Jump to content

Marielle Caerndow

Resident
  • Content Count

    254
  • Joined

  • Last visited

Posts posted by Marielle Caerndow


  1. As how I see it, it won't be much different as it's now.

    Even now, someone can wear a pair of 300-, 400- or 500-prim boots andcause performance issues.

    Same will applyfor mesh. Some creators will build with insane high poly-count (instead using textures for details). 

    On the other hand, mesh will allow much more performance-friendly builds.

    Just a test: I made a pair of boots, built with 8 sculpties. For the viewer's render engine it counts with 15.000 triangles (!). (One sculpty already counts nearly 2.000 triangles).

    Now I built the same  boots as mesh: counts only 1.700 triangles. This means, the mesh version is nearly 90% more efficient.


  2. Daniel, I can confirm this.

    Tested the colladas from 3dsMax2011 ( => collada 1.4.0), upload works.

    Tested the colladas (same, simple model) when exported from Max2012 (=> collada 1.4.1) did not upload (completely crashed the viewer)


  3. From my own tests, I can crash the mesh viewer when uploading dae-files created with 3dsMax2012.

    The reason in that Autodesk  ships a newer collada exporter with Max2012. Max2011 uses collada 4.0, Max2012 has 4.1.

    Maybe this is the same what happens when you convert your model from AutoCad, it ends as (incompatible) collada 4.1 file. 


  4. I agree with the basic problem: (old) items with hundreds of size-scripts are simply 'out-of-date' from the technological point of view. Today we have smarter options to do the resizing with just one script.

    But I can't agree with the solution to allow deleting scripts from no-mod items, sadly. Just one example: A creator who offers demos of her/his products. The demo might be no-mod and including a script that disables the items after the demo-period. Allowing to remove scripts from such demos would break content.


  5. Maya, I just tested it with a friend. Here it worked.

    Ava1: created box (yes-copy/no-mod/yes-trans). Add 1 control-script ( yes-copy/no-mod/yes-trans). The whole object is now still yes-copy/no-mod/yes-trans. Now adding a second 'dummy-script' with yes-copy/no-mod/no-trans)....the whole object gets no-trans as expected. The control-scipt was scripted to remove the dummy-script when touched. Now sent the object to a Ava2.

    Ava2 : Rezzes the object, it's still yes-copy/no-mod/no-trans. Ava2 touched it and the control-script removed the no-trans-dummy-script. Result: as expected, the whole object becomes yes-trans.

    To mention again, Ava2 removed the tranfer-block-script when the object was rezzed in-world (not attached) ...maybe this is the difference.


  6. as I understand it, you could do:

    1. Add a no-trans/yes-copy script into the object. This makes the whole object no-trans. And since the scipt is yes-copy, it's can't be removed from the object other than by a scripted call. That's why next...

    2. Add a scripted function that removes this no-trans script.

    As result, your first owner get a no-trans object (because of the no-trans dummy-script). Means, the object must be rezzed by the first owner to remove this no-trans script.

    After the script has been removed, the whole object becomes yes-trans again.

    (( test this trick with an alt before going in production. If everything done as I said (adding a yes-copy/no-trans script + script call to remove it), I should work. No guarantee, though.))

×
×
  • Create New...