-
Ever wanted an RSS feed of all your favorite gaming news sites? Go check out our new Gaming Headlines feed! Read more about it here.
-
We have made minor adjustments to how the search bar works on ResetEra. You can read about the changes here.
Indie Game Development |OT 2018| Come, help, learn, show and tell!
- Thread starter Popstar
- Start date
- Status
FISH was at Glitch-Festa this weekend, a small expo in Curitiba - Brazil and I had the chance to watch random people play it, this was a great opportunity to gather feedback and see how people play it. One thing I noticed for instance is that most players were trying to drop fruits from the trees by shooting at them, so I decided to add that to the game, why not?
(gif was recorded with the brazilian portuguese version)
Anyway, everyone that played it seem to have liked, even though the game was pretty hard for people that are not used to old FPS, plenty of kids played and enjoyed it too, which was a surprise to me. Based on that I tried re-balancing the weapons and enemies and the easier difficulties should feel easier now. There's a demo available with those changes already: https://8ah.itch.io/fish-demo
(gif was recorded with the brazilian portuguese version)
Anyway, everyone that played it seem to have liked, even though the game was pretty hard for people that are not used to old FPS, plenty of kids played and enjoyed it too, which was a surprise to me. Based on that I tried re-balancing the weapons and enemies and the easier difficulties should feel easier now. There's a demo available with those changes already: https://8ah.itch.io/fish-demo
Thanks! :D looks like i need more than just luck at the moment lol
FISH was at Glitch-Festa this weekend, a small expo in Curitiba - Brazil and I had the chance to watch random people play it, this was a great opportunity to gather feedback and see how people play it. One thing I noticed for instance is that most players were trying to drop fruits from the trees by shooting at them, so I decided to add that to the game, why not?
(gif was recorded with the brazilian portuguese version)
Anyway, everyone that played it seem to have liked, even though the game was pretty hard for people that are not used to old FPS, plenty of kids played and enjoyed it too, which was a surprise to me. Based on that I tried re-balancing the weapons and enemies and the easier difficulties should feel easier now. There's a demo available with those changes already: https://8ah.itch.io/fish-demo
that looks pretty fun! good move on the fruit addition
You have to be way more specific. Maybe transport puzzles refers to something specific, but to me it could be anything. Do you mind sharing a little more about what your game is and what kind of puzzles you are looking for? Even if I can't find a resource for you, I could spit ball ideas to you.
You bokeh my heart! :D
Yeah, I've got some pixelized bokeh running for 2d sprites! As said last time,
my pixelized defocus isn't able to pull out bokeh (it is, but needs eons of
resolution unless I find a cool trick), so I switched to an explicit method
in tagging the pixels which should receive some bokeh, which can be automated
by looking for the very bright pixels in an image and use those (can be done
on a lower resolution). These pixels then undergo a convolution with a blur
spot having the shape of what ever aperture you want (circular used here).
The blur spot's size varies with depth and should match your defocus blur at
the given depth. The spot size can be computed with the help of the thin lens
equation. This will give some cool nice and soft approximation to some bokeh.
However, I needed it all pixelized, because a soft one won't match with the
pixelized stuff of mine. So I've programmed sort of a pixelixed convolution
and have matched the blur spots with the pixelixed defocus. Bam! Works! Sure,
some smooth bokeh would be cool as well esp. if the given object is strongly
defocused with the object being a gain blur and the bokeh be smooth and
bright giving this very nice contrast. That's something I will work on further
down, making sort of a transition from pixelixed to smooth bokeh.
Anyhow. In the animation above the orange pixel is tagged for bokeh, but the
red one up right isn't. As can be seen, when the sprite moves into defocus the
blur spot of the red one breaks down rather rapidly whereas the orange one
holds up tightly and blows up into a bokeh the more the sprite gets out of
focus.
PS: The sprite is due embalm.
embalm: Will stick with those a bit further. They don't suffer from
starvation that much. xD Yeah, keep going! :)
Last edited:
I started building out my game world; it's based on a town home / apartment community from google earth references. Its not that large of an area (its about 2sq kilometers), but I plan on creating a sewer system and modeling all (or most) the home interiors. I am hoping all that plus the smaller scale of the game characters will make the world feel large enough. Here is a quick scale/camera test with a couple of the outdoor assets I've been working.
Scale Test said:
Getting better. The pixelized blur spot now fades into a smooth one with
distance. Quite nice. But I don't know if that works good with multiple
layers. I also emphasized the outer rim of the blur spot somewhat to simulate
an uneasy bokeh (uneasy because it will produce odd patterns if there are too
many of them). Next I will try to layer them on different sprites and see how
the visibility plays out.
Getting better. The pixelized blur spot now fades into a smooth one with
distance. Quite nice. But I don't know if that works good with multiple
layers. I also emphasized the outer rim of the blur spot somewhat to simulate
an uneasy bokeh (uneasy because it will produce odd patterns if there are too
many of them). Next I will try to layer them on different sprites and see how
the visibility plays out.
yeah thats looking better mate! starting to work out after all
^ Better than the initial version. Yeah, lots of tweaks necessary. But I don't
like the smooth bokeh that much despite it would be superior for any smooth
blur but doesn't match with the pixelized version. And I don't think it will
work with multiple layers where you have, for example, a pixelized defocused
sprite in the front but a smooth bokeh from a strongly defocused sprite in
the back. That's not going to match, I would say.
Well, I like the following version much more despite it isn't up to par
representing the bokeh like the smooth version does but fits the pixel look
much better.
I need to work a bit harder on the the blur spot's pixel spread to better
approximate the smooth one.
like the smooth bokeh that much despite it would be superior for any smooth
blur but doesn't match with the pixelized version. And I don't think it will
work with multiple layers where you have, for example, a pixelized defocused
sprite in the front but a smooth bokeh from a strongly defocused sprite in
the back. That's not going to match, I would say.
Well, I like the following version much more despite it isn't up to par
representing the bokeh like the smooth version does but fits the pixel look
much better.
I need to work a bit harder on the the blur spot's pixel spread to better
approximate the smooth one.
Tinkering around with all that bokeh stuff I ask myself what's the take on
HDR sprites among 2d artists? For, it would be much easier for me to threshold
the bright pixels tagging them for bokeh. But a more compatible way would be
to use yet another texture marking said pixels. But an HDR solution could be
interesting nevertheless. Guess for example a sprite reflects some strong
light resp. bright objects with the bokeh adjusting accordingly. Hmm. Anyone?
HDR sprites among 2d artists? For, it would be much easier for me to threshold
the bright pixels tagging them for bokeh. But a more compatible way would be
to use yet another texture marking said pixels. But an HDR solution could be
interesting nevertheless. Guess for example a sprite reflects some strong
light resp. bright objects with the bokeh adjusting accordingly. Hmm. Anyone?
We want to make a new visual novel/rpg engine. We're not going to make a new game with the zombie code from Army 1. Trying to make something as visually effective as RenPy or Tyranno. We'd also try for Twitch support (modification, etc.)
So, we could either build this new engine in Game Maker or Unity, or go the insane option and go straight for Lumberyard, which already has built-in Twitch support. Thoughts?
So, we could either build this new engine in Game Maker or Unity, or go the insane option and go straight for Lumberyard, which already has built-in Twitch support. Thoughts?
So, I finally found the time to work a little bit on both that polygonal model of mine and the scripting side of things (which I enjoy much more honestly).
Regarding the model, I texture painted half of the trousers. It's not a masterpiece but I'm not entirely dissatisfied either. Anyway I'm happy that I'm making some progress and that I'm learning a lot of stuff in the process too.
Allow me to post another screenshot of the overall level that I've been working on, since I kinda like its vibe.
Regarding the scripts, I've reworked the way interactive objects work. Now I have one single abstract class which defines an abstract method Interact(). Then I have many subclasses, one for each interactive object, like UnlockedDoor, LightSwitch etc., which implement said abstract method. This way in order to deploy any interactive object I just need to associate the script of the appropriate sublcass with the game object. When the Player object gets near to an interactive object, I simply cast the object returned by the collider to an InteractiveObject and then I call the abstract method Interact(). Before these changes I had an absurd system which consisted in looking into a public String associated with the game object which defined the appropriate class (UnlockedDoor, LightSwitch etc.) and then switching using said String as parameter: each different case would result in a different cast to the specific object. Horrible.
