Support NeoGAF

[Yoshi]

Though I am continuing work on my SRPG, progress has been a bit slow, because now that most of the programming is done, I was always thinking about how to realise a 3D platformer, so I decided to further split up my limited ressources into two projects and to start programming a 3D platformer as well. This has the added advantage that a good friend of mine is really, really hot for that as well and has numerous ideas for level creation, so I can actually concentrate on programming for quite a while and do not have to focus on level design too much (though I also like to do that, it takes alot of time for me to come up with fun ideas). I've first worked on the two things most important I think, character physics and camera system. I have actually decided to not use anything premade in Unity, in order to have full control over everything, because I have often experienced some issues with 3D projects made in Unity as a player myself.

However, so much for the background, now my actual question: Is there any good and easy to use (for a non-artist) Voxel-editor to create Unity models, preferrably a free one, that you can recommend?

I really don't want to delve into the whole 3d model creation too much, though I would still like to have a nice look for the game that I can realistically achieve. So Voxel art springs to my mind as an obvious way to easily model game characters and level elements that should also be reasonably efficient, because I would like to put the game on Switch (if Nintendo permits), Xbox One and Wii U, and I really don't want any performance issues on either platform at all. Currently the game is using just cubes as characters and is running very fluidly on Wii U, so I hope to be able to keep up that performance, get a reasonably nice looking game and not to delve into complicated modelling.

I thought I'd just quote myself here, because I reconsidered this after making one Voxel character and seeing that it looked considerably worse than what I could make in Sculptris (other than crashes on saving really nice and intuitive program!) and still required way more polygons. My duo of main characters is now done, as well as the camera (sans collision detection) now I can focus on collision detection for characters and camera, before building a level editor. One question though: I am using models that only have very primitive textures, oftentimes no textures, but just flat colours. However, the Unity lighting makes them look all shiny, is there an easy way to get a more, let's say modest lighting like for instance in Mario 64? The game gets a bit of a PS2-era esthetic through that plastic look, but I'd probably prefer a more N64-looking look.

 

[LordRaptor]

One question though: I am using models that only have very primitive textures, oftentimes no textures, but just flat colours. However, the Unity lighting makes them look all shiny, is there an easy way to get a more, let's say modest lighting like for instance in Mario 64? The game gets a bit of a PS2-era esthetic through that plastic look, but I'd probably prefer a more N64-looking look.

You probably want a custom shader to do that - the N64 IIRC used Gouraud shading (represents lighting on model vertices), and Unitys basic Diffuse shaders use Phong shading (represents lighting per pixel).

If you don't have the technical knowledge to write your own shaders (I know I don't!) you could try using some of the existing shaders and tweak to get the results you want. Setting 0 metallic and 0 smoothness in the Standard shader will give you flat non-reflective colouring on your objects and still make use of PBR elsewhere - you'd just be making objects that behave similarly to coloured wood in terms of lighting (but more so).

If you really have no need for any PBR elements, legacy/diffuse or legacy/diffuse_fast will give you basic non-specular texture objects, and you can just set a colour in the material instead of a texture to get an effect similar to what you want (and you'll probably find its really performant too).

If you reeeeeaaaaaallly don't need any light interactions at all, Unlit shader will completely ignore any lighting (to the extent you can just remove lights from your scene completely) but you'd need to fake lighting with a projector doing blob shadows or similar. Again, you'd find this to be crazy performant, modern machines can throw around hundreds of thousands of unlit polys.

 

[Minamu]

I must be using the attenuation settings in Unity wrong, or they're completely broken :/ The default setting is a min distance of 1, which means that if you're within 1 meter of the audio source, the audio volume is 100%. The default max distance is 500 and is the distance between the min distance and where the sound is completely silent. At least that's how I've interpreted it? It seems to work sort of fine at these default values but some of my audio sources can be heard as clear as day from like 50 meters even though their max distance is 15. I must've misunderstood something or coded something really strange with the audio volumes.

 

[Dascu]

I must be using the attenuation settings in Unity wrong, or they're completely broken :/ The default setting is a min distance of 1, which means that if you're within 1 meter of the audio source, the audio volume is 100%. The default max distance is 500 and is the distance between the min distance and where the sound is completely silent. At least that's how I've interpreted it? It seems to work sort of fine at these default values but some of my audio sources can be heard as clear as day from like 50 meters even though their max distance is 15. I must've misunderstood something or coded something really strange with the audio volumes.

Are the sound effects set as 3D Sound?

 

[Violet_0]

nvm

 

[Pehesse]

It's #gifsaturday!

 

[orioto]

4mo gifs, slower than the real thing oviously

It was pretty complex cause we need separate wing animation to keep the flow, then different body animations. But dragonfly are pretty stiff to begin with and i'd like to keep that vibe. But there is work to do.

Aaaw shit Pehesse don't post this kind of animation before mine

 

[Pehesse]

4mo gifs, slower than the real thing oviously

It was pretty complex cause we need separate wing animation to keep the flow, then different body animations. But dragonfly are pretty stiff to begin with and i'd like to keep that vibe. But there is work to do.

