Look at how it works in the "Poulson" core. The need for high compiler optimization is gone and it works in a similar manner to any other in-order processor.
You didn't real the comment. Especially the part where he said "Also it isn't as though MS is dropping Itanium on the floor. They are just not going to support it in their next server OS. You can get Server 2008R2 for Itanium and that'll be supported until 7/10/2018 minimum (they've extended support dates before, but never shortened them). So they are keeping support on going, just announcing it is coming to an end so you've time to plan."
First off, the notion that Itanium is "dying" is ridiculous - or at least just as ridiculous as the idea that SPARC is dying. Power is the only high-end RISC processor that's really thriving. Both IA64 and SPARC bring in hundreds of millions of dollars per quarter, although their revnue is slowly dwindling:
Second, comparing Oracle suddenly killing support to Microsoft and Red Hat killing support is ridiculous. Red Hat is continuing to develop the 5.x tree for IA64, despite the fact that maybe 5% of Itanium customers ran RHEL. Oracle, on the other hand, is just suddenly saying "No more. Nada." despite the fact that they build key apps for all three HP Itanium operating systems (Rdb for VMS, Oracle for HP-UX, Tuxedo for Nonstop.) There's also the fact that Oracle has its own competing UNIX OS and processor, one that hasn't performed particularly well in comparison to Power or Itanium for several years. The whole thing just looks like Oracle is being a bully.
The thing is, other vendors are using it. Huawei and Inspur announced they're developing new Itanium machines earlier this year; Hitachi and Mitsubishi resell HP's machines. NEC and Bull also use Itanium to run their proprietary ACOS and GCOS mainframe operating systems. I think these vendors would probably get pissy if HP got exclusive control of the architecture.
Microsoft and Red Hat had negligible market share on Itanium. Most VMS and HP-UX customers run Oracle products, and Oracle is a direct competitor for servers and operating systems with HP. The whole thing looks wildly anticompetitive.
I'm in favor of punishing him because he has no honor. He was pissed off at the Army so he tried to shaft the government in any way possible, all while hiding behind what he thought was a convenient shield of anonymity. Now he has to pay the piper, and everyone's praising what a hero he is for dumping large numbers of sensitive documents. Almost none of them contained actual evidence of crimes. Most of them were, essentially, gossip, and there's no way in hell he actually read all of them before leaking them. He knew the consequences for what he was doing, knew the criminal penalties involved, and he should (and will) face a court-martial for it.
It's expected that if a soldier determines that his orders are illegal, he'll face a court-martial to prove it. A junior enlisted man can't just say "No, Captain, I don't want to do that," and expect "Hmm. Okay. I'll ask someone else." as a response.
I seriously doubt that, since it has a locked-down app store and sandboxed applications (fairly restricted API's, no native code.) It should be at least as secure as iPhone OS, if not more so.
PowerPC is oriented towards embedded (PPC 4xx) and manycore supercomputing (BlueGene) workloads, and isn't really ideal for desktops at this time. POWER7 is IBM's flagship server processor, and outperforms anything x86 by quite a considerable margin (admittedly while drawing 200W.)
Ha, one of those people that thinks there's a clean, perfect RISC architecture inside Intel CPU cores.
First off, everything is microcoded. Power is microcoded. SPARC64 is microcoded. Itanium isn't, but it's an oddball in that regard. Microcode just lets you hide implementation details and potentially simplify internal design.
Internal microcode isn't necessarily fun to play with. Look up the articles on RealWorldTech on the guts of Transmeta's CPU's if you're interested, and that used a significantly higher-level microcode than Intel does, from what I understand. Most of x86's microcode is related to stuff like turning direct memory references in instructions into load/store instructions. Taking that away does not make the chip magically faster. If you took away the decoder, and just ran on the metal, you'd probably encounter something that wasn't really any faster and felt vaguely like an x86-like microcontroller or a DSP from a programming perspective.
No, what made apps "just work" was the fact that Apple licensed a fast PowerPC emulator (QuickTransit.) Objective-C compiles to fast native code, not a bytecode environment.
Shocking though it may appear, there are actually some distinctions between Social Security (a program to create, essentially, a nationalized retirement system that has long since fallen into the realm of poor planning) and Wall Street (a road; also, a vaguely defined bogeyman for the far left who need something to rage against.)
Oh, and GP never mentioned Wall Street. Fucking leftist nutjobs.
ROK, Japan, the ROC when we're playing nice. Thailand. Australia. Arguably Poland and some of the other Eastern European states.
Israel is the most civilized place in the Middle East, but they would shaft the United States without a second thought if they determined it was beneficial.
Slashdot Mirror
Tremulous has an in-depth class/upgrade/building system if I recall. It's actually pretty unique, but not really my cup of tea.
There is roughly zero overlap between what makes a good HPC processor and what makes a good datacenter processor.
Hint: AVX throughput matters almost none when running an SQL server, but looks very good on Linpack.
Look at how it works in the "Poulson" core. The need for high compiler optimization is gone and it works in a similar manner to any other in-order processor.
You didn't real the comment. Especially the part where he said "Also it isn't as though MS is dropping Itanium on the floor. They are just not going to support it in their next server OS. You can get Server 2008R2 for Itanium and that'll be supported until 7/10/2018 minimum (they've extended support dates before, but never shortened them). So they are keeping support on going, just announcing it is coming to an end so you've time to plan."
