Re: THE problem with xconq (long)

[Hans Ronne <hronne at pp dot sbbs dot se>]

>I have been following XConq from the days of XConq 5 quite passionately.
>
>Sadly, the last version I downloaded is 7.3.1. It has been a long time now
>I did not manage to do a multiplayer game of XConq due to the networking
>bugs and I have not yet heared on any progress on the issue. The code
>seems way too tricky for me to untangle.
>
>If XConq is to ever live again, forget all interface and eye candy issues
>and fix the networking code. It was designed for multiplayer and it has no
>life without it.

I couldn't agree more. The main problem, however, is not that the network
code could not be fixed. Stan fixed most of the problems back in 2000
before he started on the SDL interface. I fixed some more network problems
last year, and also added network code to the Mac version. In fact, I spent
a lot more time on the network code than on other stuff during the last
year I was actively working on xconq. This was because I came to the same
conclusion as you: without a working network code xconq would not survive.

The network game is now almost playable again. If you are lucky, you can
play an entire game of a simple module such as flattops.g without falling
out of sync. Most of the remaining problems are related to AIs, so if you
don't use AIs at all (and stick to games without Indepedents) you can avoid
them. If you download 7.4.1 or even better the latest CVS sources you can
see for yourself that things have improved.

THE problem with xconq is that nobody is working on it any more. Every
project needs a critical mass of interested developers in order to keep it
alive, particularly if it is done on a purely hobbyist basis. In fact, only
two people (Stan and I) actively worked on xconq for the last several
years, though Jim Kingdon also made some valuable contributions. We both
lost interest in working on xconq early last year and xconq development all
but ceased after that. I can't speak for Stan, but for me the main problem
was the absence of other people that were interested in this project.
Meaning interested in the coding, not just in playing xconq.

I think the only way for xconq to survive is if some more people get
involved in the project. That would at least encourage me to return to it.
I'm not going to work on it alone. I'm not sure how to achieve this,
though. Several people on this list have stated that they would start to
work on one thing or another (Windows interface, ai code, even the network
code) but nothing ever came out of it. I don't know what the problem is.
Perhaps the learning curve is too steep. Or perhaps the people who enjoy
playing xconq (presumably the people on this list) are not the same people
that would enjoy hacking it. Maybe we should search for new xconq hackers
in some other forum that is dedicated to game development? Any ideas are
welcome.

Going back to the network problems, I don't think reverting to the last
multiplayer version is a viable option. We had a long thread about the
network problems on this list last year, and that was one of the things we
discussed. The multiplayer version used the old X11 interface, which does
not even exist any more. It was first replaced by the tcltk interface on
unix, and then later again by the SDL interface (which was never completed,
though). Moreoever, there was never a working multiplayer version on the
Mac. This brings up another problem with xconq: its focus on a minor
platform (sorry if I hurt somebody). The reason why Stan started to work on
an SDL interface is that it will run under both Unix, MacOS and Windows. I
think that running under MacOS and Windows is just as important for xconq
survival as the network code. Therefore, getting the old X11 interface to
work again is not a solution. The future is instead the SDL interface which
is actually quite good (not as good as the Mac interface but clearly
superior to the tcltk and X11 interfaces). Too bad it was never completed.

So this is where we stand today. If we could get a few more people
seriously involved in this project, I'm sure several of the key tasks
(network code, working Windows port and/or a fully functional SDL
interface) could be completed within a year. Most of the work has already
been done. If Stan is not interested, I could probably coordinate such an
effort. I know enough about the xconq sources to tell others where to start
hacking. But I can't do everything myself.

Hans


Hans Ronne

hronne@pp.sbbs.se