Synnah

Ah, so Unreal's animation blueprints have two components to them; the Event Graph, which you use to control the logic of the animations and do things on a frame-by-frame basis, and the Anim Graph, where you set up a series of state machines to control how to move from one animation to another with conditions like 'when the walk speed is greater than 50%, transition from a walking animation to a running one' and 'at the end of this animation, return to the the idle animation'. In the Event Graph for my raven's animations, I have code for picking one of the head positions at random, but there still needs to be a state machine saying that you can move from any head position to any other head position, which is what that massive mess above is doing. I did think about just having two states, each of which selects a head position at random, but on the surface it looked like that wasn't possible. The system seems to be set up such that each state is a single animation, but I might be wrong about this.

Because there are two ravens in the game, but you control only one at a time (And because I spend a lot of time watching birds in my garden), I wanted to make sure they have an interesting idle loop when they're not moving. I still have faint plans for a game about birds at some point, so I wanted to prove to myself that I can animate a convincing bird. There's the base idle pose, the exaggerated breathing animation that all game characters need, and then I made poses for 11 different head positions. I blended these onto the final animation by just animating the head bone, and then set up a system for choosing a head position at random every second or so (When you switch from one animation to another in Unreal, even if it's just a single pose, you can set a blend time so it interpolates between the two poses). Of course, I then needed to make sure it was possible to transition from every head pose to every other head pose, which ended up looking like this: I'm pretty sure there's a much cleaner and smarter way of doing this, but it's done now. After I got the head movement working, I added in two full-body animations that play every 6-10 seconds. Those also animate the head bone, so I had to turn the head blending off when they are playing. And that's basically it! As far as I can tell, the only animations I have left are jumping, double jump, glide, falling and singing. I want to finish those off today, and then go back to building the rest of the mechanics, because I feel like I'm falling behind a bit on those.

Oh hey look. Need to animate the jump soon.

Just a quick update before I go to bed; today I worked on rigging and animating Mythalore's raven model. Here's how it looked after a bit of tweaking of the mesh, and then rigging it: And in motion, with a preliminary hop cycle: Not really happy with this (The feet need to be desynchronised, and there's too much bobbing on the head), but it'll do for the time being. Tomorrow I'll set up a basic animation network and go back to the movement code.

Hello! I wasted a lot of time yesterday trying to set up a Perforce server, but I couldn't get it working; I'm still annoyed about that, because while we're using BitBucket and it works, it would be nice to have solved the Perforce issue, for larger projects in future. Anyway, here's a GIF of my progress from today: (Those are ravens, honest!) Regarding the mechanics, we used God of War's node puzzles as a basis, but went for a kind of Lost Vikings twist. You switch between two ravens, Hugin and Munin (Odin's ravens, 'thought' and 'memory'); Hugin has more mobility (Double jump), and can sing songs that are used to open doors and trigger switches, but only Munin can remember the songs, and must teach them to Hugin. Basically, you move Munin into the optimal position to reach the goal before Hugin finishes singing. I don't think I'm explaining this very well, but it should become apparent soon enough. Mythalore and I are splitting the work as follows: He's doing 3D modelling and level building/design, and I'm doing coding/mechanics, animation and music. I want to do some interesting stuff with the music, where the songs that Hugin sings become part of the soundtrack once the node is lit up, so that a song builds up as you complete a level. So far I've got movement, jumping, double jumps and gliding working, as well as the ability to switch between the two ravens. Rather than use UE4's built-in character class, which is way too much of a black box to me, I'm writing all of the movement code myself, so I think it's going to take a bit of effort to make it feel good. Next thing I need to do is handle height changes on the ground, because at the moment a gentle slope is enough to stop the ravens dead.

Oh noooooo! Fixed it for you.

I 100% approve of this image. Also, probably going to make something for this jam, after skipping WJ6!

Augh, you're right! Delete my entire existence.

I wonder if Justin has ever read the Homestuck... EDIT: I just realised that before I posted this I had 420 posts, and now I've ruined that.

