If somebody told me two years ago that I'd still be working on my main character today I would have laughed at them but, well, here we are.
I feel particularly clever for this trick even though it's not perfect, it gets me about 90% of the way I need to go. I wanted the player to be able to aim attacks up and down, and I wanted the animations to have good consistency, and I also wanted to not spend an hour adapting each one. So here we are, a simple transform constraint applied to nine or ten different bones in various amounts, and suddenly I can aim up or down with the torso/head/arms while still have the hair and skirt bones act like gravity goes downward.
Special thanks to Jobbs for believing in me even though he was having a hard time seeing it when I described it to him, and for not hating that I @ him in every post.
thank you :DThat's very cool indeed and makes it much more dynamic. Good job
If somebody told me two years ago that I'd still be working on my main character today I would have laughed at them but, well, here we are.
I feel particularly clever for this trick even though it's not perfect, it gets me about 90% of the way I need to go. I wanted the player to be able to aim attacks up and down, and I wanted the animations to have good consistency, and I also wanted to not spend an hour adapting each one. So here we are, a simple transform constraint applied to nine or ten different bones in various amounts, and suddenly I can aim up or down with the torso/head/arms while still have the hair and skirt bones act like gravity goes downward.
Special thanks to Jobbs for believing in me even though he was having a hard time seeing it when I described it to him, and for not hating that I @ him in every post.
If somebody told me two years ago that I'd still be working on my main character today I would have laughed at them but, well, here we are.
I feel particularly clever for this trick even though it's not perfect, it gets me about 90% of the way I need to go. I wanted the player to be able to aim attacks up and down, and I wanted the animations to have good consistency, and I also wanted to not spend an hour adapting each one. So here we are, a simple transform constraint applied to nine or ten different bones in various amounts, and suddenly I can aim up or down with the torso/head/arms while still have the hair and skirt bones act like gravity goes downward.
Special thanks to Jobbs for believing in me even though he was having a hard time seeing it when I described it to him, and for not hating that I @ him in every post.
If somebody told me two years ago that I'd still be working on my main character today I would have laughed at them but, well, here we are.
I feel particularly clever for this trick even though it's not perfect, it gets me about 90% of the way I need to go. I wanted the player to be able to aim attacks up and down, and I wanted the animations to have good consistency, and I also wanted to not spend an hour adapting each one. So here we are, a simple transform constraint applied to nine or ten different bones in various amounts, and suddenly I can aim up or down with the torso/head/arms while still have the hair and skirt bones act like gravity goes downward.
Special thanks to Jobbs for believing in me even though he was having a hard time seeing it when I described it to him, and for not hating that I @ him in every post.
Impressive writeup, thanks for posting :D I might try it tomorrow, I hope your friend likes it lol. You sound like an awesome friend youself if you go to this length to make a unique gift man!
That's just in animator. In Unity I'll control that bone from script using left stick inputThis is really nice, properly dynamic. Is the goal to have enemy positions dictate "look direction", or will combat actually be based on mouse position? If it's the latter, how will that translate to a controller?
