PeteD

lol, no, good joke That's put a smile on my face that should last for at least an hour Pete

Looks scary

I didn't know there were any 48k ones Still might not be enough though Pete

800 is THE Atari to me. It's what my friend had when I had a C64 and I fell in love with the look of it. Only 8k though yeah? nooo way would anything I'm working on fit in that Pete

As I'm getting into coding the A8 I suppose some real hardware should be used at some point What is the best kit to look out for on ebay etc? 800xe/xl? 65xe? Once I've got something what are the options for connecting it to a PC for file download? Pete

What I did in one of my routines was: use only one character for background (like grass, dirt, stuff like that...), so background is same for all software sprites... and dont do shifting but lookup tables... self modifying code to set correct lookup table addresses depending on X and 4,and then simple: ldx sprite lda table,x to get correct shifted value... same thing with masks... it is 4 cycles per byte more but still better then shifting... Yeah but we were talking about exploding fist (if you read the whole thread), I know it was totally off topic in here but it started because of something José said, it's a totally different problem to handle. You can't build the background into the sprites like you could with say Gauntlet. The colour lookup is something I've used before too but can get fairly complex eg ldx sprite, lda table,x is ok for sprites that are 1 byte wide (of course you also need more than 1 lda because they've crossed the char barrier) but if they're two then you lose the bits that you're "shifting out" by reading them from a table so instead you have to say ok have 2 tables so you can read 2 bytes from the x (which becomes 3 because of course you're now into using 3 chars width) but wait, what if the sprite is wider (c64 sprite 3 bytes size) etc etc It's all swings and roundabouts and of course is very useful in some situations, not so in others. I'm quite happy to write 10 softsprite routines and do a benchmark test for them all, pick whichever is fastest in that game. Pete

José, I think we can all tell how enthusiastic you are about the A8 and you obviously really want to contribute. I don't know if this is generosity or if you just have to somehow get your name displayed in an A8 game. Either way I think by now you may have found out you're going about it the wrong way. The best thing to do would be to check the programming threads (either if they pop up here or in the actual programming section of the forum) and ask if you can help in some way. Tell people what your skills are, can you actually draw A8 graphics from scratch or only edit existing stuff? That way instead of doing something that nobody can use (and causing all this argument that's been going on) you'll have something to work on that is to the specifications of the people doing the coding. I've already said a few times I'd be happy to include your work in the next Fist "follow up" game I want to do. All it needs is a background drawn with G2F. If you have the skills, please use them As for learning to program, if you have a logical brain then it's farily easy. There aren't a lot of instructions, get data from memory, put data to memory, add/subtract (increment and decrement), some logical functions (AND, OR etc) and some logical tests (is this value 0, or equal to a value you test it with, or negative etc). As TMR said, there are a lot of people around to help, not just here but you have the whole internet to draw information from and all the programming tools you could need to make life easier than it used to be. Really, don't give up, just find out how you can contribute instead of trying to force your ideas on people. Even if this isn't your intention it is how your posts come across to most people. Pete

Thanks guys, of course it wasn't my intention for the avatar to be the C= eating the A8 logo. I'd thought of the idea to combine both logos last week (me being a C64 guy venturing into A8 land) but if it involves graphics manipulation my skills are that of a 5 year old I mentioned it to Ste'86 last night and 10 mins later there it was (thanks again Ste). Pete

Alternatively it could be seen as an Atari analy raping a C64 Pete

Yeah, got the 4 pixels per char seriously embedded in my brain now that's why only 4 sets of char definitions. Think of this scenario. The 2 guys are standing offset by 2 pixels and I only have 1 set of data so both of them have to not just have a load background, load data, load (or create) mask, mask, OR, store etc but also 4 shifts to get the bits offset to the right place. That's gonna be fairly nasty. So, keep the 4 copies of the background (320 bytes per char line) * height * 4. hardware scroll the screen up to the point where they're crossing (on a char height boundary of course) and change the "bank" of background definitions to the "offset by 2" . Now there's no shifting at all for the one guy. The other guy, in this case, because he's now offset by 2 pixels as well because of the hardware shift is suddenly also on a char boundary again. no shifting through data for him either APART from where his data crosses into the offset chars. That's of course best case scenario but at least you only ever HAVE to shift all the data for one guy. Pete

