Daylight Study and Final Gather

[CJI]

Hi Guys,

 

I have been doing a very quick daylight study using the daylight system and final gather. Im so frustrated that precalculation of the final gather map doesn't work with a static camera. Luckily this is just a very light massing model with little sophistication of materials and modeling so that the map can just be calculated at render time for each image. But this method would literally take weeks on something of higher sophistication.

 

I was wondering how others approach this problem. And as a footnote when (if ever) is mental images/autodesk going to add a method for final gather to calculate each sample over time (or something to that effect). wouldn't it be nice if they said "tomorrow...and it's free" but then christmas has already past so i guess that world doesn't exist. Autodesk...Free???...hahaha.

 

Ok, rant over...i think i need a holiday already lol.

[pfbreton]

In 3ds max 2010, we have added the necessary tools to let you bake FG maps over a series of frames, in separate files (001.fgm, 002.fgm etc.)

 

Have you looked at this? Maybe you are running an earlier version?

[CJI]

Ah well, now see thats why upgrading can be good. I am running 2009 and that would be why what im trying to do doesn't work on my version.

 

I feel bad about my rant now. Thanks for letting me know that. I will go rant at my boss now for an upgrade. haha.

 

Anyone fancy a piece of humble pie... :-D

[CJI]

Hi Brian,

 

I should have been a little more specific. When i meant daylight analysis i meant a daylight cycle to assess solar shading. The camera is static, but the light moves. Thats what causes the issue in final gather in max 2009. As Pierre said, this has apparently been implemented now, i just have the wrong version.

 

I wonder if simply moving the camera by a tiny amount, say 5mm, would make a difference and allow the precalc to work without really effecting the viewpoint?

[pfbreton]

so long as ONLY the cam

is moving you can easily get good results for Daylight Analysis

 

The problem is that at every frame, the sun is probably animated, resulting in different lighting conditions per frame. This cannot be acheived without flickering in 2009. You need one FG solution per frame and interpolate across them during render time to eliminate flickering in that case.

[CJI]

Yeah i know 2010 was a good update, unfortunately my work just missed the update so it's unlikely we will be updating any time soon. bit gutted but i suppopse i will have to muddle through.

 

Anyone fancy donating to the I need an update cause? Its called Uddie!...Updated! lol

 

Cheers

 

Curtis

[CJI]

No No, something i would have to do on my own but even with that it wouldn't be possible to pre calc the final gather for this type of animation in max 2009. Render farm could handle all the rendering though and calc each frame as it goes which would be great! Luckily i do have access to a few machines so if things did get intense i can spread the load. Its a small light model though so render time including calc is still 8 seconds at most so im not complaining. Until i'm asked to do a full model with textures, then i will complain. ALOT!