JLJac on December 09, 2013, 01:25:03 PM:

Update 209
Added the ability to give items to pups. If you place an object on the ground in front of them, the one who's closest will go to the item and pick it up.

Pups will be able to use weapons in panicked situations, but likely it won't help them much, and the feature is mostly intended to be used as a way of carrying extra items. Now at least the pups are good for something.

In order to retrieve an item from a pup you can pick it up in the piggyback mode, and then hit the action button again. Then the item is taken from the pup, and put in your hands. If you put the pup down with the standard release item command, down+fire, it'll jump off still holding its item.

JLJac on December 10, 2013, 12:24:34 PM:

Hi and welcome!

The difficulty is actually the games' biggest issue right now, IMO. The game single player campaign doesn't really suffice for the type of game play you describe, where you get into a comfortable pace of finishing levels one by one - rather I want you to have to retry each hunt a few times before you make it through. The problem is that right now you have to retry waaaay to many times. I'm looking into ways of making it easier.

Multiplayer mode offers some replayability, I think. Me and my cousin have logged many many hours of competitive play during the development of the game, and we are not really getting tired of it. Which may be because the game has been developing and changing, of course

The thing with the pups is actually that they don't help you. They are there as my take on a "story", taking care of them is an optional side quest, and the only quest in the game except staying alive yourself. So if you don't want to be bothered, you can just ignore them. Actually an important part of the narrative will likely be that you can't save all three of them, but will have to prioritize. I think it would be interesting to have events like that, and the decisions are yours, ie not a pre-scripted cut scene.

Update 210
Misc pup stuff. What happens when a pup is off-screen holding an item? What happens when you enter a short cut, with a pup in the piggyback-position, and the pup is holding an item? What should it look like when a pup is carrying a spear? Etc etc.

jamesprimate on December 10, 2013, 02:52:30 PM:

having played the game a lot, i agree. even solo! the enemy AI is really challenging, so each new hunt is a different experience. you can't just "figure out the trick" and win. even if you are good, often the lizards will just straight up out-maneuver you. even now i die constantly and in amusing new ways XD

 

JLJac on December 11, 2013, 07:23:08 AM:

Hehe too easy isn't going to be a problem!

Actually the very thing you identify as a problem is the thing I want to achieve. Many games have tried to incorporate "moral choices", but many times it's just a question of choosing your flavor of progress, red or blue sprites. For example I remember bioshock having the options of "saving" or "harvesting" little sisters (my memory is a bit foggy as i didn't own the game, just saw it at a friend's) but both choices were beneficial. This is, I figure, because they didn't want to have the conflict between emotional stuff and the "play brain" as you call it. In real life though, as I see it, the very definition of a moral choice is going against your "play brain".


My idea is that this internal conflict in the player will enhance the mood of the game. The game describe a very harsh world, and choices such as "should I save these children or let them starve" are very harsh choices to go with it. If you are the "inputs outputs" kind of player, the option is always available to ignore them, though!

JLJac on December 12, 2013, 10:06:50 AM:

Thanks!

Heh, you just saved him, he was going to be replaced, but now I think I might keep him. The colors will be yellow, black and green-ish, as it stands.

Update 211
Worked some on the pup AI. Instead of being hardcoded into special behaviours for lizard, players, other pups the pups are able to have a few template attitudes towards other enteties, NEUTRAL, FRIENDLY, FLEE and HOSTILE.

This means that in a situation where you kill one of the pups, the others might consider you something to FLEE from or be HOSTILE towards, and then the same AI as when those behaviors are applied to a lizard can kick in.

Here's a horrible chaotic gif of me and the pups spawning, with the pups having the "HOSTILE" attitude towards me.



As you can see the green one is quick to grab a spear and kill me. The others are just trying to flee. You may also notice that the green pup doesn't throw the spear until there are no siblings between it and its target - me.

Now, a situation where the pups are hostile like this likely won't occur in the game, unless you murder one of the pups in front of the others while no lizards are even around. But the same behavior can, as said, be applied to lizards as well.

