Test my MH map, DONE, made for bots.

Tutorials and discussions about Mapping - Introduce your own ones!
User avatar
EvilGrins
Godlike
Posts: 6381
Joined: Thu Jun 30, 2011 8:12 pm
Personal rank: God of Fudge
Location: Palo Alto, CA
Contact:

Re: Test my MH map, done. made for bots.

Post by EvilGrins » Sat Feb 14, 2015 8:07 pm

sektor2111 wrote:
EvilGrins wrote:
sektor2111 wrote:I think UT doesn't allow player on Player spawn-telefrag (because I did no see it)
Oh, it does. I've seen it plenty on lots of maps.
Example, please
A stretch back you edited one my guhzillion edits of a map to display monster-pathing to me, you named it MH-DefendingLand. Spawn points are a little bit in the air on the top of the Hill.

If you start the map with 9 or more players, those spawning into the game often telefrag each other because that's more players than starting spawn points.
http://unreal-games.livejournal.com/
Image
medor wrote:Replace Skaarj with EvilGrins :mrgreen:

User avatar
sektor2111
Godlike
Posts: 4140
Joined: Sun May 09, 2010 6:15 pm
Location: On the roof.

Re: Test my MH map, done. made for bots.

Post by sektor2111 » Sat Feb 14, 2015 10:32 pm

You cannot start map with 9 players. That's a server ! That's means HUMAN, I know that you have an obsession for Bot (like me) but PlayerPawn doesn't telefrag PlayerPawn in MH, only probably in DM game-types. I couldn't see that in MH where acts a single team and I was playing in more servers. Bot can kill - telefrag Monster, Cannon, Human, whatever pawn. Human is a bit limited. Anyway I'll recheck this in servers with more people Not starting a local play with Bot. Bot will always kill you if you stay around.

BTW because I mentioned On-Line play, that Mixer mover looks nasty On-Line it is rotating insane. I think an AttachMover will be different... I'm not sure if is not affected by those zones.

Red_Fist
Godlike
Posts: 1538
Joined: Sun Oct 05, 2008 3:31 am

Re: Test my MH map, done. made for bots.

Post by Red_Fist » Sun Feb 15, 2015 12:03 am

Hmm interesting, it is just a regular mover with rotation set, or maybe a "rotation mover" can't remember. It should work online. Not sure if those zones can affect it in any way be weird if they were.

There is one thing, it seems the spawnpoints can only be 16 per tag-event, the stock ones or the one Higor made. so if anyone crams in more than 16 spawnpoints per tag it don't do any good. I am going to check the gasbag ones I have, I think I used 32 of them.

Well I already have another version, I chopped chunks out of the pillars on the ramps and readjusted the paths a little bit. it looks a little structurly inaccurate but not to bad.

I used deco instead of nonsolid brushes for the endcaps on my florescent lights.

Added a trigger that I forgot to do for the two teleports on the pair of bluetop pillars. I had to put teleporters on the pillars and turn them on later on the ramp because skaarj or other pawns could fall or jump off the ramp and land on the pillar. In which the bots could not finish that crop of enemies and never go up the ramp. They only teleport to the ground, not up.
Last edited by Red_Fist on Sun Feb 15, 2015 12:17 am, edited 1 time in total.
Binary Space Partitioning

User avatar
FraGnBraG
Inhuman
Posts: 837
Joined: Sun Jun 13, 2010 5:13 pm
Personal rank: Good news everyone!
Location: Canada
Contact:

Re: Test my MH map, done. made for bots.

Post by FraGnBraG » Sun Feb 15, 2015 12:08 am

attach movers work online, it's gradual movers that definitely DO NOT ...

Red_Fist
Godlike
Posts: 1538
Joined: Sun Oct 05, 2008 3:31 am

Re: Test my MH map, done. made for bots.

Post by Red_Fist » Sun Feb 15, 2015 7:38 am

"There are located some LiftExit-LiftCenter-LiftExit combos which indeed helps in small nasty places but this time they are loading paths useless. I found pathing bugs from ramps and they might work without to ruin alarms. Try to avoid BlockPath to next WayPoint if there is no alternate route available because Bot will stop attacking - else will scan aggressive all Path-Network wandering. If is killed in Pupae zone Bot has no reason to turn in area only if is curious about some item."

That is the problem, the lift exits and the liftcenter forces the bot to unconditionally walk up the ramp. If I don't do that they never hit the next waypoint because there is another blocked path still blocking the next waypoint they don't even want to go up there , because it has a blocked path in front of it.

So I wanted the bots to ONLY go to the next ramp once the total amount of enemies have been killed and then turn off the blocked path. By using the liftexit and lift center, they walk all the way up to the end of a ramp, hitting the waypoint by forced paths. Once they kill all the pawns, and have hit the waypoint BEFORE the next waypoint, they then proceed again by hitting it via the liftexits,

it's a catch 22, they have to stay blocked until the crop of pawns die. but they will never want to go up the ramp because the blocked path is still blocking the next waypoint. (that is why you see two waypoints on each corner) The first one they can touch before the blocked path by force and egg on fights for pawns, using the lift exits. The second one around the corner will be unblocked once the pawns die, repeat, repeat.
Now everything would work right if all it was, was to go to a point and the next point like the buttons. But here I am making it conditional so they must be forced into walking into the latest waypoint. It really don't even need the movers, I only added those incase they hit the next waypoint by mistake from fighting as a precaution.

