Rotating IES files?

Come here for help & support.
Post Reply
7 posts • Page 1 of 1
BartS
Posts: 7
Joined: Wed Sep 21, 2011 9:53 pm

Rotating IES files?

Post by BartS » Tue Sep 01, 2020 1:23 am

After several years, I'm returning to Indigo Renderer, and I'm happy to see lots of improvements since my last use (~2015!). but I'm running into an issue.

For a new project, I have to use non-rotationally symmetrical IES files. That proves to be a problem. I have no problem (well, a small one - see below) using IES files, and they seem to render quite well. But once the IES file is *not* rotationally symmetrical, I cannot find how to control this rotation.

In more detail - for e.g. a streetlight, I could use a simple square, emitting straight down. The IES profile of such a light would typically spread the light along the street, rather than projecting a circle on the street. But *how* do I specify the rotation of the IES file around that vertical axis?

In other words - if that street runs east-to-west, and my light profile on the ground 'happens to appear' as north-south oriented, how do I change this?

Some background - I'm using 3DsMax to create my model; and using emitting Indigo materials to create the light sources. I simply assign these to an object. Although I cannot find how to attach an IES file this way, I made a small script to add it to the exported XML file before rendering. (that's the small issue - it's slightly inconvenient).

I have tried UV maps and rotating those, and another suggestion I found was to rotate the pivot; but nothing seems to make any difference.

So, is there a solution?

Thanks!

User avatar
thesquirell
Posts: 428
Joined: Fri Aug 29, 2014 3:49 am
Location: Novi Sad, Serbia

Re: Rotating IES files?

Post by thesquirell » Tue Sep 01, 2020 3:03 am

Hey there BartS,

Welcome back. Glad you're enjoying new Indigo. Make sure to give GPU a try, it's amazingly fast, and it even supports SSS!

As for the IES profiles, I was able to change the direction of the spread by simply rotating the emitter object. Bear in mind that this way of controlling the direction of the spread only applies to the angels 0 and 90, meaning that you can change the direction only North-South, or East-West, without incremental values in between. I will go ahead and check some other ways of doing this, possibly by using 3rd party software (like IESViewer), but I'm not getting my hopes up.

As for the record, I'm not using 3ds Max, but Cinema 4D, not sure if that makes any difference. Anyway, try rotating your object 90 degrees, not 89, because that's the switching point.
IES_rotation.jpg

BartS
Posts: 7
Joined: Wed Sep 21, 2011 9:53 pm

Re: Rotating IES files?

Post by BartS » Fri Sep 04, 2020 1:07 am

Hi,

thanks for the answer. I tried rotating the object by 90 degrees, but i got some unexpected results.
To be more precise, I made a cube, floating in front of a wall; and emitting on the 4 sides that do not face directly towards (or away from) the wall. All sides have the same material, and obviously, I expected to see 4 identical patterns on the wall.

But that's not what I saw.

3 out of 4 (!) were the same, the 4th direction had an entirely different profile. As is somehow that side had a rotated IES emission.

But stuff got weirder. When I rotated the entire cube by 90 degrees (along an axis through the cube, perpendicular to the wall), the whole situation obviously looked identical. But now TWO sides had a different emission profile, and TWO still had the original one.

To me this is utterly confusing. It seems that an emitting surface with an IES file simply gets 'some' rotation of the IES profile; without the user having any control over it. Yet, a full control (so not just increments of 90 degrees) is essential in some situations. In the streetlight example, imagine a twisting road with multiple lightposts along that road - it should be possible to align the lamps with that road, no matter the direction (north/east/south/west) of that road.

The same applies by the way for e.g. headlights of cars; these too have a non-rotationally-symmetrical profile. Most of the light goes down slightly, not exactly straight ahead But how do you specify this, given an IES file? What is 'up or down' in that case - if the "vertical" in that case is ahead of the car?

User avatar
polygonmode
Posts: 10
Joined: Fri Sep 07, 2012 5:48 am
Location: Berlin
Contact:

Re: Rotating IES files?

Post by polygonmode » Sat Feb 20, 2021 12:19 am

Hey BartS,

i have a very similar problem with some IES-files. I am just encountering this problem again with an IES-file of a wallwasher. Either it looks to the North or to the East. There is no way, the wallwasher can be rotated in different directions. Not by 90 degrees or anything in between.

This is very sad.

Is there any idea or solution? Other users have to also face the same problem from time to time.

Please can anybody try the attached file to rotate it in any direction?

Carsten
Attachments
wallwash_60cm.zip
(2.69 KiB) Downloaded 241 times
check my site: www.polygonmode.com

User avatar
OnoSendai
Developer
Posts: 6241
Joined: Sat May 20, 2006 6:16 pm
Location: Wellington, NZ
Contact:

Re: Rotating IES files?

Post by OnoSendai » Wed Apr 07, 2021 1:38 am

This is a limitation of Indigo, which I'm just working on removing. See attached render. (Note the rotated emitting quads)

In the render below the directional basis for the IES emission is based on the surface normal, and the UV mapping. (e.g. the IES is oriented in the direction of increasing u coordinate)
Attachments
rotated_IES.jpg

User avatar
pixie
Indigo 100
Posts: 2332
Joined: Sat Dec 29, 2007 4:54 am
Location: Away from paradise
3D Software: Cinema 4D
Contact:

Re: Rotating IES files?

Post by pixie » Thu Apr 08, 2021 4:22 am

Damn sexy ies lights! xD

User avatar
polygonmode
Posts: 10
Joined: Fri Sep 07, 2012 5:48 am
Location: Berlin
Contact:

Re: Rotating IES files?

Post by polygonmode » Fri Oct 08, 2021 3:37 am

OnoSendai wrote:
Wed Apr 07, 2021 1:38 am
This is a limitation of Indigo, which I'm just working on removing. See attached render. (Note the rotated emitting quads)

In the render below the directional basis for the IES emission is based on the surface normal, and the UV mapping. (e.g. the IES is oriented in the direction of increasing u coordinate)
This is really looking promising! Do you have news on this topic??? I am struggling with these asymmetrical IES again. So help or news would be very very welcome :)
check my site: www.polygonmode.com

Post Reply
7 posts • Page 1 of 1

Who is online

Users browsing this forum: Baidu [Spider] and 9 guests