Ah ..hehe....well i am glad I translated the dutch tweakers.net article in to english.....so more people could see for themselves how badly the "Sh"iTanium performed in todays applications. Without doubt it will run IA-64 programs pretty well, but by supporting IA-32 (x86) they have made a bad move imho.
- They claim the processor is only for high end workstations and servers - Why do those need IA-32 ? For backwards compatibility they should have opted for a software solution, like alpha and its Windows NT. If you add x86 hardware support on a non-x86 cpu, you are bound to get beaten down by reviewers and the public unless its as good as normal IA-32 cpu's like p3 and Athlon. Make a cpu IA-64 only and I think nobody will complain...it is for servers, the software will be for servers only and nobody cares, except the people who should care nameley compaq dell hp etc . Alpha and PA-RISC cpu's arent so popular because they offer x86 compatibility ...they have their own native software . IA-64 Linux and UNIX are available, and so is/will be software wich can be compiled or ported with ease .
Well ive said it enough now but i still think its stupid . Just introduce a new CPU, - And DONT drop the word "x86", and instead say "PA-RISC, SPARC" ...say "Comparable or better clockrates", "64 Bit", "All new instruction set", "Microsoft and linux support" and yada there you go......no tweakers complaining about how niffy it's x86 core is, or how bad it will be performing. Just hurds of sysops and admins who will be like ...ah..64 bit...whistler..mysql!:)
Not to mention if they'd put their multiprocessing on level, because I understood the Sh...eh...Itanium still divides busbandwith by the number of cpu's on a given board.
Well enough for now...ill be heading home , trying to get my system up and running...IBM hdd dead because of short-cirquit .
FHWL, Michael da Kota |
