When Trying To Light an Interior In Iray..
404nicg
Posts: 270
in The Commons
What's the best way to emulate the sun coming in through the windows? I'm not having any luck using the default dome that comes with Iray.
Daz 3D is part of 
Connect
DAZ Productions, Inc.
7533 S Center View Ct #4664
West Jordan, UT 84084
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2025 Daz Productions Inc. All Rights Reserved.
Comments
Could you be more specific? Maybe show an example and relate what settings you are using?
Among other things:
Make sure the window is actually transparent (might seem obvious, but hey). Maybe make the entire window invisible until you get close to what you want.
If you are on the default settings, you don't have a sun simulation, you have a fairly even background image. If you go to Sun/Sky environment or remove the Environmental map you'll get something like a sun with directed sunlight.
Currently lighting via just windows is hard (it can be done but it will take forever).
Thankfully there are many things to make things easier,
Use the sunsky and a spot light: In the render settings tab>environment make sure there is no environment map add a spot light make it so you are looking through the spotlight in your viewport (top right) and position the spot light such that is is looking through the window you want the sun to shine through from a good distance, in render settings>environment set the "SS sun node" to your spotlight. you should now have direct sunlight coming through the window.
If you want a well lit room cheat and add some fill lights inside. As it turns out having light start outside come through a window bounce aff the floor, bounce of a wall and bounce off another wall or 3 takes a lot of calculation, it can be done, but it will take forever to render. Add some nice fill lights in the room (a spotlight with the geometry set to to rectangle and a large area) and place them off stage in logical positions (If your window is on the right put a stronger fill light off to the right and a weaker one to the left) Tins will mean more of the scene is being lit directly, which goes much faster.
Sure here's what i'm getting and the settings which i believe are pretty close to the default
Do the DAZ default lights work well in iray at all?
Yes...the Daz default lights, since one of the 4.8 betas, are context sensetive. So, depending on which renderer is used, they are the correct lights for it.
So using Iray exposes the Iray specific controls for them...in which case makes them the basic Iray lights.
In 99% of the interior shots i render, i set the ceiling to emissive and use fairly low lumens - unless the ceiling is geometrically complex (high poly count), then i'll just use a plane. This acts as a fill light and you can simply boost the environment intensity to offset the ambient light and make the areas near windows brighter. I don't often use photometrics as i prefer softer shadows, but they do render quite quickly. Just my 2c.
You can also just give a light planar geometry of the right size.
Another option is to use invisible lights inside the room. Some spheres with emissive shader and Cutout 0.000001 can do the trick to create some ambient light.
Sometimes I do a combination for a single light source, which I find helps.
Fore example, you have a wall light that you've made the bulb emmissive. To get it to put off enough light, you have to crank up the lumins on it, which washes it out a lot. So what I emded up doing was setting the lumin between 50,000 and 100,000 then adding a daz default light in the same spot with another 100,000-150,000 lumin.
Another way...don't worry about what is outside the window (draw dome off...just use the light) then adjust your tonemapping to get a decently lit room. Then in post, add a view out the window.
That's the tone mapping settings.
Tone mapping = exposure controls in a camera. There are no 'auto exposure' settings in Iray.
Just as a note, using Lumens as the units for lighting in Iray makes for large values in most cases.....this is due to it having to scale it based on the rough surface area of the emissive object. Which means, if the object is not very small, huge amounts of lumens have to be set for it to even show up. It's better to use Candela/m^2 or Candela/cm^2, which lets you know based on the size of the object (roughly, since it's based on surface area) how many candela are being emitted.
Also, using the latter settings, if you scale the light-source, the light intensity will remain consistent. If you use the former (lumens) and you then scale the light source, it will get dimmer and dimmer (same amount of luminous flux, distributed over a larger surface area.)
Also, based on 360° emission, it is roughly 1 candela to 12 lumens. It varies considerably if the emissive surface isn't spherical, or only emits in certain directions.