Beg you pardon, what do you mean by engine here? Since you mention building it in Unity...
Anyway RenPy is opensource, no? Perhaps it could be a good staring point.
Regarding the model, I texture painted half of the trousers. It's not a masterpiece but I'm not entirely dissatisfied either. Anyway I'm happy that I'm making some progress and that I'm learning a lot of stuff in the process too.
Allow me to post another screenshot of the overall level that I've been working on, since I kinda like its vibe.
Regarding the scripts, I've reworked the way interactive objects work. Now I have one single abstract class which defines an abstract method Interact(). Then I have many subclasses, one for each interactive object, like UnlockedDoor, LightSwitch etc., which implement said abstract method. This way in order to deploy any interactive object I just need to associate the script of the appropriate sublcass with the game object. When the Player object gets near to an interactive object, I simply cast the object returned by the collider to an InteractiveObject and then I call the abstract method Interact(). Before these changes I had an absurd system which consisted in looking into a public String associated with the game object which defined the appropriate class (UnlockedDoor, LightSwitch etc.) and then switching using said String as parameter: each different case would result in a different cast to the specific object. Horrible.
Beg you pardon, what do you mean by engine here? Since you mention building it in Unity...
Anyway RenPy is opensource, no? Perhaps it could be a good staring point.
I apologize for the confusion. "Engine" was 100% not the right use of words and just makes me look like a klutz. A modification for visual novel creation is I guess a better phrase.
We want to move on from Ren'Py to something new. We've found that we are rather strangled by the current codebase that we have, and would like to add features that, at this point, RenPy does not allow.
We want to move on from Ren'Py to something new. We've found that we are rather strangled by the current codebase that we have, and would like to add features that, at this point, RenPy does not allow.
Ah I see. Well, I don't know about Twitch integration, but Unity would allow you to get the ball rolling easily, so to say. On the other hand, depending on the features that you are planning for the engine, it may be a little bit overkill since I imagine there would be some overhead involved in comparison to a more bare-bones approach, like building everything from scratch using a game library and then integrating Twitch using its api. Anyway if it is a visual novel you're planning to make, I'd imagine that the systems requirements would be quite low anyway, so using something like Unity sounds like a reasonable choice. In the asset store I found this plugin, which may help you with the Twitch integration, tough I can't vouch for its quality since I've never used it.
Not sure that'll fit what you're looking for, but there's also the Visual Novel maker, from the RPG maker team, supposedly coming on november 16th, that might be interesting to look out for?
Here's an old gif of the entrance to the second explorable area in my game, the Emerald Waters:
It's an old gif in the sense that the fish reaction icon has been replaced with a 16-bit looking one.
Fish is kinda aggressive, yeah :)
Looks fun! I like the art style choice for an FPS.
It's an old gif in the sense that the fish reaction icon has been replaced with a 16-bit looking one.
Fish is kinda aggressive, yeah :)
FISH was at Glitch-Festa this weekend, a small expo in Curitiba - Brazil and I had the chance to watch random people play it, this was a great opportunity to gather feedback and see how people play it. One thing I noticed for instance is that most players were trying to drop fruits from the trees by shooting at them, so I decided to add that to the game, why not?
(gif was recorded with the brazilian portuguese version)
Anyway, everyone that played it seem to have liked, even though the game was pretty hard for people that are not used to old FPS, plenty of kids played and enjoyed it too, which was a surprise to me. Based on that I tried re-balancing the weapons and enemies and the easier difficulties should feel easier now. There's a demo available with those changes already: https://8ah.itch.io/fish-demo
Looks fun! I like the art style choice for an FPS.
Ah of course, the amount of hours I've put into those games is borderline unhealthy; so yes I know exactly which part! (I may well steal the idea myself having seen yours!)
