Here's what I propose.
A general set of benchmarks for UNIX-like systems (Linux, *BSD, OS X, Solaris, AIX, HP-UX) that is GPL'd and tests the functionality common to all platforms. It would comprise a core set of microbenchmarks written in C that would test integer performance, floating-point performance, memory bandwidth, cache performance, etc. Next would be processes and threads, disk I/O and caching, TCP/IP networking. On top of that would be applications such as databases, webserving, number-crunching (image and sound processing, streaming), Java VM performace, X11, some high-level interpreted languages such as PERL and Python, shell performance etc. Each area would have a score, and I reckon we should normalise each score to a low-end reference platform, e.g. a PIII-1000, 512MB RAM, IDE 5400 RPM disk etc.
It'll be a huge project but I can donate a few lines of code here and there.
Forget Windows. It's going away soon.
I hope so. I am looking forward to getting 200FPS (1024x768 32 bit colour) in Quake 3 using software rendering mode:-)
I'm still using a K6-2 500 for my main box. It has 512MB of ECC RAM at 100MHz and the level 2 cache on the board is 2MB. The CPU has 64k of level 1 cache (that's 8 times that of the Pentium IV, which only has 8k!). I used to have an NVidia TNT2 Ultra (the AGP is only 2x) and got about 30fps on quake 3 at 1027x768, 32-bit colour. I upgraded to a GeForce4 MX440 and now I get around 45fps at the same resolution, and ~35fps at 1600x1200, 32-bit colour. I've found that cache size and hard disk performace are very important in terms of overall system performance. A fast hard disk is a great performance boost. Linux kernels (2.4.20) take around 12 minutes to build on this machine on a 7200rpm Western Digital hard disk on an ATA100 interface. So, the whole thing's quite useable still and it's in its fourth year of operation (it started out with 64MB RAM, 2.2 kernel and a 400MHz K6-2).
I'd say that was informative and maybe insightful. Someone needs to get their finger out and develop a comprehensive, relevant and useful set of Linux (and UNIX in general) benchmarks for these platforms, especially since Linux is gaining so much market share. Just how does a SQL server benchmark on Windows 2k relate to what I do on my AMD Slackware box? How does a Windows game using Direct X have any relevance to OpenGL applications? So, come on, who's going to do it? Who's going to give us some Free (GPL preferably) benchmarking software for Linux and other UNIX-like operating systems? I can write C. I'd gladly contribute a few hundred lines of code.
..Sigh here we go:
I take it you have a vested interest in itanic?
The people who buy these things know this and can deal with it. Remember, these are not crammed in like Mini-ATX towers (like the one under your desk). They're deployed by professionals in a professional environment with standards for this stuff.
For your information I have a 64-bit dual processor Sun Ultra 60 and a Dell PC under my desk. I also run several multi-processor 64-bit servers used daily for building Open Source and Free software.
See above. Porting will and has happened. If the logic can be presented that the company will either save or gain money by upgrading to this hardware then it will happen. It just makes business sense.
What logic is this? How can completely recompiling, and retesting and reimplementing your infrastructure save money? It's not obvious. Please explain. PA RISC who? Alpha who? You need to understand that IT departments invest for the long-haul, you won't see too many more shiney new Alphas being purchased not because they're bad but because C[T|I]Os know they're a doomed platform.
And why are they doomed? Not for any reason other than they provided the biggest threat to the (inferior) itanic, so they were artificially removed from the market place to try to boost itanic's position. There are a whole load of angry and dis-satisfied former HP, Compaq and DEC customers who are being forced to change simply for change's sake, because it suits intel's plans for world domination, and not for solid technical reasons. It itanic is so wonderful, it should have been left to compete on its supposed technical merits. I'm afraid this is a case where the politicians, marketeers and sales droids have ruined a lot of good technology and a lot of good business.
You don't need to say it because you can't say it. At least not yet. I too doubt that Itanium will be a hugh smash in the commodity arena. Not because it's inferior (I'm not arguing that either way) but because the money isn't there.
Athlon 64 will succeed simply because it is an evolutionary improvement on existing technology. There is nothing to lose on the customer's part. They can go on using their existing software, only a bit faster than before.
Intel should not, as a matter of course, invest heavily on gcc for itanium. They have their own highly optimising compilers that they can sell and make money on. Itanium is not aimed at the geek Open Source/Free Software/hobbyist market. It is aimed at big corporations with big IT budgets who buy an IT solution, comprising an enterprise-level OS and commercial applications, licensed on a per-user or per-cpu (or a per-mips in IBM's case) basis. intel will license its compiler to the application developers i.e. the Oracles and SAPs of this world. That's where the money is, and you can bet that's what intel is doing.
