Bruce Lee
The steps of this solver node have a bug.

When its value is floating point, the solver node will not be calculated.


In particular, can this step be designed to be the same as the steps of a group node? Like the steps in the range of the group nodes.

For example, we hope that every time you play n times, iterates m times. Can only be achieved every time n times, only iteration 1 times.
  This in the actual production will be very convenient.

thank peter!!!
Quote 0 0
Bruce Lee
If m and n are designed to be array type. That is, the user can use the input array to control the value of m and n, that will achieve more results.
Quote 0 0
Bruce Lee
hi peter. Do you forget this post? :)
Quote 0 0
pshipkov
Ah, this one slipped through the cracks. Thanks for reminding me.

I considered similar idea in the past, but there is a limitation imposed on us by how the dependency graph is evaluated.
We have to step through the timeline for the networks outside the node to be evaluated. If you think about it, you will see what i mean.

About the other problem:
Do you mean stepping on subframes ?
Quote 0 0
Bruce Lee
In fact, we can break the time node, with the expression to achieve.
E.g,

if (frame% 20 <8) {inTime = frame;}


This can be done to play 20 times, which will calculate the iteration 8 times.

However, mel running faster than C + +, but also need to write code. If you can achieve the steps of the group node, I think, will be very good.

Thank Peter! !!
Quote 0 0
pshipkov
I am not sure i fallow you the breaking of the time node. Can you provide an example scene ?
If i get it right we need to time warp not just the solver node, but all other nodes connected to it.
Or maybe i am missing your point here.
Let me know.
Thanks.
Quote 0 0
Bruce Lee
I believe that after you read my project documents, will understand what I mean.

Thank peter
Quote 0 0

Add a Website Forum to your website.