I'm still working away peppering Corsair's Cove with props that should help guide the player in the right direction. I'm also finding that very small props like grass and pebbles do wonders to break up the transition between bigger objects and terrain. Next thing to do is finish of the right side of the rear of this image and then the cave area!
Oh, that looks really nice. I love platformer/exploration/collectathon games and I just want to enter its world when I see one of them. Have you thought of using some kind of GI? I always like the extra color that it gives these type of games.
putting some things together
Nice art-style, looks lovely
oh hey.. this is shaping up well now :) the caterpillar is super speed!putting some things together
Thanks
Yeah somewhere along the line the caterpillar turned into a horse. I foresee going back later and slowing everything down when I make faster creatures
Hey guys, whats up!
I'm new to the development of games. I would like to know what is the best free tool available to create and learn about it. Is the free-trial version of GameMakerStudio 2 good for that purpose? I think for now, I'll stick to 2D games.
Thanks in advance
I'm new to the development of games. I would like to know what is the best free tool available to create and learn about it. Is the free-trial version of GameMakerStudio 2 good for that purpose? I think for now, I'll stick to 2D games.
Thanks in advance
You say you're new to game development, but do you have any programming experience at all? If so in which language? Also, have you read the OP?
I am a big fan of game maker (I only have version 1). I released my game on both Steam and PS4 and their support was super helpful when I ported my game to PS4. So if you want to release on multiple platforms it is really useful ( no Switch support yet though)
Their forum is really active and no question is too stupid.
If you decide to use game maker I recommend you learn to code and not use the drag and drop. It will be worth it in the end.
LET THE DESPAIR FLOW THROUGH YOU
I think i've settled on just vastly reducing the complexity of enemy types, and falling back on ways to reuse animations. This was something I specifically didn't want to do..
...but when it dawns on you that nearly every enemy needs stand, run, jump, fall, damaged, launched, floored, and stunned before you even start doing attacks....
I have 30+ enemies all with at least 3 different attacks, and at least 2 different death animations, plus staggers, walk etc.. most of them have over a 100 frames. (Only done with around 20 of them)
I know I am an idiot and that it is really an expensive way of making games. However, I am making a game I want to play myself and I really do not like games with too few enemy types.
I decided a static sprite for the drop shadow was kinda boring, so I made it more or less match the animation data of the character using the positions of the feet and head bones with different weights. It was kind of a chore to set up but it looks a lot better, so I'm happy with it.
Looking mighty fuckin' legit. Nice.I decided a static sprite for the drop shadow was kinda boring, so I made it more or less match the animation data of the character using the positions of the feet and head bones with different weights. It was kind of a chore to set up but it looks a lot better, so I'm happy with it.
May I ask what made it so difficult? I'm interested in doing a project with UE4 eventually.Holy crap the hoops i had to jump through to get this "working"...
Iunno if I'm gonna keep going with this project, making it function properly in UE4 is a nightmare.
For some reason the gif isn't working on my end so here is in case you can't see it. https://media.giphy.com/media/xT0xeAsSawkebnjtHG/giphy.gif
In UE4 moving an AI is typically just an "AI Move To" node with an end location, but that wouldn't work because of a number of reasons. The AI have trouble pushing past other AI so they get stuck, so I had to use physics on them and push them towards their location. But using physics on this many separate actors, all with their own draw calls and pathfinding, absolutely tanked the framerate. I tried maybe half a dozen different ways to fix this but eventually found one that mostly worked. The draw calls were the biggest problem so to address that I needed to make each mesh an Instanced Static Mesh object inside of one actor, and made each of their collisions into part of the same actor as well. To manage them all I had to put all of their attributes into an array, Collisions Array, Path Array, Location Array, Velocity Array etc. To command each one I used loops for each element of an array and marked which index was being modified and passed it to all the other arrays. When one of the enemies dies and the instance is removed, the remaining instances have all of their indexes shifted to close the gap which means i have to shift all of the other arrays to match (sometimes this just doesn't work when the events don't synchronize properly). To spread out the pathfinding I made it so only 10% of the group claculates a path and moves on each frame, so the full group would have calculated after 10 frames. This means each enemy responds poorly to corners and will awkwardly lurch forward every fraction of a second instead of moving smoothly. To ease the load on the physics calculations I turned the sub frame physics and velocity iterations all the way down to zero, so if the framerate gets too low sometimes they just bounce off each other and go flying or fall through the floor. On top of all this if I wanted to animate them I wouldn't be able to use skeleton animation or shape keys, I would probably have to bake animations into world space vertex transform textures and fade them in and out in the material graph to move from pose to pose which just sounds like a huge pain :|. The turrets work fine and were fun to program, I even got some projectile arc measuring and movement prediction type stuff in there to make them more accurate, but the horde is a glitchy mess that barely works and I don't know how to fix it. I gave it the ol' college try but it would take a much more experienced programmer than me to make this work.
