VRay Sun, Sky and Physical Camera video tut + Bonus Script!

[teabag]

[aaron-cds]

my 2 cents ... watching the videos, I got the impression that the .255 value relates to the 385 value when the sun is directly overhead, which is a starting point. When the sun moves, you'll adjust the iso value to your preference. That's how I understood it.

[studioDIM]

Lele, you will, some pages more all side effects of your method, that cause problems, are confirmed:

Micha, the only visible "side effect" in this thread is you.
You're a FEW inches under my skin by now.
The only really glowing thing of all these "four pages" is your utter presumption,asymmetrical lack of understanding of the English language (that, or you have serious issues maintaining concentration through a post. English comes fluent enough when you want to be sarcastic though. hence the asymmetry), and unwillingness to put yourself in the same position as everyone else before starting to blabber about things.

But let's go point by point:

- indirect light bounces are affected (my and Wouters tests show it)

Yes, that is PRECISELY the bloody point of the whole tutorial: to affect the way light is distributed.
Now you got it.
Glad for you.

- reflections are limited and ~100% like mirror is not possible (your next image will show it)

How can a multiplier of 0.255 lead to a PERFECT MIRROR?
You ought to quote me on that thing you say i said.
I merely showed you it DOES behave differently with a physcam, a gamma ramp and strong light intensities.
And what looks nearly white in the rendered pixel is instead a lot dimmer when the material is prepared.

- the visible sky and all visible emitters are brighten by factor 4 (your examples show it)

Yes, second point of the bloody tut, geezer.
Wake up and smell the coffee.

- the side effects are independent from sky/sun/cam

Indeed, they're a LOT more dependent upon the user.
I think you showed us well enough, there.
Doubting a color pick is beyond words, in MY book.

Lele

[studioDIM]

lele,
I came up with + / - 2.0 float at the brightest sun lit chair area. I used the linear multiply with gamma set to 2.2 for color mapping. The materials are all at 1.0. The materials were too dark at .255 or .5 and when I changed the exposure to lighten the interior better the windows got completely blown out.

Yes, that's what i gathered from your original post, and what made me wonder.
What i do to control the sky blowing out (which of course it does if we lower the diffuse coefficient: it's at its darkest given an exposure when the diffuse coefficients are at 1) is to put some fog within the glasses.
Adds nothing to rendertimes, but allows me to control the burn amount exactly.
The dimming of GI is negligible compared to the dimming of the blow-out, in my bar preliminary tests of a few posts back.

my 2 cents ... watching the videos, I got the impression that the .255 value relates to the 385 value when the sun is directly overhead, which is a starting point. When the sun moves, you'll adjust the iso value to your preference. That's how I understood it.

Yes, thankyou Aaron, you got it spot on.
It's meant to be a starting point: i wanted to stress that materials COULD be a lot darker, and still work nicely enough under those lighting/exposure conditions to represent a given colour, with a few benefits, and some more work.
I say so in the videos, and said so a FEW times now in this thread...
FEEL FREE TO EXPERIMENT! (possibly without trying to lecture the lecturer: he much prefers coherent discussion, I heard...)

Lele

[aaron-cds]

perhaps I'm trying too hard, but... or maybe I should just go with what looks good. so here's my question.

If I've got the sun at a +/- 30 degree angle from the horizon and I set the default material to a vraycolor white at .255, should the float value of a lit surface perpendicular to the vraysun still be a float value of 1.0 in order to be the most accurate?

Do we always need a 1.0 float value for exterior scenes for correct exposure regardless of the sun angle?

[studioDIM]

Well, a value of 1.0 only ensures you are "wasting" no colour informations.
Wether that is the most accurate, or better looking, i wouldn't know
Besides, I'd say you ought to go with what looks good, in general terms.
You might wish for an overblown area, or a very underexposed picture, even in real life.
As of now, i so wish for a nice looking pint (or two. need to steam off... was it discernible, you say? ) i might just be heading off.

Laters!

Lele

[DaForce]

Well said Lele... all of it.

[teabag]

Lele: I hope u did not get too drunk ....but i guess that depends on the amount of steam u gathered here yesterday

Micha: Man i really think u should let this thread go. If u start your own thread, most of the people on this forum (al least the ones u didnt make mad...and that is said nicely) will be more likely to discuss your questions. I am very much trying to understand what the guys here are trying to come up with and your stupid quotes, replies and answers make it one big mess that one needs to filter through....for a newcomer on this forum....well done...u probably made a bad image of yourself, which is a shame because from your portfolio ive seen, u probably have some interesting ideas, hints to offer (or not) in certain fields....please do not post in this thread anymore....become a spectator instead..at least until u get MAX. Thanks very much for understanding.

