This is a little test of unlimited clay rewritten from scratch: at the surface you see the same, but this is entirely a new engine under the hood, and it’s getting in its final form.
I’m currently facing serious communication issues as most of my channels have been squeezed out: only emails remain available but I get little answers from my email questions…I feel that I can hardly follow the open source development pace… I’m not getting enough feedback, perhaps email is not the way and also no one is forced to answer me, I understand we all have problems and projects that drain our time, i’s just that I thought the development would be more dynamic and collaborative…
I’m feeling alone and isolated, and I don’t blame anyone from the blender community for that, it’s just the piece of reality that was assigned to me. I’m loosing the momentum and seem like for getting attention that is all that matters , but I don’t code for that, I code because of the joy of being useful, of making something that people from everywhere could use, and being able to feel proud of that, because I need to feel part of something bigger than me and my difficult environment, because I want a place to render my creativity and learn in the process experiences and knowledge that otherwise would be impossible to get and also because of all the people that have trusted and helped me. Forgive this rant, being this a blog, sometimes I need to express my feelings because otherwise it could became a cold technical blog and in the future, when i read it back I could remember how I was feeling at that time and hope could say: – I was wrong
last Sunday we finished building the roof!!!!!!!! It was a very hard task because we have less than a day to finish it (otherwise we get leaks during rains). We started around 9:00 AM and ended up around 6:00 PM. We are very happy because the hardest part in the rebuilding of my house is over!! The rest of the things like floor, water pipes, doors, windows, linux
could come later because they don’t need to be installed all at once. Shortly after finish it a small hurricane visited us but ended up as a tropical storm and do more good than harm
Thanks to everyone that one way or another has made this possible!
One of the last frontiers of particle fluids simulation engine was implementing plastic behaviors (previously some elastic features were implemented using Hooke’s style spring forces) but plastic remained challenging, and Stephen has been working on it since the release of particle fluids, he has ended up with a very nice and clever implementation that only lack speed. We were sure that it was a correct implementation but something in the code was eating up CPU cycles a lot, so last week I finally had some time to work again in particle fluids with him and we made some important progresses
I added the option for displaying springs between particles, this proves to be quite useful to see what’s going on in the simulation and also could be pretty didactic and added the option for controlling the amount of springs per particle since it was the most performance eater, previously we were going for the raw implementation that takes into account the maximum number of springs per particle possible , but even the most seasoned soft body/cloth/springs simulators out there, for speed reasons, implement only a few springs per particle.
As a result speed is increased by several factors and make it user ready
A very important and nice thing that happened recently was that thanks to Jahka´s efforts the particle system is now much more stable than before: shortly after the integration of particle fluids we get lots of reports regarding random particle explosion in the most simple simulation set ups, making nearly unusable the new particle simulation features without increasing the subframe property that at the same time increase the simulation time. This was an issue inherent to the particle system in general (the collision related code), not particle fluids, but with this kind of physics it gets more evident than with other particle physics types due to its non linear nature that amplifies smaller unsatbility issues.
Well , I tested with great pleasure a new build and the problem was solved … here´s a quick and dirt set up with 100 k particles with near no stabilization trick and even then is beautiful, without the collision code corrections this was impossible to get.
Thanks Jahka !