• 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.

GAF Indie Game Development Thread 2: High Res Work for Low Res Pay

Status
Not open for further replies.
K

Kalentan

Member
So I'm going to be honest. I think my spriting is what is killing me when it comes to making a game. I wanted, and still do, make a Souls/Zelda inspiried game, but then I look at my sprites and realize they look terrible as soon as I try and draw them doing attack animations.

F5Nb.png


I made this which shows the "newest" to the "oldest" and I'm almost tempted to go back to the original since it feels like it's the only one that could have proper attack animations, but it's also very simple and not very appealing. The middle one almost looks like it might be the best, but I started with that originally, found I needed to make the arms longer and that's what created the "newest" one.

It's driving me crazy and I haven't made any progress in days because of it.
 
M

mourad louali

Member
I'm not a fan of the Rayman design but ultimately, you should chose the one that looks better animated.

As a little bit of advice, you shouldn't waste days on this. Draw some placeholders sprites really fast and move on. Work on something else until you made up your mind.
 
K

kode80

Neo Member
Kalentan said:
I made this which shows the "newest" to the "oldest" and I'm almost tempted to go back to the original since it feels like it's the only one that could have proper attack animations, but it's also very simple and not very appealing. The middle one almost looks like it might be the best, but I started with that originally, found I needed to make the arms longer and that's what created the "newest" one.

It's driving me crazy and I haven't made any progress in days because of it.
Click to expand...
When working on sprite animations, I focus on getting the animation down first and then cleaning up the individual frames once I'm happy with the movement.

Duplicate a frame, do (very) rough cut-paste-reposition of the various body parts etc, rinse + repeat.

Also, try limiting your key frames at first and then insert frames between where you feel the animation needs to be smoothed out.
 
N

Nezzhil

Member
SpacePirate Ridley said:
Going to try the Ludum Dare this weekend with a new programmer (and the old one too, lets see if they can work together).

We met the new programmer at the EGX rezzed this past week, he is super cool and funny and seems to undertand Unity really well.
If the old and new programmers work well together we could incorporate him into the bigger games we are making, and that would be great.



Yup!
Click to expand...

That sounds great! I would like to play your game, what is your LD account?
¡Mucha suerte!

Also, after some time without any news, our game Tankr has finally an online mode with a basic loobies to play with your friends on Steam. Now we need to test, test and test it to make sure that the netcode will work fine with our peer to peer system (we're developing the game without money, so a hosting service is out of our possibilities). If it works fine we'll add it to the final build before the game launch.
 
T

timetokill

Banned
Kalentan said:
So I'm going to be honest. I think my spriting is what is killing me when it comes to making a game. I wanted, and still do, make a Souls/Zelda inspiried game, but then I look at my sprites and realize they look terrible as soon as I try and draw them doing attack animations.

F5Nb.png


I made this which shows the "newest" to the "oldest" and I'm almost tempted to go back to the original since it feels like it's the only one that could have proper attack animations, but it's also very simple and not very appealing. The middle one almost looks like it might be the best, but I started with that originally, found I needed to make the arms longer and that's what created the "newest" one.

It's driving me crazy and I haven't made any progress in days because of it.
Click to expand...

Okay, so this is a tough thing. I understand that you want your game to "look" as cool as it does in your head, but it's tough because you're not really an artist. I have the same issues.

Here's what I would suggest.

bRex2Hi.png


Look at the original Legend of Zelda. The attack frame is literally one frame for each direction, and they overlay the correct sword (or wand) onto the attack frame as necessary.

Now, Legend of Zelda is awesome, and this limited art style didn't stop them from making an incredible game. Similarly, you should embrace your limitations. Go as simple as possible so that you can make the art quickly and get to making your game.

If all goes well, when the game is nearing completion and you're happy with it, you can hire an artist to do better spritework for you, and you can go back and update your art. If not, hey, you still have a completed game, which is a hell of a thing and something you can be proud of.

But don't let the areas you're lacking in stop you from making your game.
 
S

Slamtastic

Member
timetokill said:
The really difficult part of what I want is getting this explicit combo:
- 3D voxel models
- Drawn "without lighting info" so they appear flat -- each individual voxel should be a solid color on all sides
- But with lighting shading, like a toon shader -- so voxels within the model have gradations of color based on lighting

I managed to get a weird version of what I want to work, but alas it only works on spheres:
EoVkpUC.png
Click to expand...


I really want to achieve that too.

