Announcement

Collapse
No announcement yet.

Physical Camera 3.3 General Questions

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Physical Camera 3.3 General Questions

    Two general questions/comments:

    1) Will the Physical camera add guess Horizontal tilt? It appears to be cut from Autodesk's release

    2) The physical camera defaults to drastically different exposure values than the original VRay physical camera or any real camera:

    Physical Camera:
    F-8
    SS 1000.0
    ISO 102465.0

    Previous VRay Physical Camera:
    F-8
    SS 200
    ISO 100

  • #2
    Since now Physical Camera is a part of 3DSMax, Autodesk should provide a better assistance on such requests.
    Svetlozar Draganov | Senior Manager 3D Support | contact us
    Chaos & Enscape & Cylindo are now one!

    Comment


    • #3
      Originally posted by svetlozar.draganov View Post
      Since now Physical Camera is a part of 3DSMax, Autodesk should provide a better assistance on such requests.
      Hahahaha, Autodesk provide assistance? good one.

      Comment


      • #4
        This was the point of my post chadstevens. Chaos group handed Autodesk a really good camera system and Autodesk mangled it with unnecessary tabs and removed features.
        I would love to see the clean layout of the original camera back in Chaos group's hands since this will be never be updated again. (I know you can access this via script)
        I am trying not to be negative but Autodesk after almost 5+ years of shelving Max they seemed to have finally showed some sort of real development in 2016.

        I'll will post this on the Autodesk forums and update if I get any results.

        Comment


        • #5
          Update. Autodesk has created a support request about this. I will update if they send feedback.

          After some fiddling with the new physical camera settings it appears by default the physical camera exposure is overexposed compared to the original Vray Physical Camera (using the same camera settings).
          The only way to get the exposure to match you need to change the Environment exposure control Physical Scale to 750 candelas / RGB.

          Does Chaos group have a recommended workflow for this since we will need to either use a different physical scale or accept the difference and adjust exposure control to suite.

          I am mainly concerned since we regularly match photos to real values.

          Comment


          • #6
            Glad to hear that you havfe found a way to match both cameras, I have tried it here but I still got a different result between both(Phys.Camera is slightly brighter).
            Does that setup works always in your environment? If it is we might considered it as a useful workflow to match the exposure between both cameras.
            Svetlozar Draganov | Senior Manager 3D Support | contact us
            Chaos & Enscape & Cylindo are now one!

            Comment


            • #7
              Thanks Svetlozar,
              Actually I tried this again and it is a bit confusing what is happening. Each time I tried to replicate the values they did not seem to match.
              So no it did not work in different environments. Is it buggy or am I doing this wrong?

              Comment


              • #8
                Update:

                Autodesk has confirmed they are using there own formula for Max's physical camera exposure. So this means Max's Physical camera and Pre 3.3 Vray Physical cameras will never match exposure with the same settings.

                I presume will need to accept the brighter exposure and adjust to work with those values going forward?

                I still do not get why the ISO and Shutter Speed default values are really high numbers but they assured me this is normal.

                Here is the thread: http://forums.autodesk.com/t5/3ds-ma...ra/m-p/6001868

                Comment


                • #9
                  Originally posted by jneilon View Post
                  Autodesk has confirmed they are using there own formula for Max's physical camera exposure. So this means Max's Physical camera and Pre 3.3 Vray Physical cameras will never match exposure with the same settings.

                  I presume will need to accept the brighter exposure and adjust to work with those values going forward?
                  I ran a quick test sampling different f-stop values with an old vray camera and the new Phys. Cam. and the results are exactly linear, see below. I sampled the float values for each camera and just did a simple division between the terms.

                  f-stop VrayCam PhysCam b/a a/b
                  16 0.03 0.039 1.3 0.769231
                  8 0.119 0.155 1.302521 0.767742
                  5.6 0.244 0.317 1.29918 0.769716
                  4 0.477 0.62 1.29979 0.769355
                  2.8 0.974 1.268 1.301848 0.768139
                  2 1.9 2.488 1.309474 0.763666
                  1.4 3.89 5.073 1.304113 0.766805
                  Average: 1.302418 0.767808
                  All that's needed to match the Phys. cam to the old cam is a simple multiplier for either the light values or even easier the camera ISO. The Physical camera overexposes by 30.5 percent so to get a perfect match (within a few decimal places) just divide the ISO by 1.302 (or multiply it by .768 ). Let me know if I've overlooked something.
                  Last edited by dlparisi; 22-01-2016, 02:33 PM.
                  www.dpict3d.com - "That's a very nice rendering, Dave. I think you've improved a great deal." - HAL9000... At least I have one fan.

                  Comment


                  • #10
                    Originally posted by dlparisi View Post
                    All that's needed to match the Phys. cam to the old cam is a simple multiplier for either the light values or even easier the camera ISO. The Physical camera overexposes by 30.5 percent so to get a perfect match (within a few decimal places) just divide the ISO by 1.302 (or multiply it by .768 ). Let me know if I've overlooked something.
                    Thanks, I just ran the same test and thats correct it is roughly 1.3 times brighter.
                    Last edited by jneilon; 22-01-2016, 03:29 PM.

                    Comment

                    Working...
                    X