Aaaw shit Pehesse don't post this kind of animation before mine

I post quantity, you post quality - I wouldn't know where to start with yours, especially with that type of painterly rendering and level of detail! :-D

Separating the wings does sound like the best course of action here!

 

[orioto]

I post quantity, you post quality - I wouldn't know where to start with yours, especially with that type of painterly rendering and level of detail! :-D

Separating the wings does sound like the best course of action here!

Quantity you're crazy This is friggin' unrivaled in term of body animation variety and details here for a game! That's amazing you can do that on your own.

 

[Minamu]

Are the sound effects set as 3D Sound?

The Spatial Blend is set to 3D, yes, unless there are more settings elsewhere. There used to be a 3D button on the actual sound files back in the day but it seems to have been removed in newer Unity versions.

Edit: For example, I have an item with an audio source that I can throw across the map. The Spatial Blend is 3D, the min distance is 5 units, and the max distance 20. And yet, I can hear the sounds from anywhere on the map, albeit quieter with distance, but never zero, even though the map is 132 units long. Makes no sense. The audio source is on the object and my only listener is on the player prefab.

Edit2: Okay now I get it. Max distance isn't actually zero volume... It's just the barrier where the volume will stop lowering, and from there, the audio volume stays the same regardless of farther distances from the listener. Maybe I need change the rolloff setting? I'm using the default logarithmic rolloff.

 

[Blizzard]

It's #gifsaturday!

In that first one, are you holding the camera button down? The floor vanishes when I feel like the camera should keep it visible. Other than that, looking good!

 

[Pehesse]

In that first one, are you holding the camera button down? The floor vanishes when I feel like the camera should keep it visible. Other than that, looking good!

Thanks! And you're right, I pan the camera around in that first one, but the floor disappears mostly because the gifs are cropped - the actual prototype aspect ratio is 16:9, with a lot more visible space to the right/left and a little on top/bottom!

In fact, if anyone is interested in trying out the current prototype, you can access it here (playable directly from your browser):
pehesse.fr/honey/pachacuti

Here's an overview of what's available:

-the prototype is super simple and very limited, you can basically only move around, but there is no environmental interaction, or complex platforming logic and motions (no moving platforms, balancing stuff, tightropes, etc).

-playable on both gamepad (analog stick only) and keyboard (still rough implementation)

-movement (analog stick/directional arrows, shift to walk): move around, look up/down, crouch, move while crouched, hug walls, fall through multiple platforms in sequence

-jump (gamepad=A, keyboad=spacebar) = jump combo, eject from wall, edge grab+multiple climb speeds, vertical jump from wall stick

-roll/slide (gamepad=B, keyboard=ctrl) = roll or slide depending on your speed (stop/slow => roll, med/fast => slide)

-attack (gamepad=X, keyboard=W) = kick attack, on ground or in mid-air

-blade attack (gamepad=Y, keyboard=X) = blade attack on ground, wall stick when hugging wall, air blade when wall-stick'd

-glide (gamepad=LB, keyboard=Q) = hold while falling to slow down your descent

The obvious big thing missing is the ability to shift into the Warrior... and the Warrior herself. That's basically what comes next, though before that, I want to take the time to debug a bit the current moves, so I'm looking for help and feedback for that! If you notice anything wrong with the prototype, be sure to send me feedback (PMs, twitter, mail, anywhere really).

Here's a list of everything I need to look into short-term, please feel free to comment on any of those if you have thoughts or experience to share regarding any of them:

- allow jumping sooner after the landings of the jump recoil and jump double recoil

- issue with the sprite mirrored after a vertical jump near an edgehang collision zone, sprite will appear in the wrong direction for a few frames

- issue with the sprite being displaced after an edgeclimb on a slope, the sprite might appear at the wrong coordinates for a single frame

- issue with slopes leading into solid ground: depending on the speed of motion, this leads into air-time/landing (this one is giving me the most trouble, as it's a normal consequence of C2's platforming behaviors and I haven't figured out a few to work around it yet)

- when wall stick'd and panning camera, might add additional flavor anims for looking ahead/back

- rework wall jump to make jump arc more fluid, there's a movement hitch when changing directions in mid-air

- some issues with the sprite/stance not resetting after some movement edge cases with controller at rest, need to identify and fix them

There are also a few others technical things I want to rework before moving forward (replace my current collision box displacement technique with a delta-time based one to make it frame independant, and replace the current control mappings with variable controlled ones to allow for control remapping). If you have any other thoughts or feedback from your experience with the prototype, please share anything that comes to mind!

 

[JulianImp]

More Unity shenanigans, this time related to ProCamera2D.

We had previously noted that Drop Alive had some pretty bad slowdown on some PCs, but never got around to looking into what the source was until today. Turns out it's the fact that the game used ProCamera2D's parallax system, which creates a camera for every parallax layer in use. That works fine on smaller resolutions (such as when using editor mode), but the framerate went from 90 to 15-19 fps on average when playing the game on my retina MacBook with the game tab maximized (so it wasn't even fullscreen).

