Follow Slashdot stories on Twitter

 


Forgot your password?
Close
typodupeerror
×
Hardware

3rd Party PPC Machines from IBM specs 230

Posted by Hemos from the buildyer-own dept.
florian writes "Macweek is reporting that Prophet Systems, a division of Eternal Computing, is building a sub-$1,000 CHRP desktop system. " Looks like an excellent deal-they are aiming it at Linux and Darwin folks, though I'd like to see the PPC version of Be helped out by this as well.
This discussion has been archived. No new comments can be posted.

3rd Party PPC Machines from IBM specs More

3rd Party PPC Machines from IBM specs

Comments Filter:
  • Mac users who have interestes in darwin and mac os X are not the bread and butter of the apple universe.

    Splitting themselves into two companies and opening competition by forcing software makers to lisence os goodies from the hardware company whould probably of made the physical platform far far more successfull.

    Maybe the sofware companie would of died.
    woopie. Mac os X is just NeXTSTEP with a (IMHO UGLY)facelift. I'd rather have NeXT alive than Apple Software Co. Mac OS itself, well.. sucks, and even apple knows it.

    In the mean time, they've taken a hurky jerky muddled buisness plan mixed up with software and hardware that haven't been significantly ahead in years (and have probably been behind for a couple now), significantly overpriced it (even now, though they're getting better), and decided that they invented sillicon and have sued everyone in sight while fiercely defending a niche market that noone can quite define with users more rabid than even the true linux shock troopers out there.

    I love their hardware.
    I hate their prices.
    Their "buisness" is too stupid for comment.
  • No, it just opens more doors to less populated rooms. But at least you help move the crowd into unventured territory. The only reason vendors support x86 for the most part is because, for the most part, thats what people use! The PPC is a superior chip (lets not argue and just take it as a premise of this discussion), and so I'd like to support a movement to accelerate it's adoption into the computer community. I mean, if your argument is valid, why have so many people switched to Linux over the past few years? Less vendor support, but superior (again, presumably) platform! Venture forth and the vendors shall follow!
  • That may be true for now, but the more people who buy and use the machines, the more likely good programs will be ported to the platform. The PPC Linux distribution has enough software on it now to make a kick-arse Web server, or SMTP, or DHCP, or all the above all at once.

    I agree with SirSlud. Such boxen make great second machines. Can you imagine how well the Gimp would run on it?
  • I can say that when I worked for Palindrome (now defunct tape backup software company), we had NT Alpha, and NT PPC ports in the works for Palindrome Storage Manager. Microsoft's wishy-washiness for the other platforms strongly aided in the destruction of those projects. Sigh. In those years, it really looked like MS WAS going to take over the world, and even our Unix port was crushed in an act of PHB-ness, and our ROOTS in Novell were rapidly disintegrating (although that was more Novell's fault - crappy developer support).

    Remembering those years - 96, 97. . . things really are better these days. Though I don't really care if Novell makes a comeback or not, it certainly is refreshing to see this Linux movement grow and take hold, and even see Macintosh recover somewhat.

    "The number of suckers born each minute doubles every 18 months."
    -jafac's law
  • In Research Triangle Park, just like Red Hat.

    Wait, so is pharmaceutical giant Glaxo-Wellcome ... they must all be working on some kinda computer-based drug thing that will ruin us all ...

    Yeah, IBM has been there for years. RTP is a development zone, the government of NC gives 'em incentives and also tech-types flock there ... umm, here so there's a good labour pool. It's little more than a coincidence, K?

  • Last I knew, they tried to propose that to the cloners, but they all rejected the idea of actually paying enought to cover their costs...
  • Quality software? 0% on a Mac? BullSH*T ..what are you Be peeps smoking? I don't think anyone has made a Photoshop 5.5 / ImageReady 2.0 bundle for Be.. does Starcraft run natively on Be? Oh damn! Does Be have Poser 4, Flash 4, Adobe GoLive 4, Connectix Virtual Gamestation? I think not! Get your head out of your ass before you post highly explosive flamebait.

    -----
    Linux user: if (nt == unstable) { switchTo.linux() }
  • Isn't NT & Linux "comsumer" enough? Though I just read on PC Week that Microsoft is going to themselves suspend development for Alpha past the next service pack... story to come
  • Not true. It had less to do with Microsoft than you think. From what I've gathered, the NT port to PPC was killed because of several reasons, none of which were related to "billy bob wanted it"

    Basically, it boiled down to Motorola and IBM not wanting it bad enough.
  • You can't keep piling crap on top of a deadening architecture. Eventually, it becomes more economical to start from scratch or at least a more recent starting point.

    That's what Be tried to do, and look at how popular they've turned out. That's what Apple originally planned with Rhapsody, and they had to back track on that in order to convince developers to update their products. MS Windows 2000 has been plagued with compatiblity concerns...

    Given MS & Intel's revenues and profits, it's a fair statement to say that the computing world values compatibility over performance/reliablitly/price/whatever other concern.
  • What Can't a Mac do? Well, the main BIG problem is that it cannot run a given network protocol over more than one hardware type. i.e. Appletalk cannot be run over Localtalk and ethernet at the same time. This makes it hard to share certain devices on a network. Also, the networking could be vastly improved on the MacOS.

    These CHRP machines look wonderful. I would be VERY interested in obtaining a 4 or 8 way PPC CHRP system.
  • Need to look more carefully. Thanks
  • Not all of them...

    Applix and Loki are shipping PPC versions.

    Applix is shipping Alpha versions.

    It really is a nice indication of what companies are really Linux companies, or want to be, versus those that are supporting Linux out of desperation (like Corel).

    Once they've got a working Linux version, they really don't have any excuse not to have Alpha and PPC versions.
  • That was all stated in relation to commerial/proprietary software. I don't think that all the software I want to use or their equivilants will be available in source code form at any time in the near future, if ever.

    Like always, I myself advocate using the best tool for the job, even if it means paying for it

  • It's more than that (Score:3)

    by binarybits ( 11068 ) on Tuesday August 24, 1999 @02:34PM (#1727769) Homepage
    Mac OS X Server is just that--a server OS. It is little more than a modified version of the Next OS with some network apps like Apache and WebObjects bundled in. It is not suitable as a consumer OS.

    OS X Client is the future of the mainstream Mac OS, and it will feature a dramatically improved feature set for non-technical users, including...

