SmartStockBots mutator

[Buggie]

Usual players will not survive. One high skilled bot enough. But if you unlucky, there can be others, but this is pure random.

---

Automatically merged

Also this reduce variety of gameplay. if always same team with same set - they always act in same way. so better play against different teams. It is more interesting.

[Que]

Can we have option of using ALL Skilled Bots
If we want to?

For competitive servers.

[Buggie]

For competive server you just make all godlike and this all. Godlike is maximal level for bots. But you can lost in gameplay variety.

For example WarCow weak when on foot, but very like vehicles and kill you when inside it.
DP|Farham is not greatly skilled in terms of skill points but very noticeable among other human bots with same skill. And so on.

So bigger skill points not necessarily mean more interested gameplay.

Even with humans - if all play same in same way with same tricks it can be boring. Variety is important things.

Also it is reason why exists things like AlternatePaths and so on.

---

Automatically merged

Also in code picked all bots of bigger skill and choose random one from it - exact by this reason.

Because all time play against Xan can be boring. So there be WarBoss and XanMk2 at least for some variety. At least in names.

Biggest problem of bots - they not learn anything.
This mean - if you able win - it is forever.
If you catch them on choke spot, you will able do this again and again.
Which not happen with humans.
You almost never know what humans do next time, since they able learn.

I not found good way for change it, but have some ideas like change weight of AlternatePaths depending on success or loose for bring flags.
But this require a big count number of attempts for work. And there many pitfalls about it.

---

Automatically merged

I have interesting idea about bot aim. Unfortunately this can not be implemented without replace it, so not suit to this mutator.

Idea consist in add bot aim pinged. Instead of use instant aim with zero ping and artificial aimerror, use prediction.
Bot use your past position on some moment, like 20-50 ms ago (configurable), and try predict your next pos. This pos is be aim pos with lesser aimerror from now.
This lead to more human aim.

If you move predictable - bot can more easier hit you. If you move unpredictable - bot not able so easy hit you.
Currently for bot this doesn't matter. On shot moment all world stopped and bot not care what you move be before that. Just pure random of aimerror.

Also this approach possible able fix well-known problem about dependency aim from distance to target.
Humans aim bad on long distance, good on middle and very good on close, but bad on very close. So if you get very close - hard to hit you.
Bots not experience this. Dependency from aim straight linear. If you very close bot hit you in 100% times. Which not human-like behavior.

But if use prediction, bot will not able predict your movement properly on very close range, since small mistake can make big change in result pos.

Unfortunately this can not be done with stock bots.

[Old UT Veteran]

Biggest problem of bots - they not learn anything.
This mean - if you able win - it is forever.
If you catch them on choke spot, you will able do this again and again.
Which not happen with humans.
You almost never know what humans do next time, since they able learn.

Well, question is, what would you think they would need to learn?
In some racing game, I remember bots learning where and how to turn to have better time around corners. This was from seeing and tracking player after time.
In some fighting game, bot AI would start to copy which controller buttons player would use the most - they develop similar fight styles by mimicking the player.

In comparison to UT, bots would need to mimic human behavior or play out learned sequences. Say for example player use dodge-right a lot with flak cannon - bot do the same but now its exploitable.

For navigation, I think learning has benefits. For everything else... how do you teach bot where to throw a charged bio gun? Do a shock combo when player not visible? Shoot ripper so it bounces to a specific spot? Or fire 6 rockets at a spot?

godlike ut2004 bots took a cheat path: always dodge projectiles that are not hitscan
Unless you got good at hitscan, you would have a hard time killing the bot.

Compare that aspect above to real players, feels like most of them in this game:
Never take prolonged fights - always keep the opponent guessing.
Know timing of inventory, which bot already mastered.
focus more on survival than killing the enemy first - whoever wears down first through their health and armor loses.

Humans are also not tied to touching a path node during combat every 2 seconds

[Buggie]

I more talk about CTF. But it can apply to DM too but in less aspect.

Let's imagine bot take flag. And return same way. Where it be 100% killed. And this can happen 100% times. Bot not learn use another way.

Or in DM distance goes more close, bot switch to Flak and die. Every time. Because lost time for change weapons. Not learn anything.

