Just saw something pop up on the marketplace called Turn Based Strategy Grids and got all excited thinking you’d made it to release. I was very sad to see what it actually was… Any chance for an ETA on release for this?
Hehe, I got excited there for a moment as well. That’s 's blueprint, and I think it’s a pretty cool solution as well. We’ve been working in parallel for a long time, and though I started before him he submitted it before me, so it’s only fair that he got on the marketplace first. I had hoped that my blueprint would be on the marketplace today as well, though. I’m still not worried that I’m going to lose a lot of buyers, because our solutions are pretty different from each other.
I’m through the voting process, and I submitted all my files to Epic last weekend. I haven’t heard anything from them since, though I just got a cryptic mail from the site that hosts the marketplace distribution agreement forms that the form I had signed had been voided. No clue why, though I guess I must have forgotten to sign something. I’ve mailed Epic about it without getting any answer, but then they’ve probably taken the weekend off. Epic will be doing maintenance on the marketplace next week, so my blueprint will probably be up in two weeks time. I’m kind of bummed out that I have to wait that long, but I’ve decided to see this as an opportunity, and will spend the next couple of weeks adding even more cool features before release and blow people away when it finally is released 
I’ve got some really cool stuff planned 
Waiting for cool stuff is fine with me!
I hope they’ll sort this out soon, I can’t wait to implement this in my project 
Ok, time for some of the aforementioned cool stuff I’m (literally) taking the plugin to the next level, as I’ve taken the pathfinding into “the third dimension”.
I had implemented the basics of height differences before, which was pretty easy. The tricky part was having the blueprint automatically generate areas of walkability based on the height differences between nodes, but now that too works like a charm. Making levels with height is still done through drag and drop and the units takes the height of the terrain into consideration when calculating visibility. I’ve thrown together a small skirmish using BSP and the default UE4 materials to show the new feature in action. I hope you enjoy
Wow man. Awesome job … Epic, approve this already and take my money. 8-}
Thanks! After watching the video again I really want to implement some ragdoll physics. Seeing those little guys tumbling off the walls would have been awesome. I’d leave adding in Wilhelm screams to the individual developers, though 
LOL … I was thinking the same thing. 8-}
I really like what I am seeing.
So - got a few questions for you -
- Is there any form of cover system planned?
- Will this work with Unreal Engine 4 terrain?
- So far we have seen just 4 v 4 units moving around and fighting each other. Is there any chance of getting larger amounts of units or groups of units?
To answer your questions:
-
There is already a system in place for making sure units cannot attack units in total cover, since it blocks line of sight. This can be disabled in games that do not use cover. If you’re thinking of something akin to XCOM where enemies can be visible, but have partial cover that make them harder but not impossible to hit, that’s a bit too game specific for what’s currently on my to-do list. It wouldn’t be too hard to add for a user, but I have to focus on getting more universal systems in place first.
-
It will, actually, but there are still a few issues here I’d like to adress before I would recommend this to users. I’m planning to add automatic grid height generation based on an underlying landscape in the future, but at the moment making the grid would have to be partially manual work. Also, the tiles that show where a unit can walk to do not currently conform to the underlying landscape, meaning that they would clip through slopes in the terrain. It would work, but it wouldn’t look very good yet.
-
Certainly. I’ve used smaller groups of units in the preview videos so that I can show a complete combat in a short amount of time. The amount of units you can use will be limited by their polycount long before there would be any slowdown due to the specifics of my plugin. It’s been a while since I’ve done any serious stress testing, but if you look at my old WIP thread the system ran well on grids with close to 100 units and 40000 tiles.
-
Good to know it will be possible for us to add the system on our own.
-
Is something like that also going to happen when a unit is walking on stairs? (I mean a model of stairs, not bsp stairs)
There is already a system in Place that calculates how many tiles there are between a pawn and its target. This could be used to create a hit percentage based on range. One solution for partial cover could be to have certain obstacles that block a custom trace channel, so if the regular visibility trace is not blocked but the other trace is you could reduce the hit chance.
Yes, it clips through stairs with steps shorter than the tile width at the moment. There are several possible solutions to this problem, and I’m currently trying to find out which one would be the cleanest and most efficient.
Awesome Job, man. Posting in here so i can’t miss the release of it. Also, i may have missed it, but can this calculate on random generate tiles in game, for exemple, in a reconstructive cave?
Do you mean if there is a built-in system to create random dungeons and similar? Not at the moment, no. My intention is to make the plugin all purpose with all sorts of turn based games, and procedural generation is quite genre specific. Procedural generation is one of my favorite things to make, however, so I’ll definately add an example at some point. Making your own procedural genration system shouldn’t be that hard either. Everything is already set up to generate maps based on arrays of tiles, so if you feed the appropriate data into those arrays you could quickly have some procedural generation going
This is looking fantastic. Good job!
Just a quick question. There’s nothing in your stuff that would present this being used in a iOS project, correct? You aren’t do any special rendering calls or such? Would just like to make sure. Cheers,
J^2
No, rendering is very basic. The only thing I know of rendering-wise that could be problematic is that I’m using semi transparent materials for the instanced meshes that display tiles in range, which is slow on some phones. Changing these to solid materials is trivial for the user, however.
I must add, though, that I have only tested my framework on desktop computers so far. While I don’t think there’s anything actually preventing the system running on other hardware it’s worth noting that while I’ve optimized the pathfinding thoroughly, calculating long paths is still a pretty demanding operation. I would therefore not currently recommend using the framework on devices with significantly less processing power than modern desktops and laptops, such as mobile phones; at least not if you want to calculate very long paths.
I’m currently working on a couple of options for the pathfinding that should make calculating long paths nearly instant even for much weaker hardware. I won’t be able to finish this until after release next week, however, as I have to focus on making video tutorials.
Well, I’ll be able to tell you within a day or two of the pack finally releasing how well it runs on an iPad 4 and iPhone 6 Plus.
J^2
Heh, ok, but you’re doing so at your own risk, and you’ll at the very least have to modify the camera and touch controls. I’ll start working on adding mobile and tablet support myself soon after release.
WOW. Impressive work indeed. Look forward to seeing it in the marketplace.