Welcome, Guest. Please login or register.

Login with username, password and session length

 
Advanced search

1411593 Posts in 69386 Topics- by 58444 Members - Latest Member: FightingFoxGame

May 07, 2024, 10:18:48 AM

Need hosting? Check out Digital Ocean
(more details in this thread)
TIGSource ForumsCommunityDevLogsWobbledogs
Pages: 1 ... 18 19 [20] 21 22 ... 37
Print
Author Topic: Wobbledogs  (Read 232959 times)
Wingz
Level 0
*


View Profile
« Reply #380 on: November 06, 2016, 12:45:42 PM »

The other thing that I wanna stress is that while I call these things "dogs" and the name of the game implies that they're "dogs", they're pretty much their own species and I've never intended to turn them into literal dogs. It's funny to me to refer to things as "dogs" even if they obviously aren't and I'm focusing way more on making these guys appealing and fun than I am on making them specifically dog-like.

---

So I've obviously been doing a little bit of visual exploration recently. I sat down and studied some Nintendo games for a little bit, specifically Super Mario Galaxy 2 and Super Mario 3D World, and learned some interesting stuff!

The main thing that stands out to me about the look of these games is that they're at once detailed and simplistic. The best way I can describe it is it's like you're playing with dolls, plastic bodies with real clothes and hair in miniaturized environments. Simple, readable shapes for models and large-scale texture detail, but fine noise textures on most objects (specifically in the case of the environment), and lots of detail in select places, specifically fur, clothes, particles, and (in the case of Super Mario 3D World) grass. Models generally also get a ton of plastic-y shine, and in some cases soft glow.



This image, while obviously promotional, is kind of a perfect representation. Characters look like hard plastic, but the cat suit itself has hyper detailed fur. It's kind of a beautiful contrast.

So yeah, idk. Most of this is probably obvious to anyone who's played these games but I'd never really looked closely before. I'm not planning on aping the style exactly but there's lots to learn from it!

On that note, I experimented a little bit with texture today and made a candy dog.

Also, I put in a new base expression that might be sorta familiar.

i think the book "Understanding Comics" could probably shine some light on the effectiveness of those super mario designs.
http://i.imgur.com/6F57xOc.png
http://i.imgur.com/O6s9jfP.png
sort of like relateability being directly related to how appealing a characters design is
if you condense a character to just its most important features, those important features define the character more and we understand the character more clearly (our minds like simple things because we understand them faster)
hope that might help

(i dont know how, or if its possible, to thumbnail images on this forum, so i just linked them to save space)
Logged
ActualDog
Level 3
***



View Profile WWW
« Reply #381 on: November 11, 2016, 01:22:49 PM »

The queen idea feels kind of strange. I'd feel better about a machine that combines 2 eggs into one.  Smiley

I definitely expect to do iteration with how fertilization works, but this is the easiest option by far so it makes sense to try it out first. I can also see some sort of combination approach where this is the default but you can get objects later on that let you take more direct control.

i think the book "Understanding Comics" could probably shine some light on the effectiveness of those super mario designs.
http://i.imgur.com/6F57xOc.png
http://i.imgur.com/O6s9jfP.png
sort of like relateability being directly related to how appealing a characters design is
if you condense a character to just its most important features, those important features define the character more and we understand the character more clearly (our minds like simple things because we understand them faster)
hope that might help

(i dont know how, or if its possible, to thumbnail images on this forum, so i just linked them to save space)

Yeah those are concepts I've been keeping in mind for this project, especially as I've been updating the look of the dogs. It's interesting stuff for sure!

---

Not a ton to say since last time. Adding a second liquid for sweat made me aware of some limitations of my current system so I ended up going back to liquids a bit and I've now got them mixing. As objects get come into contact with different liquids, they combine. The materials and physics properties slowly mix over prolonged exposure and although there are only two liquids in the game right now (sweat and grease), I want to have more and I think this'll give some fun combinations going forward.



Puddle graphics will almost certainly change at some point but it's not anywhere near the top of my priority list right now.

I've also started doing some concepting for revamped pen builder UI. This stuff is hard for me but I think it's gonna turn out pretty nice. It'll probably go through more passes later on, but I'm excited to have something less temporary in place! I'll show more once I'm further.

Oh, and I accidentally broke dog AI about a month back but I fixed that recently and now dogs can eat again, which is very exciting.


