IA64 machines ahave already hit the market (last year) and been withdrawn. You could buy a single processor machine from Dell for about $25k. They stopped selling them recently.
Sun threads are what POSIX threads are based on. It's funny how everyone is looking forward to this M:N threading model, when it has been shown in the field to be a hinderance rather than a help. 1:1 threading is simpler, more efficient and provides much better over-all performance.
Interesting. At my old nuclear powerstation (which just closed at the end of March) out primary reactor temperature monitoring system was a 16k word (16-bit) Honeywell 316 (it did both reactors). Two years ago it was replaced by a pair of PDP 11's:-) a month or two before the Honeywell died:-) Reactor protection was by analogue circuits. There were 3 main guardlines comprising relays and amplifiers, and a double 2-out-of-3 diverse guarline which was all solid state. Lightning and arc welding were notorious for tripping the reactors. They were magnox (C02 cooled, natural uranium) and it woul be at least a day before they'd be back up to anything like full power.
It's funny. I've heard screaming hedgemonkies, I mean "environmentalists", on the news talking about "polluting outer space with our radiation" (seriously), and that launching radioactive fuel is dangerous because it could fall back to earth and pollute us all. What people don't realise is that nuclear reactor fuel is relatively benign before it has been exposed to a neutron flux i.e. before the reactor has been started for the first time. The logical thing to do would be to transport into space some fuel pellets or rods to be put into a reactor in orbit (perhaps even assembled there), give it a kick with a chemical engine to get it into a really high orbit and then _start the reactor_ to give nuclear propulsion. I'd bet you any amount of money that the gamma and neutron flux from the engine would be negligible as measured on earth compared to what we get from the sun and space naturally i.e. you'd be hard pushed to detect it. But the Greenpeace and Friends of the Earth imbiciles will still get angry about it...
>This means the AMD cpus run IA32 code much quicker, but the intel 64 bit cpus are quicker when running native code. At least thats how I understand it, in laymans terms.
Wrong! The AMD Hammer runs 64-bit and 32-bit code at the same (full) speed since the 64-bot stuff is a logical and transparent extension of the 32-bit stuff. The 64-bit instruction set is RISCy in that to keep it simple and keep the speed up, they only implemented the simpler and more useful instructions in 64-bits. Go to www.x86-64.org and look at what they did to the register set (twice as wide, twice the number of registers, i.e. the 386 registers are the upper-right quarter of the entire register set) etc. Oh, and doesn't it have 11 execution units and run at 2GHz?
Reactor SCRAM? Nuclear subs have "battle shorts" which short-circuit the reactor protection so that, when in a tricky situation, the reator will not trip or "SCRAM". Civillian nuclear reactors do not have this facility, since they do not get involved in underwater confrontations with the enemy. Although, the crazy Russian RBMK (and I dare say their VVERs) have reactor protection which _can_ be disabled. This was one of the causes of the Chernobyl accident.
...because 90% of the world thinks that M$ (which has only recently become 32-bit clean) is the be-all and end-all od computing. Yes it's a rant, but there's probably a grain if truth in there. The 386 came out in '85. It took M$ until about 1993(?) until NT3.51 Most people were only ever interested in an OS if they could run their DOS 2.0 apps unmodified, specifically Lotus-1-2-3, Norton Commander, Borland Sidekick etc. Technological superiority was never an issue. Hence, the humble Xenix was passed by, as was OS/2, etc..
The rest, as they say, is history (my dad wouldn't let me have a 68k-based (internally 32-bit) Amiga or ST because "it doesn't run DOS and doesn't run Lotus and has Mickey-Mouse grpahics which are only for playing games which is what hooligans and junkies do)
Kid, when you become parents, please take heed, and don't be a dick as described above.
Sad isn't it? What he doesn't mention, is that most Linux people have gcc, and last time I looked, the object code produced by gcc on IA64 was +20% of the speed of the intel compiler. This isn't a criticism of gcc, it's just that the IA64 arch. is so different that you absolutely _must_ have the intel compiler to get any performance out of it.
In the article he mentions that itanic can execute IA32 code _and_ PA-RISC code natively, as well as its own, but these features will be taken away sometime in the future. Does anyone remember the leaked benchmarks that showed the itanic executing IA32 code at roughly 10% of the speed of an equivalently-clocked PIII? I wonder how it shapes up on PA-RISC performance? It has to offer some sort of advantage over existing chips, or no one will buy it. On the other hand, maybe its tremendous heat dissipation will reduce drastically when they remove all that circuitry for running IA32 and PA-RISC code. Which leads me to think, why didn't they invest the time and money in software technology like dynamic recompilation, which Apple did very successfully when they made the transition from 69k to PPC?
