• Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.

Indie Game Development Thread 3: Indie Jones and the Template of Doom

Dr. Buni

Member
Hi, folks. I have been working on a game for a while now, but I am taking my sweet time as I am learning as I develop it. So far I only have the few characters of the game drawn in regular art, not on pixel and stuff. Still, I thought it was about time to drop the temporary sprites I made and start working on the actual sprites for my game.

After considering my options, cost and scope, I think this kind of sprite (below) is what would simultaneously fit my project and be within the scope of my abilities. My question is if there is a term for this kind of pixel art?

ac8bc93e042ebbfbec82427eaac40181.jpg


Please don't say "ugly" xD. Thank you.
 

embalm

Member
Hi, folks. I have been working on a game for a while now, but I am taking my sweet time as I am learning as I develop it. So far I only have the few characters of the game drawn in regular art, not on pixel and stuff. Still, I thought it was about time to drop the temporary sprites I made and start working on the actual sprites for my game.

After considering my options, cost and scope, I think this kind of sprite (below) is what would simultaneously fit my project and be within the scope of my abilities. My question is if there is a term for this kind of pixel art?

ac8bc93e042ebbfbec82427eaac40181.jpg


Please don't say "ugly" xD. Thank you.
Cool.

I don't know the terminology, but I like the style.
 
Hey all, I'm curious about something my team and I have been talking about recently. Should of asked sooner since we have to make a decision tomorrow, but here goes;


How do you guys feel about showing off and trying to sell your game at an expo?


For some background, we were thinking about going to GDEX as an exhibitor. Main concern for us is money and just how effective going to an expo can be. We went last year and I feel like it didn't all that much for us. Granted, the game is a lot better now than it was last year, and we're in Early Access to, so maybe it could work out better for us on the second go around. At the same time, if it doesn't do much for us, then we put in the time and money for nothing at all. I want to go because I think it'd be fun to do and we'd get feedback right there from people playing, but I don't think it'll work out how we hope it will. Any thoughts and advice are appreciated!
 

Situacao

Member
If you open up your exported sprites in another program, like Paint.NET, Photoshop, etc. do they look correct or incorrect? Also, what file format are you exporting to from After Effects?

Yep, they look correct, it's just in Unity with the Linear color space that they look bad. They come out as PNG's.
 

Astrael

Member
Hey all, I'm curious about something my team and I have been talking about recently. Should of asked sooner since we have to make a decision tomorrow, but here goes;


How do you guys feel about showing off and trying to sell your game at an expo?


For some background, we were thinking about going to GDEX as an exhibitor. Main concern for us is money and just how effective going to an expo can be. We went last year and I feel like it didn't all that much for us. Granted, the game is a lot better now than it was last year, and we're in Early Access to, so maybe it could work out better for us on the second go around. At the same time, if it doesn't do much for us, then we put in the time and money for nothing at all. I want to go because I think it'd be fun to do and we'd get feedback right there from people playing, but I don't think it'll work out how we hope it will. Any thoughts and advice are appreciated!

I haven't got far enough to do this myself, but I plan to when I have at least a demo build ready. Even though you went last year to little benefit, I do think it's a worthwhile investment to be honest. Many otherwise smaller and/or niche games I've seen over the years that could have easily been overlooked (even in their own genre) got noticed because they did expos and shows, and I think the risk-reward is mostly geared in your favor if you have a display people can sit down and interact with like you had mentioned and eventually get word of mouth going. You said the game is much further along and better now so I think it'll attract more attention by that merit alone. Plus, people often go to these things to try out new games and it also helps to display your own seriousness for your project. You have faith in your game, then I don't see why you wouldn't want to get as many eyes on it as you can, and get people talking sooner than later (plus like you said, it could be fun on its own).

Maybe you can also try analyzing it economically, in terms of how many units you need to sell to cover the cost of the trip. If you think breaking even with that amount of games sold is not an issue and doesn't hurt overall sales profits too badly then I say go for it.

Well, I have no business experience so these are really just random thoughts. Good luck with whatever you decide. :)
 

Mik2121

Member
Hi guys!

