This is long, but I think it is necessary to paint a clear picture of the
nature of the problem I am having.
I have been operating SO2R casually for a few years. I have been using a
homebrew interface, 2xIC-765 (each with computer control), and NA running on
an old 486DX66. The computer has an I/O expansion card in it, giving it 4
COM PORTS, each with it's own IRQ, and 2 LPT ports, each with it's own
address. (The two LPT ports share IRQ 7, but this has never been a problem.)
The computer boots to DOS 6.2, and has Windows 3.11 for easy (read:
networked) shuffling around of log files, NA upgrades, etc.
Sitting next to it, I have a Gateway 2000 Pentium II/400MHz, with the
identical COM/LPT PORT setup. (Ask me how difficult it was to get a newer
computer, with ethernet card, USB bridge, soundcard, and 4 COM ports, and 2
LPT ports to *all have their own IRQs - yeesh!). I put the I/O expansion
setup in the newer computer at the same time "just because I could" and if I
ever decided to make it my main logging computer, it was ready,
hardware-wise. Besides, I learned a thing or two along the way. This
computer was mainly used for Internet browsing, e-mail, and other GUI
applications.
A month or two ago, I decided to make the Gateway the "radio" computer. It
was a seamless operation - I disconnected all the wires, exchanged computer
towers, and voila - it worked. (The Gateway machine runs Windows 98SE, so I
set it up to boot to "DOS" first, then to Windows, with the <win> command,
per a YCCC article I found on contesting.com.) All has been well.
However, a couple of days ago, I noticed that if I switch radios (with the
CTRL-R command), AND ONE RADIO IS IN SSB AND THE OTHER IS IN CW, the whole
enchilada freezes up. (I verified this with three .CRT files and three .QDF
files). I thought I fixed it, but it turns out I only managed to not
replicate the same set of circumstances. Today, trying to chase the few KY
QSO participants, I ran into it again, and figured out what the problem
really was.
Having had some experience troubleshooting, I systematically eliminated
variables by disconnecting potential conflicting devices and making "config"
changes. It got to the point that it would still freeze up if:
1. nothing was connected to any COM or LPT port;
2. the only hardware NA was configured to "look" for was the SO2R box.
To me, this meant it was between the computer and NA. I then went back one
version of NA at a time (back to 10.52), each in it's own directory and they
all did the same thing.
Then I pulled the I/O card, and re-configured LPT2 for SPP - Standard
Parallel Port support, also disabling the DMA accessibility. I thought that
the "expanded" LPT support might be confusing NA and/or the system.
It STILL does it.
I just checked it on the old 486 and everything works fine, SO2R in mixed
modes, just like I have operated before. I am going to swap the I/O cards
(though they are identical). Maybe one is damaged and I do not know it, yet.
I'll let everybody know if that helps.
If you made it all the way to the bottom of this message, congratulations,
and thank you for not just deleting it or ignoring it.
Any ideas? This one really baffles me . . .
Don
WX3M
|