Linux has excellent threading support, and has had it since at least 1.3, maybe even 1.1. And it is not micro-kernel-based, though MkLinux is a Linux port to the Mach microkernel. And real threading can be quite easily be achieved on top of a microkernel, cf NextSTEP, WinNT, QNX, MkLinux, etc. So I'm afraid all your assumptions were wrong (including Linux being (originally) designed with portability in mind).:)
So where are people supposed to get eight-way (or four-way or two-way) Athlong motherboards and systems? The Profusion is out now, Athlon SMP is expected to debut on the market next year, from what I've seen.
What part of raw CPU is it that implies need for >32GB memory? Or >4GB, for that matter? There are plenty of tasks that require lots of CPU but will do quite fine on a 32 bit platform. Like, say, compiling kernels.;)
I can see using an old computer, but one with 256 bytes of memory? How do you word process on that, when you can't even store a full page in memory? Dump it to tape each line? Someone should interview this guy.
Not that benchmarks are really very relevant, but the Alpha 21264 at 667MHz beats a PIII Xeon 550MHz by nearly 4x on SPECMark, and a 600MHz Athlon by about 2.5x. I'd say that's a pretty significant speed advantage.
Re:Er, there is an alternative explanation...
on
Ixnay WinNT on Alpha
·
· Score: 1
The people who they're trying to sell Alpha systems to and the people who buy Celerons to run overclocked in dual boards are generally not the same people. And anyway, to get close to, say, an XP1000 at 667MHz, you need a quad 550MHz PIII Xeon (and that is assuming your task scales to at least 80% on four processors), so at those performance levels, price/performance for the Alpha is better than for Intel (quad Xeon boxes are very expensive).
You're forgetting the IA64 architecture here, and the M68k (like the Dragonball in Palms). There are probably other chips (vapor or existing) that aren't marketed as RISC, as well.
I'm not sure how much of the kernel (if any) can be pre-empted, but the I thought the general definition of pre-emptive multitasking was that tasks are pre-empted, and that they certainly are. And Linux does have kernel-level threads, at least since 2.0, so it a multi-threaded app will scale over multiple processors.
SGI are trying to increase their sales volume with the Visual Workstations, not just replace sales of IRIX boxes on a one-for-one basis. Therefore this may well increase the market share of SGI-built Unix-like OS running computers, despite SGI selling computers with NT on as well. Assuming they succeed.
Can't work that way, the DIVX player is supposed to be usable even if your phone service is down for a while. Therefore the decryption must take place with information from the disc and the player only. Unless Circuit City is lying about being able to play discs even if you can't phone in right then, of course. However, since the decryption is probably done by dedicated custom hardware, it's a bit difficult to reverse engineer it.
Because they can no longer terminate the license at will. Unless you want to be able to sue Apple for patent infringement, I see no danger of termination of the license. If you fear that you will have to sue Apple for patent infringement, stay away from the APSL. Otherwise, the termination clause is no problem.
Do you have any idea how little power is needed for something like this? It probably uses something on the order of one or a few hundred mW, which is lost in the noise from processor, memory and hard drive in a laptop. Might matter in a palmtop, but then this mouse isn't intended to be used with palmtops. Then again, this is a Microsoft product, so who knows?
They may reuse the ideas, but they can't reuse the content, and it is the content that is most expensive to produce, not the ideas. Especially not the ideas that drive most TV shows.
There is a huge difference between the freeness of TV programs and free software. Think free speech, not free beer, as they say. Though that is not really the important difference. The important difference is in the degree of reuse that is possible. What makes this difference important is that a TV show can't really make use of content from earlier TV shows to any large extent, because the audience will not be as willing to spend their time watching old content. Therefore every TV show has to have mostly new content, which leads to high production costs. With free software, we have a situation where production costs will go down the more free software there is. There will still be a cost, of course, but it is quite possible that this cost will be low enough that we can entirely scrap proprietary software development. This is what the free software people are hoping and believing will happen, I think. At least it is what I am hoping for.
I wasn't being sarcastic about people who think there is a place for proprietary software. I was being sarcastic about people who paint free software (and in this case, cheap hardware) as being somehow evil and opposed to choice, the free market, etc. If a company can't take the competition from free software or cheap hardware, it is they who need to change, not the people providing free software or cheap hardware.
Yes! He's right! Let's all petition our respective governments to mandate a minimum price on software and hardware, to protect the industry. Otherwise all innovation will end and we will all live in straw huts, eating mud, within ten years!
There are specs for 3D rendering in there. Not full specs, but enough to get decent 3D, I think. What is missing is specs for the WARP Engine, which is some sort of processor for doing geometry setup and stuff. Or that's what it sounds like.
They're a hell of a lot faster at drawing, too. And a large number of text processing tasks. And image processing. And a lot of other things. Of course, you can call all this maths, because computers are mathematical machines. OTOH, it may be possible to make a mathematical model of the brain, too.
Slashdot Mirror
Linux has excellent threading support, and has had it since at least 1.3, maybe even 1.1. And it is not micro-kernel-based, though MkLinux is a Linux port to the Mach microkernel. And real threading can be quite easily be achieved on top of a microkernel, cf NextSTEP, WinNT, QNX, MkLinux, etc. :)
So I'm afraid all your assumptions were wrong (including Linux being (originally) designed with portability in mind).