    -Carbon, which will allow existing Mac apps to run in a preemptive environment without modification

    -Quartz, which is a PDF-based graphics model, supporting high-end graphics rendering abilities built into the OS

    -First-class java support built into the default install

    -Better hiding of the BSD internals so that new users don't have to know Unix to use their Macs.

    -Ports of all the functionality of the current OS so that existing users will be able to interact with the OS in a predicatble way.

    -A revamped finder

    -Updated internals. They are switching to a new version of the BSD kernal (3.0 comes to mind)

    -Gobs of other cool stuff I can't think of offhand.

    And there probably will be a command line buried in the OS somewhere, but users will not be required to use it under any circumstances. Power users will be able to use it if they wish, however. Target release date is early 2000. They've already seeded a developer release, with a second due in a matter of weeks.
  • Ahh just as I thought.. a guy who really isn't interested in doing anything more than fiddling with his GUI. Okay well since doing WORK isn't your kind of thing than well let's list some realities to you:
    Mac interface customization: Kaleidoscope. Shareware.. thousands of custom interfaces.
    BASH and Xemacs : I don't know what the hell you would need these for since you can find text editors for the Mac like gum on sidewalks. BBEdit is a frontrunner off the top of my head. The fact is, if I needed a text editor, I'm better off scribbling on Notepad or Stickies.
    Games? Over 7 of the top 10 games listed by PCData research are available or are being ported to the Mac.
    Compilers and all that sh*t? Go to the store and buy it. Don't just sit at home and assume all your life.
    Faster hard drive? Take out the old Mac drive and put a new one in. Dang, are you that blunted? Hell, it's easier to format/partition a Mac drive anyway.
    Run Scripts and schedule tasks: AppleScript, nuff said. The Mac can even turn itself on/off and run jobs without additional hardware

    Like I said, I'm usually addressing people who haven't really used a Mac (one that that was built in the last three years). Oh yeah, I'm writing this in Netscape on a beige G3/233 Mac while working on an 89 MB Photoshop 5.5 file on multiple screens while serving out Internet access using the shareware IPNetrouter (mixed Mac/PC clients) and sending out web pages via the Web Sharing control panel. Plus this stock config is connected to the Novell network via stock AppleTalk and has a CD-RW, scanner, Zip, Jaz, two external HDs, and a Syquest44 plugged in. No sweat. No effort. No jiggling configs. Just turn it on and work. That's what my boss pays me to do. Not f*ck around with my "window managers." They work just fine.

    -----
    Linux user: if (nt == unstable) { switchTo.linux() }
  • For those who don't know, LinuxPPC and BeOS won't run natively on Macs or Mac clones. [be.com] I installed LinuxPPC on my Blue G3 and had to make it a dual-boot between MacOS and Linux. Why? Because the damn machine won't boot into anything except MacOS! LinuxPPC waits until the MacOS starts to boot the computer up, then hijacks the system. The sameannoying setup applied when I installed BeOS on my PowerComputing clone.

    Now I love my MacOS and all, but I'd like to be able to run BeOS and Linux without having MacOS hiding underneath somewhere. If these PowerPC boards actually make it to the shelves, I'll be first in line.
  • I thought that the next generation of Apple Motherboards, the ones in the Sawtooth G3/G4 towers didn't have physical ROM on the board. That it was going to be more of the ROM-in-RAM like the iMac has.

    I also read something on Friday...MacOSRumors or MacWeek or somewhere that IBM got one of these monsters to boot from 8.1 a while back.
  • Don't get me wrong, I love the Athlon too, in fact, I just bought one... But the thing's a friggen microwave! It draws 50W! Hotter than the PIII. So yes, it's a nice chip at a nice price, but to be honest, I'm a bit worried about the damn thing burning a hole through my motherboard. OTOH, it should keep my apartment nice and toasty up in the cold darkness of Syracuse.
    The PPC, however, is a truly elegant chip. Certainly on par w/ the Athlon (Altivec routines however will blow it out of the water) and it draws far less power, only 5W for the G3 which makes it quite good for both desktops and laptops, unlike the Athlon. The G4 (which btw, is probably coming at Seybold) only draws slightly more power and is also fine for laptops. Beyond that, there's a lot of Intel hardware in CHRP which not only drives the prices down a lot, but also makes it possable to walk into "mom and pop parts stores" and walk out with a "Bag o' parts". Finally, IBM's chip volume goes way up and the price of the chip not only comes down too, but the R&D goes up (and the PPC is a vastly untapped archecture.)
  • I agree with you about Apple's irrelevance. IBM for years has used PPC technologies in their AIX boxen. They also have beefed-up versions of the chips on their S/390's. (What I wouldn't do to get one of those chips on a motherboard!) I doubt IBM consulted with Motorola or Apple to do that...
  • Can you imagine ProTools/24 with the 888 and having it not being dragged down by MacOS or *shudder* NT.

    I'll take it a step further. Let's say a 4-way SMP G3 ProTools for LinuxPPC Digital Audio Workstation.

    I won't even get into the G4 or what a 128-bit Altivec unit could do for ProTools.


    God, I'm getting the shakes just thinking about it.

    Don Negro
  • Can someone post appoximate specs and prices for these two boxes? Their website was down when I tried to figure this out ..
  • check it out:

    http://macweek.zdnet.com/1999/08/22/rfitues.html

    interesting no?
  • Hi,

    Couldn't help but notice that you feel somewhat... dissatisfied with the current versions of Debian Lignux. I tried installing it a while back, and encountered a ton of problems in dependencies (dselect is a REAL BITCH) and the like.

    I tried RedHat, and couldn't stand that (it is the antithesis of proper, clean Unix design). FreeBSD was somewhat cleaner, but still had install problems. It also had "it worked fine yesterday, but it's hopelessly broken today, and I changed *NOTHING*" problems and is, therefore, not a product I care to run.

    I tried OpenBSD, and stuck. It is tight, clean, consistent, and secure. Everything just *works*, and works *well*.

    Give it a try: OpenBSD [openbsd.org]

    --Corey
  • Not wanting it bad enough... HAH!

    Microsoft told Moto, IBM and Apple they wanted $100Mil for the NT5 port, each! Moto MCG never even sold near that many retail NT and Mac PPC boxes, IBM hell, what did they want NT for , they had AIX, and Apple, DUH!!
    Andy wanted PPC kicked out of his sandbox, and Bill did the kicking. Who know what Bill got in return, probably Intel dropped some Internet software project, they were big in networking and evangelizing net access in that timeframe.
  • You've forgotten the Jobs factor.