It'll never be the success that intel and HP envision for it and here's why. First, it's too hot and too expensive. Secondly is doesn't have any applications. I don't mean Gnome and KDE, I mean the sort of applications that big corporations run. Thirdly it isn't backwards-compatible with any existing architectures. You can't just take your binaries over and run them, at least not at full speed. Applications will need to be ported and retested. This is not insignificant in time, effort and cost. Fourthly, most people who want 64-bit in the corporate world already have it in the form of SPARC, Power, PA RISC and Alpha. Why should they change to an unproven, immature "jam tomorrow" architecture given their working and reliable systems already in use? I'm afraid intel missed the boat by about 10 years. If they'd brought out a 64-bit RISC at the same time as SPARC, MIPS, Alpha and Power they might have stood a chance. It's a turkey, and apart from a few niches (e.g. number-crunching super computers) it's doomed to failure. I don't even need to mention how Athlon 64/Opteron will eat its lunch in the commodity sector of the market.
So, PCs have 3DNow!, SSE and SSE2 depending on what processor you have. I have observed factor-of-ten speed-ups of certain code using hand-crafted 3DNow! vs. GCC floating-point. I wonder how fast his algorithm would be if implemented in 3Dnow! or SSE? I bet my rusty old K6-2/500 could put in a reasonable showing at his benchmark.
Well, Linux may not be "lock-upable" but many of the libraries which run on top of it which are used in applications are not completely "Free Software" and by that I mean not GPL or LGPL. Some have BSD/X11-type licenses. One could envision a scenario in which Microsoft could embrace one of these libraries (e.g. from mono) and add their own code. They would be perfectly within their rights to do so. Applications could be written to use that code. Microsoft could then stop distributing the source, supplying binary-only libraries. Then they could break the API and add their own undocumented stuff. Or even worse, they could claim that some of it was patented and start demanding that everyone who uses it pay them a license fee. At this point they have rendered a whole load of other people's software useless, just as they have done before to Windows developers, effectively wiping out the competition.
Other languages which use the JVM
on
The Future of Java?
·
· Score: -1, Redundant
Yes, and before then things will really hot up. India is also planning on sending humans into space. The next race for the moon could be between India and China. Will the US go one better? Or will they continue to rest on their laurels?
China wants to send a man into space in October this year
"x86 compatibility is worthless ina high end 64-bit machine" - Indeed. That's why AMD is targetting the low end. The medium to high end has already been 64-bit for the best part of a decade with well established, mature technologies, especially UltraSPARC. itanium just can not hope to take any significant portion of this market simply because it is 10 years too late. No one wants to throw out their entire working, reliable infrastructure for something new and unproven. They'd have to change OS, port all their software, re-test everything, retrain their technical staff, buy new middle ware and applications, upgrade their air-conditioning etc. etc. and all for what? Intel's ambitions? Where is the benfit to the customer?
I can't wait for the first Scotsperson to reach earth orbit, amid a fanfare of bagpipes, accordions and fiddles. The tartan of the Black Watch will be worn with pride by said astronaut as she circles the earth at 250 miles up, bag of liquidised haggis in hand being consumed through a straw. As the shuttle passes Edinburgh the Capitol's inhabitants and Scots from all over the world will raise their glasses of Glenmorangie in a toast to celebrate this gloroius achievement. A gigantic Saltire will be floated on Loch Lomond, visible from space. Burns will be quoted. Jimmy Shand will be exhumed and Ricky Fulton will provide the comedy. Oh wait, that's Hogmanay...
They should do all their praying _before_ launch, after all, that's the most dangerous part. In fact, why doesn't NASA employ a load of people representing all of the major world religions to pray for the shuttle's safety before each lift-off?
Maybe shareware on Linux would be a way to distribute relatively inexpensive niche programs that use patented algorithms, such as wavelet encoding for image compression. Because money is involved, the patent could be used and paid for legitimately. Obviously this doesn't help the fight against software patents...
That's nonsense. They all ran at the same speed, 3.54MHz. The ZX81, the Spectrum's predecessor was slower, at 3.25MHz. I had (still have) a ZX81 and a Spectrum 128.