Did you try another tarball? Did it work? My E450 Solaris 8 build and 4-way PIII Xeon Solaris 8 build both worked fine with the same./configure options and gmake -j 5 bootstrap.
Are you running on x86 or SPARC? I've built various versions of gcc-3.0.x and 3.1-xxxxx on SPARC and intel multiprocessor boxen on Solaris 8 and 9. What C compiler are you using to start the bootstrap build?
I have the Freeware Companion CD installed, so am using gcc-2.95.3.
I'm building it make -j 3 bootstrap on my Ultra 60 using Solaris 9. It's been going for ages and hasn't broken yet. I've currently got gcc-2.95.3 installed. I also set it building on my humble K6-2 500 under Linux at home this morning.
Re:GCC (any version) in a nutshell
on
GCC 3.1 Released
·
· Score: 1, Interesting
How did you get to be +1 Informative with a troll of this kind? GCC's code generator has improved in leaps and bounds recently. It may not generate the fastest code on all platforms, but it generates reliable code that is good enough. If you really are interested in performance, a little bit of profiling and clever coding goes a long way.
I suspect the poster was poking fun at the sort of logic employed my Microsoft in its public statements and of many government officials who appear to be on Microsoft's side.
Slashdot Mirror
IA64 machines ahave already hit the market (last year) and been withdrawn. You could buy a single processor machine from Dell for about $25k. They stopped selling them recently.
Sun threads are what POSIX threads are based on.
It's funny how everyone is looking forward to this M:N threading model, when it has been shown in the field to be a hinderance rather than a help. 1:1 threading is simpler, more efficient and provides much better over-all performance.
Interesting. At my old nuclear powerstation (which just closed at the end of March) out primary reactor temperature monitoring system was a 16k word (16-bit) Honeywell 316 (it did both reactors). Two years ago it was replaced by a pair of PDP 11's :-) a month or two before the Honeywell died :-)
Reactor protection was by analogue circuits. There were 3 main guardlines comprising relays and amplifiers, and a double 2-out-of-3 diverse guarline which was all solid state.
Lightning and arc welding were notorious for tripping the reactors. They were magnox (C02 cooled, natural uranium) and it woul be at least a day before they'd be back up to anything like full power.
3. I have served on a submarine and know that of which I speak.
Cool!
...and trying out 1000+ themes, skins, alpha-transparent widget sets and embedded python scripts...
It's funny. I've heard screaming hedgemonkies, I mean "environmentalists", on the news talking about "polluting outer space with our radiation" (seriously), and that launching radioactive fuel is dangerous because it could fall back to earth and pollute us all. What people don't realise is that nuclear reactor fuel is relatively benign before it has been exposed to a neutron flux i.e. before the reactor has been started for the first time. The logical thing to do would be to transport into space some fuel pellets or rods to be put into a reactor in orbit (perhaps even assembled there), give it a kick with a chemical engine to get it into a really high orbit and then _start the reactor_ to give nuclear propulsion. I'd bet you any amount of money that the gamma and neutron flux from the engine would be negligible as measured on earth compared to what we get from the sun and space naturally i.e. you'd be hard pushed to detect it.
But the Greenpeace and Friends of the Earth imbiciles will still get angry about it...
>This means the AMD cpus run IA32 code much quicker, but the intel 64 bit cpus are quicker when running native code.
At least thats how I understand it, in laymans terms.
Wrong! The AMD Hammer runs 64-bit and 32-bit code at the same (full) speed since the 64-bot stuff is a logical and transparent extension of the 32-bit stuff. The 64-bit instruction set is RISCy in that to keep it simple and keep the speed up, they only implemented the simpler and more useful instructions in 64-bits.
Go to www.x86-64.org and look at what they did to the register set (twice as wide, twice the number of registers, i.e. the 386 registers are the upper-right quarter of the entire register set) etc.
Oh, and doesn't it have 11 execution units and run at 2GHz?
Reactor SCRAM?
Nuclear subs have "battle shorts" which short-circuit the reactor protection so that, when in a tricky situation, the reator will not trip or "SCRAM". Civillian nuclear reactors do not have this facility, since they do not get involved in underwater confrontations with the enemy. Although, the crazy Russian RBMK (and I dare say their VVERs) have reactor protection which _can_ be disabled. This was one of the causes of the Chernobyl accident.
