UT Zombies3-Beta
- Thread starter MrLoathsome
- Start date
-
Two Factor Authentication is now available on BeyondUnreal Forums. To configure it, visit your Profile and look for the "Two Step Verification" option on the left side. We can send codes via email (may be slower) or you can set up any TOTP Authenticator app on your phone (Authy, Google Authenticator, etc) to deliver codes. It is highly recommended that you configure this to keep your account safe.
This.
The whole idea of what the mod is supposed to be is being totally lost. You guys will just be better off to stop "fixing" these and model up some new ones that look and behave and have anims like you want. TWT was meant to play a certain way and that's almost completely changed now.
There's nothing wrong with creating new stuff like this but allow the mod to be what it is and not try to shoehorn it into places you want it to go. It's a lot like taking the DeadCity Zombies and making them slow and removing their jump...it's just not "them" any more. If it were me I'd use all your code you have and replace the zombies with some re-rigged player meshes. Any number of people here could whip up the anims for you in an evening and you would be much happier with where you are going.
Mr. GoPostal is almost 100% correct.
My goal in tweaking this has only been to try and improve performance and
eliminate any errors or warnings.
I did randomize the walking speed slightly, but made no other major
changes to the basic zombie behavior in the last update. (Zombies4)
The plan for my next update, would just involve tweaking the zombie AI so
they will interact with monsters, in the exact same "George Romero" night of
the living dead zombies fashion that they had originally.
Also plan on testing what the effect that strength value has on things
when I get around to looking at this in a week or 2. (Still got a few things
to finish up 1st...)
Anybody have any idea as to what range of values I should use for testing
that out? I haven't ever done anything with models before, but that info
would be helpful in tweaking the scripts.
(or I will figure it out the hard way as usual)
If the next re-write of the zombies works out like I want, they will be able
to use the same death effects as any other ScriptedPawns.
(i.e. effects generated by mutators such as MoreGore or MoreBlood)
Any suggestions that make the actual zombie pawns drastically different
than the originals will not be considered. (Other than any better or improved
textures, if they still fit with the original concept well.)
Note, some of the above ideas regarding dynamically generating the number
of zombies spawned by the mutator are already in the plan for Zombies5.
(You will have the choice via INI variable to set fixed or dynamic number)
This thing will require a considerable amount of testing before I release
anything new on it.
G.P. Let me know if this all sounds ok to you. If the max walking speed
I have in the current range, or anything else about it needs adjustment,
let me know. Full source for everything I modified was included with
the Zombies4 release.
Keep in mind, that even though the mutator and the ZombiePawns will be
bundled together, they are basically 2 different objects.
** I noticed while looking at the ZombiePawns source yesterday that
it has a default set for SightRadius=100000.00
Is that not an insanely high value? Or is SightRadius not using the same
UU values as everything else?
If not, wouldn't something in the 2000-5000 range work better for default
SightRadius?
Arnox, this thing does have a bit of code that make the zombies attempt
to spawn out of the line of sight of any players.
Depending upon the map configuration, and how fast you are looking around
you may still see them spawn at times. (More so if online)
I will check to make sure that is still working when I get back to this.
(Still deep in another project atm...)
Note, that zombies still might spawn directly behind you, assuming the bit
of code that is doing those checks is working the way I remember.
Not sure if I would want to adjust that or not.....
Not sure if I would want to adjust that or not.....
Definitely not!
It has made me jump a few times after clearing my way into a building, then turning around to see I'm blocked in.
Note.
Any suggestions I make regarding this mod (and most others), are chosen on the basis that they should not (in theory) add any resources or bulk to it, that aren't already in UT (other than more lines of script).
I like to keep things running fast, as I run an insane amount of mutators, and some produce a fair amount of slowdown when used together.
I'm sure having all the zombies glowing before "popping" would slow the game, but if 1 in 20 or 30 were "infected" and the gibs could damage you, I think it would add more tension if you were in a room with 1 of them in a hoard. This would not change their behaviour or any other factor, just how you deal with them.
