Monsterhunt2 'lite' version

[JackGriffin]

It's sad to see but there is a define decline in the number of MH servers running. One of the chief complaints there was with MH2 was that it took too robust a server to run it and guys with home setups got too much lag and crashing. I hope to address this within a couple of weeks.

I've begun work on a project I'll call Monsterhunt2Lite. This will be a stripped down version of MH2 with custom weaponry, most of the extra admin stuff, player stats, and relics removed. I'm optimizing the code to run as cleanly as possible so that this mod can be used with only the extras you want. It's my hope that it will function cleanly on a home server once I get it done and allow for some other servers to come back.

What I'd like from you is errors you see that the mod still generates, ideas about things that may still need fixed, or just general feedback about what you'd do differently. This will not be a replacement for MH2 but an alternative for admins who may want to run a weekend server or just like to have one from home and not have to pay the server rental fee.

I'll also be taking a long, hard look at the monsterhunt mod itself and making some improvements there too. I'd like to even add some limited bot support in if that's possible without adding overhead. I'll know more as I test. Anyway, I look forward to your ideas and thoughts.

[Higor]

If there's something that lags servers, are excessive Ticks and Iterations on them.
I'd look into that first, if some of these iterations aren't exactly critical for every tick, you might reduce them using simple FRand()'s.

Too many load operations ingame can cause lots of overhead, try precaching everything (like spawning an actor holding all references) at map start for servers/clients.

[JackGriffin]

I'm glad you showed some interest in this Higor. I did have a couple of questions to ask you later on about bot support.

MH's main problem is a poorly coded base. Because of this base we are stuck with making adjustments in replacement functions and using timers to make the monsters behave. All MH maps have actors that belong to this base so I was stuck with making changes externally. I've learned a lot though in the last couple of years and I see things that could get improved upon. I always shied away from bot support as MH is not really a bot friendly game but with the servers starting to be less used I think it might be time to try bots out and see if they can be useful.

First though I want to see if I can do some more tweaking with scriptedpawn. There are times (depending on your inventory) that they do not attack you on sight. I'd like to fix that and make them have a better aggression scale depending on the number of players in the map. There are a lot of map errors too like badly coded counters, factories, movers, triggers, etc that make a mess of things. I'm gonna try to fix those that I can. Oh, and I'll finally add a working translator for all the maps with translator events in them. I've been wanting to do that for a very long time. Almost all MH players have no idea they are even there.

[Higor]

I used to do lots of MonsterHunt runs as well as 5 other ppl from my zone, but the map choice always ended up in huge cubes filled with titans.
It would seem that I play MonsterHunt because I really wanted to revive the my old Unreal experiences, I even coded a halfassed MissionTeamGame to play Unreal and RTNP maps on UT + BotZ.
When doing that, I also had the issue that Monsters won't attack other that ScriptedPawns and PlayerPawns, of course, since I had BotZ I simply solved it on the BotZ side (BotZ has monster enemy attacking player? 50% chance monster enemy will be switched to BotZ, 70% if player has low health).

From what I've seen, Monsterhunt does something similar (using the PawnList to make some of the monsters attack bots instead of players).

After these years, I got that the only way to solve this, is to create ScriptedPawn subclass.
Name it MHScriptedPawn, modify SetEnemy() as well as some other Key events (maybe monster AI scripted sequences can be added, or adding other properties like bDontAttackPlayers for showcase monsters)
Then duplicate one by one all the monsters as subclass of MHScriptedPawn.

EDIT:
Since there are existing maps already, one way to solve this, would be to use a custom actor replacement to make exact copies of monsters when replacing them (only when class=class-to-replace, duplicate tag, health, and visual appearances as well as patrol, ai stuff)

[JackGriffin]

If you ever feel like playing a good coop unreal game I have an entire coop server download with a custom controller I made that gives you a lot of options to play with. Matter of fact it was my pet project about a year ago but sadly very few people play any coop.

I tried the way you mentioned on the scriptedpawn subclass but it became a real strain on the server in some maps like ATAA because there are just so damn many vars to assign that mappers change. I ended up accessing SetEnemy to stop monster fighting and I'll do a bit more in this update to ensure aggression.

[JackGriffin]

An update and a question.

I stripped the mod of any need for a custom PRI class so it can be run with normal Val Av or any combination of mutators that you find your server can tolerate. I'll explain this much more in the readme but expect a mainstream mod instead of MH2 with bells and whistles. Some sacrifices have to be made with the candy that was included but it sure runs leaner and faster. I think in the end this will do fine on a home connection so long as the mutators added to it don't go overboard.