Logged

rj
Level 10
*****


bad, yells


View Profile WWW
« Reply #382 on: November 11, 2016, 06:41:44 PM »

the more alien these dogs get the better imho
Logged

Bricabrac
Level 2
**


Fail again. Fail better.


View Profile
« Reply #383 on: November 13, 2016, 06:53:40 AM »

Quote
The idea right now is that you can designate a dog as the "queen". If an egg is laid while there's a queen around, it'll automatically get fertilized by it and essentially result in a crossbreed dog (or at least the chance for a crossbreed dog). I'm not sure yet how this'll feel in the long run, but for now it's at least nice to have SOME way to crossbreed dogs through the game itself without using debug features.
Think it would sound more natural if you call the designated dog a king and not a queen; it works this way for lots of animals, like fishes: they make the eggs, and then the male wanders over and fertilizes them.
Speaking of fishes, clown fishes are able to change sex from male to female to fill the colony's needs.

Quote
although there are only two liquids in the game right now (sweat and grease), I want to have more and I think this'll give some fun combinations going forward.
orange/blue gels from Portal 2, please.
Logged

Selling Sunlight - Wandering Merchant RPG
TitoOliveira
Level 2
**



View Profile WWW
« Reply #384 on: November 13, 2016, 08:26:16 PM »

Hah! Nice to come back here and these guys look like actual dogs now
Logged


flipswitchx
Level 3
***



View Profile WWW
« Reply #385 on: November 13, 2016, 11:07:33 PM »

Love the new looks of the 'dogs'.  Beer!
Logged

ActualDog
Level 3
***



View Profile WWW
« Reply #386 on: November 18, 2016, 06:29:42 PM »

the more alien these dogs get the better imho

Thanks yeah that's kinda the idea. I ultimately want them to be similar enough to dogs that you know roughly what to expect but different enough that you feel like you're learning about and discovering a new species.

Think it would sound more natural if you call the designated dog a king and not a queen; it works this way for lots of animals, like fishes: they make the eggs, and then the male wanders over and fertilizes them.
Speaking of fishes, clown fishes are able to change sex from male to female to fill the colony's needs.

I get your point but like, they're definitely queens. I can't just change the way nature operates.

Hah! Nice to come back here and these guys look like actual dogs now

Love the new looks of the 'dogs'.  Beer!

Yay! Glad you like them!

---

I've been a little quiet because I've been working on a Construction GUI overhaul, which doesn't really lend itself to progress reports. That said, I finally got most of it in, which feels so, so good.



There's still some stuff I need to do to finish up this pass, there are some additional features I know I'll need, and I kind of expect this to undergo further iterations as time goes on, but it's still a pretty big step for me to get this much working.

I also did a few goofs.




« Last Edit: November 18, 2016, 07:26:21 PM by ActualDog » Logged

io3 creations
Level 10
*****



View Profile WWW
« Reply #387 on: November 19, 2016, 10:16:09 AM »


Based on this zero-G environment, the "dogs" are definitely an alien species. Smiley

Just a question: have you thought of a tagline/elevator pitch yet?

Logged

JobLeonard
Level 10
*****



View Profile
« Reply #388 on: November 19, 2016, 10:35:32 AM »

I suggest just showing the gif of a wobbledog laying an egg. If that doesn't convince them they'd never get it anyway
Logged
io3 creations
Level 10
*****



View Profile WWW
« Reply #389 on: November 19, 2016, 11:14:43 AM »

I suggest just showing the gif of a wobbledog laying an egg. If that doesn't convince them they'd never get it anyway


That might work for the Goat Simulator approach
0:00 - 1:00
but some (e.g. journalists) will still want to know "What am I doing?"  Wink
1:59 - 2:40
as seen in:



Logged

ActualDog
Level 3
***



View Profile WWW
« Reply #390 on: November 21, 2016, 01:38:17 PM »

My idea of what the game is has changed a lot since the beginning of the project, though it's finally starting to solidify, so I haven't put a lot of thought into a specific pitch/tagline yet. I should probably come up with some stuff soon (and update the first page of this thread), but I'm not super worried about it at this exact moment.

"A genetic ant farm but with dogs instead of ants."
Logged

ActualDog
Level 3
***



View Profile WWW
« Reply #391 on: November 23, 2016, 07:18:22 PM »

GUI Hell isn't over yet, but progress is still being made. I've been fixing up the new build GUI and it's feeling fairly solid at this point. I've also started in on some sort of object info screen.



The idea's that picking something up will let you see relevant info about it and perform certain contextual actions like selling.



