>
Since short every time I use Arachne 1.61 on my Windows free Pentium II with
DrDos 7.03 I get "Error 6, Invalid Opcode."
This error is given by EMM386.exe.
>
That sounds like an invalid processor instruction, most likely caused by the
program getting lost, outside the program code area. I have experience with
assembler language programming for DOS, to a lesser extent for OS/2. C language
is notorious for null pointer assignments. I don't know whether to blame
Arachne, EMM386.EXE, other parts of DR-DOS 7.03, or DOS TCP/IP. I have had the
computer hang sometimes with UKA_PPP or DOS Lynx386 and come back in year 1994
or 2094 after rebooting, though this hasn't yet occurred with Arachne.
Another thing that can happen with a program getting lost is a BOUND violation,
activating int 5, which due to Intel and IBM getting their signals crossed, is
also the print-screen interrupt. This explains why I would sometimes get caught
in a print-screen loop, where the screen contents are printed repeatedly and
endlessly until I intervene, as by turning off the printer and rebooting.
>
As I recall, an opcode is a hex instruction code. I might be technically
wrong on this point because it has been more than 20 years since I worked
with opcodes. I used to have to load a series of "opcodes" into
micro-processors by using a hex keypad. This was my introduction to what
is now known as "computer programming". I didn't work for very long at
this method of computer programming because it became obsolete very fast.
>
to Sam Heywood: Was that raw machine language that you programmed? I heard
about the original computer programmers programming machine language bit by bit,
setting a lot of switches to the on or off position.