The solution? I nuked all parallax cameras from orbit, and replaced them with the script I had built for my freelance project that does parallax by moving a small set of particular objects against the camera's movement. The result was a way better (albeit still kind of iffy) 25-35 FPS on a maximized game tab, and nuking a GUI camera as well made things even more stable.

TL;DR: It seems like ProCamera2D's parallax implementation using multiple cameras murders your game's performance on higher screen resolutions, so I'd recommend coming up with your own parallax scrolling system instead of using ProCamera's.

 

[MrHoot]

Working on some special animation right now (the shopkeeper's abode) but I sketched earlier this week the first basic enemy for the game and will spend next weeks' animating it and implementing the first combat aspects to the game. Especially because i'm getting jealous of all ya'll animations and progress (i've been rather lazy this week =p)

The blood fae will basically be the most common enemy but will be dangerous in swarms. Some will be able to swarm you in melee, others might try to cast spells from afar

 

[Fou-Lu]

I was getting frustated with my spirtes and animations not coming out how I wanted them, so right now I am making a prototype in RPG Maker VX Ace with all the default assets and just a couple of scripts I wrote. I know it's a really simple program, but the game idea I have doesn't need any more than this. I do want to move on to bigger and better things, but before that I want to COMPLETE a project.

It's a dungeon delve/mystery game with enemies that don't respawn, Zelda-like puzzles and adventure game-like item use. Hoping to make it a 2 or 3 hour game that could be a prologue to a larger game. After it's done I'll look into making assets. So far I have the opening cutscene and the first screen of the game finished. The first screen I am trying to set up to organically teach you about investigating things.

 

[Yoshi]

You probably want a custom shader to do that - the N64 IIRC used Gouraud shading (represents lighting on model vertices), and Unitys basic Diffuse shaders use Phong shading (represents lighting per pixel).

If you don't have the technical knowledge to write your own shaders (I know I don't!) you could try using some of the existing shaders and tweak to get the results you want. Setting 0 metallic and 0 smoothness in the Standard shader will give you flat non-reflective colouring on your objects and still make use of PBR elsewhere - you'd just be making objects that behave similarly to coloured wood in terms of lighting (but more so).

If you really have no need for any PBR elements, legacy/diffuse or legacy/diffuse_fast will give you basic non-specular texture objects, and you can just set a colour in the material instead of a texture to get an effect similar to what you want (and you'll probably find its really performant too).

If you reeeeeaaaaaallly don't need any light interactions at all, Unlit shader will completely ignore any lighting (to the extent you can just remove lights from your scene completely) but you'd need to fake lighting with a projector doing blob shadows or similar. Again, you'd find this to be crazy performant, modern machines can throw around hundreds of thousands of unlit polys.

Thank you, for now I have chosen the first solution, but it's great to know of the other options as well. Also, you have saved me from programming my own blob shadows, the Unity ones are actually competent with some adjusting .

 

[Cute Dragon Here]

So, I'm now halfway implementing item data as a ScriptableObject. Is it all right if I put in declarations of custom enums inside the object class itself? I need to have a way to distinguish between item types quickly (these enums are used for item categorization and item effects) (Unity 5.4)

Or am I putting in way too much junk?

 

[Makai]

So, I'm now halfway implementing item data as a ScriptableObject. Is it all right if I put in declarations of custom enums inside the object class itself? I need to have a way to distinguish between item types quickly (these enums are used for item categorization and item effects) (Unity 5.4)

Or am I putting in way too much junk?

Up to you. No right way to code. I would recommend putting the majority of your logic in straight-up C# classes, though - you don't need to inherit MonoBehaviour/ScripableObject/etc for most things.

 

[Cute Dragon Here]

Up to you. No right way to code. I would recommend putting the majority of your logic in straight-up C# classes, though - you don't need to inherit MonoBehaviour/ScripableObject/etc for most things.

Ah, I was going through this linked video tutorial and applying what I've learned. It seemed to be a pretty good fit for implementing different item types - the intent is to write a single object class inheriting ScriptableObject, and then creating a whole lot of item representations based on it. Something to do with implementing a functional item system with unique names, descriptions, image, and effects.

 

[Pehesse]

Anyone? I would really appreciate (and kinda need) your feedback!

I've fixed issues with jumping at the border of platforms, and another with landing->sliding below low height ceilings that would result in getting stuck. There's also a "climbing" marker in mid-air in the bottom right area that can allow you to be suspended in nothingness - though that one isn't so much a problem as a leftover from the previous LD!

If you encounter any other issues, it'd be super helpful to know so I can get on fixing them ASAP while I can still tell what's what from the script!

 

[LordKasual]

Anyone? I would really appreciate (and kinda need) your feedback!

I've fixed issues with jumping at the border of platforms, and another with landing->sliding below low height ceilings that would result in getting stuck. There's also a "climbing" marker in mid-air in the bottom right area that can allow you to be suspended in nothingness - though that one isn't so much a problem as a leftover from the previous LD!

