Your Depth Pass is Wrong.

May 31, 2016 - By 

Lil’ Coffee Breakdown: All Motion Blur and Depth of Field was added in After Effects.

I realize that the title here is a bit bold, but in my case (and possibly) yours, this is very true. I ran into a fellow artist who at an event who pleaded with me to cover this issue. This artist did not have a giant render farm, nor did he have a multi-GPU setup using Octane. He simply wanted to know the correct workflow for doing Depth of Field and Motion Blur in After Effects. A method that didn’t result in artifacts or other anomalies. So, as I promised, here it is.

“A WHOLE tutorial about the Depth Pass?”

Yes and no. This tutorial is about saving a TON of render time by NOT having to render your Depth of Field and Motion Blur in C4D’s Physical Renderer. With a couple of After Effect plugins and the correct workflow, you can save yourself potentially hours of rendering.

What you will learn:

  • How to add Motion Blur and Depth of Field to your C4D render in After Effects
  • What is needed out of a Depth Pass to achieve proper results with Frischluft Lenscare
  • What a “Position” pass or WPP is and how it can be used to generate a correct Depth pass
  • How to set up a multi-layer EXR output for After Effects
  • How to properly set up your After Effects comp with Frischluft Depth of Field and RSMB plugins (see links below)

Tools you’ll need:

Let’s Get to the Tutorial*

* So Ihab in the comments found a mistake, so thanks Ihab! Looks like I grabbed the regular RSMB plugin and not the Pro Vector version which reads your vector pass. Sorry about that folks. Grabbing this Pro Vector version and choosing your vector pass will give you more accurate motion blur results.

 


 

Posted In:  
144  comments
144 Comments
  • How about hair? I tried and the result is far from good 😀

    • In trickier situations I usually go with a double sized z pass (still without any antialiasing), scaled down to the final size in After Effects but using the lowest quality “quality and sampling” toggle so it remains un-antialiased, and playing around with the “collapse transformations” toggle until it looks right. There are also situations where motion blurring the depth pass helps.
      If that still doesn’t work two options remain: a deep comp solution (but as far as I know After Effects doesn’t support it, you’ll have to go with Nuke, Fusion, Natron…) or depth of field in-render.

      • Can I ask for more information – what is deep compositing – everybody talks about it, but what is the difference from the regular multi-pass comp?

        • Deep compositing is like the World Position Pass on steroids. I cannot even begin to fully explain it here, but I suggest going to fxguide.com and researching it there. Very cool stuff indeed.

  • Tyler Scheitlin May 31, 2016 at 10:49 am

    Rendering your utility passes (motion vector, depth, etc…) with Standard renderer makes them even faster. There is a small pixel variation that I haven’t figured out yet but once the composite is finished it’s barely noticeable.

  • Why do you composite a light grey solid in the background when you completely clamp it with the levels adjustment?

    Regardless, a very helpful tutorial!

  • Another thing, the passes would NOT be affected by the adjustment layers since plug-ins only read the layer source, excluding any effects that are made in the current comp.

    • I’m not sure I completely follow you. Can you elaborate? I must confess I don’t use After Effects as much as I used to, so you may be correct.

      • There’s no need to put your FX source layers above the adjustment layers. They can’t be read by RSMB or Depth of Field if they’re under the BG or disabled – that’d cause recursion (and probably crash AE). Turn ’em off and bottom of the stack.

        Extra points for Ctrl+Alt+Shift+Down Arrow

  • Haleluia! I know what I’m doing tonight, learnin’ stuff! Thanks Chad 8o)

  • Amazing tutorial! Thanks a lot! You can have faster .EXR in AE, just right click the sequence in the Project Window – Interpret Footage – More options.. (under “Other Options”) – Activate “Cache Channels”

  • Wow, my depth pass was so wrong. This has been one of the most helpful tutorials on this site to date.

  • Great information Chad!!