    If these PPC boards are widely used, and Darwin is ported over, then a couple of things could happen.

    1) MacOS X (the GUI and stuff above Darwin) will be somehow keyed to work only on Apple platforms. This can likely be subverted, and if it is:

    2) Further development of Darwin is closed-source. Darwin and the MacOS will become proprietary. Future versions of Darwin will run only on Apple platforms and, most likely, binary formats, APIs, and the like will change to ensure that the folks still hacking on the older versions of Darwin are at an evolutionary dead-end.

    --Corey
  • That's what Be tried to do, and look at how popular they've turned out.

    Be didn't catch on for a number of reasons. For one thing, they're developing an OS targeted at multimedia but not making it for the main multimedia platform(Macs). For another, they have yet to get any big graphics software company (Quark, Adobe, Macromedia etc) to port any of their products to BEOS.

    That's what Apple originally planned with Rhapsody, and they had to back track on that in order to convince developers to update their products.

    Not exactly. They revised Rhapsody because it did not provide a good upgrade path for software developers. For an application to take advantage of the modern OS buzzwords, large parts of the code would have to be rewritten. Carbon lets developers keep most of their old code while allowing them to move to OSX.

    Given MS & Intel's revenues and profits, it's a fair statement to say that the computing world values compatibility over performance/reliablitly/price/whatever other concern.

    Of course the fact that both MS and Intel use questionable tactics against thier compeditors wouldn't have anything to do with thier dominance, would it?
  • Finally, a good affordable way to get off the x86. And with the x86's 'out-phasing' by Intel sooner or later, it'll be nice to have a system to keep developing on.

    And hey, this should really improve the port quality for the PPC versions of anything.
  • One of the great myths perpetuated by Apple and others is that cloning had to die for Apple to live. The fact is that the cloners had already agreed to pay much higher licensing fee's when Apple pulled licensing.
  • I'd love to see their claims come true and this computer exist, but it sounds too good to be true, especially to a Mac user.

    Isn't RedHat located just around the corner from them?
  • Actually, this looks largely useless to the average Mac user, because as the article says they'd have to have a license from Apple and I imagine they'd need Apple ROMS and various tripe like that.

    We'd be right back where we were a few years ago when Apple had to get rid of the clones that were ruining their business.

  • Ground Rules -- Read B4 Posting (Score:5)

    by Anonymous Coward on Tuesday August 24, 1999 @09:41AM (#1727796)
    In order to improve signal-to-noise ratio for this article, the following ground rules have been enacted:

    1. Anyone mentioning Beowulf will be forced to write a doctoral thesis on parallel/cluster computing theory. You will also be required to install and configure a working Furby cluster.

    2. Anyone declaring the superiority of Linux/Darwin/BeOS without calm, rational supporting evidence will be forced to hold a cigarette in their mouth while we light it with a military-issue flamethrower.

    Now let the games begin.
  • You say:

    anyway I hate the idea of linux which is usually a do it yourself thing having "problems" with anything (meaning it is basically broken except to Stephen Hawking or Albert Einstein).

    Then, you say:

    Proud Debian GNU/Linux Slink 2.1 (modified) user.

    Huh?

    --
    Interested in XFMail? New XFMail home page [slappy.org]

  • Yeah the no web page thing made me think this is a very small company. Never heard of them. But then I never heard of that other PPC computer company mentioned a month or two back.

    If it happens, great. More PPCs in the market is good for every everybody. If it doesn't happen, well, I can't miss what never existed.

    I'd love to run LinuxPPC and Be for under $1K.
  • Wouldn't getting an Alpha be an alternative to x86?
  • Yup, it will be absolutely useless to
    the average mac user.

    Apple ruined it's own buisness through
    greed long long ago.

    Killing the clones saved apples buiseness,
    sorta, but they shoulda split into a hardware
    and software company eons ago.
  • But you are using an Intel box, no? So, why would the problems be any worse on a non-Intel box?

    FWIW, I haven't had any major problems with hardware support in the 5 years I've been using Linux (only full-time in the last 1-1/2 years).

    I use Linux at work on a P200; at home on a Celeron 300A (@450).

    I have just now gotten my hands on a PowerMac and have LinuxPPC on it. I have to say: it has been mostly a pleasant experience (only problem is trying to run a custom kernel I built; it panics on boot. I'll figure it out, though). The machine is quite responsive, even though it's an older machine. It seems less jerky than an Intel box.

    I'm definitely considering a PPC motherboard for my next box, if the price is right.

    --
    Interested in XFMail? New XFMail home page [slappy.org]

  • I would also be cool if apple or moto or ibm would pay to get gcc up and optimized on linux-powerpc specifically.

    Apple is using egcs to build Mac OS X. Mac OS X Server was built with gcc 2.7.2.1 (who comes up with these version numbers?). I bet they're putting a lot of optimization work into it.

    I bet their changes would port to Linux pretty easily.

  • I'm all over this one. Finally the power of the PPC chip without paying the price for the resource hogging of MacOS and Apple's hardware monopoly.

    Maybe for my next upgrade ... ? It would certainly be nice to have a price-for-price competitor for Intel and AMD. And it's RISC to boot. Maybe it's time I finally got a /seperate/ box for the Linux side of my computational-adventures.

    Yipee!
  • From what I understand, you can modify OpenFirmware to bypass the Mac OS at bootup, but I have absolutely no idea on how to do this.
  • Hey is MacOS X out?? I got one of those crappy mac catalogs in the mail last week something like MacMall or MacWeek or something..(i'm geeky enough to scan even maccatalogs for cool crap :)) anyway they had MacOS X selling on some of their machines and available separately (for like $400 (not exactly umm a cost effective OS)). I scanned thru it looking for an availability date and found none. However, from the looks of this thread MacOS X is still vaguely hypothetical which means either that I was having a mighty fine hallucination or Jobs is personally sending me catalogs from the future...?
  • i do agree that the 'open-source' (APSL) bottom layer of MacOS X, darwin, could in theory be easily ported to a CHRP-based motherboard.

    but i don't see all of MacOS coming to these new boxes any time soon.

