Image Map Image Map
Page 2 of 2 FirstFirst 12
Results 11 to 16 of 16

Thread: OSI Challenger 1P loading problems

  1. #11
    Join Date
    Dec 2011
    Location
    Dallas, TX
    Posts
    147

    Default

    Also, you may want to check out the OSI forum. Nama (also on this forum) and Mark have made progress getting the Gotek emulator working with OSI, with HxC and FlashFloppy firmware and a data separator circuit. Mark is working on getting the WinOSI emulator to understand the HFE format images, which will provide a way to bring up and OSI disk system from scratch.

    With the 610 board, you should be able to have a floppy up and running.

  2. #12
    Join Date
    Jan 2011
    Location
    Vancouver, BC
    Posts
    3,378
    Blog Entries
    1

    Default

    Quote Originally Posted by dfnr2 View Post
    Microsoft BASIC has a 72-character input buffer. When that buffer overflows, BASIC tries to output a beep (BELL, ASCII 7). That character is the right half of a leftward-going submarine, character 7.

    Almost certainly, this is not a problem with your computer. If the FLAC file is synthetic, it's possible that it could be missing some characters. Try generating the files yourself.

    David Beazley has a python program to encode KCS here: http://www.dabeaz.com/py-kcs/index.html

    If you like FORTH, ED has a FORTH program to do it. Sadly, his page appears to have gone dark, but it is mirrored here: http://dxforth.mirrors.minimaltype.com

    Microsoft BASIC does a crude memory test on cold start, similar to what you describe. If it reports 7423 bytes free on startup, you have 8K of RAM present and probably good (but not thoroughly tested.)

    Cheers,

    Dave
    Thanks Dave,

    The system reports 31kish on startup of BASIC, and yeah, I think the RAM is pretty good. In fact, when I was reinstalling the first 8k I noticed it had suddenly dropped to 6139 bytes free or something similar. I found the memory chart from the OSI service manual and found the exact spot it was out.. 1 pin had bent in. So I think memory wise we are ok.

    I attempted to reincode a .bas file into .wav using KCS 0.8. First though, I went into the .BAS file. I noticed when I opened it in Notepad it was all one line, although in Notepad++ it showed individual lines for some reason. So I went through and added carriage returns. This eliminated the half sub problem and it was doing linefeeds, but at the wrong part of the line. Sort of 10 characters or so in from the line number. Not sure why that's happening. I encoded at 600baud (which my system can handle, it has a toggle for it). I also was being lazy last night and encoded in DOSBox rather than on an actual WinXP or lower machine (since KCS doesn't run on 64 bit OS). I don't know if that was a factor. I think it's mostly figuring out how to set up the .BAS text file so that KCS correctly places the linefeeds. I read on Tezza's site he had to adjust things with kcs with different flags to get it right.

    Anyway, I'm glad I know now that my 600 RAMs are ok.

  3. #13
    Join Date
    Dec 2011
    Location
    Dallas, TX
    Posts
    147

    Default

    That's interesting. I would expect the <CR><LF> to show up as in the file. One thing to consider is that the line feed causes the screen to scroll, which is basically a 2K memory copy, and since there are no interrupts, any characters that come in during the scroll are dropped. Increasing the speed to 600 baud make the problem worse. The fix is to add a bunch of NULLs in between the <CR><LF> and the start of the new line.

  4. #14
    Join Date
    Jan 2011
    Location
    Vancouver, BC
    Posts
    3,378
    Blog Entries
    1

    Default

    Quote Originally Posted by dfnr2 View Post
    That's interesting. I would expect the <CR><LF> to show up as in the file. One thing to consider is that the line feed causes the screen to scroll, which is basically a 2K memory copy, and since there are no interrupts, any characters that come in during the scroll are dropped. Increasing the speed to 600 baud make the problem worse. The fix is to add a bunch of NULLs in between the <CR><LF> and the start of the new line.
    You wouldn't happen to have any sample .BAS text files that show how you do that right? I'm having trouble picturing how you add the nulls to something like that.

  5. #15
    Join Date
    Dec 2011
    Location
    Dallas, TX
    Posts
    147

    Default

    Sure. Here's an Aardvark game, Seawolfe. It just happens to be the first sample I found with the NULL chars inserted. You'll have to view it with a hex editor. You'll see the three NULs before every line, and CR-LF after every line.
    Attached Files Attached Files

  6. #16
    Join Date
    Jan 2011
    Location
    Vancouver, BC
    Posts
    3,378
    Blog Entries
    1

    Default

    Thanks Dave!!

    That worked. I ran KCS on it with the -N5 flag and encoded at 300baud.. worked like a charm.

    But wow.. was that game slow.. haha The 600 does *not* move stuff around as fast as the Sol-20 does.

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •