AA3D- Ascii-Art 3D engine / News: Recent posts

Our project activity reached a new record! 0%. Obviously, I have failed to keep buzz up about this project. I guess I'm not working on an instant messanger or a bittorrent client though. Shouldn't expect too many hits, I suppose. <g> Anyhow, I seriously do plan to get a 1.0alpha release ready this week. I can't talk features, but I'll be dissatisfied with less than perspective viewing matrices and improved stack implementations. Plugging any detectable memory leaks would be good too, but that's more of a beta thing. <g>... read more

Summer semester ends this week and I'm behind on this project. It's so frustrating. I'm going to just spend a day this weekend sorting this out. Whatever I can get stable on Sat. will be the 1.0
release. The big problem with curses has been that
though it works under windows's cmd.exe, it doesn't display inside MinGW (our windows gcc). Very strange, and clumbsy to develop with. Should be fine though.... read more

Well, PDCurses is acting strangely under MinGW. I should be developing on Linux, but that's another whole story. Anyhow, it runs OK from cmd.exe, but in the MinGW rxvt(or whatever) terminal it just hangs. Very strange...
Summer classes are doing the finals week thing, so I hope I don't fork up and not get everthing done.
If only I could fork...

<pre>
void my_time( PERSON *me )
{
if (fork()==0)
catch_up_on_school( me, &class_list );
else if (fork()==0)
finish_release( me, "aa3d", "0.9 alpha" );
else if (fork()==0)
organize( me, "ACM Team for november" );
else nanosleep( (timespec *)FOREVER, NULL );
}... read more

Everything seems to be going basically according to plan. The hope is for 0.8 this weekend. The good part about that is that I'm calling .8 alpha, because we'll have the curses control in for standard sizes. That leaves the big hurdle to 1.0 the documentation. I'm a bad, bad software engineer. In case you don't know yet, I'm just a hacker. Documenting is just not something I think about. Anyhow, I WILL put some docs together for 1.0.

It goes to show you no matter how many years you spend coding, pointers can still trip you up. Long story short, it's embarrassing that the matrix stack bug came down to needing:

*stack = next
instead of:
stack = &next;

*sigh*

We went 0.5dev this morning, despite some problems in the matrix stack engine that cause the example program not to work properly, but its important to release when significant changes have been made.

Who knows. Maybe the community will fix my matrix stacks. heh

Seriously, this will be fixed, and a whole slew of new features will be in for 1.0, which may be the next release.

I've been hacking around with the internals and some of the library interface and it's interesting to discover clever ways of doing things, then realizing SGI already does them in OpenGL. I have to wonder if I really knew a matrix stack gives better transformation control, or whether I just remembered subconsciously about how OGL handles its matrices. Same story with the whole concept of implementing a state machine: It's the only reasonable way to approach the problem. It's late; I ramble. 1.0 may be sooner than expected. Still have to get the curses going.

Well, the wonderful people here at SourceForge have accepted our project. The prototype is doing pretty well. I'll try to work up a feature list and a more detailed final design. I've got class this summer, so things may be irradic.