When I saw the game Voxelnauts I really liked that style and wanted to try to make a "3D 16 bit sprite" game. For that game, the shading is done manually in the voxel editor from a fixed imagined light source and the game is unlit, but I too was wondering if it was possible to have dynamically lit voxel models where each voxel was unlit while changing its color to darker or lighter based on lighting.

Could you share how you achieved in so it worked on the sphere? Whatever level of shader experience (and voxel modeling skills, and unity knowledge, and general game dev ability) you have, I have much less, but I'd still like to look at it and see if I can learn.
 
T

timetokill

Banned
Slamtastic said:
I really want to achieve that too.

When I saw the game Voxelnauts I really liked that style and wanted to try to make a "3D 16 bit sprite" game. For that game, the shading is done manually in the voxel editor from a fixed imagined light source and the game is unlit, but I too was wondering if it was possible to have dynamically lit voxel models where each voxel was unlit while changing its color to darker or lighter based on lighting.

Could you share how you achieved in so it worked on the sphere? Whatever level of shader experience (and voxel modeling skills, and unity knowledge, and general game dev ability) you have, I have much less, but I'd still like to look at it and see if I can learn.
Click to expand...

Sure thing, as long as you promise to share if you figure it out :)

What I did was run a method that redid the normals of the vertices in the model based on a vector from the center of the model to the vector's position. What this does it make it so each normal basically points "away" from the center of the model. That's why it works for spheres (and "kind of works" for cones and cylinders) but not much beyond that. Then I have a shader that does a sort of toon-shading esque thing that just does light gradations in steps. Anything complex won't look right at all, not to mention if you have things "stacked" the break between individual models will become very apparent:

K28Iq0m.png

In this image, the green cylinder is made of 4 "blocks", and it's very obvious where that is based on the cut in half (horizontally) around the cylinder.

One thing I'm considering is to create a little editor that can do more fancy normal adjustments so that it can work for other model types. But on the other hand I really don't want to have to do a bunch of hand-edited shit to make everything work ;)
 
S

Slamtastic

Member
timetokill said:
Sure thing, as long as you promise to share if you figure it out :)

...
Click to expand...

What I've thought about while looking at the workings of getting voxels from an editor to a game engine, is that you can't do much to get the effect we want after the stage where the voxel data is converted to standard geometry for screen output.

The idea would be to change the color of a voxel to a darker one while keeping the output unlit, but that means the color changing step has to be done earlier.

After being imported by a voxel tool like PicaVoxel in Unity, the results that you are working with and applying shaders to, the thing that ends up in the game, is the standard triangle geometry that has been the basis of 3D graphics for so many years now.

0.png
Click to expand...

At the point of there being vertices and normals to modify, there is no voxel to change the color of.

The way I think it has to be done is when the object is at the stage of an array of defined voxel structures containing XYZ, RGB, and Alpha information.

If you write the renderer to take in the entries of the voxel object's array and generate poly cubes corresponding to the position and color of each voxel in your file saved from your MagicaVoxel or Qubicle, then the color passed into the renderer should be the output of a lighting function that takes into account the coordinates of parent object in the game world and the relative coordinates of the voxel itself inside it, the light object's position and intensity value, color and angle, etc.

I think way to get the best results for the style is to write your own renderer that pulls the voxel format file for the object directly, applies lighting first, then does the conversion to geometry, rather than making shaders for regular poly geometry that was generated from voxel data.

Of course, that's way beyond my experience and knowledge level, but at a conceptual level that's my thinking.

PicaVoxel also says it's scriptable in a way that lets you access the voxels themselves and delete/add them, so you could try using that to change their color before the conversion to meshes.
 
S

superNESjoe

Member
timetokill said:
Okay, so this is a tough thing. I understand that you want your game to "look" as cool as it does in your head, but it's tough because you're not really an artist. I have the same issues.

Here's what I would suggest.

bRex2Hi.png


Look at the original Legend of Zelda. The attack frame is literally one frame for each direction, and they overlay the correct sword (or wand) onto the attack frame as necessary.

Now, Legend of Zelda is awesome, and this limited art style didn't stop them from making an incredible game. Similarly, you should embrace your limitations. Go as simple as possible so that you can make the art quickly and get to making your game.

If all goes well, when the game is nearing completion and you're happy with it, you can hire an artist to do better spritework for you, and you can go back and update your art. If not, hey, you still have a completed game, which is a hell of a thing and something you can be proud of.

But don't let the areas you're lacking in stop you from making your game.
Click to expand...