I worked on another Game Jam at work and we each made something during our short summer vacation (mostly the last weekend of it).
I tried working on some small local multiplayer game for 4 players where you use light beams (the white beams) to reflect light and attack other players.

giphy.gif


The players have a shield that they can also use to defend themselves from other players, and their HP is shown in the center (the shiny circles).
I also tried to add something fun, so I made some prisms that you can hit and get your beam scattered in two directions.

giphy.gif

(You can see the red and then the green players using it)

What do you guys think? It's not too deep, but for a couple days thing I'm quite happy with how it ended up looking!

https://www.youtube.com/watch?v=voDwrh3u8ok
 

Situacao

Member
Okay, so it can be kinda a pain to debug this stuff.

In the Unity import settings, is sRGB checked?
Are your PNG files RGBA or RGB/greyscale using intensity as alpha?
If you run your PNG files through pngcrush to remove any color correction data does that change anything?
If you use a different file format, do you get a different result?

EDIT: what blend mode are you using in Unity?

Yep, sRGB is checked.
RGBA, I'm using alpha as transparency.
I'm gonna try that later, I'll report with the results!
What other file format can I use if I need the transparency?

I'm using the default blend mode for Sprites, as this GameObject contains a SpriteRenderer with the default sprites material.
 

missile

Member
Okay, so it can be kinda a pain to debug this stuff.

In the Unity import settings, is sRGB checked?
Are your PNG files RGBA or RGB/greyscale using intensity as alpha?
If you run your PNG files through pngcrush to remove any color correction data does that change anything?
If you use a different file format, do you get a different result?

EDIT: what blend mode are you using in Unity?

Yep, they look correct, it's just in Unity with the Linear color space that they look bad. They come out as PNG's.


Don't know if that works, but if Unity works in Linear color space, then you
may need to linearize the PNG sprites first upon loading, if they are saved
as sRGB;

Code:
sRGB_to_linear(float *a)
{
  for (int i = 0; i < 3; i++)
  {
    if(a[i] <= 0.04045f)
      a[i] /= 12.92f;
    else
      a[i] = powf((a[i] + 0.055f)/1.055f, 2.4f);
  }
}

but you could also use powf(x, 2.4f) or similar if performance is at stakes.
 

HelloMeow

Member
Hi, folks. I have been working on a game for a while now, but I am taking my sweet time as I am learning as I develop it. So far I only have the few characters of the game drawn in regular art, not on pixel and stuff. Still, I thought it was about time to drop the temporary sprites I made and start working on the actual sprites for my game.

After considering my options, cost and scope, I think this kind of sprite (below) is what would simultaneously fit my project and be within the scope of my abilities. My question is if there is a term for this kind of pixel art?

ac8bc93e042ebbfbec82427eaac40181.jpg


Please don't say "ugly" xD. Thank you.

I think it looks great. Reminds me of Hyper Light Drifter. I'd love to see more games with that style.
 
Yep, sRGB is checked.
RGBA, I'm using alpha as transparency.
I'm gonna try that later, I'll report with the results!
What other file format can I use if I need the transparency?

I'm using the default blend mode for Sprites, as this GameObject contains a SpriteRenderer with the default sprites material.
From here https://docs.unity3d.com/Manual/LinearRendering-LinearTextures.html
It seems if your sprites are exported in linear color space you need to have sRGB off, if they are gamma corrected you have to have it turned on.
Don't know if that works, but if Unity works in Linear color space, then you
may need to linearize the PNG sprites first upon loading, if they are saved
as sRGB;

Code:
sRGB_to_linear(float *a)
{
  for (int i = 0; i < 3; i++)
  {
    if(a[i] <= 0.04045f)
      a[i] /= 12.92f;
    else
      a[i] = powf((a[i] + 0.055f)/1.055f, 2.4f);
  }
}

but you could also use powf(x, 2.4f) or similar if performance is at stakes.
Unity should "degamma" your pngs already if you are working in linear and the textures are marked as sRGB space. I believe it does it on building rather than on loading, which wouldn't make much sense.

The code snippet is actually faster than powf(x, 2.4f) since it only has to do the expensive pow lookup if you are above a threshold.

If you want to speed up this algorithm you just need to precompute all 256 (if your colour channels are 8 bit) values to an array.
 

