PeteD

ah right apologies. I should read the thread through! yes that wouldn't work unless you want the buddah as a super low res block of pixels. I've been looking at the fist screens in between working on other stuff. A couple of the screen shots from the original and sequels barely need any underlays. It'd take me a good while to work out a usable solution with the colour usage. The next stage would be the # of chars actually available to the backgrounds after # required for the soft sprites. hehe not to worry Yeah, the screen José has been looking it is one of the simpler ones and with a bit of juggling with the method he's come up with it should be fairly easy. But I'm a sucker for giving myself hard things to do (like the other 2 complex backgrounds), Ste (86) keeps telling me to just do a version where the sprites don't cross the background but nooooo that's not in my nature I said once this is sorted (if ever) I'll use that method to do a 4 player version (somehow). Doing stuff like this is the only thing that gets me excited about coding these days. I've been working coding (and coding at home before that) since the mid 80s and am kind of bored with the "every day" aspects of it now. Pete

Yup, understood that, I mentioned it further back in the thread as the 1st way I'd come up with to do the red and white suits (hence me making sure they really were as wide as I'd come to believe when at quad size), what I'm saying is José has said to use them for the Buddha, just 2 players and 2 missiles, quadded up to get the 2 shades of grey. Surely then you're limited in resolution to 4 pixel wide? Ugly, not gonna happen Pete

Ahh now I'd understood your PM thing from before I'll now ask surely then Mr Buddha will look ugly with his quad sized graphics? Pete

I'd looked at Barbarian yesterday as another one to do once the code for this was done. despite looking kind of the same the Barbarian stuff is a lot simpler to do. Well in terms of working it all out in the first place anyway. Software sprites in themselves are easy. Any coder can do them, of course different techniques will lead to different numbers on screen so you tend to tailor code to cpu free vs ram available etc. The problem with Fist isn't the software sprite stuff its the fact that to make the backgrounds look like the C64 verison they'd have to be so complex with mutliple usage of pmgs, colour changes, midline changes etc that if you're not careful your sprites will destroy the background when they go over it. I think I counted 11 extra colours + the 5 for the guys in any 1 line on Fist. Barbarian has maybe 6-7 in total. I think maybe Fist was a bad choice to start as a 1st A8 project I'm sure we all know all this anyway it's just I don't want to get stuck comparing Fist to Barbarian when it doesn't have the same issues to resolve. Pete

Up Buda basement line(PF1-Pink) change to Green (on the tree) (Players never get to this area) Yes they do Try a somersault, or even just a flying kick. I'll try to put some code together tomorrow just to display 2 guys (no animation needed or anything) moving up/down on the screen to the height they do in the game, and a display list so I can try to incorporate some ideas from this discussion. Pete

I understand your ideas, mostly, (see my question above for one query). Yes, 2 players x quad size should be enough for the 2 guys. It's ok to want to end the ping-pong but like I say, it's better to work on the most complex screen first. There are a lot more changes needed per line to get that one to work. This one is easy in comparison. Pete

PM3 - Quadruple size - Light Grey - for Buda Will PM3 be wide enough? 2 pixels x4 = 8 hmm I think Mr Buddha is fatter than that

To be honest I'd barely looked at that screen as it didn't seem that much of a problem. You've got to work on the most complex one first then the others should be a lot easier to do afterwards. I'm talking about the 1st one you posted at the top of the thread. But please try that 1st one, if you can do that and allow free movement of the players over it, without any clash then I'll start coding the game tomorrow Pete

http://www.mini-itx.com/projects/atari800/ Is that just blasphemous to post on this forum? "Desecration of A8 prompts angry mob to attack"

Yeah I was deciding between the 4 and 5 colour modes but was worried about if you start messing with the 5th colour (especially for the sprites) then you have to start swapping chars and maybe altering their colour when things start crossing the background. It's yet another thing to think about. There is also a lot more crossing of the players than you think. Also if you use all the colours just for the 2 players you might have all the PMGs free for the background but then you have to use the 5 colours from the players in that background where they cross. That may or may not be a bad thing, I haven't thought about it yet Pete

Cool, thanks I was confused about 320 wide screen and presumed you had sprites that matched that res. I did wonder why I read that you can fill the screen width with quad expanded pmgs. So that's ok using 2 would cover the width of each Fist guy (them being 64 hires pixels, 32 multicolour). I'm just too used to counting pixels as pixels and not colour clocks. Pete

lol This is why I want to attempt to write an intelligent converter that at least will try to do the grunt work for you but the more I look at Fist screenshots the more my brain hurts with possible combinations. There are just so many things you can do on A8 to attempt to get more colours that it just makes it even harder than C64 because it's impossible to decide where to start. Some questions about A8 hardware though.. The Players and Missiles, players are 8 "bits" wide, but how many pixels? I know you can expand them to 4x but I don't know if each bit represents 1 hires pixel or 2. The Fist sprites are a maximum of 65 pixels That's the guy when he's lying down ( I think that's the widest one) I think losing a bit of his middle or leg wouldn't kill him so we'll call him 64 pixels wide How many players to cover him? The "software screen modes". Are they actually useful? Would using one of them that gives 160x192 be a reasonable thing to do or do they just flicker too much? I image especially on PAL machines the interlace would be too noticeable. Pete

hehe You've got most of the points from my last post before I even posted it What I thought of for Fist was to make an arbitrary colour the background colour (doesn't really matter to my brain to begin with) Yellow to be the next colour, then black then pink. Then use the background (transparent) colour and put 2 expanded pmgs "beneath" the sprites, one white the other red. So we've used 3 of the 5 colours, gained 2 more from 2 pmgs. Now it gets interesting. The minimum extra colours needed in the area the players cross are (and this is ignoring the colours in the 2 pmgs because you cant be sure they're going to be in the correct place), dark green, light green, dark brown, medium brown, white, dark gray, medium gray, light gray, red, dark blue, light blue. 11 more colours per scanline. That's where my brain begins to explode So maybe its not the best way to do it using 4 colours and 2 pmgs for the players? maybe some different combination would be better? but this is what needs working out before anything else is done or you'll end up with code that suddenly becomes useless. Pete

