Its strange that I can't find any discussion about problem with the type properties editor because I encounter a lot of problem with it when creating custom operator (VEX BUILDER) that make it literally unusable.

- Error message not updating
Say I have an error with my code, I edited the code and applied the changes, the node doesn't update and I need to disconnect the node and reconnect again in order to see if the new code is valid or not.

- Inconsistency between the code and the node
Sometimes all the input/output plug of the node just disappear in the network view even though all parameter in the properties editor are still intact, and there is no way I could make it reappear again.

- input/output resetting randomly
After editing and applying the code, houdini just freak out and it remove all the input/output parameter and again there is no way I could recover them, this drive me crazy when I have spent quite some time setting up the interface.

Is it just me having these problems?

I am using 9.5.241 64bits

I believe many Houdini learner are starving for tutorial like this, but really impossible to participate the class, it would be great if the organizer can release the course material so that we all can learn from.

Hi again, It seems that I need to recreate the DOPS network after changing the unit setup in order for the unit to take effects. Not very convenient, it would be nice if there is some kind of reset opiton. Well at least I could make it work now.

Thanks again.

andrewlowell
You could put everything in groups … merge them, scale them, then extract the groups before the sim.

Or, you could Adjust the Edit/Preferences/Hip file options

Hi, thanks for the reply. I did try changing the unit setting in hip file options, I even change it to rather extreme (like 0.001) just hoping to see the difference and get rid of the slow-motion animation, but there is no difference whatsoever. Is there some extra things I need to do?

Hi all, I am doing a dops simulation, but the result look very slow-motion, is there anyway to “scale” the scene down? I try setting the unit setup to something like centimeter, but it make no difference to the simulation. Could someone give me a pointer on the proper workflow regarding this?

(Another little thing is that is the “Play every frame, but never faster than FPS” option in the Global Animation Options works? It seems it do play faster than FPS even a check this option.)

Thanks.

Thanks guy for all the helps. Things are much clear now.

@Michael: Thanks for taking the time preparing the files, they are very helpful. Actually, I think they should be added to the upvector/copy sop helpcard.

Hi, Bernard, thanks for the solution! It works fine, much appreciated.

Though I am still keen to know why it doesn't work in my original case, or more generally, how to control the orientation of the copied geo. I have feeling the problem will come back to haunt me in the future…

Hi, I want to instance a square to points of a curve in a way that the square align/orient their axis along the curve, but the box skewed in a weird way, the doc mentioned the orient attribute but I am not clear how to use it, hope someone can give me some helps.

Thanks in advanced

Hi, Kuba, thanks for the suggestion, it works fine.

(well, I used group sop because I seems to me it's the most straight forward way to do the task and I thought it could accept this kind of pattern matching. Anyway, thanks a lot.)

hmm..I want to rename a series of group in the following fashion:

myGroup -> mygroup

But the group sop seems unable to recognize "mygroup", am I doing something wrong?

Thanks.

ahhh…miss the obvious thing. Thanks a lot for the help!

Cheers.

I have a attribute transfer node with a animated distance threshold value so that the geometry acquire point color gradually through time (with a blend width), but it seems that dop can't import the value properly, is it an unsupported action?

I have attached an example file for your ref.

Thanks.

I see, so I could do my own calculation with the point() expression too. I would try that out too. Thanks for the explanation. Much appreciated.

Cheers

Hi, Antoine, your suggestion works fine. Thanks a lot. Though I wonder why it is possible to refer to the position directly like “position of $PT5” or “position of $PT(x-1)” . It will be much convenient.

Cheers

Hi, thanks for the reply. And I read the help file, didn't quite get it…May be I didn't say clearly what my problem is. As an example, I have a sphere, I want to create a point attribute to store the distance between pt1 and pt2, pt 2 and pt 3, pt 3 and pt 4, etc. for each pt 1, 2, 3… of that sphere respectively .

Sorry for my bad English.

I want to calculate the distance of “current” point with its "previous' point and store the result in a custom point attribute for each point, but how could I access the position of ‘other’ points? Something like P-x?

Hope it make sense..

Thanks in advanced.

Hi Michael, thanks for the suggestion, it works I am so dumb…

It would be handy if we can set the no. directly though.

Thanks again.

Sorry, it seems a easy thing to do, but I just can't figure it out…, hope someone could help.

I try to make a RBD glue object to give birth exactly one (or a specific amount of) particle whenever any chunks inside the glue object collide with something. I have already extracted all the impact data I need to POP and give birth particles from it, what I can't do is control the no. of particles emit from each source (i.e. impact position). *if* there is only ONE source, I could keyframe the activation field, but since the chunks collide at different frame, how could I control all the potential particles sources so that they will only emit one particle?

Any help would be appreciated.

Thanks