One other tip, in render settings>tone mapping you can set "crush blacks" to 0, this will lighten up the shadows. Doesn't look great in every scene, but I fint it often works pretty well.
I find that strange...if a sphere is 1 cm or 1 mtr diameter and set to 100 lumens, both will emit 100 lumens of light this goes for Watts also. It is the other settings you mention that takes the size of the object in to the equation.
The Lumen is the unit of luminous flux, it is the 'total quantity of visible light emitted' from a surface. Which means, if the surface becomes larger, while the number of lumens remains constant, less light is emitted per unit of surface area.
The Candela, is the unit of luminous intensity, and is 'luminous power per unit solid angle emitted by a point light source in a particular direction.' Hence, the candela takes into account radial distribution, the lumen does not.
It's confusing, I know. But it has to do with how these units relate. To wit:
1 Candela x 1 Steradian = 1 Lumen
Thus when the unit is Candela/m^2, you get the radial dispersion over a given area. So it scales as the emitting surface scales. But with a standard Lumen, it doesn't. If your object were an infintesimally small sphere, 1 Lumen would emit about 1/12th of a Candela (4π steradians in a full sphere.) But as the sphere grew, it's surface area becomes a dividing factor, and the luminous flux emitted at every point on the sphere will reduce (while the total light emitted remains constant). So the light emitted in any given direction from the surface is reduced. Imagine a normal 60w light bulb that is a meter in diameter (but isn't filiment based, just surface)......it would still consume 60w, but the light coming out from it would be distributed over a much larger area. So the bulb would appear very dim....so dim, it would only be noticeable as emitting light in a very dark room. (Area of normal sized bulb is about 49π cm^2, where as a 1m sphere has about 204 TIMES that surface area, so the luminous flux emitted at any given area would be that of a 1/4 w light bulb.)
The size of the emitter shouldf affect how bright it loos from a set distance (you are seeing the inverse square law made solid) but the light reaching any point outside the sphere should surely not depend on the size - at least for any point sufficenetly dsitant, as the cone defined by point closer to the surface and and lines tangent to the surface would enclose a smaller area as the surface grew.
This is where it gets counter intuitive. The amount of light at any given distance from the emitter will be constant (for that distance.) However, the light amount incident at any given POINT at a given distance varies based on the size of the surface (due to the light being emitted perpendicularly to that surface.) As a visualization exercise: Imagine your light source as now being big enough that the surface is ALMOST touching your distant point in question. How much of the light now misses the object entirely? How much actually hits it? The amount of light leaving the emitter surface is constant, but its distribution area is now larger, thus the light per steradian is smaller.
Also, remember the Candela, being a unit that includes the angular section as part of the unit ( 1 Cd = 1 Lm/Steradian ) normalizes the quantity of light per angular measure.
Another visualization exercise: Imagine a point light source INSIDE your actual emitter surface (which we'll choose as a sphere for simplicity.) This point light is emitting 1 lumen total. The amount of light emitted from the emitter surface is 1 lumen as well.....same amount of light. But the intensity of the light at any point on the emitter surface depends on how far from the center point light the surface is, specifically 1 lumen / total area (the total area in this case is 4πr², the surface area of a sphere.) So the intensity of light being emitted at any given point on the surface of the emitter depends a LOT on its shape and size.
Now consider the beam of light emitted from a given quad in a surface mesh, and it's direction, as a 4-sided pyramid. It starts at the point light center, through the emitter surface mesh, and out into the scene. The amount of light is constant (since the angles won't vary as we scale the surface up) but the intensity of the light changes. Same as the fall-off per distance, so at double the distance, it is 1/4 the intensity.But since that surface is our ACTUAL emitter (not our point source for the exercise) the lumens per angle are being emitted from a larger area as the size of mesh emitter increases. So the total light being emitted is divided over the surface. Thus, larger surface = less light per unit of surface area.
