Home Artists Posts Import Register

Content

Hey y'all! It's time for that goo again. I've been working on a fun little project for hot bolt summer chapter 3 that really lends itself to being looked at as an example of the fluid sim process I've been developing. This project has been a short little thing I've worked on for just a few evenings, so getting a pretty nice result.

If you wanna read my previous adventures with fluid simulation and approximation, check out part 1, part 2, part 3 and part 4!

The process of doing simulated effects is very different to my animation process, and if it goes on too long I absolutely hate it, because it's far more analytical and less artistic and expressive. However, that makes it a great candidate for breaking down in a post like this! I can actually explain pretty much the entire process:

First, I do a simulation at a coarser voxel size. This is important to quickly make sure the fluid is being emitted in the right amounts at the right times, and to get a general idea of where it will flow. Due to viscosity, friction and surface tension the fluid will behave very differently at this resolution, so you can't infer too much from its behaviour.

I actually really like how it runs down her leg here - it's not very cum-like, but it does look cool. At this point I'm also broadly ignoring the amount of cum, just getting an idea for what the sim is going to behave like. If things don't look right in this step the animation can be re-worked since simulating is quite cheap, so we're checking to see if anthing is getting in the way, causing collision to freak out, or producing results we don't want. We do have to ignore things like leakage, sticking and other small interactions at this stage, though.

After that, we want to do a shorter sim at a higher resolution to determine the fluid's velocity. The resolution has to be near our final resolution otherwise it won't be accurate, so we can't afford to simulate many frames without wasting a bunch of time. Thankfully, if it's wrong it's usually pretty obvious quite quickly.

Here we can see it's way too fast, jetting out and splattering on the domain edge. This has to be a trial and error process, since every change to the resolution will change the behaviour of the fluid, so 8 m/s emission speed at lower resolution will lead to a lot more and faster fluid than the same speed at a higher resolution where the fluid is able to pull on itself more densely. Once we've tested it though, refining the emission speed gets easier as we can take educated guesses.

Next we need to refine the amount of fluid, and its properties. At this point our animation and emission velocity are locked in, so all we're looking to tweak is things like viscosity, surface tension and stickiness. We can also adjust when the emitters are active to modulate the amount of fluid, as long as our velocity is predictable.

Here the quantity is way too much, and while we're getting some nice globs, the fluid isn't sticky or viscous enough. That's great info! We can turn up the viscosity, and that will have the bonus effect of reducing the flow, and therefore the amount of fluid. Stickiness is a bit harder to solve, but what's overwhelming the friction and attractive forces here is the sheer amount of cum. As we can see on the screen left thigh there, the fluid does stick nicely before more is added and it falls off with a jerk under its own weight. So we can pretty confidently diagnose a lack of viscosity!

it's also a a lot more clear that cum is actually being created inside her, and dripping out from the inside of her mesh, seemingly appearing directly out of her belly. Not good! In the final version I fixed that by adding a big outflow object around where her womb would be... kind of appropriate!

With this simulation for the first time I've experimented with turning up the sheeting strength and I'm amazed that it hasn't broken! Sheeting is great for the kind of drippy divergences of fluid seen here, but since it adds fluid particles it can quickly get out of control and create a weird-looking multiplying effect where the fluid grows as it moves.

--

With that, I think I can declare this workflow for doing cum and other fluids pretty mature! It's comfortable now to add fluids to animations quite regularly, without it eating a week or more of my life. Look forward to a sticky future!


Files

Comments

Iora

Really great work! And it's looking pretty good, too. Thanks for the breakdown of how you're doing all this, I think it's pretty interesting. There's a lot of not-great or almost-good fluid simulations out there, always room for a better one.

Nightwing

"Here the quantity is way too much" da hell you mean too much, that's just right, lol.