This is a pretty awesome post. Makes using Unity feel really possible for me, as somebody who's only had a small amount of experience with it and was sort of spooked.First off...shout out to all you pros.So a while back there was a thread on a vid Jim Sterling made (Thank God For Him). It was about unity asset flips and how a baby can make a unity game, put it on steam, and make some money. I jokingly said I would make a meme game. 2 people suggested I document it, a while passed and I never got to it cuz reasons, but recently my friends birthday was coming up. I have no money cuz #gradschool, so I decided to make a video game starring him. I also had been making alot of custom birthday gifts (music videos, photoshops, etc).I decided to stuff a few things he likes in the game (Doom, Nintendo, Memes) and some things he hates (The Golden State Warriors, basketball for those who are clueless like me). The result is the game below.You play as my friend, who is sleeping. A spirit talks to him and end u up on a random island where you must collect Dramond Green Coins(basketball player) to unlock a gotta go fast sonic running section, then transition to a FPS battle with a cyberdemon that shoots Stephen Curry(basket ball player) homing mistles at you. Meanwhile hordes of uganda knuckles spawn every so often to distract you while
shouting "You do not no de wae!" I've documented everything I've done and how long it took me step by step with some notes. Keep in mind I have no experience in Unity, 3D modeling, art, animation etc. (I have an engineering background tho)
1) Download & Install Unity. 20min
2) Download & Install Blender. 20min
3) Skim over(AKA rapdily tap the right arrow key) for blender modeling fundamentals. 2hrs
4) Model Level. 2hrs
5) Skim over videos on texturing in blender. 30min
6) Texture level. 2hrs
7) Make new Unity Project and get adjusted to UI. 20min
8) Search for third person asset on store(Invector or something like that) and test it. 35min
9) Download and use Makehuman to make character model of my friend. 1hr15min
10) Figure out how to swap my custom model with asset flipped third person thing, and test and tweak. 50min
12) Character can't be controlled in midair platformer style so read over code in assets. 1hr
13) Read Unity Manual on some basic transform/physics API stuff then add code for mid air control. 30min
14) Import/setup my level model into scene and tweak level and copy objects around to make bigger level. 1hr5min
15) Add level textures and test out level. 45min
16) Model and Texture coin then place coins around level. 35min
17) Read Unity Manual on animations and animate Coin. 20min
18) Read Unity Manual on how gameobjects work and interact then write pickup script. 40min
19) Use palm tree and water asset, and place around level. 30min
20) Download sound effects and music, and add code to make them work after reading Unity manual about audio. 1hr
21) Animate a floating platform for cheap shitty floating platforms. (I think u should probs code this, but im lazy) 20min
22) Research how to make cutscenes and make 2 cutscenes to unlock next portion of level, and a secret coin. 1hr15min
23) Code up secret coin puzzle and 2nd portion of the level(rly short meme related thing) after learning how particle systems 'and user input works in manual. 1hr15min
24) Test and tweek things out up to this point. 1hr
25) Realize game is too short, and want to add FPS segment to game so download FPS asset. 20min
26) Read FPS Constructor asset manual (yeah it has a manual), and add/test basic functionality into scene, and fix a shit ton of bugs cuz it was filled with em. 2hr
27) Find cyberdemon model for final boss segment. Its not rigged so I skim over rigging/animation tutorials for blender. 1hr30min
28) Rig model and Animate intro cutscene for boss. 5hrs <---- yeah this sucked ass. God bless you artists
29) Import into Unity, and create cutscene and platformer to FPS mode code. 1hr
30) Code up boss behavior after reading Unity manual on animation and physics stuff. 2hrs
31) Don't want to spend another 5hrs animating so I learn about mixamo which automatically animates thins, and use it for in game boss animations. 30min
32) Test and tweak things. 1hr
33) Model, texture and code cyberdemon rocket behavior. 30min
34) Get uganda knuckles model, use mixamo for animaton, and use included FPS damage detection script, code basic "AI". 40 min
35) Create and add intro, and outro cutscene after downloading lightworks video editor(free). 1hr
36) Figure out, and then fight with the lightmapper, and occlusion culler. 3hrs........(NO UR OBJECT IS TOO BIG UNITY!!!!!)
37) Test everything myself and make minor tweaks. Realize Im spending too much time, and finish things up. 1hr
Total time....40hours soooooo yeaaahhh thats how long it takes to make a shitty asset flipped unity game with no experience.If I learned anything its that Unity is great (except for the lightmapper), but god damn does steam need to step up on curation. The game is janky as all fuck, but only because I had no idea what I was doing. Unity as a tool does exactly what it was intended to do. Make makeing games(and any subset of games shitty or not shitty) easy. The blame is def not with Unity, but rather with steam. As a curation/drm tool steam fails miserably because ...well its obivously not working lol. As for performance... cant really say. Haven't looked into it. The game runs like ass, but only because I
just learned how to use this thing. Def could make my objects smaller. Organize materials n shizz. Runs fine on my PC, but if you throw it at a Intel 530HD on a Mac or something then its like at 30fps @ 2k on medium. I did the bare minimum to make sure it runs on a surface Pro 3 (my friends PC) All I know is that if you try you can actually make Unity look pretty and perform well. (Assault on Android Cactus, Cuphead, etc). Now to go try out Unreal Engine lol....maybe for another
friends birthday tho. lol
DL link:
PC: https://drive.google.com/open?id=1yOU70RWSTWgzgW3OdhkbbLWUDcnPM1S3
MAC: https://drive.google.com/open?id=1G8iOWkmVzhoOECtgXRi5rc2kKDMoAExw
Great job sabrina :DIf somebody told me two years ago that I'd still be working on my main character today I would have laughed at them but, well, here we are.
