Imagination Technology Buys MIPS 56
New submitter HalWasRight writes "After years of struggle, MIPS Technologies — the original RISC processor company — is being sold to Imagination Technologies, best known for its popular mobile GPUs. Part of the deal included MIPS divesting much of its non-processor related patents to a group that includes ARM. This deal could change the landscape in the battle for mobile sockets."
MIPS press release, Imagination press release.
So they were still alive? (Score:3)
I thought MIPS was dead. And looking back a lot of architectures has been running out of steam. Motorola 68000, PA-Risc, Digital Alpha to name a few...
Right now there are only three or four architectures that delivers some punch, x86, ARM, Sparc and Itanium. But the last is only alive due to HP and Sparc is kept alive by Oracle so far.
PowerPC (Score:3)
Re: (Score:3)
Re: (Score:2)
Is that an Iris in your pocket, or are you just glad to see me? :-)
Re: (Score:2, Informative)
Think broader than PowerPC. If you include the entire Power architecture (PowerPC, Cell and POWER), you have a huge market, mostly through POWER. POWER is the dominant architecture in big iron. Keep in mind SPARC and Itanium are only picking up the odd discarded breadcrumbs from POWER and they're still (barely) managing to stay in the game. The POWER market is absolutely essential. Basically any task you have that can't be easily distributed over a couple hundred puny x86 stations, POWER is the first place
Which POWER? (Score:2)
Re: (Score:1)
Re:So they were still alive? (Score:5, Interesting)
Re: (Score:2)
How about posting a detailed review of that laptop with pics?
It would be MUCH more interesting than most recent Slashdot articles!
Re: (Score:2)
Re: (Score:3)
I'm not going to write a review, at least for now. The first reason being, of course, laziness :-j However, I don't think a review at this point would make any justice.
It is, in some ways, an unfinished product. The stock distro fails to take full advantage of the hardware, IMHO, and doing the same with a custom distro is tricky because of some driver issues. Drivers are, in fact, on their way, so a review after a few months would make much more sense. Unfortunately, the machine will feel a little outdat
Re: (Score:2)
Correct. Some of the Chinese Android tablets use MIPS. I saw a cheap one but did not buy since I was not sure how much of the Android market applications would work on it.
Re: (Score:2)
It could have been the fault of the hacks who hacked together the image; but my one experience with a MIPS based Android tablet was... dire.
I didn't even get far enough to learn what Android market applications didn't work on it, since most of the default Android applications would crash if you tried to do anything serious with them(and by 'serious' I mean "load google.com with the Android browser", that sort of strenuous work).
I have no reason to assume that MIPS is fundamentally incapable of running Andro
Re: (Score:2)
Yeah, networking equipment, such as routers, and set top boxes often use MIPS. The CPU has unfortunately lost the game console market to POWER based CPUs (which are definitely worthy). Set top boxes should be adequate to keep MIPS going, if not growing, for a while.
I have read about some Android tablets being based on the XBurst, but have no idea about how they actually run.
Re:So they were still alive? (Score:5, Insightful)
Among those you missed are Power / PowerPC and
MIPS.
MIPS is very-much alive, thanks to China. They're actively developing home-grown MIPS CPUs, and paying license fees to MIPS as well. MIPS CPUs have always had higher DMIPS/MHz than ARM CPUs, and generally compete with PowerPC in the embedded space for anything needing a good bit of performance.
Cheap MIPS chips in China mean lots of inexpensive products are coming out with MIPS CPUs in them, such as the Alpha 400 and the Novo7
http://www.geeks.com/details.asp?invtid=ALPHA-400 [geeks.com]
http://www.mobile.slashdot.org/story/11/12/06/0359235/sub-100-android-40-tablet-coming-soon [slashdot.org]
POWER will be around for a good long time... IBM isn't willing to let their own platform and cash-cow go away, and they've sunk enough money into it to keep it highly competitive. PowerPC will likely be around a good long time as well... Freescale has quite a focus on their PowerPC chips, and their performance is damn respectable.
SPARC has a bigger customer base than Oracle. Hitachi will probably keep making them no matter what. They've made supercomputers out of them, and they can scale down to embedded applications quite easily.
Itanium is an interesting case... Everybody but HP who jumped onto Intel's 64-bit CPU has died a painful death (see: SGI). Their proprietary systems all require Itanium CPUs, with no sign of HP-UX, Tru64, OpenVMS, etc., being ported to any other architecture. This even though Intel deperately wants to kill off the architecture. HP has killed off all their proprietary CPU lines, and ported their software to Itantium with immense effort, so I don't see where they can go from here. ARM sure doesn't have the horsepower for high-end servers, and switching to x64 would eat their proprietary hardware margins, and probably make them a joke... SPARC and POWER seem like the only possible options, sort of resurrecting DEC's Alpha CPUs. It would be incredibly ironic.
SPARC and Itanium (Score:3)
Commenting on SPARC and Itanium...
There was a time when SPARC was popular way beyond just Sun Unixstations. They were used in some networking gear (mainly from Sun), and even in a laptop from Tadpole. Also, there were some other companies that made SPARCstations - Integrix and Tatung, and companies that made SPARCS other than Fujitsu or Sun - Ross Technologies. Not sure what they do now. But Suns would have been a great platform for Linux - a pity that Oracle didn't deem it fit to offer Oracle Linux a
Re: (Score:2)
It's fortunate that they did not... Linux thrives on commodity hardware, so x64 is the better option all-around. But back before x64 caught on, they might have gotten some traction.
