On Sat, 22 Sep 2001, Rhialto wrote:
> I remember reading that the tape reading routines are able to adjust to
> variations in tape speed quite a bit. So even though the various models
> use different timings, (in theory) they should all be able to read all
> those tapes.
No, the PAL C128 cannot read tapes written by the PAL VIC-20. But I've
found pulse widths that work on PAL C128 and PAL VIC-20 (whose frequencies
differ the most). I haven't tested those pulse widths with a 264 series
machine yet.
> I know from studying the tape reading routines that they are quite
> complicated, using various counters and timers and interrupt vectors. I
> think they are the most difficult part of the whole kernal to
> understand. I never completely succeeded anyway.
Me neither, and also the documentation in "Mapping the C64" is partly
unclear. Maybe the book author didn't completely understand it either,
and just relied on information he got from Commodore and other sources.
The receiving part is the most complicated, since everything happens in a
single interrupt handler routine.
BTW, I have now changed the software to use XON/XOFF handshaking. It
seems that there must be a bug in the RTS/CTS handshaking of the Linux
serial line driver (at least with kernel 2.4.2), since I can write() a
8-kilobyte block with no problems when XON/XOFF is used. With RTS/CTS,
it'd just ignore the handshaking during the write() operation. On Solaris
2.7 aka SunOS 5.7, there are some problems with the serial line interface,
and we're downloading a recommended patch cluster right now. The specific
patch (105924-10) is only available to contract customers. We're also
trying to compile the program for AIX 4.2 soon.
Marko
Message was sent through the cbm-hackers mailing list
Archive generated by hypermail 2.1.1.