JLJac on April 29, 2015, 12:33:00 PM:

Thanks for the kind words! Yeah, we've really gotten into a good synergy now

I don't really think there's any incentive to kill them, if they're dead they'll slowly lose their light whereas if they're alive they'll glow forever. It's maybe a little easier to hit them with a spear than just trying to catch them, but if there are a few of them they're generally not that hard to catch.

JLJac on April 30, 2015, 07:44:47 AM:

Thanks everyone and welcome to you first timers

@fall_ark, actually I imagine them as being unable to simply turn the light off - it does get dim when they're charging, but that's more because they focus their energy inwards somehow rather than actually muting it. Also, if they could turn their light off to mess with you they wouldn't be very useful as a light source, making it sort of impossible to progress through the game haha!

#Facial expressions, thing is that I actually want this thing to be cartoony, I just fail miserably because most creatures are huge gross monstrosities that don't really have a face at all, or have a solid expressionless face such as the Vulture. So whenever I'm doing a creature where I can actually show of some stylized expressions, I'm really stoked. We'll have to see, maybe at the end of the day the inconsistency becomes a bit too much and I'll have to tone it down. But so far I actually like the contrast - it's not an entirely uncommon trope in comics and anime that some characters (that you're supposed to identify more with) have stylized expressions, whereas monsters and other mysterious creatures have less stylized faces that are harder to read ~
    o.0    >.<    x.x    ^.^

Update 421
Starting to think about assets for the new dark region. This one we had to think quite a lot about, because the darkness has some particular qualities. Basically it's two conflicting factors. On one hand, the more little details, the better the shadows look. "Porous" tiles with lots of holes and stuff look good in the second layer, because you get to see the shadows move around in them which really displays the volume.

On the other hand, the darkness makes the level way harder to read. Mostly this is because the darkness breaks up the simple rule the rest of the game follows - dark = close, light = far away. Usually you can look at a room and assume that the dark parts are terrain you'll collide with, and the light parts are backdrop. The light sources in a dark room however are in the room - and in order for that to work visually I had to make it so that stuff in front of the light source were darkened so that they would look back-lit. This means that with the light source we don't just have a dark-light gradient, we have a dark-light-dark gradient.

This works in most cases, but it is admittedly not quite as clear as the ordinary palette. Also, it gets way worse the more little details there are, and the less solid the tiles. Soooo, the same factors that made the light look good.

My idea for a solution to this is that we'll be rolling with big, chunky blocks, but on the blocks there are protruding patterns that the shadows can play with. I think this will get both birds with one stone - because the general contours of the blocks are big and square you quickly get an idea of what's terrain and what's not, but if you look closer inside those big blocks of color there are smaller variations of shadow play.

Here are some sketches:



And here's a "final" (huge quotation marks there, everything always end up hundreds of kilometers away from the "final" sketches) concept:



Basically the area will consist of of big boxes, that are somehow bolted together with huge clamps. These clamps will run on the exterior of the boxes, and leave shadows. Then the boxes themselves will have a patterned surface, that will also get some shadows going. I'm currently mocking up some tiles, let's see how it goes!

JLJac on May 01, 2015, 01:04:12 AM:

Thanks and welcome!

@tortoiseandcrow, yeah, exacly the stuff I had in mind!

Update 422
A bit of a roller coaster today. I finally had a mockup of a protruding pattern tile, threw it in the game, and it looked... awful!

This gif is pretty huge so I'll here's a gfy instead:
http://gfycat.com/ScarySickBlacklemur

Also it looks awful so it might as well be behind a link XD

Basically what happens is that in the third layer, where shadow displacement is large, the fact that the shader isn't doing proper raytracing becomes evident as light is able to sneak in behind the pattern. This makes it look like the pattern is a free floating grate in front of the wall, rather than some detail on it.

I was kind of bummed at this point, and started thinking about whether I could go with some other solution. I didn't want to mess with the light shader, because the light shader was running at decent performance and looking good in all the other situations, so I felt I could pretty much only do damage.