DemonNite

Member
I've been tidying up the House a bit more by adding details such as coving and panneling

6HgERuFh.png


Also started to fill out one of the bedrooms in the House (master)

XnSWQVQh.png


and heres a look at the workshop room found at the back end of the House via the Kitchen

UKl96zCh.png


One of the longest (tedious) things to do though had to be connecting up all the light/fan switches to all the lights in the House. It almost felt like doing it for real lol
 

missile

Member
From here https://docs.unity3d.com/Manual/LinearRendering-LinearTextures.html
It seems if your sprites are exported in linear color space you need to have sRGB off, if they are gamma corrected you have to have it turned on.
Would had been my next post, just wanted to exclude my case.

Unity should "degamma" your pngs already if you are working in linear and the textures are marked as sRGB space. I believe it does it on building rather than on loading, which wouldn't make much sense.
...
Would had been my second post, just wanted to see if what I've written fixed
anything. If not, it must be the inverse problem/solution.

... The code snippet is actually faster than powf(x, 2.4f) since it only has to do the expensive pow lookup if you are above a threshold. ...
Will be faster in some scene (when below threshold), sure, but for scenes
with most of the pixels above threshold the if-statement has some cost depending
on the architecture.

... If you want to speed up this algorithm you just need to precompute all 256 (if you are in an 8 bit color space) values to an array.
Yep. Isn't an option for me (working in 32-bit colors space).
 
Would had been by seconds post, just wanted to see if what I've written fixed
anything. If not, it must be the inverse problem/solution.

Will be faster in some scene (when below threshold), sure, but for scenes
with most of the pixel above thesholds the if-statement has some cost depending
on the architecture.
I find it hard to imagine that an if check will have a significant overhead compared to a pow call in any architecture. A lot of channel values will be below that threshold.
Yep. Isn't an option for me (working in 32-bit colors space).
I meant 8 bits per colour channel, you surely don't have 32 bits per colour channel?
 

missile

Member
I find it hard to imagine that an if check will have a significant overhead compared to a pow call in any architecture. A lot of channel values will be below that threshold.
...
Say pow is computed in hardware with the architecture not featuring branch
prediction or may have one and may miss-predict the branch depending on the
incoming data. May add up (the stall).

... I meant 8 bits per colour channel, you surely don't have 32 bits per colour channel?
Currently, yes (homebrew), but will cut down later on depending on the
quality/precision needed. It's for developing some HDR algorithms (chromatic
adaptation and stuff).
 
Say pow is computed in hardware with the architecture not featuring branch
prediction resp. may miss-predict the branch depending on the incoming data.
May add up (the stall).
So an architecture that is deliberately designed to prove a point? Your "optimization" had a worse algorithmic speed and was less accurate.
Currently, yes (homebrew), but will cut down later on depending on the
quality/precision needed. It's for developing some HDR algorithms (chromatic
adaptation and stuff).
Bonkers, are you using textures with that colour depth, how are you creating them?
 

missile

Member
So an architecture that is deliberately designed to prove a point? ...
Not really. Just wanted to point out that such can be the case, graphics
hardware comes to mind. However, even for normal arichtectures the if-statement
can be a problem (well my HDR data is way beyond that threshold most if the
time) considering compiler optimization, i.e. the if-statement is a dependency
and may influence optimal code generation for a couple of architectures resp.
compilers.

... Bonkers, are you using textures with that colour depth, how are you creating them?
I think you should have anticipated by now that I generate the data myself. ;)
Edit: from math, algorithmically
 
Not really. Just wanted to point out that such can be the case, graphics
hardware comes to mind. However, even for normal arichtectures the if-statement
can be a problem (well my HDR data is way beyond that threshold most if the
time) considering compiler optimization, i.e. the if-statement is a dependency
and may influence optimal code generation for a couple of architectures/compiles.
But so is the pow call, it's famously massively expensive, and we were talking about a specific case.
I think you should have anticipate by now that I generate the data myself. ;)
Well then it isn't really relevant. We were talking about importing a texture.
 

missile

Member
But so is the pow call, it's famously massively expensive, and we were talking about a specific case. ...
The code issuing sequence might be different without the if-statement leading
to a perhaps better fill of the (dual etc. issuing-) pipelines. I don't think
it's such a clear cut. And depending on most of my HDR data I generate, the
non-if (simple powf) version is a bit faster because the jump turns out to be
wrong most of the time. However, for random data it may be the other way
around.

