i would love to see a displaced groundplane that simulates a hill (for example), using something such as 3dtext to cast a shadow. purely to show how you can apply 3d text to an image and displace the shadow to match the image
i was looking on the internet about how i can fix the rectangle shadows from my fog in the 3d scene, and the internet brought me to you again. it looks like all complicated problems lead to your channel. The problem was in the openGL option. You talk like you actually created this software... do you have a special higher education in Davinci and Graphics, otherwise i don't understand how you know this? By the way, i have 3d animation/scene. If i switch away from the OpenGL, will i loose a lot? You said about the depth of field? what does it do to my animation?
Hey VFX, do you know if it's possible to turn your 2D plane, into a crude 3D object. So lets say for eg. you had a human standing (from a delta keyed out green screen) could you tell the program to round out where the colours are, that'd give a kind of rounded out legs, arms, body etc. which in a crude way would make it 3D and better for shadows. kind of like how it's done with text but for a 2D object that alpha channels in it's background.
Yes, you can deform the geometry with a displace3D node which you can feed with a depth map. You can create the depth map for example by using the mask that you have and then blur the edges to get a soft transition between foreground (white mask) and further back (black area). You have to make sure the 3D plane has enough subdivisions so that the geometry can reproduce the detail that you need.
I'm lost at the very beginning. I have a set up like him (only it's a cartoon character with a "green screen" that is keyed out, so he stands as a "solid" object). And I just don't get any shadow out of it. And that is back at the 3 minute mark before he goes way beyond my payscale with all the other tricks.
As always, you answer specific and technical questions that many people ask themselves with Davinci Resolve. I have a few questions if you have time to answer them: Is it better to have a photorealistic rendering to use OpenGL rather than Sofware renderer? (knowing that Open GL seems more limited with shadows and maybe the management of transparency).
Please, if you place a image on a plane or project to a plane how do you adjust the brightness and contrast of that projected image? And is there a way to select a part of this projected image so you can adjust that selection only? Your course is so full I have been trying out weeks now and keep going back to re-find things. thanks Alistair
Hi Alistair, How about changing brightness of the image before projecting? E.g. brightness and contrast tool. Then you can also use masks in 2d to select parts of it. In 3D there are a few more things that effect the brightness. If you have lights enabled then the lights obviously decide how bright it gets. In addition there is the diffuse color of the material settings of the shape3d or imageplane3d. If you have it set to white it just used the texture input unchanged but if you set it to grey it will darken the incoming texture. If I'm not mistaken it's a multiply operation.
Thank you I created a copy of the media in and on both have brightness and contrast and on one a ellipse both into merge so I can alter the brightness of either the ellipse or the copy two independently. Also your tip on the material to grey also works for over all adjustment
Where did this happen? I'm multiplying the shadow, so the texture can get only darker, not brighter and am otherwise using the original scene, so not sure where there's a corona?
@@VFXstudy The corona appears where the shadow does not darken the background behind the object. The Depth of Field will blur the edge of the shadow in the background, revealing the cut out object casting the shadow. The closer the the shadow is to the focus plane the less it is visible, which I guess is why it stays invisible in your example. It should become visible at the dogs tail or butt when you crank up the defocus (or when you shift the focus plane to the shadow behind the dog and the blur of the dog's card reveals more of the background.
@@Schmidtelpunkt oh you have a point there. But how do we solve this now? Maybe going one step further and creating passes from the full shadows without the foreground objects and then again putting the foreground object in front for the openGL renderer. Might loose some flexibility though... Any better idea?
@@VFXstudy Yep, that or projecting the full shadow pass on everything but the shadowcasting object (by limiting by object ID). Either way, this tutorial is a great way to give some insight into the catcher material, which otherwise is a rather abstract concept.
Hi Saad, was down with the flu for over a week and my voice still doesn't sound like I'll be recordining for another few days. But I have something lined up already. So starting from next week there'll be a nice multi part series I'm quite excited about...
What do you mean? Actually more lights support shadows now and the Hardware render has become a bit more flexible - but you still need to decide where you want your shadow and where not - so there's the "catching part". Unless you want it default out of there renderer, then let's hope we get new render engines via USD soon....
This is the ultimate shadow casting tutorial. I have many plans for this one and can't wait to get busy this weekend!
Thank you very much for the tutorial, you just answered my question in the comments and also did a tutorial of my question, you are awesome!!
You're welcome! This one triggered some interesting thoughts and fun playing around so I decided it was worth a tutorial.
This is WAY over my current skills level, but look forward to following your channel to learn more!
This one is more advanced. Feel free to check out my Free beginner course on vfxstudy.com if you're looking for a place to get started.
every clip is a present! thx!
i would love to see a displaced groundplane that simulates a hill (for example), using something such as 3dtext to cast a shadow.
purely to show how you can apply 3d text to an image and displace the shadow to match the image
Dear FriendYou are amazingKeep it upMay god bless your Kind heat
Excellent, thank you!
i was looking on the internet about how i can fix the rectangle shadows from my fog in the 3d scene, and the internet brought me to you again. it looks like all complicated problems lead to your channel. The problem was in the openGL option. You talk like you actually created this software... do you have a special higher education in Davinci and Graphics, otherwise i don't understand how you know this?