Slashdot Mirror
Here's what I propose. A general set of benchmarks for UNIX-like systems (Linux, *BSD, OS X, Solaris, AIX, HP-UX) that is GPL'd and tests the functionality common to all platforms. It would comprise a core set of microbenchmarks written in C that would test integer performance, floating-point performance, memory bandwidth, cache performance, etc. Next would be processes and threads, disk I/O and caching, TCP/IP networking. On top of that would be applications such as databases, webserving, number-crunching (image and sound processing, streaming), Java VM performace, X11, some high-level interpreted languages such as PERL and Python, shell performance etc. Each area would have a score, and I reckon we should normalise each score to a low-end reference platform, e.g. a PIII-1000, 512MB RAM, IDE 5400 RPM disk etc. It'll be a huge project but I can donate a few lines of code here and there. Forget Windows. It's going away soon.
I'm still using a K6-2 500 for my main box. It has 512MB of ECC RAM at 100MHz and the level 2 cache on the board is 2MB. The CPU has 64k of level 1 cache (that's 8 times that of the Pentium IV, which only has 8k!). I used to have an NVidia TNT2 Ultra (the AGP is only 2x) and got about 30fps on quake 3 at 1027x768, 32-bit colour. I upgraded to a GeForce4 MX440 and now I get around 45fps at the same resolution, and ~35fps at 1600x1200, 32-bit colour. I've found that cache size and hard disk performace are very important in terms of overall system performance. A fast hard disk is a great performance boost. Linux kernels (2.4.20) take around 12 minutes to build on this machine on a 7200rpm Western Digital hard disk on an ATA100 interface. So, the whole thing's quite useable still and it's in its fourth year of operation (it started out with 64MB RAM, 2.2 kernel and a 400MHz K6-2).
I'd say that was informative and maybe insightful. Someone needs to get their finger out and develop a comprehensive, relevant and useful set of Linux (and UNIX in general) benchmarks for these platforms, especially since Linux is gaining so much market share. Just how does a SQL server benchmark on Windows 2k relate to what I do on my AMD Slackware box? How does a Windows game using Direct X have any relevance to OpenGL applications? So, come on, who's going to do it? Who's going to give us some Free (GPL preferably) benchmarking software for Linux and other UNIX-like operating systems? I can write C. I'd gladly contribute a few hundred lines of code.
I take it you have a vested interest in itanic?
The people who buy these things know this and can deal with it. Remember, these are not crammed in like Mini-ATX towers (like the one under your desk). They're deployed by professionals in a professional environment with standards for this stuff.
For your information I have a 64-bit dual processor Sun Ultra 60 and a Dell PC under my desk. I also run several multi-processor 64-bit servers used daily for building Open Source and Free software.
See above. Porting will and has happened. If the logic can be presented that the company will either save or gain money by upgrading to this hardware then it will happen. It just makes business sense.
What logic is this? How can completely recompiling, and retesting and reimplementing your infrastructure save money? It's not obvious. Please explain.
PA RISC who? Alpha who? You need to understand that IT departments invest for the long-haul, you won't see too many more shiney new Alphas being purchased not because they're bad but because C[T|I]Os know they're a doomed platform.
And why are they doomed? Not for any reason other than they provided the biggest threat to the (inferior) itanic, so they were artificially removed from the market place to try to boost itanic's position. There are a whole load of angry and dis-satisfied former HP, Compaq and DEC customers who are being forced to change simply for change's sake, because it suits intel's plans for world domination, and not for solid technical reasons. It itanic is so wonderful, it should have been left to compete on its supposed technical merits. I'm afraid this is a case where the politicians, marketeers and sales droids have ruined a lot of good technology and a lot of good business.
You don't need to say it because you can't say it. At least not yet. I too doubt that Itanium will be a hugh smash in the commodity arena. Not because it's inferior (I'm not arguing that either way) but because the money isn't there.
Athlon 64 will succeed simply because it is an evolutionary improvement on existing technology. There is nothing to lose on the customer's part. They can go on using their existing software, only a bit faster than before.
Intel should not, as a matter of course, invest heavily on gcc for itanium. They have their own highly optimising compilers that they can sell and make money on. Itanium is not aimed at the geek Open Source/Free Software/hobbyist market. It is aimed at big corporations with big IT budgets who buy an IT solution, comprising an enterprise-level OS and commercial applications, licensed on a per-user or per-cpu (or a per-mips in IBM's case) basis. intel will license its compiler to the application developers i.e. the Oracles and SAPs of this world. That's where the money is, and you can bet that's what intel is doing.
