A83: TSE nearing completion

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

A83: TSE nearing completion


    We are aproaching the day of the apocalypse, someone is about to
release a new 'incompatible' shell. Soon (translates 'eventually'), TSE
is going to be brought forth from the depths of hell. For days and
nights (havn't slept for 72 hrs exlcuding a few short naps) I have
slaved over MS-DOS edit turning a dream (nightmare?) into code (about
1.5kb of it). I have consulted the holy books (TI-83+ SDK, AsmGuru, and
How to program the Z80) and predicted what is to come. But enough of the
occult...
    Since the first TSE thread came to this list the project has
undergone 2 major redesigns and 1 total rewrite. I'd estimate it's now
about 75% complete. Since I first posted my fanciful ideas on shell
design I've cut back on the useless stuff (PDM, Libraries, Event
Handlers), made if feasable (Relocation Tables), and added some useful
features (Task preservation). I expect to release it in about a week
(hopefully) and more info will appear (as if by magic) then.
    But first, I need some help... I need the following four routines:-

_tseSmallSprite - A routine to draw variable height, 8 pixel wide
sprites. I could get this by modifying the movax routine, is the movax
routine the most optomised?

_tseLargeSprite - A routine to draw variable height and width sprites. I
could also modify the movax one but I don't know how optomised the
result would be. Has anyone already coded a suitable routine that I
could 'borrow'?

_tseRandom - Something like ionRandom. I could make this using rom calls
to the system floating point rountines and the put it in a register with
_convop1. However, that would be very slow and some games use this
routine a lot! Does anyone know how ION does this?

_tseFastcopy - This should be fairly easy to do. But ionFastcopy claims
it's 'FAST!' (presumably in comparison to the equivilent rom call), so I
was wondering if there was some kind of trick to it?

Any help with these would be greatly appreciated... Particulaly with the
Large Sprite and Random routines. If you can provide me with a routine
and I use it then of course I'll give you credit for it in the docs,
etc...

Thanks in advance
--Robin Kay--
References: