[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Is Linuxgames still alive?

Christian Reiniger wrote:

> >>> The linuxgames.org site - and this list - currently remind me
> >>> a little bit of the late lamented Freedows project; there
> >>> seems to be more talk about how to organize the site than
> >>> about code.
> I think our main problem is that most of us are simply busy with their
> respective game sdk. Keeping a website up-to-date, preparing code examples,
> writing tutorials/editorials/... is time consuming. And it's especially
> hard if it has to be done *in addition* to some "main duty".

It looks like people are mostly interested in code, but I for one would
like to see a lot more docs or specs.  Actually, I was surprised to find
on linuxgames.com (the 'other' site) a list of "porters" : there seem to
be people who know all or at least a lot about some topic (say
networking, sound dev/dsp, X) that they are confident enough to
say "just gimme your code and I'll port it".
  Where are all these people ?  Wouldn't it be nice if they could
produce a number of articles about porting code ?  I believe it
would particularly be a good idea to write a number of
"Moving from Windows/DX to Linux" papers; virtually all
big-buck PC (ie, nonPSX) games are for Windows/DX these
days, and if a site can explain how all the Linux equivalents of
common Windows rip-offs^H^H^H^H^H^H^H OS-concepts
like multithreading, networking etc work, it'd be a MAJOR hit.

 For example, I'm part of a team working on a really big game,
ie the 'real' thing with publishers and marketing and the whole
booha.  It's about multithreading, using DirectPlay for
networking, DirectSound for music and these for graphics :
- all-software, only DirectX for final output (easy to do with X)
- all-software with MMX asm.  Oops, gcc can't dig this.  A paper on
an easy solution would be great (say, how to compile with VC++
and link under Linux - i386 only, obviously)
- part software, part DirectX blitting.  Could be done with X.
But how, without digging up the 6 O'Reilly volumes ?..
SDKs can help here, but again don't deliver actual insight,
and may be too restrictive in their licensing (ie, GPL = no go)
- someday, full D3D.  Funky, 2D in 3D.  No X here I guess. :)
- Glide.  I already moved this to Linux, I guess this isn't worth
being mentioned on the site, it's trivial...
 If each and everyone of these components can be ported,
according to the list of volunteers, a good and solid doc on how
to do it may be better than just having it done by someone or just
uploading a bunch of samplecode.
(btw : I know about semaphores, X, sockets, fork etc
but it's still a big Unknown Realm thing imho to jump into
porting without porting-specific docs.  I really hoped these'd
be on LinuxGames.org...)

> Well, there are quite some people on this list that want to get into Linux
> game programming, right? My suggestion is: team up, agree <grin> to write
> some very simple game (important!!) for now and simply start desinging and
> coding.

This will produce new stuff, not ports.
I guess the people at Firaxis (Alpha Centauri) or SCI (Carmagdn2) who
are "looking into" a port are hardcore freaks who already know linux
inside out from their spare time projects.  But I don't have those
resources, without a "here's how to do it" site it would just take too
time, and I think the vast majority of Windows game coders out there are
in this position.

..My 0.02 Euro


-=<..alle bugjes zwemmen in het water..>=-