After a bit of trying different stuff I decided to give the light shader a go after all. First I tried some sort of ray tracing solution, which would've brought the texture grabs up to like, 12 per pass, and looked proper awful. Then I tried another approach - which was basically to invert the highlight code. The shadows travel far and wide, but the highlights just ask a pixel one or two steps closer to the light source from the pixel being processed whether it's at a larger depth, and if so makes a highlight. I turned this upside down and created a "lowlight", basically a thing that asks the neighboring pixel if it should be casting a shadow at the current pixel. This is actually the same principle as the ordinary shadows, but it doesn't really scale with distance and is generally much smaller and subtler. Then I disallowed the ordinary shadows from being cast on surfaces that are close to them so they wouldn't intervene, and got an acceptable result:



This does admittedly introduce another texture grab in the shader, bringing it up to 4 or 5, but the shader still seems to be able to handle 50-100 light sources which I hope and believe is more than we'll need. Now back to art again!

JLJac on May 01, 2015, 04:32:49 AM:

Yeah the grating effect is cool, and we're gonna use that as well - but I also wanted the standard bumpmap look. Slowly getting there, I think - here's a comparison of two coloration models I've been trying:

http://gfycat.com/WebbedDentalErne

The lantern mouse behavior is because the fleeing AI is really, really, really bad. I cry tears of shame every time I think of it. I just can't figure out how to make a fleeing behavior using A* path finding -_- I'll definitely have to get back in there some time haha.

jamesprimate on May 01, 2015, 08:14:21 AM:

ha! i think that specific comparison was more geared toward me, who has a strong opinion of the differences of course. (left is vastly preferred for my purposes)

Since Joar is working on Shadow Urban tile stuff, and ive wrapped up what I can do with Garbage Wastes for the time being, im back at hacking away at the audio. (despite my upstairs neighbor noisily renovating his apartment!) I had promised a rough audio demo, so here is what we've got going on so far:





The volume is low here for audio nerd purposes (need that headroom!) You may notice its a much more naturalistic sounding direction from the more "gamey" audio of the lingo. There are a few "characteristic rain world" sounds, like the super jump, bat screeches and terrain impact "thud", that i kept from the previous version because with a little polish and contextual application they totally still work. and i like them!

From a technical standpoint its nothing terribly exciting, though we are probably doing 20x more sample instances and procedural implementation than a platformer usually has. basically every audio trigger (of the approximately 1 million that Joar set so far) has around 5-20 samples attached to it, and each of those samples is randomized by pitch and volume within a certain narrow range. This is mostly just because i get way too involved and micromanagy with this sort of stuff, but the world sounds much more alive and deep when you dont hear so many repeated sounds. i get completely taken out of immersion when i hear mechanically repeated sounds in game audio, so im stoked to get the opportunity to take the time and do it right.

anyway, another few hundred triggers, some polish and the same level of detail applied to the ambient audio, i think we'll have a pretty nice aural experience for you guys

here is an older lingo demo clip for comparison, and to slap me upside the head with how much progression has been made (warning the audio is much louder here):

jamesprimate on May 01, 2015, 10:51:32 AM:

Thx thx for the feedback! Yes the super jump and backflip sounds are definitely worth further consideration. Joar wants to keep some nod to the cartoony aspects, so that's where that comes from, though striking the right balance between "cartoon" and "nature documentary" is going to take some fiddling. We have yet to implement the ambient (atmospheric) audio engine, so I'm looking forward to hearing how this current soundset works in that context.

Audio design is a weird process. It's like erosion smoothing away the corners of rocks in a stream.

