| |
Bitbreaker
Registered: Oct 2002 Posts: 510 |
new ifli editor
As the existing editors drive me nuts more and more (most of all Funpaint, due to corrupted files when packer is on, buggy color-clash handling, sucky load/save-routines, ...) i decided to do a new editor, lean but working.
yet implemented functions:
- working color-clash handling
- optimizing of block, if colors are wasted
- exchange color in blockline/block
- draw single color/pattern
- load/save (kernalroutines only, so that it will also work with all kind of fancy drive hardware)
- quick color selection by 0-f (col1) and shift+0-f (col2)
- copy/paste block
- blockwise movement by cursor, pixelwise movement by joystick
Is there anything else that would be reasonable to be implemented? Otherwise, i'd release that tool soon, including source.
Btw. a MCI-mode would be more or less easy to implement aswell. |
|
... 29 posts hidden. Click here to view all posts.... |
| |
chatGPZ
Registered: Dec 2001 Posts: 11523 |
http://www.canberra.edu.au/~scott/C=Hacking/C-Hacking10/C-Hacki..
easy to adapt for c64 (if i could do it, you should be able to do it too =D) |
| |
JackAsser
Registered: Jun 2002 Posts: 2038 |
Quote: somewhat ontopic: can someone put up on codebase the cia method of stable timing?
Well, the problem with sprites over FLI is trickier than a simple "CIA stable timing". The main problem here is that you need to hit $d011 directly after the sprite fetches. And as with all stable timing methods you can not compensate the variance where there are sprite fetches. So with FLI you have to stabalize BEFORE the sprite fetches (simple using CIA or what ever method u prefer), THEN you must manually NOP the appropriate amount of cycles to get past the sprite fetches (which are dynamic in timing) and then write to $d011. This is the exact problem i tackled in The Wild Bunch with the vector balls over 4x4-FLI. For the interested I could post the source or whatever. :D
The above of course only applies if you want full FLI-width. As I stated before; if you can accept somewhat smaller width of the FLI area you can simply ignore all timing issues.
|
| |
Oswald
Registered: Apr 2002 Posts: 5127 |
"This is the exact problem i tackled in The Wild Bunch with the vector balls over 4x4-FLI."
for that I sometimes look at you as a serious masochist :D thanx for the explanation now I fully understand it tho. btw another solution is software sprites ;) |
| |
MagerValp
Registered: Dec 2001 Posts: 1082 |
You can also make the sprite timing static by multiplexing or stretching a single sprite across the whole screen.
|
| |
saehn Account closed
Registered: Apr 2006 Posts: 44 |
NTSC/PAL modes or versions would be awesome :D |
| |
Bitbreaker
Registered: Oct 2002 Posts: 510 |
So i wrote a ifli to 4x4 decoding routine today, and if i move the bounding box blockwise within the pic, there is enough potential to cheat by moving a whole block and only decoding the resulting gap. So scrolling works rather smooth. I have to see as next what i can inherit from the other code (boundingbox handling) to save more space and then move display to 4th bank.
@saehn
hm, not sure if ntsc is my focus, but as i'll also include the source when i'm done, it should be easy for others to fix (i guess most of all waste another 2 cycles :-) ) |
| |
Bitbreaker
Registered: Oct 2002 Posts: 510 |
Okay, seems like Clarence was faster with his editor (guess that was the secret Oswald was talking about?) :-) However i have still continued work on my editor and implemented the 4x4 thingy now, and well, it looks pretty cool. So now it is possible to switch between 4x4 and 8x8 mode to edit. Just a few minor things have to be fixed now. Might work on a mouse support as next, as i have the chance to test it in the real hardware. |
| |
chatGPZ
Registered: Dec 2001 Posts: 11523 |
an option to enable/disable the d016 shift would be nice... and maybe also one for two colorrams. |
| |
Copyfault
Registered: Dec 2001 Posts: 487 |
Quote: an option to enable/disable the d016 shift would be nice... and maybe also one for two colorrams.
...plus an option to toggle MC/Hires (just one $D016-bit more to cope with;)). This would be close to the ILM-Editor... tell me if you come up with some reasonable put_pixel_routine for the mixed lines (e.g. Fr1=MC, Fr2=Hires).
Copyfault |
| |
Bitbreaker
Registered: Oct 2002 Posts: 510 |
Well, this evening i'll do a proper 1351 support first of all. The rest is kind of optional i guess. What i wanted to reduce is the pain in the ass when painting, not adding a bucket of fancy bling bling stuff :-) Changing the editor afterwards and doing derivates for all kind of other modes should be rather easy, so all that needs to be changed is the calculation of line/block and color-checks (either if they need to check 2 or 4 cols)
|
Previous - 1 | 2 | 3 | 4 - Next |