This is really solid advice. Embrace all of your shortcomings and limitations and make your game around those. If you're not a very skilled or experienced artist, then make really simple art. Your vision or idea for art direction might have to be compromised, but you'll find yourself much happier with it if you sculpt within reason. You'll find a nice common ground that allows you to put out art you're happy with in a reasonable amount of time.

Don't overexert yourself and let your project suffer over something like that.

Also, wanted to hop in and share that GunWorld 2 is on Steam Greenlight right now. I should be releasing the game on Xbox One within the next month, but a PC release is up in the air at this point. I'm committed to it if I get through Greenlight, but I've had tremendous success on consoles and essentially zero success on PC so for the mean time it made way more sense to focus exclusively on console. However, for the (very few) people who bought the original on Steam, I'd like to offer the option to them. Any time I've sold a game through Humble Widget or another store all I got were tweets or emails about how they refused to buy it until it was on Steam, so I'm not releasing on PC otherwise.
 
B

Blizzard

Banned
SpacePirate Ridley said:
So we had a great idea yesterday night with the Ludum Dare title, Shapeshift.

Hoping all works well this days and we can get it finished.
Click to expand...
I kind of disliked it (along with all but 2 of the final themes) and felt sad and uninspired and wanted to quit last night.

I hardly got anything done, but today I feel a bit better and I think I'll try some more.

This time around, not ONE of the final round themes had more upvotes than downvotes.
 
S

SpacePirate Ridley

Member
Blizzard said:
I kind of disliked it (along with all but 2 of the final themes) and felt sad and uninspired and wanted to quit last night.

I hardly got anything done, but today I feel a bit better and I think I'll try some more.

This time around, not ONE of the final round themes had more upvotes than downvotes.
Click to expand...

Shapeshift was pretty crappy as something to choose, there were more interesting things. Thankfully one of the ideas we were thinking to make with other of the finalists themes could work also with shapeshift. But yeah, pretty uninspired choice, you can see the people posting in the ludum dare website yhe typical first thinking boring ideas about the theme. Reminds me of evolution, the first one I did with other guys some years ago, some of exactly the same type of games im seeing now were done that time.
And the second choice was 2 colors, thats another of the typical things were we already have some great indie games using that ideas.

HP_Wuvcraft said:
Oh yeah that's right! Good luck to all the Ludum Dare Warriors today!
Click to expand...
Thanks!
 
J

Jumplion

Member
Blizzard said:
I kind of disliked it (along with all but 2 of the final themes) and felt sad and uninspired and wanted to quit last night.

I hardly got anything done, but today I feel a bit better and I think I'll try some more.

This time around, not ONE of the final round themes had more upvotes than downvotes.
Click to expand...

Yeah, I wasn't quite feeling this theme either, all my ideas I had never quite grabbed me and what I did go with kind of turned into a janky mess of idontknowwhat. It kind of necessitates either a "circle, triangle, square" style for non-artists.

I'm downgrading my LD experience into a "do some productive asset working-dare" to experiment with some things I wanted to before, so hopefully I got some neat stuff to show for it later.
 
M

MR BLACKMAGIK

Member
Have you guys seen Soda Drinker Pro? The game looks so terrible it honestly inspires me to try and get a game published on Xbox One. Seriously, if a game that looks like it was made in MS Paint can do it then anything can!

Am I missing something here or is it that easy? The Wiki page for Soda Drinker Pro says it was made in one day. I'm assuming that was at a Game Jam and there have since been additions but still. Thoughts?
 
S

SpacePirate Ridley

Member
HP_Wuvcraft said:
Please do tell us stories later.
Click to expand...

Of course!


So we are 5 people in our group, we are going pretty well with time right now.
The first night we designed the document for the game and did the brainstorming throwing some ideas. At the end we got with one we were pretty happy to theme it around the shape shifting theme (although we didnt really like the theme when it appeared and there where a few minutes hating the guts of it lol).

So we created the shapeshifting element into a brawling futuristic sports game (2 vs 2 multiplayer game) where you have to change the rival by converting him into a ball (something similar to Samus' morphball) and trying to get to the yard of your oppenent, fight and shapeshift the rival again to score (something similar to Nidhogg about trying to fight your way to your rival's field).

Yesterday was the prorgrammers works of working with the camera movements and characters animation, while the artists, me included, worked creating the world, characters (created by my good friend and excellent artist, Laura, and I love them!) and menus (although mostly mockups). The pixelart is done by another good friend of mine, and another excellent artist (he was the pixelartist and animator of Gods will be Watching) and good friend Jon.

