[Richard]

I wish the planet not to be a sphere but it is.

That's what they want you to believe.

I understand the wish, but it really is a matter of post processing and not "Multilight".

But why does it HAVE to be so? That is my wish!!

And don't forget part of this wish was not just for that of IBL channels (which I understand I could be barking up the wrong tree!) but also the option to blend skydome / phys sky! The best results I've ever got with maxwell was on a few tests that I ran rendering two versions of the scene one with Physical sky and one with skydome and then blending in PS! The results were stunning though not practical due to the dual render times! Softened lighting and shadows, it gave it a real HDRI feel where you had intimate control of the shadows!

This wish I should reinforce is certainly the far bigger one and that which I was refering to in regards to an often requested feature! I might post it as a separate wish! A back it with some tests to show the result.

[giacob]

but also moitor gamma and burn are post process features.... arent them?
nontheless we have slider for adjusting them while rendering !!!!

[krotala]

"Exporting a Maxwell Physical Sky as HDRI will easily prove it works, too. "

Will that render faster than regular physical sky?

[Richard]

"Exporting a Maxwell Physical Sky as HDRI will easily prove it works, too. "

Will that render faster than regular physical sky?

According to Tom "yes"! And he's generally on the mark!

[krotala]

According to my test it doesn't. Cooked both images for 20min and the physical sky comes out cleaner even though the IBL one comes with a higher benchmark score.
Maybe I didn't cook the HDRI long enough, but it is a 2000x2000 image at SL 16. It looks clean enough. Could be I doing something wrong?

[JorisMX]

IF you use 2.0 it is perfectly easy to adjust reflection and other values in Post.

You might be disappointed though, its the boring way all the other biased renderers work:

Render your Beauty Pass. This will take quite long given a somewhat complex scenery as well as resolution.

Change Materials and render seperate passes (w/o ML!!!) for

1. Reflection (black material w/ spec layer),
2. AO (using a clay render 1 Layer grey Rough about 80-90)

You may want to do more passes and perhaps do some with skydome instead of a HDRi
The final tuning over the image can now be done inside a compositing app or photoshop

Some people might say that one render takes so long that the multiple passes make this a never ending story.

Im currently rendering a few concept scenes in this menthod up to Full HD Res where my beautypass took perhaps 10 hrs to render.
I had a REF and AO pass (make sure to use grey and not white!!! Otherwise it will take too long to clear) done in a bout 30-50 minutes rendertime
(I was trying to see how much difference it would make from the 30 min progress, there wasnt much difference in my scene..)

I know alot of us are trying to achieve the great "This is direct MX Output with no post" moment.
And if you can, I think alot of people envy you.

You must be either:

1. Incredibly talentend and/or experienced
2. Having way too much time on your hands

My workflow has increased alot and the flexibility you gain is imense.
You can also save time if you are willing to do other passes with the Renderer that comes with your Modling app.

I know this isnt what the guy who started this thread was after, but its basicly the same and alot more fun than moving 4 sliders


In case somebody is interested, I could do a PDF or Tutorial or something on the entire progress

[devista]

please, I´m interested

[Richard]

ME TOO!

Though with that I will say I've always had the philosophy that any workaround is a sign that something is either broken or insufficient in it's feature set!

This actually made me think about Tom's initial comments questioning a need for ML control of these channels I would consider these channels of the same importance as those of any other light, sure they can be adjusted prior to rendering but then again so can any other light!

For me I struggle so much with getting these setting right and it is evidenced here with a suggested and implemented workaround that others do too! I may fall into the less talented bracket for sure (reality check) but that doesn't mean I don't want features that at least allow me to look like I'm in that set!

Thanks Joris!

[JorisMX]

Hey Guys,

I'm happy to hear about the interest on such a TUT.
I'll try and box something up for you, can't promise when though - got a lot of projects pending and running.

Richard, I do understand what you mean about using "workarounds".

But this is a technique and not something people do to make up for poor rendering skills.
As I said before the main Issue, something we all have to deal with every day, is workflow speed.
So if I can gain control, save time and leave room for client corrections... why not?

The drive and will for perfection is something fueling all of us.
But until you turn into Stack, Herve or even Mr. Alex Roman you might want to split things up
in a way that allows more room for control and possibility.

Instead of being frustrated because your direct doesnt look like so & so

This is why I believe its great to do this.
By chopping up Materials, Environments & Lighting Setups into different parts you are able to concentrate on them.
After some practice you will understand how these different attributes of your scene come together and interact.

Multilight is very interesting since it allows you to create Versions off one ML rendering.
It can be very interesting in Compositing as well.

Alot of people render out a GI pass, AO then each light is rendered out seperately so you're able to control the strength in Comp lateron.
We can do this with a ML rendered MXI. So in a very blunt way, we are already compositing the second we are using ML.

I want to add that of course you're starting to leave the sphere of unbiased rendering by including techniques like these.
Anyhow enough yadda yadda, I will hopefully have something for you soon.

Perhaps I'll just start it out as a simple PDF Paper and lateron do a big Tutorial kinda of thing.

[Richard]

Thanks again Joris!

I think though mate one thing to always be considered is that of production time!

Rendering takes long enough it's self, to add time in post pro that could be achieved to even 90% close to the ultimate effect for some is close enough. In the one man design office who counts on minimal work time any single functionality thats reduces overheads of workaround and post processing is not just a luxury but almost one of nesessity!!!

I understand your point but being someone who utilises MR to support their designs not to project ones shear skills in rendering I support anything that not just improves my output but reduces my production overheads!