You make very in depth tutorials! I think you should have way more views than you have currently! Anyone willing to learn 3D in DaVinci should watch your videos. Even though they are long, they are all worth it! Thanks for sharing your knowledge and keep up the good work!
Great usage of stacking planes in 3D, same technique can be used for doing long shadows, I'll have to remember to also try abusing camera projection to fake pseudo extrusion with stacks. This would be so much easier if the weld tool was able to bridge the gaps but alas.
Do you mean "stacking" with multiple projectors somehow? Or just projecting onto a stack of planes? It's not yet clear in my mind... but if you work something out please share. With Fusion there are always 5 things to try out for every problem :-)
Very nice! I'm surprised DR does not do this out of the box with solid Z sides of user selected color(s) or simply inferring color from 2D side. I keep looking for some other approach, perhaps something using recent changes to DR... but every tutorial I encounter either copies your video or takes an approach that seems quite involved, not the auto-magic approach that Text 3D provides (but for images). By the way, videos I've seen which copy your approach have all been created recently after your video so congrats on leading the way... and for being good enough that others creators copy you! 👍
There is a new and better way now in 18.6 you can take Polygons into shape tools with sPolygon and then take them into 3d space via Extrude3D. That's creating true 3d geometry from 2d vector graphics
@@VFXstudyThank you... for that approach, how are you getting the image into sPolygons? I have tried import SVG which comes in as normal polygons/shapes which must then be converted to sPolygons which offhand has seemed like a PIA in my testing for not only complex SVGs but even simpler ones. In one case so far, I opted back to your Duplicate approach as it was good enough and far easier to manage than all the copy/paste/clicking with the aforementioned... and for all that copy/paste/clicking, I was running into errors... for example, one SVG had XOR or inverting blitting or masking which was in some cases was coming through to the sPolygons after copy/paste but was getting lost in some cases. To be clear, when that XOR'ing or masking was working, it's not visible as an enabled feature in the sPolygon Inspector.... but it nevertheless seems like sPolygons, under the covers, support such but just not via Inspector UI that I could see. Anyway, despite such support, SVG import to copy/paste to sPoloygon seems to lose that in some "random" cases. I shelved it for a non-critical case and just used Duplicate. Maybe you have a totally different approach w/sPoloygon but this is all to say, your Duplicate approach, which, yes, has issues compared to pure 3D, is still quite viable as evidenced by everyone still copying your 4 year old video. 😬
This is all native from just an image, if you need more Depth control consider the font hack that lets you "type" your "image" as text and allows easier usage of stuff like bender 3D(3Bn). WSL or bmd forum should have posts on it? Roughly: convert image to svg to a font ,font added to system. Use text-3D(3Txt) node with font set to the custom font. Considerations: search "convert image to font" to learn the process, use something like calligraphr service for alot of characters. May need to convert to SVG no matter what, optionally go black&white, remember transparency, and a large font set can be very slow. For stuff like social media icons there are several *opensource* projects that have them as brand icon fonts (fontawesome, fontello on github). Though you may have trouble finding the character code,unicode value, or ligature, however most services may generate a zip with a demo html file to literally copy and paste between programs.
That's a cool hack as well. Haven't thought of it. May get tricky with multicoloured images, but guess you could still project onto it or texture it. Though when starting to use external tools to create your own fonts, I'm wondering if I wouldn't directly go for tools that convert to a 3D mesh. Either way, thanks for adding another trick to the list :-)
Thanks for the tutorial! I tried it in a different way for a project (5 png on top of each other with some space in the Z axis, with a light to cast shadow, and trying to simulate a "multi-layer 3d logo"), but the shadows are so jaggy. I'm sure I'm missing an option to activate.
@@VFXstudy Thank you so much for this!!! I stopped working on my little project because it drove me NUTS for days. I've tested every single others options (on the 3D object itself), and didn't even saw the Shadows "drawer" on the light tool! (I mainly use the standalone Fusion 9 version). That's why I like your channel. You go into small details which can be use for so many other things than the main subject of the video. I love the technical stuff!