Now a Candela is the unit of light per unit of surface angle (which at a given fixed distance is a fixed surface area.) This means that regardless of the SIZE of the emitter, the same intensity is emitted per surface angle. In order for this to hold constant while the emitter itself increases in size, the number of lumens has to increase as well.
It's hard to describe without good illustrations....and I can't find any. May have to make some tonight.....
The critical thing to remember is that Lumens and Candelas do NOT measure the same thing. Lumens are the total QUANTITY of light emitted, whereas Candelas are the INTENSITY of light per solid angle emitted.
Not much difference when dealing with sound (volume and loudness are not the same...).
Thanks, I should have made my post more of a question than a statement. The concept of how much light could reach a point was what I was trying to cover when I burbled about cones.
Well as that contradicts what I have read on these forums I am now intrigued to what is correct. But please leave out the math and keep it to plain English which will help many of us hobbyists. So what you are saying is that Lumen and Watts are size dependent where as the other are not???
I found this link to be helpful in showing the difference: http://follyfool.deviantart.com/art/Daz-Iray-Emission-Types-550609853
Ugh :( and here I thought it was just different measurements for the same thing (such as feet, inches, etc.)
In the context of 3D and Iray, yes.
Lumens (light amount) and Watts (energy) are total light energy emitted. In ALL directions from the surface.
Candelas (light intensity) is the concentration of light being emitted PER UNIT of angular solid volume.
Candela/cm² and Candlea/m² are derived values based on how much surface area is covered by the 'angular solid volume.' And it represents the concentration of light emitted from every such 'unit' of surface area.
Math actually helps a bit here. What is a lumen?
If you have a unit sphere (1 ft, or 1m, both work) centered on a 'standard candle' (1 candlepower, approximated as a point light), and you take 1 square (ft or m) on that sphere's surface, there is 1 Lumen passing through that area (which is 1 steradian, if the area is circular)
Therefore, total lumens passing through the unit sphere is ~12.57 lumens. ( surface area of a sphere is 4πr², so unit sphere has 12.57 square units area )
But Candela are uniform through out the surface, regardless of distance. 1 Candela is the intensity on each and every square unit of the surface.
Lumens are 'luminous flux'. How much (quantity) of light is passing through a given area.
Candela are 'luminous intensity'. How bright (quantity per area) is the light at a given distance over a certain area.
(there's also Lux, which is different, and deals with how the light illuminates the surface. But it's confusing enough already, and we can't directly control that....it's part of the Iray shading BRDF functionality.)
yeah I know what a lumen, lux and candela are as I used to sell a lot of stadium and factory lighting years ago and it is still up there in my head. I know what the inverse law etc too. I just wanted to clarify the contradiction of information, Thanks
Maybe the confusion comes from the varied ways emissive lights are used. One use is as scene illumination (generally to be avoided for reasons discussed in this thread), another is in-scene elements, such as TV monitors. The emissive may or may not use a map, but the results are the same in any case.
For in-scene elements, using the cd/* units allows you to maintain the same apparent brightness regardless of the size of the surface. See the sample figure. I use cd/cm^2 on a 4 and 8 meter plane. I prefer cd/cm^2 because A), the cm is the default units of measure for all lighting in C|S, so it's easier to keep things straight, and B) it doesn't require astronomical luminance values. In fact, for the top examples the luminance is set at just 2 -- not 2 million, just 2.
The bottom examples use lumens for the plane, with a luminance value of 500,000. Note that apparent brightness of the larger plane is less than that of the smaller plane, though the luminance values are the same. The effects are readily observed.
Oh sweet Tobor thanks for that. I was about to test this myself but I have been rendering all day. And thanks to hphoenix for correcting my understanding.
So, this all sounds like I should have used pin lights to simulate candelights, right than emissive lighting with the flame surface on the props, on a recent picture I did (in a 3DL version, I used AOA ambients at the candels).
This was very helpful thank you!
That's how I do it too because IMO it gives the most realistic lighting, albeit somethimes terrible render times. For daytime interiors I often use an emmisive wall or plane with the light coming side-on like through a window.