It also even matters where I put the spawnpoints and the time interval of the spawns, they needed to be at the first half of the ramps otherwise they hangout and get stuck off the patrol points unless a bot gets there to fight, which they don't.
The whole thing is glued together with paperclips and duct tap, one little thing wrong and they won't make it to the top.

if I didn't use blocked paths they would just march right up to the exit.
Touching a creaturefactory is half hazrd it would not be reliable. Because it does not give the bots purpose and make them want to go up the ramps on their own. I don't see how that could work.
and luring them up there with pickups also does not work.
Binary Space Partitioning

User avatar
sektor2111
Godlike
Posts: 4140
Joined: Sun May 09, 2010 6:15 pm
Location: On the roof.

Re: Test my MH map, done. made for bots.

Post by sektor2111 » Sun Feb 15, 2015 9:04 am

Red_Fist wrote:That is the problem, the lift exits and the liftcenter forces the bot to unconditionally walk up the ramp. If I don't do that they never hit the next waypoint because there is another blocked path still blocking the next waypoint they don't even want to go up there , because it has a blocked path in front of it.
Doesn't make sense: Path Blocked and... No blocked ??? Then get rid of nodes or do something else. Or... if way is blocked Patrol works ? I saw patrols there. As I can see Path to alarm = None (it doesn't log which monster has problem - it's internal monster log and is simple - this time Epic did not log important things). If path to alarm is missing then I guess neither patrol is not better. That Combo is not for such stunts, such distance is too far, I'm not sure what happens with bot if is killed in that moment. And for sure mayn nodes on those ramps are too far so I don't think patrol will work.

Net-Codes - Fire game using Multi Player Menu unchecking advertise server - using DEDICATED button. Use the same version to join to than personal LAN test server - you will have first image how things goes On-Line. See how reacts Nali at player On-Line - XC_Engine doesn't block SeenPlayer through those mover-glasses. Is different and it might be bad.

SpawnPoints used more than 16 need custom things.
I was talking about PlayerStarts. MH is a Botpack game child using default routines = 16. I experimented less nodes, different paths, better respawntime at ammo and less ammo spots - removing Nodes from Teleporters because is wrong and suddenly DevPath worked properly. YES, it is doable. AT this moment I have saved a Player Crash, I cannot say what was wrong there but something bugged player movement in map originally from your link.
MH-Electrolyte.jpg
It was On-line in a separate machine using separate resources as it happens usually in servers.
Red_Fist wrote: So I wanted the bots to ONLY go to the next ramp once the total amount of enemies have been killed and then turn off the blocked path. By using the liftexit and lift center, they walk all the way up to the end of a ramp, hitting the waypoint by forced paths. Once they kill all the pawns, and have hit the waypoint BEFORE the next waypoint, they then proceed again by hitting it via the liftexits,
Here you need a custom rebel node that need to block Bot nearby. BlockPath is messing path completely. A node having a kind of boolean bAllowHuntToward that can be turned ON or OFF with a... Counter.
How goes? If Monsters aren't killed then Bot won't pass that point but they can move if are too far since path is not blocked just debated with a special condition. I can say that will work because I was toying in other piece of crap map where Bot went in water too wick getting drowned. That Node did not blocked path but denied closer Bot until he got health and Scuba placed around to be prepared for a long water way.

Red_Fist
Godlike
Posts: 1538
Joined: Sun Oct 05, 2008 3:31 am

Re: Test my MH map, done. made for bots.

Post by Red_Fist » Sun Feb 15, 2015 9:46 am

Yes the blocked path don't block pawns.

Now in UT2004 you can set the pathnodes to "forced path" that is what I am using the lift exits and liftcenter for. I also set the extracost to zero, and in this case I set the height differential to 1024 and radius to 4096 because of the ramp angle and length.
I am sure there is a slick way to do it using stock game actors to make a conditional path. but my long distances are also making it more difficult.

You say the Nali are disappearing behind the glasss? I changed it back to a brush for the control room, but I know of this problem. I didn't want the Nali to freak out from splash damage if I used a mover. So I will probably have to use a bunch of blockall's.
Binary Space Partitioning

User avatar
sektor2111
Godlike
Posts: 4140
Joined: Sun May 09, 2010 6:15 pm
Location: On the roof.

Re: Test my MH map, done. made for bots.

Post by sektor2111 » Sun Feb 15, 2015 10:30 am

Red_Fist wrote:Yes the blocked path don't block pawns.
Technically BlockMonsters blocks them. BlockedPath does BLOCKS finding path to an actor UnReachable immediately.
Try to understand AS-Mazon, Bot could go at Door for the first time but it doesn't if door is not opened yet because of BlockedPath.