This is really great! It has just the right mix of difficult controls and rapid game loop that means that learning how to play well is something you can really feel happening. Lots of really clever design touches (Like the coloured glow that lets you know exactly which enemy is going to spawn where), and it just looks and sound great! I managed 34th on the leaderboards, with 3,955! Pleased with that.

I'm really into the style of this! The visual design is really strong. It's a fun, interesting concept too!

I had a lot of fun with this! It's a very slick tower defence game, and I love the visuals and the music. I enjoyed it more when I made my own level, which turned out to be quite a challenge (Because the enemy spawning is the same for every level, shorter levels tend to be a lot more difficult), and it actually forced me to figure out the best places to put the towers, and when to upgrade them. Like you say, there's no real balance at the moment, but you can kind of make your own balance, with the level editor.

Yeah, I liked this a lot! Things I enjoyed: - The visuals. - Immediately solving the 'frog door' puzzle as soon as I realised where the samples were from. - Having the sudden realisation that the music was a cover of The Wizard. - The description of Gob Island on the sign. - That skeleton animation at the end!

Sorry you got stuck on that level! Did you realise that the onion fits on top of the mushroom? I feel like it's a little fiddly to get the game to place it on top correctly, so it might be easy to think it's not possible and give up. Thanks for playing! EDIT: Oh I think I know where you might have got stuck; That's the first level where some of the vegetables remain in the basket; if you click on it, it'll move them to the chopping board so that you can use them. I figured that might be a problem, so my quick'n'cheap solution was to make the last vegetable to spawn a large one, so that you can see it through the holes in the basket! It may not have been enough.

Hey! I'd hadn't forgotten about this. On the contrary, I've been working pretty hard on it in the month since the jam ended! I'll probably follow up with a post about why it's taken me so long, but for now, here is the game! Download on Itch.io! I never really got to implement the horror stuff that I wanted; hopefully Veggie himself is horrifying enough. There are a couple of bugs I hope to fix over the next week or so.

I completely failed at getting the game into a playable state this weekend, and rather than submit something half-broken, I figure I'll hold off for the time being. I'd really like to have something ready for next weekend, but I'm going to be away from Friday, and I've got a bunch of other work that I need to do, so I'm not super-confident. I'll leave you with this, though:

Okay, let's see how Unreal's automated physics asset looks...

'It's me, "Veggie" Panino! Welcome to my kitchen.'

This looks great! I love the sprites and tiles, and the mechanics look like fun!

I'm continuing to spend far too long on modelling, as a means of putting off going back to trying to get my placement code working. On the plus side, I've got these: I'm really happy with this style. I haven't even started on "Veggie" himself, though. I've still got three vegetables left, which I'll get finished tomorrow, and then I can make a start on modelling, rigging and animating. The chances of getting all of that done before the end of Friday are really slim, but assuming I do then I've got Saturday and Sunday to cram the rest of the game in there. Yeah, I don't have a chance. Fingers crossed for the weekend after!

I thought the rest of my vegetable-placement code would be a cinch, but I've been stuck on it for several days now; I took a break today and made some models, but I'm going to have to go back to it at some point. Snapping vegetables to a grid on the panini, and then placing them, was straightforward. However, determining where to place vegetables when you move the mouse over vegetables that are already on the panini was way more complicated than I was expecting. I'm pretty close to getting it working as intended, but there are just a few things that are going wrong, and trying to fix them was giving me a headache. So I made these: Food preparation table! Chopping board! Panini! I spent a particularly long time assembling this texture out of photos of bread, and then I scaled it down to 128x128, so I could have been a lot less precise about it. I like it though! Going to make a start on some of the vegetables now, and maybe go back to the placement code after that.

I really like the mirror box GIFs! It's a really cool visual effect, but also really cool how well your raytracer handles all of the calculations involved.

Excellent, thanks!

Okay: I've implemented a grid system, and allowed you to place veg onto the panini and return it to the chopping board. The red box is a chilli. I swear these are going to look more like actual vegetables at some point. Next thing I need to do is start checking for overlaps, so you can't place veg inside existing veg, deal with placing veg on top of existing veg, and make a system for determining when a layer is completely full of veg.

These are all off the damn charts! Shout-out to the horse-bag, and also the mildly-irate Treasure Goblin.