If you encounter any other issues, it'd be super helpful to know so I can get on fixing them ASAP while I can still tell what's what from the script!

Sure thing! The animation is really smooth. Two things i noticed after playing around for a while

- It looks like you have some kind of detection to switch her into crawling if she lands near a low opening? You should tweak it a bit to be a bit more precise. It activated on the first 2 platforms and at first i couldn't tell why it was happening. Example, If you land and slide off the first platform while in her landing animation, the collision box will cause you to not grab that ledge and slide under, and she'll land in crawling state under the passthrough platform seemingly for no reason. Also in the opening right above, jumping and landing in the opening will almost always trigger her to land in crawling.

- I'd recommend including some kind of "Charge jump" option that lets her reach Triple Jump height from standing. Or, if that isn't intended, force the triple jump progression to require forward momentum. You can short hop twice and hold the last one from standing to get the benefit of the triple jump, which allows you to just ignore the wall jump mechanic.

Working on some special animation right now (the shopkeeper's abode) but I sketched earlier this week the first basic enemy for the game and will spend next weeks' animating it and implementing the first combat aspects to the game. Especially because i'm getting jealous of all ya'll animations and progress (i've been rather lazy this week =p)

The blood fae will basically be the most common enemy but will be dangerous in swarms. Some will be able to swarm you in melee, others might try to cast spells from afar

I'm really interested to see this thing in action, looks awesome.

 

[Pehesse]

Sure thing! The animation is really smooth. Two things i noticed after playing around for a while

- It looks like you have some kind of detection to switch her into crawling if she lands near a low opening? You should tweak it a bit to be a bit more precise. It activated on the first 2 platforms and at first i couldn't tell why it was happening. Example, If you land and slide off the first platform while in her landing animation, the collision box will cause you to not grab that ledge and slide under, and she'll land in crawling state under the passthrough platform seemingly for no reason. Also in the opening right above, jumping and landing in the opening will almost always trigger her to land in crawling.

- I'd recommend including some kind of "Charge jump" option that lets her reach Triple Jump height from standing. Or, if that isn't intended, force the triple jump progression to require forward momentum. You can short hop twice and hold the last one from standing to get the benefit of the triple jump, which allows you to just ignore the wall jump mechanic.

Thanks a lot for that detailed feedback!!

I'll look into the land->crawling state according to your findings, there's definitely some improvements to make there.

EDIT: just tried it and oh my, this looks like a severe bug I added in the latest version! I'll look into right away!
EDIT2: fixed it and uploaded a new version! For anyone who tried accessing the prototype before, you'll need to force refresh (ctrl+F5) to make sure your browser redownloads the data and doesn't read it off the web cache.

For the triple jump, I didn't think to make the third available immediately as a convenience move, but this could be a nice trick! I actually don't mind it being used in place, as the LD won't always make it an optimal solution. I'm looking to have a component of "player visual expression" through the motions, and make them try to chain the moves fluidly, especially for speedrunning - so while it's possible to simply jump in place, it's definitely much less gracious and fast than doing alternate moves (in that example, jumping to the wall to rebound off it)! Still, gives me lots of food for thought, so I appreciate your ideas on that front very much, thanks again!

 

[Thom Yorke's Wobbly Eye]

I added an unlockable tennis racket to Stage Presence, just like a real touring band has.

Perfect for deflecting glowsticks!

And what works for glowsticks works for bottles of piss!

Of course it's real purpose is as an air guitar.

Bonus video of the revamped audience mind probe

 

[titch]

Tried unity last year and after a fair bit of angst i managed to create (basically copying code) a simple level with working pathfinding. Sure I had managed to get a project working and deal with the problem solving that comes with it but i didn't always understand how things worked. Unfortunately with lots of other stuff going on things kinda petered out.

About a month ago with the October break coming up and being broke I decided to give things another go.

A few weeks were spent getting back into things, messing about with either code or pixel art and i managed to pick some more skills and started to get a bit more of a grasp of what was going on especially within unity. The October holidays came and went and things were slowly moving forward but with so many things to learn at times it can be hard to focus on one thing and not get distracted. I finally created my very first project - a simple 2 player shooter in the vein of pong and had some fun with the family playing something that i had actually created.

I was mulling over where to go next - i know i now have the skills to create some simple projects but wasn't sure where to go. I read Derek Yu's book on Spelunky which i found pretty inspiring and it helped me get some focus. Not sure if anyone else went through this phase but i was finding to deal with the abstractness of coding. Trying to mentally visualize things was really tiring.

Anyway had some kind of breakthrough on Friday at work - wrote down the basis of what i wanted to code and finally got it all working as planned this morning. All it does is create a grid and generates a start at the top and a bottom at the finish and generate a random path from top to bottom without allowing backtracking. Looking at what others are generating on here it's pretty small fry but for myself its a major breakthrough and has given me the confidence to take things further.

