Announcement

Collapse
No announcement yet.

Please change!---- Vrscenes have the "camera" and "layer" strings attached .

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

  • Please change!---- Vrscenes have the "camera" and "layer" strings attached .

    VRay: 1.50.00.10915
    [feature request] The batch camera name should be appended to the end of the image filename when there are multiple renderable cameras - resolved.



    This has introduced a bug, when using the Render command the user should be able to control the export file name and not have the camera and render layer appended to it. This is problematic with render farms where you need full control over the output name.

    Old (Correct) behaviour:
    Render -r vray -s 1 -e 1 -x 1920 -y 1080 -exportFileName /tmp/test.1.vrscene -noRender file.ma
    Result: /tmp/test.1.vrscene

    New (Bug'd) behaviour:
    Render -r vray -s 1 -e 1 -x 1920 -y 1080 -exportFileName /tmp/test.1.vrscene -noRender file.ma
    Result: /tmp/test.1_camera7_masterLayer.vrscene


    Can we get rid of the Layer & Camera appending, .. cause in the mayabatch command you can specify with layer you want to convert to vrscene...

    For the Multi cam rendering, could we do it in maya as an option in render globals

    No reason anyways, that camera info should be in the "vrscene " file...

    Causes huge pipeline issuees.. Especially with Linux if you want to do rerenders , knowing the buffer is not connected yet.

    Thank you,

    O.

  • #2
    Will think about it, but we might have to leave it this way.

    Best regards,
    Vlado
    I only act like I know everything, Rogers.

    Comment


    • #3
      Just creates lots of problems for farms etc.

      Especially if you can do it with mayabatch, in production not very handy.

      O.

      Comment


      • #4
        If it was consistent, but now if you create a camera by accident in the file,
        (which 99 percent of people do in production)

        you now have to strip every camera, and make sure only 1 camera is renderable..

        Just don't get the thought process..


        O.

        Comment


        • #5
          From what I can see, the layer has always been appended, even in the original release of V-Ray for Maya.

          The camera thing is a bit odd indeed and I have to look into it more carefully.

          Best regards,
          Vlado
          I only act like I know everything, Rogers.

          Comment


          • #6
            The reason for the change is this post: http://www.chaosgroup.com/forums/vbu...xport&p=419061

            If you want to export the scene for different cameras using the -cam option, and the camera name is not added to the file name, then you will just overwrite the same .vrscene file.

            Best regards,
            Vlado
            I only act like I know everything, Rogers.

            Comment


            • #7
              Well, i agree with having multi cams in situations and rendering those, but would be handier if it just exported from maya into vrscene,
              that way the user has control in maya, now you don't have that much control over the outcoming file..

              Also this person is a windows user, so he doesn't even have to worry, cause he has a working renderbuffer,
              rather have that fixed first before doing that request..

              Its easier to make a python / mel script within maya, that exports and saves out , then changing the outputs.

              Thank you tho.

              O.

              Comment


              • #8
                Hey Vladimir,

                Just thought about it some more...

                -leave the layer append, makes sense, just masterlayer (confused me) due to -> default renderlayer naming

                -remove the camera one in the next build please,

                reason being: it 's not lowering your render times, i mean exporting a camera is 10 kb or whatever.. (not adding a difference exporting in file or sep)

                And if the reason is stereoscopic or multicam its easier to run a script that renames the layers _L and _R, and exports those out, you still have to write Vrscene files,
                so will take the same amount of time. (1 per camera, ) also making 2 files, is not that big of a deal...
                You have to have this control from maya, so lets say when it renders Images, it just renders images with a folder names L and R. , or Cam1 Cam2 in the render outputs.
                (not appending this to the "file name" but doing a directory before as option.

                However next to writing "lights" and "materials" as seperate .vrscenes it could help adding a "camera" option in the translator,
                this way if your writing everything out, you could make a python compile, and recombine the vrscene,...

                This would actually speed up workflow etc, and then the camera export would work and save workflow time.
                Lets say if your working with a big team, a layout person could reexport just the camera, and a lighter would have a new update...



                Needed some sleep,

                Cheerio...

                O.

                Comment


                • #9
                  We will probably remove the automatic addition of the camera name to the .vrscene file name; if it is needed, one can always include the <camera> tag in the .vrscene file name.

                  Best regards,
                  Vlado
                  I only act like I know everything, Rogers.

                  Comment


                  • #10
                    That be great,

                    O.

                    Comment

                    Working...
                    X