Mail Archives: djgpp/2006/12/07/23:15:10.1
Rod Pemberton wrote:
> "Alexei A. Frounze" <alexfru AT chat DOT ru> wrote in message
> news:6_OdnQRIt-7nVOrYnZ2dnUVZ_qadnZ2d AT comcast DOT com...
>> Blair Campbell wrote:
>>> On 12/6/06, Alexei A. Frounze <alexfru AT chat DOT ru> wrote:
>>>> So, what's the DJGPP's future under Vista and 64-bit Windows?
>>>>
>>> Hey yeah, are 32-bit DOS apps able to run under 64-bit Windows? I
>>> know 16-bit apps can't.
>>>
>> They can't for two reasons:
>> - there's no any kind of DOS, which they need for all I/O
>> - they start as 16-bit DOS apps anyway and only after that switch to
>> 32-bit mode
>>
>> I guess the only way to go would be to do the following:
>> - build DJGPP as win32 apps so it can run under windows (maybe just
>> like Open Watcom -- it essentially has 2 toolsets)
>> - (cross-)compile code for DOS, so that the DOS applications can
>> still be compiled, even though can't always be executed
>> - provide source-level compatibility (where possible) so that some
>> DOS applications can be compiled into functionally
>> (nearly-)identical win32 apps -- everything that explicitly deals
>> with DOS interrupts and special memory locations (BIOS data area)
>> will have to go. Some things (like console I/O) will probably be
>> possible. I wonder if it's possible to make a few functions so that
>> the direct graphical output (either to a window or to the entire
>> screen) just like in DOS...
>>
>
> Alexei,
>
> As you know I use DOS toolchains. But, I think DOS code development
> and toolchains have seen a huge decrease since Gates officially
> declared DOS to be dead with the release of XP in late 2001.
Twice dead -- win 9x is also officially dead -- no more support for it.
Though, not thrice -- it still has to work in MS' virtualization software
for a while.
> Since
> then, except for some of the early internet DOS archives, DOS sites
> and code are disappearing like a forest on fire... Those that
> haven't converted to XP have gone to Linux.
>
> I doubt that most DOS toolchains will advance much further.
And that's probably not needed much. I liked DOS for being easy to deal with
(that's not the general user thinking, though, rather low-level programmer
:) and being small. Right now it seems to me if I want to continue
experiments in the GUI area on a PC, I need to download 500 MB+ DirectX SDK
(not to mention some big compiler for Windows, well, OW isn't too big) and
write some wrapper for DirectX that would allow me to draw as easily as I
could in DOS. Yet, to test a video driver (such as VGA) I still need to go
to a VM or boot my PC into DOS or directly into my OS. That's an unnecesary
complication to me.
> Multi-platform toolchains like OW will continue to advance but not in
> the DOS area. To use DOS toolchains in the future, one will probably
> have to look into creating a full DOS emulation environment. My
> thoughts for a 64-bit ""DOS"" would be to start with a 32-bit C based
> partial emulator like DOSBOX and push it to the next level by
> combining it with any additional or missing components and emulation,
> such as DPMI or SB audio, from FreeDOS-32, QEMU, and/or DOSEMU,
> FreeDOS, etc.
Actually, it seems like Windows programming isn't such a big deal compared
to what it was some 8-10 years ago. The documentation is available, MS even
gives away stripped versions of their compilers, there're more Windows
programmers now and there's internet to share, something of a rarity or
scarcity back then.
> You're questions about Vista on a few NGs brings up an interesting
> point that I've been wanting to discuss. Do you think that long term
> Linux can compete against Microsoft given the small amount of capital
> available to Linux versus MS (7 Billion on Vista alone...) and the
> continuing rapid change or advancement in PC hardware?
I don't see that much of a difference other than it's faster and has higher
capacity. Multiprocessor systems have some issues but those aren't an
entirely different thing to program. I/O (like software) will be constantly
changing to bring higher speeds and to make more money by forcing to buy new
stuff, though not revolutionarily.
As for competing... I don't know. We have yet to see free office software
comparable to that of MS or significantly better. And since MS is going to
release (or has already released) the specs for their office documents'
formats, it's gonna be interesting to see if compatibility and other things
in OO catch up and become better. There're problems and bugs in both free
and commercial software and the product life cycles are different. It's hard
to tell whether lots of resources (people and money) is the key to the
answer. Sometimes it just doesn't work as things become overly complicated
and poorly scalable at which point the effective value of the resources
slows or stops its growth with the growth of their amont.
Btw, can someone tell me what important things happened in the Linux world
over the last 5 years or so while MS was trying to make Vista? I wasn't
keeping an eye at Linux. And that's lots of time, I'd expect some very good
and noticable changes for the end user. Any?
> (I'm not sure if DJ wants this discussed here. If he complains, I'll
> catch your reply if you post to alt.os.development.)
No prob.
Alex
- Raw text -