bpj1138

At the risk of telling you things you already know, it seems to me like the pleasure we get from programming isn't really the end product of our work (games, pretty pictures, etc) but the work itself. The mainstream protrays us as people addicted to games, getting perverted pleasure out of killing virtual monsters or whatnot, meanwhile, we're solving complex mathmatical equations, making impossible optimizations and so forth. What's even more extraordinary is that we're just as happy when other people accomplish these things. (the response to this thread proves that).

I'm just wondering if anybody did this? Put a PC into an XE case. I still think the XE was the best looking computer ever made.

Hrm. On a 8 bit computer like the Atari, the assembler assembles to a specific address, unless you try to assemble position independent code. So, whatever address your library is assembled at, that's the address you load to. You can use the C stdio functions to load binary files.. ie your code.. Then to call your library functions, you'll have to use whatever calling convention you use in your library to push whatever parameters are required and then jsr to your function, in CC65 you can use inline assembler, like: __asm__("jsr $e6c4"); so if E6C4 was the hex address of your library function it would call it.. --Bart

CC65 is the way to go. You pretty much have to mix C and assembly though, unless you don't care about speed, but that's the way any good program is written anyway, even today!

I don't understand, I thought I was starting a new club, but nobody showed up. Oh well, happy new year.

You should have seen the Donkey Kong thread a while back. Turns out a lot of the commerical games were cross assembled on mainframes. This stuff I wrote was really a calling card for you guys. I'm trying to get some people together to work on PC games. C++/OpenGL. If anybody's interested, leave a message after the beep.. --Bart

I remember reading "Assembly Language Programming for the Atari Computers" back in junior high school (I should have gotten an award for that, but nobody had any idea of what I was doing). I didn't have an assembler, so I would assemble my code on paper, then encode it into BASIC strings.. you know the drill.. Needless to say, I never got very far coding this way, but the experience taught me a lot and stayed with me. My only regret was that I never met people like me, and that kept me back all these years. Now I can't get people to help me if their life depended on it.

cyco130: Don't mind you using this thread one bit (or 8 bits even). I'm glad people are interested in this at all. If you want to use any of my code, you're welcome to, or not... My routines are optimized for mode E. This is mainly so that coordinates can be kept in one byte per component. I use a lot of zero page space too. For example, my rectangle filler uses 16 zp pointers, so it fills 16 scanlines per interation. My text draw uses the same scheme.. fills the entire character at once. I also keep a complete list of scanline addresses in higher memory. The polygon filler is pretty cool too.. it uses the line algorithm to fill scanline spans for the polygon. This works with any number of edges. This also takes up some higher mem.. like 192x2 bytes.. I gotta say programming the 6502 is a lot of fun. More fun than playing video games. --Bart

I posted this some time ago in the regular A8 forum. This is my graphics library in C/assembler. Here's the features thus far: * ANTIC mode E init code (single or double buffer) * front/back buffer flipping * horizontal/vertical line draw * symmetric bresenham's line draw * rectangle fill * polygon fill * text draw If anybody wants to add to any of this, you're welcome to it. Fake_Rotation.zip

I did open it just to see what was in there. Had hell of a time removing the shielding, but no worries, did a good job, and snaped it back in place with no problems. There was no rust or nothing like that. I think I picked it up right after it was tossed away. AX, since you got references, I guess you're the winner. Send me a message with your address and we'll try to figure out how much it will cost to ship.

I found one the other day by the side of the road. It was raining slightly, so I don't know if this thing works or not. It's just the console itself, no power supply or cables. If anybody wants this thing, let me know. You'll have to come by and pick it up. I live in Jacksonville, FL.

Plogue's already working on it (well not just POKEY), in the form of Plogue chipsounds, and nearing completion, at least it had better be because they announced it at NAMM in January this year.. http://ploguechipsounds.blogspot.com/ Though he is taking his time about it, so it'll probably be quicker to write your own Well, I don't know that much about this, but from what I understand, you just need to call "next_clock_cycle()" on the POKEY lib within the emulator, you don't need the 6502 or memory at all. In fact even the clock speed is irrelevant. The emulator eventually maps its memory to POKEY lib function like "pokey_state_changed(x,y,z,..)" which probably passes all the register values there. Incidentally, you could probably change the POKEY state every cycle, which would be impossible on a real machine where it would take a few cycles to set a new state. That's my story and I'm stickin' to it!

I thought about such a project. I was thinking of making a VSTi out of the POKEY.