ABZU seems to have done something akin to this on UE4 but that might as well be space magic as far as I understand it.
Thanks for explaining! I did something very simple but without AI long ago in the UDK:
I think what you've run into is that, as far as I know, game engines aren't typically designed to do AI and physics on a HUGE number of enemies at once. Occasionally someone will try to make a game that requires this, and as you discovered it may require very specific optimizations -- for example, creating a pool of enemies / resources / whatever, and reusing them instead of creating and destroying things. Instead of adjusting array indices, perhaps a single pool element could be marked as active (visible, alive, used) or inactive (invisible, dead, not used).
I don't really use design patterns, but if you're into that sort of thing you might read about the Flyweight pattern:
https://en.wikipedia.org/wiki/Flyweight_pattern
https://sourcemaking.com/design_patterns/flyweight
Long story short, it should be possible but yes it may require an experienced programmer (perhaps a version of you in the future). :)
Valve was asking I provide a tracklist for my 1 month project (as I was releasing the OST as a $1 DLC thing), I didn't want to for spoilers but they require it, so uploaded a preview.
This was a quickly done soundtrack for the shorter OST time, but I decided to share it here since I don't think I share my developments much here, though these are just quick previews of a few tracks:
This was a quickly done soundtrack for the shorter OST time, but I decided to share it here since I don't think I share my developments much here, though these are just quick previews of a few tracks:
I decided a static sprite for the drop shadow was kinda boring, so I made it more or less match the animation data of the character using the positions of the feet and head bones with different weights. It was kind of a chore to set up but it looks a lot better, so I'm happy with it.
That's awesome!
EDIT:
To continue the gifstravaganza, here's my own screenshot saturday from Enemies week:
Got four new enemy types in, which makes 6 total for now - enough to test and play around with for a while, so I'll go back to bug fixing and stuff, to hopefully get the prototype presentable by the end of the month for a local expo. NERVOUS!
Last edited:
Can't help, but looks cute somehow. Really.putting some things together
Clever! :)I decided a static sprite for the drop shadow was kinda boring, so I made it more or less match the animation data of the character using the positions of the feet and head bones with different weights. It was kind of a chore to set up but it looks a lot better, so I'm happy with it.
I decided a static sprite for the drop shadow was kinda boring, so I made it more or less match the animation data of the character using the positions of the feet and head bones with different weights. It was kind of a chore to set up but it looks a lot better, so I'm happy with it.
good idea, looks nice!
First post here, Been working on some ruins not sure what to add to the Background
Welcome to the thread, then :-D This looks pretty nice already! What are you looking for as far as background elements go?
Whatever you choose to go with, you're going to be using it for a sizable chunk of time; don't just go with a popular recommendation, or what you heard is easiest / best / most flexible to use - take some time before committing to try a bunch of different engines out, go through a few tutorials, and see what fits your workflow and style, and if you foresee any obvious problems with how that workflow can be expanded, or how nicely it plays with other programs you might use.
thanks everyone ♥
Looks really good! If you're looking on advice for the background, I think it'd be worthwhile to explore a different darker or more muted color for the water in the background. You have a really good color separation for the bricks in the foreground and background, but the water is all the same color palette.First post here, Been working on some ruins not sure what to add to the Background
- Status