The pups have different personalities, and the green one is the violent, heroic and brave. When you find them he is holding a spear, and guarding the others. He always wants to carry a spear, and if he doesn't have one he'll pick up whatever weapons he can find around the levels. The HOSTILE behavior is more or less reserved for Green, he is the one responsible for protecting his siblings, and if the player betrays them or if a lizard is carrying a pack mate off towards its den he'll try to use his spear.

I've actually had him "rescue" me once, when a green lizard was carrying me away he lodged a spear in its back. Sadly I was already dead, otherwise that would have enabled me to continue playing.

The other pups are also able to use weapons, but only in panicked situations when a pursuing enemy is getting too close. At this point I don't know exactly what their personalities will be, but I'm contemplating having one of them sickly or injured, so you'll pretty quickly have to decide to get rid of it.

JLJac on December 13, 2013, 01:52:12 PM:

Thanks people!

Yeah, they should be able to get past each other easier. I'll look into a "walk above" solution. Thanks!

Update 212
Today I finished(?) the "social AI" as I call it, a horrible thing that's way too complex for a game like this. Theoretically it's able to simulate shakespearian dramas such as taking pups 1 and 2 off to the side, place 1 in the sight of 3, who's a little bit to the side, killing 1, but make sure that 3 thinks that 2 were closest to 1 at the time, making 3 hate 2.  Because of bad communication between the game and the player this whole system will basically look like "a weird bug happened, where one pup suddenly became aggressive" but it was fun to make lol.

Each pup has, for each other bear creature, a value -100 to 100 for how much they like that creature. They also have a temporary "like" value, on the same scale, that slowly adjusts to the base value. This is used for a situation like if you throw a rock at a pup - then they will all get a very negative temporary value towards you, meaning that they'll flee. But the base value is only lowered slightly, meaning that after the initial shock settles, they'll come back to you and be your friends again. But the base value is still lowered, so if you throw a lot of rocks, they'll eventually lose their temper and start to avoid you for real.

The relationship will only be able to heal (by throwing up food, rescue from lizards) until a certain degree of hostility, after which they will never trust you again. But that is more or less reserved for spearing them to death one by one in a room where no lizards are around.

By now we have an engine that can understand that a violent situation is going on, who is the victim, who is the offender, who is able to see the crime, etc. And, as I mention above, if they are not able to see the offender they'll make their own assumptions about who did it, which will probably not show up in the game, but it was fairly easy to implement, so well. They can also understand if an act of violence was an accident, by looking if there are enemies close to the offender or victim, and if so react less harshly to the crime.

If you commit an act of violence towards a pup, the other's will react depending on how much they like that specific individual. If they were enemies with it, they won't react as negatively.

It's a mess. But hopefully we won't see much of that mess, because it's not that usual for players and pups to hit each other with weapons - most of the times those situations are accidents, which the violence recognition engine accounts for. But if you are a crazy evil person, the game is able to understand that and react accordingly, which I'm happy with.

JLJac on December 15, 2013, 08:20:29 AM:

Yeah, the black could be the sickly one... only problem is that since it's the hardest to see, if it's also sickly, the choice of which one to dump first might get too easy.

The sickly one developing is interesting, but I won't ever be able to make them behave in a way that's beneficial to you I'm just not good enough for that kind of AI.

Update 213
As the pups fleeing behavior was just too pathetic (running into a corner an shivering there) and fleeing behavior is one of the things you might actually want them to be able to do, I updated it a bit. The big difference is that they are able to move between rooms by themselves.

For each of the room's exits, a dijkstra map is created, mapping the distance from each pup-available tile to each exit, using pup-possible types of movement:



When a pup is threatened, it looks at the dijkstra map values for the tile it's on, and the tile it's fleeing from. Then it can decide on which exits it is closer to than the enemy. If there are several, it can compare them, and go for the one where the difference in distance is the greatest. Then it's just a question of running! This is exactly the kind of fleeing behavior that the flies already use, but with hives rather than exits.