Hopefully this time i will keep the momentum going and continue to build my skillset.

 

[liliththepale]

if anyone is interested in trying out the current prototype, you can access it here (playable directly from your browser):
pehesse.fr/honey/pachacuti

I really like a lot of what is here. The animation is great and the moveset is robust and fun.

Having said that, the base movement feels a little bit more slippery than I personally like (but note that this is coming from someone who thinks Mario is too slippery).

I like that the camera is trying to be predictive but too often it predicts wrong. Like when you stop running the camera hangs there for a second, seemingly on the off chance that you'll pick up running right away again, but then the way it lurches back to center as you slide to a halt is upsetting. I would either have the camera snap back as soon as you stop holding input or wait until after the character has come to a complete stop so that the player and camera aren't travelling in opposite directions. Contrary to that, I think the way the camera juts out when you slide is too sudden or two extreme.

Also there are a couple edge cases you might want to handle or guard against:

 

[Pehesse]

I really like a lot of what is here. The animation is great and the moveset is robust and fun.

Having said that, the base movement feels a little bit more slippery than I personally like (but note that this is coming from someone who thinks Mario is too slippery).

I like that the camera is trying to be predictive but too often it predicts wrong. Like when you stop running the camera hangs there for a second, seemingly on the off chance that you'll pick up running right away again, but then the way it lurches back to center as you slide to a halt is upsetting. I would either have the camera snap back as soon as you stop holding input or wait until after the character has come to a complete stop so that the player and camera aren't travelling in opposite directions. Contrary to that, I think the way the camera juts out when you slide is too sudden or two extreme.

Also there are a couple edge cases you might want to handle or guard against:

Got nothing to add, except thanks a lot!! All of it is very interesting and I'll make the appropriate fixes ASAP!

That looks stellar.

Afraid I have no guidance to offer for voicefiles, but thanks! Still needs lots of work, but with everyone's help here I hope it'll turn into something cool eventually

 

[HP_Wuvcraft]

I seek guidance.

What is the best template for voice files?

A) "line # global_name_line # character" [i.e. "1 perry 1"]

B) "line # global_start of line" [i.e. "1 hello"]

C) "line # character_name_line # global" [i.e. "4 morris 5000"]

It's #gifsaturday!

That looks stellar.

 

[wwm0nkey]

Got a little bit done in the time I had to work on it. Mostly getting ready to test platforming stuff but I wont got fully into that until I finish up my own 3rd person controller script. Though I am also testing out a trap/boss ability.

 

[HP_Wuvcraft]

Any SteamDevGAF, are you able to access the backend? I have not been able to all day.

 

[DemonNite]

Any SteamDevGAF, are you able to access the backend? I have not been able to all day.

just tried it and nope, doesn't work for me at the moment.

Anyone? I would really appreciate (and kinda need) your feedback!

I've fixed issues with jumping at the border of platforms, and another with landing->sliding below low height ceilings that would result in getting stuck. There's also a "climbing" marker in mid-air in the bottom right area that can allow you to be suspended in nothingness - though that one isn't so much a problem as a leftover from the previous LD!

If you encounter any other issues, it'd be super helpful to know so I can get on fixing them ASAP while I can still tell what's what from the script!

This looks great. Just gave it a go and a few things popped up (sorry if it has already been mentioned in a different way).

- On the first platform if you drop down and keep the down key held she will get stuck in her falling frame

- When you climb down from a platform edge and drop to the lower ground, her positioning is off (you are not in the right position to jump up and climb). Maybe its a personal preference but it may become a problem if platforms below only have a short edge

- When crawling through the low height parts the camera tracking judders if you don't move

- The small thing green platform section (after the slide) feels off to me. I felt that I should be able to jump up and grab/climb but couldn't but at least one allowed me (far top right)

Hope that helps

 

[Pehesse]

Any SteamDevGAF, are you able to access the backend? I have not been able to all day.

just tried it and nope, doesn't work for me at the moment.

Doesn't work for me either!

This looks great. Just gave it a go and a few things popped up (sorry if it has already been mentioned in a different way).

- On the first platform if you drop down and keep the down key held she will get stuck in her falling frame

- When you climb down from a platform edge and drop to the lower ground, her positioning is off (you are not in the right position to jump up and climb). Maybe its a personal preference but it may become a problem if platforms below only have a short edge

- When crawling through the low height parts the camera tracking judders if you don't move

- The small thing green platform section (after the slide) feels off to me. I felt that I should be able to jump up and grab/climb but couldn't but at least one allowed me (far top right)

Hope that helps

Ah, that's fantastic! Thanks a lot, all of that needs fixing and I have good ideas where to start thanks to your descriptions.
For the last thing (platforming pit with green platforms), if we're talking about the same one, that's actually not surprising you felt that way: it's a rough test area for me to find out how it works without edge climbing markers, so it's not possible to hang on these platforms at all (except the top right one, from the right). I'm considering making different types of platforms, some you can climb on and some you can't, as it would both solve many LD issues and help create some level variety - only trick is going to find the right visual way to convey the info so players are never confused about which is which!

