Sep 11

There are many ways of using a Vray physical camera, but it is sometimes confusing which settings to use especially with f stop, shutter speed and ISO and what they do. Below I have attempted to clarify what each does and some basic values for each after initially setting up a datum point for your lighting, which always helps in keeping your exposure settings fairly consistent between shots.

1. SET A DATUM POINT FOR YOUR LIGHTS

Having a datum point for lighting is always a good idea, especially when sharing files as lighting setups can then be shared from one file to another. So you could either set your HDRI for your GI environment to a multiplier of 1 (This should make the sun around 0.015 for it’s multiplier) and calibrate everything to that or our preferred method which is to set the vray sun to a multiplier of 1. We’ve found the GI spread to be more robust with the sun set to 1, and most of the Vray camera settings come close to real world figures which does help get ball park figures matching things like your background plates, if you have them.

Of course the only problem with changing the sun multiplier to 1, is it changes all your (GI & Reflection) maps to about multiplier 30, i.e. they just appear white within the material browser, and you can’t render through the viewports; only through a vray camera.

2. VRAY CAMERA SETUP

So a solution is to change your exposure through your vray camera using the following outlined settings.

First of all identify what you want to do through your Vray camera, and this will affect how to put in these values highlighted in orange above. I’ve made a table below that might help adjust these values a bit more logically depending on what you want the Vray camera to do on the right.

What do you want through the Vray Camera?
F stop
shutter speed
ISO
Exposure only
ADJUST to correct exposure
fix at 1 (keep same as ISO)
fix at 1 (keep same as shutter speed)
Exposure & Motion Blur
ADJUST to correct exposure
fix at desired motion blur speed, i.e. 50= 1/50sec
keep same as shutter speed
Exposure & Motion Blur & 3D Depth of Field
fix at desired depth, i.e. 2.6 is a lot of depth blur and 22 or higher not very much
fix at desired motion blur speed, i.e. 50= 1/50sec
ADJUST to correct exposure

.

These values don’t always relate to real world camera settings of course but it helps to at least place them in these real world ranges especially with f stop ranges that in real life range from about 2.6-36 depending on your lens. Motion blur also mimics reality with 1/50 (or 50) giving a fairly decent motion blur just like you’d expect with a real camera.

Rendering times also increase with motion blur and 3D depth of field depending on the number of subdivs as highlighted in orange above. The lower number you can get away with the better for rendering times, but the blur will become more noisy.

ALTERNATIVES

There are a few notes in relation to the above, many of which depends what you do in post-production. Because motion blur and 3D depth blur increase render times, it can become more desirable to do these in post instead to save time. In these cases you would use the f stop for exposure only. The only thing I haven’t listed below is vignetting which of course can be done quite easily in post-production so can be left off for that stage.

MOTION BLUR

So the 1st option is for motion blur. Through most compositing programs, like Fusion and Nuke, they will have a 2D motion blur which is average at best although there is a great plugin called realsmart motion blur which does a great job for things like panning shots etc. But this might not work on faster moving objects within scenes, so for this you can output a separate render element called a velocity pass that can easily be used in compositing programs to blur some pixels more than others depending on the rendering element’s velocity, which it colours differently through the RGB channels.

Remember also motion blur won’t work through the usual camera settings as in the dialog below. This can only be done through motion blur on the vray camera or the options above.

3D DEPTH BLUR

Also under your render elements you can output is something called z-depth. Measure your scene’s maximum depth and put this in the max field under the render elements dialog when z-depth is selected. This will render you a black and white mask where white is foreground and black is background. You can then use this through compositing programs to blur the black bits, but not the white bits and of course varying degrees in between. Hey presto, cheap depth blur.

The only problem is extreme foreground elements where you will find it’ll blur these edges too much and you get a soft edge around these foreground objects. This is because it should be blurring what is behind the object, but obviously this won’t work because it hasn’t got this information. A solution to this is rendering extreme foreground objects separately and blurring them as a separate pass in post-production, or if you can spare the render time do it in 3D.

I hope the above helps, any questions let me know below and I’ll try and answer.

Post to Twitter Post to Delicious Post to Digg Post to Facebook Post to LinkedIn