecture Guide [hp.com] says:

    Binary compatibility between PA-RISC and IA-64 is handled through dynamic object code translation. This process is very efficient because there is such a high degree of correspondence between PA-RISC and IA-64 instructions. HP's performance studies show that on average the dynamic translator only spends 1-2% of its time in translation with 98-99% of the time spent executing native code. The dynamic translator actually performs optimizations on the translated code to take advantage of IA-64's wider instructions, and performance features such as predication, speculation and large register sets. In addition, if an application has been aggressively optimized for PA-RISC, some of the benefit of the optimizations will carry over to IA-64. In fact, an aggressively optimized PA-RISC application may actually perform faster on IA-64 using the dynamic translator than the same application recompiled at a low level of optimization on an IA-64 compiler. Of course, the best performance will result from a high level of optimization using a good native compiler.

    The dynamic translator is designed to run all non-kernel intrusive code, handling both 64-bit and 32-bit instructions. This means operating systems and device drivers typically would not be supported, but all other applications will run. HP's dynamic translator will be bundled with all versions of HP-UX sold for IA-64 systems. When HP-UX encounters code compiled for PA-RISC, it will automatically and transparently invoke the dynamic translator which will allow the code to run on IA-64 without any intervention. Correctness of the dynamic translator has been verified with the same testing regimen used to validate PA-RISC processors.

    If I remember correctly, HP used binary-to-binary translation to move code for the stack-based 16-bit HP 3000 machines to the PA-RISC-based 32-bit HP 3000 machines, so they've done this before.

  • Like I said in a different comment, prior to the 21264, PA-RISC's descimated Alpha's in terms of SPEC perforance per clock cycle... Alpha made up the difference with higher speeds. With the advent of the 21264, which is much faster at a given speed than the 21164 (I don't use Alpha's, am not a chip person, and therefore don't know why - i presume more pipes and a larger cache) the Alpha makes up some ground on the performance per MHz front, but more than makes up the ground due to sheer speed.
  • IBM and Motorola are simply expanding, they aren't moving away from the desktop market. Motorola has a roadmap out which includes the G5. AFAIK, there are no plans for Apple moving to any other architecture, although the BSD base of MacOSX (Darwin) would probably make it very easy.

    --
    Harvey
  • Be encouraging.
  • Alphas come out slightly faster, as was mentioned, due to clockspeed. PA-RISCs are faster clock-per clock (tho 21364 hasn't come out yet...)

    But Alphas are very different machines than HPs. Alphas are pure number crunching machines. I don't think anyone's using HPs who doesn't have legacy reasons to do so. HP-UX is a pretty miserable UNIX, and the machines themselves are quite expensive (moreso than Alphas, I believe). Alphas are relatively cheap as workstations go. HPs tend to be popular in engineering applications, where the people have been using them for a long time and are very comfortable with them.
  • hmm, a high end? The SGI port doesn't work well, and guess what, SGI gave it to the community in '96, and almost nothing has happened.

    Really, any high end chip is supported, because well, its powerful enough to get attention. That doesn't mean the port is good, though, or even functional. There are many times where you need to try both NetBSD and Linux on a platform in order to determine who you want to go with, because both vary.
  • Ehh, not even worth it. I love bigass RWD cars too. I had a '71 Challenger w/ 360 4 speed. I do all my own mechanic. Loved that car. But I still wouldn't drive it everyday! That's what FWD shitboxes are for!

    I never considered ANY 455 to be a real engine, especially those slugs they'd put in those 6.3 TA's. 302 Furds are like belly-buttons: everyone has one. How genuine is that?

    So according to your post, you probably made up a 32-way 286/8 SMP box to run Linux? With 128 X 1MB SIMMS? heheh and 100 X 40MB Western Digital Drives? Those were the good old days indeed...
  • Nice site.Thanks to the well trained midgets. OK-how do you rate the posts?I ask because I'm interested in top marks.;-)
  • You're missing something. G4 is still a 32bit PowerPC processor. Maybe you are thinking of their graphics chips, which are ATI Rage128s.

  • Re:PA-RISC and iTanium (Score:4)

    by Guy Harris ( 3803 ) <guy@alum.mit.edu> on Thursday October 07, 1999 @07:29PM (#1630384)
    I believe either Merced or McKinnely will have NATIVE PA-RISC support, rather than emulated support as is the case of x86...

    "Native" and "emulated" in what sense? Merced (and, I suspect, McKinley) will be able to directly run x86 code; in what sense is that "emulated" rather than "native"? (The latest Microprocessor Report has a story on Intel's presentation on Merced at the latest Microprocessor Forum; it says

    Sharangpani's presentation shed little new light on Merced's IA-32 portion. As disclosed earlier, IA-32 code and data share the same caches and execute in the same function units. [Presumably "same" means "same as what IA-64 code and data use". -gh] When in IA-32 mode, the processor fetches x86 instructions into a separate decoding and scheduling unit that reorders the instructions and executes them using the native execution core. [The article doesn't say that they'll be translated to native instructions. -gh] We expect the IA-32 decoding and scheduling unit to be similar to Pentium III's front end.

    As for PA-RISC code, HP's IA-64 documentation online (see my other comment in this thread for a reference) says that the chip won't execute PA-RISC code; software will translate PA-RISC code to IA-64 code, and the IA-64 code is what will be executed. In what sense is that "native" rather than "emulated"?

    A while back, I did some calculating of chips based on their SPEC performance, and MHz for MHz, the PA-RISC series is the fastest line of chips.

    But if it's easier to make, say, a 21264 run at a given clock rate than it is to make a PA-8500 run at that clock rate, "Alpha wins in the end" regardless - SPEC/MHz isn't necessarily a figure of merit in and of itself.

  • Today's high end chips are tomorrow's consumer items and the day after's scrap. No-one introduces new chips at the consumer level.

  • Or will Apple just move to the IA-64 architecture?

    Are you kidding? Not only would it entail a huge amount of effort to move to IA-64, IA-64 would be a huge step backwards from the G4. There is literally no reason for Apple to move to IA-64.

  • iTanium, the cpu with the cool translucent die. Comes in 5 fruity flavors.
    Don't forget the RISC-K, for accellerating your porn downloads...

Slashdot Top Deals

To write good code is a worthy challenge, and a source of civilized delight. -- stolen and paraphrased from William Safire

Close