This is still very rough, of course. It'll be snappier in the end, I need to do work to ensure you never accidentally sell something (eek), and I'm not super happy with the specific flow yet but I want to get all the functionality in asap so I can get all my systems in place and refine. Next big step is that I gotta start designing a way to actually determine how much dogs and eggs are worth, which will be interesting.
Logged

marcgfx
Level 8
***


if you don't comment, who will?


View Profile WWW
« Reply #392 on: November 23, 2016, 10:49:38 PM »

GUI hell, hell yeah I understand that. It's astonishingly difficult to make it appealing and functional without being overloaded.

giving dogs a value might be quite difficult, maybe you can allow some online exchange between players. then the players would set the price.
who wouldnt pay thousands for something beautiful like this?
Logged

Zrcalo
TIGBaby
*


HOW DOES I ART?


View Profile
« Reply #393 on: November 24, 2016, 01:34:39 AM »

I know with me, as someone who exclusively plays pet breeding sims, adding a value is really good.
Any incentive to cull the population and reap awards is going to be the best installment.

re: flightrising's exalt feature, and the OWF on wajas. ..and well.. Gryffs.. you just delete them. rip.

best thing is, you can set up a value trade economy and then make tons of money through microtransactions by selling exclusive genes and breed abilities.

One of the pet sites I'm on charges $20 for a one time use gene item that applies a marking (not a color. just a marking.) to the pet. thing is, in the long run, that $20 could give you a massive advantage in the game, as you can breed out critters with that marking. The only thing you cant change is the color and sex of your critters in the game.. so you'll have to breed those out and hope you get the right gender to continue the lines.

(I'm talking about flightrising, but there's tons of other pet sims out there. they're just 2d and not interactive.)

this is what a basic looks like:



and this is what it looks like after $40 of your monies has been vaporized.



its a lot more complicated than that, but this is what 99% of pet sims are like online. 2d png art.


(also sorry if what I've been saying has been stated before in this thread.)

probably one of the best things you could do would be to just allow currency exchange with all the other pet sim sites. ie "its ok you can do it." and then they just trade money on their own. nothing needs to be installed for that. its moreof the whole "I'm on chickensmoothie (other pet site) and if I trade my brown hippo for my friend's spotted wobbledog I wont get banned on chickensmoothie"

I mean, if its a social game. if not, it can be silly like sims and be a really cute DIY sort of a deal where you can just simulate silly things. But I really like the idea of breeding and pet sims.

really
really

reaaaally
like it.

(thats all I play ever.)
« Last Edit: November 24, 2016, 01:56:07 AM by Zrcalo » Logged

Wobbledogs is my fandom.
io3 creations
Level 10
*****



View Profile WWW
« Reply #394 on: November 24, 2016, 10:15:49 AM »

"A genetic ant farm but with dogs instead of ants."
Having a queen makes sense now.
Logged

ActualDog
Level 3
***



View Profile WWW
« Reply #395 on: December 01, 2016, 08:05:55 PM »

giving dogs a value might be quite difficult, maybe you can allow some online exchange between players. then the players would set the price.
who wouldnt pay thousands for something beautiful like this?

I really really love the idea of online dog trading, but I'm already worried about having bit off more than I can chew and I just don't think I can justify trying to add an online/connected component. I'm aiming to have dogs be sharable some way (possibly through just their genetic codes, but maybe eventually with some more compact method), but it won't really figure into the game itself, so I gotta approach this economy thing differently!

...

Thanks for the breakdown! I wanna avoid any sort of microtransactions or online connectivity stuff (this won't be a social game), but it's still really interesting to see this stuff talked about; I know these games exist but I've never really played them. I might check some of these out later to see what they do in more depth.

--

Blergh.

I've been rewriting existing systems, which is always fun.

Just a warning, I really butcher my "genealogical" terminology here...

So, I was starting to design the dog value system and quickly realized that I was gonna need it to work very closely with the genetics systems I have going on. Not all mutations should be worth the same amount, and the value of a mutation probably shouldn't just be tied to how different it is from the base value. Most of my genetics code has been around since the early days of this project, or at least built off systems that were around back then, and it's all kind of difficult to work with and very spread out. It would've been possible to re-use everything, but there were some features I had issues with and over the long run re-writing this stuff is gonna be a net positive.

The main things I wanted this updated system to be able to do that the previous system could not were:

  • Have directly query-able genetic properties.
  • Be at least somewhat backwards-compatible.
  • Reduce complexity for adding new genetic properties.

