Slashdot Mirror
Speculations Intel's Next Generation
An anonymous reader writes "The Inquirer speculates about the next generation Intel chip. It's low power, 64 bit, multi core (up to 16?) and the real reason for the Apple switch."
← Back to Stories (view on slashdot.org)
If they're announcing an archtecture this radical at next week's IDF, what are the chances that it will be available and running well in time for Apple's announced timeline for desktops?
Or is Apple going to sell a lesser version first, in which case why haven't they already switched over to selling it to early adopters already. Yes there really are people who buy systems and wait for the applications to arrive later.
"It's the height of ridiculousness to say for those 9 lines you get hundreds of millions."
If a VLIW X86 processor had a "native" mode, one would have to wonder if Apple's Rosetta technology could compile directly to it instead of X86. I mean, it would seem dumb to JIT-compile to X86, which in turn is translated to VLIW.
Any sect, cult, or religion will legislate its creed into law if it acquires the political power to do so.
...that we will see, eventually...
1. Four cores standard
2. Chips pluggable to the mobo like Atari cartridges to eight CPUs
3. Mobos as blades to passive backplanes
4. Home blade servers and thin clients.
I think in the end we'll see low-end, mid-range, and high-end blade everything in the future with modularity being the way of everything.
But that's just my speculation.
If my grammar and spelling are off, I am [distracted/tired/careless] (take your pick)
He's been smoking some seriously strong weed to come up with the crazy ass ideas in that article.
Ian Ameline
That's good for sun, because they sell server stuff, but for other kinds of workloads this approach is very innefficient. See the Piranha research paper, by Barroso et al.
The Raven
Huh? HP approached Intel with the EPIC architecture, as it was based on their next generation PA/RISC research. The HP/Intel allience is a refinement of the Super-Parallel Processor Architecture (SP-PA). There was no swindling or bluffing HP. If anything, one could say Intel was tricked since they dropped their x86-64 designs, lost focus on x86 in general, and invested billions in Itanium. Try actually reading some of the history next time.
The failure of the other architecture is not just Intel's successful marketing, but also in the lack of interest to continue development by the respective companies.
"Open Source?" - Press any key to continue
I don't know if Intel bluffed HP or HP bluffed Intel. What I do know is that HP's customers were the ones who got fooled. When the announcement of their joint processor development was made to HP customers in the mid-90s, the promise was that the result was only a couple of years in the future and would be PA-RISC compatible. Compatible meaning that it would run existing code without modification. That's what I heard them say to in a presentation to one of their larger customers. Ten plus years later it still isn't here and isn't compatible.
Assuming that the article is generaly correct this upcoming processor will be able to morph to other architectures. Could this mean that we can have some sort of native (or at least semi-native) JVM or .Net processor? I am not certain whether implementing a java virtual machine on hardware is feasible but this would be an interesting possibility.
Or it could be that the software JVM of today produces good enough native code for any architecture (x86, ultrasparc, ppc) that it makes it pointless to try to implement a machine that interprets the classes directly?
There's a better explanation of why the Inq article's speculation is bogus here:
i on=detail&PostNum=3655&Thread=3&entryID=55310&room ID=11
http://www.realworldtech.com/forums/index.cfm?act
I find it seriously doubtful that the next intel chip (at least the one that apple is planning to use) will use a translation layer. Think about it; if it had such a capability, why would they (Apple) go through all the trouble to program an incredibly slow emulator when they could pressure intel to make it processor/firmware reliant and avoid the panic from the intel switch? Granted, this may be the performance boost they are banking on in rosetta, but I still fail to see how it could not be completely confined to a lower level of the system and transparent, as well as make it better performing (I'm probably wrong, but I thought that Transmeta had a working PowerPC translator working for their chips, but maybe it was just hypothetical propaganda). So, either Apple is not being its typically self-centered, well, self, or they are pulling the ultimate double-reverse psychology marketing to make us exercise our (your) code foo and think there will be a nonexistent change! Seriously, though, why go through all this trouble when there exists a way to, with a bit more input from both parties, they could avoid this transition problem apple zealots (myself included)are squirming about? Unless IBM has some IP issues with its processors..... Bashing and Speculation in 3.... 2.... 1.....
The writer of the article does have a point about Intel's next gen chips. It does lay in line with the current fab expansions, BUT (there's always a but), there could of course be other aspects of it. As far as I know, Apple won't struggle to beat the fastest PC:s in the world no more, simply because they never will. Therefore, I doubt that such move would be based purely on better Intel chips. Instead, I would suggest that PPC architecture wasn't pleasing Apple just that much. After all, PPC is unable to keep up with Intel/AMD nowadays and supposedly, Apple could be worried about this. On the other hand, Apple is so well known for surprising the masses. What if old Stevie woke up one day with that big idea in his mind of switching over to x86 architecture, just to make a move on Microsoft? What if they actually expand to non-Apple certified hardware? I wonder if they could make more money on such choice, and really, I doubt that people would stop buying iMacs because of this. That's still a matter of taste, since iMacs come with simplicity and style. PC:s generally don't. Last but not least, COSTS. Apple is expected to announce their own developed iPods. Nowadays, they sell iPods with peripherals from Samsung, Toshiba, etcetera. So what it comes down to is, perhaps this move was purely so that it could make an extra buck or two? I'm pretty sure that Intel would love to make a sweet deal with Apple as it did with Dell. Depple? FINALLY!
Full Tilt
If you haven't been paying attention, the Inquirer is one of a new pile of pseudo-news websites posting ridiculous garbage with sensationalist headlines and plenty of ads. Nothing to see here, please move along.
I think you hit on it. I was talking to some guys from Freescale recently about processor offerings for one of our new board designs and the topic somehow got sidetracked on Apple's switch to Intel. They told me that nobody was really making much money selling processors to Apple. They had to invest a lot of $$$ into R&D to continue cranking out new chips for Apple, and Apple wasn't willing to pay much for the chips. As a result, a business decision was made to focus R&D in the area that had the biggest payoff -- the embedded market. Bottom line: IBM & Freescale were planning to target all new processor designs at the embedded market not the desktop market, and Apple knew it. x86 is really the only processor out there that gets designed specifically for the desktop market. To stay competitive, Apple had to move to x86.
As a result, a business decision was made to focus R&D in the area that had the biggest payoff -- the embedded market.
You know, that feels very much like a deja vu. Same thing happened with M68K.
If J.K.R wrote Windows: Puteulanus fenestra mortalis!
In any case, the fact that everyone wants to jump to 64 without testing the waters very carefully first is seriously foolish. I know I'm not the only one who feels this way -- Microsoft's Windows speech recognition subsystem refuses to run on any 64 bit architecture unless all of the OS and applications are strapped to 32 bit mode.
This is possibly worse than five years ago when people were paying absurd premiums to go from 800 MHz to 1.3 Ghz with RAM speeds stagnant. At least then you got something more from algorithms which weren't memory access-bound. From 32 to 64 is a significant step backwards in many cases.