It'll never be the success that intel and HP envision for it and here's why. First, it's too hot and too expensive. Secondly is doesn't have any applications. I don't mean Gnome and KDE, I mean the sort of applications that big corporations run. Thirdly it isn't backwards-compatible with any existing architectures. You can't just take your binaries over and run them, at least not at full speed. Applications will need to be ported and retested. This is not insignificant in time, effort and cost. Fourthly, most people who want 64-bit in the corporate world already have it in the form of SPARC, Power, PA RISC and Alpha. Why should they change to an unproven, immature "jam tomorrow" architecture given their working and reliable systems already in use? I'm afraid intel missed the boat by about 10 years. If they'd brought out a 64-bit RISC at the same time as SPARC, MIPS, Alpha and Power they might have stood a chance. It's a turkey, and apart from a few niches (e.g. number-crunching super computers) it's doomed to failure. I don't even need to mention how Athlon 64/Opteron will eat its lunch in the commodity sector of the market.
So, PCs have 3DNow!, SSE and SSE2 depending on what processor you have. I have observed factor-of-ten speed-ups of certain code using hand-crafted 3DNow! vs. GCC floating-point. I wonder how fast his algorithm would be if implemented in 3Dnow! or SSE? I bet my rusty old K6-2/500 could put in a reasonable showing at his benchmark.
Well, Linux may not be "lock-upable" but many of the libraries which run on top of it which are used in applications are not completely "Free Software" and by that I mean not GPL or LGPL. Some have BSD/X11-type licenses. One could envision a scenario in which Microsoft could embrace one of these libraries (e.g. from mono) and add their own code. They would be perfectly within their rights to do so. Applications could be written to use that code. Microsoft could then stop distributing the source, supplying binary-only libraries. Then they could break the API and add their own undocumented stuff. Or even worse, they could claim that some of it was patented and start demanding that everyone who uses it pay them a license fee. At this point they have rendered a whole load of other people's software useless, just as they have done before to Windows developers, effectively wiping out the competition.
He asks if anyone knows of any other languages that compile down to Java byte code (i.e. run on the JVM). Here's a list of 160 implementations of various languages for starters. Google rules.
Yes, and before then things will really hot up. India is also planning on sending humans into space. The next race for the moon could be between India and China. Will the US go one better? Or will they continue to rest on their laurels? China wants to send a man into space in October this year
FORTH was and still is such a combination. It's compact, easy to implement and incredibly fast and felxible.
"x86 compatibility is worthless ina high end 64-bit machine" - Indeed. That's why AMD is targetting the low end. The medium to high end has already been 64-bit for the best part of a decade with well established, mature technologies, especially UltraSPARC. itanium just can not hope to take any significant portion of this market simply because it is 10 years too late. No one wants to throw out their entire working, reliable infrastructure for something new and unproven. They'd have to change OS, port all their software, re-test everything, retrain their technical staff, buy new middle ware and applications, upgrade their air-conditioning etc. etc. and all for what? Intel's ambitions? Where is the benfit to the customer?
I can't wait for the first Scotsperson to reach earth orbit, amid a fanfare of bagpipes, accordions and fiddles. The tartan of the Black Watch will be worn with pride by said astronaut as she circles the earth at 250 miles up, bag of liquidised haggis in hand being consumed through a straw. As the shuttle passes Edinburgh the Capitol's inhabitants and Scots from all over the world will raise their glasses of Glenmorangie in a toast to celebrate this gloroius achievement. A gigantic Saltire will be floated on Loch Lomond, visible from space. Burns will be quoted. Jimmy Shand will be exhumed and Ricky Fulton will provide the comedy. Oh wait, that's Hogmanay...
They should do all their praying _before_ launch, after all, that's the most dangerous part. In fact, why doesn't NASA employ a load of people representing all of the major world religions to pray for the shuttle's safety before each lift-off?
They should be put in the stocks for a day or too, so that all the village peasants can throw eggs and rotten tomatoes at them. That'll learn 'em.
How could anyone forget him. And what about Chuck Moore, inventor of FORTH? :-)
Maybe shareware on Linux would be a way to distribute relatively inexpensive niche programs that use patented algorithms, such as wavelet encoding for image compression. Because money is involved, the patent could be used and paid for legitimately. Obviously this doesn't help the fight against software patents...
Well, I thought he was bad in Waynes World, but in Austin Powers, he's really funny.
Christian Science. That has always struck me as being a bit of an oxymoron.
Sell the widows and go into South American zinc.
Has someone decided to go mad since it's Friday 13th? Is this some kind of automated troll "spamming?"
In soviet Russia, Soviet Russia jokes are tired of YOU!
But you speak Portugese and have the unsafest roads in Europe.
That's nonsense. They all ran at the same speed, 3.54MHz. The ZX81, the Spectrum's predecessor was slower, at 3.25MHz. I had (still have) a ZX81 and a Spectrum 128.
No, they had a Z80A running at 3.54MHz.