First off, the notion that Itanium is "dying" is ridiculous - or at least just as ridiculous as the idea that SPARC is dying. Power is the only high-end RISC processor that's really thriving. Both IA64 and SPARC bring in hundreds of millions of dollars per quarter, although their revnue is slowly dwindling:
http://smarterquestions.org/wp-content/uploads/2011/10/UNIX_Revenue_08-2011.png
Second, comparing Oracle suddenly killing support to Microsoft and Red Hat killing support is ridiculous. Red Hat is continuing to develop the 5.x tree for IA64, despite the fact that maybe 5% of Itanium customers ran RHEL. Oracle, on the other hand, is just suddenly saying "No more. Nada." despite the fact that they build key apps for all three HP Itanium operating systems (Rdb for VMS, Oracle for HP-UX, Tuxedo for Nonstop.) There's also the fact that Oracle has its own competing UNIX OS and processor, one that hasn't performed particularly well in comparison to Power or Itanium for several years. The whole thing just looks like Oracle is being a bully.
The thing is, other vendors are using it. Huawei and Inspur announced they're developing new Itanium machines earlier this year; Hitachi and Mitsubishi resell HP's machines. NEC and Bull also use Itanium to run their proprietary ACOS and GCOS mainframe operating systems. I think these vendors would probably get pissy if HP got exclusive control of the architecture.
Microsoft and Red Hat had negligible market share on Itanium. Most VMS and HP-UX customers run Oracle products, and Oracle is a direct competitor for servers and operating systems with HP. The whole thing looks wildly anticompetitive.
I'm in favor of punishing him because he has no honor. He was pissed off at the Army so he tried to shaft the government in any way possible, all while hiding behind what he thought was a convenient shield of anonymity. Now he has to pay the piper, and everyone's praising what a hero he is for dumping large numbers of sensitive documents. Almost none of them contained actual evidence of crimes. Most of them were, essentially, gossip, and there's no way in hell he actually read all of them before leaking them. He knew the consequences for what he was doing, knew the criminal penalties involved, and he should (and will) face a court-martial for it.
The Arab Spring revolutions are over. Does Libya or Egypt now have a free and democratic government?
It's expected that if a soldier determines that his orders are illegal, he'll face a court-martial to prove it. A junior enlisted man can't just say "No, Captain, I don't want to do that," and expect "Hmm. Okay. I'll ask someone else." as a response.
So now something done by a random Texan company falls into the category of "crimes of the US government?"
I seriously doubt that, since it has a locked-down app store and sandboxed applications (fairly restricted API's, no native code.) It should be at least as secure as iPhone OS, if not more so.
It seems unlikely that there wouldn't be NDA's involved with this program, despite your apparent inclination toward paranoia.
Calling Mac OS X "BSD based" is a stretch.
PowerPC is oriented towards embedded (PPC 4xx) and manycore supercomputing (BlueGene) workloads, and isn't really ideal for desktops at this time. POWER7 is IBM's flagship server processor, and outperforms anything x86 by quite a considerable margin (admittedly while drawing 200W.)
What alternative is there?
Wind and solar aren't economical, and solar has toxic waste problems of its own.
Coal is dirty, contributes to global warming, and is a dependence on fossil fuels.
Natural gas is better, but has a lot of the same issues.
Geothermal is hypothetically pretty cool, but it's expensive and there's that annoying problem with earthquakes.
Hydro probably will never be able to cover close to all of the world's energy needs, and floods places.
Fusion is thirty years away, and has been since at least 1960.
What am I missing?
So it's fine to piss away $443mn because it's less than $70bn?
Ha, one of those people that thinks there's a clean, perfect RISC architecture inside Intel CPU cores.
First off, everything is microcoded. Power is microcoded. SPARC64 is microcoded. Itanium isn't, but it's an oddball in that regard. Microcode just lets you hide implementation details and potentially simplify internal design.
Internal microcode isn't necessarily fun to play with. Look up the articles on RealWorldTech on the guts of Transmeta's CPU's if you're interested, and that used a significantly higher-level microcode than Intel does, from what I understand. Most of x86's microcode is related to stuff like turning direct memory references in instructions into load/store instructions. Taking that away does not make the chip magically faster. If you took away the decoder, and just ran on the metal, you'd probably encounter something that wasn't really any faster and felt vaguely like an x86-like microcontroller or a DSP from a programming perspective.
No, what made apps "just work" was the fact that Apple licensed a fast PowerPC emulator (QuickTransit.) Objective-C compiles to fast native code, not a bytecode environment.
The US is the largest manufacturer in the world (including a large arms industry!) and is barely inflationary at the moment.
Shocking though it may appear, there are actually some distinctions between Social Security (a program to create, essentially, a nationalized retirement system that has long since fallen into the realm of poor planning) and Wall Street (a road; also, a vaguely defined bogeyman for the far left who need something to rage against.)
Oh, and GP never mentioned Wall Street. Fucking leftist nutjobs.
ROK, Japan, the ROC when we're playing nice. Thailand. Australia. Arguably Poland and some of the other Eastern European states.
Israel is the most civilized place in the Middle East, but they would shaft the United States without a second thought if they determined it was beneficial.
The first paragraph of that article was one of the stupidest things I've ever read.
It originally was a video card (Larrabee project), but things didn't look good for consumer performance and they repositioned it.
Intel claims it will be released as a commercial product in the near future.