Announcement

Collapse
No announcement yet.

Can you xref a simulation?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #16
    So, been testing - by "known issue apart from the particle positions rendering wrong", do you mean this?

    Click image for larger version

Name:	water.jpg
Views:	56
Size:	700.3 KB
ID:	1005720

    I just opened a working file from 2018 in 2019 with the nightly Phoenix from last night, and its like this!! If t his is a known error, then I'll try xrefing the scene and trying to see if it replicates.

    Thanks

    Comment


    • #17
      I can confirm that xref'ing the scene above, while has weird foam locations, does render exactly the same, so that's great!

      I also tried merging the xref in, and it also rendered the same as the two other files. So maybe that's fixed?

      My tests were only with a single frame, so I cant confirm its working for an ani.

      Comment


      • #18
        Yup, that's the foam thing - we gotta completely rework the way the hidden PG [] nodes work in order for that to be fixed, but that's just one of a line of issues with these nodes that just make it obvious that we need to do this soon. I hope in the next month or so I'll start working on this, but it's gonna remain broken for a while...

        As far as merging - I can't remember if it was an object xref or scene xref, but there will be an issue only if you are using a paths which contains a $(...) Phoenix macro. You can workaround the problem if you have an absolute path to the caches in the original scene.

        Cheers!
        Svetlin Nikolov, Ex Phoenix team lead

        Comment


        • #19
          Ok, cool. Until the foam issue is fixed, the nightly is unusable, as it happened outside of the xref setup. I'll stick to the current release, but very keen to see this workflow added, as it will be so much better for reusing sims!

          Comment


          • #20
            Ah, wait, you mean you don't use xrefs when you get the shifted foam? In that case, it's something different that should be fixed in a different way and can and should happen much sooner.
            Svetlin Nikolov, Ex Phoenix team lead

            Comment


            • #21
              Yes, so I took a 2018 working file, that was fine, and just opened in 2019 and rendered, at it was like the above.

              Comment


              • #22
                Oh, definitely gotta fix that ASAP - did you have different Phoenix versions for 2018 and 2019? Also, can you try a Phoenix nightly from today - these last 2-3 days I caught some issues in the nightlies that appeared due to the changes to referencing.

                If a 1 Aug's nightly still renders that way, can you send that scene over together with a single cache file? Only the simulator and particle shader can remain in case you have any geometry that should not be shared publicly.

                Thank you!
                Svetlin Nikolov, Ex Phoenix team lead

                Comment


                • #23
                  I got the 1st August nightly this morning. And yes the 2018 is the current release, and the 2019 nightly. Can you get me an FTP?

                  Comment


                  • #24
                    Affirmative, can you please drop me a mail at my forum user name + at chaosgroup.com and I will send you the FTP credentials.

                    Thanks very much for helping track this down!
                    Svetlin Nikolov, Ex Phoenix team lead

                    Comment


                    • #25
                      Oh, I see what this is - the particle group lost its node handle if you merged a scene into this one.

                      You can fix this scene by deleting "PG [Foam] of system [PhoenixFDLiquid001]" and picking the Simulator's Foam particle system from the particle shader once again.

                      Meanwhile, I'll try to make this not happen again.

                      Thanks once again!
                      Svetlin Nikolov, Ex Phoenix team lead

                      Comment


                      • #26
                        Hey, indeed that has fixed it!

                        Comment

                        Working...
                        X