    For the motion blur, any specific reason you didn’t use RSMB Pro Vectors since you had the motion vectors pass? I’m guessing since the background is simple for the particular shot you were working on, leaving Alt Motion to none would yield the exact same result based on motion estimation of the moving pixels vs the actual RG vector data.

    On a separate note, would love to see more from you on Arnold!!

    Thanks again for the tutorial!

    • Thanks for this comment, you are absolutely right. I SHOULD have used the pro rsmb plugin and accessed the motion vector pass, after all, that’s why I rendered it 😉 I guess I was moving too quickly to think properly. Good thing these elements were simple and it worked with the regular RSMB! ha! Thank you so much for bringing this up Ihab! I added the correct workflow back to the bottom of the tutorial.

      • I was super curious to see you do this, because I thought it’s not possible to do both dof AND motion blur using passes. If you do dof first, then the motion vector pass is not going to fit your new pixel values, and vice versa. Have you tested this workflow using both passes at once?

        • Yep I sure have. There is no magic technique and every shot is different, but if you recognize what you can get away with, you can usually make it work. This particular shot was very easy for this method. Small objects falling with an alpha, no transparency, not a ton of speculars, it was a great use case for this technique. If the shot had been a bunch of falling glass marbles with an elaborate background, I may have approached it differently. In production, you often have to be able to recognize where you can cheat and where you have to bite the bullet and let the computer do it’s thing.

      • By the way: if you’re using the motion vector pass with the “rsmb Pro vectors” effect (which is designed for exactly that) – you’ll have to key out the black from the mv pass first (using a luma key).

  • Thank you Chad.

  • Awesome tutorial Chad!

    Would the outcome be different if you used the standard AE Camera Lens Blur effect? Are you able to get the same look with AE standard plugins? Not only am I against time, but budget as well, lol. Thanks!

    • You’d be able to get it pretty close, but Frischluft (pronounced Frish-Looft) is just a better plugin, so you get a much nicer blur from it.

    • I don’t have Frischluft Lenscare but got a pretty good result with Camera Lens Blur. I also left this comment on the Vimeo page.

      The scale parameter on the Position Pass seems to be a very hard number to find for most of us. I even animated it from .001 (the minimum C4D will accept) up to 20000. It still was a solid blue from beginning to end. What I eventually did was go to Edit > Scale Project and made it 1% of its original size. Then, I got some good results with 0.001 as the scale number in the Position Pass. I suspect Chad didn’t go through this torture because he said his scene was very small at the beginning of the tutorial.

  • That wpp is a very cool tip. I still cannot find a solution for accurate DOF onto reflections when using a depth pass though.
    Post DOF can be a great time saver but the fact that reflections are not taken into account by any z-pass makes that workflow a bit of a joke in my opinion.
    Nevertheless, nice tut.

    • Sean, I agree. Nothing beats the real thing, but sometimes the clock can work against you and you are forced to cheat. 😉 thanks for the comment!

    • I read on the Arnold documentation that DoF passes cannot actually take reflective surfaces into account – it’s a 3D software thing, you’re not mental!

      • It is a general 3d software thing but what makes it even worse in c4d is that the standard renderer doesn’t take reflections into account even when rendering the dof onto the RGB image!! so it is technically impossible to have correct dof onto a reflective surfaces with the standard renderer no matter which way you decide to go: straight or multipass.
        Not sure why maxon doesn’t just dump the default render engine.

  • Is there a rule of thumb to follow for finding the best scale for the Position Pass?

    Great tutorial!

    • No real rule of thumb beside needing to see a nice range in that blue channel.

    • Cinema wants you to model everything huge – the standard box is 2m wide! So use sense here – you know a car is 5m long, a toothbrush is 15cm long – so drop a cube in that is big enough to contain everything (or select all and look in Size in the Coordinates panel) to get an idea of what scale your scene is. Use the automatic scene scaler inside of Ctrl+D to reduce/expand your size if need be…

  • This was very helpful. Thanks!

