Headline FeedsLinuxInsider Talkback
|
|
See Full Story
For the last three weeks I've been talking about the impact the new Sony, Toshiba and IBM cell processor is likely to have on Linux desktop and datacenter computing. The bottom line there is that this thing is fast, inexpensive and deeply reflective of very fundamental IBM ideas about how computing should be managed and delivered. It's going to be a winner, probably the biggest thing to hit computing since IBM's decision to use the Intel 8088 led Bill Gates to drop Xenix in favor of an early CP/M release with kernel separation hacked out.
Re: Apple: Up the Market Without a CPU
Posted by: muzza 2004-07-12 01:02:14 In reply to: Paul Murphy
What is "adequate management bandwidth" supposed to mean? "Not enough money"? Why not just say so?
Re: Apple: Up the Market Without a CPU
Posted by: Paul_Murphy 2004-07-12 08:24:57 In reply to: muzza
"not enough management bandwidth" means that they're operating at or past the limit of their management's abilities.
Sun, in particular, is strong at the very top and strong at the bottom but has a horribly weak middle management layer.
For either of these companies to take on another workload would be to over extend tremendously.
Sun, in particular, is strong at the very top and strong at the bottom but has a horribly weak middle management layer.
For either of these companies to take on another workload would be to over extend tremendously.
Re: Apple: Up the Market Without a CPU
Posted by: Tech_Writer 2004-07-08 23:25:11 In reply to: Paul Murphy
Makes me think about the SNAPPLE jokes that were popular a few years back when the fortunes of Apple and SUN were reversed. The author neglected to mention the fact that the Next Corporation did a version of Openstep for the Sparc architecture and that SUN's internal politics kept such from ever being supported in a meaningful way. The joke of an interface known as CDE was promoted in favor of Openstep. Then there's the Macintosh Application Environment for Solaris...Had Steve Jobs not killed the Mac OS clones, Apple would now be an interesting footnote in the Wiki history of PCs.
If memory serves, IBM also had a license to sell the PC version of Openstep/NextStep and never managed to do anything with it for this OS just added to IBM's sad mid-90's OS story: OS2, Windows, AIX, Taligent, Pink, etc.
Anyone remember "Project Star Trek?"
The author should have also mentioned that SUN's having quite a few problems with it's CPU roadmap and had to kill its next generation Ultra chip in favor of one developed by Fujitsu. SUN's move to LINUX is one of the brighter things it's done in the last year. The faster SUN can move towards standardizing upon the Gnome desktop for Linux, Sparc Solaris and PC Solaris the better.
The idea of a SUNRAY for Apple's OSX is interesting, but what's the point, especially if you can use the Apple Remote Desktop in conjunction with VNC on just about any box you might have laying around?
Apple and IBM have had an interesting relationship that's spanned close to 15 years. Remember things like OpenDoc, PReP/CHRP hardware and the like? IBM's chip games are not news to anyone at Apple/NeXT. On the other side of the fence, even Intel has had to pull a U-turn with respect to clock speeds and CPU designs. Least there wasn't mention of AMD and it's 64 bit wonder.
That IBM and Apple didn't hit 3 ghz is not that big of a deal. This kind of schedule slip is SOP in the world of Tech - and is not a sign that the sky is falling or that Apple needs to find yet another savior. Odds are the R&D going into IBM's new cell chip will likely power iBooks in 2006. Any takers on this particular bet?
PS Apple was doing fat binaries when Microsoft was still trying to transition folks from 16 to 32 bit applications. At the same time, Next was doing quad fat binaries...
If memory serves, IBM also had a license to sell the PC version of Openstep/NextStep and never managed to do anything with it for this OS just added to IBM's sad mid-90's OS story: OS2, Windows, AIX, Taligent, Pink, etc.
Anyone remember "Project Star Trek?"
The author should have also mentioned that SUN's having quite a few problems with it's CPU roadmap and had to kill its next generation Ultra chip in favor of one developed by Fujitsu. SUN's move to LINUX is one of the brighter things it's done in the last year. The faster SUN can move towards standardizing upon the Gnome desktop for Linux, Sparc Solaris and PC Solaris the better.