By the way, i have 3d animation/scene. If i switch away from the OpenGL, will i loose a lot? You said about the depth of field? what does it do to my animation?
amazing!!!! thank you
Hey VFX, do you know if it's possible to turn your 2D plane, into a crude 3D object. So lets say for eg. you had a human standing (from a delta keyed out green screen)
could you tell the program to round out where the colours are, that'd give a kind of rounded out legs, arms, body etc.
which in a crude way would make it 3D and better for shadows.
kind of like how it's done with text but for a 2D object that alpha channels in it's background.
Yes, you can deform the geometry with a displace3D node which you can feed with a depth map. You can create the depth map for example by using the mask that you have and then blur the edges to get a soft transition between foreground (white mask) and further back (black area). You have to make sure the 3D plane has enough subdivisions so that the geometry can reproduce the detail that you need.
1:33 I have Software Renderer on anywhere I could find, and it still renders a rectangular shadow, what else could it be?
Nevermind, I must have accidentally enabled Non-Transmittive Materials in my Spotlight node.
Hi Bernd, it works great for still, but if you move the camera you have to link the projector and the camera for correct shadow.
I copied the camera to create the projector - copy together with all animations. Instant copy would also do or link with expressions.
I'm lost at the very beginning. I have a set up like him (only it's a cartoon character with a "green screen" that is keyed out, so he stands as a "solid" object). And I just don't get any shadow out of it. And that is back at the 3 minute mark before he goes way beyond my payscale with all the other tricks.
As always, you answer specific and technical questions that many people ask themselves with Davinci Resolve. I have a few questions if you have time to answer them: Is it better to have a photorealistic rendering to use OpenGL rather than Sofware renderer? (knowing that Open GL seems more limited with shadows and maybe the management of transparency).
Please, if you place a image on a plane or project to a plane how do you adjust the brightness and contrast of that projected image?
And
is there a way to select a part of this projected image so you can adjust that selection only?
Your course is so full I have been trying out weeks now and keep going back to re-find things.
thanks
Alistair
Hi Alistair,
How about changing brightness of the image before projecting? E.g. brightness and contrast tool. Then you can also use masks in 2d to select parts of it.
In 3D there are a few more things that effect the brightness. If you have lights enabled then the lights obviously decide how bright it gets. In addition there is the diffuse color of the material settings of the shape3d or imageplane3d. If you have it set to white it just used the texture input unchanged but if you set it to grey it will darken the incoming texture. If I'm not mistaken it's a multiply operation.
Thank you I created a copy of the media in and on both have brightness and contrast and on one a ellipse both into merge so I can alter the brightness of either the ellipse or the copy two independently.
Also your tip on the material to grey also works for over all adjustment
But doesn't the cut off shadow lead to a bright corona around the object at places with greater difference in depth?
Where did this happen?
I'm multiplying the shadow, so the texture can get only darker, not brighter and am otherwise using the original scene, so not sure where there's a corona?
@@VFXstudy The corona appears where the shadow does not darken the background behind the object. The Depth of Field will blur the edge of the shadow in the background, revealing the cut out object casting the shadow. The closer the the shadow is to the focus plane the less it is visible, which I guess is why it stays invisible in your example. It should become visible at the dogs tail or butt when you crank up the defocus (or when you shift the focus plane to the shadow behind the dog and the blur of the dog's card reveals more of the background.
@@Schmidtelpunkt oh you have a point there. But how do we solve this now? Maybe going one step further and creating passes from the full shadows without the foreground objects and then again putting the foreground object in front for the openGL renderer. Might loose some flexibility though... Any better idea?
@@VFXstudy Yep, that or projecting the full shadow pass on everything but the shadowcasting object (by limiting by object ID).
Either way, this tutorial is a great way to give some insight into the catcher material, which otherwise is a rather abstract concept.
@@VFXstudy Bernd, do you think you could show us what you have in mind in that regard with a video? Thanks!
Genius
🔥
Hey Bern. Very informative channel and I like your "nice guy" approach. What tablet are you using?
I have a Wacom intous tablet, one of the simplest and smallest models that was available 4 years ago and it does the job for me.
Great :-) thx
Super Video!!!! Mach doch mal ein Video auf deutsch;)
Hey Bernd! New video coming soon I hope? Hope all is well.
Hi Saad, was down with the flu for over a week and my voice still doesn't sound like I'll be recordining for another few days. But I have something lined up already. So starting from next week there'll be a nice multi part series I'm quite excited about...
@@VFXstudy it's that time of the year when everyone is getting sick. Hope you are better soon! Looking forward to your next series!
Can you please do a tutorial or quick tips about compositing stock footages like explotions with world position pass?
Five years have passed and Davinci still hasn't learned how to catch a shadow without displaying a plane. Five years....
What do you mean? Actually more lights support shadows now and the Hardware render has become a bit more flexible - but you still need to decide where you want your shadow and where not - so there's the "catching part". Unless you want it default out of there renderer, then let's hope we get new render engines via USD soon....