Hi Matt,<br><br><div class="gmail_quote">On Mon, Feb 25, 2013 at 6:59 AM, Matthew Turk <span dir="ltr"><<a href="mailto:matthewturk@gmail.com" target="_blank">matthewturk@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Hi Jill,<br>
<div><div class="h5"><br>
On Mon, Feb 25, 2013 at 6:32 AM, Matthew Turk <<a href="mailto:matthewturk@gmail.com">matthewturk@gmail.com</a>> wrote:<br>
> On Mon, Feb 25, 2013 at 1:43 AM, Jill P. Naiman <<a href="mailto:jnaiman@ucolick.org">jnaiman@ucolick.org</a>> wrote:<br>
>> Hi Matt,<br>
>><br>
>> I've tried a few different things with PLY files and no luck. Actually, I'm<br>
>> not 100% sure if Blender really likes PLY files anyway - you can't even<br>
>> export color easily in PLY format in Blender. So, I've given up on PLY and<br>
>> have hacked together a bit of a messy OBJ exporter:<br>
>> <a href="http://paste.yt-project.org/show/3205/" target="_blank">http://paste.yt-project.org/show/3205/</a><br>
><br>
> This looks nice. We should incorporate this, or a modified version,<br>
> into the AMRSurface object. We used to have a .obj file format<br>
> exporter, but I (incorrectly, it seems) thought the PLY format, which<br>
> can be in binary and is thus much smaller, had feature parity. Would<br>
> you be up for making this a bit more general, and then adding it to<br>
> the AMRSurface object?<br>
><br></div></div></blockquote><div><br>I would be up for making a general version, perhaps even with the correct "face" coloring, but given that I'm pretty new to python and yt (... and Blender... and OBJ formatting...) it might not be very pretty, but it will at least work. Hope that is ok.<br>
<br>As far as PLY for Blender, I think it can in theory work, but it would require some updates to the PLY importer/exporter in there, and just outputting things to OBJ seemed like the path of least resistance. Plus, the current Blender PLY importer seems to require float, not binary formatting, and I'm not sure how easily that can be changed.<br>
<br>Also, I couldn't seem to get transparencies to change in sketchfab using PLY (with the code you posted a bit ago, not sure you had success), and having not attempted it, I'm sure it will be easier with OBJ formatting :)<br>
</div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div><div class="h5">
>><br>
>> I'm probably doing the colors wrong, since there's likely a nicer way to get<br>
>> actual "face" values then just taking the value at 1/3 of the vertex values,<br>
>> but it worked, so I didn't mess with it too much.<br>
><br>
> There is a simpler way, but honestly it's pretty non-intuitive and I<br>
> think the face/vertex sampling in the AMRSurface is a bit clunky. If<br>
> you do:<br>
><br>
> surf.get_data(field, "face")<br>
><br>
> it will sample at the faces rather than the vertices. This is also<br>
> the default behavior for getting data through the dict-like method:<br>
><br>
> surf[field]<br>
><br>
> which should be face sampled.<br>
><br>
>><br>
>> A few other things: (1) in theory, this can set transparencies, but I<br>
>> haven't got that working as of yet, (2) the default specular (ambient?)<br>
>> color settings make for a bit of a washed-out view in sketchfab:<br>
>> <a href="http://skfb.ly/54jged0b" target="_blank">http://skfb.ly/54jged0b</a>, (3) the .mtl file is a bit too long since I'm<br>
>> essentially setting each face as a different material, when one would just<br>
>> like to use 1 material per color and set all the faces with that color on a<br>
>> specific surface plot to that material, but I haven't got that bit working<br>
>> well either.<br>
><br>
> (2) Is this something that can be controlled by the materials editor<br>
> on Sketchfab?<br>
<br></div></div></blockquote><div><br>In theory, yes, but you have 255 colors per surface, and I think you would have to change the transparency of each of the 255 by hand. It looks like there is a way to group sets of materials in OBJ files, but I'm not sure if sketchfab supports applying transparency changes to groups of materials instead of individual materials.<br>
</div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div><div class="h5">
</div></div>Quick followup -- by changing the "Rendering Options" to "Shadeless" I<br>
was able to see things a bit better on Sketchfab.<br>
<br></blockquote><div><br>It seems that turning off the ambient and specular color is all
that is required to make the sketchfab upload look good w/o having to mess with it:
<a href="http://skfb.ly/54jhe10b">http://skfb.ly/54jhe10b</a> Besides, if you want to do something silly like make it super shiny, I think you need shading turned on :)<br><br> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
-Matt<br>
<div class="HOEnZb"><div class="h5"><br>
> (3) Ah, interesting. One could do just 255 materials, for instance,<br>
> and then that is now the colormap.<br>
><br>
> This is definitely very interesting, and would be a huge upgrade for<br>
> our Sketchfab / Blender / etc capabilities. Thank you for sharing all<br>
> of this!<br>
><br>
> -Matt<br>
><br>
>><br>
>> Thanks for the info, and please let me know if any of that doesn't make<br>
>> sense.<br>
>> -Jill<br>
>><br>
>><br>
>> _______________________________________________<br>
>> yt-users mailing list<br>
>> <a href="mailto:yt-users@lists.spacepope.org">yt-users@lists.spacepope.org</a><br>
>> <a href="http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org" target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org</a><br>
_______________________________________________<br>
yt-users mailing list<br>
<a href="mailto:yt-users@lists.spacepope.org">yt-users@lists.spacepope.org</a><br>
<a href="http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org" target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org</a><br>
</div></div></blockquote></div><br>