 

[Geminosity]

All the amazing 2D stuff in here makes me think I should probably try my hand at some myself but for some reason everything I keep making is 3d

Funny that I've spent the last few days dabbling with a prototype for an idea involving procedural braided mazes only to find out there's a procedural jam going on. Guess I should check it out and see if I can kill two birds with one stone

edit:

Tried unity last year and after a fair bit of angst i managed to create (basically copying code) a simple level with working pathfinding. Sure I had managed to get a project working and deal with the problem solving that comes with it but i didn't always understand how things worked. Unfortunately with lots of other stuff going on things kinda petered out.

About a month ago with the October break coming up and being broke I decided to give things another go.

A few weeks were spent getting back into things, messing about with either code or pixel art and i managed to pick some more skills and started to get a bit more of a grasp of what was going on especially within unity. The October holidays came and went and things were slowly moving forward but with so many things to learn at times it can be hard to focus on one thing and not get distracted. I finally created my very first project - a simple 2 player shooter in the vein of pong and had some fun with the family playing something that i had actually created.

I was mulling over where to go next - i know i now have the skills to create some simple projects but wasn't sure where to go. I read Derek Yu's book on Spelunky which i found pretty inspiring and it helped me get some focus. Not sure if anyone else went through this phase but i was finding to deal with the abstractness of coding. Trying to mentally visualize things was really tiring.

Anyway had some kind of breakthrough on Friday at work - wrote down the basis of what i wanted to code and finally got it all working as planned this morning. All it does is create a grid and generates a start at the top and a bottom at the finish and generate a random path from top to bottom without allowing backtracking. Looking at what others are generating on here it's pretty small fry but for myself its a major breakthrough and has given me the confidence to take things further.

Hopefully this time i will keep the momentum going and continue to build my skillset.

Good that you're still going at it and making progress :3

If you're doing something procedural you could always jump on the procedural jam I linked above as a way of tying it to a deadline or the like and get something out there
Also, if you're doing procedural paths then you should check out the list of various methods on Jamis Buck's blog. It breaks them all down and shows you them in lil demo applet things so it's handy for getting ideas.
For some reason arguably the best method isn't on that list, but it's covered quite nicely here: http://weblog.jamisbuck.org/2011/1/27/maze-generation-growing-tree-algorithm

 

[titch]

All the amazing 2D stuff in here makes me think I should probably try my hand at some myself but for some reason everything I keep making is 3d

Funny that I've spent the last few days dabbling with a prototype for an idea involving procedural braided mazes only to find out there's a procedural jam going on. Guess I should check it out and see if I can kill two birds with one stone

edit:
Good that you're still going at it and making progress :3

If you're doing something procedural you could always jump on the procedural jam I linked above as a way of tying it to a deadline or the like and get something out there
Also, if you're doing procedural paths then you should check out the list of various methods on Jamis Buck's blog. It breaks them all down and shows you them in lil demo applet things so it's handy for getting ideas.
For some reason arguably the best method isn't on that list, but it's covered quite nicely here: http://weblog.jamisbuck.org/2011/1/27/maze-generation-growing-tree-algorithm

Funnily enough ive been contemplating entering a jam as a way to help my focus.

Not sure if i could get something more worthwhile up and running for the w/e though, shame the main deadline is so soon.

Cheers for the links - always good to have a variety of sources of information

 

[Cute Dragon Here]

Personally I use ScriptableObjects to make items representations and save them as assets, in that way I can drag them and even save them in a Resources folder so I can instantiate them from code. After that you can just keep your inventory in a single list with strings of the names of the items (or a struct with the name and the number).

Interesting...

I've have my Items as a ScriptableObject now and have made my first item. Now, to put them in the game proper - implement some sort of item list, preferably sortable, and some way to display what the player is holding (good for debugging too)

(Probably going to be "smart" and offer sorting by category in addition to A-Z and quantity.)

 

[Geminosity]

Funnily enough ive been contemplating entering a jam as a way to help my focus.

Not sure if i could get something more worthwhile up and running for the w/e though, shame the main deadline is so soon.

Cheers for the links - always good to have a variety of sources of information

No worries :3

I find jams good for just getting simple stuff done and practising the arts of scope, planning, scheduling and actually finishing things (according ot derek yu, it's a skill that improves with practice after all ;D )

 

[MrHoot]

So I really should've been working on combat and more interactions in game but I was feeling down last week so to change my mind I did mostly concept, and worked on the first shopkeeper scene for the game. There's a bunch of different animal shopkeepers (animals, faes) and I thought it would be funny to start with this...odd fellow

Also I started streaming art and animation a lot more on picarto.tv/hootstreams . It's not much about advertising but more that I don't have a lot of people to talk to most of the time so feel free to drop by i play comfy music or sometimes videos like plinkett reviews or little documentaries on the side

 

[Makai]

woah

 

[Pehesse]

So I really should've been working on combat and more interactions in game but I was feeling down last week so to change my mind I did mostly concept, and worked on the first shopkeeper scene for the game. There's a bunch of different animal shopkeepers (animals, faes) and I thought it would be funny to start with this...odd fellow

Also I started streaming art and animation a lot more on picarto.tv/hootstreams . It's not much about advertising but more that I don't have a lot of people to talk to most of the time so feel free to drop by i play comfy music or sometimes videos like plinkett reviews or little documentaries on the side

It's not truly Vanillaware inspired if it's not in a seducing pose!

