wb6rqn@yojna1.UUCP (Brian Lloyd) (11/19/85)
Many people are now interested in packet radio and are trying to decide which TNC to buy. Well there are many options and your choice must be made based on the hardware and software features you desire. There are two main TNC designs both from TAPR. They are the TNC1 and the TNC2. The TNC1 is currently available from AEA as the PKT-1, and from Heath as the HD-4040. There are other sources for boards and parts if you are motivated to "roll you own". The TNC2, also from TAPR, is available from AEA as the PK-80, and from GLB in kit form as the PK-2A. Your inteneded use will will determine the one you should buy. The TNC2's seem to be best suited to the job of interfacing to a dumb terminal. The TNC1 standard firmware is OK in this respect, but I feel that the latest release of TNC2 firmware is better than the standard TNC1 firmware. There is a firmware release from Ron Raikes, WA8DED, that makes the TNC1 my current choice. The DED firmware (as it is called) supports multiple connections simultaneously, and has a special host interface. This host mode greatly simplifies interfacing the TNC1 to your computer. Note, this interface does not look at all like the TNC command structure so you will have to write your own TNC to Computer interface program. There are two other TNCs that need some mention here. They are the Kantronics and the GLB PK-1. Both of these units share the same system for generating the packets. These units use a software frame generation scheme rather than the hardware approach used in the TNC1 and the TNC2. The problem with the software approach is that it is slow and difficult to modify. In the future we will be using higher speeds, probably 9600bps locally, with the network backbone running at 56Kbps or faster. None of the software based TNCs can be adapted to the faster speeds. Also, you are at the mercy of the hardware vendor, either Kantronics or GLB, for software updates. It is unlikely that Joe Ham and Computer Hacker will be creating new firmware for these units the way that Ron Raikes did for the TNC1. What way is best? That is a matter of opinion. I currently lean toward the TNC1 with the DED code, but I am not fond of the 6809 processor it sports. I prefer the TNC2 hardware, but am lukewarm about the current firmware available on it. I am not interested in either the GLB PK-1 or the Kantronics. If price and features are important, and you want it wired and tested, get the AEA PK-80. It will sell for about $220. If you are a hardware and software hacker, the Xerox 820 boards are interesting. The X820 can be turned into a full-blown CP/M system and TNC for about $350. The packet software for the X820 was written by Phil Karn and is in the public domain (FYI, the TNC2 development took place on the X820). Another potential way to go, if you have an IBM-PC, is to use the dual packet port board developed by Terry Fox, WB4JFI. This board is in the prototype stages but several wire-wrap versions exist. This board consists of a Zilog 8530-SCC with two world modem chips. There you have it, a not-so-comprehensive rundown of what is available :-). If you have any questions, let me know. 73 de Brian Lloyd, WB6RQN ...!seismo!umcp-cs!cvl!umd5!obpa1!yojna1!wb6rqn
wb6rqn@yojna1.UUCP (Brian Lloyd) (11/20/85)
> > If you are a hardware and software hacker, the Xerox 820 boards are > interesting. The X820 can be turned into a full-blown CP/M system and TNC for > about $350. The packet software for the X820 was written by Phil Karn and is > in the public domain (FYI, the TNC2 development took place on the X820). > *** REPLACE THIS LINE WITH YOUR MESSAGE *** I was incorrect in one of the above statements. Phil Karn's code is not in the public domain, it is copyrighted. "Hams" are free to copy it and use it for non-commercial purposes so long as the author is credited. Sorry for the misinformation. 73 de Brian Lloyd, WB6RQN ...!seismo!umcp-cs!cvl!umd5!obpa1!yojna1!wb6rqn