The Athlon is faster than the G4. Not only because it has nearly twice the clock speed, but it also has a better branchpredictor, which is important because it has a very deep pipeline. G4, on the other hand only has a 4-stage pipeline. That's why it won't ramp beyond 500 Mhz. It also has more execution units and is capable of issuing 6 ops (Athlons internal RISC-like ops) per cycle whereas the G4 can issue 2 + one branch op. It has twice the L1 cache, although it's only 2-way associative where the G4 is 8-way. The Athlon has twice the bus speed. Etc, etc... Satisfied?
Steve said you couldn't by a faster personal computer. Higher clock speed !== faster execution.
No, but lower clock speed != faster execution either. Are you trying to tell me that a Athlon 800 is slower than a G4 450? Sorry, but it isn't. The Athlon would have better support for high-performance peripherals too (3d accelerators, faster memory, hard disks, etc...)
Why use a vector based windowing system on a monitor that can only display squares and blocks efficiently.
What do you do if you have a piece of graphic that you want to look the same on a 800x800 monitor, a 1600x1200 monitor and a 800 dpi laser printer? Display systems like DPS and Quartz do this easily and totally transparent to the developer.
This is why Display PS was so damn cool back then and why Quartz's Display PDF will be even cooler.
Why?? Apple dropped DPS mainly becuase of licensing issues. PDF is better than PS for what it was designed for (document distribution), but PS is more powerful. Everything that was in that PDF compositing demo can be done in DPS.
Apart from all the gui's-suck comments, quartz (the technology itself) is a brilliant piece of work! A built-in 2d renderer with support for PDF with all that entails. Yes, Aqua is glitzy and flashy and probably not as intuitive as the old interface, but the technology underneath it is what calls my attention.
*That* I consider innovation. I wish we had more of that in Linux...
Innovation? They bought NeXT, and wrapped it in a pretty GUI with lots of colors. You call that innovation?
I suppose someone will now announce a Open-Source project to *copy* Quartz'z functions...
"Free" Solaris comes with SunX, which has licensed Display Postscript, which kicks the shit out of that nasty hack display ghostscript.
Of course. SunX has DPS as an X extension, whereas Display Ghostscript is a DPS/NX agent, which means that it talks to the X server through the usual X protocol. It doesn't require a DPS extension at the server, so it can use any X display. Of course, this is much slower than having DPS integrated with the display server.
There is a free DPS/X extension under development, but it's in a very early stage. When it's ready, i'll change my sig.
This is great news. IMO, GNUStep is (will be) more important than KDE or GNOME, even though they get more attention. OpenStep is a well designed spec. Programming for it is great. The only thing that matches it could possibly be BeOS (both are object oriented and similar in many respects, though BeOS is multi-threaded to the max). I would say OPENSTEP is nicer, much because of the language (Objective-C). If you look closely at the BeOS API and then the OPENSTEP API, you notice that BeOS have borrowed a lot from NeXT. Did you know that BeOS originally didn't have a desktop (Just like NeXT), and instead a dock? (Just like NeXT)
There are other similarities too: NeXT was started by a former Apple Executive. Be was started by a former Apple Executive.
NeXT was originally making a computer and an OS. Be was originally making a computer and an OS.
NeXT had to leave the hardware bussines. Instead, they made an Intel version of their OS and became a pure software company. Be had to leave the hardware bussines. Instead, they made an Intel version of their OS and became a pure software company.
When Apple realised they couldn't make a modern OS by themselves, they decided that they would by an OS vendor. They choice was between NeXT and Be.
I was under the impression that smalltalk was a basis for extending C to "C with classes" and then subsequently to C++.
You're almost right!
Objective-C is like C with Smalltalk's object model, that's probably what you've heard of ("C with classes"). C++ is different, and uses an object model similar to that of Simula. An OO purist will tell you that it's not really object-oriented, but that's more of a religious issue.
One thing that we may find, however, is that a certain architecture is emulated better than x86 (i.e. the PowerPC, ARM, or Alpha architecture may be easier to translate into native VLIW) Therefore it may be a better idea to run Linx over PPC/ARM/Alpha code-morphing software on a Transmeta chip (or maybe just specific type of Transmeta chip works better, etc., etc.)
I have thought about this also. I don't want this old legacy stuff, although the core is all cutting-edge and all that.