  • nice tutorial thx

  • Thanks!
    Your style of tutorial is bringing another layer of good vibes to this place !!
    Because you love pushing cg render to artistic photographic “awssom”’ness”, if you don’t know this tool http://peregrinelabs.com/bokeh/ I really encourage you to visit!
    and Pixar’s Umbrella was an amazing example of using there tool https://vimeo.com/82101119.

    and actually for more inside making off and sharing of the creative process things about Umbrella :
    https://vimeo.com/user3018211/videos/sort:date

    enjoy!

    • Oh yes, I LOVE that plugin. However, I’m not a Nuke user so I admire it from afar. I wish they would release an OFX version. Thanks for the comment!

  • This is very nice. You guys should do more compositing tips/tricks!

  • Thanks!
    Does it work well with Vray Render?

    • I haven’t tried in Vray for C4D, but I can tell you that VRay for 3ds max outputs a correct Zpass and does not need this workaround.

      • I had the same problem in vray for C4D. And there is no “Consider for antialiasing” checkbox in pass settings, to correct this problem

        • Vray do an amazing job for Zdepth Pass. All you have to do is un-check Filtering, and it’ll give you Non-Antialiased edges. Works great with Lenscare.

          • Yah, most 3rd party renderers do this correctly. I hope someday Maxon fixes this issue.

  • This, Sir, is really helpful, thank you very much!!

  • Cool technique. Never thought about using the PPass for Dof.

    There is also a nice option that comes with the ProExr Plug in(which should now be part of every installation of AFX; before you could go to the fnord website and download ProExr for free) which ExtractOR is part of. If you look under File there should be an entry called “Create ProExr Layer Comps” which splits up all Multi Layer Exrs into there individual Layers and you don’t have to manually go in and set everything up yourself.

    You could also get rid of the black in the position depth pass using Umult from Red Giant

    To use the WPP pass even more in AFX there are 2 nice additional Plug Ins on Aescripts that adds features you can also find in Nuke and Fusion. (but you need to set those renders to World)
    http://aescripts.com/pixel-cloud/
    http://aescripts.com/position-matte/

    Best regards

  • Thank you soooo much Chad I never would have thought about using that technique. I JUST finished a project yesterday where I could have used this information!

    When I use the pro motion vector in AE the motion blur gets all screwed. Maybe you can do a followup on how to use motion vectors because I have never been able to make the motion blur look just right.

    I just did a test with the motion vectors with pro vectors but it looks all blocky. It looks way better with normal RSMB.

    • I did actually screw up in the video and chose the wrong RSMB. I forgot to use the Pro version which lets you actually use the vector pass (brain fart). BUT, that being said, you don’t ALWAYS need a vector pass (as I clearly displayed in the video). ha ha ha. As long as the alt motion source is clear and has an alpha, it can sometimes be better for you.

  • can we have your source files, this tutorial was amazing and we have trying to finds a fix for our fuzzy edges with DOF for ever

  • Thanks for the great tips!

    Few questions:
    – any reason to choose premultiplied over straight alpha? What are your thoughts on this?
    – if you were to use both motion vectors (not regular rsmb) and zdepth, how does After Effects handle the order? I assume you put zdepth first, but in both cases the second plugin adds to the results of the first one, but use unaffected maps. There could be consequences because, for instance, RSMB assumes all pixels to be where the motion vector map tells it, but in reality the image has been affected by DOF already.

    • You’re welcome!

      -first answer: I prefer premult, always have.

      – second answer: I’m not sure how AE handles the order. I use the method I’ve laid out in the video after many years of trial and error. It seems to work for me most of the time.

  • Cinema 4d is such a poor program. People having to jump through hoops to get the basics to work. Instead of bringing out all new features – just fix the basics Maxon, its ridiculous.

    • Not sure I fully agree with you here. Yes, this is a simple thing that should be fixed, but there are plenty of other features in C4D i WISH i had when I was using max/maya. There is no perfect solution.

