[Cosmasad]

Good morning everyone,

We are a group of architects working in Washington, D.C. on a restaurant interior. There are 110 different lights in the restaurant, -- 50 tracklights with MR 16's, 40 downlights with floods, and 40 table lamps. We are following up on a message that we posted in another part of the forum.

We understand the concept of the emitter material, and have been told that when creating a light, it is best applied to a sphere 1 cm diameter. This is all well and good, but intuitively it seems that a sphere that has emitter material applied to it would radiate light evenly in all directions. It would glow in all directions. We have been told, however, that "the sphere is like any other geometry, you can manipulate its position and rotation -- the IES material will follow where the sphere points once it is applied." But the sphere doesn't point... it is perfectly round. Then Daniel posted that "if you are in studio you should immediately see the ies distribution diagram in the viewport (blue dots) after applying the emitter material to the sphere(s)." It seems that the ies function is not supported in Rhino yet. Can anyone tell us if there is a workaround for this. Should we be working in Studio instead?

Somebody else said that "the normals of the Sphere will determine where the light is pointed -- after you apply the material just move and rotate the sphere like a flashlight/spotlight..." but again-- a sphere glows in all directions evenly. Could it be that people are using the word sphere to mean a circular surface 1 cm in diameter?

Your assistance with this would be appreciated.

Thank you.


Cosmasad

[JDHill]

The advice about using spheres does not apply in the case of Rhino, because geometry in Rhino exists directly in world space -- it uses no concept of local coordinate systems as some other modeling apps do. If you are using the release preview 2.5.1 plugin linked above (http://www.maxwellrender.com/forum/view ... 05&t=35883) then you can try this:

1. make a new IES emitter material
2. make a Rhino spotlight
3. assign the material to the spotlight

Since a spotlight has a position/direction, the plugin is able to build a coordinate system for it, and this allows Maxwell to consider the light to be pointing in a particular direction.

[Cosmasad]

Good morning Jeremy,

As a relative novice to Rhino (I have been working with it for a few months) and only about a week into Maxwell, I appreciate your patience with my questions. I have a restaurant interior that I have built a model for and it has a large number of lights: 50 overhead tracklights (50w MR16 halogen tracklights), 40 overhead downlights (recessed ceiling canlights with Par30 halogen bulbs) and 40 table lamps with compact fluorescent bulbs. The tracklights need to be angled in specific directions, the downlights point straight down, and the table lamps radiate light through their shades. In the background there are two large windows with daylight pouring in.

Now, I have two questions relating to this. The general community (Users Forum, Dario at Technical Support, and the Maxwell Render Manual) say that I should use a sphere 1cm in diameter when I want to create a light that is linked to an IES emitter material. You seem to feel that I shouldn't. I will take your word for it. Does that mean that I should create bulbs that have the same shape and size (anotherwords no 1cm spheres) as the real bulbs we are specifying and assign the material to them? If so, should I opague the back of the bulb so light doesn't leak out the back or will the IES file control the direction?

My second question has to do with a feature mentioned by another user forum member (Daniel aka Polynurb): an IES distribution diagram "drawn in the viewport with blue dots" that seems to be a feature of Maxwell Studio but it isn't clear whether it is available in the Rhino plugin. This feature apparently would help me point the lights. If this is true (that the feature is not available in the Rhino plugin) do you think that I should be working in Studio instead of the Rhino plugin, or is there a workaround in the Rhino plugin?

I'm going to try to say this nicely -- but it may not sound nice -- I have to say I am surprised that there aren't more clear step-by-step instructions on this either in the Manual or on the Think website. It is such a basic subject and I have had to spend three days trying to figure it out.

Anyway, I really appreciate your help. We are a group of architects who very much want to master Maxwell. It is a wonderful program. We just need to find our way through it.

Thank you.

Cosmas

[JDHill]

Now, I have two questions relating to this. The general community (Users Forum, Dario at Technical Support, and the Maxwell Render Manual) say that I should use a sphere 1cm in diameter when I want to create a light that is linked to an IES emitter material. You seem to feel that I shouldn't. I will take your word for it.

That's probably a safe bet -- I write this plugin. They are telling you to use a sphere elsewhere because that makes sense elsewhere; not in Rhino though, because as I wrote previously, a sphere in Rhino has no local coordinate system. The suggestion I gave you above was to use a Rhino spotlight with an IES-based Maxwell emitter material assigned; it is the direction of the spotlight which will determine where the IES lobe points.

Does that mean that I should create bulbs that have the same shape and size (anotherwords no 1cm spheres) as the real bulbs we are specifying and assign the material to them? If so, should I opague the back of the bulb so light doesn't leak out the back or will the IES file control the direction?

