Will developers of free software be able to afford a Pathscale license? Or has GCC x86-64 code generation progressed to where that's not strictly necessary? gcc on x86-64 is "good enough" for most people, and can produce significant performance gains on Opteron over 32-bit compiled code.
In comparison, the last I heard about gcc on itanium was that it could only get about 20% of the performance of intel's proprietary compiler, and gcc still can't vectorise. Things may have changed, and the gcc-compiled itanic object code may be better now, but who knows.
The thing is, there is a highly-optimising compiler for Opteron (from Pathscale) if you want to pay for it and a "good enough" one (gcc) if you don't. If you're doing serious numbercrunching on any architecture, you must consider the best compilers, otherwise all that (expensive) hardware is wasted. Most times, that means paying for a specialist compiler.
I recommend you look at the gcc website for more information about platform-specifics.
t will take much longer than next year (if ever) when Opteron starts to become "mainstream". Opteron is already mainstream.
application availablility along with essentially non-existant tier-1 vendor support.
Let me put it this way: you are living under a rock.
what's the point of upgrading your hardware if you only want to run 32bit x86 binaries? The performance on legacy code alone is enough to warrant the upgrade.
Show me such scalability with an opteron! Cray will be doing so soon, I believe with 10,000+ Opterons in one machine, and I'm sure you'll be pleasantly surprised.
no *good* AMD64 compilers Ah, so you haven't heard of Pathscale then?
It took 11 years for 32bit operating systems to finally displace 16bit operating systems. Your prediction of 32bit PCs being laughed at by December 2004 is probably a little too radical.
Sun and DEC brought out 64-bit hardware and Operating Systems around 1994-1995. This is now 2004. I don't think my prediction is radical at all.
The big problem with the Pentium isn't that it's only 32 bits wide, it's that the instruction set is so poorly designed. It's complex and hard to execute quickly, doesn't have enough registers, REALLY doesn't have enough floating point registers, has too high a cost to transferring data between the CPU and FPU, and huge chunks of silicon have to be used to cover for these faults.
AMD has solved these problems in the Opteron (and Athlon 64) by doubling the number of integer registers (to 16) and making them all general-purpose, and by implementing SSE2 (again, with 16 registers compared to the Pentium 4's 8) for fast floating point. SSE2 allows you to explicitly code two double-precision floating point adds or multiplies to execute in a single clock cycle (or 4 single precision ones). It also does scalar FP, and does it much faster than the legacy x87 floating point hardware (which is still there for backwards compatability).
Intel will very likely release a 64-bit x86 processor, or kludge unit for Pentium V, this year (just like the math coprocessor was prior to the 486).
However consider this:
AMD has been shipping Opteron for nearly a year already, and ports of the main OSs (including Windows and Linux) have been done, with others already working in the lab. It also runs old 32-bit OSs with no change. It will run legacy x86 code at full speed along side new 64-bit code. It is more efficient in terms of useful work done per clock cycle compared to Pentium 4 and Xeon. It scales better in multi-way systems (very important in workstations and serves) : the logic is built in. Xeon does not have this (and plain P4 is limited to single-way). It has a built in memory controller. It has twice as many registers. It's very inexpensive. Go and look up your favourite component retailer right now and compare an Opteron to a Xeon (and even the "high-end" Pentium 4).
The only place AMD may have trouble selling is to the ignorant masses who buy on MHz (or GHz) from highstreet stores, and pay too much.
The corporate world is more clued-up, and so are the enthusiasts and power-users.
Even if AMD does not knock intel off of it's perch, there is a huge potential market for Opteron. Several major corporations are behind Opteron. They've committed to it. It's going to be big business. 2004 will see a radical change in the hardware business. I predict that in the second half of this year, people will laugh a 32-bit PeeCees. They will be obsolete and bargain-basement by then.
I'm sure you could get away with a pimp-slap followed by "clearly I'm happy to drop you if you keep this up"
I don't know. In this case she would probably have gone around telling all her friends that I'd beat her or something, and there was this man we knew who was prone to random acts of vigilante violence when he decided someone deserved it. Anyway, it was all a long time ago, and I'm sure she's had her fair share of intolerant men by now.
