Adventures Of Lolo
Re: Adventures Of Lolo
If you're looking for more flash, the water animation appears to be quite long. I find water looks pretty good with three frames and you animate it like 0,1,2,1,repeat.
Re: Adventures Of Lolo
Yeah 5 tiles+2 shadow tiles on the water. Saving 192 bytes at 3+1 would be just fine like you say. I was wrong about the 7 frames on walking, I already had it at 5 frames I guess so need space somewhere else like the water. Global animation might also be a good candidate for ram tile color swaps too then you only need 1, I'll see how much many cycles are left worst case before I'd try that. Got the dying music and floor cleared music converted plus a rudimentary GUI similar in operation(wont be as "fancy") as sokoban world. Progress is slowly being made, as always hacks perpetually being redone .
Re: Adventures Of Lolo
More rhetorical comments.
*Water is updated now 3 frames+1 shadow, animated 0,1,2,1, as Paul suggested looks good. Saved 256 bytes.
*Title Screen drastically simplified+other cuts, much more plain but will have ram tile effects or something to make it interesting. SAVED 3.5K, the silver bullet. Otherwise project was doomed. Necessitated big redesign.
*THE FINAL redesign #3. AKA Ridiculous Frustration! Collision much slower/complicated but uses no redundant tiles instead of sprites, had to do it or give up on bootloader (ie. screw project). Practically written this game 3 times now. SDL version scrapped! The space is available now and whatever code works is final. Then write a cocky postmortem saying it was all too easy Should have done bomberman....
*Water is updated now 3 frames+1 shadow, animated 0,1,2,1, as Paul suggested looks good. Saved 256 bytes.
*Title Screen drastically simplified+other cuts, much more plain but will have ram tile effects or something to make it interesting. SAVED 3.5K, the silver bullet. Otherwise project was doomed. Necessitated big redesign.
*THE FINAL redesign #3. AKA Ridiculous Frustration! Collision much slower/complicated but uses no redundant tiles instead of sprites, had to do it or give up on bootloader (ie. screw project). Practically written this game 3 times now. SDL version scrapped! The space is available now and whatever code works is final. Then write a cocky postmortem saying it was all too easy Should have done bomberman....
Re: Adventures Of Lolo
Good to hear, Lee. Lolo sounds like it's back on track.
I think one of the biggest savings you can make is to build your title screen out of tiles you used to make your levels. It sounds like you've done something similar to claw back 3.5k.
I think one of the biggest savings you can make is to build your title screen out of tiles you used to make your levels. It sounds like you've done something similar to claw back 3.5k.
Re: Adventures Of Lolo
[rant]
Exactly the case here, game tiles make the title fit in with everything pretty well too. Still some title specific tiles, but they are so simplistic I am confident I could actually have none and replicate them with ram tiles. The ability to fill and draw colored lines and curves into ram tiles is a good reusable way to make some pretty respectable graphics. As example draw up a simple title like normal, make a map of it. Then modify the ram tiles to look like the tiles you used(slow,so WaitVsync(1)), then set that map to vram(not +RAM_TILES_COUNT). Tedious but as long as it doesn't cost 64 bytes of function calls per tile it would be worth it, I'll test.
Also thinking about that for text in a game that doesn't really have text in gameplay. Why waste (26*64)=1.7k on alpha characters, instead just do it like a digital clock. Store an 8 and have a function to modify that 8 in ram tiles as needed per different character. Sounds sloppy especially trying to change things later, assuming the code would even be smaller than the tiles! It would hopefully be small but definitely slow so only when you have time to spare like GUI. Something like printramtiles(...) would handle all this stuff internally, loading up characters as needed so you could have a screen full of text just as easily as 1 string. I'm not totally there yet but slowly I'm putting together a little library that should facilitate such things that at least I have faced in a design. Everything would need a really transparent interface to be useful to others. Just be a magic black box the developer doesn't have to understand to utilize, let's you know what ram tiles are still available, etc. It could still be tedious to design with, perhaps what I should really do is put everything together into a GUI library with menu effects,text,functions hooked to menu choices. Sometimes I think I just like to complicate things until I run out of time, or I think I'm writing a PC game so who knows . Fun to think about if nothing else.