I'm sure you see what I mean.
I am just throwing ideas in the air, but I do quite like to know where my ideas are workable or not, or if I am completely getting things wrong.
This is usually a good way for me to see if I have been learning anything
However this mod finishes up, I am already more than happy with the excellent work you have done bringing this mod back...to......life? Death?
Un-life?....
Un-death
I must not have read that well 1st time I looked at it.
Seriously? Completely changed? Have you even tried the Zombies4 version
or looked at the new source code for the mutator and pawn?
You would have to be watching things very closely to notice any difference
in the way these play.
Almost all the changes I have done so far merely fix a bunch of problems
and dead code that was in the Mallzombies source you provided.
To tell you the truth, so much in that was broken I was sort of surprised they
even ran at all.
At this point, you should be able to add even more zombies without hurting
performance.
Basically none of the suggestions or ideas others have been posting in this
thread were used in the current Zombies4 version of the thing.
I am confused....
Behind you is fine. Right in front of your face is very annoying. For exampleNot sure if I would want to adjust that or not.....
Me: Yeah, playing around with my BFG9000, wooooo!
*zombie spawns in front of me while charging up*
Me: OMG WT- BOOM!
Dead.
Or with any explosive weapon for that matter. Just sayin'.
If I revisit this I will do some more testing on that to make sure the line of sight check is working.
Unless you are on a map with a very low spawnpoint/pathnode count, that
shouldn't happen. (The Zombie mutator builds a list of those locations, and
uses them all for the zombie spawnpoints.)
If I do update this, it will only be to add support so the zombies will cope
with other types of scripted pawns instead of ignoring them or freezing.
Any other changes would only include further bug fixes and other
"under the hood" stuff.
If having alternate death effects for the zombies turns out to be workable, they
will be optional via the ini file and the default will remain as is.
(or, they might end up with the same death effects as any other monster.... Dunno yet.)
My goal all along has just been to make them function well independently
in any map, either via the mutator or as standalone scriptedpawn that
could be added by other mutators or however.
Other than the very slight randomization in walking speed, I was thinking
the Zombies4 version and the Zombies included with GP's MallZombies map
were virtually indistinguishable as far as appearance and behavior.
Although I did re-write a large portion of the TWT_ZombiePawns.u class
including a lot of the state code, the only 2 things you should notice are
very subtle.
1. They will now change enemy if another player/bot bumps or shoots them.
2. They now can cope with maps that have waterzones in them.
Last edited:
The map I was using was Deck16 but admittedly, I cranked up the zombie count to like 300 or something.
Yes, that would explain it. 300 is quite a bit for a map that size.
Things happen fast on smaller maps, and if a whole lot is going on and
you are moving while the check happens, they can spawn in front of you.
If you set the CheckRate variable lower, it might reduce how often that
occurs, but if you start noticing any lag or performance issues, turn it back
up a bit... (or reduce the # of zombies somewhat)
Just wanted to clarify something after skimming thru the last page or 2
here.
None of the code I mentioned above that I had to fix was written by
GoPostal. When I said I was surprised it ran at all, I was referring to
the original TWT_Zombies.
I did compare the original TWT_Zombies code to the version GP included
with his MallZombies map before I did the Zombies4 version.
He cleaned it up very well, and got it working so that his map had no
problems at all. I just had to look at both the mutator and pawn
code much closer to make it work as an independent package.
All the other dead, broken code I was referring to, is 11 years old.
I have just been bored lately and looked at it a lot closer than he
needed to.
We both tried to contact those guys. The original TWT_Zombies
was great, but had issues. I could tell the coder had just taken or
was taking a CS101 class in data structures. Lots of cool Linked List
stuff, and stack pushes and pulls, that were doing nothing, and not
needed anyway... They did call it a beta. At least the version I looked
at.
GP fixed the beta considerably, but there was still a lot that needed done,
and I am just trying to optimize it as much as I can. (I am bored...)