    Apple's revenue centers on their high-performance hardware, with software playing at a slightly lower level of importance. as far as business sense goes, and it's been proven that Apple has more than their fair share of that, relicensing their OS would cause fracturing problems. we all know that, aside from its shotty core, windows' main source of problems is trying to support all intel-based hardware. keeping the MacOS on custom hardware is necessary to maintain their high level of performance.

    if you want to keep consumer interaction with the low-level os minimal, custom hardware is the only way to go.

    linux excels in this fractured hardware market because you can tweak to your heart's content.

    Mac boxes are too highly priced and specialized to be anything other than what they are: Mac boxes.

    i prefer to see the release of the CHRP motherboards as a boon to the linux, bsd, and BeOS communities.

    i'll openly admit to being an avid mac user, as well as dabbling in *nix and Be.

    if i pick up one of the new boxes, maybe i'll be able to see Be rocket on a G3 or G4 for the first time.
  • I know! Especially since BeOS has about 100 more apps than MacOS. Oh wait, no it doesn't. Nevermind.
  • True, the Mac OS isn't as buzzword compliant as Linux and Windows...

    No opensource, pre-emptive multitasking, protected memory, open source, object oriented archetecture... Just a very easy to set up and use system. I gave my mom a PC running Windows 98. Now that she's actually doing stuff on it beyond just looking at email, I wish I'd gotten her an iMac instead. Linux? Never.

    There's more to an OS than Marketing. Apple made a great system in the Mac Operating System... Fortunately they're wiser than MSFT in realizing they needed to rebuild everything from the ground up.

    As for hating their prices. Go nickle and dime with the rest of the PC crowd. You get what you pay for in when you buy Apple. PowerPC. Plug & Play. Great case designs (8500 & earlier not withstanding!).
  • >How would you network them?

    Easy. Duct tape.

  • Hate to say it, but it's true.

    If Steve decides to keep Apple hardware closed, these machines will have the life expectancy of a NeXT cube on the open market. The geeks-only segment will be the only folks who care, and that group still isn't large enough to carry a serious manufacturer.

    Sorry to burst bubbles here.

    But, now that Apple need not bail out clone manufacturers, there's hope. IMHO, they should concentrate on morphing their image from a hardware company to a systems company..

    And if these machines run MacOS 8/9/X and LinuxPPC, I can't wait to drive the necessary ten miles to pick one up. =-)
    ("Y'all can go screw Silicon Valley; we have.. Morrisville!")
    -----

  • Hey, I earn much more than i spend on my software. Seems quite justifiable to me.
  • Well, to be fair to IBM, someone there decided to hand that project off to Taligent. We all know what a mistake THAT turned out to be. . .

    "The number of suckers born each minute doubles every 18 months."
    -jafac's law
  • Right, that should have been "with minor modification." Carbon is a subset of the existing Mac OS API's, and Apple provides a tool called the Carbon Dater that checks your code and tells you which API calls are no longer supported. In theory, the deprecated APIs are seldom used, so developers should be able to "tune up" theis apps in days rather than months. But this allows Apple to eliminate enough of the incompatible OS calls to allow apps to run in the preemptive multitasking, protected memory environment.

    I believe Adobe managed to tune up Photoshop in a matter of weeks, so it seems to work. If everything goes as planned, it'll be a nearly painless transition, and developers will be able to switch over very quickly. It's gonna be very cool.
  • Ok, it's flamebait, but I'll bite anyway.

    #1, as to the bit on 'thought', you obviously seem devoid of it. Regardless of interpretation, there are things called Facts. And in the Article in mention, it is said that these machines are NOT treading on Apple territory. This is a Fact.

    Had you read my response carefully and even (gasp) the article itself, you would have noticed the section on how these AREN'T Apple clones. Apple machines use ROM's (or used to anyway).

    Calling these machines Apple boxes just because they are PowerPC boxen is like calling every x86 a Windows box. In fact, IBM has been using these chips extensively in their own machines and Apple hasn't gone ballistic yet.

    As to your conspiracy theories, Apple is going to take over the world just as quickly as AOL/Sun/Netscape. You are obviously 'blind' to the fact that all Apple is doing is trying to make a business again. I don't agree with how they're doing it, but honestly, you claiming Jobs is going to become Gates is about as founded as someone saying Linus will. Because simply put, the leverage isn't there for it to happen.

    And your assumption that I'm an Apple lackey is sorely misplaced. I can't stand Apple. I can't stand the way they make expensive boxen, I can't stand the way they sit on Quicktime codecs. I don't like Apple.

    Like it or not, IBM made the specs available to motherboard makers. So Apple can't do SFA (sh*t f*ck all) about it. The only way Apple will crush these as competition is by making OS X either need ROM's or just not work with CHRA. Like MacOS.
  • Imagine having to decide between:
    XXX Software for Redhat Linux on x86
    XXX Software for Redhat Linux on SPARC
    XXX Software for Redhat Linux on Alpha
    XXX Software for Redhat Linux on PPC
    XXX Software for Debian on x86
    XXX Software for Slackware on x86
    XXX Software for Corel Linux on whatever chips they'd use
    XXX Software for Linux PPC

    No problem:

    ./configure
    make
    make install

    commercial != proprietary

    If it's proprietary (closed), I'm likely not all that interested anyhow.

    --
    Interested in XFMail? New XFMail home page [slappy.org]

  • set DISPLAY=macinabox:0.0?

    THE X WINDOW SYSTEM IS NOT UNIX.

    Mac OS X Server does support remote windowing through a Display PostScript and Mach messaging mechanism. It does not (thank God!) use X-Windows.

  • Not exactly. The 603 does SMP perfectly well (the BeBox screamed, and it was only 2 by 75 MHz).

    The G3 doesnt because of the way it accesses the backside cache, or at least that was the blurb i read somewhere (probably macosrumors) from Moto.
  • Whoops, you are of course correct, I was just throwing data out quick before I left work :)
  • That is only because Microsoft noticed the political infighting and lack of customers and wanted the money as a commitment from the parties involved. They declined, so no more PPC NT development. Had they been unified, and brought enough customers to the bench, Microsoft would have bent over backwards to help. As you said, "...what did they want NT for, they had AIX, and Apple".

