Indigo 1.1.3

[OnoSendai]

Hi all,
Another small release, just fixing bugs and tweaking the new embedded mesh format.

32 bit:
http://www.indigorenderer.com/joomla/in ... &Itemid=62

changelog:
1.1.3
* changed embedded_2 format yet again
* changed default name of pigm/pigs in save dialog to be same as igm/igs
* --unpack now doesn't overwrite existing files, and can unpack to absolute paths
* packing packs displacement maps now
* fixed crash on exposition of UV set without providing actual UV data.
* fixed --unpack to make subdirs properly
* new version 2 of IGI format: added render time.
* total render time and samples/sec now displayed correctly when a render is resumed.
* added command line tonemapping with the --tonemap option
* reenabled optional specification of vertex shading normals in the scene file
* reenabled fatal error on non unit-length normals.
* added merge_vertices_with_same_pos_and_normal option for mesh, defaults to true.
* removed step_null_blend option, added step_blend option for blend materials.

The embedded_2 format looks like:

Code: Select all

<mesh>
		<name>mesh2</name>

		<normal_smoothing>true</normal_smoothing>

		<embedded_2>
			<expose_uv_set>
				<index>0</index>
				<name>albedo</name>
			</expose_uv_set>

			<used_material_name>mat_a</used_material_name>
			<used_material_name>mat_b</used_material_name>

			<v p="-1.5 0 0" />
			<v p="-1.5 0 2" />
			<v p="1.5 0 2" />
			<v p="1.5 0 0" />

			<uvs uv0="0 0" uv1="0 0" />
			<uvs uv0="0 1" uv1="0 100" />
			<uvs uv0="1 1" uv1="100 0" />
			<uvs uv0="1 0" uv1="100 100" />

			<t v="0 2 1" uv="0 2 1" m="0"/> <!-- this triangle uses mat_a -->
			<t v="0 3 2" uv="0 3 2" m="1"/> <!-- this triangle uses mat_b -->
			
		</embedded_2>
	</mesh>

[manitwo]

* total render time and samples/sec now displayed correctly when a render is resumed.

finally
thanks ono

[OnoSendai]

hehe, yeah that issue took a while to fix

[CTZn]

Thanks !

[cpfresh]

Great update Ono, thanks!

[Pinko5]

Tnx ONO!
Luca

[PureSpider]

Finally the render times :>
Thanks!

[aleksandera]

Thanks!

[aleksandera]

Blendigo 1.1.1, simple cube and sun scene.

[palawat]

* total render time and samples/sec now displayed correctly when a render is resumed.

Does that refer to the situation when one did run Indigo on slave with resume file path?

The reason I ask is that, I tried <indigo.exe scene.igs -r scene.igi -n s -h hostname:7777> on slave which make Indigo on the master box kinda confuse and wrongly display total render time and s/pix. Nowadays I use <indigo.exe -n s -h hostname:7777> on slave.

When you said it's been fixed, that kinda makes me thinking I might have used the improper command for the whole time. Which one should I use to resume rendering on slave?

Sorry, if it looks like I'm panicking. I really DO.

[Kram1032]

thanks

[OnoSendai]

* total render time and samples/sec now displayed correctly when a render is resumed.

Does that refer to the situation when one did run Indigo on slave with resume file path?

The reason I ask is that, I tried <indigo.exe scene.igs -r scene.igi -n s -h hostname:7777> on slave which make Indigo on the master box kinda confuse and wrongly display total render time and s/pix. Nowadays I use <indigo.exe -n s -h hostname:7777> on slave.

When you said it's been fixed, that kinda makes me thinking I might have used the improper command for the whole time. Which one should I use to resume rendering on slave?

Sorry, if it looks like I'm panicking. I really DO.

You should only use the resume '-r' command on the master indigo process.

[palawat]

I see.

Thank you very much, Ono. The panicking is out, now.

[Kosmokrator]

respect..............to the MASTER!!!!!!!!!!1

[CTZn]

* reenabled optional specification of vertex shading normals in the scene file

What should I do in order to render a cube with sharp edges ? It's confusing because you define geometry normals and Indigo seems to loose them... is that because of merge_vertices_with_same_pos_and_normal ?

edit: Mmm no because 1.1.2 renders hard edges as smooth, 1.1.1 and below does as expected.

edit2: maybe the exporter is faulty here, was meant for 1.1.1

edit 3: oh well a cube is a bad example, what about a carglass it's surface is smooth while schematically it's borders are sharp (that's a theorical example, any partially smooth object would fit).