Index
> Main > Going crazy using AMD64 or INTEL64Goto page 1, 2 Next |
| Going crazy using AMD64 or INTEL64 | |||||||||||||||||||||
|
|||||||||||||||||||||
| Total Votes : 17 |
| Author |
|
|
revolution 13 Oct 2009, 03:22
What help are you asking for?
If you have specific questions then feel free to ask them. |
|||
13 Oct 2009, 03:22 |
|
|
LocoDelAssembly 13 Oct 2009, 03:55
Note that currently the term Intel64 refers to the same architecture AMD64 does: http://en.wikipedia.org/wiki/X86-64
|
|||
|
13 Oct 2009, 03:55 |
|
|
Remy Vincent 13 Oct 2009, 06:51
revolution wrote: If you have specific questions then feel free to ask them. It is about working more slowy than expected because of more effcient instruction cache, and also how could one be sure that his using a 64 then 128 then 256 procesor, because you could only work with one 64 128 256 DOCUMENTATION ( for example ) _________________ Groups lower your IQ |
|||
|
13 Oct 2009, 06:51 |
|
|
Remy Vincent 13 Oct 2009, 07:02
LocoDelAssembly wrote: ... http://en.wikipedia.org/wiki/X86-64 that is c r a z y ! MENUETOS64 is mentioned, between OPENBSD64 and SOLARIS64, so an """very small""" operating system is trying some kind of competition with operating system more used. I Hope Menuet OS could reach the top of the list at wikipedia X86-64 topic! _________________ Groups lower your IQ |
|||
|
13 Oct 2009, 07:02 |
|
|
revolution 13 Oct 2009, 07:35
Remy Vincent: I don't know what you are asking. We are not likely to see 128bit processors any time soon, if at all. And 256bit processors are not ever likely to be useful in any practical sense. Unless you are thinking of SSE/AVX?
What do you mean when you say "more eff[i]cient instruction cache"? Which is more efficient than what? How are you testing the instruction cache efficiency? |
|||
|
13 Oct 2009, 07:35 |
|
|
pal 13 Oct 2009, 10:09
revolution wrote: We are not likely to see 128bit processors any time soon, if at all. And 256bit processors are not ever likely to be useful in any practical sense. Unless you are thinking of SSE/AVX? You sure about that? Check this out: Code: http://www.pcpro.co.uk/news/enterprise/352270/microsoft-leaks-details-of-windows-8-and-windows-9 I guess they mean native 128 bit, otherwise it would be a bugger to program, let alone if it is even possible. 2012 at the earliest, so we have a wait, but we could have native 128-bit GPRs, unless I misinterpreted the article. |
|||
|
13 Oct 2009, 10:09 |
|
|
Madis731 13 Oct 2009, 11:54
@pal:
That amazes me quite a bit. I thought 128-bit was as far away from 64 as 64 was to 32!?!? Why don't WE hear anything about future CPUs? Intel has many CPUs coming out, but none with 128-bit GPRs. Maybe they still mean the SSE-compatibility. Don't you think? |
|||
|
13 Oct 2009, 11:54 |
|
|
r22 13 Oct 2009, 13:35
pal wrote:
The 128-bit reference in the Windows 8/9 info leak is presumed to refer to their 128-bit file system (The one that was supposed to be released with Vista). But who knows what'll happen a decade from now. I predict 128-bit architecture by 2019 !!! |
|||
|
13 Oct 2009, 13:35 |
|
|
revolution 13 Oct 2009, 13:49
I predict not before 2069. 60 years for 30 doublings means ~2^64 bytes in the bigger systems. But even then I would be very sceptical that there are still 30 more doublings remaining in the future, most likely far less than 30 before the "wall" is reached and things level off.
|
|||
|
13 Oct 2009, 13:49 |
|
|
pal 13 Oct 2009, 13:55
I didn't think they meant native 128-bit CPUs. It is interesting though as, according to the article, they are talking with the major hardware production companies - Intel, AMD, IBM etc. I don't see why they would have to talk to them if they were just using SSE or the upcoming AVX or something alike it.
The reason that I don't think they mean 128-bit GPRs is because, as Madis said, there was a huge gap between 32 and 64 bit. Most people are still using 32-bit OSs, even though all major CPUs being produced now are capable of supporting 64-bit. Another thing is many people don't have enough memory to make 64-bit useful (4GB). I only have 3GB of RAM but I like 64-bit programming when I can as it is much better with all of the registers, sizes etc. Since this is true, how many people will have enough memory to take advantage of 128-bit hardware. I don't know how many GB that would be though (16 exabytes right?), bit I doubt many have that much (apart from super computers...). Interesting that Microsoft are creating a 128-bit file system. That must be a bitch to code right? Although it must offer quite a few advantages. |
|||
|
13 Oct 2009, 13:55 |
|
|
revolution 13 Oct 2009, 14:03
Maybe MS just wanted to beat UFS and claim bragging rights for the largest capable FS.