But, without changing the basic concept.
I did post this at 1 other site, and got 1 request for a new feature.
They were requesting an option to make the zombies move faster....
This was my reply:
If I do get a Zombies5 done, one of the changes will be to reduce slightly
the maximum random walking speed Zombies4 is using.
The point I was trying to make wasn't to put you on the defensive Loathe. What I'm trying to reiterate back to you is that admittedly you have greatly changed the code for the zombies. The last I looked at the code it was profoundly changed, to the point that it's a new mod using the TWT zombies model. All I was trying to say is that you should stop making more versions of this mod and make your own zombies. Then you could have clean, crisp textures on models with no drop-off, death and walking animations, a true hopping-up jump to cope with ledges...
All I'm saying is that you should maybe respect the work that the original coder did a little more than what is being done. I did quite a lot of changes on them myself but embedded them into a single map only. I also took great pains so that any casual player would immediately see that my zombie version was visually different, there were many new skins and sounds. You are up to v5 now and Joe Player isn't going to see any differences among the mods at all, including my version.
Three or four people are going to care about improved config files. Three or four hundred are going to see visually changed zombies. As it is now it's a mess of versions with no clearly distinguishable features for the casual mapper/player. They don't care that they can control spawn rate or that the zombies spawn more efficiently. They just want hordes of zombies to spawn so that they can be mowed down, cannon fodder if you will.
I do respect your work. I also love the old TWT's, as do quite a few other people here. I mean this in the kindest of ways but "go make your own". Everyone is going to be happier. I'll help you rig up a couple of existing models if you want, just pick the ones you want to use and I'll make the anims.
All I'm saying is that you should maybe respect the work that the original coder did a little more than what is being done. I did quite a lot of changes on them myself but embedded them into a single map only. I also took great pains so that any casual player would immediately see that my zombie version was visually different, there were many new skins and sounds. You are up to v5 now and Joe Player isn't going to see any differences among the mods at all, including my version.
Three or four people are going to care about improved config files. Three or four hundred are going to see visually changed zombies. As it is now it's a mess of versions with no clearly distinguishable features for the casual mapper/player. They don't care that they can control spawn rate or that the zombies spawn more efficiently. They just want hordes of zombies to spawn so that they can be mowed down, cannon fodder if you will.
I do respect your work. I also love the old TWT's, as do quite a few other people here. I mean this in the kindest of ways but "go make your own". Everyone is going to be happier. I'll help you rig up a couple of existing models if you want, just pick the ones you want to use and I'll make the anims.
GoPostal, I believe you are contradicting yourself. "Respecting one's work" =/= saying "Go make your own". Frankly, you changed the original TWT zombies for your map so you didn't make your own -- so SOD OFF instead of saying that line to the topic author.
You make a monster and release it.
I come along, reskin it and mylevel it into a map. Someone else comes along and release several "updates" for your entire monster mod with plans to do more, making fundamental updates to the entire code.
Those are NOT the same thing. I purposely kept my altered zombies from functioning outside my map.
I have spoken with Loathe in private, expressing my concerns over more and more versions. He's beta testing using the public and making a mess of the monsters in the process. MH maps will be (if not already are) made with different versions of this, thereby ensuring the propogation of the various versions. Need I point to any number of mods that have done this already?
Stop releasing a version each change, do some serious testing (which I have offered to help with, and have done), and release something that is finished. I love MH arguable more than anyone and the single biggest threat to it is mod conflict. OFC I'm going to point this out where I see it.
I come along, reskin it and mylevel it into a map. Someone else comes along and release several "updates" for your entire monster mod with plans to do more, making fundamental updates to the entire code.
Those are NOT the same thing. I purposely kept my altered zombies from functioning outside my map.
I have spoken with Loathe in private, expressing my concerns over more and more versions. He's beta testing using the public and making a mess of the monsters in the process. MH maps will be (if not already are) made with different versions of this, thereby ensuring the propogation of the various versions. Need I point to any number of mods that have done this already?
