[Bf-python] RenderEngine API - fractional frame

[Doug Hammond]

Thanks for the hints Matt.

Regards,
Doug.


On Sat, Aug 6, 2011 at 12:37 AM, Matt Ebb <matt at mke3.net> wrote:

> Yes it works, and no there isn't another accurate way because there can be
> dependencies within objects that need to be calculated to get the right
> values (drivers etc).
>
> In my exporter I don't continually scrub through frames, rather I
> precalculate it all in one go beforehand, to limit the number of times it
> needs to set the subframe. I store all these animated values in a dictionary
> - this comes at the expense of memory, especially if you've got lots of
> high-poly deforming objects, but it's not too bad so far. At least with
> renderman you can export an object to an archive once individually to avoid
> doing the entire scene in one go.
>
> See here:
> http://www.pasteall.org/23730/python
>
> matt
>
>
>
> On Sat, Aug 6, 2011 at 9:20 AM, Doug Hammond <doug.hammond at luxrender.net>wrote:
>
>> I'm also having issues supporting motion blur for external render engines.
>>
>> Does frame_set(integer_frame,fractional frame) actually work from a
>> RenderEngine.render() context?
>> (all RNA write and operator calls are disabled).
>>
>> Is there any other way to get object transforms for arbitrary time values
>> instead of continually scrubbing through frames on export?
>> (which is a really slow and tedious method to use; feels really dirty).
>>
>> Regards,
>> Doug Hammond
>>
>>
>>
>> On Wed, Jul 27, 2011 at 3:27 AM, Matt Ebb <matt at mke3.net> wrote:
>>
>>> Hi, what you've written below is looking at the geometry data itself. If
>>> you've animated a Plane object moving you'll probably need to print the
>>> plane's object level transformation - should be something like
>>> bpy.data.objects["Plane"].location[0]
>>>
>>> bpy.data.meshes is looking at the mesh datablocks (geometry level data).
>>>
>>> cheers
>>>
>>> Matt
>>>
>>>
>>>
>>> On Wed, Jul 27, 2011 at 12:07 PM, Nicholas Yue <yue.nicholas at gmail.com>wrote:
>>>
>>>> On 26 July 2011 22:00, Nicholas Yue <yue.nicholas at gmail.com> wrote:
>>>> > I ran a test query vert.co[] data of a simple plane with 4 points at
>>>> > fractional frame, the returned value is the same as on the integer
>>>> frame.
>>>> >
>>>> > Looks like either I may have missed calling some required
>>>> update-status
>>>> > function as even basic geometry (non-sim) is not quiet right.
>>>> >
>>>>
>>>> Here is the printout from the Blender Python console
>>>>
>>>> >>> scene.frame_set(2,0.2)
>>>> >>> print(bpy.data.meshes["Plane"].vertices[0].co)
>>>> Vector((0.9999999403953552, 0.9999999403953552, 0.0))
>>>>
>>>> >>> scene.frame_set(3)
>>>> >>> print(bpy.data.meshes["Plane"].vertices[0].co)
>>>> Vector((0.9999999403953552, 0.9999999403953552, 0.0))
>>>> _______________________________________________
>>>> Bf-python mailing list
>>>> Bf-python at blender.org
>>>> http://lists.blender.org/mailman/listinfo/bf-python
>>>>
>>>
>>>
>>> _______________________________________________
>>> Bf-python mailing list
>>> Bf-python at blender.org
>>> http://lists.blender.org/mailman/listinfo/bf-python
>>>
>>>
>>
>> _______________________________________________
>> Bf-python mailing list
>> Bf-python at blender.org
>> http://lists.blender.org/mailman/listinfo/bf-python
>>
>>
>
> _______________________________________________
> Bf-python mailing list
> Bf-python at blender.org
> http://lists.blender.org/mailman/listinfo/bf-python
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.blender.org/pipermail/bf-python/attachments/20110806/d9830a00/attachment.html>