I feel particularly clever for this trick even though it's not perfect, it gets me about 90% of the way I need to go. I wanted the player to be able to aim attacks up and down, and I wanted the animations to have good consistency, and I also wanted to not spend an hour adapting each one. So here we are, a simple transform constraint applied to nine or ten different bones in various amounts, and suddenly I can aim up or down with the torso/head/arms while still have the hair and skirt bones act like gravity goes downward.
Special thanks to Jobbs for believing in me even though he was having a hard time seeing it when I described it to him, and for not hating that I @ him in every post.
Thank you! It's 2D. I've just moved and scaled the bones a lot to get twisting and foreshortening.That is really well animated! Im having a genuinely difficult time figuring out if that is a 3D model or 2D assets "marionetted" together. (I assume its the latter though based on what I've seen of your game before). If it is 2D, im not normally a fan of that style of animation as it can sometimes end up looking a little soulless, but you've done a fantastic job of adding detail and bringing character to it! Keep up the good work!
Sure.Great job sabrina :D
I hope you can give us a closer look at what exactly you did to make this work! We could all learn from it!
Thank you! It's 2D. I've just moved and scaled the bones a lot to get twisting and foreshortening.
Sure.
The main trick is in that in some places where the bone hierarchy normally goes parent -> child it now goes parent -> rotator -> child. The rotator bone isn't keyed in animation, just uses what Spine calls a transform constraint to partially or fully match the rotation of the aiming bone. The reason for this is so that I can key an animation in the child bone and not have to mess with appropriately offsetting each keyed value to aim differently. Effectively the rotation of the rotator and child bones get added together to create a new angle at runtime. So the waist gets about 40% of the angle, the upper torso gets almost 30% more, the neck gets 20%, and the head gets the final 10% of the angle added to it. This provides a reasonably natural distribution of the bending. Then the hair pieces get negative the full amount to compensate.
The other somewhat unique thing I'm doing for foreshortening the sword is I have a parent bone that doesn't have any translation or rotation keyed in it, just scale. Then the child bone keys rotation. The effect is I can turn the circular rotation of the sword into an elliptical rotation.
Thanks for your help. Right now I'm struggling on how to manage and what to say in our discord. I'm supposed to be the one in charge, but I'm so indecisive and unsure. I don't want to end up saying the wrong thing having people leave. I'm so nervous right now.Maybe do a random set of cards. Have a set for mechanics and another for setting that you are willing to do. Then mix the two sets up separately and pull one from each. So you have like Dystopian Shoot'em Up, or Fantasy First Person Shooter.
If somebody told me two years ago that I'd still be working on my main character today I would have laughed at them but, well, here we are.
I feel particularly clever for this trick even though it's not perfect, it gets me about 90% of the way I need to go. I wanted the player to be able to aim attacks up and down, and I wanted the animations to have good consistency, and I also wanted to not spend an hour adapting each one. So here we are, a simple transform constraint applied to nine or ten different bones in various amounts, and suddenly I can aim up or down with the torso/head/arms while still have the hair and skirt bones act like gravity goes downward.
Special thanks to Jobbs for believing in me even though he was having a hard time seeing it when I described it to him, and for not hating that I @ him in every post.
The 2d stuff has definitely improved if it's been several years for you, yeah. I haven't made any platformers myself though.It has been several years since I tried using Unity, and at the time I had a terrible time trying to make a 2D game simply because the controls were so floaty and I couldn't find a way to get them to control sharply. Is it fairly easy for a newcomer these days to get a 2D game started that has a more classic feel such as Mega Man, or Celeste as the most recent example?