    Motorola DID decide to do Windows CE on embedded PowerPC, and has committed itself to supporting it. How long will that last? All I got to say is do a search on Slashdot for Psion or Motorola and figure it out yourself.
  • i.e. are all the gnu utilities easy to port over, do we get a bash prompt? can we compile ssh,gcc??
    > Take a look over at http://www.peanuts.org/peanuts/MacOSX/ [peanuts.org] for stuff that's already been ported, and replace MacOSX with NEXTSTEP [peanuts.org] and OpenStep [peanuts.org] for software written for or ported to NeXTstep and OPENSTEP. On my NeXTstation at home I have ssh and bash running just fine, and when I get around to it egcs 1.1.2 will be installed .br>
    Oh it sucks that it won't run 'carbon' tho. :(
    Well, it may once Carbon is done, but it wasn't released with Carbon. Subtle difference :)
  • Yah... But it won't work with those flashy iMacs or Blue G3's because the Linux/PPC Kernel has broken OF headers which allow it only to boot on older machines (Apple fixed part of there broken OF to make it more standard -- and Linux/PPC developers haven't kept up.)

    The reason for this is, it would seem little of them have interest in wasting time to do this -- after all why bother, when the Mac OS comes free with these computers, and BootX works just as good as open firmware and is easy to use.

    The fact is, people that use Linux-only Blue G3's want to be able to use OpenFirmware -- any people interested in this project?

    Finally, CHRP machines boot right out of OF into Linux.
  • OpenFirmware on the Mac equals BIOS on the PC.
    quik on the Mac equals lilo on the PC.
    Linux/PPC Kernel on the Mac equal Linux/x86 Kernel on the PPC.

    So it works basically the same way -- it should be noted that at least on Power Macintoshes, OpenFirmware booting has fallen out of favor of Linux/PPC developers -- so they aren't actively updating quik / Linux/PPC Kernel for better support of newer macs (blue G3/iMac) with less-brain dead OpenFirmware.

    It's not that it's difficult either -- NetBSD fully supports OF booting on the iMac and the BlueG3 -- just Linux/PPC developer have no interest and continuing active development on a cryptic boot language -- when people can use something that came (free of cost -- not free as in freedom) with there PowerMac Blue/iMac G3.

  • here (Score:1)

    by Vryl ( 31994 )
    Here [merriweb.com.au] it is.

    Until we bunged AIX back on it, it was running NT ppc. It even had perl!

    Nice pics too . . .

    -- Reverend Vryl

  • You're thinking of MacOS X Server (previously known as Rhapsody). As for MacOS X Consumer (or Workstation, or whatever you want to call it) I think it's a few months away.

    And hopefully It's price will be much much lower than $400.... ;)

  • Ummm... on most Mac's, the only part of MacOS that starts booting is the part thats in ROM, which is the equivalent of the BIOS on a PC.

    You can get Linux to boot without a BIOS on a PC too, but why the hell would you want to? LinuxPPC can be installed with that init tool that lets you dual boot, or you can easily configure the machine to boot straight to Linux. I've run LinuxPPC on two different Macs after MacOS was removed from the system.
  • Linux for the most part came to being because it ran on "commodity" hardware. You'd have never seen it take the form it has if it were developed for Sparc or Alpha initially. Why? Because all the developers already had x86 boxen in their basements. People switch to Linux because they can try it on their current computer, rather than needing to go buy a whole other box to see if they'll actually like it.

    x86 will always be the platform of choice for Linux, in my eyes. And yes, G3 definetly womps on Pentiums... But the G3 is constrained to single CPU machines. G4's are around the corner, but how much will they close the distance between PPC and Alpha?

    I'd rather see Linux settle onto x86 and Alpha rather than run out and try to support every platform. That's probably contrary to the mindset that got it to where it is today, but... I want to see more commercial apps appearing. That'll only really happen once linux is settled into it's place in the universe.

    Imagine having to decide between:
    XXX Software for Redhat Linux on x86
    XXX Software for Redhat Linux on SPARC
    XXX Software for Redhat Linux on Alpha
    XXX Software for Redhat Linux on PPC
    XXX Software for Debian on x86
    XXX Software for Slackware on x86
    XXX Software for Corel Linux on whatever chips they'd use
    XXX Software for Linux PPC

    etc... etc... etc...

    This only serves to slow down the application development, in my eyes...

    I could be wrong... That's jsut me though
  • (Wiping bloody bits of bone, hair and gray matter from face)

    Hell, I'd be happy with Cool Edit Pro and drivers for my Isis!

    GadgetLabs seems the most likely to produce something for Be/nux someday, sometime...


  • oh yeah, that would solve that problem

  • Yea you are right, it was Max OS X Server. But that is the one w/ the unix core right? It was advertised as the first mac w/ true preemptive multitasking etc etc.. (no mention of unix tho explicitly) I gathered the server bit was just that it came w/ apache and the obligatory marketing buzzwordage :)
  • This is not completely accurate (at least for LinuxPPC). You can use OpenFirmware. The LinuxPPC web pages do focus on the BootX (needs macOS) solution, but OF is completely feasible
  • Are you kidding? Geeks support many otherwise marginal companies. VA Linux Systems (or whatever they call themselves now) comes to mind.

    But maybe you are defining "serious manufacturer" as "a manufacturer everyone has heard of", which in this context is circular. If everyone has heard of them then they aren't just for geeks, are they?
    ---
    Put Hemos through English 101!
    "An armed society is a polite society" -- Robert Heinlein
  • But Linux is developed to be multi-platform. I mean, with respect to it's place in the universe, it's meant to have as little arch dependant code as possible. Ergo, building apps for Linux shouldn't be /too/ hard to do with a multi-arch goal in mind.

    This was just my understanding, however. I'm not too in tune with the arch issues with respect to programming, but shouldn't we be trying to achieve this goal? I mean, theres no sense in supporting Linux as a Multi-arch OS if we don't develop apps with a multi-arch mindset anyhow. Right?

    I think the multi-arch commercial app support will come when companies realize users don't want to be tied down to one arch. I find it difficult to believe 100$-500$-a-pop shrink-wrap software companies can't store their arch-dependant logic in seperate code from the main logic of the application, thus enabling them to do ports with a fair amount of ease?

    Emagic (www.emagic.de) will be getting a lesson in this approach quite soon .. their Emagic Logic Audio (music sequencing software) product will soon support BeOS, on the heels of merging their Mac/Windows code about a year or so(?) ago. And that's different OSes, nevermind arches too.

