[SUPPORT] Advanced Turn Based Tile Toolkit

[Monokkel]

Again, apologies for for the late answers. I’ve been serving as best man in my brother’s wedding in France and the past few days have been very busy. Things should return to normality very soon and I’ll be back to giving prompt replies.

@Selentic;748041

It’s possible, but I can’t think of anything I added they would have come from either. I was also chasing down another warning previously about a tile grid actor i think it was called, and the only information I could find was a thread you made last year ish on answerhub with the same warning where you said it was some ancient asset that had been out of the toolkit for years. Since it doesn’t say where the reference is coming from I can’t tell, but I can’t imagine what it’s from.

I’ll give a fresh project a check, but it’s not like it matters much anyway since as far as I can tell it causes absolutely no issues, just triggers me to see it in the log.

hey while im here, have a picture

https://imgtc.com/i/6TsD8jV.png

Looks fantastic! Looking forward to seeing it in motion.

I’ll need to look at this myself when I return home at the end of this week. I do no have any likely guesses for the error message without checking. Good that it does not cause any actual issues, but I will of course still need this find a fix to prevent message log spam.

@MatthewDouglas;748121

Hello @

It Seems like there is a minor error in your project regarding waking units that I was wondering if you can help me with some insight on where it is actually happening and maybe how to fix it. Its nothing major, but would like to remove it as it is annoying to see haha.

So this error happens if you one shot someone before they wake from the player. What seems to happen is that the enemy tries to wake itself, but since its destroyed, it has no reference to the enemy and some errors appear. The “!” still appears over where the enemy use to be and the sound goes off, but once I stop the project, that is when all the errors appear.

It looks like somehow the find wake units get called before the enemy is destroyed, so it is stored in the wake units array, but when it comes time to actually wake the reference is change to null because it was destroyed. I just can’t seem to pin down when it actually happend because sometimes the debugger on blueprints seems to not work and I can not see when things get called even though I know they do at some point (Debugger just completely skips and does not stop at breakpoint). I see that there is some code in the kill unit where it says if enemy’s health is less then 0, dont wake it. Unfortunately it seems not to work and ive tried putting that in many different areas.

Anyways, I figure I let you know, because I did alot of adjustments to the code, so to make sure It was not because I changed something I created a new project with just your template and recreated the scenario on the Jungle Raid Map (Same erros appear). I was hoping that maybe you could check it out and see what you think. Any Assistance would be appreciated.

Thanks

I believe your assessment of the cause of this issue is correct. I’ve fixed this bug in my internal build already. To fix this you will need to properly remove the reference to the unit in the woke units array when the unit is destroyed. You can use the OnDestroyed default event in the unit blueprint for this. You also need to remove the awake status effect actor. Also OnDestroyed loop through the status effects array of the unit and destroy all actors referenced in the array. I believe that should do the trick.

@AceD;749145

Hello @,

i have a few questions about your ATBTT:

1. Is it possible to use the ATBTT without pathfinding, and just using the WASD-direction-keys instead, for realtime grid based movement like up and down, left and right?
   The player should walk from tile to tile and push or pull objects from tile to tile
2. Is it possible to locate a single tile in the world, to check, on which tile the player or a mesh is located right now for undo movement purposes?

Thanks in advance!

Hello AceD,

sure, it should be possible to use WASD-movement. If your game has a locked camera view you could do something like this:

When pressing W, get the index variable of the player unit. From the vector field array in BP_Grid_Manager get the vector at this index - GridSizeX. Move the unit to this location through a lerp or other means. At the end of movement update the index of the unit to the index of this location (index - GridSizeX). For moving right use index + 1, left index - 1 and down index + GridSizeX.

If you have a rotating camera you will need to double a bit of math to translate the forward vector of the camera into a direction on the grid, but after that the process is similar.

For your second question, use the Vector to index function in BP_Grid_Manager to translate a location into a grid index.

@Bridgeburners;749244

Hi ,

Thanks for answering my previous questions.

I have another one, though. Is there any way to make the cost of moving to a tile truly zero? I think you set “0” as a shorthand for increasing the cost, somehow.

I ask because I created portals, and I want the cost to jump between portals to be zero. I did this by making it so that if two tiles are portals, then I add each of them to the other’s respective “edge array” and “edge array integer” elements. In those edge array element values, I set the cost to 1, so it takes one move point to jump from one portal to the other, which I don’t want. However, if I set the cost to 0 then, for some reason, it costs two movement points to jump the portal.

Hmm, good question. In older versions of the toolkit I used 0 to represent adjacent tiles that could not be moved between. Back then what you are asking would be difficult to implement.

With the current toolkit I don’t see why this should not be doable. With any such basic changed to toolkit functionality I have to be careful, though. So many other processes depend on the pathfinding function that it is som hard to predict the consequences of making such alterations.

I’m not sure what is causing edges with a cost of 0 to act like having a costly of two, but I will look into it. I think this should be solvable.

Again apologies for the delay in replying providing support. My life will soon return to normal and I’ll be able to provide quicker and more thorough support.