"Sure I hit my wife - when she deserves it!" (this is probably less of a taboo than it should be)
I hate to rain on your parade but... Violence should be unacceptable except in self-defense.
In this country (UK) for about 10 years now there has been a campaign to stop "violence against women" (by men) i.e. Domestic Violence. There has been no such campaign agains violence against men by women. In the media, women are portrayed as heroes when they thump men.
Look, I'll cut out the crap and come straight to the point.
One night my then girlfriend went ape at me in a fit of madness, ripped out half my hair and beat me black and blue.
I didn't hit back and I'm glad I didn't,
In that case I deserved a go at self-defense, not revenge, but I prefered to curl up into a ball and take the beatings.
So you see, the guy was right on the money in his article. It's what people will let you say, what they will let you believe and what they allow us to believe is important.
Ho hum. Unfortunately, we live in a culture where being able to argue the toss about various passages from Shakspeare is held in higher esteem than being able to apply Newton's Laws to real situations, or where arguing the toss about various "issues" is more highly regarded than making rational, impartial, scientific observations and assessments. Nyeah.
Back in the day, we were forced to use sneaker.net (TM). It worked quite well, even on MS-DOS workstations with 512k RAM, and the 80286 processor and still works to this day. Reliability is so-so, and speed can be poor, but nowadays with technological progress transfer rates can be the orders of gigabytes per second, but latencies are large (tens of seconds upwards to several days). One downside was the propagation of viruses, but distribution of code across platforms by source and proper protected mode operating systems with selectable user privileges make viruses less dangerous.
If you would just put up and shut up like the fat cats want you to, you'd buy a new copy of the movie each time you moved zone. That's what it's about, plain and simple. I know some people who will do this sort of thing and accept it. Some people really are like that. People are not rational. People will allow their convenieces and even their rights to be eroded for a quiet life in the short term.
What a great deed and astounding achievement. My hat is off to this man. I hope he has success with his next project, gettiong Open Source software into the hands of every school child in Scotland. He has an uphill battle (they are so conservative about these things it's unbelievable) but I think he has what it takes to achieve his goal.
This man may just have radically altered the course of Scottish society. He is bringing enlightenment to thousands. This could be the best thing to happen to Scotland this century.
Save your money. You'd be better off with an Opteron. Even a 1.4GHz should keep up with that P4 2.8GHz. Buy a dual board and pop in an extra CPU later when the prices drop.
Solaris Free and Open Source Software
on
Solaris 9 x86 Review
·
· Score: 2, Informative
You can get a binary version (for x86) of GCC from any of the sunfreeware.com mirrors.
The Solaris Companion Software CD comes in the Solaris Media Kit and contains many Free and Open Source packages compiled and packaged for Solaris. They install under/opt/sfw and include gcc-2.95.3 and gcc-3.3.x, gdb, ddd, KDE, GIMP, Emacs, vim, Python, MySQL, you name it, it's probably there. You can download the ISO images for Solaris 8 and 9. Note that it is updated to correspond with the current Update release of Solaris.
GNOME has been shipped with Solaris for quite some time now. Another thing that people don't realise is that some Open Source software is installed under/usr/sfw by default as well.
I hear that athlon, and even dual-athlon, systems work with it. but it depends on your motherboard, etc, etc.
Indeed. I have a dual P4 Xeon running Solaris, an Athlon XP2400+, some SPARC boxes and some 4-way Dell PIII and P4 Xeons all running Solaris. Most of them have pretty standard chipsets and peripherals, nothing terribly fancy or expensive. I even have one with accelerated NVidia graphics. We also have some multi-way Opterons running Solaris.
There's an awful lot of anti-Solaris FUD about.
The cool things about the Opterons is that the intel compatibility is so good, existing 32-bit OSes run totally unmodified and withoud problems on them.
I'm pretty sure the new SGI runs without futzing, not sure if it's a 106 cpu box
Wrong. To get SGI's "scalability" in Linux you need SGI's closed source Linux kernel extensions. That's right, they're not GPL or even Open Source. You have to complie in their own special "shim" layer into the kernel.
As for the user space, I'd imagine they have many of their own non-standard utilities to drive it all.