Today we will start to change the mockups into the real game, and we will test with the programmers the real game to see if its any fun to play (thats the thing we most fear which might make everything coming down from our happy go lucky state of liking how things are working now at the moment).
That and the music, that we dont really have anyone at the moment.

So this are 2 mockups, one for the main menu, and another is the play field (the center section, then is the field yard of each team with different platfoms placement helping the defending team).
Everything will be tranformed into pixel art.
menumock-up6nru3.png

pitchitcej.png
Click to expand...
 
H

HandsomeCharles

Member
Quick question for those of you who have either worked with Xbox (One/360) or more specifically, supported those controllers on PC (Working with Unity, but I assume this issue crops up elsewhere).

So, Axis' 3 and 6. These things just seem like a pain in the ass to me. They both report the input from both analogue triggers. Although I can use Axis' 9/10 to get each trigger separately this makes it really annoying to make a keybinding system that just waits for user input, as their default state (of 3/6) isn't "0" and it keeps "interrupting" my system.

I managed to make a workaround using some #if platform stuff, but it still feels rather hacky. I was just wondering if anyone else had come up against difficulties like this? And if so, what was your solution?
 
E

ephemeral

Member
HandsomeCharles said:
Quick question for those of you who have either worked with Xbox (One/360) or more specifically, supported those controllers on PC (Working with Unity, but I assume this issue crops up elsewhere).

So, Axis' 3 and 6. These things just seem like a pain in the ass to me. They both report the input from both analogue triggers. Although I can use Axis' 9/10 to get each trigger separately this makes it really annoying to make a keybinding system that just waits for user input, as their default state (of 3/6) isn't "0" and it keeps "interrupting" my system.

I managed to make a workaround using some #if platform stuff, but it still feels rather hacky. I was just wondering if anyone else had come up against difficulties like this? And if so, what was your solution?
Click to expand...

I'm interested in this too. Mind sharing your "hacky" solution?

I've been thinking about using kickstarter but I'm afraid that generating and sending out keys to the potential backers will be a huge time waste. Is there any sort of script that does this automatically?
 
H

HandsomeCharles

Member
ephemeral said:
I'm interested in this too. Mind sharing your "hacky" solution?

I've been thinking about using kickstarter but I'm afraid that generating and sending out keys to the potential backers will be a huge time waste. Is there any sort of script that does this automatically?
Click to expand...

Here's a snippet of the listening code. Ill explain afterwards (This probably isn't the best way to do it either, but it worked for me)

Code: 
                    //Axis loop
                    for (int i = 1; i < 11; i++ )
                    {
                        //Clear stringbuilder
                        m_StringBuilder.Remove(0, m_StringBuilder.Length);

#if UNITY_STANDALONE_WIN || UNITY_EDITOR_WIN || UNITY_XBOXONE

                        //Hacky skip for shitey Xbox axis 3 and 6 (triggers -1 to +1)
                        if (i == 3 || i == 6)
                        {
                            bool _bXbox = false;
                            for ( int j = 0; j < Input.GetJoystickNames().Length; j++)
                            {
                                if (Input.GetJoystickNames()[j].Contains("Xbox"))
                                {
                                    _bXbox = true;
                                }
                            }

                            if(_bXbox)
                            {
                                continue;
                            }
                        }
#endif

                        //Make new stirng
                        m_StringBuilder.Append("Axis_");
                        m_StringBuilder.Append(i);

                        //Attempt to create new binds

                        bool _bCreatedBind = CreateBindFromInput(eBindType.eBT_Axis);


                        //If the bind was created, Bail and reset state
                        if(_bCreatedBind)
                        {
                            ExitBindingProcess();
                            return;
                        }
                    }

So, In my attempts to make a "Universal" input manager, I set up a potential 11 Axis'. This part of the code essentially just iterates over them, and then calls a function that checks for input (when given an axis name) and creates a binding.

What I do to skip 3 and 6, is to just do an #ifdef (#if in c#) to check the platform. If it's something "Windows-y" then I do a horrible little hack of finding the controller names. If those names contain the word "Xbox", then I skip that iteration.

Now, this is shit because I can't guarantee what the names of the controllers are going to be. It's fine for official ones. I know they have the word "Xbox" in them, but if someone is using a 3rd party controller I have no idea what the name might actually be.

