Multi mod server - can it work?

[Pileyrei]

Many years ago I had a multimod server - it ran NIUT (everyone has same weapon, changes every 30 seconds), MH, Fragball and Chaos UT.

I managed to get it fairly busy at times but often it would sit empty.

Two questions:

1. Can a multi mod server ever be successful?
It seems servers need to be mainly one thing - DM, CTF, MH, etc to stand a chance of being popular.

2. If you were to run a multimod server, which mods would you install or consider installing?

Thanks
Piley

[papercoffee]

which mods would you install or consider installing?

The Damage mutator... to add more oomph into a shot and to make self-damage go away.

[SC]-[WARTZ_{HoF}]

I have been hosting UT99 servers since 2004 and not once has a multimode server ever done well for me. Clients tend to play one game-type or another and trying to mix them together on one hosted game server just doesn't work. You need to dedicate one gametype per server and this is coming from years of experience hosting. Wish you luck.

[Hook]

I am in same boat as Schwartz above.
I agree with him.
We recently completely changed out our CLD multi mod server that mars007 runs.
Popularity of CLD skyrocketed!

[Pileyrei]

Thanks for the replies.
I do agree that a multimod server will probably fail.
Was just curious if anyone had successfully got one working.

[Feralidragon]

Multi-mod is tricky to do well.
I have seen servers which pulled it off successfully, both in UT99 and UT2004, but they were mainly around vehicles (SLV in the case of UT99, and Assault and Onslaught in the case of UT2004), so it seems to me that the gametype itself can differ successfully, but not the actual "core" of the server (SLV and vehicles in these cases).

So, I would say that you should restrict yourself to one main mod overall, and then different gametypes, but all which may provide a similar level of gameplay (for example, CTF goes well with with JB and TDM).

[sektor2111]

I have two (stealth and temporary) sort of "multi-servers" more exactly servers with 3 game types each, and they do not really crash often because there I run ONLY MY OWN stuff and NOTHING from BotPack, no Nexgen, no crapped timers, no messed up stock repositories, no NewNet fantasies - that's the way of doing "multi-servers" aka multi-game servers from my point of view - not so useful for popularity, each player is pulling votes where wants causing conflicts and argues with others and all that stuff...

Such a sample of server is an old XC server done as a "How To XC" hosted by Medor - also running only custom MH games - and to mention small tech details:
- NONE of game-types involved in any server will NEVER need to load useless packages for Players not even MonsterHunt. If player is playing a game, will not need to load other game-types as packages for nothing;
- ANY game-type is self packaged AUTOMATED by Engine - ignore those docs saying things like "ServerPackages=MonsterHunt" or such because there are not even needed - maps are calling them as well - and for sure it's only an useless packages load for such "multi-servers" making player to wait ages for downloading files;
- All Downloads means Skins, New HUDS from mutators, New weapons, New pawns, New decorations which must be mapped to not be invisible, but NOT game-types and Neither Server-Side things or "ServerActors".

[sektor2111]

Question(s) from me for you (any - PickAnyTarget) about these options for "multi-servers".

1) Do you think that is useful for you a small toy able to do some packages mapping I mean automatically adding ServerPackages depending on game configuration and not all packages for all games at time ?
I'm asking because I've done a Packager Server-Side tool, mapping packages for certain game-type only when said game is loaded using the power from XC_Engine - yeah, it is addressing XCGE based servers. This will prevent player to download X packages for game-types which are not running but mapping desired ones. Then, you can make it more complex if you need more stuff... By example a DM scoreboard is not used in MH or MA or other custom game-type but player is downloading it because it will be needed in a sudden DM game-change voted.
If yes, probably it will go into XC section but I repeat, it's aiming XC Multi-Servers. I did it because I have done a configuration where I got tired of downloading junk files for nothing so I reduced downloading of packages in order to speed up joining time.

2) Another tool which I've done not really only for XC Servers is a "Delayed Loader". I'm talking about a Server-Side mutator able to load other Actors/Mutators but... having a delay.
Purpose of this thing.
I was hacking all crap from some paths network in a few Maps. At a moment iterations developed did not helped me and I went to a slow solution. The problem here comes with Relics type mutators, SwarmSpawn types and such. Simply mutators based on Navigation must wait until hacks/tweaks are operated in Level and being initialized AFTER this task as long as they do not have a problem if they do start after let's say 3 seconds. This one has logs for figuring if something is wrong in configuration reporting packages loaded. List has 64 places for such delayed mutators - I'm not sure if there are more needs. I tested this thing for mutators addressing server but which have to wait at least 2 seconds before to start and which are supporting a delayed start.
If you wanna check these and/or to develop them I'll drop them around...