OpenBIOS is another interesting free BIOS project for PeeCees which aims to implement the Open Firmware (IEEE 1275-1994) standard, as used on Sun SPARC machines and Apple Power PC machines.
It's interesting because Open Firmware is based around a FORTH interpreter, using which high-level BIOS code is implemented. This code is portable across different binary architectures. This has interesting implications for the initialisation of peripherals. It also means you can program your own BIOS at a command line at system start up.
There are loads of other uses, and it's already an establishged Open standard, and has been in use for well over a decade.
Linux Throws On a Spaghetti-Strap Dress and Comments Mockingly on Stodgy Brits
I'd rather be slightly plump, well-fed and contented than scrawny, edgy and looking like an emaciated anorexic skeleton to conform to your Hollywood stereotype thank you very much. Those "tart's breakfasts" will catch up with you one day...
Slashdot Mirror
gcc on x86-64 is "good enough" for most people, and can produce significant performance gains on Opteron over 32-bit compiled code.
In comparison, the last I heard about gcc on itanium was that it could only get about 20% of the performance of intel's proprietary compiler, and gcc still can't vectorise. Things may have changed, and the gcc-compiled itanic object code may be better now, but who knows.
The thing is, there is a highly-optimising compiler for Opteron (from Pathscale) if you want to pay for it and a "good enough" one (gcc) if you don't. If you're doing serious numbercrunching on any architecture, you must consider the best compilers, otherwise all that (expensive) hardware is wasted. Most times, that means paying for a specialist compiler.
I recommend you look at the gcc website for more information about platform-specifics.
Opteron is already mainstream.
application availablility along with essentially non-existant tier-1 vendor support.
Let me put it this way: you are living under a rock.
what's the point of upgrading your hardware if you only want to run 32bit x86 binaries?
The performance on legacy code alone is enough to warrant the upgrade.
Show me such scalability with an opteron!
Cray will be doing so soon, I believe with 10,000+ Opterons in one machine, and I'm sure you'll be pleasantly surprised.
no *good* AMD64 compilers
Ah, so you haven't heard of Pathscale then?
This year promises to be very interesting indeed.
Sun and DEC brought out 64-bit hardware and Operating Systems around 1994-1995. This is now 2004. I don't think my prediction is radical at all.
AMD has solved these problems in the Opteron (and Athlon 64) by doubling the number of integer registers (to 16) and making them all general-purpose, and by implementing SSE2 (again, with 16 registers compared to the Pentium 4's 8) for fast floating point. SSE2 allows you to explicitly code two double-precision floating point adds or multiplies to execute in a single clock cycle (or 4 single precision ones). It also does scalar FP, and does it much faster than the legacy x87 floating point hardware (which is still there for backwards compatability).
However consider this:
AMD has been shipping Opteron for nearly a year already, and ports of the main OSs (including Windows and Linux) have been done, with others already working in the lab. It also runs old 32-bit OSs with no change. It will run legacy x86 code at full speed along side new 64-bit code. It is more efficient in terms of useful work done per clock cycle compared to Pentium 4 and Xeon. It scales better in multi-way systems (very important in workstations and serves) : the logic is built in. Xeon does not have this (and plain P4 is limited to single-way). It has a built in memory controller. It has twice as many registers. It's very inexpensive. Go and look up your favourite component retailer right now and compare an Opteron to a Xeon (and even the "high-end" Pentium 4).
The only place AMD may have trouble selling is to the ignorant masses who buy on MHz (or GHz) from highstreet stores, and pay too much.
The corporate world is more clued-up, and so are the enthusiasts and power-users.
Even if AMD does not knock intel off of it's perch, there is a huge potential market for Opteron. Several major corporations are behind Opteron. They've committed to it. It's going to be big business. 2004 will see a radical change in the hardware business. I predict that in the second half of this year, people will laugh a 32-bit PeeCees. They will be obsolete and bargain-basement by then.
I don't know. In this case she would probably have gone around telling all her friends that I'd beat her or something, and there was this man we knew who was prone to random acts of vigilante violence when he decided someone deserved it. Anyway, it was all a long time ago, and I'm sure she's had her fair share of intolerant men by now.
