If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. |
|
|
Thread Tools | Display Modes |
#1
|
|||
|
|||
New Itanium info
Some new Itanium info is out, http://goo.gl/vBg63 gives details on the eight
core 32nm version of the chip just announced. |
#2
|
|||
|
|||
New Itanium info
On Feb 25, 10:13*am, Bill Davidsen wrote:
Some new Itanium info is out,http://goo.gl/vBg63gives details on the eight core 32nm version of the chip just announced. That is by far the most optimistic take on the new Itanium I've read so far. The article implies that a key obstacle to adoption of the chip is software optimization. That may be a formally correct statement, but the possibility of routinely producing good code for this chip at any plausible cost seems to be a pipe dream, even if Intel could jigger its compiler to produce could spec numbers. If this chip really were socket-compatible with QPI Xeon's, that could be a game-changer, but I haven't seen language that seems sufficiently specific to draw conclusions. Robert. |
#3
|
|||
|
|||
New Itanium info
On Fri, 4 Mar 2011 15:03:54 -0800 (PST), Robert Myers
wrote: [...] If this chip really were socket-compatible with QPI Xeon's, that could be a game-changer, but I haven't seen language that seems sufficiently specific to draw conclusions. I'm curious how you came to your "could be a game-changer" conclusion. By all appearances, it's the architecture and the mountain of baggage it brings that has been pretty soundly rejected by all but a very few. Hardware implementation details seem to be well below the noise floor wrt significance... Cheers |
#4
|
|||
|
|||
New Itanium info
On Mar 4, 6:54*pm, daytripper wrote:
On Fri, 4 Mar 2011 15:03:54 -0800 (PST), Robert Myers wrote: [...] If this chip really were socket-compatible with QPI Xeon's, that could be a game-changer, but I haven't seen language that seems sufficiently specific to draw conclusions. I'm curious how you came to your "could be a game-changer" conclusion. By all appearances, it's the architecture and the mountain of baggage it brings that has been pretty soundly rejected by all but a very few. Hardware implementation details seem to be well below the noise floor wrt significance... People have x86 software and they have enterprise software. If you're IBM, you satisfy that need by selling expensive plugins that allow you to include PC-compatible chips into your, um, mainframe installation. If you're HP, potentially you say, here's you box. Run whatever you like. Robert. |
#5
|
|||
|
|||
New Itanium info
On Fri, 4 Mar 2011 17:56:41 -0800 (PST), Robert Myers
wrote: On Mar 4, 6:54*pm, daytripper wrote: On Fri, 4 Mar 2011 15:03:54 -0800 (PST), Robert Myers wrote: [...] If this chip really were socket-compatible with QPI Xeon's, that could be a game-changer, but I haven't seen language that seems sufficiently specific to draw conclusions. I'm curious how you came to your "could be a game-changer" conclusion. By all appearances, it's the architecture and the mountain of baggage it brings that has been pretty soundly rejected by all but a very few. Hardware implementation details seem to be well below the noise floor wrt significance... People have x86 software and they have enterprise software. If you're IBM, you satisfy that need by selling expensive plugins that allow you to include PC-compatible chips into your, um, mainframe installation. If you're HP, potentially you say, here's you box. Run whatever you like. Robert. Ok, I got you. And that's a concept that has been pitched often - usually as a bridge solution - but actual implementations have been rare to say the least. And not to beat on the recently deceased, but if dual-architecture ability was a sought-after feature, Advanced Modular Solutions might still be around... Cheers |
#6
|
|||
|
|||
New Itanium info
On Mar 4, 10:23*pm, daytripper wrote:
Ok, I got you. And that's a concept that has been pitched often - usually as a bridge solution - but actual implementations have been rare to say the least. And not to beat on the recently deceased, but if dual-architecture ability was a sought-after feature, Advanced Modular Solutions might still be around.... If I were HP, I'd be hanging onto that for dear life. There isn't much else to sell. Robert. |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
z/OS and Itanium | Robert Myers | Intel | 1 | January 25th 07 10:26 PM |
CPU Info? Itanium..... | Suzanne | Intel | 1 | August 29th 06 11:15 PM |
We NEED an Itanium 3! | John Savard | Intel | 15 | April 7th 04 01:15 PM |
Itanium experts- Building Itanium 1 systems from old parts | Matt Simis | Intel | 5 | December 20th 03 02:41 PM |
Itanium Experts - Building Itanium 1 systems (parts)? | Matt Simis | General | 1 | December 18th 03 07:02 PM |