Stop releasing a version each change, do some serious testing (which I have offered to help with, and have done), and release something that is finished. I love MH arguable more than anyone and the single biggest threat to it is mod conflict. OFC I'm going to point this out where I see it.
Really?
I released the Zombies3 BETA, and removed it from the download after
just a few days soon as 2 or 3 people had grabbed it for testing.
Zombies4 was the only version released other than that.
Hardly a "mess of versions".
The 3 Beta_GP development versions, I only sent to you, and 1 other
person for testing.
Zombies5 was just going to be the next name if I was going to release
a version where the pawns were tweaked so they would interact with
monsters the exact same way they used to with player and bots.
Guess that wont be happening now.
When it is finished, Loathsome.Zombies mod sounds good to me as it is also a stand alone mutator, like the original.
consider the others still beta anyway.
It has now obviously become a separate programming branch, so maybe it's time for a rename, but it is important to keep the version history reflecting all these relevant changes.
as it is also a stand alone mutator, like the original.consider the others still beta anyway.
It has now obviously become a separate programming branch, so maybe it's time for a rename, but it is important to keep the version history reflecting all these relevant changes.
No I didn't.
Look, Loathe does really nice work. His code is clean, his mods function as intended and I'm a fan of what he does. He's just given to making many multiple versions of the same mod and releasing them instead of doing rigorous testing himself, essentially offloading the testing phase to the playing public.
There are several coders that do this, one only needs to look at the release thread to see "version 34ampv2a1" type renames across multiple mods. Hell, I'm guilty of it too as much as anyone else. However I will do strenuous off-and online testing before release to try to mitigate the problems.
The one big exception to this style of dev is monster (and weapons to some extent) creation. Once a monster or monster mod is anywhere on the net it immediately gets tossed into a map by someone. This means that the mod version will live forever in redirects and you are forced to save it.
So you can see that releasing multiple versions of a monster will always lead to serious issues on MH servers. Mappers find an older original file that is broken or incomplete and either use it or try to "fix" it. The Spinner is a perfect example of this. The beta version of that monster is everywhere with multiple "fixed" versions and virtually everyone ignoring the actual (really) fixed one.
I'm truly sorry Loathe is angry, I genuinely like him. It's just there are issues with how he is dealing with these and it's best to hash them out now before it gets (more) out of hand. If you can't tolerate criticism, best not dive into development. I've gotten MUCH more direct and harsh critiques of my work and behavior but you don't see me getting mad and taking my ball home. The bottom line is that yeah, I'm a jerk but that doesn't mean there isn't a problem.
Look, Loathe does really nice work. His code is clean, his mods function as intended and I'm a fan of what he does. He's just given to making many multiple versions of the same mod and releasing them instead of doing rigorous testing himself, essentially offloading the testing phase to the playing public.
There are several coders that do this, one only needs to look at the release thread to see "version 34ampv2a1" type renames across multiple mods. Hell, I'm guilty of it too as much as anyone else. However I will do strenuous off-and online testing before release to try to mitigate the problems.
The one big exception to this style of dev is monster (and weapons to some extent) creation. Once a monster or monster mod is anywhere on the net it immediately gets tossed into a map by someone. This means that the mod version will live forever in redirects and you are forced to save it.
So you can see that releasing multiple versions of a monster will always lead to serious issues on MH servers. Mappers find an older original file that is broken or incomplete and either use it or try to "fix" it. The Spinner is a perfect example of this. The beta version of that monster is everywhere with multiple "fixed" versions and virtually everyone ignoring the actual (really) fixed one.
I'm truly sorry Loathe is angry, I genuinely like him. It's just there are issues with how he is dealing with these and it's best to hash them out now before it gets (more) out of hand. If you can't tolerate criticism, best not dive into development. I've gotten MUCH more direct and harsh critiques of my work and behavior but you don't see me getting mad and taking my ball home. The bottom line is that yeah, I'm a jerk but that doesn't mean there isn't a problem.