Spoiler

(/jk, just in case)

... but yeah, looking good :-D

In other Pacha news: this is what giving feedback gets you!

New charge jump move, courtesy of LordKasual, and lots of fixes and improvements (camera, jumpthrough, and most other things mentioned here by Lilith and K Monkey)!

I've also reworked the wall jump to be "sticky", ie: not require additional input from the user to adhere to the wall, so it's possible to simply press "jump" repeatedly when the opposing surfaces are close enough. Will update the prototype soon for those curious to try it out!

EDIT: prototype is up, same link as before! Should display version number 0.015.2, if it doesn't, force refresh it until it does!
Still need to adjust the wall jump air vectors as they feel weird, but I hope I've addressed most of what had been found out so far - now, to find some more stuff that goes wrong! :-D

 

[MrHoot]

It's not truly Vanillaware inspired if it's not in a seducing pose!

Spoiler

(/jk, just in case)

... but yeah, looking good :-D

In other Pacha news: this is what giving feedback gets you!

New charge jump move, courtesy of LordKasual, and lots of fixes and improvements (camera, jumpthrough, and most other things mentioned here by Lilith and K Monkey)!

I've also reworked the wall jump to be "sticky", ie: not require additional input from the user to adhere to the wall, so it's possible to simply press "jump" repeatedly when the opposing surfaces are close enough. Will update the prototype soon for those curious to try it out!

EDIT: prototype is up, same link as before! Should display version number 0.015.2, if it doesn't, force refresh it until it does!
Still need to adjust the wall jump air vectors as they feel weird, but I hope I've addressed most of what had been found out so far - now, to find some more stuff that goes wrong! :-D

Bahah, well it's only one of the shokpeeps (there will also be a fox, a cat and cutie fairies, among others...the fairies will fill the cute quota =p)

Also played through the prototype and really awesome job ! Is the charged jump already in ?

I also maybe noticed an oddity maybe with the animation, if you do a succession of short jumps while staying in place it might look a little bit janky (like she tries to do her flips but since she lacks the time it gets interrupted)

 

[Klart]

So I really should've been working on combat and more interactions in game but I was feeling down last week so to change my mind I did mostly concept, and worked on the first shopkeeper scene for the game. There's a bunch of different animal shopkeepers (animals, faes) and I thought it would be funny to start with this...odd fellow

That looks really great! What kind of game are you making?

 

[Pehesse]

Bahah, well it's only one of the shokpeeps (there will also be a fox, a cat and cutie fairies, among others...the fairies will fill the cute quota =p)

Also played through the prototype and really awesome job ! Is the charged jump already in ?

I also maybe noticed an oddity maybe with the animation, if you do a succession of short jumps while staying in place it might look a little bit janky (like she tries to do her flips but since she lacks the time it gets interrupted)

Owls are cute, too :-D But yeah, this looks great already!

For Pacha's prototype: indeed, in 0.015.2, the charge jump is available! You need to crouch, then hold "glide" (Q on keyboard, LB on gamepad) then jump when it's charged enough.
Should work on both, though gamepad really feels a lot better and if you can, I recommend playing the prototype that way - I'll need to rethink the keyboard commands at some point!

For the animation, yeah, I see what you mean, but I'd rather go for reactivity to player input - if I prioritize the animation to let them play out, it'll give the game a feeling of unresponsiveness. I don't mind the anims being skipped, I'd rather them feel like rewards if the players lets them play out entirely!

 

[TheKroge]

Been a while since I posted! But there are a bunch of things going on in the game now, one namely... an interactive stock market!

You can blow up a corporations space station and you can see the stock prices drop. Or help them in a big enough mission and see it rise.

 

[ken_matthews]

Anybody know what is going on?

The texture looks great in Painter, but gets distorted in UE4 (also happens in Maya). It is the same mesh and same texture maps.

 

[VisceralBowl]

Anybody know what is going on?

The texture looks great in Painter, but gets distorted in UE4 (also happens in Maya). It is the same mesh and same texture maps.

Did you try to convert to tris before importing to UE4? No idea really.

 

[ken_matthews]

Did you try to convert to tris before importing to UE4? No idea really.

I usually let UE4 triangulate on import but I will give that a go later tonight when I get home. Might not be the issue though because it happens in Maya also.

 

[LordKasual]

I am so happy right now.

