Zdepth Pass Issue

Hi Guys,

I have been using Fusion for a while and had no problem using Z-depth pass coming from Maya/Arnold and piping it up in Fusion to get the accurate result. However, It's been two days that I can't achieve the same result anymore.

I have searched the web for any possible solution ( and there are not many ) to see maybe I have missed something. But no luck!

I have included a snapshot of the result and both Beauty and Z pass EXR in the dropbox below so you might want to try as well.

( I can't share the Dropbox URL) So I have just included the screen capture

<a href="https://ibb.co/mMn33k"><img src="https://thumb.ibb.co/mMn33k/Capture.jpg" alt="Capture" border="0"></a>
The strange part is, when I open old fusion files where they were working fine, they do look strange too!!!

Appreciate the help.

The image is so small I can't see what I'm looking at.

The image is so small I can't see what I'm looking at.

My Apologies, Please visit the dropbox link for screenshot and EXR file.
https://www.dropbox.com/sh/pqk3ovtyytq1 ... M_9La?dl=0

Most likely, your z depth pass is not making it to the depth blur, perhaps due to an error in the Boolean setup or within the channels of your EXR.

The check to do is route your depth blur to a viewer and drag the PICK from depth-blurs control panel over the image. You should see the Z depth number change in relation to expected depth. If not, it's because the depth channel connected to the blur does not contain what you think it does.

Hard to be more specific as I don't know what is in the two "untitled" exr files in your set up. Are they both the same file containing both RGB and Z depth? Or are they two different files one with RGB and the other with a depth map? If so, is the depth map contained in a dedicated EXR depth channel or in the RGB data?

Your current set up is for RGB pictures coming from the upper "untitled.exr" file; the Z depth contained within the RGB data of the lower "untitled.exr" file. Have a poke around and you'll probably find the Z Depth is either not routed correctly or not rendered correctly.

**edit. Don't use "lightness" in your boolean routing. If you know your z depth is in the RGB channels, then select one of the R, G or B channels directly.

DC

Most likely, your z depth pass is not making it to the depth blur, perhaps due to an error in the Boolean setup or within the channels of your EXR.

The check to do is route your depth blur to a viewer and drag the PICK from depth-blurs control panel over the image. You should see the Z depth number change in relation to expected depth. If not, it's because the depth channel connected to the blur does not contain what you think it does.

Hard to be more specific as I don't know what is in the two "untitled" exr files in your set up. Are they both the same file containing both RGB and Z depth? Or are they two different files one with RGB and the other with a depth map? If so, is the depth map contained in a dedicated EXR depth channel or in the RGB data?

Your current set up is for RGB pictures coming from the upper "untitled.exr" file; the Z depth contained within the RGB data of the lower "untitled.exr" file. Have a poke around and you'll probably find the Z Depth is either not routed correctly or not rendered correctly.

**edit. Don't use "lightness" in your boolean routing. If you know your z depth is in the RGB channels, then select one of the R, G or B channels directly.

DC

Thanks for the help. However, Everything is piped correctly and I see the Z values. As I mentioned earlier, something has been wrong in Fusion, since the older files are also not functioning correctly.

Depth blur is working in Fusion 9 / 9.01 so perhaps something else has changed. I have a very similar set up to yours working.
DC

I can't test the images right now, but the result has the looks symptomatic to RGB being blurred but not the alpha. This gives a halo around the sharp original shape.

Maybe the problem is reverse.

Depth blur is blurring the alpha but the RGB values of the edge pixels are not being multiplied by the blurred alpha.

Try an "Alpha Multiply" node after the depth blur and see what happens.

Might be worth raising this with support. I downloaded your images and also had the difficulties you encountered. I also tried rendering out a similar scene from blender and got to the same place. It seems that the depth blur is always adding some form of an overlaid blur, even at the point of focus. Any amount of combining with alpha doesn't provide a solution. I haven't seen this before but it might well be hardly noticeable where small blur values are used and not on objects with alpha.
DC

I think I got a better result with this setup:

Code: Select all

{
   Tools = ordered() {
      Loader1 = Loader {
         Clips = {
            Clip {
               ID = "Clip1",
               Filename = "E:\\temp\\Fusion\\Z-issue\\Beauty.exr",
               FormatID = "OpenEXRFormat",
               StartFrame = -1,
               LengthSetManually = true,
               TrimIn = 0,
               TrimOut = 0,
               ExtendFirst = 0,
               ExtendLast = 0,
               Loop = 0,
               AspectMode = 0,
               Depth = 0,
               TimeCode = 0,
               GlobalStart = 0,
               GlobalEnd = 0
            }
         },
         Inputs = {
            ["Gamut.SLogVersion"] = Input { Value = FuID { "SLog2" }, },
            ["Clip1.OpenEXRFormat.RedName"] = Input { Value = FuID { "R" }, },
            ["Clip1.OpenEXRFormat.GreenName"] = Input { Value = FuID { "G" }, },
            ["Clip1.OpenEXRFormat.BlueName"] = Input { Value = FuID { "B" }, },
            ["Clip1.OpenEXRFormat.AlphaName"] = Input { Value = FuID { "A" }, },
         },
         ViewInfo = OperatorInfo { Pos = { 606.475, 44.3024 } },
      },
      Loader2 = Loader {
         Clips = {
            Clip {
               ID = "Clip1",
               Filename = "E:\\temp\\Fusion\\Z-issue\\Z.exr",
               FormatID = "OpenEXRFormat",
               StartFrame = -1,
               LengthSetManually = true,
               TrimIn = 0,
               TrimOut = 0,
               ExtendFirst = 0,
               ExtendLast = 0,
               Loop = 0,
               AspectMode = 0,
               Depth = 0,
               TimeCode = 0,
               GlobalStart = 0,
               GlobalEnd = 0
            }
         },
         Inputs = {
            ["Gamut.SLogVersion"] = Input { Value = FuID { "SLog2" }, },
            ["Clip1.OpenEXRFormat.Channels"] = Input { Value = 1, },
            ["Clip1.OpenEXRFormat.RedName"] = Input { Value = FuID { "Y" }, },
            ["Clip1.OpenEXRFormat.GreenName"] = Input { Value = FuID { "Y" }, },
            ["Clip1.OpenEXRFormat.BlueName"] = Input { Value = FuID { "Y" }, },
         },
         ViewInfo = OperatorInfo { Pos = { 422.518, 131.179 } },
      },
      ChannelBooleans1 = ChannelBoolean {
         CtrlWZoom = false,
         Inputs = {
            ToRed = Input { Value = 5, },
            ToGreen = Input { Value = 6, },
            ToBlue = Input { Value = 7, },
            ToAlpha = Input { Value = 8, },
            EnableExtraChannels = Input { Value = 1, },
            ToZBuffer = Input { Value = 10, },
            Background = Input {
               SourceOp = "Loader1",
               Source = "Output",
            },
            Foreground = Input {
               SourceOp = "Loader2",
               Source = "Output",
            },
         },
         ViewInfo = OperatorInfo { Pos = { 630.839, 136.531 } },
      },
      AlphaDivide1 = AlphaDivide {
         Inputs = {
            Input = Input {
               SourceOp = "ChannelBooleans1",
               Source = "Output",
            },
         },
         ViewInfo = OperatorInfo { Pos = { 740.839, 136.531 } },
      },
      AlphaMultiply1 = AlphaMultiply {
         Inputs = {
            Input = Input {
               SourceOp = "DepthBlur1",
               Source = "Output",
            },
         },
         ViewInfo = OperatorInfo { Pos = { 841.105, 198.346 } },
      },
      DepthBlur1 = DepthBlur {
         Inputs = {
            Filter = Input { Value = 2, },
            XBlurSize = Input { Value = 10, },
            FocalPoint = Input { Value = 131, },
            ZScale = Input { Value = 47, },
            Input = Input {
               SourceOp = "AlphaDivide1",
               Source = "Output",
            },
         },
         ViewInfo = OperatorInfo { Pos = { 850.839, 140.765 } },
      },
      Background1 = Background {
         Inputs = {
            Width = Input { Value = 960, },
            Height = Input { Value = 540, },
            Depth = Input { Value = 4, },
            ["Gamut.SLogVersion"] = Input { Value = FuID { "SLog2" }, },
         },
         ViewInfo = OperatorInfo { Pos = { 690, 279 } },
      },
      Merge1 = Merge {
         Inputs = {
            Background = Input {
               SourceOp = "Background1",
               Source = "Output",
            },
            Foreground = Input {
               SourceOp = "AlphaMultiply1",
               Source = "Output",
            },
            PerformDepthMerge = Input { Value = 0, },
         },
         ViewInfo = OperatorInfo { Pos = { 838.87, 280.117 } },
      }
   }
}

composite.jpg (51.66 KiB) Viewed 4877 times

Clipboard01.jpg (16.76 KiB) Viewed 4877 times

Yes I was trying something similar. But your closest sphere still has a halo around it and it is this halo that always seems to exist no matter where the point of focus is. I tested a scene which didn't have an alpha channel (i.e. the full frame was filled with a CGI scene) and that base halo still seemed to exist.

DC

It seems to be a some kind of general depth slicing based z defocus artefact, which is exaggerated by dark background. I get similar halos in Nuke and with some fiddling couldn't help it much. When image BG is brighter (above 18% gray), the sharp edge dissapears and result looks much nicer.

Why the sharp edge appears, don't know. It shouldn't be there, so I guess it is some kind of side effect of depth slicing because it is more pronounced on some spheres and is not related to the distance of the sphere.

Thank you guys for giving this a try!

yet I haven't found a good reason as for why this suddenly happening.

I assume this is happening in Maya/Arnold.

Im checking my previous Z renders and they all look 0 to 1 or 1.5 but now in the sample, I share with you I'm seeing 0 to 9999.999999999

This is very frustrating!