Thinking Particles

#501

Yeah, I’ll publish some BBs on my site soon too, crazy busy here also.

#502

Hey guys,

Here is a Normalize op Fabian and me did today. This basically replaces the ValueToTime op, but this one is much simpler. This operator re-maps a range of values to another range.

You can download it here:
http://www.nalvfx.com.ar/TP_Normalize.ms

         TP_Normalize.ms belongs: 
         C:\Program Files\Autodesk\3ds Max 2011\Scripts\Startup

Cheers!
Nahuel

#503

Cool :slight_smile: Setting up collection of custom nodes somewhere would be great :slight_smile:

#504

I should have some time this weekend to go through and set up the first post with a list of custom script ops, blackboxes and max scripts for TP. If you guys want, I think it would be a good spot for a massive list of TP links. PM me if you want me to link anything in particular. Anyone know if there is a Character limit to one Post here? This might be an issue.

It would be great if everyone stuck to a naming convention for the max/thi/ms files.So we know what version of TP and Max the author originated them from.ex.
MyBadAssTPNode_2010_TP4_Alex

Would be awesome to have zipped files accompanied with a text file with the authors info and words of warning about the files. I know it seems a little anal but really helps down the road when you have a bunch of files in a folder and trying to find or debug a file.
What do ya guys think?

#505

You guys could always add a wiki page and dump them there, it will hold attachments. Since that thing gets so heavily used and all. It is easy for a repository.

EDIT: Good to see the importance of naming convention being addressed too :wink: :slight_smile:

Although since it is windows I would go:

TpVersion_Date_NodeName_Build

That way explorer will order by version, then date, then name, and build number. So opening the import dialog you will see all tp3 ops first then listed by date, then tp4 ops, in order of date, ect. much easier on the brain.

#506

Yea I was thinking a wiki might be better, that way there can be a few moderators, rather than just one person.

Can you still create a CG wiki here? went to try and get a page going…looks like all the html is messed up. If anyone has any suggestions on where to start a wiki, i don’t mind getting it going.

yea, would make more sense to have the TP/Max/FileName/version #/Author naming order. Though i’d just use windows detail to sort by date, just personal pref i guess :stuck_out_tongue:

#507

Oh I just noticed they removed the wiki button from the menubar! The wiki is still there though, hmm maybe they are considering removing it? It is too bad it is a good medium for transport, not that it really got utilized here.

You have to use wiki markup language it is super simple, for instance a header is labeled like this =Header=. Paragraph text is just text, there is no special markup for it. Search MediaWiki and there is a codex.

About the naming no big, anything by the way of a standard is good. It is totally difficult to find a solid naming convention that works for everyone, I have been trying for years :smiley: Our media servers literally run off of a single folder, no directory tree, with a shit-ton of files from the last 8 years, it is such a PITA!

#508

Hmmm… how about the TP vimeo group forum then?
That way I can add additional moderators.
Any other suggestions for a wiki place?

#509

I managed to create a TP wiki here:

http://wiki.cgsociety.org/index.php/ThinkingParticles

But it’s just text. It doesn’t create sub-menus. Can you edit the wiki?

#510

Thanks Nahuel! :beer:

I couldn’t create a page for some reason, but looks like people can edit it fine.

Sweet lets get this going!

#511

Hi, I’ve installed the Demo version for both 2011 and 2012 and am seeing a problem in each. Basically, the Create rollout doesn’t populate the operator buttons when I select categroies, and the right-click flyouts don’t don’t display anything. Has anyone encountered this problem? It seems rather odd.

Opening the sample scenes cause no problems and I can see operators normally. I just can’t create any new ones.

#512

Guys help me.
Collision map op don’t see the fragmented geo from Volume Breaker op ?

#513

ur vb particles need to be intersected with ur object that u want to generate the collision map

#514

Thanks Aboubakr. I find where I’m wrong when watched your file.

My particles intersect object , but when I read cebas doc’s this string misleading me.
"“All other objects below will collide with the first in the list to create the collision map.”
So I added below first object in ColMap list , all my original geo from “Obj to particle”
:beer:

P/s/ One more question. What the best way to bake TP in geometry ?

#515

Keep your eyes open at Siggraph about that :slight_smile:

#516

Nevermind, fr-proxy is not supported in TP

#517

Hey guys,

I am pretty comfortable with tp now,but I am having some problems with caching, not with the process but with the file sizes. It seems that when i cache, let’s say a 3 stage volume break activated by a group,the cache size is reasonable if I sample per frame but when I up the samples to 500+ which seems to be needed for any production quality breaking the cache files are huge(20gig+ for a 100frames). Not sure why the cache is bigger as it only needs to store the same info as when I cache per frame.I can workaround by exporting but that takes away the greatness of tp caching. Is this an issue for everyone else or is it just me?

Cheers

Craig

#518

It does save all the subsamples. 500 means more then 21 samples per frame, how fast are your objects moving? I’ve never gone above 240, and that’s with in quite fast moving stuff.

#519

I agree with Hristo, if you have to up the subsamples that much, to get a good collision, there is definitely something wrong with you setup. Of course unless your objects are moving like 2000 units per frame. :slight_smile: Check out my “new” collision system.

http://www.vimeo.com/27314307

I’m using “half frame sampling” on this one and it’s still pretty accurate. Perhaps try to set the voxelmode to real facecollision, depending on your geometry or just increase the voxel/edge samples. Sometimes a collision looks wrong because the friction is too high.

#520

Thanks guys, all my groups voxel grids are set to 1(which as I understand is face collisions) and my frictions are set to 10…guess i should look at edge samples or up the sampling in the sc node.

Thanks again,

Craig