Significant Pure Assembler Application In MASM ?
in [ASM]
|
Prev: aeBIOS Test Request
Next: CMOVcc vs. Jcc
From: //o//annabee on 7 Oct 2007 00:39 On Sat, 06 Oct 2007 22:00:49 +0200, santosh <santosh.k83(a)gmail.com> wrote: > //\\\\o//\\\\annabee wrote: >> On Sat, 06 Oct 2007 20:50:21 +0200, Betov <betov(a)free.fr> wrote: > >>> Oh, i have no doubt that 64-bit will become main stream, with times >>> going. What i fail to understand is the reason why the actual >>> Assemblers were so fast at upgrading, whereas, twenty years after the >>> death of DOS, there still exist a significative number of guys >>> programming for DOS, nowadays. Of course, it is quite easy to >>> implement, but... what for? You say "Multimedia"? Well, as if XMMX >>> did not exist under Win32. > >> 1) I am lazy >> 2) I want to experiement with 64 using RosAsm. >> 3) Wolfgangeclaims it has huge advantages. >> 4) If "reason" has anything todo with it, we would not be using >> computers. >> >> We would be realming the streets naked..,,,,,,,, > > In Norway? Do you want to catch your cold of death? :) > >> :)) :)) In the summer we have quite nice temp, and the winters, until the golf stream has died it may become warmer.... Anyway, if the serious rumors about the pole melting is true, (Which I now must consider a fact) and is indeed caused by human additive activity... I am beginning to wonder if a PC on every desktop will be viable in the future. At all. My AMD64 generates so much heat I can use it to warm my "bunglalow". I am not using it. Simple as that. Each time I now want to play a game, I have to face the fact that this is now allmost to consider a luxury equal to a crime. If theese thing are factual, I will have to reconsider all of my life again. I hope I am wrong, I hope they are wrong. But this is not what i hear everyday. The party may soon be over..... and at least I can no longer justify many things that I did not consider before. Tell me I am wrong, and I will listen. >> >> Admit it Ren�. Programming is such a stupid activity, it REQUIRES a >> geek todo it. > > Programming is logic. Applied logic is science. Science is a method to > satisfy innate human curiosity. It's nothing to do with "geeks." People > have been observing nature and seeking to understand it, and in the > process, improve their lives, ever since the dawn of man. It's also an > inert and externally propelled phenomena in lower organisms, through > the combination of mutation and natural selection. > > Even religion is an effort to understand our existence. It's just that > unlike science, it does requires blind faith and dispenses with > empirical consistency. On the other hand it "explains" everything there > is to explain in one stroke, while science moves towards that goal > tortuously slowly. > >> Now shut up and go and implement 64 bit in RosAsm. What we all crave. > > Do you mean port RosAsm to 64-bits or enable 64-bit code generation? I > think the former is unnecessary. Yes, to be able to generate 64 bit apps.
From: Rosario on 7 Oct 2007 03:03 In data Sat, 06 Oct 2007 21:13:22 GMT, Robert Redelmeier scrisse: >Charles Crayne <ccrayne(a)crayne.org> wrote in part: >> On 06 Oct 2007 07:48:22 GMT Betov <betov(a)free.fr> wrote: >>> Considering the various actual ones, is there some >>> probability for having, later, a Processor emulating, by >>> default, any other Processor as supported by the main OSes, >>> say, on a PC and a embeeded Phone? (So that porting from >>> here to there would no more exist). >> >> Technically possible, but economically unlikely. The current >> marketing requirements for chip design seem to be increasing the >> raw speed without a corresponding increase in power consumption. > >The "flagship" x86s from Intel and AMD certainly follow this. >They cannot increase electrical power much due to physical laws >such as heat removal and Ohm's law (current vs voltage drop). > >However, there are VIA and other embedded x86 CPUs around. >And the fabulous success of ARM points the way. Annual sales >in the 100s of millions, and ASP/mm2 probably well above all >x86 except the very top end. > >I could see a world with small stateless appliance boxes. >Walk up, plug in a memory stick and it boots your OS into >your execution and data environment. so the poor low level programmer have to rewrite the same programme each environments or he/she have or use a very high hll language like java ... i think it is better that hardware people has some consensus for a minimal cpu mode (30 to 50 instructions n� x registers of 8,16,32,[64?]bits registers, AND, OR, PUSH , JMPS, CALL, etc etc and fpu too) or a minimal enviromets () so define an assembly language for that cpu; so c or c++ compiler can produce assembly for that cpu and so to be portable by hex, and assembly programmers can have a cross cpu to program the same the Operative Systems people can define a set of command cross OS for doing "visual" programming all this because the cost of software seems bigger than cost of hardware so programs have to be maximus portable but possible this is an error, i write errors >-- Robert
From: Betov on 7 Oct 2007 03:05 santosh <santosh.k83(a)gmail.com> �crivait news:fe8s3n$70p$1(a)aioe.org: > But now you've boxed yourself with Windows! At least make it output ELF > files. You don't need to port the whole IDE to Linux, just the > assembler alone would do. You seem to be missing the concept of RosAsm. Again, the real point is with *integration* inside a single IDE for RAD. Without this, RosAsm would just mean nothing at all. So, the strategy i am actually considering is: * Re-organizing RosAsm Source for grouping all OS calls into a dedicated TITLE (for maintaince, in case GTK would not be that reliable, at a backward compatibility point of view). * Port RosAsm itself to GTK. There exists a "GTK for Windows", that should make the job a bit easier, saving from having to deal with Linux. * Implement the ELF Format. * Compile RosAsm as an elf. My estimation of the work time is around two years. Probably way more because of my health. And, if two or more years, Ubuntu is still in the same state as today, this will mean that Linux will never win more than its current 1.5% or 2% of the market place (against 90% for Windows). Considering the actual number of RosAsm users (around 20, i know about), this makes: 20 / 90 * 2 = around... zero users. Ubuntu is a more risky bet than ReactOS was 10 years ago, in my opinion. And as i am becoming too old for the job, this bet sems to me... scaring. Also, there is the Debugger problem: Without an *Integrated Source Level Debugger*, RosAsm would mean nothing, and, as a matter of fact, GTK does evidently not offer any Debug Api like Windows does. Maybe those Api exist somewhere else in Ubuntu (i suppose not), but if a real Processor level Debugger is too be re-written, this is again, a couple of years of works to be add to the price to pay. Betov. < http://rosasm.org >
