[comp.dcom.modems] Size of UUCP 'g' acks

gandrews@netcom.COM (Greg Andrews) (04/07/91)

In article <7151.27F9618A@zswamp.fidonet.org> root@zswamp.fidonet.org (Geoffrey Welsh) writes:
>
>  [talking about the HST modulation]
>
>   It wasn't designed specifically for UUCP-g, but it definitely was designed 
>for protocols (both windowed and not) which sent ACK packets which were small 
>in comparison to the data packets, and UUCP-g does fall into that broad 
>category, doesn't it?
>

Uucp 'g' ACKs are six bytes long.  Given that an ACK is sent for every
packet, the receiver might manage to stuff TWO acks in before the modem
checks its buffer.  Twelve bytes could be enough to make them thrash the
line turnarounds.  Maybe even six bytes...

When I worked at Crosstalk, the Remote2 remote control program was having
trouble with HSTs.  The program's ACK packets were just big enough to
trick the modems into turning the line around, causing long interactive
delays.  I left the company a short while later, so I don't know if the
program was changed to avoid that problem.

-- 
.------------------------------------------------------------------------.
|  Greg Andrews   |       UUCP: {apple,amdahl,claris}!netcom!gandrews    |
|                 |   Internet: gandrews@netcom.COM                      |
`------------------------------------------------------------------------'