Well then it isn't really relevant. We were talking about importing a texture.
You just wanted to know.
 

Situacao

Member
Thank you for all the answers!

From here https://docs.unity3d.com/Manual/LinearRendering-LinearTextures.html
It seems if your sprites are exported in linear color space you need to have sRGB off, if they are gamma corrected you have to have it turned on.

The thing is, I tried turning it off. Doing this, however, only made the texture look brighter and ended up not removing the artifacts and color banding.


Don't know if that works, but if Unity works in Linear color space, then you
may need to linearize the PNG sprites first upon loading, if they are saved
as sRGB;

Code:
sRGB_to_linear(float *a)
{
  for (int i = 0; i < 3; i++)
  {
    if(a[i] <= 0.04045f)
      a[i] /= 12.92f;
    else
      a[i] = powf((a[i] + 0.055f)/1.055f, 2.4f);
  }
}

but you could also use powf(x, 2.4f) or similar if performance is at stakes.

Just to be sure I understood, the float "a" is each pixel of the texture?

You can try TGA as an alternative to PNG if needing transparency.

Roger, will try that too!
 

DemonNite

Member
did a small test to see if I could animate the skybox by switching out textures and adjusting the exposure

WickedVengefulChanticleer.gif


I may revisit this as a "nice to have" when I have nothing else more important to do
 

Situacao

Member
Ok, so I'm back with the results of the tests you guys suggested:

- Changing the texture from PNG to TGA and running it through pngcrush both did no good. The sprite came up with the same color banding as before.
- Running it through missile's function did give me some results, even though it didn't fully solve the problem. It's much better, but you can still clearly see color banding on the borders. Is there any way one can still improve this?

Here are the 3 versions:

Linear
Ra2VipH.png


Gamma (Correct version)
conKGmf.png


After sRGBtoLinear
y3G9got.png


EDIT: The forum kinda altered the look of the sRGBtoLinear version. If you Right Click+View Image, it will look like it does on the engine.
 

Darklor01

Might need to stop sniffing glue
A buddy of mine with lots of experience working in the game industry decided to try and make a solo indie project. It's an FPS with elements of Dark Souls, Borderlands, and Doom for example. He has worked on Subnautica, Warframe, Call Of Duty Blackops, was the lead Dev on the official Doom 2 expansion No Rest For The Living.

Thought you guys might like to give it a go. There is a free demo download with some decent length. There is no current option for purchase since it is very much a work in progress.

https://steemit.com/gaming/@thecast...the-first-demo-build-of-my-game-project-stray

ezgif.com-video-to-gif%20(19).gif

ezgif.com-video-to-gif%20(18).gif

ezgif.com-video-to-gif%20(20).gif

ezgif.com-video-to-gif%20(23).gif

ezgif.com-video-to-gif%20(24).gif

ezgif.com-video-to-gif%20(31).gif
 

Blizzard

Banned
You're wrong. Fuck off.

(I normally ignore spammers but not only could you not be bothered to do more than cut and paste the entire post in multiple threads, but you also included over 100MB of GIFs. For fuck's sake.)
Good points. Also, considering it involves a lead developer from a DOOM game, it looks massively derivative to the point of a cyberdemon in that last GIF.
 
This is one of the threads I keep returning to.

Generally, I'm not as excited about the current day AAA output, as I once used to be. However, whenever I glance at some of these projects that Independent GAF are working on, I sense an essence that most current AAA games seem to lack.

There is a sense of a genuine drive from whats showcased here, regardless of scale.

I sincerely hope you all stay true to yourselves and stay committed to what you've initiated.
 

-COOLIO-

The Everyman
This is one of the threads I keep returning to.

Generally, I'm not as excited about the current day AAA output, as I once used to be. However, whenever I glance at some of these projects that Independent GAF are working on, I sense an essence that most current AAA games seem to lack.

