ahah!! I just got a sample scene of my own working much thanks to Nikolaus Gradwohl’s super simplistic sample scenes. This is actually a lot faster than I thought it would be to generate the simulation. Very impressed… now off to test more.
@Pyroevil: Hi man, thanks for a toy 
First few results https://vimeo.com/66812201
As someone wrote here - it would by great to have more presets.
I still can’t say I know what slider to move to get where I want to be (especially with “E” sliders).
It is not your fault (I was never be able to easy use Cloth or Fluid sim), so presets could help.
(Or maybe video like Daniel Kreuter did http://www.youtube.com/watch?v=u1dCP_aQLzI&list=UU9NTBQja_r2NI5e56PvuSAA&index=26
Just to see value and behavior of object with this value.)
Note: All sample files works, only pasta’s molecules flies around.
Making preset can be hard because it’s depends on many parameters the users put ( numbers of particles … volumes … shapes … ). I can check for that but not sure I get something with good result that can help people. But what I can do soon as possible is doing a tutorial or better docs to explain parameters.
E sliders is about to make Expension different from Compression. Just imagine spring between particles. You can make it broken when it’s compress to 5% but only broken when expand to 20% ( and you can make is more stiff to compress than expand … etc ). If the “E” option is not activated , compression and expansion have the sames settings.
EDIT: really nice test ! I can put your video on my blog ? Thanks for testing samples files , I check pasta machines blend when I have time.
Ad presets - sorry, you’r right. I forgot you use particle weight preset. Not script preset.
Your note reminds me one think from your doc (4.8) to test behavior of setup with small amount of particles Do you mean low grid (res) and than use hires, because as you said now - it gives different result …
Ad “E” - thanks for explanation, this note was related more to the fact, that E factors takes my problems exponentionaly Becuase I’m not sure if result is not as I expected because of my setup, beta version, or any other aspect That is why I was asking for something like Daniels video, where I can clearly see what exactly value does in video.
I know this kind of docs is huge time-consumer.
It was more note to an air if someone who is friendly whith script can release something like this.
Ad Video - for sure, just if you can download it from there, because I will delete it in a few days.
I would love to get to something more interesting
EDIT: I add some short description to each part, that I left as note in my file names.
Now I see it is not complete info, weight and search is missing (so its a bit useless), all other parameters should be default.
(Weightwas 1 or 2 thousand and search length 3,51-relative)
Thanks
(and thanks to Patrick for OSX build)
… after a few hours of experimentting … I’m an “one button” user, for sure … I prefer to have one button for consistency (0=liquid - 0,5=gummy - 1=hard) second for breaking (0=fragile - 1=unbreakable) + random (to get breaking into parts or like sand)
This system is too much for me … most of the changes in setup drastically change result … impossibility to experiment in lowers and than switch for hires only for more detailed result is killing.
Changes on subframes too. In dynamic paint subframes just fills “holes” between frames, to get more steps for smooth texture. But in this addon result gives me a new behavior of particles. Changes on randomization of any value, shoot me to the beginning of experimenting …
Maybe too frustrations in one day … or it is over my skills … or there are some tricks how to get it that I don’ know …
Here are some tests (uploaded for a few days:
(Sorry for bigger size - zero patience. It is better to download (with right click) and watch later.)
Search Length
https://dl.dropboxusercontent.com/u/55113627/test-Molecular/SearchLenght_1-100.mov
Stiff
https://dl.dropboxusercontent.com/u/55113627/test-Molecular/01-Stiff_1-0.mov
Damping
https://dl.dropboxusercontent.com/u/55113627/test-Molecular/02-Damping_1-0.mov
Broken
https://dl.dropboxusercontent.com/u/55113627/test-Molecular/03-Broken_0-1.mov
For fun, small change on Link Friction
https://dl.dropboxusercontent.com/u/55113627/test-Molecular/LinkFriction%3D0.05.mov
Subframes (I changed only subframe from 4 to 8)
https://dl.dropboxusercontent.com/u/55113627/test-Molecular/Magnetic_force-res25_4subframe.mov
https://dl.dropboxusercontent.com/u/55113627/test-Molecular/Magnetic_force-res25_8subframe.mov
I vklidu !
One button solution don’t work in all case ! 
I always joke with my coworker about that be telling " where is the button MakeItBeautiful " in XSI ?
I try to make it simple and user friendly , because I’m not just a pseudo-coder , but at first I’m a artist. I play with Lagoa a bit and it’s really more complex. I can try to help you a bit by viewing your videos:
Lenght : I never increase it high has that. It’s the searching area to make “spring links” between particles. I put it to 3.5 or 4 for many of my test. More than taht make more links between particles and make structure more solid , but not the better way to did it.
Stiffness: It think you understand it well. Is how hard the spring like are. 1.0 mean it’s fully hard … if that not achieved the hardness you searching for , it’s because your substep is to low and the algorythm don’t have enough iteration to do it well.