It has been several years since I tried using Unity, and at the time I had a terrible time trying to make a 2D game simply because the controls were so floaty and I couldn't find a way to get them to control sharply. Is it fairly easy for a newcomer these days to get a 2D game started that has a more classic feel such as Mega Man, or Celeste as the most recent example?
Ooh, neat. I've never looked into the asset store. No problem spending money if it saves me time. :)Those classic controls are not that easy to pull off, specially with modern engines. You propably need to build the whole system ground up, basic unity physics nly get you halfway. You can also try to check out the Unity asset store for platformer systems, I bet theres something that would get you started.
I gave the store a quick search and found this Corgi Engine - 2D + 2.5D Platformer system, looks exactly what you are after. There are probably free similar starter kits as well if the cost is an issue to you. Unity Asset store is absolute gem :)
Another week, another set of anim sketches down... a little behind schedule, but there are only a few left so I should be done with those next week, so maybe I'll have actual integrated blue sketch footage next week! :-D
I haven't really posted here in a while because my game project reached a point late last year where I determined proceeding in the old engine wasn't going to be feasible - It just wasn't going to perform well enough for users (and there was really no reasonable way of porting to consoles from that engine).
Once I started learning Unity and Spine 2D I decided to re-examine everything about the game, visually and gameplay-wise. One of the new things is melee.
I'm going to add more FX and stuff to spruce this up, but for now it's just nice that I got it working. Sorry for the large gif size.
I haven't really posted here in a while because my game project reached a point late last year where I determined proceeding in the old engine wasn't going to be feasible - It just wasn't going to perform well enough for users (and there was really no reasonable way of porting to consoles from that engine).
Once I started learning Unity and Spine 2D I decided to re-examine everything about the game, visually and gameplay-wise. One of the new things is melee.
I'm going to add more FX and stuff to spruce this up, but for now it's just nice that I got it working. Sorry for the large gif size.
Very impressive, as usual! It looks so much more dense than the previous version, and all this background animation really sells atmosphere.
Regarding moving to Unity, I'm in a similar position where I already know C2 won't allow console development, so I'll have to port eventually (likely to Unity), but I still want to develop a working final version of the game for PC users first using C2, and use that as a base to port after receiving feedback. Not sure if it's the best move, or if I should switch before first release like you... it'll depend on tech issues encountered along the way, I suppose.
Are you planning to release the Stencyl version of Ghost Song sometime down the line, as I seem to recall you did with the very first prototypes way back when?
it will be really great to hear a post mortem on the switch to Unity when things have settled. Would love to hear the ups and downs of the approach :)
Great work btw ! ;)
Thank you. :)
I really miss scripting in Stencyl. Scripting in C# is basically the same -- It's more the same than I ever thought it would be -- But the visual language just felt more intuitive. Staring at code remains difficult for me and it takes me a bit longer to sort out. Other than that, though, with Unity and Spine 2D it's like experiencing the world with brand new eyes. There are so many things that I can easily do now that never would have been possible before. It's like growing up.
Regarding your project and your plan to finish it in C2 -- My initial reaction is that if you're going to change you should do it sooner rather than later. The initial learning phase might be a bit of a drag, but once you get the hang of it I suspect you'll have a similar reaction as I did and kick yourself for waiting so long.
I have no plans to release the Stencyl game. There are a multitude of reasons why I don't think that'd be a good idea.
Started on a new area, would love all the feedback i can get on it.
Right almost done with the beehive level and I am pretty proud of how this is turning out! Lots of turns and twists to keep the player entertained. Thanks to the honey current gimmick, it really is now starting to feel like a giant beehive too!
I've been on a bit of a self doubt when it comes to indie deving recently but now that everything is starting to feel CONNECTED I'm getting pumped up again!
If somebody told me two years ago that I'd still be working on my main character today I would have laughed at them but, well, here we are.
I feel particularly clever for this trick even though it's not perfect, it gets me about 90% of the way I need to go. I wanted the player to be able to aim attacks up and down, and I wanted the animations to have good consistency, and I also wanted to not spend an hour adapting each one. So here we are, a simple transform constraint applied to nine or ten different bones in various amounts, and suddenly I can aim up or down with the torso/head/arms while still have the hair and skirt bones act like gravity goes downward.