You could design a "pseudo-architecture", that is optimized to be translated in realtime by code morphing software, rather than executed in hardware, like current architectures. That is what Elbrus did with their E2K.
There is alot of glitzy information now available about Crusoe VLIW, a core instruction set that is nothing like x86 and the code morphing software. But the actually technical nitty gritty seems to be lacking. Can a program get access to the core instruction set thus bypassing the code morphing?
No, no, please! That would be a disaster!! The hole point of this architecture is to get rid of this compability mess. They've already done two different instruction sets. Every new processor from Transmeta will likely use a new instruction set, optimized for whatever the processor is designed for. Don't you see the advantage of this?? Well Transmeta does, so they won't be releasing their compiler, or any specs.
Elbrus E2K uses a similar translation technique, and is said to have seven times the performance of an alpha. So it's not just about power, Transmeta will probably make faster chips in the future. It's probably not a coincidence that Dave Ditzel used to work with Elbrus in Russia back when he was working for Sun. Personally, i think Dave Ditzel is a bit embarassed that the Crusoes isn't faster. This guy used to do chips for UNIX workstations that nobody could afford, back when we where using c64 and spectrum. Look out for fast chips from Transmeta in the future. And as the article points out, there are some hints in that direction.
It's interesting to speculate on how the Crusoe will affect the Itanium, since in many ways they are very similar products.
They're both VLIW, but apart from that they're not very similar. Crusoe's performance is probably not very impressive, but it's har to know, since they haven't released ANY information whatsoever, other than "it will play DVD and MP3".
My understanding of Itanium is that it's a 64-bit VLIW processor that will have some capability for emulating x86 instructions. Sounds a lot like Crusoe.
Itanium's x86 emulation is hardware-based. Itanium does have software-based emulation, a remotely similar to Transmeta's, for running PA-RISC binaries.
More interesting is going to be watching what happens with the patents that Transmeta has on low-level emulation. Is Intel going to be able to keep Itanium from sucking on x86 code without getting a license from Transmeta?
As i understand the patents, they don't have any patents on the emulation, just the special hardware that supports it. The itanium does suck on x86, and it seems intel can't do much about it. They're of course hoping on good native support.
Depending on the performance of the Crusoe in the field, expect the Itanium (whenever it actually gets released) to sell for only a fraction of what Intel usually try to fleece off it's customers for new releases. If they don't, it's dead.
No. Itanium is initially targeted for real-high-end servers, and will NOT be cheap. It will be fast and power-hungry though. Crusoe is a COMPLETELY different market. You can't compare the two.
It's going to be slower than a P3, anyway, and companies are going to start asking if it's not better to just upgrade to the Crusoe instead.
Expect the itanium to be about twice as fast as a P3, and a lot more expensive. You don't upgrade to the Crusoe. Remember, both the P3 and the K7 will kick Crusoes ass, real hard, when it comes to performance.
I imagine the Alpha (horribly overpriced as it is) to take a knock from this, too.
Err...no. The alpha consumes about 60W IIRC. It's 60 times more than the Crusoe. And it's WAY faster. Again, completely different markets. You're saying that the fastest chip in the world will take a knock from a low-end chip from a start-up? I don't think so.
Isn't it easier jsut to type `make install' and have something downloaded, configured, installed and running right away! No conflicting libs, troubleshooting that much in depth just to get it working with different or missing libs, different distrobution, etc..?
I'd prefer to type "apt-get install" and have it downloaded, configured and installed, dependencies resolved, old config files saved/replaced/diffed/whatever.
FreeBSD is nice, but there's nothing like Debian. (My system reached 100% potato status last night, so i'm a little excited.)
Since I'm really not up to date with these things, anyone who is have any thoughts on whether this was pushed ahead because of the Debian code freeze? I would hope no, and it probably is mere coincidence, but just scanning the stories here, this stuck out at me.
Why should the FreeBSD maintainers care about the Debian freeze?
Why is everybody freezing in the winter??
BTW, the Transmeta homepage is updated. Now that is news!!
Mozilla is licensed under the Mozilla Public License, which is certified Open Source.
True for a significant part of it, but some of it is Netscape Public License. Although it doesn't restrict the use of the source code, it includes som clause that gives Netscape special privilegies.
Mozilla has nothing to do with Netscape whatsoever, except for a majority of the authors, the backwards-plugin functionality, and the fact that Mozilla will be incorporated into Netscape 5.