From: santosh on 7 Oct 2007 03:27 //\\\\o//\\\\annabee wrote: > santosh wrote: <snip about looks of s/w> Well, I agree that if other more important technical issues are done, then there is nothing wrong in improving the visual appeal of the s/w. I don't agree that considering the functionality alone is an "elitist" view, and I don't agree with your reasons for the success of MSN. >> Also keep in mind that much of commercial s/w development and >> computer usage, especially by lay people took place first in the USA >> and parts of Europe and only later on, percolated to the rest of the >> world. It so happens that the USA is "home territory" for Microsoft, >> so that fact enabled them to market their OS intensively, especially >> towards their target audience who didn't really know squat about >> computers, programming, and the existence of technically superior, if >> academic, alternatives. >> >> Of course by the time Linux was competitive enough, Windows had >> already achieved a suffocating monopoly on PCs. > >> If anyone is to blame for this, it's Europe. Technically and >> scientifically Europe was, and is, close enough to the USA. Why did >> no one in Europe produce an alternative to Windows? Europe was the >> birthplace of modern science, but I'm surprised that they are not >> competing, (friendly), with the USA, but rather, meekly >> getting "assimilated" with it? > > I have asked this question many many times. I do not know. Seems also > extremly strange to me. First of course, writing an OS like windows is > _not_ a simple task. Its more like a miracle they wore able todo it at > all, quite frankly. No. The first versions of Windows were relatively simple and built on top of existing software like DOS. A complex piece of s/w is always written and improved incrementally and by numerous programmers. It's almost impossible for a single person to write an OS similar in functionality to Windows and the UNIXes. But as a coordinated team, it can be done. It's been proved by Linux and BSD, essentially team efforts. > Anyway, Linux _must_ get good at selling itself. > It should have an _working_ installer for windows first, able to do > the entire install, so simple that anyone can do it. What are you talking about? All the current mainstream distributions like Ubuntu, Fedora, Debian etc., have installers that even a child could use. It was actually easier than installing Windows XP. > I tried the MenuOS installer, both crashed here. Do you mean MenuetOS? I'm talking about practical alternatives to Windows, not "hobby" OSes. > I tried Ubunty. An ok _beginning_. It more than a "beginning." Tell me one reasonable thing that you can do with your Windows that I cannot do with my Ubuntu here. The fact is that you're uncomfortable because you do not know Linux. That's all right. Every time we use a new s/w we don't know about it. But we can always learn it soon enough. Just get a Linux usage book and you'll be up and running soon. > I will soon try it again. Sooner or later, maybe I > move to it. But I am not moving permanently until it as least as > stable as NT, and as easy to configure and setup. I don't know about ease of configuration, (though things are _vastly_ improved since the early days, I admit that some configuration issues still are obscure to sort out), but as for stability, I found no difference between Linux and Windows NT. They are both very hard to crash and are very stable. You and Betov seem to have the gift of crashing Linux everytime you put your finger on it? > We ALLREADY had all > those pains in theese years and years with windows. I am not going > todo that again. It is ONLY fun the first time, and the fun went away > with OS crash number 100001, back in 1995. Well, as the creator of GPL says, "are you willing to give enough importance to freedom that you're willing to sacrifice some bells-and-whistles to use Free Software, or is it that you cannot be bothered to actally take action on your ideals?" [The text in quotes is a paraphrase, not an actual quote, of RMS] <snip> > You are basically wrong in thinking NASM has better syntax then > RosAsm. It is close, but RosAsm is closer to the better one. And > RosAsm naturally allows for far faster devs, and for easy structuring > the large sources. Can you point out any syntax elements that you consider are superior in RosAsm, as compared with NASM? Not features of the IDE, but of the core language. <snip>
From: Betov on 7 Oct 2007 03:28
//\\\\o//\\\\annabee <w(a)w.w.w> �crivait news:op.tzs7segkin6out(a)darth-fpsr: > I tried Ubunty. An ok > _beginning_. I will soon try it again. Sooner or later, maybe I move > to it. But I am not moving permanently until it as least as stable as > NT, and as easy to configure and setup. You are missing the point that the Windows users DO NOT install Windows: They all buy a Computer with Windows inside. Where the things are changing, is that, because of the oncoming of Ubuntu (which is close to usable), several gears sellers are beginning to sell PCs with Ubuntu on board. In this case, only minor problems are encounted. Even less for a real flat user, who wouldn't try to modify anything, and just live happy with what comes out of the box. About the look&feel, did you saw screen-shots of the new 3D desktop? Will Ubuntu really succeed is another question, which answer does not really depends on Ubuntu, but on the market behaviour. There is a hope, in my opinion, because almost everybody (gear producers included) want the death of MicroSoft, for many different reasons, which all resolve down to the monopol abuses problems. So, the situation, on the market place, could as well switch dramaticaly in the next year, or keep the way it is since 20 years... for 20 more years. Betov. < http://rosasm.org > |