If Gordon ... is accurate, then what is the purpose of the BBS?

None practically. It is the most inefficient AX.25 packet arrangement possible.

AX.25 connected packet depends on line-by-line ACKS. This works on a two-way balanced link when both stations hear each other reliably and without interference. But the Earth to ISS path is anything but balanced. It is solid ISS to earth, but the other way is probably 10% if that much. Thus, for each packet delivered, there are 10 RETRIES or more than 20 packets on channel to deliver EACH LINE.

But it is even worse than that. The ISS uses AX25V2L2 which means it does not even send a RETRY without first sending a "what was your last heard line number" packet. SO it will not even retry a single line until TWO more packets have been exchanged. Now multiply that by the 10% channel performance and it might take 20, 30 or 40 packets to deliver just ONE LINE.

That is why if someone logs onto the BBS, that the pass is more or less wasted. The logee never gets a successful download due to all the bazillions of inefficient overhead, and no one else gets to make any digipeated contacts either.

Just listen to a BBS pass, and its wall to wall packets, but only a line or two are actually delivered in most cases. The only true use of the BBS is possible if the ground station is using a few kilowatts of ERP to make sure that there are no missing ACKS!

Something like that,

Bob, WB4APR

On Mon, 25 Jul 2011 19:30:50 -0400 Diane Bruce db@db.net wrote:

On Mon, Jul 25, 2011 at 03:43:03PM -0700, Ted wrote:

...

Curious..

If Gordon (always ready with a cheerful comment) is accurate, then what is the purpose of the BBS? Also, Kevin is far from 'selfish'...he is just using

The BBS is a side effect of the commercial equipment (Kenwood) they took up there. Many of us do think it should be disabled.

Time for a quick description, for those who may be less familiar with how packet BBSes work.

Many of us are running packet radio BBSes or even just mailboxes, on modern computing hardware. Now this hardware with 16- and 32-bit processors running at clock speeds of over 100MHz, up to a megabyte or even more of RAM and maybe as much as a couple of gigabytes (that's 1024 megabytes) of mass storage is often capable of handling many simultaneous connections. Users can send messages to each other directly or via mailboxes, while other users are simultaneously reading messages or browsing saved bulletins - perhaps as many as eight users at once!

By comparison, the 8-bit 4MHz microprocessors used in most normal TNCs can really only cope with a single connection at a time. Once someone is in connected mode, everything else gets rejected. Some of the more modern TNCs can accept very large amounts of RAM (the KPC-3 can have 32 kilobytes), but most have much less than that. The TNCs built into radios tend to be fairly simpleminded beasts, with crude firmware and very little RAM for storing messages and connections.

The UI packets, on the other hand, are merely received and, if they are valid, repeated (almost verbatim) a moment later. There is a slight modification to the header to indicate that they've been through a digipeater, but the message content remains unmodified. Since the TNC has no need to track any state for these packets - UI is not a connected mode - then it can handle digipeating packets for any stations it can hear.

...

the available resources. Helpful education is far more useful than lame name

It's the way Gordon is, he's a nice chap online (on IRC). Don't take it the wrong way. The pair of you should go have a beer together or something. He's just a typical dour direct Scot, but he will drink you under the table. ;-)

Thanks, that's the nicest thing anyone's said all day.