Fixing this is on my radar, it’s just going to be a heck of a process to do correctly. If I were to add rims, I’d love to do it correctly and let people pay to upgrade theirs.
1 Like
You could just keep what you have now and use meshes.
3 Likes
I was dreading this all day but I made the IK work with any angle (before it only works if all the joints were at 0 rotation relative to the origin, aka the torso). I don’t know how else to explain it other than ‘relative solution’.
Now I found a solution and it’s actually about 57% faster. I realized that if I pack everything into one CFrame*Vector3 that I could extract a bunch of data from it that would be much cheaper than what I’m currently doing.
Old: 
Basically if I rotated the torso in any way before the IK would fail to recognize that change. Now it just includes it. The dankest part (imo) is that it only updates the CurrentAngle and it only does that when the Target’s position relative to the torso changes, meaning that if your hand animation is complete the IK won’t do any more equations to find that elbow angle.
I just used renderstepped+noise to rotate the rig to show what I mean by ‘relative solution’.
3 Likes
The real question is, when will you finish your revamps and actually make the KC game as great as what you’re showing off. I’m eager to see what happens when you put it all together in a polished state.
2 Likes
That is exactly why I have nothing to show off myself.
1 Like
I can confidently say this is the last time I have to redo these rigs. I’ve tried a ton of different stuff. Some of it failed, some of it didn’t. I’ve thrown out a lot of optimizations that ended up failing and have rewritten code to accommodate for new working optimizations. I could have ‘finished’ the game months ago with a ton of content and such but then my 96-player servers (at least 64, 96 would be ideal!) wouldn’t be possible. I’d have to go back and re-animate so many guns, characters, taunts, etc… and I’d have a lot of characters to re-rig, tons of code to re-write and a crapton of work to throw out.
I’ve rewritten so many portions of the game so many times but I’ve slowly been checking off a made-up checklist “Final animation format? Check. Final weapon system? Check. Final Control script? Check”. Next is “Final character rigs” and “Final foot planting”. At this rate (5-14 hours a day) I’m able to say that the game will be ready to really “play” by this Christmas.
Also I just like polish and the standards are really high for this game 
They aren’t even set by me either, lito said something along the lines of “the reason PF stuck to the simpler body types is because when you add things like fingers, legs, arm, etc… (human-form) players expect to see a lot higher quality and that takes time and effort”. Personally I think it’s hella fun kinda “pioneering” this stuff on roblox.
2 Likes
I look forward to this Christmas then. Although, I find a even 64 player count to be next to impossible. It just seems too impractical for what the engine is capable of. I’m not the only developer that finds your player number goal to be really ambitious.
But I still paid for KC, and im waiting to see just how far you can push it.
If you think that’s nuts you should hear how many vehicles and destructible buildings I’m adding.
There’s a reason the game is being squeezed for every ounce of extra performance
First draft of a prone state on the apoc 2 characters. Lots of tweaking to do but it turned out fairly well. Ignore the terrible hat and backpack offsets. Go ahead and forget you saw the limb popping too, that’ll all be fixed soon enough q_q.
17 Likes
This is what I do, except instead of starting a new side project I just decide to add another thing to my game, or remake some part of it.
But this is exactly the reason I decided I have to finish my current game, no matter how it turns out or how long it takes. I have had SOOO many ideas for awesome new games that in my opinion would be far superior to the one I’m working on, but I’m sure that if I started on that idea I’d have another idea, and another idea, and another idea…
Been doing a lot of character work for a game I’m working on with a few friends. Animations are still a WIP but looking pretty decent so far.
https://i.gyazo.com/f437f854aec48125ea69977a8eb03b50.mp4
https://i.gyazo.com/aeb2a5d727f949072a2d7ccf0f64461d.mp4
https://i.gyazo.com/bedd452ea5d2c1dd6ac89d0de6e6b133.mp4
I wonder why official R15 isn’t using sphere joint connections, nonetheless I’ve made a pretty good alternative if I do say so myself.
14 Likes
love that first person view.
4 Likes
I like the animations, hand motions seem too stiff though. Could use a more swinging motion as arm swinging is more momentum based and when it gets close to the max point it gradually slows down.
Also when walking backwards and sideways I feel like the arms from the elbows down should raise up a little bit.
Just going by my own walking posture.
1 Like
Try putting spaces between details to prevent the inside from appearing without clicking it 
Most of the magic is in the legs right now, the arms are currently on a pretty rigid track. Still working out a lot of the motions since it’s all procedural, no keyframes are being used.
Makes sense. Keep up the good work!
Because that’s not how limbs work 
Try pulling your arm out, with hand opened pointing up. Now rotate your forearm by 90 degrees on left/right
Good luck in breaking it
Am I doing this right?
Also, that is most certainly how limbs work. The shoulders, hips, and even pelvis use ball and socket joints. Elbows and knees also use a similar type of joint, although it only allows for rotation on one axis.
1 Like