Special thanks to Jobbs for believing in me even though he was having a hard time seeing it when I described it to him, and for not hating that I @ him in every post.
I haven't really posted here in a while because my game project reached a point late last year where I determined proceeding in the old engine wasn't going to be feasible - It just wasn't going to perform well enough for users (and there was really no reasonable way of porting to consoles from that engine).
Once I started learning Unity and Spine 2D I decided to re-examine everything about the game, visually and gameplay-wise. One of the new things is melee.
I'm going to add more FX and stuff to spruce this up, but for now it's just nice that I got it working. Sorry for the large gif size.
You can switch back and forth.Cool stuff, as many have said. But also, do you dual wield the sword and rod?
good call. Thank you. I've traditionally animated her face to be more expressive, but I was mostly focused on hand and legwork here.Can I make a tiny suggestion that I think would improve the feeling of the animations for relatively little work? Give the eyes some more movement, specifically:
1) Make the character look up when attacking up, and down when attacking down. That is, make her move not just her head, but her eyes too. This may mean actually moving her head less, to compensate.
2) Make her eyes move when idle, ideally imitating eye saccades (jerky, sudden eye movement that is how people's eyes work). In my experience this is what sets apart decent animation from lifelike one, even in professional games and animation.
good call. Thank you. I've traditionally animated her face to be more expressive, but I was mostly focused on hand and legwork here.
WELP, should put the "don't be afraid to ask" advice into practice!
Is there anyone here who does pixel art/sprite work who can give a... what's lower than a novice? well, who can give ME and anyone who's interested some tips on getting started on sprite work? It's kinda tough not know where to start or how to get things started, and all I could come up with was picking up drawing again(used to draw when I was younger, but quit because i'm an idiot), and recreating sprites using piskelapp(my profile pic of terry doing buster wolf was done there, and I was pretty proud of it). Any advise/tips would truly be appreciated from you talented folk here! I'm sorry if this has been asked before, but i'm a little new here and don't know how to search through the pages here haha
Finally got Ancient Greece Trap up to par with the rest of the levels visually, now back to proper level design work!
Oh jeez, thanks so much for the tutorials! What you described is exactly what I did with the Terry sprite I did for my profile picture, so good to hear that I did SOMETHING right haha.I love to draw pixel art from time to time, but I'm average at best, so I'm sure there are people here who will give you better advices than me. My top advice is search inspiration in other pixelart, download and zoom them to look how they are drawn and copy their style. Don't be ashamed for copy them, you're learning.
Other tip, try different drawing tools to find a confortable workflow for you.
Some awesome tutorials are from artists and game developers: Derek Yu , finalbossblues, Tsugumo and Pedro from Miniboss(click into each title of Pedro's tutorials to learn more about each topic).
Also, you can find a lot more here.
Line 271: error X3065: object assignments are not allowed inside functions
float4 main_fragment (float2 tex : TEXCOORD, uniform sampler2D s0 : TEXUNIT0, uniform input IN) : COLOR
{
float2 pos=Warp(tex.xy*(IN.texture_size.xy/IN.video_size.xy))*(IN.video_size.xy/IN.texture_size.xy);
s0_global = s0;
IN_global = IN;
float3 outColor = Tri(pos);
//Add Bloom
outColor.rgb+=Bloom(pos)*bloomAmount;
if(shadowMask)
outColor.rgb*=Mask1(floor(tex.xy*(IN.texture_size.xy/IN.video_size.xy)*IN.output_size.xy)+float2(0.5,0.5));
return float4(ToSrgb(outColor.rgb),1.0);
}
s0_global = s0;
Speaking of "it never hurts to ask," I don't suppose anyone wants to help me figure out an issue with a shader? I posted about it on Stack Overflow a couple days ago and nobody's responded.
