Re: Networked xconq

[Jim Kingdon <kingdon at panix dot com>]

> 1. When did you test the network code? This spring or last year? Was
>    your xconq version up to date?

February 2001, xconq from CVS.  This is the same tests mentioned at
http://sources.redhat.com/ml/xconq7/2001/msg00092.html (plus Stan and
I did the same, maybe last summer or something, and got roughly
similar results).

> 2. What game did you play? Sequential or non-sequential?

Tried it both ways.  Got more errors with non-sequential but even with
sequential we'd only last 20 or 40 turns.  In the sequential case the
error seemed to be at the end of a turn.

> 3. How many participants? What kind of connection?

Two sides, two players (both human).  Derek and I had Cable modem and
DSL (I think), with the internet in the middle.

> 4. How many sides were under AI control? Did you switch the AI on or
>    off?

None; no.  With Stan I did find that switching on the AI (or setting
it for a particular unit), would immediately cause loss of sync.  But
that's not the loss of sync I'm worried about.

> 5. If you did see a synch error, please provide details about when it
> occurred and what (if anything) you think may have triggered it.

I can't really prove this, but I suspect the sync error probably
happened when we were both typing somewhat rapidly.  The kind of thing
which wouldn't happen in testing (especially if the testing only
involves one human) but which would happen pretty much every time in
actual play.

> 6. Did you experience any of the other problems described above?

I've seen the "unit it inside a dead transport" (including in
non-networked play) but I agree, in my experience it seems to be
harmless.

The other one that sounds vaguely familar is "Timed out while waiting
for ack." - I'm not sure I was able to recover once this happened.
But it has been a while, so I'm not sure that was the exact message
I'm thinking of.

Anyone interesting in playing an over-the-internet game, let me know
and we can try to set up a time (I'll be off the net from July 14 to
28, though).  If I remember right, Stan's debugging tactic after loss
of sync would be to have both players save the game and then run a
diff on them.