OTOY Forums

I'm having some difficulty understanding how to accurately control some attributes in Octane. So far, absorption/scattering scale and aperture. The documentation doesn't go into any detail.

In Octane, aperture is controlled by a slider that goes form 0-100. They don't seem to be real-world units because, opposite of a real camera, raising the aperture number in Octane narrows the DoF. So what are these numbers based on? How do I accurately match background plates that are shot a certain f-stop without eyeballing it every time?

The absorption/scattering scale is also on a seemingly arbitrary range. I'm used to setting scattering depth in real-wrold units. How does this relate to the Octane control? There was a post a while back when medium nodes were introduced that said this:

The unit for absorption and scattering is m^-1 (or 1/m). If you have a measured value in cm^-1 you can convert it to m^-1 by
multiplying it by 100. For example: suppose you want to enter 4.5 cm^-1. This equals 450 m^-1. You can enter this value by
setting the scale to 1000 and the absorption texture to 0.45.

I honestly don't know what to do with that information.

Hi John

In Octane, aperture is controlled by a slider that goes form 0-100. They don't seem to be real-world units because, opposite of a real camera, raising the aperture number in Octane narrows the DoF. So what are these numbers based on? How do I accurately match background plates that are shot a certain f-stop without eyeballing it every time?

Aperture is defined in Octane Standalone (Script->Open Editor->"?" Browse API->Items Tab->NT_CAM_THINLENS->P_APERTURE) as...."The radius of the lens opening [cm]."

The last time I used a 'real' camera, opening the aperture increased the DOF effect - which is the same as what is happening in Octane.

I think you want the "relative" aperture (based on the fstop), but Octane's aperture is 'linear' (ie. a measured amount). See http://www.tomshardware.com/forum/33195 ... rture-stop for more info. I'm sure other's in this forum will have a more exact answer.

The absorption/scattering scale is also on a seemingly arbitrary range. I'm used to setting scattering depth in real-wrold units. How does this relate to the Octane control? There was a post a while back when medium nodes were introduced that said this:

I haven't yet looked at the conversion of Modo absortption/scatter to the Octane equivalent - however I am sure they are different. Scatter is more complex because you have the absorption offsetting the scatter to an extent, and then that "amount" is multiplied by the "Scale".

To best match the Octane Scatter to the Modo Scatter, set the Octane Transmission color to the Modo Subsurface Color, the Modo Front Weighting = Octane Phase (50% front = phase 0), set the Absorption to black (so 0), set the Scatter to white (1), then the Modo Scatter distance will be some proportion to the Octane Scatter Scale.

I went through the above process to find the conversion formula from Modo Scatter Distance to Octane Scale, and it is not so easy - since I couldn't get exact enough light intensity matches between the Modo and Octane with emitters, and I couldn't get Modo Scattering to work with HDR lighting. But you can see what I got below....

Hope that helps. It's early days for the Modo plugin - so I'm sure there will be answers to these given more time....

Paul

face_off wrote:Aperture is defined in Octane Standalone (Script->Open Editor->"?" Browse API->Items Tab->NT_CAM_THINLENS->P_APERTURE) as...."The radius of the lens opening [cm]."

The last time I used a 'real' camera, opening the aperture increased the DOF effect - which is the same as what is happening in Octane.

I think you want the "relative" aperture (based on the fstop), but Octane's aperture is 'linear' (ie. a measured amount). See http://www.tomshardware.com/forum/33195 ... rture-stop for more info. I'm sure other's in this forum will have a more exact answer.

Ah-ha~ it's the iris diameter. At some point it might be helpful to have that in the documentation. Although, that could be directed to OTOY as much as anyone. Actually, my whole post could have been directed toward them, I just thought I'd start in the MODO plugin forum since that's how I'm using Octane. So, the next two questions are: 1) If I'm working in English (feet/inch) units, is my aperture setting being read in cm or inches?, and 2) Is there a way to include the extra calculation (aperture / focal length) to get the f-stop units that people are used to having both in photography and in other rendering packages? Maybe that's best directed at OTOY core team?

To best match the Octane Scatter to the Modo Scatter, set the Octane Transmission color to the Modo Subsurface Color, the Modo Front Weighting = Octane Phase (50% front = phase 0), set the Absorption to black (so 0), set the Scatter to white (1), then the Modo Scatter distance will be some proportion to the Octane Scatter Scale.