Please look twice at alarms. I found more alarms with the same TAG - this is wrong. Pawn takes in account FIRST alarm which might be over a BlockedPath, will discard alarm if path is denied.

Red_Fist
Godlike
Posts: 1538
Joined: Sun Oct 05, 2008 3:31 am

Re: Test my MH map, done. made for bots.

Post by Red_Fist » Sun Feb 15, 2015 5:36 pm

"Please look twice at alarms. I found more alarms with the same TAG - this is wrong. Pawn takes in account FIRST alarm which might be over a BlockedPath, will discard alarm if path is denied."

Yes I was wondering about that, I can check that out.
Binary Space Partitioning

Red_Fist
Godlike
Posts: 1538
Joined: Sun Oct 05, 2008 3:31 am

Re: Test my MH map, done. made for bots.

Post by Red_Fist » Sun Mar 15, 2015 6:58 pm

I still am working on a slightly fixed version of this map.

AND I took my textures and made a black and white (grayscale) and using colored lights for another version.
This other version is a more traditional MH, just kill the current crop of enemies and opens door. Keep going and weapons are unlocked. A lot more enemies more spam more like a MH for more players. Crude pathnodes ,bots can F-Off lolzzz on this map.

I think I found some of the lag problems in the large room, and found a few other things bad on the first map from working on the newer map.
This new one will have the exit always open (if you can get to it without cheating) hoooohahahaaaarggg. :twisted: :twisted: :twisted:
Binary Space Partitioning

User avatar
Radi
Adept
Posts: 362
Joined: Tue Apr 16, 2013 12:41 pm
Location: DeGie - PL

Re: Test my MH map, done. made for bots.

Post by Radi » Sun Mar 15, 2015 7:35 pm

Im still waiting for final version :) how much you make me wait for it ;p ?
I can download it now but where is link ?

Red_Fist
Godlike
Posts: 1538
Joined: Sun Oct 05, 2008 3:31 am

Re: Test my MH map, done. made for bots.

Post by Red_Fist » Sun Mar 15, 2015 10:56 pm

Radi wrote:Im still waiting for final version :) how much you make me wait for it ;p ?
I can download it now but where is link ?
The new one will replace the old one, and I will edit the thread better for a link.
Hmm I got side tracked into messing around in this other level.

All I need to do is watch the last room a few more times using one bot. Need to check a few snags for the pawns going down the ramps. I fixed a lot of things even though it won't have much to see or change. I needed paths, so in certain areas I put back the original spawn points. In other spots I needed Higors no path type.

Got rid of an extra un-needed 16 spawnpoints for the gasbags. And tested with one bot a lot and fixed some mover lighting, just a bunch of things. So it will be a week, I will try to be done in a week, next Sunday.
Binary Space Partitioning

Red_Fist
Godlike
Posts: 1538
Joined: Sun Oct 05, 2008 3:31 am

Re: Test my MH map, done. made for bots.

Post by Red_Fist » Fri Mar 20, 2015 11:43 pm

Ah ha !

I brought up the FPS in the large ramp room a lot. I would get down to 23 to 29 FPS, now I can't get below 35 and really trying to find a spot that low is hard.
I always keep vsynch on, so my top frame rate stays the same. However while scanning the room before I click to play starting flying around I would say 96% or higher stays at my highest (72hz) in that whole room. so at least now with enemies it shouldn't drag down below playing easily or easier, with much less lag.

It was the one large ramp brush, I now have add brushes all separate for the ramps and it's pillar supports. So I moved the center pillars of the ramps to the edge now. I should have just remade the ramps a long time ago. :loool:

So I will have to change it for the other map I am doing as well.
Binary Space Partitioning

User avatar
FraGnBraG
Inhuman
Posts: 837
Joined: Sun Jun 13, 2010 5:13 pm
Personal rank: Good news everyone!
Location: Canada
Contact:

Re: Test my MH map, done. made for bots.

Post by FraGnBraG » Sat Mar 21, 2015 5:27 am

imo, you should try getting rid of the HSD you have on all those surfaces... that _is_ a FR killer, no matter what system spec you have .. thing is, you don't really notice the difference if it's a darkish room (which it is) ...

Red_Fist
Godlike
Posts: 1538
Joined: Sun Oct 05, 2008 3:31 am

Re: Test my MH map, done. made for bots.

Post by Red_Fist » Sat Mar 21, 2015 6:20 pm

FraGnBraG wrote:imo, you should try getting rid of the HSD you have on all those surfaces... that _is_ a FR killer, no matter what system spec you have .. thing is, you don't really notice the difference if it's a darkish room (which it is) ...
I did that and it had little effect. I made the pillars regular and the ceiling, along with the new brush(s).
Right now it stays at my vsynch rate almost the whole map. Got some slowdown in that mixing room but very difficult to know why. Other than that I looked all over the place (no pawns just fly before playing with time demo), looking at different angles and it's about as fast as it can be.
Binary Space Partitioning

Post Reply