Oh, come on! Mozilla has a LOT to do with Netscape. Everybody knows that.
Mozilla is quite modular.
True.
Mozilla is completely cross platform. If the OS exists, Mozilla will probably have no problems running on it.
You have to port it, and it's not trivial. Granted, 97% of mozilla is completely platform-independent.
Mozilla is a very insanely complex program, which is probably why it doesn't have a lot of outside hackers. It's not a weekend project, and the codebase is enormously sophisticated.
True.
Mozilla will do to-the-standard CSS1, HTML4.0, DOM Level 1, XML, a great majority of CSS2 (although this is not promised), and the latest bastard variant of Javascript.
IIRC, it will do ECMAScript, which is the standardized version.
Mozilla will not have crypo. That's for binary-only vendors, or your own project.
True.
The entire UI (ENTIRE) will be themeable. So if you don't care for the UI, don't bitch, because theme support is coming soon, and you will be able to write your own.
It has been themable for ages.
Mozilla is going to be large because that is how it is. Don't bitch about bloat, because none of the weekend-project HTML widgets your favorite toolkit sports are able to do everything Mozilla does yet.
It's not going to be large compared to IE or NS, but maybe if you compare to lynx...
Mozilla is not going to force any particular Java Virtual Machine, HTML editor, or mailer on you. Although the comes-with editor and mailer are extremely nice this time around.
Well, those components are in the binaries, so you have to download them. Not the JVM, but messenger and composer. However, they don't get loaded before you use them, and you can compile your own version without them if you want.
Mozilla will be a compile-it-yourself type thing if you so desire.
Slashdot Mirror
The Athlon is faster than the G4. Not only because it has nearly twice the clock speed, but it also has a better branchpredictor, which is important because it has a very deep pipeline. G4, on the other hand only has a 4-stage pipeline. That's why it won't ramp beyond 500 Mhz.
It also has more execution units and is capable of issuing 6 ops (Athlons internal RISC-like ops) per cycle whereas the G4 can issue 2 + one branch op. It has twice the L1 cache, although it's only 2-way associative where the G4 is 8-way. The Athlon has twice the bus speed. Etc, etc... Satisfied?
Steve said you couldn't by a faster personal computer. Higher clock speed !== faster execution.
No, but lower clock speed != faster execution either. Are you trying to tell me that a Athlon 800 is slower than a G4 450? Sorry, but it isn't. The Athlon would have better support for high-performance peripherals too (3d accelerators, faster memory, hard disks, etc...)
Why use a vector based windowing system on a monitor that can only display squares and blocks efficiently.
What do you do if you have a piece of graphic that you want to look the same on a 800x800 monitor, a 1600x1200 monitor and a 800 dpi laser printer? Display systems like DPS and Quartz do this easily and totally transparent to the developer.
This is why Display PS was so damn cool back then and why Quartz's Display PDF will be even cooler.
Why?? Apple dropped DPS mainly becuase of licensing issues. PDF is better than PS for what it was designed for (document distribution), but PS is more powerful. Everything that was in that PDF compositing demo can be done in DPS.
Apart from all the gui's-suck comments, quartz (the technology itself) is a brilliant piece of work! A built-in 2d renderer with support for PDF with all that entails. Yes, Aqua is glitzy and flashy and probably not as intuitive as the old interface, but the technology underneath it is what calls my attention.
*That* I consider innovation. I wish we had more of that in Linux...
Innovation? They bought NeXT, and wrapped it in a pretty GUI with lots of colors. You call that innovation?
I suppose someone will now announce a Open-Source project to *copy* Quartz'z functions...
Do you mean something like DGS or X/DPS???
The "3rd generation" GUI's has been around for at least a decade...
"Free" Solaris comes with SunX, which has licensed Display Postscript, which kicks the shit out of that nasty hack display ghostscript.
Of course. SunX has DPS as an X extension, whereas Display Ghostscript is a DPS/NX agent, which means that it talks to the X server through the usual X protocol. It doesn't require a DPS extension at the server, so it can use any X display. Of course, this is much slower than having DPS integrated with the display server.
There is a free DPS/X extension under development, but it's in a very early stage. When it's ready, i'll change my sig.
This is great news. IMO, GNUStep is (will be) more important than KDE or GNOME, even though they get more attention. OpenStep is a well designed spec. Programming for it is great. The only thing that matches it could possibly be BeOS (both are object oriented and similar in many respects, though BeOS is multi-threaded to the max).
