Indigo v0.7 test 2

[manitwo]

Indigo 0.7t1




Indigo 0.7t2

[Zom-B]

Hey Nick first bug(s) found

It seems that difuse Materials with textures still need the <colour> Tag, without I get an Error message!

Code: Select all

    <material>
        <name>mat_ground</name>
        <diffuse>
	<colour>1 1 1</colour>
            <albedo_texture>
                <uv_set>albedo</uv_set>
                <path>fabric.png</path>
                <exponent>2.2</exponent>
            </albedo_texture>
        </diffuse>
    </material>

For my wine Tests Scene I get 12 to 16 PANIC: tryed to exit from medium when not inside it.
So display took always 413s to 416s!!!
(0.7t1 just had rare PANIC's once in a while)

** EDIT **
OK the Problem for this situation is the supersampling!
Without it I only get 3 to 4 Panic's and display takes "only" about 46s,
this is still a lot more Panic than in 0.7t1...

BTW:
Why did you raised "max_depth" to 10000 (since 0.7t2) and lowered "max_change" to 0.01 (since 0.7t1),
is this for better SSS calculation??

[Wedge]

Radiance, try setting super sample factor to 1 in the inifile to disable it.

I have noticed it makes bloom take quite a bit longer to compute on a scene with one plane, two uvspeheres, and a cube. I can only imagine how longer it takes when the scene is way more complex.

And also I played with super sample factor by setting it quite high, and this simple scene took about 377 mb of ram (like Nick said, uses a lot of ram) and also I waited for a couple minutes to not see it finish computing bloom. I closed it manually.

So I think it has to do with super sample.

[OnoSendai]

The bloom will take longer as super_sample_factor is increased.
This is because the bloom is computed on the larger internal buffer, which depending on the super_sample_factor, may be 2 times wider, 3 times wider etc..
Your best options if you want to use bloom for now are:
* don't set super_sample_factor above 2.
* make the bloom radius smaller.
* increase the display period, so the bloom is computed as often.

[OnoSendai]

Updated link to point to new build, which fixes the blend_material issue,
and also fixes a couple of problems with the Henyey Greenstein phase function.

[oodmb]

Thank you so much!

i am currently about 3 hours 22 miniutes into a multilayered skin model using this build and now i am wondering what problems were fixed in the phase function in the more recent update and if its worth stopping my current render and starting it up again with the new one. from what is rendered so far, there doesnt appear to be any problem.

[OnoSendai]

Heh, the Henyey Greenstein sampling was completely busted, sorry

[OnoSendai]

ZomB:
max depth was raised because it's entirely possible to have more than 1000 scatters inside a heavily scattering medium.

[manitwo]

[Roger]

First of all, Nick, thanks for all your effort in this program, it's getting better and better! Of course, a huge thanks to everyone else who helped in the development and testing of Indigo! And a big hug to Wedge for making the Blender exporting script! Three cookies for all of you!

Now. Is it just me or are the meshlights far more bright in 7.2 in comparision with 6.6? My scene is pretty overexposed now. Also I don't get SSS to work, but that's probebly because I'm a moron.

[tungee]

ONO thanx thanx man(unbiased)!


Ps: i wanna donate for da server, but have no credit card; any tipps or help (Geht auch bareinzahlung?)

[u3dreal]

Great Ono ,

thanks for the next generation ..

cheers

u3dreal

[CTZn]

Roger, the way <burn> works has changed since 0.7t1. It's generally supposed to be over 1 (2-3 would be a good default I think).

[oodmb]

if you like, i could probably set up a donate to indigo thing on my site that has gotten about 200 hits total ever.
AND YAY ONO
YAY SSS
YAY VERY SHINY MONKEY

[Phil]

Hi,

if you like, i could probably set up a donate to indigo thing on my site that has gotten about 200 hits total ever.
AND YAY ONO
YAY SSS
YAY VERY SHINY MONKEY

oodmb, your website does not work with my latest mozilla

Regards, Phil