Damping: It’s a damping effect on spring link. Links with high stiffness but no damping react like rubber and bounce indefinetly. High damping with low stiffness react more like clay. It’s help to give stability to high spring value to , so put it at zero is not a good idea.
Broken : it’s how much a link can be compressed/extended before they be broken expressed in %
Subframe : it’s the most important part of the script. it’s how many time the script make iteration to achieve the setting you put below. Sometime I put it at 16 … 32 or higher to get hard surface or when I have a lot of particles. 4 is a minimum.
Preset is not really possible. Because all depends of what you have ( the shape of your geometry , the numbers of particles … etc ) and what you want to get. But it’s why I give some example blend to give a idea of setting I put to give some effect.
Latest videos on my channel :
Cloth test:
Hair test:
@spatdsdg - I’m off to try to get up to speed with this incredible looking addon. I can’t find reference to Nikolaus Gradwhol’s scenes in this thread. Can I ask where you found them and are they public? In the meantime, I’ll study Pyroevil’s sample files (but they look pretty advanced). Thanks!
This is looking great! Seems alot of effort went into this script. I can’t wait for a project to use it on, and with the new particle mesher things are looking great.
That link looks very good. And I have learned a lot studying your examples as well! I’m looking at the dirtpile now… Might be up all night!
Thanks!
Love the addon ! Really appreciate all the work you’ve been putting into this.
Can I get an answer on this question please? - What is the expected simulation time for the Ropes sample file. I think it would be helpful to know the sim time to make sure it’s all working correctly on my pc.
Also what is best practice for testing the simulations, which settings to reduce, etc.
Thanks!
Regarding the simulation time, I noticed that the Start Molecular Simulation button changes to a status bar after you push it (a great idea). It counts down the time remaining, and with the tests I’ve done, it is pretty accurate. I’ve tried it on very simple tests and also on Pyroevils dirtpile sample. So you could try loading the rope simulation, starting the simulation, and immediately checking the time remaining. That would give some idea and would be specific to your machine. Then just escape out of the simulation.
One thing I’ve found is that its best to un-tick the “Render At Ending” button when experimenting.
Another thing that perhaps should be obvious is that the frames that get counted are something like YourTimeline * (mol_substeps + 1). So for example, if you use the default number of frames of 250, and your mol_substeps is 4, you will see the frames count up to 1250 (not 250). However, you will still end up with a 250 frame animation. I hope this makes sense and that I’ve got it right. Its my first (of many) evenings with this addon…
Correct PyroEvil - right guy. His web site has the video + the basic blend files. http://www.local-guru.net/blog/2013/5/23/molecular-experiment-6---helix
Hey Pyroevil,
really great work with your plugin - I’m just experimenting with it and am already very positive about taking it into account in our studio pipeline…really cool stuff!
At first, everything was exploding like hell for me - I followed your docs and set normal velocity to 0, looked carefully to not place the particles too near to each other…but everything was still exploding.
To make it short: The issue was, that I set the particles size always to 1 to have a exact copy of the original object. Even scaling this object (in object mode) didn’t stop the particles from exploding. I thought it maybe helpful for people that see this issue too to check that…
Greetings and keep on developing, Thomas
Yeah, with grid emission, it definitely helps to view the emitter object (cube) and particles in wireframe ortho view and adjust the size slider until the particles just touch. Then the factors Pyroevil gives in his very good documentation make more sense. I’m talking about the 2.01 (relative) etc he suggests for ‘search length’. I was able to calm things down and get some good linking once I ensured that the particle size and grid resolution were consistent with the visible dimensions of the emitter (just like Pyroevil suggests in his docs !! From there, it was much easier to tweak values and look at their effects.
So far, I just have a problem trying to pre-calulate what the particle size should be given the dimensions of my dupliobject and my emitter, but that’s more a lack of understanding on my part regarding how Blender stacks all the various transforms and whether things like ‘scale’ get applied or not.
i tried your sript and get the following error on window 32 bits
i’v install it in the local SVN addon folder
i can see the script name in the addon window but cannot Activte it !
File
0blend25\0beta56991\blender-2.67-r56991-win32\2.67\scripts\a
ddons\molecular_init_.py", line 32, in <module>
from molecular import cmolcore
mportError: DLL load failed: The specified module could not be found.
SVN 56 991
any idea how to correct this ?
thanks
[/COLOR]
I don’t know why but few windows 32bit users get this issue. I install it on my old Windows XP 32bit computer and it’s work great … so it’s hard for me to fix that if I cannot reproduce 
Hello everyone, I just wanted to share some tests I did. I used unrealistic values, just playing around with each particle system weight to get the desired result.
Man thats brutal! Poor Suzane :).
How much time did the simulations take calculating?