I would say OPENSTEP is nicer, much because of the language (Objective-C).
If you look closely at the BeOS API and then the OPENSTEP API, you notice that BeOS have borrowed a lot from NeXT. Did you know that BeOS originally didn't have a desktop (Just like NeXT), and instead a dock? (Just like NeXT)
There are other similarities too:
NeXT was started by a former Apple Executive.
Be was started by a former Apple Executive.
NeXT was originally making a computer and an OS.
Be was originally making a computer and an OS.
NeXT had to leave the hardware bussines. Instead, they made an Intel version of their OS and became a pure software company.
Be had to leave the hardware bussines. Instead, they made an Intel version of their OS and became a pure software company.
When Apple realised they couldn't make a modern OS by themselves, they decided that they would by an OS vendor. They choice was between NeXT and Be.
What we really want is portman.
Hey, man is already available on Linux. There is no need to port man.
I was under the impression that smalltalk was a basis for extending C to "C with classes" and then subsequently to C++.
You're almost right!
Objective-C is like C with Smalltalk's object model, that's probably what you've heard of ("C with classes"). C++ is different, and uses an object model similar to that of Simula. An OO purist will tell you that it's not really object-oriented, but that's more of a religious issue.
My norwegian is not the best in the world, but here it comes:
Should it be illegal to crack protection codes?
A. Yes, that's why the codes are there.
B. No, the media giants is overprotecting themselves.
C. Only if you use it for commercial purposes.
One thing that we may find, however, is that a certain architecture is emulated better than x86 (i.e. the PowerPC, ARM, or Alpha architecture may be easier to translate into native VLIW) Therefore it may be a better idea to run Linx over PPC/ARM/Alpha code-morphing software on a Transmeta chip (or maybe just specific type of Transmeta chip works better, etc., etc.)
I have thought about this also. I don't want this old legacy stuff, although the core is all cutting-edge and all that.
You could design a "pseudo-architecture", that is optimized to be translated in realtime by code morphing software, rather than executed in hardware, like current architectures. That is what Elbrus did with their E2K.
There is alot of glitzy information now available about Crusoe VLIW, a core instruction set that is nothing like x86 and the code morphing software. But the actually technical nitty gritty seems to be lacking. Can a program get access to the core instruction set thus bypassing the code morphing?
No, no, please! That would be a disaster!! The hole point of this architecture is to get rid of this compability mess. They've already done two different instruction sets. Every new processor from Transmeta will likely use a new instruction set, optimized for whatever the processor is designed for. Don't you see the advantage of this?? Well Transmeta does, so they won't be releasing their compiler, or any specs.
Elbrus E2K uses a similar translation technique, and is said to have seven times the performance of an alpha. So it's not just about power, Transmeta will probably make faster chips in the future. It's probably not a coincidence that Dave Ditzel used to work with Elbrus in Russia back when he was working for Sun. Personally, i think Dave Ditzel is a bit embarassed that the Crusoes isn't faster. This guy used to do chips for UNIX workstations that nobody could afford, back when we where using c64 and spectrum. Look out for fast chips from Transmeta in the future. And as the article points out, there are some hints in that direction.
It's interesting to speculate on how the Crusoe will affect the Itanium, since in many ways they are very similar products.
They're both VLIW, but apart from that they're not very similar. Crusoe's performance is probably not very impressive, but it's har to know, since they haven't released ANY information whatsoever, other than "it will play DVD and MP3".
My understanding of Itanium is that it's a 64-bit VLIW processor that will have some capability for emulating x86 instructions. Sounds a lot like Crusoe.
Itanium's x86 emulation is hardware-based. Itanium does have software-based emulation, a remotely similar to Transmeta's, for running PA-RISC binaries.
More interesting is going to be watching what happens with the patents that Transmeta has on low-level emulation. Is Intel going to be able to keep Itanium from sucking on x86 code without getting a license from Transmeta?
As i understand the patents, they don't have any patents on the emulation, just the special hardware that supports it. The itanium does suck on x86, and it seems intel can't do much about it. They're of course hoping on good native support.
Depending on the performance of the Crusoe in the field, expect the Itanium (whenever it actually gets released) to sell for only a fraction of what Intel usually try to fleece off it's customers for new releases. If they don't, it's dead.