[/rant]
Exactly the case here, game tiles make the title fit in with everything pretty well too. Still some title specific tiles, but they are so simplistic I am confident I could actually have none and replicate them with ram tiles. The ability to fill and draw colored lines and curves into ram tiles is a good reusable way to make some pretty respectable graphics. As example draw up a simple title like normal, make a map of it. Then modify the ram tiles to look like the tiles you used(slow,so WaitVsync(1)), then set that map to vram(not +RAM_TILES_COUNT). Tedious but as long as it doesn't cost 64 bytes of function calls per tile it would be worth it, I'll test.
Also thinking about that for text in a game that doesn't really have text in gameplay. Why waste (26*64)=1.7k on alpha characters, instead just do it like a digital clock. Store an 8 and have a function to modify that 8 in ram tiles as needed per different character. Sounds sloppy especially trying to change things later, assuming the code would even be smaller than the tiles! It would hopefully be small but definitely slow so only when you have time to spare like GUI. Something like printramtiles(...) would handle all this stuff internally, loading up characters as needed so you could have a screen full of text just as easily as 1 string. I'm not totally there yet but slowly I'm putting together a little library that should facilitate such things that at least I have faced in a design. Everything would need a really transparent interface to be useful to others. Just be a magic black box the developer doesn't have to understand to utilize, let's you know what ram tiles are still available, etc. It could still be tedious to design with, perhaps what I should really do is put everything together into a GUI library with menu effects,text,functions hooked to menu choices. Sometimes I think I just like to complicate things until I run out of time, or I think I'm writing a PC game so who knows . Fun to think about if nothing else.
[/rant]
Re: Adventures Of Lolo
Hehehe, 7-segments display, that's actually a clever packing idea! 1 byte per character instead of 64 (1 bit per "segment")! Now if just had more RAM...Why waste (26*64)=1.7k on alpha characters, instead just do it like a digital clock. Store an 8 and have a function to modify that 8 in ram tiles as needed per different character. Sounds sloppy especially trying to change things later, assuming the code would even be smaller than the tiles! It would hopefully be small but definitely slow so only when you have time to spare like GUI.
-Uze
Re: Adventures Of Lolo
Hey guys, anyone that is interested I put together a few notes on ram tiles effects http://uzebox.org/wiki/index.php?title= ... cts_Primer It's quite basic functionality at this point but their might be some information that would be useful. Sokoworld and Lolo used for example effects with generic code. I also put up a small GUI intro aimed at some of the newer guys, all my tutorial type stuff will be at http://uzebox.org/wiki/index.php?title=Weber%27s_Rants. Hopefully a sound effects tutorial soon, easy to do from NSF perhaps a tool is in order. Also here's a newer http://uzebox.org/wiki/images/a/a9/Oldlolo.hex lolo hex, keeping in mind this was after a large collision overhaul. Medusa killing lolo crashes! Eggs are messed up.Some basic GUI stuff added, few sound effects, etc. Press start for in game menu. Can't understand why this game has become so difficult!
Re: Adventures Of Lolo
There's a deal of work there, Lee. I'll probably give some ram tile effects a go for my next game.
Re: Adventures Of Lolo
Good to hear, please document anything you come up with! It would be nice to have an organized collection of code people could just plug into their apps, without too much trouble.
Also, I finally got around to figuring out how to do sound effects(how are you guys doing them?) and I'm really quite happy with the results for Lolo(was easy even). I made a rough draft for my approach on this, might even be useful to some experienced Uzers. Mainly because I really think we are light on documentation in the community. We're probably losing a lot of potential from the new members, if they could only just get a running start on everything.
Also, I finally got around to figuring out how to do sound effects(how are you guys doing them?) and I'm really quite happy with the results for Lolo(was easy even). I made a rough draft for my approach on this, might even be useful to some experienced Uzers. Mainly because I really think we are light on documentation in the community. We're probably losing a lot of potential from the new members, if they could only just get a running start on everything.
Re: Adventures Of Lolo
Wow, this is reeealy nice! I had already write up on how to make sound effects on the wiki, but your info if more detailed with pictures and everything. Thank a lot for taking time to write this, I'm sure it will be *very* helpful to many. Btw, could Anvil be actually usable to create music? Does it export to Midi format 0 and converted with my Java tool?
-Uze
-Uze