Physical equivalences of Light

[areyal]

Hi all

After trying Octane I'm very impressed with lights.

I have a few questions:

- I'd like to know what is the phisical equivalence for the variable "power" in light materials (if it can be established).
It wolud be nice to know the equivalence in lm/m2 or other units.
That would be great.

- I don`t know if there is a way to tweak the control of exposure in order to obtain a image (for example in gray scale) calibrated to know what is the iluminace or luminance of the scene (I think luminance would be easier but iluminance is the key for me).
As Octane is an unbiased renderer it would be great to have such possibility.
For example you can do that with 3D Studio with "radiosity". It is quicker but less precisse and the most important thing is that it cannot pay attention at specular reflections/refractions and caustics.

- I've noticed that the power variable of emission has this range: 0,01 to 100
I find it a bit restricted, because I use to render exteriors, and when I have a few point lights and a "wall of light" in the same scene, the wall has much more surface.

Thanks

[necko77]

Hi all

After trying Octane I'm very impressed with lights.

I have a few questions:

- I'd like to know what is the phisical equivalence for the variable "power" in light materials (if it can be established).
It wolud be nice to know the equivalence in lm/m2 or other units.
That would be great.

- I don`t know if there is a way to tweak the control of exposure in order to obtain a image (for example in gray scale) calibrated to know what is the iluminace or luminance of the scene (I think luminance would be easier but iluminance is the key for me).
As Octane is an unbiased renderer it would be great to have such possibility.
For example you can do that with 3D Studio with "radiosity". It is quicker but less precisse and the most important thing is that it cannot pay attention at specular reflections/refractions and caustics.

- I've noticed that the power variable of emission has this range: 0,01 to 100
I find it a bit restricted, because I use to render exteriors, and when I have a few point lights and a "wall of light" in the same scene, the wall has much more surface.

Thanks

First, u must know for what some room(space) will be used
Then you will have to decide how much foot candles you need. (resotrant, room, office, parking etc etc
Then you do lighning scenario (on paper) just to see what part of room (space) will need illumination and how much, and what effect you want to achive
After that u will have to use IES lightning cos with that it is possible to see how much lumens is goign to reach floor, celing, walls...and you can choose light form catalogs and you will recive lamp in 3d + how that lamp distrubute light
I can see great renders of interior in octane but none of them are correct !

Octane is far away from lighning desing program and its lightning is not for some serius work, it is far away from it. It is very important when octane team implement IES lighning in octane render that we r able to see info about how much light (lumens) arrive to floor, celing, walls and objects. Without it, octane is for children, they can be impresed with its lighning which means nothing in real world scenario.

radiocity or what ever is need cos of its speed of calculation of light
So if you want to do lightning design in octane now, u will make render which looks perfect but u will not be able to tell what light to use to achive that lightning scenario and thats why octane render is wrong tool for lightning which must be changed. If real results are needed then you must use IES lighning in interior and exterior renderings.

[areyal]

Well, I don't need a precise calculation (for that there are other kind of software of course)

I think that the fact that Octane is an unbiased renderer can have the advantage of making an aproximation about values, only to know the order of magnitude where you are.

Radiosity is fast and accurate in terms of physical values, but inaccurate in terms of image. Before Octane I only used radiosity in 3D Studio, mainly because raytracing was too slow. Also it was too imprecise in terms of real world to know where you are moving on.

I think the point of Octane is a medium term: you have a raytrace renderer that is fast and good in terms of image quality, and as it is unbiased it could give you an idea about real lighting only describing a few equivalences. That would be enaugh form me until IES lighting arrives to Octane.

I had the opportunity of try iRay of MentalImages for 3D Studio, and after some tests I concluded that it is physically accurate. It delivers same results that radiosity, but with good image quality. That didn't happen with the biased renderer (mental ray) which they said it was precisse.

With the same simple scene, I could reproduce the same results in Octane and iRay (except for some material because I didn't tune it enough). If you have interest I can post the two images.
The only difference is that I couldn't say to Octane "so many lm" or "lm/m2", I had to tune it manually.
The point is that for that scene (100 passes) Octane was cooking for half the time of iRay and it delivered less noise. Also important is that I could tune the image in octane in realtime.
That taking into account that iRay didn't supported some graphic card that Octane did and iRay was using 100% of my quad processor plus GPU while Octane was using oly 25% CPU + GPU

[necko77]

The only difference is that I couldn't say to Octane "so many lm" or "lm/m2", I had to tune it manually.

thats why we need IES lights, to tell octane how light distribute itself over some space
im architect, and i need this a lot as other architects,interior designers and people whos doing renderings for them
when i say that i want to use 6 Sento letto 40,Wall-mounted luminaires u r saying that you will be bale to tweak that. Let say you can, but u will loose two days on tweaking and u will not get accurate results
I want to import real world lights in my projects and get accurate results, bias or not bias it dosnt matter to me and to see it lumens
U dont need it and i dont need to know it, be constructive, write what u need, dont be destructive it is really needed option for octane cos it will be more spreded software. more cool about octane

[areyal]

Ok, ok, I understand, but don't missunderstand my words please.
I'm really pleased with octane (otherway I wouldn't buy it).

Octane is a beta program so I talk about how it is and can be, and the problems I have. The program is cool or not because of itself, not for what I say about it.

In resume, i only wanted to know "aproximately how many lm/m2 is POWER=1" for example. Not because the result is bad, but because it is easier for me to predict and choose the power in long term scenes. To put references to physical units in the camera (ISO, exposure, aperture) is not so useful if there are no physical units in the lights.
(I understand that it is so at this time, I'm not being a destroyer).

A thing I disagree with you: You don't need IES to be physically accurate. You need IES if you want to be physically accurate with commercial luminaries in the scene.


PD. The fact is that I find Octane very revolutionary, mainly for the concept of the interface.

[necko77]

great, its nice to see that we understand eachother

i dont think that you are destroyer and as light designer i need IES cos i want to be physically accurate with commercial luminaries in the scene, thats right !
In my projects if im doing light i always use existing products cos when clients likes result i can tell them what to buy to get that particular result...IES is real timesaver.
ofc, we need lm/m2 or footcandles or lx
this octane power seting of light is for kids, not for serius work

cee ya !

[radiance]

Hey,

Octane does not have radiometric inputs for light intensity, however that is quite simple a feature to add.
I have already said we will add IES support soon, so we can probably add some radiometric controls for defining area light intensity too. (lm2 / watt / efficiency)

Radiance

[areyal]

nice!

That would be very useful if you don't have a IES profile of a lamp (or it is a specific system).

cheers

[necko77]

Hey,

Octane does not have radiometric inputs for light intensity, however that is quite simple a feature to add.
I have already said we will add IES support soon, so we can probably add some radiometric controls for defining area light intensity too. (lm2 / watt / efficiency)

Radiance

thx man !

[suhail_spa]

Hey,

Octane does not have radiometric inputs for light intensity, however that is quite simple a feature to add.
I have already said we will add IES support soon, so we can probably add some radiometric controls for defining area light intensity too. (lm2 / watt / efficiency)

Radiance

this is really amazing!
thanks a lot Radiance..
you rock