No. Itanium is initially targeted for real-high-end servers, and will NOT be cheap. It will be fast and power-hungry though. Crusoe is a COMPLETELY different market. You can't compare the two.
It's going to be slower than a P3, anyway, and companies are going to start asking if it's not better to just upgrade to the Crusoe instead.
Expect the itanium to be about twice as fast as a P3, and a lot more expensive. You don't upgrade to the Crusoe. Remember, both the P3 and the K7 will kick Crusoes ass, real hard, when it comes to performance.
I imagine the Alpha (horribly overpriced as it is) to take a knock from this, too.
Err...no. The alpha consumes about 60W IIRC. It's 60 times more than the Crusoe. And it's WAY faster. Again, completely different markets. You're saying that the fastest chip in the world will take a knock from a low-end chip from a start-up? I don't think so.
Isn't it easier jsut to type `make install' and have something downloaded, configured, installed and running right away! No conflicting libs, troubleshooting that much in depth just to get it working with different or missing libs, different distrobution, etc..?
I'd prefer to type "apt-get install" and have it downloaded, configured and installed, dependencies resolved, old config files saved/replaced/diffed/whatever.
FreeBSD is nice, but there's nothing like Debian. (My system reached 100% potato status last night, so i'm a little excited.)
[BAD JOKE]
London drugs??
Isn't that where they sell packaged versions of Enlightenment and Xfree86??
[/BAD JOKE]
And I'm not sure where you got that 15 day quote.
Uhmm... Slashdot? So it must be true.
Since I'm really not up to date with these things, anyone who is have any thoughts on whether this was pushed ahead because of the Debian code freeze?
I would hope no, and it probably is mere coincidence, but just scanning the stories here, this stuck out at me.
Why should the FreeBSD maintainers care about the Debian freeze?
Why is everybody freezing in the winter??
BTW, the Transmeta homepage is updated. Now that is news!!
don't moderate me down, it's a waste of time!
Don't worry, i think the moderators are out of points. It's called a DOM (Denial of moderation) attack.
Anyone know what's new in 4.0??
Anyway, what's up with that two week freeze period? Isn't that a little short?
So write a new interface, not a new tool.
WTF has this to do with C++?? Do you suggest that Adobe rewrite Premiere in Python??
It should be a good idea too confirm this before donating any money.
I think he's hunting for (Score:3, interesting).
Comments:
Mozilla is licensed under the Mozilla Public License, which is certified Open Source.
True for a significant part of it, but some of it is Netscape Public License. Although it doesn't restrict the use of the source code, it includes som clause that gives Netscape special privilegies.
Mozilla has nothing to do with Netscape whatsoever, except for a majority of the authors, the backwards-plugin functionality, and the fact that Mozilla will be incorporated into Netscape 5.
Oh, come on! Mozilla has a LOT to do with Netscape. Everybody knows that.
Mozilla is quite modular.
True.
Mozilla is completely cross platform. If the OS exists, Mozilla will probably have no problems running on it.
You have to port it, and it's not trivial. Granted, 97% of mozilla is completely platform-independent.
Mozilla is a very insanely complex program, which is probably why it doesn't have a lot of outside hackers. It's not a weekend project, and the codebase is enormously sophisticated.
True.
Mozilla will do to-the-standard CSS1, HTML4.0, DOM Level 1, XML, a great majority of CSS2 (although this is not promised), and the latest bastard variant of Javascript.
IIRC, it will do ECMAScript, which is the standardized version.
Mozilla will not have crypo. That's for binary-only vendors, or your own project.
True.
The entire UI (ENTIRE) will be themeable. So if you don't care for the UI, don't bitch, because theme support is coming soon, and you will be able to write your own.
It has been themable for ages.
Mozilla is going to be large because that is how it is. Don't bitch about bloat, because none of the weekend-project HTML widgets your favorite toolkit sports are able to do everything Mozilla does yet.
It's not going to be large compared to IE or NS, but maybe if you compare to lynx...
Mozilla is not going to force any particular Java Virtual Machine, HTML editor, or mailer on you. Although the comes-with editor and mailer are extremely nice this time around.
Well, those components are in the binaries, so you have to download them. Not the JVM, but messenger and composer. However, they don't get loaded before you use them, and you can compile your own version without them if you want.
Mozilla will be a compile-it-yourself type thing if you so desire.
True.