So, i've programmed in a new style of visual effect that takes advantage of mesh deformation (similar to the effect used on grass) in order to get more mileage out of my avaliable horsepower. It's such a versatile trick, and allows me to do way more with this engine than I originally thought I could. I try to create organic looking effects, which is why i've strayed away from limiting myself to pixel-drawn VFX...and this is pretty much my holy grail thus far.

It's particularly good at things like fire, smoke, aura, anything that has a high-density of particles but isn't particularly voluminous. But best of all, it does so without driving up the sprite count too high.

These gifs are actually from yesterday. Today at work on my downtime, i've refined it a tiny bit, and they look a bit more natural. For reference, this gif had only 3 objects being created every...i think every half second? On a regular interval. Which is a big deal for me, as a simple fireball spell created and maintained way more than that.

But most importantly, the effects that use this technique are now also affected by the wind system that causes the grass to move! Using this, i'm able to fake effects that would otherwise require a crazy amount of particles / animation frames to look right...specifically, i originally planned for Fire spells to be able to dynamically burn the environment (grass, trees, ect.) although I dialed it down because it took too much to have the effect look right without tanking the framerate. But now I can go crazy.

When the wind blows, the fire sways in its direction, and objects that can catch fire have their own Flammability attribute. Looks a bit rough in screens, but is more pleasing in motion. As an added bonus, i've applied a simple dynamic filter to the grass objects that darkens it the longer it's under the flame. Though sadly, I don't think I can use it on tiled backdrops, but for specific areas i'll be able to tailor what can be burned.

The quality of the effect is pretty much limited by the subdivisions in the mesh, which is currently only 2x2. If i bumped it up and just ran it through a loop, i'd be able to remove the hard edges and smooth out the texture when it bends too much. But compared to what I was using before, it still looks so much better. And scales infinitely better.

Burn status currently does Max HP% damage and stacks. Stacking is meant to be done with successive fire spells (resource costly), but now that i've put this feature back in, I can have burning environments stack it quickly at low cost to the player (and over a potentially huge amount of space) adding a bit of strategy.

 

[ZServ]

EDIT: prototype is up, same link as before! Should display version number 0.015.2, if it doesn't, force refresh it until it does!
Still need to adjust the wall jump air vectors as they feel weird, but I hope I've addressed most of what had been found out so far - now, to find some more stuff that goes wrong! :-D

I was thinking about your post a couple pages back about animations, and keeping them under a certain size in the RAM. Have you considered doing any tricks with your drawing to maximize the RAM? The one that came to mind immediately was "puppeteering" the common animations.

So, the upper legs, blue, would be the same source image, but the rear ones would get a color-pass first to make them appear darker. You'd have a SHITTON more images for your PC for like, one action. Like, if you normally have 30 frames for a walking animation, that would become 30 frames for each joint, which is a *ton* more, but across walking, running, jumping, falling-- y'know, various actions, you may end up with less frames, due to not having to redraw each limb and having it take up that space in the RAM. Although, I'm not sure how you'd do this, and it seems very, very complicated.

But, an idea, nonetheless.

 

[LordKasual]

Home, gifs of the same Aura effect with the wind turned on

And fire propagation and burning

 

[ken_matthews]

Did you try to convert to tris before importing to UE4? No idea really.

You were on the right track! Apparently Substance Painter triangulates a mesh differently than UE4 and Maya. Triangulating prior to texturing in Painter fixes the issue. Thanks for the suggestion.

 

[Pehesse]

I was thinking about your post a couple pages back about animations, and keeping them under a certain size in the RAM. Have you considered doing any tricks with your drawing to maximize the RAM? The one that came to mind immediately was "puppeteering" the common animations.

So, the upper legs, blue, would be the same source image, but the rear ones would get a color-pass first to make them appear darker. You'd have a SHITTON more images for your PC for like, one action. Like, if you normally have 30 frames for a walking animation, that would become 30 frames for each joint, which is a *ton* more, but across walking, running, jumping, falling-- y'know, various actions, you may end up with less frames, due to not having to redraw each limb and having it take up that space in the RAM. Although, I'm not sure how you'd do this, and it seems very, very complicated.

But, an idea, nonetheless.

Oh, much appreciated!

If it were any other project, I'd agree, but the main issue I have with this solution is from an aesthetical standpoint: technically, it makes perfect sense, but having anything puppet like would profoundly impact the syle of animation, and would take away from the entire point of making it that way in the first place, which is one of the core reasons for using this specific kind of animation. So while RAM usage is definitely an issue, I don't think cutting corners on the character animation style is the way to go, at least not for the aesthetical aim of the project - I'd rather find ways to optimize backgrounds and logic calls!

But mostly, my issue with RAM usage isn't so much the actual RAM usage, rather the warning message C2 has when going over its arbitrary limit. If I can find how to remove that warning, the whole thing will get that much easier, as I believe C2 is more than able to handle what I intend to throw at it!

(In fact, just to test, I tried to see if I could play the prototype on a mobile browser, and it works perfectly - of course, there isn't much yet, but the RAM usage is already supposedly quite high according to C2! I've recorded videos of it that I'll post ASAP!)