[Tier 3] Juno's Post-Workout Protein Project Breakdown (Patreon)

Animation 5 (Juno's Post-Workout Protein) Was not intended to be such a big project. I had a simple idea of Junkerqueen and Juno in a reddish-pink locker-room. It was originally intended to be like 2 scenes after the intro sequence. However, the more I worked on it, the more it just kept expanding and expanding, kind of like Juno. I also really liked the environment I made for it so I didn't want to waste it on like a quick 1-2 minute animation. Anyways here is how I got things started and where it went from there and the problems that I had along the way.

1. Character Rigging

This is pretty much the beginning to every project right now because all of the characters don't have any jiggle cages set up. For both Juno and JQ I set up breast cages, thigh cages, and butt cages. In the end I never used JQ's butt and thigh cages. WAIT, I almost forgot I also made a cage for Juno's belly to simulate subtle jiggle and the inflation which is something I have never done before.

Also, I know I have talked a lot about jiggle physics, but I think I finally may have mastered it. The trick is to unbind then rebind the surface deform modifier after the beginning of the cloth physics' baked sequence on the cage mesh. That sounds confusing as hell so here it is simplified. To get jiggle physics you need a cage around the part of the body that's gonna be moving. You also need a surface deform modifier on the character mesh with a specific vertex group so that area follows the cage and mimics its physics. My problem before is I was binding it at frame 0, so (lets say the breasts) would just instantly inflate and go all weird because it would be perfectly matching the physics AND shape of the cage. If you rebind the surface deform, let's say at frame 50, after baking from frame 0, then the breasts will only mimic the physics and not the shape. Sorry again if that is confusing but it is very important for jiggle physics. (That is also how I got the balls to move so well in animation 6)I also needed the characters to be sweaty/wet, like how they are in animation 4 (Ashe and Illari). So, I applied the same material with a mix shader setup. But I did have one problem with JQ, it would not work on her head, all it did is make it kind of shiny. I got around this problem by simply enabling her choker so no one can see the weird border of moisture on her neck.2. Building the Scene

Going into this project I wanted to do the classic locker room setting because the characters can already be naked. In my mind I had an idea for small red lockers in a pinkish-salmon colored locker room. When I finally completed the scene, I was really happy with how it turned out because it literally felt like I took an image from my brain and brought it to real life. I made most of the objects and some of the materials, but a lot was royalty free to use from BlenderKit. I feel like the lighting could have maybe been a little bit better but I think it is fine. I am no expert in lighting at all.

3. Animation

Animation for this project was generally straightforward as always. The thing that I struggle with is setting up the poses. This was also my first time animating speech and moving mouths when talking so that was very interesting. When I first started this project, the opening intro scene was immediately animated out and finished (animated the mouth later on). Then instead of going right into a sex scene I wanted things to start out a little slow and give the audience a little anticipation I guess with the whole smacking on belly and insertion scene. Next, I had to animate scene 3 which is where JQ climaxes at the end. That whole part turned out much better than I was hoping. Scenes 4 and 5 were really fun to animate and setup. Juno's breast collisions on the locker were a bit tricky to setup just because everything kind of had to be the perfect distance. And the inflation was pretty easy, just keyframed the pressure on the belly cage mesh so it would inflate when I want.

I also played around with lattice's for this project and the vertex groups that they would affect. I used them to give Juno the stomach buldge and I also used them in the scene above to give her thighs a little bit of a ripple effect.Another small thing I did was give Juno a red face / a blushing effect. This was really easy, just made a new pink material and hand painted it on her face using an image texture and a mix shader on her head material. I made it fade in by attaching the mix shader of the blush to another mix shader so I could keyframe the factor value. The setup is pretty messy, but it worked:

Both of the final scenes had a lot of sticky liquid and strands which I also feel like I may have mastered. Basically, to setup a liquid strand between 2 characters I start with making a low poly cylinder, scaling it down, and stretching it out to its desired length.

I then add an edge loop or 2 very close to the edges of the cylinder. and then scale up the end vertices, apply a subdivision modifier, and shade smooth.

