Bulletins from the Pacific Packet Radio Society - page 019

(3) A socket for burning EPROMs (with separate 25 volt power supply (3 9volt batteries and regulator) ), or EEPROMS as the price drops. This would allow downloading new versions into EPROM over the net.

(4) Option for battery backup RAM. This would allow downloading the controller software into RAM and having it keep.

(5) Options for multiple terminal devices attached to the board..eg - 1 serial port, 1 parallel port (AT THE SAME TIME!!!) - 2 serial ports This is pretty easy to support in software (as two HDLC addresses, for example) but can't be done with the current board. I want to connect my Apple as a host and my terminal as a terminal, to the same board. It'd also be possible to run a printer off one and a terminal off the other -- getting listings in the "spare" time when the terminal wasn't generating packets.

( 6) Bring all the CPU pins out to an edge connector, so we don't have to resort to awful kludges to extend the board when we want to do further experimenting or production use. Like, ever try adding memory or 1/0 to a VADCG TNC? Best idea I've come up with is piggybacking into the CPU socket ... eech.

(7) A set of switches (DIP) or a hex input dial which is readable by the software. (The DIP switch on the VADCG TNC is nice, but unfortunately useless -- a jumper would've been better -- how often do you change your modem baud rate???)

(8) Possibly use a more popular processor. Z80's are easy to find; 8085's are hard to find (even in Silicon Valley, they don't carry them at Radio Stack). Even better would be a newer generation processor like the 6809, which would also be compatible with the Swedish FORTH efforts.

(9) Full kit (w/parts) and/or A&T and/or A&T & packaged with modem and radio would be nice. I'm at the moment stuck off the air because I can't debug my board because finding loaner scopes is not easy, and I'm not a hardware hacker.

(10) A real timer chip should be included, or at least a socket for one. We'll need it for accurate timekeeping for time-division muxing schemes, which will soon be required with multi repeaters on one frequency at 1200 baud. Else congestion from hidden terminals (which rptr can hear but you can't, thus you transmit and collide) becomes a real mess.

(11) Parallel ports should be connected to ADIP header, rather than to a DB25. You can connect a DIP header to anything pretty easily, while DB25's are expensive and painful to use when there's no obvious reason to do so (like, when you're connecting to an async serial device).

Click for Original Graphic Image of this page.