Thanks for the correction. Guess I had hard drives on the brain...
Re: (Score:2)
I still maintain that OpenVMS and NonStop have a lot of life left in them, with no sign of any platform coming along to replace it. And I also believe that running the same OS on a new architecture and only needing to recompile your in-house applications is vastly easier than throwing away decades of work and rewriting everything for an entirely different OS, and Linux (or any other common OS you can name) is in no way even remotely a workable substitute for OpenVMS or NonStop.
I always thought it was a mist
Re: (Score:2)
I thought MIPS was dead.
You see them (or probably don't see them) mostly in embedded processors: network gear, printers etc., although ARM seem to be cutting into that market recently.
MIPS business model (Score:2)
Re: (Score:2)
I remember reading that some instructions on the 68000 like SUB or XOR were actually faster than CLR. The other companies were mashed up when Microsoft announced around the mid 1990's that "UNIX was legacy and Windows NT is the future". The manufacturers caved into the hype, brought out Windows boxes, and reduced themselves to what were the industry called "box packagers" who just took components, slapped them together, and bundled in an OS, then shipped the box. HP, DEC, Digital ended up just competing aga
Re: (Score:2)
Along with the other examples so far, MIPS was the architecture of choice for the supercomputer company SiCortex, which I was very sad to see go.
Re: (Score:2)
MIPS had been licensing its architectures rather than building their own, much like ARM. And also like ARM they had a 16-bit instruction set variant.
68000 is still around in the form of Coldfire system-on-chips from Freescale.
There's more to the world of CPUs than PCs and high end workstations.
Re: (Score:2)
Re: (Score:1)
Sad, really (Score:4, Insightful)
Of all the assembly languages I learned in college, MIPS was by far the best design.
Virtually any assembly is better than x86 (Score:5, Interesting)
Which has turned into a complete dogs dinner in the last couple of decades first with the FPU stack based instructions being glued to a register based architecture then all the endless psuedo parallel long word instruction sets intel have thrown at their cpus like confetti. Its long been said that compilers make a better job of generating assembler than a human, but these days I doubt if a human could even grok the entire x86 instruction set to start with. I gave up long ago.
Re: (Score:2, Funny)
Re: (Score:2)
Might have missed the comment title, there.
Re: (Score:1)
Thanks, I didn't realize that the MIPS architecture used the x86 instruction set.
It doesn't. He made that classic mistake of starting his comment in the title - which is annoying. Read his title as part of the comment and then what he says makes more sense.
Re: (Score:3)
That was the original battle of CISC vs. RISC in the 1990's. CPU designers did a survey of how often every instruction was referenced by various compiler writers. Most of the time it was move, arithmetic, function call and conditional branching instructions, with the more complex ones used very rarely if at all.
And it was just as efficient doing floating-point in software as it was implementing custom instructions, simply because it wasn't possible to get all the transistors onto a single chip. They had to
Re: (Score:1)
Nothing like writing MIPS assembly on an SGI using XSPIM
Re: (Score:2)
Re: (Score:2)
the original RISC? (Score:2)
Shirley that was Acorn with the original ARM in the 1980s??
Re:the original RISC? (Score:5, Funny)
Don't call me Shirley!
Re: (Score:1)
Shirley that was Acorn with the original ARM in the 1980s??
I think the point was that MIPS was the first company founded (1984) for the sole purpose of marketing RISC processors. The ARM equivalent would be ARM Holdings (est 1990.)
The first company to make a RISC processor would be IBM, which produced the ROMP in 1981.
Re: (Score:2)
ARM chips were made by Acorn originally (the A stood for "acorn"). ARM Holdings is just the name given to the spun-off chunk off Acorn that survived the company going bankrupt.
I believe Acorn were developing ARM as early as 1983 (so says Wikipedia), meaning they pre-date MIPS Technologies as a RISC-making company. Although Wikipedia tells me that the original academic origins of MIPS pre-date ARM, so the technology (rather than the company) can claim to be "first".
Re:the original RISC? (Score:4, Informative)
D. A. Patterson and C. H. Sequin, "RISC I: A reduced instruction set VLSI computer," in Proc. 8th Annu. Symp. Comput. Architecture, Minneapolis, MN, May 1981, pp. 443-457.
The architectural principle is generally credited to John Cocke, on the IBM 801 minicomputer, though others published first. Some (like me) might argue it goes back to Seymour Cray's designs at Control Data for the CDC6600 PPU but those definitely did not have a minimal instruction set, just a small one with a 6-bit opcode field.
Re: (Score:2)
The early days of MIPS (Score:5, Interesting)
Fun (if you're into that sort of thing) discussion of the early days of MIPS at the Computer History Museum in 2011. http://www.computerhistory.org/events/video/?videoid=3paiCK3dlK0 [computerhistory.org]
I've taken a few MIPS courses (Score:2)
Very fun introduction to assembly and the language has a small API. I even bought an ARM textbook to learn more about it. I'd just love to have an excuse to use it. Something very satisfying about working at that low a level.