Now that the shape is done, you select the end verts on one side, press ctrl h, hook to a new object and do the same for the other side. Now you can attach each end to wherever you want, and make it stick/follow the character by giving it a child of constraint. Select the target as the character mesh, then select the vertex group for that area. You might have to make a new vertex group but that is very easy. Also, there might be a better way to attach it to a character, but this is the way I found. However, this is just for a strand that stretches. If you want a strand with physics then you need to add more edges loops so it can bend, make a Pin vertex group on each end, then give it cloth physics. And of course, a material for whatever type of liquid it is.

Also to get all the cream on their bodies I used Nyl's built in vertex paint method that comes with the characters, along with the method that I have been using since my first animation where I hand paint over the characters texture. The liquid simulations for this project were also pretty fun, and I did not have any problems. There were 4 total (the plug popping out, the small squirt in scene 2, the flowing in scene 3, and the flowing in the final scene). Each of them were creating using simple inflow objects, keyframing when they flow, and making sure the initial velocity or collisions were right. For the final liquid simulation, I did utilize physics again like what I did with animation 5. I basically made an invisible funnel that followed the bottom of Juno. That is what gave the stream that kind of squirt effect when she would be pushed down. I don't know if it was really needed but I think it looked ok.

4. Problems....

Eevee caused me a lot of pain during this project. Raytracing looks so good but holy it can cause a lot of problems when it comes to animation and moving objects. I will include some gifs to show the problems that I faced. But to sum it up, eevee can basically make weird, unrealistic shadow or reflection artifacts. Because of this I had 3 big noticeable problems throughout the project. The first was when it zoomed in on JQ crossing her arms, the shadow on her arm would also noticeable creep upwards. This is because eevee doesn't render what is outside the frame so I had to fix this by enabling overscan so it would render farther than what is seen.

The next problem I had was where certain parts of the character would make VERY unrealistic, weird shadows on the ceiling. This was fixed by going into the ceiling material settings and changing the rendered method from dithered to blended. So raytracing would basically ignore the roof (I think)

I was able to fix these 2 problems, but I could not fix the issue in the final scene where the camera pans down from JQ's back to Juno, and Juno's butt weirdly fades in. It goes from all shadowed to bright. I tested an insane overscan amount of 10% but it did not fix it. So, it is still there and very noticeable in the final product...Another problem I had was when I needed collisions on Juno's back in scene 4 so JQ's breasts would collide with her back and not swing around through Juno. At first duplicated a large portion of her back then did a child of constraint so it would stick and follow her back, then I gave that duplicated back object collisions. Yeah, don't do this lol. It created a ton of problems where my surface deform modifiers would not work anymore, and me being a dumbass I did not know it was the back object I attached to Junos back for collisions. So, I spent hours trying to fix it and remaking cage meshes and surface deforms etc. I ended up just hand animated the back object so it would follow. All of these problems usually take me hours to fix, and it is pretty funny every time because they usually have the easiest fix such as changing one small setting lol. 5. Postproduction & Audio

I wanted this animation to be pretty bright and vibrant so getting the adjustment layer in Premiere-Pro just right did take a bit of time, but I think it came out good. I also commissioned a voice actor for the first time which was also really cool. And my gosh it probably saved me hours or even days. The fact I had to do barely any voice audio editing was such a relief compared to past projects. Overall, however, the audio did take me days because this was a 3-minute video with like 5 different scenes. Here is what my whole Premiere timeline looked like in the end:

Also, the credits were very fun to make. I was originally just going to do a classic black screen with my Patreon and the credits but instead I chose to quickly animate the map with no one in it, where the lights turn off to really show that it is over, they left. Maybe more protein for another day...6. Final Thoughts:

I love and hate when I run into problems because they are painful and they take a while to fix, but I love finding the solution because then I know how to deal with problems in the future. I will also be commissioning VAs much more from now on because of how much time and suffering they save me from. The only problem is they can be slightly expensive which is 100% understandable but I am broke lol. Anyways I enjoyed making this animation and I am very happy with how it came out and holy fucking hell 12k likes on Twitter is actually insane. Thanks to everyone who supports me or even just saw the video. I think learning blender and animation is really fun and I love the progress I have made so far. More animations to come! And as always, I am still learning so anything in this project breakdown could be showcasing the wrong way of doing something, and if you have any questions or suggestions or even fixes to my problems then let me know!