What does the 3d extrude tool do? This seems like such a convoluted way to do something so simple and the more duplicates you do the worse the render. This isn't even a work around thing in After Effects. This shouldn't be so difficult.
This is an old video with workarounds that are no longer necessary since 18.6 The Extrude3D was released with v18.6 and allows you to take a shape tool into 3D space. And there is sPolygon that allows you to take a polygon into the system of shape tools so you can effectively go from Polygon all the way to 3D with just two tools. See my 18.6 update video for this.
Hi Bernd, Great stuff, thanks. More guidance on Ambient Occlusion would be great but I'm also hoping you'd find time to do something on Materials/Textures and how they work with Bumpmap and CookTolerance/Phong etc. I always seem to struggle with resolution and sizing! Thanks!
I think the replicate3d node might also work. That's the one you use to copy onto the nodes of a 3D geometry, right? Had no need for that here, but as long as you get the geometry, I suppose it should work.
Hallo Bernd! Klasse Video! Mir hat noch gefehlt, wie du die Animation...also das herauswachsen der Logos aus dem Boden animiert hast. Kannst du das noch erklären?
Einfach im 3D Raum die Z-Koordinate der entsprechenden Shape 3D objekte animiert und die einzelnen Logos so quasi aus der Hintergrundebene herausgefahren. Kannst auch geren meine Fusion comp von der Webseite herunterladen und anschauen.
Hey Bernd, great tutorial! thank you for sharing! would love something on the differences of nuke and fusion, and how to find the appropriate similar tools in fusion. As i'm learning a node based compositiing, I keep trying to do something and can't find the tool in fusion. Thanks you again for sharing your info
Thanks, Buon for the suggestion. It's a great idea, but something that might take a while to realize. I watched some Nuke tutorials before and tried to recreate some stuff I saw with Fusion for exercise, but I'm far from calling myself proficient with Nuke. I might look deeper into it since some students from that background have specific questions, so I should learn more - although much more often people come from After Effects... :)
so, ive done same as you say on your video.. but i have a Big Question ofr you pal.... After all that.. Were or How im Supposed to do to render this image in to Chroma Key Rotating Image? Because when i tried to rotate, the image, It does.. nor Not all the Images but just the ones created whit Z Offeset... hope you Respond... because you just show us how to make all that.. but not to render the 3d image... or make it to rotate
Well the 3D to 2D rendering is done in a Render3D node. I assume I added one in the example file but if not, just add after the merge3d. Then you have your image in 2D. From there you can transform it with a transform node if you like. In Fusion Studio you can save it to disk through a saver node, in DaVinci Resolve you would add a MediaOut node to bring your forward to Color and Deliver Page.
@@VFXstudy il try.. thank you so much! im barely new to Resolve.. and thanks to you.. Fusion Doesnt look Like Quantic Work..... i love davinci resolve.. i want to make 3d Images rotating, to Make it Green screen orChroma key. or diferent Models in 3D. Anyway... Have a nice Day
Not sure about blender, but I assume if there is a simple fader it would first require a vector graphic so there are some vertices to extend from. Except for the roto splines Fusions 2D stuff is mostly based on raster image. I suppose that's why having that simple slider may be more difficult.
Thank you for this video! I have an issue that I can't, for the image I have, have just the image and not also the black background that the 2D image is on. Is there a way to strip out just my image from the black background? Hope that makes sense...
I cannot move around the object with my mouse in the preview like you do. When I move my mouse it moves the object and not the view. How can I do it like you do please ? thks !
You can attach an image texture directly to the green texture input of any shape3d node. It gets wrapped around the shape. Alternatively you can add shaders from the existing templates or create your own using the material nodes like blinn, cook-torrance etc. For more on that you might want to watch my materials in Fusion tutorial
Came here wishing there was something like a "solidify modifier" :( That said. For a simple scene like this, even if you never used Blender before, it will probably be a lot easier to get a great result. (Resolve will be easier to get an ok result like this, but I can't imagine how much work it'd take to get close to the default Principled BSDF of Blender + some free HDRi).
Yeah the full rendering with raytracing and image based lighting g etc.is not something for Fusion. If you have 3d meshes then this here is also not needed and you can load fbx meshes directly. But sometimes having a simple way for some added titles or simple graphics without going to Blender or similar is quite nice.