And speaking of water, to tortoiseandcrows comment, yes  indeed! we just discussed adding a set of triggers for impacts and etc in specific environments, so wet splishes, dusty fuffs, metallic clangs, etc. this will be layered with the impact sound just as you described.

jamesprimate on May 02, 2015, 07:49:40 AM (Last Edit: May 02, 2015, 08:22:51 AM):

ah this is fun, thank you! no one ever gets into debates about audio design philosophy around here 

in defense of le Mario Bros jumps, one of the reasons they remain to be used in many games is readability. in a sound environment like this where most of the mid-ground aural texture will be composed of various ambiguous white/pink noise elements (bat flaps, wind, gravel, crunching steps, water, etc etc), having a distinct palette for important moves is key, as it functions as significant "aural UI". To the player in a chaotic sound environment, a white noise "whoosh" is useless in a sea of white noise "wooshes".

also there is a point to be made about diagetic audio vs experiential audio. in my design philosophy, conveying the mood and a satisfying "experience" of the action / sound / music / etc. takes a slight precedence over some sort of cohesion ideal: just like with the visual UI debate, it is a game and sometimes there will need to be game sounds. the reason i kept this specific sproing from the lingo alpha is that it *feels really good* in-game, and i can say this comfortably without bias since its not "my sound". I just like it!

*BUT* these are much more general comments about these topics! I think in this case its very possible were just so used to hearing that "pounce" sound (after playing it for 3 years) that it can't possibly seem out of place to us. Our ears are TIN for this specific case! So if its sticking out then its sticking out, and i value your opinions, especially as you both went to great lengths to describe the context for them. You guys are the best!

In this version, that sound is a blend of the original alpha sproing sound tempered with a filtered white noise air woosh like Teod describes, so ill def try some re-leveling and blending of sporings keeping these opinions in mind. Im sure there is a balance where we can subtly satisfy our sproing lust and ALSO make it not sound like mickey mouse just dropped randomly into the screen for everybody else. And if not, other options will be explored!    

edit:

oh! i wanted to talk about this a little, as Joar had said something similar. with such a large world that we have going on, in my opinion uniformly coating everything with a similar aural texture would be a huge missed opportunity for immersion, especially since its so easy to implement a more dynamic solution! its those subtle differences that add so much IMHO

and ill be honest, there is no way my obsessive audio micromanagement tendencies would let such a thing fly XD

JLJac on May 04, 2015, 10:41:06 AM:

Good talk! I kind of agree on that jumping sound in particular, which might have something to do with me knowing the unprofessional circumstances it was created under haha! Think it was something like me blowing on my laptop mic and then messing around with a dynamic pitch bend in Cool Edit Pro Anyways, on the more general cartoony/naturalistic discussion, I think this is a really interesting topic. I won't lie, I think it could totally work out to just lean all the way over on the naturalistic side - realism can't really be wrong. It can be a bit boring however. The art and animation is always straddling this balance of nature documentary and cartoon - and though I know many are more interested in the nature doc side of it, I do believe that it would be worse off without the cartoon element.

It's like onions in cooking, no-one likes to just munch down on an onion, but it's not as simple as onions=bad and removing them makes the dish better. The onion does unexpected stuff when cooked, mixes stuff around, brings out the other tastes. Similarly it tastes better if you add salt, but if you add a lot of salt you hit diminishing returns. In the same way I think the cartoony elements in Rain World add some value. The natural first reaction is "ewww cartoony", but if it was removed it would be obvious that something was missing.

It would be cool if the audio could strike the same balance the animation is already attempting. It's a fine balance though, and might take some tuning. That jump sound might be like biting into a rather large piece of un-cooked onion, to return to the analogy. Stuff should obviously not be jarring, it should feel like a smooth blend, but a smooth blend with some unexpected undertones is more interesting than just straight up one thing or the other.

Update 422
I'm preparing for the big region connect/divide structural change where the game will actually become open world, one way or another. I want to know what I'm doing with this one, as it's a rather important piece of the under-the-hood infrastructure. In the mean time while getting ready, I've been doing some assorted messing around. Today I did a lot of small things - touching up the light shader, streamlining the player controls for some situations, fixing a few issues with the player animation, hunting down a few bugs, etc. Maintenance work. I have also been working on assets for the dark region, but you'll have to wait for James to start posting screen shots of that stuff. I think this one is going to be pretty interesting visually! Maybe I will do a little bit more misc stuff, but I feel like a game plan for the "let's make this an actual open world game" process is taking shape, so I'll get in there hopefully tomorrow or the day after.

jamesprimate on May 05, 2015, 10:21:20 AM:

OMFG so after a NIGHTMARE 2 days of trying to get kickstarters new update system to reliably load gifs, here is the messy result:
https://www.kickstarter.com/projects/rain-world/project-rain-world/posts/1220116

and now it turns out that old gifs on previous updates are broken as well. wats going on kickstarter 

JLJac on May 05, 2015, 11:25:46 AM:

 

Still, a good update! Two critters in one month :D

On my end, connecting regions! More news as I make some significant progress.