That said, Ironically this issue wouldn't actually exist if I was creating the game for the Xbox. Even though a variety of controllers could be used, they'd all conform to the same standard and I could just skip those Axis' without doing any kind of string check. The problem arises because If I'm on the PC, some controllers will have totally legitimate Axis 3/6, and I don't want to exclude those ones accidentally.

This is the best-fit that I can think of right now, but if anyone else knows a better solution I'd really like to hear it!


Also, If you'd like to see how I've done my Input Manager, then take a look at this Link. It's a little tutorial I made (But it's not great so I apologise in advance). I also need to update it so that it doesn't use "foreach", as that is going to be generating a shitload of garbage... Can't wait for that new GC....
 
S

SpacePirate Ridley

Member
Our character designer, Lauren Eavle, just finished the 4 playable characters for our Ludum Dare game. We only have two teams for the game, as we dont really have more time.
The orange one is The Charging Balls and the blue one are The Crazy Vikings. The shoulders pads look like Samus as the shift into the playable ball when they get hit too much, so the rival can get you as the ball and score using your body.
eileenrodriguezzpqjn.png
larslundstromx7o5s.png


brynjargunvaldssonjrrgd.png
misterhightowerpppdo.png


How are the gaffers participating in the Ludum doing?
 
E

ephemeral

Member
HandsomeCharles said:
Here's a snippet of the listening code. Ill explain afterwards (This probably isn't the best way to do it either, but it worked for me)

Code: 
                    //Axis loop
                    for (int i = 1; i < 11; i++ )
                    {
                        //Clear stringbuilder
                        m_StringBuilder.Remove(0, m_StringBuilder.Length);

#if UNITY_STANDALONE_WIN || UNITY_EDITOR_WIN || UNITY_XBOXONE

                        //Hacky skip for shitey Xbox axis 3 and 6 (triggers -1 to +1)
                        if (i == 3 || i == 6)
                        {
                            bool _bXbox = false;
                            for ( int j = 0; j < Input.GetJoystickNames().Length; j++)
                            {
                                if (Input.GetJoystickNames()[j].Contains("Xbox"))
                                {
                                    _bXbox = true;
                                }
                            }

                            if(_bXbox)
                            {
                                continue;
                            }
                        }
#endif

                        //Make new stirng
                        m_StringBuilder.Append("Axis_");
                        m_StringBuilder.Append(i);

                        //Attempt to create new binds

                        bool _bCreatedBind = CreateBindFromInput(eBindType.eBT_Axis);


                        //If the bind was created, Bail and reset state
                        if(_bCreatedBind)
                        {
                            ExitBindingProcess();
                            return;
                        }
                    }

So, In my attempts to make a "Universal" input manager, I set up a potential 11 Axis'. This part of the code essentially just iterates over them, and then calls a function that checks for input (when given an axis name) and creates a binding.

What I do to skip 3 and 6, is to just do an #ifdef (#if in c#) to check the platform. If it's something "Windows-y" then I do a horrible little hack of finding the controller names. If those names contain the word "Xbox", then I skip that iteration.

Now, this is shit because I can't guarantee what the names of the controllers are going to be. It's fine for official ones. I know they have the word "Xbox" in them, but if someone is using a 3rd party controller I have no idea what the name might actually be.

That said, Ironically this issue wouldn't actually exist if I was creating the game for the Xbox. Even though a variety of controllers could be used, they'd all conform to the same standard and I could just skip those Axis' without doing any kind of string check. The problem arises because If I'm on the PC, some controllers will have totally legitimate Axis 3/6, and I don't want to exclude those ones accidentally.

This is the best-fit that I can think of right now, but if anyone else knows a better solution I'd really like to hear it!


Also, If you'd like to see how I've done my Input Manager, then take a look at this Link. It's a little tutorial I made (But it's not great so I apologise in advance). I also need to update it so that it doesn't use "foreach", as that is going to be generating a shitload of garbage... Can't wait for that new GC....
Click to expand...

Thanks for sharing. What do you mean with this btw:

"Although I can use Axis' 9/10 to get each trigger separately this makes it really annoying to make a keybinding system that just waits for user input, as their default state (of 3/6) isn't "0" and it keeps "interrupting" my system."

I haven't thought that much about using a system that supports different controllers, gonna be a pain to implement.

You can make the code a bit neater by calling the function CreateBindFromInput(eBindType.eBT_Axis) in the following if statement, which also removes the garbage from the local variable.
 
P

Popstar

Member
HandsomeCharles said:
Quick question for those of you who have either worked with Xbox (One/360) or more specifically, supported those controllers on PC (Working with Unity, but I assume this issue crops up elsewhere).