@@VFXstudy Ah, it worked now. Maybe was a temporary problem there since I tried another of your links and also didn't work and my internet was working fine in other pages. Anyway, problem solved.
For which part? The best true 3d option would be to convert the image into a vector via some edge tracing and then convert that into a 3D mesh. But I don't think krokodove has such capability - what do you have in mind?
very detailed, but I don't understand why explain for 5 min what not to do? Just come to the point... I watch videos becaus what I want to see, not what I don't need to do ;)
Thanks. I'll do my best. Generally I make videos so people understand the background, not just copy a single effect without understanding. Sometimes that includes additional advice 😉
You make very in depth tutorials! I think you should have way more views than you have currently! Anyone willing to learn 3D in DaVinci should watch your videos. Even though they are long, they are all worth it! Thanks for sharing your knowledge and keep up the good work!
Thanks a lot!
Thanks for making this! Helped me out a lot!
Thanks for sharing your time and expertise!
I would like to say the same so I take Mr. Prasad words. Thank You!!
Fantastic video, thank you so much. Very easy to follow. Can't wait to use some of these nodes that I didn't even know existed!!
Thanks for the tutorial! It helped me figure out why I couldn't connect 3D nodes to 2D ones.. I was missing the renderer3d node xd
Great usage of stacking planes in 3D, same technique can be used for doing long shadows,
I'll have to remember to also try abusing camera projection to fake pseudo extrusion with stacks.
This would be so much easier if the weld tool was able to bridge the gaps but alas.
Do you mean "stacking" with multiple projectors somehow? Or just projecting onto a stack of planes? It's not yet clear in my mind... but if you work something out please share. With Fusion there are always 5 things to try out for every problem :-)
VFXstudy sorry meant stacking as in duplication really, projection was a side thought
Very nice! I'm surprised DR does not do this out of the box with solid Z sides of user selected color(s) or simply inferring color from 2D side. I keep looking for some other approach, perhaps something using recent changes to DR... but every tutorial I encounter either copies your video or takes an approach that seems quite involved, not the auto-magic approach that Text 3D provides (but for images). By the way, videos I've seen which copy your approach have all been created recently after your video so congrats on leading the way... and for being good enough that others creators copy you! 👍
There is a new and better way now in 18.6 you can take Polygons into shape tools with sPolygon and then take them into 3d space via Extrude3D. That's creating true 3d geometry from 2d vector graphics
@@VFXstudyThank you... for that approach, how are you getting the image into sPolygons?
I have tried import SVG which comes in as normal polygons/shapes which must then be converted to sPolygons which offhand has seemed like a PIA in my testing for not only complex SVGs but even simpler ones.
In one case so far, I opted back to your Duplicate approach as it was good enough and far easier to manage than all the copy/paste/clicking with the aforementioned... and for all that copy/paste/clicking, I was running into errors... for example, one SVG had XOR or inverting blitting or masking which was in some cases was coming through to the sPolygons after copy/paste but was getting lost in some cases. To be clear, when that XOR'ing or masking was working, it's not visible as an enabled feature in the sPolygon Inspector.... but it nevertheless seems like sPolygons, under the covers, support such but just not via Inspector UI that I could see. Anyway, despite such support, SVG import to copy/paste to sPoloygon seems to lose that in some "random" cases. I shelved it for a non-critical case and just used Duplicate.
Maybe you have a totally different approach w/sPoloygon but this is all to say, your Duplicate approach, which, yes, has issues compared to pure 3D, is still quite viable as evidenced by everyone still copying your 4 year old video. 😬
This is all native from just an image, if you need more Depth control consider the font hack that lets you "type" your "image" as text and allows easier usage of stuff like bender 3D(3Bn). WSL or bmd forum should have posts on it?
Roughly: convert image to svg to a font ,font added to system. Use text-3D(3Txt) node with font set to the custom font.
Considerations:
search "convert image to font" to learn the process,
use something like calligraphr service for alot of characters.
May need to convert to SVG no matter what, optionally go black&white, remember transparency, and a large font set can be very slow.