JLJac on May 06, 2015, 12:42:29 PM:

Region gates technically working, and I have connected the regions we already have. Still a million things that are wonky, a million things that should be added, and approximately 400 000 things that need visual polish or to have their graphics be created, after which they too will need visual polish. However, large leaps! Moving around the world between regions makes you realize it's really, really big already. I've been doing one Rain World tour of the regions we have without actually running through it, just jumping from room to room with the dev tools, and it still takes me like, 7 minutes to find my way through the maze. With enemies in, actually having to run everywhere and do the platforming challenges, and the rain cycle to think of, I think just getting from one region to the next will be quite an epic endeavor, probably spanning over several hibernation cycles. Still, we don't know a lot about exactly how it will play just yet - I literally hooked up the regions a few minutes ago. Hopefully this is where we will start to get a better grasp of what our game actually plays like though. My first impression is that it feels pretty different with the large connected world, which is cool!

jamesprimate on May 06, 2015, 02:58:49 PM:

just did a run of the quickest direct loop around the 6 finished regions (no side paths) with no creatures or rain and it took me, their creator who knows every shortcut, like 35 minutes O_O

jamesprimate on May 06, 2015, 03:19:51 PM:

12!

jamesprimate on May 06, 2015, 05:50:39 PM (Last Edit: May 06, 2015, 05:56:07 PM):

indeed! It's getting to the point where i wonder if soon we'll have to starting to think about spoilers here :/ Eh, we'll cross that bridge when we come to it.


sheesh. yeah, and to a degree i'll admit its partially overcompensation on my part.

with the game being so reliant on the emergent experience of a big, fussy web of interacting AI's, when we started out on this new version there was no real concrete way to know ahead of time if any of it would really work out and be playable. That was sort of a paralyzing thought when expectations got hyped and it was clear we'd have to make "a real-ass game" rather than just some little ecosystem toy. So I thought at the very least we could make sure that there would be Content. Then worst case scenario if the interactive AI ecosystem experiment didnt work out, people would still be getting their money's worth in exploration/adventure game. And also, I mean, it is called Rain WORLD not "Rain Area."

At this point i think we're both pretty happy with how the experiment is working out, so hopefully the large scope is a bonus that will add a lot to the replay-ability and NG+ aspects. As its open world without much in the way of arbitrary "go here to find blue key" type of barriers, the player isnt going to be dragged through every room or anything, so i dont think the size will feel like a burden (to anyone but meeee  ) I would guess a player could very easily go through the game a number of times before seeing everything, and thats always fun.

Anyway, I've just laid the foundation really, and I cant wait till we get to the "Joar adds tons of weird stuff and set pieces" part :D

JLJac on May 07, 2015, 01:27:47 PM:

Update 423

Started animating the gates. Work still in progress, I want even more going on haha!



Oh, and yes James, I expect this thing to create insanely many sound triggers :D

jamesprimate on May 08, 2015, 01:52:48 AM:

yessss! i think the mockup i did of the shelter door audio for the last trailer had something like 30 individual sound events, so IM READY 

JLJac on May 08, 2015, 11:01:40 AM:

Update 424

Working on the gates and cleaning up a bit in the new multi-world infra structure. As every big structural change, it did bring a lot of bugs with it as many parts of the games suddenly find themselves in an unfamiliar environment. For example, I earlier had rooms indexed by their room number within the region, the first one being zero. Under the new system I didn't like this, as "5" could refer to the fifth number in any of the regions - so I changed the indexing to be spanning over all of the regions, and had the world translate world-general room indexes to region-specific indexes on the fly, causing everything to crash horribly because every place in the code where I was looping over the rooms in a region suddenly referred to some unloaded rooms in the SubUrban region. Well, got that sorted out though, thanks to the search function!