Hrm, I'm surprised there is more interest in the "sequencer". No, never owned a synth in my life, except a CZ101, which I also sold for cash years ago along with my Mega 2 ST. In PG2, I did some silly stuff like when you edit a bar, and then hit '*', it will double the pattern lenth and copy the first half to second half. That's exactly 1 line of code in the program. Silly stuff like that, but pretty essential. Also, by default the sequencer is writing to the pattern, but if you hit TAB, it won't write to the pattern, BUT you can still press keys which will be heard over the pattern, kinda like "KEYBOARD MASK" or "KEYBOARD OVERRIDE". All of this is geared towards a live performance. BTW, if you stop the sequencer with 'RET', and you're in play mode, it will behave like a normal keyboard.. the notes won't be quantized.. --Bart

That's why I include the sources. If you want something different, you can make the changes yourself. I use this thing in real time, or to record a sample I can import to a digital sequencer. Doing everthing on the Atari doesn't always make sense. Look for the next version to be more synth less sequencer related.. it will feature graphics, and deal with functions more than notes. I envision a sort of loop generator, with a bunch of function displays, etc.. --Bart PS: F: thnx..

Buenos, Here's the new version of the synth/sequencer program. I made some changes (the type of stuff I think of in a drunken stupor). Noise! What a concept!.. Of course to the POKEY chip that means frequency noise. This thing can produce some bass now.. which reminds me of the song "we discovered a machine now we teach it the bass".. how true in this case.. The source code is included... I don't care what you do with it.. Put your name on it and sell it if you want. Pokey_Glider_2_0.zip --Bart

Andreas, Unfortunately, it's the same version. I suppose it would be really easy to write save routines. If anybody's interested, I can upload the sources. It's written entirely in C. I myself use it in an emulator, so I can always just save the machine state. I was also thinking of recording realtime user key input along with the pattern, because that is how I use it. I first write a short pattern, like 16 or 32 notes. Then I let it play use the octave keys like notes, or just flip some bits (mainly the filter) in real time. Lastly, I thought it might also be useful to allow "screen switching", so you could write a few patterns, and then switch between them in real time, again, the timing could be recorded. That way, you could switch in the middle of a pattern, or even a note. So, as you can see, I'm trying to create a unique realtime program, that is simple to use, and doesn't box the user into fixed patterns, and complicated syntax, as most trackers do (in my opinion).. --Bart

I think the POKEY was/is a very unique chip, with its "divide by N" and "bit masking" technology. By comparison, the SID is very conventional. I'm surprised the POKEY wasn't picked up by techno artists back in the day like the TB303. People that bash such instruments are simply grumpy snobs, much like old people that hate anything electronic. 'nuff said

Buenos, I noticed some posts on the subject of sound, so I thought I'd post this up here. It's a very simple synth/sequencer. That being said, it does allow you to access most of the POKEY chip bits, with one exception, namely it uses the 2 channel mode ("tied" 16 bit mode) as opposed to 4 channel mode. That's it! Hope you enjoy it! A8 --Bart pg.zip

I always liked the N64. I thought it was a neat machine, and I guess the MIPS processor is pretty cool, none of that dumbass reverse byte ordering, flat memory model, etc. Then you got 3D hardware, and so forth. --Bart

I agree with your sentiment, Bob.

If you're reading this, you're either looking for a job (good luck with that) or one of the thousands of pessimists that will undoubtedly make their negative opinions known. Regardless, I'm going to give this my best shot. I'm an Atari kid, both the 8-bit and the ST, though like many of you I've graduated or been forced to move on to the PC/Unix. I've always wanted to write games, and have been working towards that goal for many years. I'm sure you know the story by now. I've had some minor professional success in the computer world, working for some web companies and such, all the while looking for people that I respected and wanted to work with. Well, honestly, I've never found those people, neither among my coworkers or personal friends. So what was left? I had to go on the internet to look for them. Of course I soon found out that people on the Internet didn't trust one another. Nobody was willing to take a chance and work with someone they can't talk to in person. There was a lot of talent, but no trust. So I figured out that there had to be some common link/trust between people on the Internet otherwise nothing would jell. Then it hit me.. There is trust among people who have had that common 8-bit computer experience. If somebody tells me "I can write 6502 code", they're automaticly accepted (at least by me). It's a bit like the computer/game industry itself. They only hire within the family too. Ok, I'm gonna leave it there. What is the project? I don't know. This isn't about me, and I'm not gonna set those parameters for everybody. I'm just starting the conversation. Give us/me your ideas. One thing I would say from the outset is that even though we love our Ataris so, we have to be realistic, and use a platform that's a bit more mainstream. That's all for now. Let me/us know your thoughts..

Thank you all for the kind words. You're welcome to use the sources in any way possible. I'm not great at naming things, but what I was getting at was that there were some 512 "vertexes" being rotated, but actually it's only the shape verticies that are rotated (by the use of tables), the interior points are "implicitly" calculated from those verticies by using a static fractal trace (Sierpinski et al). LOL.. --Bart

Decided to brush up on my 6502 assembly language. It's been a while.. http://www.geocities.com/bpj1138/Fake_Rotation.zip --Bart

Thanks for the info on the monitor. Yup, I'm using the non-curses version (hopefully not for long). Lastly, I think I found a bug (this may be old too), but the PAL/NTSC flag doesn't seem to be set correctly.. Cheerio.. --Bart