    Garret
  • umm... Linux/PPC couldn't care about if you have an ROM or not your machine -- it never reads it at all -- neither does OpenFirmware.

    The main problem with Linux on Macs is:

    - They are quite expensive -- for obvious reasons -- they are brand name computers, and have licensing fees doing with the Mac OS.

    - Macs traditionally have had weaker OpenFirmware compared to CHRP and IBM machines -- since OpenFirmware wasn't much of a use for Apple -- since why would an Apple machine need to select OS (besides Mac OS X vs. Mac OS Classic) or Password Protection.

    - You are limited in selection when you buy an Apple machine -- CHRP gives you many more choices.

    Apple is working hard to improve their weeknesses -- and it's showing up -- look at iMac -- cheap, nice looking, and a half decent OF implementation.
  • Lets not forget Apple helped design CHRP (and is currently selling CHRP-like Macs -- Blue G3 / iBook come to mind).

    So plug and play works without problems, no matter what OS use choose on the PowerPC -- Plug in that SCSI hard drive -- and mount it in Linux -- no fidling at all. Plug in that ethernet card in a PCI slot -- make sure you have the correct drivers compiled into the kernel -- and your ready to rock.

    Linux is much easier to set up on the PowerPC -- inheritably due to the much better hardware design -- doesn't matter if it's PowerMac, CHRP or RS/6000.

    PowerPC hardware is designed to make adding hardware as easy and do-able as possible -- it doesn't matter of the OS.
  • You are seeing more stuffing making it to PowerPC Linux everyday -- it's not a difficult port from one Linux to another -- and it's a cheap way to sell more copies of product -- with mimual testing / support.

    Stuff that is here now or is being promised:

    - Netscape Communicator 4.6.1 is here now.

    - Applixware (latest version) for glibc 2.1 is being ported. The last version for glibc 1.99 is out.

    - CTP/Civilization is out and bunch of other commerical Linux games are being ported, including Riven (yes!).

    - IBM DB/2 Database software is avalible for Linux/PPC.

    - JDK 1.2 is out today.

    - Adobe Acrobat Reader 4.0 is Comming Soon.

    - SheepShaver 2.1 for running Mac OS apps is almost ready for release.

    - RealPlayer G2 for Linux/PPC is under study of Real Inc., to see if such a port would be worth while -- figuring how much the Mac OS port of RealPlayer G2 sucks, people would really love a solid Linux/PPC version of this.

    - BOCH (in .ppc.rpm format) for Linux/PPC glibc 2.1. Run great PC x86 software on your Linux/PPC including Windows.

    - And a bunch others that I can't yet think of.

    And yes there is all of that great GNU software out there for the PowerPC including:

    - KDE / GNOME desktop enviroments.
    - XFree86/XF68_FBDev X Server
    - Mac-On-Linux. Run Mac OS or extra copy of Linux/PPC or MkLinux in a seprate console -- this is great stuff.
    - KOffice/AbiWord/Gnumeric/GIMP/Apache and many, many other great ones.

    So you aren't by far limited on Linux/PPC -- you'll find yourself with lots of great software. And great friendly user base and friendly developers.
  • Even if they did code to check to make sure for Apple hardware -- people would figure how to get around it -- using methods like recycling ROMs from first generation G3s and soldering them on to daughter board for CHRP boxes -- and using some kind of OpenFirmware hack to get around them. Or maybe just keep a software image in memory -- illegal yes -- but people would try it (remember the Mac-clones of early 1980's.)

    And there might be extentions that crack Mac OS X so it doesn't do this check or that it lie to Mac OS X about the installed hardware.

    If there is an challenge, crackers will often break it quickly -- no matter how hard / well Apple designs it.

  • Apple is dependant on hardware sales to fuel software development. It is now and has been for sometime extremely difficult if not impossible to imagine a way that they could separated their businesses and suceed.

    Seems to be that it would be easy: just make the license for MacOS be high enough to cover the cost. The price of Apple Macs wouldn't need to change, since presumably what they charge already covers the cost of both the hardware and the software. On the other hand, the cloners might end up having to pay $500 pre license of MacOS. (I'm pulling that dollar figure out of my ass, but you get the idea...)

    Tying the hardware and software products together is an admission that they are overcharging their customers. There's no other reason to do it.

    Well, actually, I can think of one other reason: If they charged a fair price (e.g. $500) for MacOS sales, they might fear piracy. Thus, they sell the Mac hardware as a dongle for the software. A lot of people have the demented idea that operating systems should be free or cheap, in spite of the cost of its development. (e.g. Linux is free, and Windoze is sold at a loss to lock people into buying/renting other MS products.)


    ---
  • "I would also be cool if apple or moto or ibm would pay to get gcc up and optimized on linux-powerpc specifically."

    Well, at least Motorola has contributed dozens of various patches to the Linux/PPC effort -- including some to egcs/gcc -- for better PowerPC support now and for future support of Altivec processors.

    While, I don't think Motorola has full time programmers working on patching / improving PowerPC-Linux, I do know for a fact that they have people, at least part time / free time / slacking of from work time, working on getting Linux supported better on the PowerPC, and better on hardware they ship. They made Linux/PPC on CHRP possible, and helped get it on there embeded systems.

    Many of Motorola employees, would love the day they can get Windows NT boxes off there desks -- many which run on Intel machines (which you can see why Motorola employees dislike there Intel Window NT boxes).

    Don't believe me -- check the Linux/PPC Mailing Lists (for old patches) or the Motorola Computing group Linux website.
  • The reason the 750 (G3) is not often used in dual processor designs is that it lacks the S-bit (snoop bit) which allows the cache to be snooped. This can be gotten around in software. Synergy Microsystems has a good example of such a design (4 way 750) running Linux in SMP mode. -Matt Motorola Computer Group
  • Actually there are people working on optimizing Linux for PowerPC. They range from the performance group at SPS, to the MCG porting group which optimizes specifically for our embedded VME and CPCI boards. At the very least, Linux is an important part of Motorola Computer Group's strategy to sell more hardware.

    -Matt at MCG

  • PowerPC won't ever drop to the prices you see in x86 land. Why? 10x as many users.

