Re: [freesci-develop] Would like to contribute...

[James Albert]

Sorry I haven't been in touch, my yahoo account is acting up.
Just wanted to mention that SDL has an undocumented picture scaling routine.  I 
forget what it's called but you can check the graphics header files for the 
function call.  It says not to use it because it's unstable, but I didn't have 
any problems with it (although it dosen't play nicely with an OpenGL buffer 
understandably).

- Jim :-)

Christoph Reichenbach <address@hidden> wrote: Kevin,


On Thu, Nov 17, 2005 at 08:21:49PM -0800, kevin wrote:
> I am an good c++ programmer, with a will to see  this thing finished.

  Any help on FreeSCI is greatly welcome! Though I should point out at this
point that FreeSCI is implemented in C, not in C++ engine, partially for
reasons of portability.

> Seems
> it's been years gone by now.

  It has been almost two years now, yes. We are planning on releasing our
current CVS state before the two years are complete, however ;-)

> But I dont know  the history behind what has
> gone on internally with this project.

  Let's just say that many of the people involved in it have less spare time
on their hands than they used to have.

> I am  mostly interested in having it
> extract the resources out of the later  SC1 sierra games such as kq5 and
> up. I have scoured the internet on  brians decompiler and this one. And it
> seems SC1 is a big problem.

  SCI1 itself is actually not that big of a problem conceptually-- though
there are some issues we still need to take care of for SCI1.1 (pixmap
scaling, script/heap resource separation, audio resources etc.)

>   Forgive my ignorance, but what is the difficulty so far? Can it pull
>   bitmaps, wav files, hotspot maps, and the scripts decompiled at least?  Or
>   is it just emulating the IDE a problem?

  I'm not sure what you mean by the IDE, so I can't comment on that. If you
try the latest Glutton CVS, you will notice that we can extract pixmaps, play
music and sound effects (modulo some bugs), and generally interpret the
scripts, however. We can also decompress resources for inspection (cf. the
'scidump' tool, though I personally find the built-in debugger in FreeSCI to
be more useful for most tasks), but there is no translation mechanism into,
say, MIDI for music, or PNG for graphics, mostly because no-one has bothered
to implement that (some of the longer-standing FreeSCI users here may recall
that we used to support that, "back in the days"). I believe that there's
still a tool in CVS which dumps all scripts present in a game into readable
text files (entry points, SCI class declarations, instances, method
definitions, all that stuff), but I also haven't tested that in a while.

  I can give you a very incomplete list of "what's wrong" with Glutton CVS at
the moment, which everyone else is invited to extend:


* Bugs, bugs, bugs. For example, background images behave strangely when
  running in scaled mode, and odd semantics of the SCI port system (a simple
  "window manager" system, if you will) screw up graphics etc.
* Savegames for the new VM have not been implemented yet. We may have someone
  working on that in the near future, depending on external factors.
* Boundary polygon checks have not been implemented yet (partially due to
  patent issues). We may soon have one person working on that, though,
  partially depending on external factors (*nudge*).
* Audio resources (which I believe are also of relevance to KQ5) and, in
  particular, their cue mechanism are not supported at the moment.
* We have no configuration subsystem; even though many choices are reasonably
  flexible in theory, decisions are compiled in ATM.
* Dynamic type information in the heap is not complete, making the garbage
  collector more conservative than it would have to be.

And probably lots of other things that I forgot about.

I should note that most development discussion takes place on IRC in #freesci
in OPN or freenode or whatever it's called these days (irc.debian.org works
fine for me); you're welcome to check out our (unsorted, not searchable) logs
at http://www-plan.cs.colorado.edu/creichen/freesci-logs, and join the
discussion, or we can continue through the mailing list.

  In any case, if you are serious about wanting to help, we would greatly
appreciate any help you could give us!


  Thanks for your interest!

-- Christoph



_______________________________________________
FreeSCI-develop mailing list
address@hidden
http://lists.nongnu.org/mailman/listinfo/freesci-develop



__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com