Announcement

Collapse
No announcement yet.

Light Cache - difference between Flythrough mode and Single Frame / Use Camera Path

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

  • Light Cache - difference between Flythrough mode and Single Frame / Use Camera Path

    I was having errors with flythrough mode to precalc a Light Cache. I'm seeing some posts about adaptive lights issues -- so maybe it's related.

    Instead of using fly through, I tried precalcing the light cache using Single Frame mode and Use Camera Path, since I could've sworn I read something about this on the forums a long time ago. I've rendered half the animation so far, and it seems to be working. Which begs the question?

    What is the difference between Flythrough Mode and Single Frame with Use Camera Path. They sound very similar when reading the docs. Is Single Frame/Use Camera Path a legitimate way to precalc? Is there a problem with doing it this way?


  • #2
    You can check VRay help, it is well described there : https://docs.chaosgroup.com/display/...Cache+Settings
    Available for remote work.
    My LinkedIn: https://www.linkedin.com/in/olegbudeanu/

    Comment


    • #3
      I read the docs as I stated in my post, and no, it is not well-described there. It makes Use Camera Path sound exactly like Fly Through. But Is it exactly the same or similar? It doesn't make sense for it to be an option, if it's exactly the same so there must be at least a subtle difference which is not explained in the docs.

      Can it be used in place of Fly through in most cases? Is there a disadvantage to doing that?

      My questions are very specific and the docs do not answer them.

      Comment


      • #4
        When you choose "Fly-through" mode and you render a sequence, the light cache will be computed just once for the first frame, and then for the rest of the frames in the same render, it will reuse the cache in memory. This means that the only thing that can be animated is the camera.

        When you turn on the "Use camera path" option, the light cache will still be calculated for each frame, but it will be more consistent from frame to frame which would potentially reduce flickering. In this mode you can have moving camera or objects - it doesn't matter.

        Since we added the "retrace" option in recent builds though, it makes less sense to use any of these options. Most of the time, you can just leave the defaults and only increase the light cache subdivs if needed. The speed-up you get from the adaptive lights is usually more than enough to compensate the recalculation of the light cache for each frame.

        Best regards,
        Vlado
        I only act like I know everything, Rogers.

        Comment


        • #5
          Thank you Vlado... I appreciate your input but I still am a bit confused.


          Since we added the "retrace" option in recent builds though, it makes less sense to use any of these options.
          Are you saying there is no real advantage to Precalculating a light cache now ?

          I used Single Frame with Use Camera Path to precalc and it seems like it worked quite well (for an animation with only a moving camera). I'm not sure how it would have differed from Flythrough -- but I was getting those dark spot errors so that's why I tried this method.

          Lastly, why would you NOT turn on "Use Camera Path" -- if it reduces flickering and creates a more consistent look, why ever turn it off?



          Comment


          • #6
            Vlado correct me if I am wrong.
            If you use the use camera path, your single frame light cache samples will be spread over the entire area that the camera path covers.
            So for example if your camera has a long path to travel and goes to many parts of the scene, your samples are very spread out over a large area, that may not even be seen in the frame you are actually rendering.
            To get the equivalent samples in a specific area you would need to increase your light cache subdivs.
            Flythrough will have much higher concentrations of samples in the areas where each precalculated frame was.
            Gavin Jeoffreys
            Freelance 3D Generalist

            Comment


            • #7
              Originally posted by davision View Post
              Thank you Vlado... I appreciate your input but I still am a bit confused.




              Are you saying there is no real advantage to Precalculating a light cache now ?

              I used Single Frame with Use Camera Path to precalc and it seems like it worked quite well (for an animation with only a moving camera). I'm not sure how it would have differed from Flythrough -- but I was getting those dark spot errors so that's why I tried this method.

              Lastly, why would you NOT turn on "Use Camera Path" -- if it reduces flickering and creates a more consistent look, why ever turn it off?


              I think the difference is that you can have moving objects with "camera path" option, though you need much more subdivisions.
              With FlyThrough you can get away with fewer subdivisions because the only thing moving is the camera.
              And as usual - it depends on what is happening in your scene.
              I recently finished 90 sec animation with moving camera only. Camera path gave me slightly worse result than flythrough with the same amount of subdivs being used.
              Available for remote work.
              My LinkedIn: https://www.linkedin.com/in/olegbudeanu/

              Comment


              • #8
                Originally posted by Raven View Post
                If you use the use camera path, your single frame light cache samples will be spread over the entire area that the camera path covers. So for example if your camera has a long path to travel and goes to many parts of the scene, your samples are very spread out over a large area, that may not even be seen in the frame you are actually rendering. To get the equivalent samples in a specific area you would need to increase your light cache subdivs.
                This is exactly right. It is also the exact same description for the fly-though mode. Only difference is that one is calculated per frame to allow for moving objects (if you have any), and the other one is calculated just once and assumes the camera is the only thing that moves.

                Flythrough will have much higher concentrations of samples in the areas where each precalculated frame was.
                There is no difference in the samples distribution between the two modes.

                Like I said, V-Ray 3.6 will bring some improvements to both of these modes, but in my tests, I could get equally good results with just regular single-frame light cache.

                Best regards,
                Vlado

                I only act like I know everything, Rogers.

                Comment


                • #9
                  I am doing an exterior flyaround a building and using the old style I have always used (pre calc and save light cache using flythrough and pre calc irradiance with multiframe incremental) I get those dark squares everywhere. So I have tried using just light cache and brute force for each frame but I get flickering materials here and there. I have played with settings (use camera path on and off, screen or world scale, tried ir or brute force for primary) but I am still getting flickering. Any suggestions? I am embarrassed to ask this but can you maybe walk me through how I should setup an exterior animation with 3.5? Is there a new tutorial? I'm still trying to figure it out on my own but I'm stressing for time now...
                  Last edited by stevirtvr; 09-06-2017, 12:32 PM.
                  You may as well be doing something while you're doing nothing.

                  Comment

                  Working...
                  X