What are you working on currently? [2016]

Maybe, but I’m basing it off of how Mass Effect does it:

And also, moving it closer to the shoulder would mean I had to extend the fully-extent left arm even further to reach the grip. Rotating the stock towards the right shoulder and the front of the gun left would make it easier to grip, but I have the rotation locked so the barrel is always pointing in front of the character (nobody shoots off to the side)

1 Like

Couldn’t quite get a standard crouch animation you’d expect from a game due to so much height of the character being in the torso, but I did manage to come up with something that looked believable I think:

R15 is interesting to animate with. My only wishes were that I could control the fingers and that the animation editor supported IK.

10 Likes

I’d suggest having multiple hand meshes with different poses welded to the normal hand (with the same offset so the wrist matches up), or just having a file mesh on the hand with the meshID being changed whenever you want to change finger position.
it’s a crude and basic system, but it works(maybe, i’m tired and not thinking very good rn)

That would work great for static poses, but not transitioning between them. For instance, imagine an animation where I insert a fresh magazine, let go, make a fist with my hand, and then slam the magazine into the gun from the bottom. I’d have to get from “gripping mag” to “open hand” to “fist” and back. In order for the animation to look fluid, I’d need a lot of frames in between and that means lots of unique meshes. That’s a lot to preload (don’t want invisible hands during reload) and IIRC swapping out meshes like that isn’t performance-friendly.

Something that’s been on my mind is breaking the hand mesh into multiple pieces so I can control each segment of the fingers. That would turn the R15 into R45, so I’d have to think about how to prevent that from becoming a performance issue – perhaps only distinguishing fingers on the player’s character.

1 Like

r45 on only locally?

1 Like

I’m not necessarily worried about replication – I believe the purpose of joints is to stop replicating Part1’s CFrame and instead only send updates to the joint, saving the need to constantly replicate CFrame and only send changes over while the joint is being animated. Though I’m not sure on this – maybe @zeuxcg can correct or affirm this.

Pre-unions/meshes, I would see guns that were 100 parts and caused serious performance issues, even though the gun components remained static once equipped. I’m not sure why this was (components were CanCollide=false and joints were not being changed), but I imagine a similar situation would happen with the fingers.

King’s Landing in SmoothPlastic, we’re nearing completion :slight_smile:

18 Likes

I think you guys might enjoy what I’m working on, and I’ve been looking forward to showing off what I’ve got so far.

I’ll start this off by showing everyone this screenshot.

Now some of you might be thinking “Oooh neat, he put the R15 version of his character into Blender. I wonder how long it took him to assemble that.”

But heres where things get CRAZY.
I didn’t assemble this myself, it was automatically generated using a program that I wrote entirely in C#.

For the past few days, I’ve been working on rewriting my Rbx2Source program from the ground up. For those of you who don’t know about it, Rbx2Source is a tool I made that ports roblox characters, and other items into the Source Engine.

This new version of the tool rips the .mesh files and models directly from the website, parses them into data that my program can read, and does some crazy stuff to build it into an actual mesh with skeleton data. Its compatible with both R6, and R15, and it works with pretty much any catalog assets you throw at it.

Now I’m not going to go into the technical details of how all this works, because that would add at least 5000 more words to this post.

The program will eventually be open source once its completed, but theres still a lot of stuff I have to do to make it user friendly.

This has been a huge test of my endurance to get working correctly, and I’m really happy with what I’ve got so far. Kudos to @EgoMoose for porting ROBLOX’s CFrame datatype into C#, and kudos to my friend AJ (@RedTopper) for helping me reverse engineer the binary .mesh file format.

Stay tuned for more info :slight_smile:!

9 Likes

the final test is importing animaiton data back into roblox @_@

Thats next on my to-do list :smiley:
It should be fairly easy.

Taking a bit of a break from Cloven Hoof development, here’s another “upscaled” logo.

6 Likes

This has been a dream of mine. Keep up the great work! I always look forward to seeing the crazy cool things you come up with.

1 Like

Mean, I herd you were good but cow that is much better than I expected hoof you.

2 Likes

Keep this up, and I’ll have some real beef with you

Sorry for the crappy image quality (I even took them at 5k resolution!) but the lighting outside + my room is terrible so it looks bad on my phone.

Here’s my in 6th grade vs me today, almost 10 years later. Pretty glad to be surrounded by my work :smiley: For reference, the Kinetic Code poster is 72 inches (6 feet) wide. (and looks much better than my camera can show sadly :frowning: )

(click to zoom in if you want, the resolution is pretty high)

http://i.imgur.com/LtoSqEm.jpg

http://i.imgur.com/Gp4B1Iu.jpg

4 Likes

dang, I just made a stained window + glass fully by hand

For comparison resolution-wise, here’s the image dump I’m getting most these logos off of for this.

1 Like

I’m working on some next generation emergent gameplay.

http://puu.sh/szOjg/63a18bb843.webm

13 Likes

Phantom Forces has nothing on this.

die die die but then just slightly faster