So, Axis' 3 and 6. These things just seem like a pain in the ass to me. They both report the input from both analogue triggers. Although I can use Axis' 9/10 to get each trigger separately this makes it really annoying to make a keybinding system that just waits for user input, as their default state (of 3/6) isn't "0" and it keeps "interrupting" my system.

I managed to make a workaround using some #if platform stuff, but it still feels rather hacky. I was just wondering if anyone else had come up against difficulties like this? And if so, what was your solution?
Click to expand...
Not using Unity. But in my C++ framework I use RawInput for generic USB HID controllers and filter out any Xbox controllers. Those use a separate XInput path.

You can detect this with the device ID. It will contain the substring "IG_". It's still somewhat of a hack but ¯\_(&#12484;)_/¯

As an example here's the device ID for a Street Fighter IV Xbox 360 controller:
Code: 
\\?\HID#VID_1BAD&PID_F028&IG_00#7&530b005&0&0000#{4d1e55b2-f16f-11cf-88cb-001111000030}
These IDs vary based on a lot of factors, including what USB port a controller might be plugged into. But searching for the "IG_" substring is working for me so far. If anyone has any Xbox controllers lying around I'd be interested if they could check them.
 
H

HandsomeCharles

Member
ephemeral said:
Thanks for sharing. What do you mean with this btw:

"Although I can use Axis' 9/10 to get each trigger separately this makes it really annoying to make a keybinding system that just waits for user input, as their default state (of 3/6) isn't "0" and it keeps "interrupting" my system."

I haven't thought that much about using a system that supports different controllers, gonna be a pain to implement.

You can make the code a bit neater by calling the function CreateBindFromInput(eBindType.eBT_Axis) in the following if statement, which also removes the garbage from the local variable.
Click to expand...

So, what I mean is that my "listening" code listens for any of my defined Axis' inputs to be a value other than zero. When a value other than zero is found, it means the player has pressed that axis and thus that is the axis that should be used with the bind being created.

This poses a problem with numbers 3/6, as because these output a value that is for both of the analogue triggers combined, it goes between a value of -1 and +1. This means that in the "resting state" the value output by numbers 3/6 is not zero. (Which then means that it is incorrectly picked up as the bind being created, even if the player isn't pressing them).

Axis 9 and 10 are the left and right triggers again, but without this "combined" value. They simply go from zero to one (0 being not pressed, 1 being fully pressed)

(Also thanks for that little bit about the local bool, that slipped my mind for some reason :p)

Popstar said:
Not using Unity. But in my C++ framework I use RawInput for generic USB HID controllers and filter out any Xbox controllers. Those use a separate XInput path.

You can detect this with the device ID. It will contain the substring "IG_". It's still somewhat of a hack but ¯\_(&#12484;)_/¯

As an example here's the device ID for a Street Fighter IV Xbox 360 controller:
Code: 
\\?\HID#VID_1BAD&PID_F028&IG_00#7&530b005&0&0000#{4d1e55b2-f16f-11cf-88cb-001111000030}
These IDs vary based on a lot of factors, including what USB port a controller might be plugged into. But searching for the "IG_" substring is working for me so far. If anyone has any Xbox controllers lying around I'd be interested if they could check them.
Click to expand...

Thanks, that was exactly the sort of thing I was looking for. An "IG_" sub seems much more reliable than just looking for "xbox" anyway :p

I'll do some reading and see what the deal is with RawInput / XInput with Unity and try to get something sorted out :)

Just imagine, none of this would have even been a problem if those damn Axis' didn't exist :p
 
E

ephemeral

Member
HandsomeCharles said:
So, what I mean is that my "listening" code listens for any of my defined Axis' inputs to be a value other than zero. When a value other than zero is found, it means the player has pressed that axis and thus that is the axis that should be used with the bind being created.

This poses a problem with numbers 3/6, as because these output a value that is for both of the analogue triggers combined, it goes between a value of -1 and +1. This means that in the "resting state" the value output by numbers 3/6 is not zero. (Which then means that it is incorrectly picked up as the bind being created, even if the player isn't pressing them).

Axis 9 and 10 are the left and right triggers again, but without this "combined" value. They simply go from zero to one (0 being not pressed, 1 being fully pressed)

(Also thanks for that little bit about the local bool, that slipped my mind for some reason :p)




Thanks, that was exactly the sort of thing I was looking for. An "IG_" sub seems much more reliable than just looking for "xbox" anyway :p

I'll do some reading and see what the deal is with RawInput / XInput with Unity and try to get something sorted out :)

Just imagine, none of this would have even been a problem if those damn Axis' didn't exist :p
Click to expand...