So where are people supposed to get eight-way (or four-way or two-way) Athlong motherboards and systems? The Profusion is out now, Athlon SMP is expected to debut on the market next year, from what I've seen.
What part of raw CPU is it that implies need for >32GB memory? Or >4GB, for that matter? ;)
There are plenty of tasks that require lots of CPU but will do quite fine on a 32 bit platform. Like, say, compiling kernels.
I can see using an old computer, but one with 256 bytes of memory? How do you word process on that, when you can't even store a full page in memory? Dump it to tape each line? Someone should interview this guy.
Not that benchmarks are really very relevant, but the Alpha 21264 at 667MHz beats a PIII Xeon 550MHz by nearly 4x on SPECMark, and a 600MHz Athlon by about 2.5x. I'd say that's a pretty significant speed advantage.
The people who they're trying to sell Alpha systems to and the people who buy Celerons to run overclocked in dual boards are generally not the same people.
And anyway, to get close to, say, an XP1000 at 667MHz, you need a quad 550MHz PIII Xeon (and that is assuming your task scales to at least 80% on four processors), so at those performance levels, price/performance for the Alpha is better than for Intel (quad Xeon boxes are very expensive).
You're forgetting the IA64 architecture here, and the M68k (like the Dragonball in Palms). There are probably other chips (vapor or existing) that aren't marketed as RISC, as well.
Neither Java nor Visual Basic are functional. Look at LISP or Haskell or something for functional languages. They rock!
Microsoft always bashes ideas a year or so before they decide they were really good. It's a tradition.
I'm not sure how much of the kernel (if any) can be pre-empted, but the I thought the general definition of pre-emptive multitasking was that tasks are pre-empted, and that they certainly are. And Linux does have kernel-level threads, at least since 2.0, so it a multi-threaded app will scale over multiple processors.
SGI are trying to increase their sales volume with the Visual Workstations, not just replace sales of IRIX boxes on a one-for-one basis. Therefore this may well increase the market share of SGI-built Unix-like OS running computers, despite SGI selling computers with NT on as well. Assuming they succeed.
Hint: Creative own E-MU. If they want specs, they can get them without an NDA.
When, where and to who did they say this?
Fraunhofer only holds the patent on the encoding technique, decoding is not covered by the patent, so they can't do anything about players.
Can't work that way, the DIVX player is supposed to be usable even if your phone service is down for a while. Therefore the decryption must take place with information from the disc and the player only. Unless Circuit City is lying about being able to play discs even if you can't phone in right then, of course.
However, since the decryption is probably done by dedicated custom hardware, it's a bit difficult to reverse engineer it.
Because they can no longer terminate the license at will. Unless you want to be able to sue Apple for patent infringement, I see no danger of termination of the license.
If you fear that you will have to sue Apple for patent infringement, stay away from the APSL. Otherwise, the termination clause is no problem.
Do you have any idea how little power is needed for something like this? It probably uses something on the order of one or a few hundred mW, which is lost in the noise from processor, memory and hard drive in a laptop. Might matter in a palmtop, but then this mouse isn't intended to be used with palmtops.
Then again, this is a Microsoft product, so who knows?
They may reuse the ideas, but they can't reuse the content, and it is the content that is most expensive to produce, not the ideas. Especially not the ideas that drive most TV shows.
There is a huge difference between the freeness of TV programs and free software. Think free speech, not free beer, as they say. Though that is not really the important difference. The important difference is in the degree of reuse that is possible.
What makes this difference important is that a TV show can't really make use of content from earlier TV shows to any large extent, because the audience will not be as willing to spend their time watching old content. Therefore every TV show has to have mostly new content, which leads to high production costs. With free software, we have a situation where production costs will go down the more free software there is.
There will still be a cost, of course, but it is quite possible that this cost will be low enough that we can entirely scrap proprietary software development. This is what the free software people are hoping and believing will happen, I think. At least it is what I am hoping for.
I wasn't being sarcastic about people who think there is a place for proprietary software. I was being sarcastic about people who paint free software (and in this case, cheap hardware) as being somehow evil and opposed to choice, the free market, etc. If a company can't take the competition from free software or cheap hardware, it is they who need to change, not the people providing free software or cheap hardware.
Yes! He's right! Let's all petition our respective governments to mandate a minimum price on software and hardware, to protect the industry. Otherwise all innovation will end and we will all live in straw huts, eating mud, within ten years!
There are specs for 3D rendering in there. Not full specs, but enough to get decent 3D, I think.
What is missing is specs for the WARP Engine, which is some sort of processor for doing geometry setup and stuff. Or that's what it sounds like.
Building Mozilla with full debugging takes ~320 megs for me. That's too much for me to participate in development, but it's not as bad as 1 gig.
The Netscape toolbar allows you to have folders that behave like hierarchical menus, at least the Motif version does.
They're a hell of a lot faster at drawing, too. And a large number of text processing tasks. And image processing. And a lot of other things. Of course, you can call all this maths, because computers are mathematical machines. OTOH, it may be possible to make a mathematical model of the brain, too.