If you do as I suggested above, you will not need to concern yourself with this; the plugin will generate appropriate geometry for your lights at export time, and your IES-based emitter will be applied. If you do not use IES then yes, you would want to model your lights and fixtures realistically.

My second question has to do with a feature mentioned by another user forum member (Daniel aka Polynurb): an IES distribution diagram "drawn in the viewport with blue dots" that seems to be a feature of Maxwell Studio but it isn't clear whether it is available in the Rhino plugin. This feature apparently would help me point the lights. If this is true (that the feature is not available in the Rhino plugin) do you think that I should be working in Studio instead of the Rhino plugin, or is there a workaround in the Rhino plugin?

No such pre-visualiztion of the IES lobe is currently implemented in the plugin. However, you should be able to point your IES lights just fine if you do as I suggest above and create them using Rhino spot lights.

[Cosmasad]

Thank you Jeremy. I really appreciate it.

[softroom]

Further to this topic, I would appreciate any help on the following;

In order to keep down the number of multilight channels, I am used to grouping emitters by first creating meshes from nurbs primitives and then joining those meshes into one entity. With an IES emitter applied even to a joined set of spherical meshes, it seems possible to have each sphere emit correctly ... except that there's no control over direction of course!

With the trick of assigning and IES emitter to a rhino native spotlamp, I understand that it is possible to create an IES lamp that could be directed. Which sounds good, except that presumably there's then no way efficiently joining a number of these lamps into a single multilight channel. Or is there?

Also, while IES is OK for making spot lamps, is it possible to get the IES data from a linear source like a wall-washing striplight to do anything accurate if IES data can't be mapped onto anything other than a sphere

Many thanks in advance. - Oliver

[JDHill]

In order to keep down the number of multilight channels, I am used to grouping emitters by first creating meshes from nurbs primitives and then joining those meshes into one entity. With an IES emitter applied even to a joined set of spherical meshes, it seems possible to have each sphere emit correctly ... except that there's no control over direction of course!

I wouldn't have guessed this would work at all, but apparently the engine is assigning a separate IES lobe to each individual piece of the joined mesh.

With the trick of assigning and IES emitter to a rhino native spotlamp, I understand that it is possible to create an IES lamp that could be directed. Which sounds good, except that presumably there's then no way efficiently joining a number of these lamps into a single multilight channel. Or is there?

Sorry, but no, there isn't. The IES depends on having a transformation matrix applied to the mesh where it is applied. If you combine several meshes together, you naturally have only one possible transformation for the joined mesh. No matter though, since in the case of using Rhino spotlights, there is not even any way to join their meshes (i.e. the meshes generated by the plugin) together.

Also, while IES is OK for making spot lamps, is it possible to get the IES data from a linear source like a wall-washing striplight to do anything accurate if IES data can't be mapped onto anything other than a sphere

Possibly, but I am not too knowledgeable about that. You can try using a Rhino rectangular light -- the plugin will also translate those into geometry. There is really no real-world correlation when doing this, but maybe it will work for you.

[ricardo]

On the matter of IES preview, Windows 7 gives me the diagram when I set the folder to large thumbnails. I have no idea if this is native windows behaviour or installed afterwards, but I have not much more than Rhino and Maxwell that could deal with such file types.

EDIT: There are IES files for stuff other than spots: http://metalumen.com/products/ies?family=21&item=50

[ricardo]

On the material assignment side, make a small square plane and use it as emitter. The plane will have a nice normal to be pointed.
One problem I found is that these planes may render black randomly. The suggestion I got in the main Maxwell forum was to hide the emitters from camera. You probably would need to create another small emitter to shine to the camera, and that emitter should be very low intensity or hidden from GI to avoid adding to the ambient lightning.

[softroom]

Many thanks to both of you for the quick responses.

@ JD Hill - I will probably try a workaround whereby spotlamps pointing straight down are IES, and angled spotlamps are modelled.

@ Ricardo - Thanks for the metalumen link. I am assuming that when you say

On the material assignment side, make a small square plane and use it as emitter. The plane will have a nice normal to be pointed.

this is referring to a modelled light source, not an IES one.

[ricardo]

Now I went cheching what I said and you can't change the light direction by rotating the plane... The IES stands stil..

But JD, Can you assign an emitter to a Rhino spot light? The material got assigned, but didn't render.

[JDHill]

It (IES emitter applied to Rhino spotlight) seems to be working fine here, could you send me a file where it doesn't?

[ricardo]

I'll assemble one later, but as a heads up, it was V5...

EDIT: It is indeed working, I don't know what I did wrong before.
Sorry.