Indeed, and I'll say it again But I still think allowing people to do their own G2F screens is asking for trouble As for software sprites, hmm, I haven't really read/understood all of what you were getting at in that thread but you seem to be talking more about masking with the background and not taking into account that to make a screen as complex as Fist (or LN for that matter) that you're going to have to have so many chars, pmgs, colour changes etc per line that it'll be impossible to tell what is going to be on the screen when you start to draw your player. The problem with a generic software sprite solution on A8 is that it just won't cover half the circumstances you'll come up against. This is probably why most A8 games don't look as good as the C64, because it's so much more work just to do what we take for granted on C64. It isn't that it can't do it (at least in some cases) it's just that it isn't worth the time it would take for a programmer getting paid the same (probably less) than the c64 coder. I know how I'll do Fist if I take it on and it really involves a lot more work on making the background look right AND allowing the players to cross the background than any of the other code (software sprites, collisions, logic etc are all things I've done a lot of over the years). Pete

The only problem with allowing people to do their own screens in G2f is it's already so complex to do and then they would have to conform to strict rules on the number of colours/pmgs/splits/tiles etc etc that it becomes almost impossible to do. There is of course no reason why the sprites have to cross the "backgrounds" ala IK+ which makes the G2F idea a whole lot more feasible. The main reason I'm interested in doing Fist is the technical aspect of handling players which already take up 5 colours (background, black, red, white, pink) crossing a bitmap background which has even more colours. The same problem is increased by having 3 player mode (Fist+). Background, black, pink, red, white, green. 6 colours just in the play area. I think there should be an attempt at recreating Fist (like the IK+ recreation) just "beacuse". Like asking people, "why do you climb mountains?" Then using the same code base create Fist-A8 or Fist-2009 or Way of the Popping Foot Something like 3 players, allow people to import their own backgrounds but the players can't cross the background. It may even be possible to have a 4 player version. Now that would kick ass (is there an ass kick move in Fist?) Pete

Understood, G2F doesn't add anything to the hardware, but the same can be said for the "new" c64 modes which are a PITA to use unless you use the appropriate tools. Doesn't change the fact that you can't ask which is better, c64 or atari then say atari IF someone redid it. Kind of defeats the purpose. I'm looking forward to messing around with all the possibilities of the A8 hardware and seeing what I can do with it and then compare it with what I can do on C64. I'm hoping for some satisfying weeks ahead

@José The problem with some comparisons being made is you're saying, hey, with g2f it could look nicer. But we're talking at the time they were released (at least I am) not what would be possible if they were remade today because C64 coding has generally come a fair way in the meantime too. Pete

I think that's the idea, rather than everyone just list their favourite A8 games, maybe yea/nay the ones already listed. eg Gauntlet? not sure I agree with that one. They're pretty similar. C64 has more colours, runs faster (unless what I've seen of it has filtered its way out of my busy head), music is about the same, title screen c64 has more colours except the logo.. Pete

This thread can co-exist with other threads. It would make more sense to kill the first post so the topic "Atari v Commodore" would be pertinent to most posts here. True, but as I found out when I joined it's impossible to read and if you make a comment you're likely to get someone saying, "damn you! I already mentioned that 178 pages back!" or similar Just thinking it may be easier for people interested in the discussions about porting games (eg Exploding Fist), or the info on A8 for it to be started again in separate threads. Of course, keep this dinosaur going if you want to. Pete

Not checked the docs yet, I'll have a look tomorrow. If it does everything I think I need to do stuff the way I want then all the better if not I'll start up some code of my own. Hopefully it'll be good enough to save me some time though. Pete

Rescue on Fractalus, Koronis Rift, Eidolon. Elektra Glide

That's far too tidy for a "real" desk

I'm just trying to use my limited brain power to work out where you could change colours, reuse them, reuse pmgs horizontally, change data in them, colours etc etc. It's pretty much impossible to do for a human so I'm sketching out some algorithms to do it automatically. It should end up as a g2f kind of tool but a more complex, but in a way easier Like g2f it'll let you edit stuff, move bits of the picture around to limit "clashes" etc I need to work it all out from the info on available cycles per line so I know how many LDA STA combinations you can have for whatever purpose eg changing colours either of characters or pmgs, repeating pmgs, putting new data into pmgs. The list goes on and on The code can then work out colours per line, give priority to ones you tell it to (like the 2 guys in the case of fist as they have to cross from the simple yellow floor over the complex background). Hopefully if I can get it to work you'll be able to import a picture, tell it you want all the players available but it can use the missiles, you want x amount of cpu (if possible) available per line (for doing "other stuff") and let it get on with it. Or alternatively use it to draw your levels/sprites from scratch. It's going to be a shame though that Fist still wont look as pretty as IK+, but they don't have to worry about the guys going over the background. That gives you a lot more freedom. Pete *EDIT* Is what I'm contemplating above what g2f actually does? I had a quick play with it but don't have much clue what was going on lol I expected to load a gif of a c64 screen and have it give me a close approximation of the source picture. Instead i got a 4 colour nasty thing I've never been one for persisting with apps I don't understand straight away though lol

Yeah, where do you buy girls who'll play with your joystick?

Just saw it on Digg, was sure someone would post it