Cool. I have pretty much no idea how pokey works. SID, I know almost 100% (I wrote a SID emulator for GBA so had to dig deep into the hardware specs) but I'll really need someone to do music for me because even if I did know it 100% I still can't write music on a tracker or other editor. I've tried 100 times and I can't make it sound how I want so I surrender to it and give up. Pete

Yeah, apart from the weird speed and the going out of tune bits it's ok. The crickets are great Pete

José has not offered to help with the coding, in fact the only part of the original post that deals with the code, where it says about doing "something like Ripper demo" to produce more multicolour objects per scanline, is simply naive. And in the case of 8-bit development you're wrong, because the bulk of these games have one or perhaps two programmers being supported by (not dictated at by) graphics artists and musicians, check the credits on XXL's releases or Heaven's or Fandal's or mine for that matter. Indeed, especially in the "good old days" and even now Ste (STE'86, who did my new C64/A8 avatar btw) knew that he had to know EXACTLY what a coder needed before he started drawing stuff. This way he did things quickly and to spec and didn't waste his own time let alone anyone elses. The amount of times I (and Ste) have come up against artists who you could say "do all these frames in 16 colours" and they'd do them in 16 DIFFERENT colours per frame because they had no clue how the machine worked and what the limitations are. José does seem to have a good idea of what the A8 is capable of but when he goes off on the wrong track he seemingly won't listen to corrections, hence at least my frustration with the constant posting of the same pictures (from the C64 version) in the WotEF thread. Also, people who have produced nothing on any machine have no right to have a go at people like TMR who is extremely prolific, knowledgeable and has the "prods" to back up what he says. Pete

@Heaven, it uses 4 copies of the chars definitions in that area shifted 1 pixel at a time. So say the guy needs to be offset by 1 pixel, you'd smooth scroll the bottom area by that 1 pixel, move the bitmap area that they can go into to the set of data thats offset by a pixel, dump the bottom of the guy straight to the screen the softsprite the top of him but using those offset char definitions. Of course the top of the other guy has to then be shifted back the other way but so does the bottom of him and without using the smooth scroll you're likely to have to shift both of them (if it has to be done using the shifting data through chars method then thats a lot of cpu). Of course all the PMGs and stuff will need to move too but as most of the time it'll just be a simple x movement it wont take much time and any other PMGs I'm shifting around in odd positions, well they'll already have code to be moving them so changing the offset doesn't add anything. I'm not sure what saving it'll make because I've not yet decided on what approach to take with the software sprites. I can chop them into char blocks which is more work per sprite but means I could probably just have all the guys frames repeated 4 times, or keep them mostly whole, or one of another couple of ideas I've used in the past. It all depends on how much ram just 1 set of the guys frames uses and looking through the C64s ram while the game is running you can see all the sprite definitions and they take up a LOT of ram. Maybe I'd have room for 2 copies one normal and one shifted 2 pixels so you'd only ever need to shift 1 pixel. Now I've got the background stuff pretty much sorted I need to write a little converter that takes a raw dump (sounds painful) from the C64 ram so I can get something thats 2bpp and fix that up to A8 colours (you never know what bit patterns you're going to get from a C64 bitmap) and start putting the background together. Then I'll just take a couple of frames of the guys as full sized minimum rectangles just to make 100% sure they don't mess the background up in any way. Pete *edit* added a bit about shifting the PMGs in the background too just in case you've read and and thought AHA! the PMGs!!

Now you've done it, given away the secret A8 Super Fist screengrab!! lol

Then why don't you try coding an original game first rather than posting graphical mockups of games that exist on other systems and trying to 'port them'. You'll gain a lot more 'street cred' that way and people may take you more seriously. Tempest Or do as I suggested earlier. Instead of trying to come up with methods of doing graphics for a game nobody is working on, or doing graphics that the guy who wants to code it says won't work. Why not design some backgrounds for what I hope to do after Fist, the 4 player game. You yourself wanted to be able to have different backgrounds in fist, I said it's pretty much impossible but not if the players don't cross the background. You want something in an A8 game, try that. Impress us with what YOU can create, not taking something else and saying "if you changed this colour and moved that there it might work". I'd be happy to include your work in something I was working on if it's suitable. No point making up your own rules to work by. Pete

