Jump to content


+AtariAge Subscriber
  • Content Count

  • Joined

  • Last visited

Posts posted by dhe

  1. So... For those of us that are a little slow.


    1) Why would you want to piggy back a chip instead of just pull and replace? (Especially since it IS in a socket)?

    2) Is the socket itself unobtainable?

    • Like 1

  2. Gotcha, causes some of the stuff I seen on stackoverflow said, with pascal procedures and arrays - was to use pointers, as in free pascal with real pointers.

    I'd hoped I had missed a feature! 😃

  3. tp99 = No Pointers. =(


    I did do a bit of file i/o this week. I can tell you the one or two examples I found were really bad c := stream[80], f := string[1];


    Finally I got it rewritten with some sanity, using names like FILENAME[something], RECPOINT (for RECord POINTer)[something].


    Example programs are a really bad place to save a few bytes, by using one character variable names! 😃

  4. I think my ask was to broad.


    Custom sprites are fairly easy to make, there are some great tools for that. Wouldn't take more than five minutes to make one, thought you could spend all day if you wanted.


    I was thinking more something like a db - explosion or crash or whatever.

    And it would take you to some call sound routines with explosive sounds - once you have the data most sound effects are easy enough to port from extended basic to other languages.


    The other thing graphic effects - again, the example is explosions - I think Parsec does a good job of simulation an explosion when you ship gets hit.


    But yea, way to big of a task trying to create something like that.

    • Like 2

  5. I'm looking for some libraries!

    Is there a Database of sound effects and graphics published that people can just lift in to their own programs?


    The closest thing I can remember, is the Tiger Cub always had an example routine, and one who disk of routines of subroutines you could you to 'jazz up' your programs.

    • Like 3

  6. Yea, I thought about that route, but algorithmically, that makes for just about as much ugly for the shuffling routine, IF you want to keep the suits separate, if you just want a deck of cards shuffled, absolutely, that would be the way to go.

    • Like 1

  7. I've been able to spend some time with TP99v3.


    Some observations..


    I tried on Classic99 to use maxcpu. While TP99v3 handles this well, the developers must not have used the same line entry command they did in the editor (or that TI used for all the assembler cart stuff), because one quick push of the key in the linker, results in about four repeats.


    I have three sets of documentation I use, the original German manual, the manual the was distributed by LL Conner Ent (v2), and the translation of the German manual with v3 stuff added by Wolfgang. One of the items that isn't well addressed, I guess it's a given, what is the difference between functions and procedures? The bottom line is function returns values, procedures don't.


    The reason I went down that particular road, is I wanted to shuffle a deck of cards.

    Initially, I was going to create and array with 4 columns and 13 rows.

      Suits + Cards.


    Unfortunately, TP99v3 only has single dimensional arrays.


    Another way of passing arrays by name; say shuffle(hearts) or shuffle(diamonds) is with pointers. We don't have those either.


    So, I thought, I'd been a good weasel, and I would use a function. Not so fast, said the compile I can only return base types.

    Unfortunately, arrays aren't a base type.


    I got the job done, by cramming all four arrays in a procedure, buy it wasn't very elegant.


    I've grown soft having machines with 64Gigabyte or memory...


    I think I need to look at some type of chain loader, shuffle the deck and then discard the code to do it, and load the main program.


    When dealing with 32K, I've gained a new appreciation for the need to load character sets, create variables, change video and keyboard mode, and then discard all the code that did that, so you can't get on with the main event.





    • Like 4

  8. I like the idea of having a gram kracker-ish device. I'm also very interested in the language idea.

    What's the cosin(30) - then rather then running a bunch of instructions on the 9900 - little zippy cpu could have the answer back in no time.

    Not sure how much ram will be available, but a small ramdisk like provided in the minimem might also be achievable.

    • Like 1

  9. 1 hour ago, wierd_w said:

    because it can do the compression/decompression fast enough that the TI has no clue.)


    Ok, now your just messing with the old guy (the 9900) - what next? Hide his keys? 😃

  10. I have a document drafted by I guess Lou Phillip for IPC - Inter Process Communication - that seem to me, to be lost in the sands of time.

    The other term that came to head was create for JP Charlton - the MASTER DSR - getting that sorted of maybe just letting it die, might be the way to go.

       Paul said have Geneve OS doing all the dsr stuff was just a step until he could get all the cards dsr's updated to run with the the geneve.

  • Create New...