You been sniffing the fumes the itanic gives off? I mean it runs so hot it must melt something somewhere...
>c There are over 8000 packages for i386 (the most up to date architecture) - ia64 currently has about 7650 or so packages built
:-)
....and 93.2% of those are themeable IRC clients for X with alpha transparency and build-in MP3 streaming.
Sorry, couldn't resist...it's late and I have Kronenbourg 1664
Maybe the editors don't like my opinions, after all I'm skeptical of IBM.... and I use Slackware on my home-made K6-2/500.
...because 90% of the world thinks that M$ (which has only recently become 32-bit clean) is the be-all and end-all od computing.
Yes it's a rant, but there's probably a grain if truth in there.
The 386 came out in '85. It took M$ until about 1993(?) until NT3.51
Most people were only ever interested in an OS if they could run their DOS 2.0 apps unmodified, specifically Lotus-1-2-3, Norton Commander, Borland Sidekick etc.
Technological superiority was never an issue. Hence, the humble Xenix was passed by, as was OS/2, etc..
The rest, as they say, is history (my dad wouldn't let me have a 68k-based (internally 32-bit) Amiga or ST because "it doesn't run DOS and doesn't run Lotus and has Mickey-Mouse grpahics which are only for playing games which is what hooligans and junkies do)
Kid, when you become parents, please take heed, and don't be a dick as described above.
Thanks. That's very interesting. PA-RISC always was a very good design.
Indeed I do :-) :-)
My eyesight and tryp[ing ain;t what they used to nbe
Sad isn't it?
What he doesn't mention, is that most Linux people have gcc, and last time I looked, the object code produced by gcc on IA64 was +20% of the speed of the intel compiler. This isn't a criticism of gcc, it's just that the IA64 arch. is so different that you absolutely _must_ have the intel compiler to get any performance out of it.
In the article he mentions that itanic can execute IA32 code _and_ PA-RISC code natively, as well as its own, but these features will be taken away sometime in the future.
Does anyone remember the leaked benchmarks that showed the itanic executing IA32 code at roughly 10% of the speed of an equivalently-clocked PIII?
I wonder how it shapes up on PA-RISC performance?
It has to offer some sort of advantage over existing chips, or no one will buy it.
On the other hand, maybe its tremendous heat dissipation will reduce drastically when they remove all that circuitry for running IA32 and PA-RISC code.
Which leads me to think, why didn't they invest the time and money in software technology like dynamic recompilation, which Apple did very successfully when they made the transition from 69k to PPC?
Last summer at the London Linux Expo I asked this HP reseller (who had a big itanic display) "what about legacy code?"
He replied,"16-bit code?"
I sighed and moved on...
Poor you. Maybe you can swap it with someone for an electric storage heater or a gas central heating boiler? The latter would be cheaper to run.
Did you try another tarball? Did it work? ./configure options and gmake -j 5 bootstrap.
My E450 Solaris 8 build and 4-way PIII Xeon Solaris 8 build both worked fine with the same
Just kicked off a build on my S8 E450 now...
...I'll let you know how it turns out.
Are you running on x86 or SPARC?
I've built various versions of gcc-3.0.x and 3.1-xxxxx on SPARC and intel multiprocessor boxen on Solaris 8 and 9.
What C compiler are you using to start the bootstrap build?
I have the Freeware Companion CD installed, so am using gcc-2.95.3.
Here is my configure line:
../gcc-3.1/configure --prefix=/area51/trial/install/gcc-3.1 --with-ld=/usr/ccs/bin/ld --with-as=/usr/ccs/bin/as --without-gnu-ld --without-gnu-as --enable-shared
FYI I've just had my make bootstrap complete successfully on my Ultra 60 running Solaris 9, with the companion CD installed using gcc-2.95.3
I'm building it make -j 3 bootstrap on my Ultra 60 using Solaris 9. It's been going for ages and hasn't broken yet. I've currently got gcc-2.95.3 installed. I also set it building on my humble K6-2 500 under Linux at home this morning.
How did you get to be +1 Informative with a troll of this kind? GCC's code generator has improved in leaps and bounds recently. It may not generate the fastest code on all platforms, but it generates reliable code that is good enough. If you really are interested in performance, a little bit of profiling and clever coding goes a long way.
I suspect the poster was poking fun at the sort of logic employed my Microsoft in its public statements and of many government officials who appear to be on Microsoft's side.