DyDo 2.2.2, AI for skirmish games and challenges

[guciomir]

The differences are not very big and AI is not able to use them to the fullest. Skipping new paradigm bodies will make other researches come faster and I think Dydo will benefit more from it. Of course this is debatable, but it would be nice if you write something more than "rly?"

[NoQ]

The NP bodies are much faster. They can carry most weapons at full speed (125) on tracks. You can also compare heavy cannon hovers based on python and mantis: the speed boost is dramatic. And, yeah, scorpion plasmite vtols at speed 525 (:

And they're cheaper. Considering the amount of units bots produce, it is paying off.

And researching them doesn't take that much time, since power and engine upgrades are necessary anyway, and even mantis is researched very quickly.

AI is not able to use them to the fullest

Well, thats a good point. But i still think P≠NP this time

[guciomir]

AI is using both types of bodies at the same time and NP speed is nullified. If AI was building only 1 type of bodies (for example only Project or only New Paradigm) than the speed would be a factor here. But the force is mixed and given the fact that AI is grouping forces together in a numerous places, the outcome is that New Paradigm has the same speed as Project bodies, it costs less (but not much less) and has weaker HP/armor. If Dydo is able to use those bodies better, it would be awesome. (for example using NP purely for very heavy weapons). There is a lot of space for progress here.

Some more ideas:

9. In 1-1 games when AI is adjusting % of its forces deciding to build AP, AT, AR weapons, does it take only player droids into consideration? Or does it also take player defences to optimize the force? I'm worried about the following: I build a few cyborgs and a lot of defences (no bunkers), this will make Flamers/Rockets AI build mainly Flamers while it is not the most effective way to fight the player. If Dydo is able to identify player defences, it could create some weapons against them (for example mortar cyborgs are great unless the player has artillery in its base, maybe some bunker busters etc)

10. It would be nice (in a distant future) to create "Rusher General". Example, in T1 game , it builds max number of cyborg factories, researches mainly flamer technology (for flamer/rockets) or mainly machine guns (for guns paths) + improved armor and health, creates like 100 cyborgs , set them to do or die and attack player. AI does not focus on other researches, it has only 1 strategy against the player. In a 4p game, it would make a lot of trouble for the player,

11. Once more about grouping forces.

YYYY - Dydo
YYYYYYY - Dydo
YYYYYY - Dydo
----1 - empty space
----2 - empty space
----3 - empty space
XXXXX - my defences
OO - some artillery


All of my defences are firing at the first row (the nearest one) of the Dydo forces. Dydo does not engage my defences (it goes back a few steps, it gets closer a little to my defences etc, moreover high range Dydo is blocking low range Dydo to come closer and my defences have a lot of time to kill everything). Dydo units start retreating which creates more confusion for attacking Dydo forces. Only some Dydo units are firing at my defences because many are outside the range. (only nearest row is close enough). Relatively small defense is able to kill relatively big Dydo force. I think it could be improved. When Dydo is attacked by defences, it should order it forces to go to closer range for example row "----2" so all Dydo droids can attack player. It would be best if Dydo concentrate fire on 1 defence tower, 1 by 1. In the first part of the fight, when attack group is coming closer, it can be set to do or die, so retreating forces will not make confusion. Of course, player artillery will make this much harder to make because it can trigger this "come closer" event while Dydo is not grouped properly and it can screw the tactic. But I think this is worth consideration.

[NoQ]

There is a lot of space for progress here.

Agree (:

[DylanDog]

Wow this AI is great! 2 research paths are great! I have a few suggestions based just on 1 game (it is best for me to test 2p maps because it gives me a lot of info how AI is working). Advanced bases on. AI flamers rockets.

I still suggest not to play 1vs1 if you are an advanced player.

1. Dydo is using new paradigm bodies. I think it is not necessary because they are similar to Project bodies and researching them takes time.

need to check this.

2. Flamer droids should be do or die (not retreat). In many situations they retreat even though they did not engage enemy and die few seconds after. This is because my weapons have bigger range.

Possibly you are right, I will run some tests to see if with no-retreat option Flamers AI is stronger.

3. Dydo is grouping droids in a strange way. They go from point to point, group there, go back a few steps then procede. Everything is happening under my artillery fire so Dydo is losing units when they are just moving from point to point. It could have fewer grouping points and fewer go-retreat-go-retreat actions.

What DyDo does is that it "regroups" his attacking and defending teams. This is necessary to keep attacking teams together even if slower tanks are on it.

4. Flamers/Rockets AI builded tons of machinegun tanks and mortar cyborgs at the beginning of the game (no flamers nor rockets). Is it on purpose? I think it was easy for Dydo to develop minipod and flamers (it was advanced bases game) but it chose not to use it in the early part of the game.

I think this is the case because of "Advanced bases" option. Possibly due to lack of flamers templates. This should not happend in the "standard base" option.

5. Dydo had very big defence VTOL group but no attack VTOL group. When I engaged its base with my armored tanks (like 25 tanks with full armour, scourges and twin assault guns), they got destroyed by VTOLs. If Dydo attacked me with such a big VTOL group, my defences would be shattered. I suggest attacking player defences with VTOLs prior to sending ground Dydo forces. Especially attacking player artillery would be nice. Dydo attacked my base with VTOLs only once during the game and attacked only 1 Oil derric. I just rebuilded it so no loss for me (apart for some energy that would be produced)

Strange. DyDo does has attacking teams. I have increased a lot the checks DyDo does before attacking the enemy targets. Possibly your base was well defended with AA? Also enemy base is attacked only with if at least 6/7 VTOLs are available (if I remember well)

6. When does Dydo build composite (more armor and hp) technologies?

Difficult answer, it is easier to have a look yourself at the VLO files to see research paths.

7. What is your current ToDo list for Dydo? I see most points on your website are already done and it is very nice to read what you are working on.

My current plan for DyDo is to stop worling on it for some time. I need be involved in something new. I am working on a mod to make ALPHA campaigns much, much harder.

8. Will you introduce later Dydo paths Flamers+Cannons and Machineguns+Rockets? I do not advise for doing this now because it is of course better to focus on existing paths and make them perfect. But i wonder if this will happen in a distant future.

Sure.

AS always, thanks A LOT for the suggestion mate!

[guciomir]

Nice

what about points 9-11 from this page?

[DylanDog]

just wanted to say that I am currently a bit too busy to work on WZ mods...I hope things will get better beginning of next year...

[Tenoh]

Hey DD,if you still working on the mods can you try and add a command to AI,like that help command with the red spot but a bit different- id call it hold this spot command :p,if you see for example a strategic place that you want to hold no matter what then AI could go there,clean it and put defense building there with some troops for cover.