Home |
Search |
Today's Posts |
|
#1
|
|||
|
|||
Nosko S. wrote: Dana, I don't need to know, but what's the bug just for my information? http://groups.google.com/groups?hl=e...com%26rnum%3D1 Basically, the D700A TNC never sends an ack to an I-frame unless the I-frame has the Poll bit set. What this means is that the vast majority of other TNCs (as far as I can tell) will consistently send MAXFRAME number of frames to the D700, then time-out waiting for an ack, re-send the frames and set the Poll bit. It positively kills performance when using the D700 in connected mode to most BBSes and such. The problem does not appear when talkign to other D7/D700s, since the D700 *always* sets the Poll bit on every frame and does not implement a deferred ack timer at all. This is less than optimal as well. Of course, if you run in KISS mode and use a host-implemented AX.25 stack, this problem is avoided. Since the D700 is really focused on APRS which is implemented using UI frames (datagrams), it's apparently not a critical issue. Dana Myers K6JQ wrote in message s.com... ... Just remember it's not suitable for ordinary connected-mode packet cause of a blatant bug, but it's a fine APRS radio. Dana K6JQ |
Reply |
Thread Tools | Search this Thread |
Display Modes | |
|
|