[BETA] [FREEPLAY] Wood Ant Agro - End Game Lag (Recording)

[jaxmthemirc]

Since this newest patch...

Wood ants when marked to "move" and has enemies in it's path, they will not agro when in range of an enemy and will wait until they get right on top of the enemy before starting to attack. If they are "waiting" at the destination, they will attack when in agro range but not while moving. This results in alot of wood ants dying and thus wasting food.

(Not reproduced in recording)
Twice now my starter worker ants have stopped spawning despite having plenty of food. They just stop spawning. Before this patch I have been able to go right to end game without building additional worker ants. Obviously no more worker ants means no larvae being brought to your nests for more fighters... Royal guard being summoned and then the eggs hatching will respawn them, but after that they still do not respawn once killed. Not sure when exactly this starts to happen as it is in mid game or so.


Recorded the end-game lag as well. Starts to show it at hour 2 of the recording. Made some commentary on it during the final stretch even changed graphics settings. Also noticed if I pushed my ants outwards and cleared up some of the enemies, the lag alleviated itself and then all of a sudden came back. Looked to be just before the next attack wave timer started.

https://www.twitch.tv/videos/272475498

 

[MikeSlugDisco]

Thanks for the detailed reports - have passed them on to John.

 

[JohnSlugDisco]

Thanks for the report. You really put freeplay through its paces there so well done.

I have watched most of this video today and it has helped solve a couple of things so thanks for that.

The number of creatures you end up with is huge, something I am taking away from this is in the later game they need to attack you as they spawn as having huge numbers of creatures about doing nothing is going to tank the FPS for no reason.

Matt is taking a look at some optimisations, some of which are pretty large changes to how the game works. This could take some time to experiment with and we are going to have to make some decisions as to weather we need to optimise now or if we should hold out on it until after we have more content. This is an internal discussion we are going to have to have.

The Marker that would not go away was a separate but, I found the point in your video where it appears and it looks like you can place nest group markers if done on top of the tunnel. I have fixed this bug.

I am going to have a think about the wood ants. I do know what you mean they seem to keep walking when they should be firing. Could be a bug or could just be a quirk of the setup the ability has. Will look into it. Also they should not hit the lip of the tunnel like it appears they do.

Attack waves tanking the FPS makes sense as it is a load more creatures on top of the ones already on the map. I will be putting some thought into this also. I could have it that later on it favours fewer more powerful creatures, or spawns higher level ants.

Regarding the trails and the ants doing nothing at the start of the game. I will explain how they work and why this is happening.

Firstly always half of the ants assigned to a trail will remain at the trails end, the other half will form the trail to and from the start point. If your marker is underground the start point is the queen and if overground it is the surface where your tunnel comes out.

Underground the ants waiting to join the trail will wait by the queen and will be sent to the end point at a rate that should cause a constant flow of ants to and from the end which is based on the distance the marker is from the queen. Overground it will do the same only the ants will wait by the tunnel underground and the distance is based on the distance from the surface tunnel to the marker.

This system has worked well in the formicarium levels where we dictate the distance to the surface tunnel, however giving players the freedom to have a long run around underground in freeplay is leading to some ant behaviour that is not intuitive to the observing player.

We may need to revisit this in light of freeplay. Anyway thanks for the comprehensive test it has given me a lot to work with.

 

[JohnSlugDisco]

I have been investigating wood ant firing. Bare in mind this patch did not change how this works but I will explain how it is set up. Currently the ants "think" every 0.4 to 0.6 seconds, and when they think they can target an enemy. Now in an open area this is fine as the wood ant only moves a short distance in the time between thinking so it is still far away when it starts to shoot, in fact this gives a bracket where the ant could stop allowing a space for more ants to squeeze in the back.

In the case of thin corridors like we see here in the woodlouse case the tiles stop vision until the ant is fairly close. By the time it is within line of sight the half second movement may cause the ant to be on top of its target by the time it starts shooting.

I have shortened the time between thinking (to 0.2 to 0.4s). Will continue to monitor this in case it causes FPS to worsen

 

[jaxmthemirc]

John,

Thanks alot for this very detailed response, sorry I forgot to check back into these forums so did not notice.

Very happy to help and it pleases me to know the feedback is being taken so seriously by you all!