    Granted, x86 has the economy-of-scale advantage, so its development costs can be amortized over a greater number of sales. But that is only one factor in their prices. The x86 costs more to make than modern processors of similar performance. The x86 has to support more legacy and brain damage. A Pentium-III has chip real estate dedicated to performing useless and ridiculous things like "real" mode, V86 mode, 286 protected mode, segmentation, etc. Engineers spent time and money making that stuff work, too. If you bought one, then you paid for it.

    You see, economy of scale is a blessing, but a curse as well: the only reason people buy them is that they are compatable with all the previous generations. Compatability is the #1 design constraint in x86 production. Performance and cost are secondary considerations. It must be able to run code that was compiled in 1980.

    Modern processors don't have that constraint. They are easier to design, and should cost less to make. If the developers can sell enough of them to amortize the cost enough, it'll be price-competitive with the 386. In the case of the PPC, they get an easy 10% (approx) of the personal computer market, thanks to the Mac. And then there's the embedded market. Now they're going after the Linux folks too.

    Will it be enough? Maybe, maybe not. I sure wouldn't rule it out. I also know how I'm gonna vote my dollars. :-)


    ---
  • SheepShaver [sheepshaver.com] is a MacOS emulator for LinuxPPC and BeOS/PPC.


    If it will run under LinuxPPC on one of these boxes, I am there.

  • I wouldnt mint it at all. It would give the PPC platform acceptance in the eyes of the suits and help the CHRP platform. A good thing. Let MS do the hard part...
  • M$ ported Windows NT 4.0 to CHRP three years ago, and cancelled it. Just like Sun cancelled Solaris for CHRP.

    My CHRP box has a Windows icon in it's graphical OF boot menu. OF course clicking on it doesn't do anything :-) Haven't found out yet how to put a Penguin in the boot menu...

  • M$ was one of the CHRP partners, along with IBM, Moto, Apple and others. Their part of the bargain was to port NT4 to PPC.

    They never did. (Needless to say.)

    That's part of why Mac cloning died the horrible death it did. All the companies weren't intended to be competing solely for the Mac market, they were supposed to go after the NT market as well.

    As to why they didn't, one-word guess. Intel.

    I'd imagine the concept of competing against a (let's face it ) better architecture pushed by Big Blue *and* Moto (this is back in '94, when Somerset was kicking ass and taking names) scared Andy Grove shitless.

    God only knows what he offered Bill in exchange for killing the PPC port.

    Anyway, that's the skinny on that. Any further details from people in the know are, of course, appreciated.

    Don Negro
  • AFAIK, it's not so much an emulator as a run-time environment that accesses the MacOS ROMs from within BeOS/LinuxPPC. So, essentially, you're running the MacOS in a window, not emulating the API calls.

    Don Negro
  • Right thats what i meant.. (the workstation/server looked like a bundling difference than any core difference) but all of this means someone somewhere has MacOS X Server (since it is out or advertised?) So what I was wondering was how cool was it? i.e. are all the gnu utilities easy to port over, do we get a bash prompt? can we compile ssh,gcc?? (it should all be relatively straightforward since its unix right?) set DISPLAY=macinabox:0.0? A Mac that can run all the mac utils as well as having a decent set of programs would be a fantastic compromise.. for people who only want one machine for all things.. (word processing etc) Oh it sucks that it won't run 'carbon' tho. :(

    -avi
  • Porting should be easy enough, it's Linux to Linux after all... But I doubt that many people will ship source-code anytime soon. So you're locked into what whomever wants to ship. Take photoshop, for instance. For a long time the Mac version came out eons before the PC version. Now they're roughly equal with maybe even the PC version arriving first. They cost the same, too. Last I looked, Photoshop for SGI or Solaris was $2500 and still at version 3.

    What it really boils down to is a matter of PACKAGING. Redhat, as so many /.er's love to point out is worth more than SCO, SGI, and myself combined. They can go to a software company, anysoftware company, and say "So, you're developing that for Linux, eh? How about we pay you $XXXXXXX dollars and whenever you mention Linux on the packaging or in the manual, have it say "Redhat Linux" instead?"... Oh and we're going to use a slightly different directory layout than say Caldera and Debian, so be sure to hard code that info into your installation routine"

    So a company would get development money to make a version for Redhat. They'ed then need to ship a second boxed version for any other x86 Linux. Most vendors do not ship for different platforms on one medium (aside from games)... If you're going to use your software on a Mac and PC and Linux, you've gotta buy it 3 different times, as stupid as it seems. So it may turn out that if you're going to run Linux software on Alpha Linux, Redhat Linux, and LinuxPPC, you need to purchase 3 separate editions of it...

    Of course that all goes away when everyone embraces opensource. Unfortunatly.... That's going to take longer than the time it takes my wristwatch to crack 4,096,302,813,988 bit PGP keys...
  • Even if they couldn't kill the toxic fumes from the benzene would be a cause for alarm.
  • Maybe I'm missing something here but dosn't debian have a great deal of support for just as much if not more platforms than Red Hat? They also have a hurd port in the works check ftp://ftp.debian.org/pub/dists/potato/main/ and see for yourself.
  • You forgot expensive compared to cheap PCs today or one at a used place.
  • Why run on sub-standard hardware when you don't want to run sub-standard software? With such a mindset, we'd all be running Win98 on X86.

    This is what pisses me off about Be, too. They went on and on about how great the PowerPC was, how the combination of Be/PPC was top of the line all the way around, and abruptly shut up once Intel made their investment.

    Besides, the article was basically about the PowerPC platform dropping down (in some cases) to 'commodity' X86 prices.

    - Darchmare
    - Axis Mutatis, http://www.axismutatis.net
  • there is another way-- it's called Mac-on-Linux.
    http://www.ibrium.se/linux/mac_on_linux.html

    It isn't so easy to install (requires kernel recompile type stuff), it takes over the screen instead of going in a window like in sheepshaver, and doesn't work on all machines. However it's totally open-source and free and it's here now, whereas Sheepshaver isn't quite yet released for linuxppc and seems to cost $50 anyway. So until sheepshaver/linux is released, i guess you can use this instead.

    p.s. i believe the term is "hardware abstraction", not emulation, since you're using the exact same machine you're "emulating" and there's no translation between different chip instruction sets.. but it might be emulation if you run it on one of these IBM-based thingies.. i dunno.
  • amiga will be a unix tailored for masses, as well. beos isnt unix but they are making a dumb-ass frienlyd version.
  • Plug in that SCSI hard drive -- and mount it in Linux -- no fidling at all.