Bot goes to demeer in long way. On exit it 100% blown with rocket pack, since all see how bot do that and wait little and fire. Bot as train appear in exact time and die.

Bot goes to shield in trap and depressured. This learn nothing. So bot goes again and again and die again and again.

Bot run for player, which fallback and make combos. bot die. Again and again. No any learn.

Also player can use bot knowledge for timing stuff for make trap. Bot jump to shield. Step out. Combo. Frag. Can be repeat until ammo not ended. Bot will jump here again and again. No memory, no learning. Nothing.

In exact same situation bot always use same tricks. So if player found way which cover all possible action of bot here, bot in trouble. But bot not analyze. It just do it again and again. And goes be easy prey.

[Old UT Veteran]

Let's imagine bot take flag. And return same way. Where it be 100% killed. And this can happen 100% times. Bot not learn use another way.

Its not a big issue in my opinion. Its more map related. On the other hand, If player with flag wants to trick bot, that won't happen because bot always know position of flag. Best solution already exists IMO: stopping player shenanigans.

Or in DM distance goes more close, bot switch to Flak and die. Every time. Because lost time for change weapons. Not learn anything.

Mhh. Maybe mutator can hack this to decrease switch time for bots? I think ive seen more players use the rocket and flak from long range just because they panic in the moment. I think it looks more stupid than having to choose the right gun.

Bot goes to demeer in long way. On exit it 100% blown with rocket pack, since all see how bot do that and wait little and fire. Bot as train appear in exact time and die.

Something for this would be nice. In team games waiting for backup is a possible fix. On map I would place Sektor's "Hunt Node" so that bot does not go chasing through a strategic point - find another way around while in combat.

Bot goes to shield in trap and depressured. This learn nothing. So bot goes again and again and die again and again.

I'm guilty of this as well.
Map fix: placing a hunt node at DM-Pressure door so that bot does not enter chamber when in fight. Bot consciously will not enter the chamber. If bot already in the chamber, that's a different story

Bot run for player, which fallback and make combos. bot die. Again and again. No any learn.

UT4 bots do this 90% of the time. They rarely ever chase player.

[sektor2111]

As usual, is not only about Bot when it comes to intelligence. Several other external things are causing a different behavior and these are more or less map related assets or... random injections. A normal PathNode can be assisted from outside by a non-navigation actor locking it or unlocking in certain conditions - and I'm afraid that here is about human intelligence as well (that "mapper"). Map itself can be educated to deploy whatever things in certain conditions (number of pawns, number of available pickups, etc.) and here is another chapter, map's intelligence triggering smarter Bots. And then ? And then a smarter Bot means a smarter map... Note: do not expect smarter maps from cube drawers and pictures makers...

[Que]

@Buggie

Can we stop bots from attacking pawns behind Invisible-Collision-hulls?

Probably not but thought I'd ask.

[Buggie]

Bots never attack anyone behind Invisible-Collision-hulls. At least if it solid. In fact solid. Which stop bullets and movement.

Bots can attack target behind BlockAll or mover. Since FastTrace ignore both (second only on dedic server).

But proper Invisible-Collision-hulls is just regular geometry and never create such issues.

[Que]

Have a look at my map DM-Boundaries_Monsters

Bots and monsters constantly try attack each other from behind invisible collision hulls..

Is it because there is small gap?

---

Automatically merged

I will make gap air-tight and see if works.

[Buggie]

Gap can be reason. Also I want mention ICH is static geometry. For last view of your map I see queen and warlord behind mover. It is not ICH.

[Que]

Ahh yeah.

...

Exploding glass / mover .. my bad

Can we make bots aware of movers as collision hulls?

[Buggie]

Mutator can not fix this stuff. In fact it can try do something but it be really awful for performance. this issue must be solved on server side. Need implement MovingBrushTracker for dedic servers. I hope one day in some v469 patch this issue will be solved.

[Que]

https://github.com/OldUnreal/UnrealTour ... ssues/1118

See what happens

[f7r]

XC_Engine?

Moving Brush Tracker in Dedicated servers (movers block visibility checks), specific maps can be ignored.

---

Automatically merged

DM-!!!shoot_dedic_bug successfully passes.