[Micha]

Do you also apply the 0.255 multiplier to reflection, refraction and whatever other color swatches?

Again, yes and no.
Yes i normally dim reflections a lot (if diffuse was one fourth of the old value, this can be the same, or slightly higher if it needs to be a dominant reflection) , but do not change bump, for instance. That's an "unshaded" channels, in that it only depends from light /camera incidence, not so much the intensity of it.....

I leave the thread here. I see, some people are serious testing the method and don't get nice results, other people don't test it, but dicuss it in a way, I don't want to follow.
Thank you Wouter and Aaron for the serious tests.

If nobody see, that this simple example of Leles chrom balls are not chrom balls, than it makes no sense to talk about it anymore. I ask for a mirror test and got an image, that show the env at the mirror but not at the background, that show a mix of diffuse and reflective material. It's not a typical mirror.

Now, Lele jump from "dim your reflections" to "let it be at 100%" and nobody seems to note it.
So it is with all critical comments, after a while, anything is accepted, but I'm the bad guy. Bad game.

I'm curious to see images without burned out backgrounds, without burned reflections, with right multiple interactions of reflections, proper diffuse bounces and maybe some colorbleeding. Lele, you should do it. I wonder me, so much words and so less images.
And yes, multiplier 1 cause blow-out, because 100% diffuse reflection is not natural. But multiplier 0.255 cause a sky blow out and I'm curious how you catch the sky again, like it every good customer digital camera can do. If I take my old digital camera exposure at a white surface, than the sky is not burned out. And every new good camera can exposure shadow and light in the same images, like the old analog cameras.
Here some photo examples at this link. Your examples should look like winter photos with blue sky or?

But no comment to this subject by me anymore here. (Thanks Teabag for the hint. Maybe you like to test the method and to show some images here. But be careful. ) From me it was a bad idea to discuss the method and the problems that could cause by it.

[instinct]

And every new good camera can exposure shadow and light in the same images, like the old analog cameras.

Sorry, i dont know if it is just a language problem, but that is plain wrong. period. That's the reason for HDR photography in the first place. For once your link points to utterly badly retouched pics with contrast changed etc etc.... A camera (and i assume you meant consumer, not customer) can NOT expose both shadows and a bright sunny sky. You will loose information either on the shadows or on the highlights. That's the point of shooting at multiple f-stops and combine them.

And for the rest. This whole thread is about discussion. But about constructive discussion, but your posts are simply put offensive, and i can VERY well understand lele going through the roof (actually i gotta admit that i wouldnt have been able to keep calm as long as lele did). You present any comments and criticism in such an arrogant and ignorant manner. And this is the bigger problem.
Plus i didnt get your whole Point in the first place. Your first post was about the method not beeing physically correct. Wich is completely unrelevant, as it is not MEANT to be physically correct. That's like "Hey i like the new Lotus" Reply:"Nah, you're proven wrong, cause it's not a banana"

[instinct]

As you were complaining about too much words (wich is odd in discussions :P ) and too little images, here's too images from an HDR session we did. Two different exposures shot at the same time, and location :




Take a look at the Shadows under the car. Even tho the Sky is already blown out like crazy, the shadows under the car are already pitch black. And this wasnt even a cloudless sky day.
And on the dark picture all of a sudden a lot of details appear where the bright pictures shows only whites.

As a sidenote this wasnt shot with a consumer camera.

Thorsten

P.S. As a quick sidenote, these pics are untouched, directly taken from the 16bit tifs the camera outputs. Just scaled down and saved as jpg.

[DaForce]

That's like "Hey i like the new Lotus" Reply:"Nah, you're proven wrong, cause it's not a banana"

awesome, nicelly put.
Sums up the later part of this thread to a T

[Paul Oblomov]

Awesome thread. Spent a night to figure out what's the deal

BTW - anybody tried to render in multiple exposures (EV, in PS terms) and then merge all 'em to one pure HDR file ?
I tried (years ago) - and must say it's kewl Damn long, but looks like method used in real-life to get HDRI. Nowadays, plain .exr is enough. Mostly.

[teabag]

cool...
...guys lets make one big thick line here and start from begining? could anyone please take a little time and sum up all the relevant information on this topic now? Lele? Daforce? (i would ..but i got lost myself )

[Neilg]

could anyone please take a little time and sum up all the relevant information on this topic now?

If it looks good, use it. If it doesnt, try changing the numbers a bit.


Thats really all it should ever boil down to.