  • How to say Frischluft for English speakers:

    Say “Fish”, say “Fisch” the same way.

    Add an R, Say Frisch, then add luft, pronounced looft

    Ta-da!

  • Hi nice tutorial,
    can you do this technique with images too? Then you can just render one frame of the project as an JPG etc. ?

    • The process for stills is basically the same. Instead, only render one frame of the sequence and be sure to use RSMB Pro Vectors. Make sure you choose that motion vector pass in RSMB Pro Vectors as your motion source. Otherwise, you’ll get no motion blur. If you don’t need motion blur, but only the DOF, disregard the vector pass altogether.

  • Gary David /Bouton June 1, 2016 at 10:20 am

    Um, I enjoy the tutorials a lot as an owner of v16. But increasingly, you’re obliging people who watch your tutorials to buy add-ins, or the tutorial is fairly useless. Like X-Particles, which is over $400, and now a DoF plug-in and a motion blur plugin.

    Can we have more of your terrific tutorials that use only C4D, and perhaps AE?

    Please?

    Dead broke,

    Gary

    • Gary, we appreciate your comment! We try to be as inclusive as we can here at GSG, but as our audience grows, so does the need to cover broader techniques and toolsets. We hear you though, and we will continue to keep your comments in mind.

  • I liked the tute – I’d never have thought the Position Pass existed in C4D!

    I’m well versed in the passes>Frischluft>RSMB workflow, here’s my ten cents:

    I work in Linear colorspace with 32-bit colour, so naturally I want that colour depth inside of After Effects. So it makes sense to comp at 32-bit depth. A 32-bit depth matte may seem like overkill to you, but in terms of eliminating dither, and stepping, in the ranges of the pixels, it makes a lot of sense. The EXR format will scale, so usually the frames are tiny in filesize, no 16-bit lossy compression needed. The Position pass will benefit from the added complexity. Crushing the passes with Curves or Levels will show that…

    Also you need to make sure that inside of Interpret Footage, Preserve RGB is checked on for the Motion Vector pass. This is near-vital, and perhaps using RSMB instead of RSMB Pro Vectors helped in your case. I don’t know. I tend to ensure it’s on.

    Multi-pass EXR is nice if you use Nuke – however until the implementation is updated inside of AE (as in, rewriting that Fnord nonsense for 2016) it will take an absolute age to preview your shot if you are stacking extracted EXR layers.

    The WP Pass obviates the need for supersampling, but rendering Motion Vector and Depth passes at 1440p with all textures, lights, reflection, transp, AA turned off usually takes minutes with anything you could throw at it.

    I’m amazed that your comp worked with the adjustment layers – no precomps – but if you look at the intersections between the coffee beans in space, you notice that the depth matte isn’t behaving. I’d freak out and chuck it in a precomp at that point, but hey, the motion blur seems to have fixed it.

    All in it’s a lovely shot and it all looked great in the end, so please ignore all my trolling!

    • Ha, thanks for the big comment! I appreciate your 32bit preference! I find myself wanting to strangle AE in 32bit, so I usually settle for 16bit. However, when I comp in Fusion (my comper of choice), I’m always in full float.

  • Michael Tompert June 1, 2016 at 12:48 pm

    Great tutorial. That will come in real handy…

    Speaking of handy… here’s a handy pronunciation guide for Frischluft (it means “fresh air”)

    Try saying “FishLooft” then add an R after the F… “FrishLooft”

    and that’s how you pronounce Frischluft.

  • Vladan Miladinovich June 1, 2016 at 2:12 pm

    Chad, this was very informative. Thanks.
    I have couple of questions.

    1) you added grey solid under the extracted WPP. good. then you inverted it and crushed it with levels. what was the purpose of it ? In the end the very background is black . Couldn’t you achive all this by adding a solid black in ADD mode on top of everything in order to fill the alpha ?