Alright, so the reason you don't want to use axes 9 and 10 is cause you want to have that possible combined value? Makes sense then :). Thanks for sharing.
I would like the ability to read those trigger inputs as button presses, but that'll never happen :(.
 
N

Nezzhil

Member
SpacePirate Ridley said:
How are the gaffers participating in the Ludum doing?
Click to expand...
Your game idea seems amazing, I hope that you and your team have enough time to give it the good gameplay that it reserves.
We had some problems trying to find a cool idea with the Shapeshift theme, and I dont think that anyone of our team is 100% happy with the idea that we finally are developing, but we're working hard to finish it and I hope that the people have fun with it :)

Our game is about a sorcerer or a master in disguise that is a prisoner in a castle and tries to escape thanks to his ability to copy the body of any life form that is unconscious.

I dont have any screenshot in my mobile right now, but I will share some tomorrow morning.
 
H

HandsomeCharles

Member
ephemeral said:
Alright, so the reason you don't want to use axes 9 and 10 is cause you want to have that possible combined value? Makes sense then :). Thanks for sharing.
I would like the ability to read those trigger inputs as button presses, but that'll never happen :(.
Click to expand...

Ah, sorry, it's the other way around. I don't care about the combined values (Numbers 3/6), I just want to use the 0-1 values (9/10).

There's multiple ways to mock a trigger into a button press though. You can just look for any input that isn't zero, or wait until the input value is "1". That'd do the same trick.
 
P

Popstar

Member
HandsomeCharles said:
Thanks, that was exactly the sort of thing I was looking for. An "IG_" sub seems much more reliable than just looking for "xbox" anyway :p

I'll do some reading and see what the deal is with RawInput / XInput with Unity and try to get something sorted out :)

Just imagine, none of this would have even been a problem if those damn Axis' didn't exist :p
Click to expand...
After writing that I was trying to remember where I got that "IG_" hack from, and it turns out it's from Microsoft sample code. So that's probably as good as it gets. https://msdn.microsoft.com/en-us/library/ee417014(VS.85).aspx
 
S

SpacePirate Ridley

Member
Nezzhil said:
Your game idea seems amazing, I hope that you and your team have enough time to give it the good gameplay that it reserves.
We had some problems trying to find a cool idea with the Shapeshift theme, and I dont think that anyone of our team is 100% happy with the idea that we finally are developing, but we're working hard to finish it and I hope that the people have fun with it :)

Our game is about a sorcerer or a master in disguise that is a prisoner in a castle and tries to escape thanks to his ability to copy the body of any life form that is unconscious.

I dont have any screenshot in my mobile right now, but I will share some tomorrow morning.
Click to expand...

We have the characters running through the screen and trhowing the ball with physics and feels great, but still things can go wrong, but at least is looking good.

And your idea of the castle and the copy of abilities sounds great. Would love to see some screens of it.
Good luck for you and your team in the last day!
 
K

KOCMOHABT

Member
So uh, I came up with a name for the game.

I hope it's ok if i basically repost the story part which i also posted in my blog

e7U6io1.png


What is Bounty Road?

You take on the role of an infamous bounty hunter. An anonymous source offered a hefty amount of credits for the execution of the self-declared &#8220;Lord of the Desert&#8221;, a rebel leader hiding far away in the land of sand.

You gather your gang of loyal road warriors to take on the job and prepare for a long journey full of hazards.

N1Eusmm.gif


While locals, bandits and wildlife are some of the enemies you&#8217;ll encounter on this adventure, there are also rumours about other bounty hunters making their way to the desert&#8230;

Bounty Road is a combination of classic Party RPGs with the roguelike genre. You control a group of cars, which each has unique attributes depending on chassis, armament, armor and drivers/gunners.
Light motorcycles will stay just as relevant as heavier vehicles through their unique loadouts.

AAS10Lv.png


You will depend on recources like food/water, gas and spare repair parts to make your way through. The people that come along will improve in skill, but without proper tactics you won&#8217;t win the fight.

------------------------------------------------------------------------------

Also, today I implemented some high quality bloom with the implementation proposed by Jorge Jimenez in his paper for advanced post processing for Call of Duty [Jimenez14], which is basically an enhanced version of the (initial) Unreal Engine 4 bloom.