Other endeavors has included making the script that loads a region able to run over several frames, being kept alive with an Update call, rather than doing it all in one go. This is so that a region can be loaded as the game is running without causing a huge hiccup. I know you should probably use threading for this, but the little research I did in threading scared me away. One article was named "Simple example of threading" and started out "There is no such thing as a simple example of threading."  I've never before come across a phenomena in programming that can't be contained in a neat half-page example... So for now I'll just split processes over several frames instead. If I do learn how to do multithreading down the line, I assume I can put these split-up processes within the threaded class (if that's even how it works!?) and just call update on it in a loop till it's done.




On spoilers, how do you guys feel? Personally I would prefer to keep blogging, at least to some degree, as the devlog is good for my productivity. At least it feels so - writing down what you have done and have others read it confirms that you actually have done something. On the other hand I can totally see the point of it being a bit boring if every single little detail about the game is already known going into it.

I guess it depends on what your motivations for reading is? I myself read the Dwarf Fortress devlog, but I don't really play Dwarf Fortress - the project is more interesting to me conceptually than practically (you'd have to play for years to actually come across most of the phenomena described in the devlog), and with my interests geared like that the devlog is actually a superior medium to experience the project than the game itself.

On the other hand, a project that is more about the experience either as an action experience or an art experience would probably have way more impact as a game, and the spoiler discussion becomes relevant.

I can only speculate in what makes RW interesting to you guys, but I think that maybe it's somewhat in the middle of the two? There's some conceptual stuff like AI etc that might be interesting to other devs and enthusiasts, but the game is also a lot about the experience of exploration and encountering the unknown. So it's sort of straddling that line perhaps?

Let us know what your feelings about spoilers are! Obviously we are going to have a way more open discussion hanging out here than what goes out in the twitter feed etc, but the question is to what extent that should go. We could have a policy that the devlog is cards on the table, and if you want to stay unspoiled it's your own responsibility to stay away from here. That would make it possible for everyone to custom-tailor their experience, but it brings in a weird element of temptation. Another idea is that we could have a devlog black-out for a month or so, during which we will focus a lot on content creation. That way you'd still be able to follow the development of the game on a conceptual level, but you'd know that when you actually get your hands on it there will be at least some creatures and areas that can surprise you. What do you think?

jamesprimate on May 09, 2015, 06:22:29 PM:

good feedback and questions!


this sounds about right. think its worth a discussion on our end about where specifically to draw the line, but probably being vague about a couple of regions and a few beasties and omitting a few others that might drop too obvious of hints would do the trick. as so much of the game is about observation and creature interaction, spelling out "this does this" would be a shame, but thats probably easy to avoid as well.


interesting question! most of what we've done so far has been to organize the room geography, creature den placement, etc., in such a way that certain set piece-like outcomes are likely to occur. a simple example would be set-piece where you are first exposed to a new lizard type by having a pack of them chase you along a long bridge. easily done by having the lizards access to the player terrain be far enough away onscreen that the player has both visual alert and a running start before being in danger. it wont work perfectly, and sometimes not at all, but when it does it has a similar effect as scripting 4 lizards to spawn behind the player or something, etc etc. how feasible this is on a larger scale, we'll see. there are already plans for tools to make individual creatures more or less territorial, and then specify that territory, so that will definitely help structure set pieces while still being fuzzy enough to make the experiences unique.

a trade-off on all this that we have to be cognizant of is that there will certainly be people who get unfair / bad / boring experiences from us leaning on the cosmic dice roll, and they'll be like "this game is broken". so balancing the emergent with the static so its at least generally functional will be some WORK.

JLJac on May 11, 2015, 11:36:59 AM:

Hi! Hope everyone had a nice weekend!