Don't forget about the death of NTFS.  |
|||
|
13 Oct 2009, 14:03 |
|
|
Remy Vincent 13 Oct 2009, 15:37
revolution wrote: I predict not before 2069 ... My own problem is that I need the whole 256 OpCodes ( with small examples ) , and also I need the whole 128 bit OpCodes ( with small examples too ) just before I start coding in 64 bit, more longer programs. You Dudes would be appreciated to CHECK my own anticipation of the whole 256&128 bit OpCodes ... |
|||
|
13 Oct 2009, 15:37 |
|
|
Remy Vincent 13 Oct 2009, 15:40
Remy Vincent wrote: ... 256&128 bit OpCodes ... My main question was about AMD ability to achieve AMD64+AMD128+AMD256 projects, or INTEL64+INTEL128+INTEL256 projects are much more dynamic ? |
|||
|
13 Oct 2009, 15:40 |
|
|
revolution 13 Oct 2009, 15:47
AFAIK neither AMD or Intel have any such plans for 128bit (forget about 256bit, that's not going to happen). Indeed I don't know of any chip maker considering 128bit GPR's yet. I think your anticipation is very premature. Wait 60 or so years then start getting excited about 128bits.
 |
|||
|
13 Oct 2009, 15:47 |
|
|
asmfan 13 Oct 2009, 18:28
Yup, with AVX and future Windows 8 & 9 already claimed "lot-of-bits" (more than 64:)) OS we'll see the bright "lot-of-bits" future and the death of assembly coding because on new complexity and absence of jobs.
btw. have you seen anywhere asm jobs worldwide? that's it. Gone w/ lack of professionals. If you'll find one point me on pls. interesting to now who use all its (asm) powers /SSEs, AVXs/ now. However saw some volunteer proj. like x264 codec that use tons of asm x32 & x64 (nasm IIRC). Do U participate? |
|||
|
13 Oct 2009, 18:28 |
|
|
Remy Vincent 13 Oct 2009, 19:03
asmfan wrote: ... tons of asm x32 & x64 (nasm IIRC). Do U participate? May be i'm already gone asking "tons" of 64 128 256 DOCUMENTATION, but it's true that I don't hook up from 64 bit OpCodes to 256 bit OpCodes or more ... Where is the documentation ? Well, replying to "the" where is your heart question, I just say that I'm addicted to 16/32 bit 80x86 coding, so it's difficult to accept a what you did was low arguing, because I'm trying to code searching for 64 128 256 DOCUMENTATION! |
|||
|
13 Oct 2009, 19:03 |
|
|
mattst88 15 Oct 2009, 04:15
pal wrote: I didn't think they meant native 128-bit CPUs. Uhh... pal wrote: I guess they mean native 128 bit. And what the fuck is Remy Vincent going on about? I can't make any sense out of anything he's said in this entire thread. Also the first time I've ever seen anyone put quotes around the word "the". See what I did there? _________________ My x86 Instruction Reference -- includes SSE, SSE2, SSE3, SSSE3, SSE4 instructions. Assembly Programmer's Journal |
|||
|
15 Oct 2009, 04:15 |
|
|
pal 15 Oct 2009, 07:50
Yeah, the two quotes should be the other way around. I have, since the first one, changed my mind. As there is a lack of information concerning the article, there are mixed opinions to what it means. I now think that they will be using a 128-bit file system on 64-bit hardware.
I also have no idea what he is on about. P.S. mattst88: nice instruction reference website. I have been looking for yours for a while but couldn't find the website. Bookmarked  |
|||
|
15 Oct 2009, 07:50 |
|
|
mattst88 28 Oct 2009, 04:59
pal wrote: P.S. mattst88: nice instruction reference website. I have been looking for yours for a while but couldn't find the website. Bookmarked Ahh, I'm only two weeks late to respond, but thanks a lot! I really appreciate it. Maybe that'll provide me with a bit of encouragement to continue adding instructions. It's just that once I made it work, adding instructions was so monotonous. Also, if you've got any suggestions for improvements, I'd like to hear them. I think what I really should do is write a Perl script to parse the binutils source and generate the list of instructions from it, minus the descriptions. Maybe parse the intel manuals for the descriptions.
Ah, so many projects, so little time (and such a short attention span). _________________ My x86 Instruction Reference -- includes SSE, SSE2, SSE3, SSSE3, SSE4 instructions. Assembly Programmer's Journal |
|||
|
28 Oct 2009, 04:59 |
|
| Goto page 1, 2 Next < Last Thread | Next Thread > |
Forum Rules:
|