When reaching an exit, the pup object is put in stasis, and the pup is saved in the next room. Upon entering that room the pup is loaded, in a position that has roamed around the level a little bit, the number of steps determined by the amount of time since the pup was last seen.

When fleeing from players, this behavior is pretty fun. The pups run into another room, you follow, they run to another, etc.

If the pup is unable to reach any exits, or the enemy is between the pup and all reachable exits, it uses the old fleeing AI, which is horribly bad but at least enables it to move in somewhat the opposite direction. The bad AI running back and forth actually looks pretty cool in this new context though, because it only sets in when the pup is cornered, and it conveys a suitable feeling of panic.

JLJac on December 19, 2013, 12:55:30 AM:

I think I started at about fifteen, and now I'm 24, so that's...  almost ten years

Update 214
Behold the beauty and grace of nature:



Throwing up officially added!

jamesprimate on December 19, 2013, 01:26:40 AM:

AHAHAHAHAH 

JLJac on December 21, 2013, 07:55:25 AM:

Update 215
Ugh... One of those days.

James wanted me to make a t-shirt that could be used as a kickstarter reward, and because a t-shirt probably should have a logo on it I decided to revisit the text elements of the game. After all there has been some talk back and forth about that, so it could use a little improvement.

There is a little bit of a tribal theme going on in rain world, or tribal industrial you could probably call it.

Look at this level, for example. (Click for original size)


I kind of like this little bit of flavor, and would like to explore it more. What could I do with tribal/industrial?

So then I had this brilliant idea, that I would draw inspiration from mayan writing, mix it up with graffiti, and make "if mayans developed to the modern and post-modern age, did graffiti, and that graffiti was rendered through pixel art" kind of typeface. With roman letters. There's no way this could fail, right?

I started out being kind of happy with my work. It looks mayan, it looks a little graffiti and a little high-tech. Sweet, this is going to be awesome. I'm already planning all the new art assets for the game, including mayan runes in 3D and mayan high tech gargoyles to stick to buildings, etc etc.



Then I try to move the linework to photoshop and give it color. Now I encounter the problem. If I want it to be somewhat legible, I have to fill in the main letter shapes, and leave the little beads and dots and stuff as they are. Or at least somehow put the letters apart from the decorations. But. Filling in shapes takes away from their mayan look, because the mayan look is more or less based on it being only line work, no solid areas, and all the lines being of equal thickness. Finally I manage this:



And realize I've wasted two days  Contrast the old logo:

¨

and you'll see that it is both more legible, and cooler looking, which makes this new mayan adventure kind of meaningless. I haven't given the mayan typography all the finish I could've, but I think it's not really needed, it's painfully clear that it doesn't work.

Back to the drawing board.

JLJac on December 21, 2013, 08:27:11 AM:

True... Monochrome might look better, but I think it would be harder to read as the hue difference is meant to set the letter shapes apart...



 

JLJac on December 21, 2013, 10:04:26 AM:

And you guys are able to read it? Not like, decipher, but read?

This isn't really known, but there are a lot of indicators. They are able to use tools, but not manufacture them, and they don't seem to have language. I think religion is beyond them, actually.

jamesprimate on December 22, 2013, 09:53:09 PM:

i think its going pretty well too. just a thought, but i think the lettering could look really amazing in that distressed glowing neon style that you do in levels like this:


that, large, with the cat in front of it perhaps? NICE.

jamesprimate on December 28, 2013, 07:17:57 PM:

as a horribly underpowered netbook user, rainworld runs flawlessly for me even with 10 other windows open. fear not in that regard!

JLJac on December 31, 2013, 09:29:51 AM (Last Edit: December 31, 2013, 09:34:59 AM):

It's mostly running well enough, in some of the more extreme situations I can notice some slowdown on my machine. I think most people will be able to enjoy the game perfectly well

Update 215
It's been kind of quiet here, with the holidays and all - I hope to get back on track soon enough. Also James has been setting up the kickstarter, and I've been assisting him some in that. James said it might be a good idea to do a count down to opening the kickstarter, in order to get people hyped, so here we go:

THREE DAYS TO KICKSTARTING RAIN WORLD!

