Thanx to all the people who answered my question. Probably I couldn't explain what I really thought. However I'm an Open Source fan and my doubts were only about the growing power of .net. I remember the end of the 16 bits applications and looking at .net, a sort of Java clone but wider, I think that in the near future the present 32 bits apps will become old stuff. Therefore, even if I always hope to build program in assembler (that's very difficult), I think : is it worth ? (I hope my bad English is clear). What Hutch says is the answer I wished, but It seems to me that you too have doubts about that (how long ?). Will you be able to make masm32 survive to .net (or in .net) ? Thanx
Posted on 2003-07-17 14:11:48 by newbie
I'm no expert, but believe I can pass along some relevant information.

.NET is a development framework used by many high level languages in Microsoft's recent IDEs. If you were to ask if learning Microsoft's older MFC (Microsoft Foundation Classes) for C++ programming was worth it, a more solid answer could be given: "forget it, learn .NET".

Masm and assembly language programming in general is a specialized art. There are a number of reasons why someone would want to learn how to program in Masm, and a number of reasons why not to. I don't know by how much, but newer 64-bit processors will extend the instruction set and 64-bit Windows OSes will possibly radically (?) change the Win32 API calls. The basic PE file probably won't be changed in Longhorn.

Unlike most higher level languages, assembly is not cross platform, so while targeting Linux systems won't require much transition, creating Mac OS X programs is out of the question. :grin: You may be familiar with the Mono project which aims to recreate the .NET framework for Linux.

Many (but not anyone here) will say that assembly programming is already dead. It all depends on what you want to do and what you feel comfortable learning. I have found that the x86 assembly syntax is more interesting then C\C++, though now that I'm slowly learning how to use Win32 API calls, C isn't that bad. Any program that can be created via another language can ultimatly be created using assembler, though it might take more time to do it. But assembler can create many specialized programs that no other language can create.

I hope I at least touched on some aspect of your question.
Posted on 2003-07-17 16:50:32 by Masmer
Time will tell. :grin:

Assembly programming was around since 1950 and still running(2003). :)

Wow, look at the years that went by... hmm! she was there at the start of the programming era, she was there before everybody else and she will be there to the end.

When the time comes that assembly is totally dead, I myself already have found a new career aside from programming. Why? I know that it's death will spell doom to the future of every programmers.

p.s. there's only 1 secret to kill assembly and still survive thereafter... and I ain't tellin' :grin: :grin:
Posted on 2003-07-18 02:27:49 by arkane

If I was into predictions, I would guess that 32 bit binary code will outlast .NET and that is the distinction you are talking about. Whether you write in it C, C++, Pascal, Basic or assembler it is my view that it will survive longer because they collectively haqve a large user base and there is a mountain of existing software that Microsoft would not be game to try and break with catastrophic results to their operating system sales.

Many people including various size companies own DOS era software and much of it is still in use and if Microsoft try and break the backwards compartibility that has existed since the early 80s, many existing users will not buy their later versions as they cannot afford the software to run on it.

Whether .NET is useful or not, the day Microsoft tries to abandon the backwards compatibility of software that people have bought over the years, they will fail on a large scale. This factor alone will ensure that code as we know it will survive for some time to come yet.

Each attempt Microsoft have tried to kill off the old stuff or introduce radically different stuff has had trouble getting off the ground and the more they do it, the less most people trust them as they have seen it all before. DDE, OLE, COM and trying to force UNICODE on many who will never need it and while .NET may have its virtues, I am incline to see it in the same light.

Posted on 2003-07-18 06:59:10 by hutch--
given the state of the world, the real crime is forcing ASCII on those in dire need of unicode. Memory hasn't been a valid excuse since 94-95. String functions should have unicoded a decade ago, transparant to the programmer.

Dans tout fa?on le monde ne fonctionne pas tout en 'Engrish' ;)

my 0.02 ?
Posted on 2003-07-18 08:03:10 by Hiroshimator
Arkane, I think I know that 1 secret, and it involves some special expensive PIC , that is easy to use, and it has its internal interpretator of a well-known language :) That would be a disaster - people will say 'blah, my 12GHz PC is so slow, I can't play solitaire at all!'.

In my opinion, it will take at least 50 or 100 years to remove x86 asm completely, during this period of transitioon, x86 will be morphing into something extremely similar. So, for all these years you'll have to learn very few new things. I think haven't had any glimpse of CISC, is Mac code (motorolla 68000 I think) CISC? If Macs have CISC, then I trully hope the cpu manufacturers keep away from that awful code structure.

pls excuse my ingnorance 'bout other things than x86 :)

ah, I've read about the opteron - it has backward compatibility with x86, but I think it was emulated.
I suppose the chance all people to jump to new asm is like one would jump from PC to Mac :grin:
Posted on 2003-07-19 01:51:27 by Ultrano
Originally posted by Ultrano
ah, I've read about the opteron - it has backward compatibility with x86, but I think it was emulated.

Opteron (athlon 64) has full hardware support of current x86 architecture (as i know even some better in speed than Athlon XP) plus some new registers, etc... (like i386 vs. i286)
IA64 uses emulation.
But Intel is developing "secret" alternative for A64, known as "Yawmill", just "in case" :grin:
And often, when new CPU whith new "features " is released, the only way to use it the first time is assembler. Just some time after - compiler.

P.S. in russian "net" means "no" :grin:
Posted on 2003-08-13 02:10:04 by S.T.A.S.
it's rather "njet" :)
horosho :grin:
Posted on 2003-08-13 04:03:48 by Ultrano