Previous, I had a sprawling function with hard-coded generation values I had to update every time I added a gene. When I wanted to use this gene I had to unwrap it in the exact order I created it in and it was gross and finicky and I had all sorts of weird code to support this pattern.

The basic idea behind the new system is that I now define the gene like this.



There are various properties I can choose to change how each specific gene works. All gene types have a key (used for querying), a length, and the ability to auto-generate both positive and negative values. This last one is important because all my genes are modifiers, not direct setters. A gene of all zeroes needs to ALWAYS return a its minimum value, so for any genes that need to go positive as well as negative, I double them up and have both a positive and negative modifier. If I didn't do this, a single gene ranging from -1 to 2 would return a value of -1 when zeroed out. With my system, however, I have two genes, one going from 0 to 1 and one going from 0 to 2. Each returns 0 when zeroed out, and by subtracting the first's value from the second, I still have my full range.

Moving on, super genes are genes that have the ability to undergo super mutations. This is my solution for infinite properties. A gene that undergoes a super mutation has a bit added to it and a corresponding max value increase (defined by the "Super Add" property). When a super gene is queried, the game compares its current length to its expected length and returns an appropriate result.

Finally, looped genes are essentially genes that wrap around. They have a standard length and a corresponding number of loops (sometimes given through code, sometimes added through this tool). The gene's total length is its standard length multiplied by its loop number. Every time it's queried it increments an internal counter that moves the start index along. When it reaches the end, it loops back around. I use these heavily for patterns, along with a random genetic seed each dog carries with it, so they can generate near-infinitely without repeating. This is important because dog body sizes are unbounded.

All of these values are taken in by my new system when a dog is generated, and a gene is created from them. If a ready-made gene is passed in, the system checks it against these tuned values and fixes it up as best it can if it's missing anything. After all this, it goes through and maps out all the individual values so I can query them whenever I want.

The main issue is still backwards compatibility. Order is important because it affects how mutations work, whether certain traits are more likely to be carried along together or not, but once a gene is generated there's only enough information to figure out if it's missing properties, not to determine which properties are missing or if the order has changed. To that end, my genetic fixup code has no problem tacking on additional genetic properties, but absolutely cannot manage inserted properties or re-ordered properties. This is an easy problem to fix with a standard save system, but a hard one to fix when I'd ultimately like dogs to be sharable entirely through a single genetic string. For now I'm moving forward with the assumption that I'll just avoid re-org changes after ship, but it's something I'd like to come up with a real solution for at some point.

While making these changes I also discovered a few bugs and leftover debug paths that hadn't been removed, and the general flow is generally much cleaner and easier to follow now. I also finally went in and added the ability for genes to crossover multiple times during breeding. This means that it's now actually possible to get any conceivable mish-mash of properties when two dogs breed, whereas before crossover only happened once at most, so certain property combinations would never result.

After all this, I had to hook up this new system to my existing systems and rip out all the old stuff. While doing this I exposed some previously existing bugs with my leg generation systems. There was some very unfortunate stuff going on in there, just the worst. Really not good at all. After a near existential crisis trying to re-write this functionality, leg generation now works much better than before (though with a few caveats for feet I need to address later). I fixed a ton of issues with legs generating in uneven ways and with gaps between segments, and legs now also correctly enforce a minimum size.



I know this looks like chaos but you'll just have to trust me when I say this is all working exactly as intended.

Phew. Anyways, my reward for all this is I now get to go back to work on the dog value system I've been trying to start on since Monday.

Some misc gifs for your time.




Logged

JobLeonard
Level 10
*****



View Profile
« Reply #396 on: December 02, 2016, 01:57:42 AM »

Quote
I know this looks like chaos but you'll just have to trust me when I say this is all working exactly as intended.
I've followed this thread for long enough to be worried when things don't look like chaos.
Logged
whistlerat
Level 1
*



View Profile
« Reply #397 on: December 02, 2016, 02:31:46 AM »

It's incredibly fascinating to see that little snippet of how you define your genes, and to read the problems you've been coming up against. I have exactly the same issue regarding mutations and changing the size of a gene, and how to make sure that they can still reproduce with other members of the same species despite that. Right now I just flip the bits of my genetic strings for mutations and don't add anything because of this issue, because I've defined a kind of genetic interpreter which knows exactly how to interpret a particular creature's genes, but the moment I arbitrarily add or remove bits from the gene, my whole system falls apart and it would fail horrifically. Right now, my system would allow very easy sharing of creatures via their genetic string so long as both parties had the exact same interpreter I've not been able to decide how best to approach this, because being able to alter the length of genes - as you already know - would be really useful in being able to allow the creatures to mutate in a more dramatic fashion.