I'm essentially trying to adapt a RetroArch CRT shader (Timothy Lotte's public domain CRT shader) to work in Clickteam Fusion. I have never edited shader code before in my life, and Clickteam's resources on understanding shaders in general is basically nonexistant -- everything they've written is from the perspective of you already knowing how a shader works. Clickteam Fusion does, however, provide its own shader editor of sorts that at least spits out compile errors.
It started out with me talking about a compile error on twitter and a friend walking me through fixing each error one by one, until we got to nearly the end of the shader file, with this error:
That pertains to this piece of code:
Code:float4 main_fragment (float2 tex : TEXCOORD, uniform sampler2D s0 : TEXUNIT0, uniform input IN) : COLOR { float2 pos=Warp(tex.xy*(IN.texture_size.xy/IN.video_size.xy))*(IN.video_size.xy/IN.texture_size.xy); s0_global = s0; IN_global = IN; float3 outColor = Tri(pos); //Add Bloom outColor.rgb+=Bloom(pos)*bloomAmount; if(shadowMask) outColor.rgb*=Mask1(floor(tex.xy*(IN.texture_size.xy/IN.video_size.xy)*IN.output_size.xy)+float2(0.5,0.5)); return float4(ToSrgb(outColor.rgb),1.0); }
In particular, the exact line it's failing on is:
Code:s0_global = s0;
s0_global is defined as part of sampler2D, and the friend suggested the code was in HLSL format, something he wasn't that familiar with. The original file was a .cg, I know that much, and Clickteam Fusion seems to only take .fx, which is probably relevant.
I've tried just commenting out the line its failing on, but that just gives me a new, different "There were no techniques" error that doesn't even give me a line number.
I am super super deep in over my head, here.
Here's the full, complete modified Clickteam Fusion (.fx) shader file I'm working with
Here's Lotte's original CRT shader that works in RetroArch (.cg)
If anyone has any information to share, I'd love to hear it. I have a sinking feeling this is a bigger problem than just a failure on line 271 and that I'm actually a huge idiot.
I'm no expert (just had some courses on OpenGL)Speaking of "it never hurts to ask," I don't suppose anyone wants to help me figure out an issue with a shader? I posted about it on Stack Overflow a couple days ago and nobody's responded.
I'm essentially trying to adapt a RetroArch CRT shader (Timothy Lotte's public domain CRT shader) to work in Clickteam Fusion. I have never edited shader code before in my life, and Clickteam's resources on understanding shaders in general is basically nonexistant -- everything they've written is from the perspective of you already knowing how a shader works. Clickteam Fusion does, however, provide its own shader editor of sorts that at least spits out compile errors.
It started out with me talking about a compile error on twitter and a friend walking me through fixing each error one by one, until we got to nearly the end of the shader file, with this error:
That pertains to this piece of code:
Code:float4 main_fragment (float2 tex : TEXCOORD, uniform sampler2D s0 : TEXUNIT0, uniform input IN) : COLOR { float2 pos=Warp(tex.xy*(IN.texture_size.xy/IN.video_size.xy))*(IN.video_size.xy/IN.texture_size.xy); s0_global = s0; IN_global = IN; float3 outColor = Tri(pos); //Add Bloom outColor.rgb+=Bloom(pos)*bloomAmount; if(shadowMask) outColor.rgb*=Mask1(floor(tex.xy*(IN.texture_size.xy/IN.video_size.xy)*IN.output_size.xy)+float2(0.5,0.5)); return float4(ToSrgb(outColor.rgb),1.0); }
In particular, the exact line it's failing on is:
Code:s0_global = s0;
s0_global is defined as part of sampler2D, and the friend suggested the code was in HLSL format, something he wasn't that familiar with. The original file was a .cg, I know that much, and Clickteam Fusion seems to only take .fx, which is probably relevant.
I've tried just commenting out the line its failing on, but that just gives me a new, different "There were no techniques" error that doesn't even give me a line number.
I am super super deep in over my head, here.
Here's the full, complete modified Clickteam Fusion (.fx) shader file I'm working with
Here's Lotte's original CRT shader that works in RetroArch (.cg)
If anyone has any information to share, I'd love to hear it. I have a sinking feeling this is a bigger problem than just a failure on line 271 and that I'm actually a huge idiot.