Very helpful, thank you~

I went through the above process to find the conversion formula from Modo Scatter Distance to Octane Scale, and it is not so easy - since I couldn't get exact enough light intensity matches between the Modo and Octane with emitters, and I couldn't get Modo Scattering to work with HDR lighting.

You probably already tried this, but in the GI render settings, Indirect Illumination needs to be changed from the default of Direct. Indirect Affects SSS or Both will take into account Environment lights.

So, the next two questions are: 1) If I'm working in English (feet/inch) units, is my aperture setting being read in cm or inches?, and 2) Is there a way to include the extra calculation (aperture / focal length) to get the f-stop units that people are used to having both in photography and in other rendering packages?

1) It will always be cm.
2) In theory the plugin could set the aperture according to the Modo f-stop. Happy to look at this - although I couldn't get Modo to render high level's of DOF like Octane does - so could work out the exact formula. But a Modo f-stop of 1.4 seemed around about an Octane aperture of 4 or 5. If you think you have the calc - let me know, and I'll add a property to the Octane camera "Calculate aperture from Modo f-stop", which will override the Octane aperture value.

You probably already tried this, but in the GI render settings, Indirect Illumination needs to be changed from the default of Direct. Indirect Affects SSS or Both will take into account Environment lights.

I still can't really make this work for me (see below). Ideally we want a scene with IBL lighting set to 100% power, with an object with scattering which can be used as a visual guide to determine the Modo scatter distance to Octane scatter scale factor.

Paul

face_off wrote:2) In theory the plugin could set the aperture according to the Modo f-stop. Happy to look at this - although I couldn't get Modo to render high level's of DOF like Octane does - so could work out the exact formula. But a Modo f-stop of 1.4 seemed around about an Octane aperture of 4 or 5. If you think you have the calc - let me know, and I'll add a property to the Octane camera "Calculate aperture from Modo f-stop", which will override the Octane aperture value.

The formula for determining aperture diameter is: D=ƒ/N (where ƒ is the focal length and N is the f-number). So an 85mm lens at f1.4 would have a 60.71mm aperture diameter: D=85/1.4=60.71

Since Octane's P_APERTURE pin controls the radius (not diameter) and is measured in cm, the formula should be D=((ƒ/N)/2)/10. This is assuming that ƒ (focal length) is received in mm units. So, for the above example, that same lens would get an Octane aperture of ((85/1.4)/2)/10=3.03.

Here's how they look from MODO and Octane with those settings in a quick b&w test:



The back wall has a grid texture and MODO is obviously sampling it differently, making it look sharper than it should. But if you look at the way the floor lines blur and fan out as they get farther away from the camera, especially the ones to the right of the sphere, they match. As does the blurred edge of the background sphere.

Edit: I should note that my focus point is not exactly the same in both, since Octane doesn't appear to use the focus distance from the MODO camera. So I focus-picked it by hand as close as I could.

Edit 2:Swapped the variables in the formula to solve for the diameter, not the f-number, to match the correct example. For the record, to solve for aperture diameter: D=ƒ/N and to solve for f-number: N=ƒ/D

Great work John. I'll see if this can be implemented.

Edit: I should note that my focus point is not exactly the same in both, since Octane doesn't appear to use the focus distance from the MODO camera. So I focus-picked it by hand as close as I could.

You can manually type in the focus distance in the Octane Camera settings - maybe this needs a "Use Modo Focus Distance" setting too - which disables if you focus pick in the Octane Viewport.

Paul

face_off wrote:Great work John. I'll see if this can be implemented.

Edit: I should note that my focus point is not exactly the same in both, since Octane doesn't appear to use the focus distance from the MODO camera. So I focus-picked it by hand as close as I could.

You can manually type in the focus distance in the Octane Camera settings - maybe this needs a "Use Modo Focus Distance" setting too - which disables if you focus pick in the Octane Viewport.

Paul

Fantastic~ Thanks for looking into these Paul. That focus distance override would be a great addition, too.

Good thread. Using the Modo focal distance as well as f-number values would be a very welcome addition for me.

Martin - the above change is coded in the next test release - if you want a sneak peak, PM me and I'll send you a link.

Paul

Wow nice ! Love this, wich release is this in?