The only thing I can think of is to introduce 'stop markers' so that the interpreter no longer looks for 'a gene starting at index X and of length Y' but just simply 'the Nth gene', with the genes all separated by a stop marker. The gene could then be any length, except maybe 0. This should allow for compatibility between creatures with different lengths of genes.
Logged

Bricabrac
Level 2
**


Fail again. Fail better.


View Profile
« Reply #398 on: December 02, 2016, 04:33:24 AM »

Neat genetic stuff. You're my favourite mad doctor.  Gentleman

Quote
I'm aiming to have dogs be sharable some way (possibly through just their genetic codes, but maybe eventually with some more compact method), but it won't really figure into the game itself, so I gotta approach this economy thing differently!
What about passwords, or maybe pictures? The game could generate a "trading card" with some sort of infos that could be read from the software to re-create the dog in a image.
I've seen this method used in *coff* 3D custom girls *coff*, and I think Pico-8 also permits to share games by using cartridges/images.
It also evokes this playful feeling of exchanging stickers with friends in a playground.
Logged

Selling Sunlight - Wandering Merchant RPG
ActualDog
Level 3
***



View Profile WWW
« Reply #399 on: December 02, 2016, 05:47:11 PM »

I've followed this thread for long enough to be worried when things don't look like chaos.

<3

It's incredibly fascinating to see that little snippet of how you define your genes, and to read the problems you've been coming up against. I have exactly the same issue regarding mutations and changing the size of a gene, and how to make sure that they can still reproduce with other members of the same species despite that. Right now I just flip the bits of my genetic strings for mutations and don't add anything because of this issue, because I've defined a kind of genetic interpreter which knows exactly how to interpret a particular creature's genes, but the moment I arbitrarily add or remove bits from the gene, my whole system falls apart and it would fail horrifically. Right now, my system would allow very easy sharing of creatures via their genetic string so long as both parties had the exact same interpreter I've not been able to decide how best to approach this, because being able to alter the length of genes - as you already know - would be really useful in being able to allow the creatures to mutate in a more dramatic fashion.

The only thing I can think of is to introduce 'stop markers' so that the interpreter no longer looks for 'a gene starting at index X and of length Y' but just simply 'the Nth gene', with the genes all separated by a stop marker. The gene could then be any length, except maybe 0. This should allow for compatibility between creatures with different lengths of genes.

Glad it was interesting! I should probably mention that I do actually use "stop markers" for my super genes. It was the only way I could think of to get that unbounded growth and it works really well! That said, they still don't give me a solution for gene re-organization and insertion, so it's definitely not a silver bullet.

Thinking a little more about this, I suppose I could have backwards compatibility work with a sort of "version counter" gene that gets put on the front of all generated dogs. As long as it was fixed length and exempt from mutation, then every time I update the global genetics there could be code that looks at these version counters and goes through all the needed upgrades in order to arrive at the correct current version of dog! A little early for me to put this in since I don't have any dog data that needs saving, but I feel like this could end up working well.

What about passwords, or maybe pictures? The game could generate a "trading card" with some sort of infos that could be read from the software to re-create the dog in a image.
I've seen this method used in *coff* 3D custom girls *coff*, and I think Pico-8 also permits to share games by using cartridges/images.
It also evokes this playful feeling of exchanging stickers with friends in a playground.

Yeah I've thought a bit about alternative sharing methods, especially if genetic strings get too long! It'd be amazing if dogs could be shareable via a single tweet (which would mean either through an image or some sort of compacted string).
--

Tiny update before the weekend. I did some pipeline work today on the new gene stuff and I'm very happy with where it's at. I now have each genetic key get baked into a dynamically-populated enum (super simple c# code generation) when I update the gene, and all genetic properties are now query-able through enums rather than strings, which is way way cleaner and easier to maintain.

I also finally started in on value calculations. Right now I have systemic dog value increases based on leg and tail counts. Almost certainly a temporary implementation but all the systems are in place and working now, which feels great.

And here's this gif I took of a tiny tangential experiment I embarked on this afternoon that really didn't work out so well. Though good stuff to keep in mind for later when I inevitably add in genetic diseases.


Logged

Pages: 1 ... 18 19 [20] 21 22 ... 37
Print
Jump to:  

Theme orange-lt created by panic