Map assessment and pathing enhancements

[Nelsona]

Yes, it seems that there are two versions of this so called MultiCTF - CTF4. I have files and I could check what was all about...
Conclusions are simple:
Geometry: Good executions, map it's suitable even for public gaming.
Pathing: No skills... based on EYE examination...
- "BigKeg" Not linked with anything because we don know the rules concerning LiftCenter linking;
- useless nodes near Inventories and very close to walls - Automated Generation without polishing placement was never great;
- Secret Items here ? What for ? And not linked from anywhere;
- HomeBase actors - excuse me but engine doesn't speak english, it's a software not a human being - a dumb addition - I doubt that CTF4 controller uses that... and not even Monsters embedded here. HomeBase it's aiming ScriptedPawn if are correctly set - it's not the case;
- Upper Placed PathNodes and FlagBases are not helping;
- DefensePoints for defending walls ? Do they not need the right view direction ?
In other hand, Scripted Checker has reported this misery:

Code: Select all

LostData: ________________________________________________________________
LostData: ReachSpecs missing from Paths[0-15] 		= 1846.
LostData: ReachSpecs missing from upStreamPaths[0-15] 	= 1655.
FixedWithSuccess: ReachSpecs successfully attached 	= 2727.
FixFailures: ReachSpecs attaching failures - probably no place = 774.
LostData: ________________________________________________________________

Clearly it's overcrowded without anything helpful.

Let me know if this stage has annoying flaws....

CTF4-UTDMS-ThornsV2_rS704.7z

(445.9 KiB) Downloaded 112 times

[evilgrins]

This is a big one, so no rush:
https://unrealarchive.org/unreal-tourna ... cc3a9.html

Pathing, it has none. Needs it above and below.

[Nelsona]

Map actually has some paths but... as usual, the boys are not aware about the size of "PathNode" and native issues cause by their lousy placement in ramps, it has also some scientific "bOneWayPath" but settings are exactly in opposite direction. The rest goes to original borked spam data caused by overcrowding spots which was never good for original UT Editor. For completion in years 2024+ we do have all needed stuff for doing them correctly and not just guessing like in preschool.

Edit: See this...
For me was normal (XC environment)

MH-Nexus_rS1984.7z

(360.38 KiB) Downloaded 102 times

[evilgrins]

Let me know if this stage has annoying flaws....
CTF4-UTDMS-ThornsV2_rS704.7z

Looks good, but I do have one little question.

Why did you change the projectiles on the UTDMS?

[Nelsona]

I think because I used my files for editing, something went changed - actually Skaarj Projectiles doesn't kill any Skaarj, my versions are using something different.

[evilgrins]

Skaarj Projectiles doesn't kill any Skaarj

I know, which is why in team games I switch the projectiles.

[evilgrins]

Assessment, though pathing is actually pretty decent there's a couple observable bad points...

...like a bot going thru a teleport, getting to the other side, and then running in place within the teleport effect on the other end.

No pathing for the Titan, he's already hard enough to beat!

[Nelsona]

... though pathing is actually pretty decent ...

Oh yeah... until something might be causing some sickness...(I did not throw up... yet)...

WskBar_0.PNG (4.25 KiB) Viewed 2927 times

And this can be a 469-Only map - if links are logically executed... either way it's just mindlessly loaded only for showing that map has paths...
And now let's see what builders checkers are reporting (remember, it's 2024...)

UnReachableNodes: Here we have 3 points skipped, 125 Navigation Points which are too high and 1409 good ones.
.....
BreakPointsFound: These 69 navigation points aren't that used as valid paths.
....
NoIncomingPath: VisibleTeleporter8 is not having incoming connections.
NoIncomingPath: PathNode83 is not having incoming connections.
NoIncomingPath: PathNode297 is not having incoming connections.
NoIncomingPath: PathNode298 is not having incoming connections.
....
UtterlyUseless: JumpSpot0 is not having any path defined.
UtterlyUseless: JumpSpot1 is not having any path defined.
UtterlyUseless: JumpSpot2 is not having any path defined.
....
NoIncomingPath: InventorySpot657 -> UDamage0 is not having any incoming connection...
NoIncomingPath: InventorySpot658 -> Armor3 is not having any incoming connection...
NoIncomingPath: InventorySpot660 -> UT_invisibility0 is not having any incoming connection...
NoIncomingPath: InventorySpot661 -> ut_shieldbelt0 is not having any incoming connection...
Isolated: InventorySpot666 -> UDamage1 is not having any path defined...
Isolated: InventorySpot667 -> HealthPack3 is not having any path defined...
Isolated: InventorySpot714 -> SniperRifle12 is not having any path defined...
....

So that's about "decent pathing" ? I'm totally disagree with this point.

First of all more than 1000 nodes in plain UT is already questioning about routing process which stops at 1000 Nodes. The rest of routing internals are more limited but in 2021 these boys can only constantly fail and learning nothing. Clearly here it's needed probably a full rework of everything.

WskBar_1.PNG (3.76 KiB) Viewed 2926 times

WskBar_2.PNG (353.02 KiB) Viewed 2926 times

My personal Notes bellow
Since mappers from EPIC were "experts" at this point, failing in n00b style, having all information in their own yard, we cannot expect external mappers to do things as they should or even getting more closer to a "good" thing. And then images concerning car accidents shown on YouTube are demonstrating that people are not understanding physics from this planet, going often in dangerous situations where even an animal can figure out what can do VS what cannot do.

[evilgrins]

So that's about "decent pathing" ?

You know me, I'm a simple player... when bots can seemingly get around okay I think things are "decent".

Think you can save it, doc?

[Nelsona]

I was flushing junks out so far and setting up some basics - attempting to trigger Goblin to do an original job correctly. If Goblin will go crazy then it will be some custom work. In other chapter I flushed duplicated ZoneInfo actors for the same map zone. This map needs more checks not only for paths...
Target: Map must work in original without "requirements".