Nick S Bensema wrote:
>
> >Maybe. I need to devise a more efficient way to store tracks, but even
> >the kludgey, brute-force method I use now only takes about 500 bytes for
> >a decent 30+ second one lap ride, and the rest of the code is pretty
> >straightforward (eg. small) so far. I figure if I keep the sounds real
> >simple, I could do a standard 2K game with quite a few challenging
>
> The standard is 4K. Hope this helps. :-)
Yeah whatever ;) The first assembler I used didn't even have an ORG
directive. It took me a while to figure out a) that I needed to pad to
2k or 4k (oh gee, look at the sizes of all those .bin files...) and b)
how to do it exactly, correctly in debug manually after assembling.
Hmmm, nothing in Stella about that. Nope, nothing in DejaNews
either... I hadn't even heard of this list then.
So far I'm no where near needing 4K. Why can't I go all the way to 8K
minus TIA and RIOT address ranges without any bankswitching? Exactly
(in 25 words or less), what's that all about? There are lots of details
even disassembling and careful analysis won't reveal (stuff i guess
that's tied up in ancient data books I've sure never read). Programming
ROM is a little different than "regular" programming, innit? I
typically just figure things out after many hours of late night
head-scratching and best-guessing, coding, assembling, emulating, and
observing (add water and repeat)... The idea of a FAQ or basic guide to
describe the fundamentals (without once repeating any fact already,
clearly spelled out in the Stella docs) would surely get more people
going. It's too bad we're all too busy getting going ;)
> And hey, if you can figure out the Supercharger bankswitching, and can
> arrange your program around it, you got 6K.
>
> Who needs multiloads?
>
> >tracks by building from track section templates. The big problem with
> >this game is good perspectives and motion (basic algebra is a bitch in
> >6502, never mind applied geometry and physics). I suppose I should take
> >apart and study Pole Position and Enduro to try and swipe some good,
> >basic ideas to move on from, but that's the sort of thing I haven't much
> >time for these days (and got stuck on with some flaky Unix problems when
> >I thought of doing it a while back).
>
> Enduro, I believe, is 4K and should be disassemblable. Night Driver is
> 2K and should be even easier, though it sucks.
>
> I'd be surprised if either of them used actual applied geometry and
> physics. The best example of physics I can think of is Space War,
> and if my tests are right, things like the inverse square law of
> gravity apply. It's not like Gravitar. To be honest, the more I
> play Gravitar, the more I'm disappointed in the physics it uses,
> it being 8K and Space War being 2K and all that.
The physics are kinda secondary, and I'd settle for any at all. The
real thing is switching from hard-coded horizon/distance events to
elegant equations for everything.
> >I'm tempted to set this aside and just do something simpler--just to
> >have something completed, but I think the last thing I need right now is
> >yet another brand new project ;) Before starting this game I did a
> >bunch of fairly playable demos to get the feel of things, including
> >something pretty similar to Space War and a Pong game where you shoot at
> >the ball to hit it (that one was devised simply to make me use *all* the
> >graphic objects simulataneously), but felt that wasn't where I wanted to
> >go. I'll get this done someday.
>
> Some of us on the list might be interested in your demos. There is
> very little source code out there that has comments written for it
> at design time.
I'll have to re-mount my hard drive in DOS and see. Not much in the way
of comments for sure. Some of the common routines used in all my early
efforts (basic TV loop, make a sprite, make it move, shoot, tell it
collided, etc... small sections I took the time to comment well for the
benefit of the curious) I already posted to the Classic mailing list
six/seven months ago in response to other, "how do I do simple stuff"
questions. Most of the demos mutated constantly over time (wonder what
twiddling this will do???) and might well have been left in a seriously
buggy, discontinued-idea state. Indeed, most started out as wacked
effects that kinda looked like something, purposely evolved into that
something for a while, then as a result of a huge miscalcution/bug, into
some other wacked thing ;) Stream of consciousness. Most of the people
who were interested in getting going back then seem to have vanished.
Hopefully they're all too busy coding. One thing 2600 programming
definitely takes: stamina.
Now that there are a new batch of people chatting about this stuff, I'm
starting to feel like doing it more (at the expense of other things that
are suddenly losing their appeal?). The other thing 2600 programming
definitely takes: company. It can be a very lonely endeavor.
--
mor@xxxxxxx
http://www.crl.com/~mor/
--
To unsubscribe, send the word UNSUBSCRIBE in the body of a message to
stella-request@xxxxxxxxxxx