    2) Are you sure the precomps with DOF and motion vectors HAVE to be above the adjustment layers even if they are invisible ?

    thanks.

    Vladan.

    • Vladan, yeah I ended up crushing it down you are right. I guess I added that solid because I wasn’t sure how far I wanted to push the overall blur, but as you pointed out, I did end up crushing it to get a more dramatic DOF falloff. As for your second point, I’ve always done it that way (precomps on top), but I’ve been told it’s a non-issue. Good habits die hard I guess 😉 Thanks for the comment!

  • Trevor Gilchrist June 1, 2016 at 3:13 pm

    SO useful. One of the most valuable tutorials I’ve come across to date. Thank you.

  • great tutorial,thx so much

  • Yoshwar Saridam June 2, 2016 at 3:22 am

    Thank you for this really helpful tutorial especially with the WPPass. Finally I can confidently leave the .RLA workflow to get the depth pass that I’ve been using for all these years. Especially with the still unsolved bug that discards the zpass completely from RLA files if rendering with Team Render, it sucks to have to render the RLA separately in single machine.

    Just want to add a vital info that for the Motion Vector pass to work correctly with the Pro Vectors plugin, it needs to have its “Preserve RGB” checked (Interpret Footage>Color Management tab>Preserve RGB checked) otherwise the blur will not look correct.

    The Motion Vector pass also needs to use the alpha channel with a bit of Simple Choker/Minimum applied to the matte. That what works for me so far.

    Cheers

  • Hey Chad, I was playing with this workflow and I was pretty confused. When I imported my .EXR ito after effects and EXtracoR’ed my position pass, I was still getting AA on the edges. I found that alpha channel needs to be turned off in the C4D render settings to get that nice clean edge.

    Maybe I missed something in the video, maybe?

    • Hard to say without looking at your file/comp, but as far as I know, the WPP does not have any anti-aliasing.

      • I really think it is the alpha channel messing me up. I cannot find a way to make this workflow work with alpha enabled. Every render with alpha has AA on the edges of the WPP.

        • I know it’s long after the fact but we ran into this AA issue with alphas as well. The simplest is to bring in two versions of the same footage to AE – one you interpret to alpha as NONE and use that for your depth pass comp and the other you use for the RGB pass with alpha intact.

  • Great Tuts , I love these little tips and tricks

  • Thanks for the tips, I’ve never used a position pass before, I’ll need to try that method out!

    An alternate method is to use the min-max filter, found under channel effects in AE. It gets rid of the nasty aliasing that usually shows up on the edges of DOF and motion vector passes out of C4D.

    The min-max effect basically forces values to the values of neighboring pixels and comes in handy whenever I have nasty edges on my passes. It works kind of like a smart matte choker to expand and contract color values.

  • Thanks for a tutorial, Chad. In my workflow I usually just make a Z-path render using Standart with an anti-aliasing turned off which gives me same result I guess. And then I simply adjust a Depth Buffer curve inside DoF . I find this way more quick or am I wrong about the same result? Did you make any compare of both methods? Thank you.

    • I am a lazy man 😉 I really hate setting up separate render passes. I prefer to get everything packed into one scene (not setting up work around scenes) and one exr render when possible. Most of the time, I’ll render out DOF and Mblur in camera and not have to use this method. Octane and Arnold have been spoiling me 😉

      • I understand, but I was wondering more about technical terms and difference. Does both of this methods produce the same accurate result or there is something more then just personal preferences behind? 🙂

        • I don’t see any technical reason for it to be incorrect. I would say that on complex jobs with multiple shots, your workflow will take more time to set up than the one I outlined in the video. Setting up separate renders even with Takes will be more time consuming than the WPP method.

  • Where is that LUT from? It says “GSG” on it, is that a part of some pack or something?

  • Glenn Takakjian June 3, 2016 at 1:59 pm

    I was excited to try this method but am experiencing a lot of issues. I have transparencies with objects inside which make the Position Pass a little weird in the transparent areas.