There is a sense of a genuine drive from whats showcased here, regardless of scale.

I sincerely hope you all stay true to yourselves and stay committed to what you've initiated.

hear, hear
 

Astrael

Member
After considering my options, cost and scope, I think this kind of sprite (below) is what would simultaneously fit my project and be within the scope of my abilities. My question is if there is a term for this kind of pixel art?

ac8bc93e042ebbfbec82427eaac40181.jpg


Please don't say "ugly" xD. Thank you.

I like it, reminds me of low-res VGA graphics from the early 90s, but far nicer animation than they used back then.
 

andymcc

Banned
Hi, folks. I have been working on a game for a while now, but I am taking my sweet time as I am learning as I develop it. So far I only have the few characters of the game drawn in regular art, not on pixel and stuff. Still, I thought it was about time to drop the temporary sprites I made and start working on the actual sprites for my game.

After considering my options, cost and scope, I think this kind of sprite (below) is what would simultaneously fit my project and be within the scope of my abilities. My question is if there is a term for this kind of pixel art?

ac8bc93e042ebbfbec82427eaac40181.jpg


Please don't say "ugly" xD. Thank you.

V good animation.
 
Hi guys!

I worked on another Game Jam at work and we each made something during our short summer vacation (mostly the last weekend of it).
I tried working on some small local multiplayer game for 4 players where you use light beams (the white beams) to reflect light and attack other players.

giphy.gif


The players have a shield that they can also use to defend themselves from other players, and their HP is shown in the center (the shiny circles).
I also tried to add something fun, so I made some prisms that you can hit and get your beam scattered in two directions.

giphy.gif

(You can see the red and then the green players using it)

What do you guys think? It's not too deep, but for a couple days thing I'm quite happy with how it ended up looking!

https://www.youtube.com/watch?v=voDwrh3u8ok

With some gameplay polish I could see that being really cool. I love the potential risk vs reward of needing to be in a certain location to attack.
 

Dlink16

Member
Hi, folks. I have been working on a game for a while now, but I am taking my sweet time as I am learning as I develop it. So far I only have the few characters of the game drawn in regular art, not on pixel and stuff. Still, I thought it was about time to drop the temporary sprites I made and start working on the actual sprites for my game.

After considering my options, cost and scope, I think this kind of sprite (below) is what would simultaneously fit my project and be within the scope of my abilities. My question is if there is a term for this kind of pixel art?

ac8bc93e042ebbfbec82427eaac40181.jpg


Please don't say "ugly" xD. Thank you.

Like the colours, keep it up!
 
This thread is very inspiring. I'm a UX/UI designer and have been getting the itch to design a game in my free time. I think id try a beat em up game as its my favorite classic genre and there really hasn't been many released and the ones that have don't really satisfy that classic Capcom itch. I would do something simple with no RPG elements that focuses on good sound, smooth gameplay, large sprites and fun enemies. The theme of the game would be 80's/90's Pro Wrestling.

What are the best and quickest to learn engines to use for a beat em up?
 

missile

Member
Ok, so I'm back with the results of the tests you guys suggested:

- Changing the texture from PNG to TGA and running it through pngcrush both did no good. The sprite came up with the same color banding as before.
- Running it through missile's function did give me some results, even though it didn't fully solve the problem. It's much better, but you can still clearly see color banding on the borders. Is there any way one can still improve this?

Here are the 3 versions:

Linear
Ra2VipH.png


Gamma (Correct version)
conKGmf.png


After sRGBtoLinear
y3G9got.png


EDIT: The forum kinda altered the look of the sRGBtoLinear version. If you Right Click+View Image, it will look like it does on the engine.
Did some reading; what happens if you set "Bypass sRGB Sampling"?
 

missile

Member
You're wrong. Fuck off.

(I normally ignore spammers but not only could you not be bothered to do more than cut and paste the entire post in multiple threads, but you also included over 100MB of GIFs. For fuck's sake.)
That's why I want to see us in the Community section, to get rid of those
people posting their fire-and-forget ads adding nothing to the discussion.

Either that or abusing this thread that way should become a bannable theme.
 
Top Bottom