José, think what you want. I came to this forum to defend the A8 against people showing a crappy youtube video of a certain level of a game demo on the C64, I simply said it's not THAT impressive to be showing as the best the C64 could do and that the A8 could probably produce something similar. Then I noticed a discussion on Last Ninja and saw a lot of people posting screenshots of the c64 version and trying to make those work by running them through G2F then coming up with all crazy ideas on how to do masking. Of course some people had some great ideas on how to do things. I only gave my opinion in that thread that an editor would need to be made to design the levels and how to do the masking. I also expressed an interest in getting involved which I still am. I'm not planning on dying any time soon so have plenty of time to work on things. Yes, as I've said I'm pretty much a noob to A8 coding (I did some about 8 years ago, just messing around with it) but as I also said, with 25+ years experience on different platforms I think I've got the ability to work out how to use a few hardware sprites and a 2bpp char screen (it's all not much different to the C64). You kind of sound like you're telling me I don't know what I'm doing just because I haven't uploaded some screenshots of what I'm working on or something. If that's your attitude then you go ahead and keep uploading pictures from the C64 version or the spectrum or amstrad. Then I suggest you buy a book on 6502 and learn to code. If you want me to upload a picture of what I'm working on then just take a look at the complex fist background and think "it looks like that, only slightly different". I'm not an artist, all the work I've done is in data or algorithms or lists of lines/colours/bits/pmgs. I've also been working on software sprite routines and techniques to draw the 2 guys as fast as possible (eg using the hardware scroll and 4 different char sets for the lines of the background the guys cross so at least one guy doesn't need his data shifting through the chars to put him in the correct x position) but I'm sorry I'm not working fast enough for you. I'm done. Sick of arguing, sick of repeating myself.. Pete

Thanks, a little too "drummy" for my tastes but then I'm a SID snob hehe I like the originals and not remixes but I can certainly use it. How about the more mellow ingame tune? Pete

You're right I didn't know A8 but I've done enough work on different platforms over the last 25 years to be able to quickly pick up what you can do with some hardware sprites, some colour changes etc. I'm not saying don't try to help I'm saying as a non programmer you have to take the advice of a programmer because they're the ones doing the work in the end. Coming up with ideas that are impossible to code or won't work for some reason is just wasting your own time. I'd already stated I wanted to work on quite a few things before the new thread was started, if you remember it was me who requested that we close that Vs thread and start to open new topics Nobody is saying we don't want to hear ideas. I think we're hitting a language barrier here. Your English is 100 time better than my ability to speak any other language (I can barely say a few words in Welsh and I've lived here 8 years). Pete

José, I'm sorry to say that even though you do seem to be trying to help, at least in the Exploding Fist thread, you tend not to listen to what is being said and work from misconceptions, especially if you aren't the one programming something endless ideas won't help do anything but complicate what is being worked on. Maybe it would've been better to start a thread "Games I'd like to see on the A8" then if any coders were interested you could share your information with them and work TOGETHER with them to get something done. Pete

Fist music? where? gimme ! hehe Pete

There is a lot of Fisting going on in this thread.

Guys, unless you're doing it for fun it's not worth tying to work out how these screens can be done as I've already worked out a 95% copy of the c64 version. It'll have some changes from the C64, slightly different colours in some areas, small missing or redrawn bits to allow for widths of PMGs and it'll need some clever coding to stop bad clashes. José the area of possible clash is a lot bigger than 8 pixels wide, as I said run the game on an emulator, 2 player mode, walk one guy over the other. It's quite normal for this to happen when for example you try a flying kick and are too close. Also changing suit colour doesn't make it a "good looking one" IMHO. It's a good idea for what I want to do after Fist which is attempt a 4 player version where the players won't cross the background and will use the PMG behind playfield trick to produce their suit colours. It's probably better that you focus your efforts on designing some new backgrounds for that one then we can have an A8 "super fist". Pete

hmm, count them. 5 colours total. As far as the 4/5 colour clash problem I already said I could use lines of PMGs to fix any clash areas. Most of the time there may be like 1 pixel to fix. Whichever sprite has priority then he will get the most of his colour in the char. It may even be possible to flag which 8x8 areas of the guys have the most transparent area in them so that when drawing one over the other you can give the one with the most solid area of their colour the priority. Like I said there are many ways to fix these problems Pete

José, just get a c64 emu, fist, set it to 2 player mode and see where the guys can stand. They CAN stand right over the top of each other, which sucks for the clashing but as I said I'm happy to write code that uses some missiles or something to "mask" that area with the correct colours. When they cross in mid air they're usually moving quite fast and I don't think it will matter too much. If it does, I know some other tricks to help get around it, I sure don't want a game looking like a spectrum Pete