    But the main problem I am encountering is: My Position Pass once I run it through Extractor is Anti-Aliased. It looks correct in C4D but it is definitely anti-aliased in after effects. I have double and triple checked your procedure but I can’t seem to get it non- Anti-aliased.

    • Unfortunately, Frilschuft DOF is not ideal for scenes with a lot of transparency/refractions. I especially don’t recommend it if your scene has transparencies with objects inside them. Depth and Position passes do not take transparency into account. You may have to bite the bullet and use in camera DOF, but without seeing your scene I can’t say for certain.

  • I tried setting up a similar simple scene and used 1000 for my Position Pass Scale. I used the camera’s Front Map Blur distance and camera cone to determine my farthest object of interest. I also tried very small numbers too but no matter what my blue channel for depth was a solid blue with no shading at all.

    Unfortunately, I’m going to be stuck with a normal depth pass for now.

  • Great tutorial. Have struggled for years to get a good Depth pass from C4d. I found my way through Vray

    Would love to see Fog/Environment compositing.

  • Not sure if anybody asked this, but does the Position Pass even look at the camera’s focus point? If not then this is pretty huge if you can just adjust where you want the focal point t be (using lens care’s target).

    The only problem I see is that you can’t designate isolate a font blur or a rear blur.

    thanks!

    • I do not believe that the Position pass looks at the focal point.

      • Right. So if you have a scene with a moving camera and you want to stay focused on a specific object- you’re kind of screwed. Correct? Any workflows for that?

        (Still great method though)

        • Because it’s in camera space, yes you would have to “pull focus” just like you would if it were a real camera.

  • Pedro Ferreira June 8, 2016 at 6:36 am

    Thanks for the tut, these tips are gold 🙂 It took me a long time to figure this out in the past.

  • Just figured out that you said Nuke and Fusion not new confusion at 5:03. That helped my google search a lot. Lol!

    • Ha, yah I’m a Fusion guy. Can’t beat the bang for the buck. Nuke is great, don’t get me wrong, but Fusion has a free version which is awesome. Highly recommend it.

  • Wondering if there is an alternative to using EXRs… a file format that correctly support 16bit, 32 bit color info and can work in Aftereffects? Tiff? PNGs? I realize each pass would be its own sequence. And the position pass blue channel would somehow need to be isolated.

    EXRs have proven too slow in Aftereffects doing HD 8-12 minute shows all comped in AE, especially when EXRs are saved on a slow network drive and not on a fast local drive.

    Not an ideal drive set up, but its what I have to work with.

    • Actually, rendering out separate EXR files (16bit Half Float / ZIPS scanline) should work fine. This should lighten up the load significantly on your network. AE does suck when dealing with EXRs and sorta sucks for anything over 8bit. I always recommend anyone that is doing a significant amount of compositing of 3D renders, to learn a node based comper like Fusion or Nuke. Those programs handle multichannel EXRs much faster.

  • Tomás De la Rosa June 12, 2016 at 9:49 pm

    Nice but it’s a headache to set the scale for the Position Pass, can’t figure out the correct value yet, I still geting black results, please give us any rule.

    • I guess the only rule I’d give is to work in a physically correct scale. Don’t work in a scene that is mega-huge or microscopic.

  • Will this technique work with the hair render?

    Getting a depth pass for hair seems to be impossible in C4D

  • Hey Chad! Thanks for this tutorial. I’ve been trying to generate my position pass, but no matter what my settings are, i keep getting a four colored mosaic. (blue, white, magenta and cyan).
    https://goo.gl/18xooF
    Any thoughts?
    Thanks.

  • Finally a correct tutorial about DOF and MB for C4D and After effects. And it is from GSG, of course. Guys I love you so much, you have no idea.

  • These are some GOLD technics to learn! Thank you Chad!

