[aprssig] APRS Message Idea
Wes Johnston aprs at kd4rdb.comFri Mar 4 20:56:36 UTC 2005
- Previous message: [aprssig] APRS Message Idea
- Next message: [aprssig] APRS Message Idea
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Bob's right about this one though.... the decay is the way to go. One of the things I thought was so cool about TCP/IP when I started playing packet was that TCP/IP would back off exponentially when attempting to deliver a packet, as opposed to the dumber ax.25 'fixed timing with a little randomness' method. When delivering packets onto a network, with TCP/IP it would start out pretty agressive but back off quickly if the link turned out slow. AX.25 on the other hand could send a packet over a long link, and if set aggressivly enough, would send the same packet again twice before the ACK got back. Not very effecient. Also, there's another advantage to the decay. Let's say that I send a message right now. about 1 minute later, you send a message. They collide because you pressed enter at the same time my station resent a copy of my message. Now in a fixed timing system where we each retry every minute, we will always collide. In a decay timing system, my message's next TX interval is 2 minutes from now. Your's is 10 seconds from now. We don't collide again. The point is that even when our timings match up every once in a while, our interval to the next TX time is different and we won't collide again for a good long while. I wish more APRS clients used the decay model. Wes
- Previous message: [aprssig] APRS Message Idea
- Next message: [aprssig] APRS Message Idea
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the aprssig mailing list