In other news, I've added the feature of jumping on other creatures. Preciously you could collide with other creatures, but not jump on them. Now, if you press the jump button while colliding with another creature and not being like, below it, you can jump on it. The jump will propel you upwards, while the other creature is inflicted with an equal force downwards, and in the case of lizard a tiny 3 frame stun. For lizards the downwards force is also boosted a little bit compared to what the mass of the two creatures would suggest (screw conservation of energy) to make up for the fact that they're most often fairly steadily anchored to whatever they're clinging to, and you didn't see much effect otherwise.

Jumping on a creature gives that creature your kill tag, but only if no other player has a kill tag on it already. This means that you can (in theory) get a kill on a lizard by jumping on it and making it fall to the bottom of the level, but only if no other player has recently damaged it. This is because it would feel kind of cheap if you speared a lizard and it's just lying there, not quite dead yet, and then your friend comes by and nudges it with his feet and that makes the kill count as his.

EDIT: Some answers to questions, sorry guys:
@DarkWanderer - the game can either be finished with a fairly small map, or, if the kickstarter goes well, be expanded to proper metroidvania size. We'll see what happens
@NicoM - get ready to laugh... Macromedia Director. It was the only platform I was comfortable with when I started two years ago, and this was just going to be a month long project or so anyways... There isn't exactly a demo for this current game, but there is a movement prototype from when only the player character was made, in which you can move around a little and get a feel for the controls. Someone re-uploaded it a few pages back, if you do a little digging and searching you'll find it.

JLJac on December 31, 2013, 04:42:09 PM:

Whoops, there's obviously a little delay on getting projects started for them to review the project, so we can't launch until the 7'th. Sorry!
ONE WEEK TILL KICKSTARTING RAIN WORLD

and yeah, we have been trying to make the game look appealing to newcomers as well!
Happy New Year everybody!

jamesprimate on January 01, 2014, 11:46:46 AM:

great info gents, we really appreciate any advice and especially previous experience! Thanks, and I'll be in touch shortly 

JLJac on January 01, 2014, 12:05:32 PM:

Wooow! Look at this, people!


It's super awesome that you appreciate the devlog so much. It started out as a thing more or less for myself to log my progress, and it makes me very happy that it has developed into this thing that's a source of enjoyment to other people. Now that I know how appreciated the devlog is, I will definitely try to step up and give it some more love myself - I'll try to keep it more consistent, informative and entertaining, and minimize the update slips and neglect it has seen at times.

I'm not really sure if you guys nominated it because of the devlog, or because of the game being devlogged. In either case I will try to make both the game and the devlog as good as I possibly can, in order to make this project enjoyable both prior to the release of the game and after it. I have noticed that many people are interested in technical details, and I've been lazy in answering those questions. That I'll try to change. Thanks again!

In regards to the kickstarter stuff, thank you all for helping out. What you write seem to make a lot of sense, and you guys seem to have a very good understanding of what makes those projects sink or swim. James and I have been dividing the work in a way where I focus more on the strict core of developing of the game, and he is taking care of pretty much everything else (not that I let him do his work in peace). So this input of yours is most relevant in relation to James' work - I don't really know much about this stuff - and I'll be interested to hear what he makes of it. Thanks for your engagement in the project!

JLJac on January 04, 2014, 08:57:45 AM:

Thanks!

Haha yeah a crossover would be sweet! I'll see what I can do

Update 216
I'm sorry I was all like "OMG I will never again neglect this devlog" and then turned straight to neglecting the devlog haha! The reason is James and I have been working some on a kind of minimalistic website, in order to have something to connect the kickstarter to. What do you think?

joarportfolio.com/rainWorld.html

Please don't post this around or anything, we kind of want to make a reveal later in order to get attention when the kickstarter is up.

jamesprimate on January 04, 2014, 10:59:42 AM:

no way! full version will be available to *everybody*. We do have some pretty exciting reward tiers planned though. At least, IM pretty dang excited for them.

And just for the record, the crossover idea didnt come from any of us, lol