Thank you. Now will you kindly go and tell the British mass media! :-)
And remove the prostate? :-)
I hate to rain on your parade but... Violence should be unacceptable except in self-defense.
In this country (UK) for about 10 years now there has been a campaign to stop "violence against women" (by men) i.e. Domestic Violence. There has been no such campaign agains violence against men by women. In the media, women are portrayed as heroes when they thump men.
Look, I'll cut out the crap and come straight to the point.
One night my then girlfriend went ape at me in a fit of madness, ripped out half my hair and beat me black and blue.
I didn't hit back and I'm glad I didn't,
In that case I deserved a go at self-defense, not revenge, but I prefered to curl up into a ball and take the beatings.
So you see, the guy was right on the money in his article. It's what people will let you say, what they will let you believe and what they allow us to believe is important.
Ho hum. Unfortunately, we live in a culture where being able to argue the toss about various passages from Shakspeare is held in higher esteem than being able to apply Newton's Laws to real situations, or where arguing the toss about various "issues" is more highly regarded than making rational, impartial, scientific observations and assessments. Nyeah.
...and he's king of Newcastle, and hence the world.
No, I think it's "T" as in James T. Kirk.
Crocodile shooooes!
Back in the day, we were forced to use sneaker.net (TM). It worked quite well, even on MS-DOS workstations with 512k RAM, and the 80286 processor and still works to this day. Reliability is so-so, and speed can be poor, but nowadays with technological progress transfer rates can be the orders of gigabytes per second, but latencies are large (tens of seconds upwards to several days). One downside was the propagation of viruses, but distribution of code across platforms by source and proper protected mode operating systems with selectable user privileges make viruses less dangerous.
There ends my rant for the day.
This man may just have radically altered the course of Scottish society. He is bringing enlightenment to thousands. This could be the best thing to happen to Scotland this century.
Well done!
Save your money. You'd be better off with an Opteron. Even a 1.4GHz should keep up with that P4 2.8GHz. Buy a dual board and pop in an extra CPU later when the prices drop.
Isn't it more like 40?
Nyeah, here, of course!
The Solaris Companion Software CD comes in the Solaris Media Kit and contains many Free and Open Source packages compiled and packaged for Solaris. They install under /opt/sfw and include gcc-2.95.3 and gcc-3.3.x, gdb, ddd, KDE, GIMP, Emacs, vim, Python, MySQL, you name it, it's probably there. You can download the ISO images for Solaris 8 and 9. Note that it is updated to correspond with the current Update release of Solaris.
GNOME has been shipped with Solaris for quite some time now. Another thing that people don't realise is that some Open Source software is installed under /usr/sfw by default as well.
Indeed. I have a dual P4 Xeon running Solaris, an Athlon XP2400+, some SPARC boxes and some 4-way Dell PIII and P4 Xeons all running Solaris. Most of them have pretty standard chipsets and peripherals, nothing terribly fancy or expensive. I even have one with accelerated NVidia graphics. We also have some multi-way Opterons running Solaris.
There's an awful lot of anti-Solaris FUD about.
The cool things about the Opterons is that the intel compatibility is so good, existing 32-bit OSes run totally unmodified and withoud problems on them.
Wrong. To get SGI's "scalability" in Linux you need SGI's closed source Linux kernel extensions. That's right, they're not GPL or even Open Source. You have to complie in their own special "shim" layer into the kernel.
As for the user space, I'd imagine they have many of their own non-standard utilities to drive it all.
It's interesting because Open Firmware is based around a FORTH interpreter, using which high-level BIOS code is implemented. This code is portable across different binary architectures. This has interesting implications for the initialisation of peripherals. It also means you can program your own BIOS at a command line at system start up.
There are loads of other uses, and it's already an establishged Open standard, and has been in use for well over a decade.
Not quite. It's more like : "Why pay for rubbish when the good stuff is available legally for free."
Some of us realised this back in about 1994 and have never regretted having this insight.
I'd rather be slightly plump, well-fed and contented than scrawny, edgy and looking like an emaciated anorexic skeleton to conform to your Hollywood stereotype thank you very much. Those "tart's breakfasts" will catch up with you one day...