The idea of a SUNRAY for Apple's OSX is interesting, but what's the point, especially if you can use the Apple Remote Desktop in conjunction with VNC on just about any box you might have laying around?
Apple and IBM have had an interesting relationship that's spanned close to 15 years. Remember things like OpenDoc, PReP/CHRP hardware and the like? IBM's chip games are not news to anyone at Apple/NeXT. On the other side of the fence, even Intel has had to pull a U-turn with respect to clock speeds and CPU designs. Least there wasn't mention of AMD and it's 64 bit wonder.
That IBM and Apple didn't hit 3 ghz is not that big of a deal. This kind of schedule slip is SOP in the world of Tech - and is not a sign that the sky is falling or that Apple needs to find yet another savior. Odds are the R&D going into IBM's new cell chip will likely power iBooks in 2006. Any takers on this particular bet?
PS Apple was doing fat binaries when Microsoft was still trying to transition folks from 16 to 32 bit applications. At the same time, Next was doing quad fat binaries...
Re: Apple: Up the Market Without a CPU
Posted by: chaoskitty 2004-07-08 22:49:44 In reply to: Paul Murphy
The cell isn't the only way to go; Power architecture server chips will fuel much, if not most, of the PowerPC technology development.
As it stands, watt for watt, a PowerPC is a much more attractive chip. Many embedded developers have realised this, cluster builders are catching on, and large hosting companies will soon follow suit.
If the Linux community ever gets out of its myopic, x86 centric worldview, PowerPC will be the obvious choice for efficient, reliable computing.
John Klos
As it stands, watt for watt, a PowerPC is a much more attractive chip. Many embedded developers have realised this, cluster builders are catching on, and large hosting companies will soon follow suit.
If the Linux community ever gets out of its myopic, x86 centric worldview, PowerPC will be the obvious choice for efficient, reliable computing.
John Klos
Re: Apple: Up the Market Without a CPU
Posted by: panicboy 2004-07-08 14:08:10 In reply to: Paul Murphy
It may be that Apple doesn't WANT to use 3GHz G5s because it's already busy engineering products that use the cell processor. Apple already has an OS portable across 2 similar processors (and Intel, if rumors are to be believed). They're not that far past a multiyear drought brought on by their reliance on Motorola. Is it really that farfetched to believe there might be a contingency plan in case their star processor ends up on a milk carton?
But.....
You make a big point about how cell wont support altiVec but completely leave out any mention of how altiVec would/nt fit with Sun cpus and products like Niagara.
You make a big point about how cell wont support altiVec but completely leave out any mention of how altiVec would/nt fit with Sun cpus and products like Niagara.
Re: Apple: Up the Market Without a CPU
Posted by: RavenMoon 2004-07-08 12:51:56 In reply to: Paul Murphy
"Motorola's microprocessor division -- now Freescale Semiconductor -- is mostly out of the picture, despite having created the PowerPC architecture."
.
Actually IBM created the PowerPC architecture, which was based on its POWER line of CPUs, and brought it to Apple for consideration. Apple liking what they saw then got Motorola involved since they were long time partners. Thus the AIM alliance was formed.
IBM has the trademark on the name "PowerPC"
.
Actually IBM created the PowerPC architecture, which was based on its POWER line of CPUs, and brought it to Apple for consideration. Apple liking what they saw then got Motorola involved since they were long time partners. Thus the AIM alliance was formed.
IBM has the trademark on the name "PowerPC"
This is sort of true. IBM invented the POWER architecture, but Motorola modified the architecture to simplify implementation on silicon. The IBM POWER line of CPU's were multi-chip monsters that wouldn't work in a desktop machine (I think some of them still are). So Motorola trimmed a few features and instructions making the architecture implementable on a single current day chip.
They then later added the great altivec instructions and unit to the chip when silicon technology allowed them the transistor budget.
So I think it is fair to say that Motorola created the PowerPC architecture. It might have been more clear to add "based on IBM's POWER architecture"
They then later added the great altivec instructions and unit to the chip when silicon technology allowed them the transistor budget.
So I think it is fair to say that Motorola created the PowerPC architecture. It might have been more clear to add "based on IBM's POWER architecture"