    Now, I'm as excited about low-cost PPC boxes as anyone, but...fair is fair...

    That's a feature of SCSI, not of CHRP or PPC.

    Plug in a SCSI hard drive on a x86 box that has SCSI, and it's about the same. Let's not forget that new Macs (and Sun Ultra 5s and Ultra 10s -- insert Homer Simpson's girlish scream here) have IDE peripherals, along with all the baggage that entails.

    And even then, who's to say that setting SCSI ID and termination jumpers is really easier than setting IDE Master/Slave jumpers?

    --
    Interested in XFMail? New XFMail home page [slappy.org]

  • I would really like to see Be get their hands on the hardware and get some G3/4 support going. cheap PPC BeBoxes would be really cool especially for schools and such that could use a really nice multimedia OS. Someone else also mentioned better heat/cost performance for rackmounts. Large cool cheap rackmounts means cheaper clusters and servers which is good for everyone. Maybe these could go in a line of set top boxes or portables running QNX or LinuxPPC. Tres cool.
  • Sounds good, but I can't find any appropriate flashers. If anyone out there knows how to flash OpenFirmware for Linux or Be, please email me!
  • I think it was also a StrongARM based system. It was running at 160 MHz at a time when desktop systems were in the same speed range.

  • open ppc motherboard development (Score:3)

    by rillian ( 12328 ) on Tuesday August 24, 1999 @10:01AM (#1727948) Homepage
    I'd like to point out that there's an open collaboration also working on producing IBM's design. I doubt we're going to ultimately price competitive with large motherboard manufacturers, but we may be faster, and we're dedicated to keeping design improvements free and available. "Open Source" for hardware, as it were.

    It's hard to say anything concrete until IBM releases the design [phys.sfu.ca] but we're aiming for a US$500-$700 box.

    Please subscribe to our mailing list [phys.sfu.ca] if you're interested in participating.
  • I don't smoke.

    You would after the flamethrower finished with you. :-)

  • NT PPC died - because MS had enough clout and audacity to demand that the HARDWARE MANUFACTURERS do the porting of the OS to their own chip: Moto had to port to PPC, DEC (now Compaq) had to port to Alpha, etc.
    Moto couldn't/wouldn't do it for some insane reason, so MS asked Moto to pay some obscene fee to port THEIR OS to PPC. Needless to say, after NT 4, MS raised the fee, and Moto told them to stuff it.

    Now, Moto has done a number of moronic things prior and since regarding the advancement and advocacy of the PPC chip - we won't go into this sordid history here.

    But yes, back in 1993/94, the future looked very bright indeed for the PPC, and I got into Macs BECAUSE of this potential future (like them still, but rail agains the price). Although I would have thought, several years back, that if PPC went nowhere, and then would have been ressurected 5 years later, there would be no chance, because by then, Intel would have caught up.
    Boy was I wrong. Intel hasn't done shit other than raise prices, and maneuver to shut out x86 competition, and let MS take care of competition on the Sun and IBM side of things. And now PPC is poised for a resurgence.

    Only that one dickhead at Motorola (the guy who wants Intel and NT everywhere) and Steve Jobs stand in the way. They can probably kill this new PPC movement, it remains to be seen, the fortitude of these neo-CHRP cloners. Can they withstand being bought out or otherwise sleazed to death by Apple and Motorola? (the purchase - and likely subsequent destruction of Metrowerks may have been Motorola's move in that direction, with a nice side effect of hosing Apple in the process).

    The only thing I know, is now, though NT PPC would mean more PPC chips sold, I think that the ABSENCE of NT PPC does more good in the net, and furthers "the cause".

    "The number of suckers born each minute doubles every 18 months."
    -jafac's law
  • : PowerPC won't ever drop to the prices you see in x86 land. Why? 10x as many users. AMD motherboards are projected to be MUCH more expensive than BX boards for that same reason, until the volume comes. At least AMD remains compatible with Intel...

    It is certainly true that there are a lot more x86 users than Mac users, but all those iMacs and iBooks give IBM and Motorolla decent volume. If Apple keeps gaining market share and a substantial number of Linux and Be users switch, you could see 15-20% market share for the PPC in a couple of years. Given that the x86 market is split several ways, that's probably enough to bring the cost down into competitive territory, especially if the G4 is as fast as it looks like it will be.

    One particularly attractive option would be LinuxPPC labtops. Apple already has 400 MHz Powerbooks for about $3000. The PPC is a much smaller and cooler chip, and so you could see gigahertz G4 labtops by the middle of next year. That would leave Intel's anemic pertable Pentiums in the dust.

    As for the Alpha, the Mac gives the PPC a much larger market share than the Alpha, so no matter how impressive the Alpha is in theory, it's not going to come down in price unless a consumer OS becomes available for it. The only hope of higher volumes on the Alpha is Linux users, and I doubt enough will switch to make it worthwhile.
  • 1. Anyone mentioning Beowulf will be forced to write a doctoral thesis on parallel/cluster computing theory. You will also be required to install and configure a working Furby cluster.

    You'd have to modify the Furbies and remove their little coughing/cold feature. Can you imagine a several hundred Furby cluster all coughing? Chattering alone is bad enough.

    Can you rack-mount Furbies?

    How would you network them? Would they do what even the US Marines won't do - go into "battle" with a cable hanging out of their butts?

    Why do I find a "Furby Cluster" amusing?

  • Mac OS X isn't going to be just another Unix* OS. It's going to be the only one that's fit for use by the general public (read: iMac users and such). Linux will be there in a couple of years, as GNOME and KDE get better, but OS X is going to be there in 6-10 months. That's a _very_ big deal. Ease of use is one of the two major reasons Linux hasn't killed Windows yet. The other of course is apps targeted at the desktop market, and Mac OS X will do better than Linux there as well in the near future. And OS X hasn't really had any features dropped from it from the original announcement. They killed the x86 version, when it ships it'll have a _larger_ feature set when it ships than what was originally announced.

    As for Jobs missing this whole thing, he likely knew about it before IBM even told the public. The question is what he's going to do about it. It's easy to say that Evil Apple will do the Evil Things, but I wouldn't bet on any predictions about what Jobs is going to do.

    *Yes, I know it's not Unix(tm).

Slashdot Top Deals

Time is the most valuable thing a man can spend. -- Theophrastus

Close