So, from what I'm picking up here most people would like to keep the updates pretty much as they are. I think that means that we are not doing some total devlog blackout. Maybe we can walk some kind of golden... not middle road, but two-thirds-road? Like, I devlog pretty much everything I do, but if there is something small I think would be particularly fun to find out for yourself (and which is small enough to not be a big project that strangely keeps me quiet for several days), I omit it. Also, we want to include a lot of what we call "set piece rooms", which will be a special room with custom art and maybe a few lines of code that creates some interesting visual or interaction in that room. The reason for having these is that we have this huge world, and if you go exploring you should obviously be rewarded occasionally by finding something interesting. To me it seems like these could be a pretty perfect fit for something to keep quiet about - they're there specifically for the joy of exploration, and they are not mega projects that will take forever.

 
Number one and three here are sort of viable! For regions, it's James making those, and though he does use the devlog he doesn't as much as I am, so it would maybe not be too big a difference for him to just do a region sneakily without letting anyone know. Number three is sort of already happening a little bit, as I mess about with the creatures I sometimes add little features and interactions that never make it to the devlog. It's not that systematic (one per creature) but there are a few things going on with most creatures that are not on the devlog, small and very small. I could perhaps be a little bit more deliberate about this! For number two, a creature is generally at least a week-long project to implement, and the region to creature ratio is currently around 1:1... Meaning that this would basically be me making an entire new RW in secret, pushing the release twice as far and create devlog blackouts for half of that time haha! But good suggestions, I'll keep this in mind 

Update 425
Puuuuuuuh it's been a very region gate weekend indeed. Here they are:


(click to expand)

The central gate mechanism will look pretty much the same for all gates, but out to the edges there will be some room for James to add a little bit of individual flavor, and maybe hint at the regions the gate is bordering to. This one is Vanilla though, why it's just plain metal on the sides, real gates bordering to regions might look a little more interesting in those areas.

And a few gifs showing the walking through them. First of all, the gate closing behind you:



Mandatory sauna:



And passing through to the other side. Ignore the cheesy palette swap at the end of this, I still haven't entirely figured out how the palette swap between regions should work, so I just slapped on a ... palette swap. Gotta come up with something better for that.



Also, James and I have been talking about how these are going to work gameplay-wise a little bit. You might notice the water cistern up top - it's more visible when it's empty after you have passed through the gate:


(click to expand)

The gates are powered by the water in that cistern, is the idea. It's not visible in the gif because I had to crop it, but there is always water running as the doors are working to make this seem plausible. The gameplay implication of this is that you can only pass through a region gate once per rain cycle. You can technically go back to the same region twice, but you'll have to circle around and find some other way in - each specific gate only opens once. When the rain returns the water replenishes. The reason why we think this could be an interesting game mechanic is because it somewhat increases your stakes. When finding a new region you will naturally experience the excitement of exploration from entering a new area, and the idea is that this will be enhanced by making the experience a little bit nervous as well. Because with the gate shutting behind you, you don't have the option to just dip a toe in the new region and opt out if it wasn't to your liking. Instead you will have to find food and shelter (or at least a way back) in the new alien region. Most of the regions actually have a shelter and an easily accessible swarm room close by the gate, but maybe not all... So walking through one of these gates will really be equal to throwing yourself into the unknown. Hopefully this nervous feeling will be enhanced by the 40 seconds of gate mechanisms clicking and clacking and you being steamed like a dumpling, haha!

I'm really happy region gates are coming together, but there are still a couple of technicalities to be taken care of. One thing I have no idea how to solve is what to do with a creature that followed you into the gate room but ended up on the outside of the gate. Then we have a realized creature with an abstract representation that should actually be hanging out in another region from the room which the realized creature is in o__0 Going to be real interesting how I solve that one!

jamesprimate on May 11, 2015, 01:34:26 PM:

these gates are going to be rad. the shelters added so much to the mood, so its exciting to imagine experiencing these for the first time as a player, not knowing whats going on. will be so cool and moody! and im quite looking forward to sound designing the hell out of them :]

past few days ive been working on the first dark region, and this talk about spoilers has made me realize that pretty much all of the dark stuff on my end will be fairly spoiler-proof already. screenshots wont show much other than a white blob, so unless we want to do a video literally running through a room or something, it'll just have to be left to the experience. which is good! but also might make for some dry spots in my contributions to the devlog tho