  • This just isn’t working for me. Every thing in my scene is a consistent Blue. Ive been messing around with the Scale using higher and lower values and it just isn’t changing. Any ideas?

  • Thanks Chad for this.
    I ain’t having luck getting the Motion Vector pass using R18 and physical render. I only get a black EXR sequence. Changing the Motion Scale doesn’t work. Curiously if I render the same scene using the standard renderer the Motion Vector pass magically appears.
    My scene is an array of wine bottles using real world object scale {bottles are aprox. 30 cm. high}.
    Any advice?

    Thanks in advance and have a great 2017 in all your ways.

  • Excellent tutorial, Chad. I finally understand why I’ve been struggling with DOF artifacts for so long.

    But then if anti-aliasing is the culprit, why not just turn it off? Would that not also fix the problem? (Not sure if this the same question DENNI was asking.)

  • Bryn Desmond-Jones February 17, 2017 at 1:46 am

    Great tutorial! Thank you

  • Does this technique only work with the Physical renderer? What happens if you render OpenEXR’s out with the Standard Renderer? Thanks for the awesome tutorial. 🙂

  • Hello CHad,

    Have you tried to render in the RPF format? (https://www.youtube.com/watch?v=agLHxkKPDRM), seems to be a good option too.

    About OpenEXR, maybe it’s better to work in 32bits deth an put en AE/Project Settings/Color Settings as: Depth – 32 bits; Working Space – sRGB IEC61966-2.1; Linearize working Space activated. All this settings just to avoid the darkening of the position or other passes because EXR works in Linear Color workspace.

    I’ve tried with standard renderer and it’s working too.

    I’ve just tried your porpouse and works pretty good. With a floor I had to create an object buffer to use it as luma matte for the motion vector.

  • Hi guys,
    Any idea how to do this with arnold AOV’s?
    The Position pass doesn’t seem to have any options in it, according to their website is “world” by default and can’t be changed to camera.
    thanks

    • You do not need this method if you are using Arnold. The zpass coming out of arnold is totally fine. Thanks for the comment!

  • If you export RPF and use 3D Channel Extractor, the depth pass is exported correctly (unaliased). Much easier than the method mentioned in this tutorial.

    • I used your method, way easier even if i have to render the z-path seperatly with the standard renderer. Finding the good scale setting for the position pass is so tricky… I still have trouble when my camera is animated, my zpath is constantly changing even if my focus point is targetted.

  • I set the Camera Position Pass scale to the minimum (which is 0.001) but still cant see any shape of blue gradation… I tried to resize my scene placed in a nul but the keyframes don’t seem to scale down with it. I don’t know what can i do more.

  • I watched the tutorial four or five days after you uploaded it and still, although there are plenty of 3rd party renderers with nice built-in blur out there, this tutorial is so helpful and I think it’s one of the most needed tutorials for me. Always forget how to do it right and then re-watching the whole tutorial. Thank you for this amazing helpful clip!

  • using frischluft you can just change the depth buffer to take depth from the blue channel – wouldnt need to do all the extraction and then desaturation etc. I think you could just add levels effect and clamp the blue channel to your liking. seems to work ok to me. i tried both ways and saw no difference between using luminance and whites versus the blue channel only.

  • Hi all,

    Thanks for the tips Chad, love the Tut.
    Totally new to C4D, I had a lots to catch up…
    I was trying to understand (like many of us) why the WPP came out with AA and Chad didn’t face the issue in his Tuto.
    Correctly if I am wrong, but Chad used his original scene which didn’t got any alpha (this is why he did a Luma key on the WPP) as he mentioned that he kept the C4D background of his scene.
    When you don’t used Alpha pass then your WPP has no AA and work fine, like here. But if you do use an alpha channel, then you have to turn of the AA in the render option (like mentioning in the comments) for the WPP. Hope it helps?

  • Leave a Reply

    Your email address will not be published. Required fields are marked *

    Categories

    Follow us on Instagram