Huge thanks to Nelsona. He's quietly been coding and testing a great deal of things in MH and he opened his code up to me without any restrictions. With what he has done and what I have learned from my logs you can expect a very solid mod that will not crash. No, that's not a guarantee but I can promise that a lot of the existing crash problems are going to be gone with this. I have a testserver up but it's not quite ready to show (I don't want the code ripped off yet) but I'll get something public in short order.

Now my question. Carcass causes a lot of overhead in MH. Gibs and chunks and blood spray, it all contributes to a lot of the stress in wilder maps. Nelsona and I discussed this over the weekend and he showed me some ways he has used to minimize the problem (which work very well). Cleanup happens better now but visually it needs a bit of tweaking because I think it happens too fast.

When a carcass is spawned, like say a titan dies, the carcass immediately moves to State Corroding. That's when the carcass seems to shrink up some then blink out of existence. I don't like this because it's too fast. What I'm thinking is to change this to some sort of effect that's more pleasing but I'm at a loss as to what that is. Anyone got ideas?

[Creavion]

Although your MH version is meant to replace the original one, it is somehow crazy to develop at times like this new gametypes. As like with Shadows SDK most of this is somehow "pointless" if there is not anybody at the end using it for creating content (mostly maps you know OR weapon mods, both usable to enjoy a game).
Sooner or later the arrow will hit UT`s knee. With the years fading away the amount of mappers will obviously reduce even more. Yes, for me mappers are the most important people if it comes to keep a game alive, since they do not only add new content, we can assume they play them with other ones. People like Ferali or papers Food Fight team are not less important of course, but (good) weapon mods are not as numerous as maps.
We can count yourself lucky, UT1 is not as dead as its sequel releases. Because of its quality it could survive as like Quake 3 over a decade. But at some point there will be an end. If the user and content activity of this forum will equal the one of unrealdeadground, we know it is over and then I will definitely leave as well. For nearly a year I am learning 3D Modelling, making myself finally independent from UED, my currently island is more or less a "learning project" (e.g. UV Mapping).
I surely do not want to piss you off with that but you might know, I speak out the truth. At the end you need content to make your mod famous. I would wish you a lot of success, for sure. But this is not a mutator which can be just added into the game. I do not think your mod is meant to replace MH, you will need content and this VERY BADLY.
This is also the problem of Unreals 227 patches. The enhanced Unreal engine 1 is really great. Dots and smirf were able to add a shitload of good stuff into it. Alone static meshes are able to change the whole mapper experience. But at the end nobody can use it anyway (I am not joking) or do not want because of compatibility with UT. So it ends up pointless. No matter what, Unreal 227, Shadows SDK, your new MH. Without content nobody might care and it does not matter how well designed and how full it is of good features. Of course we can assume you enjoy making this stuff but it is really a big risk doing that. So to put it in a nutshell: If you could manage it, you would need a small mapping team, but this is not easily done. As you might see, I also was not able to repeat the "CMP2 wonder". I would surely not able to get a whole team together as like 3 years ago. Mappers are just ... gone.

[JackGriffin]

Creav I could not agree more. However MH2 doesn't require new maps, it works with the existing 1500+ or so regular MH maps that are around. Keeping it able to work with the existing actors in MH maps while still doing what all I wanted to do was the real tough part of this. Default MH is incredibly unstable and badly coded that it's been a beast trying to make it stable enough to move forward.

[MrLoathsome]

Bluto: "What? Over? Did you say "over"? Nothing is over until we decide it is! Was it over when the Germans bombed Pearl Harbor? Hell no!"

Just look how active our retired members are.

Good to have you back posting stuff JackG. (UScript is quite addicting I think....)

[JackGriffin]

I'm on a short leash, and it's been yanked once already so I'll be good

[Letylove49]

what news about this project ?

[EvilGrins]

MH Lite = MonsterHugs
Could be interesting.

*runs and ducks from the assembled mob*

[Fuzz_Ball]

MH = Massive Herpes?
Hey! Where is this new thing? I'd like to futz with it!

[JackGriffin]

I got to about 85% on it but I've been called out of the bullpen on another project. It's probably going to be early summer before I get back on this. Fuzz, if you are interested I'll send you the code and chat you on TS on what's left to do. I did the hard parts already, the rest is filler. You're welcome to work on it.

It might be worth your time. There are some pretty damn sweet AI fixes for the monsters in this. They behave much more like true monsters.

[Letylove49]

thanks for the answer.