Am really happy with the results. (Exaggerated in the gif of course)

oQC6ja7.gif
 
F

floatstarpx

Member
timetokill said:
What I did was run a method that redid the normals of the vertices in the model based on a vector from the center of the model to the vector's position. What this does it make it so each normal basically points "away" from the center of the model.
Click to expand...

Maybe you could combine this approach with also looking at which sides of each voxel are exposed (e.g. not touching another voxel)?
 
SeanNoonan

SeanNoonan

Member
Working long distance sucks :(

I've still not played the game... here's an older screenshot though.

OuKE9xC.png



...and the dude!

tumblr_o5t113kRs81tsc2fyo1_100.gif


I think the eyes + antenna combo are way too Bomberman D;
 
M

missile

Member
SeanNoonan said:
It might seem stupid, but I actually don't know how to use colour. I almost made this black and white :D
Click to expand...
Colors are overrated. They are used all over the place like nothing else. xD

What still gets me is making a game with a fixed and quite limited color
palette where you have to take the shades out of the same limited color
palette. This means, every color within the color palette has at most as many
shades as there are colors in the palette. An interesting aspect of this
technique is that you will get somewhat wrong coloring in the shades, but the
shades themselves are sound, i.e. they are producing a believable shade of a
main color when put in relation, i.e. when you use the main color with the
shades nearby, for example when shading an object. So you look at the object
and for some reasons you won't mind the wrong coloring because the shades are
matching to a given degree. However, the look will be different because the
coloring will be different in the shades. I'm pretty sure I'm going to use it
for my game Superstall. Hand-picking the shades resp. optimizing them for
certain picture would further improve the whole thing, as some cool artists
from the past have shown with their graphics drawn out of some very limited
color palettes. However, for 3d graphics hand-picking the shades becomes
difficult. Hence, some intelligent algorithms for produce them are in order.

I think it is better to make each color count in a game resp. relate the
colors to each other. I don't know, but all these smooth color gradients
always look cheap to me, anyone?
 
M

missile

Member
I want to inspire everyone in here with my lastest achievement!

Look at this fukkin sphere!

RV3pImS.png


I hope you feel a lot better now, now that you know that your work looks way
better than mine! xD

--

Well, yeah, looks a bit crappy. But I just thought I'll post something I'm
currently working on hard. Recently, I couldn't fell to sleep and have started
reading an old paper from Heinrich Hertz about the equations describing
electromagnetic wave propagation away from a dipole antenna -- sort of an
approximation to Maxwell's equations. His equations and its solution give some
interesting decaying function for a wave radiating out from a dipole at a
given frequency over ideal ground.

To relief myself somewhat from daily work, I thought; hey, why not try to
mimic the wave propagation and decay properties and use it as the input to my
TV simulator, man? Sounds cool, I thought. So I came up with some ideas on how
to realize it which I will now try out. My idea is that the wave not only
could be blocked by objects and reflected by them, the wave should also be
able to go through objects, if necessary. The sphere you see will get sort of
an dielectric resp. permeability property which guides the wave though the
sphere's material. Stuff like that.

If this works out the way I dreamed it up, then this may have some potential
for creating some interesting gameplay. I imagine an enemy radiating out a
wave (signal) which travels though the environment towards you depending on
the geometry of the scene. The interesting aspect is that you can
"communicate" with the enemy at a distance depending on the geometry of the
scene/level. So the enemy may distort your vision while closing in with you
having to use the environment's geometry to get shielded etc.

I will prototype his idea somewhat and see how it works, if ever. May perhaps
be complete bullshit. I've really no idea whether this will work because I
have to truncate of lot of equations to make the problem tractable, which may
render the effect towards /dev/null.
 
K

KOCMOHABT

Member
1QuUiVa.gif


Well the compression butchers a lot of the image, but I still love how the bloom compliments the moonlight reflections on the car surface <3
 
Status
Not open for further replies.

Similar threads

Solo developer behind The Falconeer and Bulwark explains why Game Pass is so great for indies:
2
54
2K
chakadave replied
NEW Xbox Indie Game Fest Demo Event: 35 new demos
11
1K
Punished Miku replied
Are there any non corporate big console game developers left?
0
117
Gamer79 replied
Indie GAF: Speed Rivals anime racing game
Game Dev Indie
0
490
BioFan replied
Capcom is holding a game development competition for Japanese students using its RE Engine
Game Dev
11
821
Dacvak replied
Top Bottom