[jep]

Matt,
just so I understand, you're saying that you take the "real world scale" model and do global scale of ".1" on it to make it work in a physically accurate manner?

I'm not sure whay you mean by Meters = Centimeters... 'cause you'd have to scale by ".01"... a scale of .1 means decimeters are equal to meters, no?

In any case, I agree, we're not getting the expected resultsif we use "real world scale"

Thanks,
Jep

[abgrafx3d]

Matt/Jep,

Did you look at the sample file in the mayall/doc/samples folder that NL provided? There is a file "testDOF1.mb" and an image "DOFTest1.tga". The units of this file are meters. Looking at the attributes of the Maxwell camera in this sample file, there is a unitConversion node = 100 which means that the default units should be cm (1m = 100 cm).

Take a look at that sample file.

I'll do a test as well to verify.

[jep]

Wayne,
Did you acutally try to test this file?

NO GO. The scene opens up in meters yet it still tries to multiply everything by 100! Then switch to CM and it mults everything by 10000!!

There seems to be some serious problems in the expression driving the focal lenth. Slelct the maxwell camera and look at the channel box on the right. The center of interest is listed correctly but, look at the focal distance... it's muliplied by .01 no matter what units you're working in. Am I missing something? Listen, I ask anyone to open up the DOF example that NL supplies in the Mayall folder and render it with NO changes - my money is on NO DOF - something is still fishy about the cameras - using 1decimeter-1 meter in my modeling seems to return the closest to real world results.... sad but true.

But maybe I'm missing a key point, which is why I ask for the support of you wiz kids in this forum!

[Brett Morgan]

I dont get any DOF with the sample file, maya units is set to centimetres.\\\

Brett

[3dtrialpractice]

MAtt
its luke email me about the dof issue.
Ive also noted like the rest of you that the weirdness happens to a camera outside of cm units, even the sensitivity of the camera translate and rotat values shift severly, but noneofthe less ive gotten just amazed at how cm renders go.
Also your sphere scene was just a tad on the extream range som that has to be set into the rendered result
so i think its to be expected som oddities from settings like:
Focal Distance: 280
Shutter: 50000000000.0000
F-stop .001

but its great that maxwell lets us atempt to get these abstract settings.
great rendered on the pillars and the bust!

[jep]

weoking in CM is a must for maxwell, but what we're saying is that real world scale of models (where 1CM = 1CM) doesn't seem to return the proper DOF. I fixed the example DOF test file so it was in CM and placed a new Maxwell camera (with no unit coneversions because it was created in CM space) in the EXACT same position as their camera and NOTHING - NO DOF.

OSCAR? PABLO? can you guys please verify?

thanks

[Brett Morgan]

so if you rescale your scene (increase), i'm sure I did a little test rescaled ecerything x10 and the dof looked better, maybe I was rendering an old scene, but my clipping plane setup in the preferences is near: 0.001 and far 100000.00 and I cant see all of my model, mayube I'm on the wrong track

Brett

[sidenimjay]

dang this is all SOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO f'ing confusing

he said she said i did u did we did damn

and still not one reply from an NL rep or a mayall rep WHAT GIVES HERE PEOPLE!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

ARE WE LESS IMPORTANT, cuz we use the high end software?????????????????????????????????????????

HOW BOUT SOME ANSWERS!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

OR AT LEAST A POST SAYING hey all were are looking into this now!

come on, im goin frickin crazy here

[tom]

hey run out, sidenimjay gone mad...hahahaha

[abgrafx3d]

Wayne,
Did you acutally try to test this file?

jep - I went back and looked at the file. I'm having the same problems you are. So I agree. I saw your post in the "Bug Reports" - definitley something wrong here with DOF!

[jep]

Brett, you scaled UP X10?
I think we were all saying that when we modeled to scale (i.e. 1 CM = 1CM) that we weren't noticing predictable depth of field - But, by scaling the scene DOWN by .1 it was showing up exactly where the near and far focal planes suggested it would. I'm not saying it's right, I'm just observing that it's predictable.

[iker]

I don't know if this would help you

http://rapidshare.de/files/881972/DOF_1 ... st.mb.html

[jep]

iker - I can't grab that file right now (at work )

Can you describe what you find please?

[iker]

Well I just found that you just can work in cms to get DOF, if you work in meters DOF doesn't work. So I think it's 1 Maya cm=1 Maxwell unit

The way I'm working is scale down all that is in real scale to fit with cms.

So I'm scaling 1m to 1 cm to get 1 Maxwell unit

[jep]

Iker,
Thanks for responding so quickly!
Please let me make sure I understand you. You're saying not only should we work in CM (agreed), but also that 1CM = 1 maxwell meter? So if we have a model built to scale, we need to scale it down by .01 to get the physically accurate DOF?

Have you tried just scaling it by .1?

Thanks for explaining!