For stuff like social media icons there are several *opensource* projects that have them as brand icon fonts (fontawesome, fontello on github). Though you may have trouble finding the character code,unicode value, or ligature, however most services may generate a zip with a demo html file to literally copy and paste between programs.
Example of calligraphers image grid www.calligraphr.com/static/quicktest/quicktest_prefilled.jpg
That's a cool hack as well. Haven't thought of it. May get tricky with multicoloured images, but guess you could still project onto it or texture it. Though when starting to use external tools to create your own fonts, I'm wondering if I wouldn't directly go for tools that convert to a 3D mesh.
Either way, thanks for adding another trick to the list :-)
Hi Bernd, Please do a tutorial on the AmbientOcclusion and the DepthBlur nodes. Thanks,
Really interesting tutorial, thanks for sharing!
You're welocme :-)
Thanks for the tutorial!
I tried it in a different way for a project (5 png on top of each other with some space in the Z axis, with a light to cast shadow, and trying to simulate a "multi-layer 3d logo"), but the shadows are so jaggy. I'm sure I'm missing an option to activate.
Check the shadow map size in the options of the spotlight. You might have to increase that number for higher resolution shadows
@@VFXstudy Thank you so much for this!!!
I stopped working on my little project because it drove me NUTS for days.
I've tested every single others options (on the 3D object itself), and didn't even saw the Shadows "drawer" on the light tool! (I mainly use the standalone Fusion 9 version).
That's why I like your channel. You go into small details which can be use for so many other things than the main subject of the video. I love the technical stuff!
Thanks Bernd, I appreciate you!
Great work! Thanks!
Thank you. Nice tips :)
You're welcome :)
What does the 3d extrude tool do? This seems like such a convoluted way to do something so simple and the more duplicates you do the worse the render. This isn't even a work around thing in After Effects. This shouldn't be so difficult.
This is an old video with workarounds that are no longer necessary since 18.6 The Extrude3D was released with v18.6 and allows you to take a shape tool into 3D space. And there is sPolygon that allows you to take a polygon into the system of shape tools so you can effectively go from Polygon all the way to 3D with just two tools. See my 18.6 update video for this.
@@VFXstudy Thanks so much. Appreciate it.
Hi Bernd, Great stuff, thanks. More guidance on Ambient Occlusion would be great but I'm also hoping you'd find time to do something on Materials/Textures and how they work with Bumpmap and CookTolerance/Phong etc. I always seem to struggle with resolution and sizing! Thanks!
Thanks for the input. Sounds like more then one tutorial, so some ideas for the near future :-)
@@VFXstudy Thanks, looking forward to them. BTW could you use Replicate instead of Duplicate here to produce a similar effect?
I think the replicate3d node might also work. That's the one you use to copy onto the nodes of a 3D geometry, right? Had no need for that here, but as long as you get the geometry, I suppose it should work.
Really great video thank you
Ok sold. LIke and SUbs . Thx for the knowledge
I like it
please more like this
Hallo Bernd! Klasse Video! Mir hat noch gefehlt, wie du die Animation...also das herauswachsen der Logos aus dem Boden animiert hast. Kannst du das noch erklären?
Einfach im 3D Raum die Z-Koordinate der entsprechenden Shape 3D objekte animiert und die einzelnen Logos so quasi aus der Hintergrundebene herausgefahren. Kannst auch geren meine Fusion comp von der Webseite herunterladen und anschauen.
Hey Bernd, great tutorial! thank you for sharing! would love something on the differences of nuke and fusion, and how to find the appropriate similar tools in fusion. As i'm learning a node based compositiing, I keep trying to do something and can't find the tool in fusion.
Thanks you again for sharing your info
Thanks, Buon for the suggestion. It's a great idea, but something that might take a while to realize. I watched some Nuke tutorials before and tried to recreate some stuff I saw with Fusion for exercise, but I'm far from calling myself proficient with Nuke. I might look deeper into it since some students from that background have specific questions, so I should learn more - although much more often people come from After Effects... :)
thanks a lot!!
so, ive done same as you say on your video.. but i have a Big Question ofr you pal.... After all that.. Were or How im Supposed to do to render this image in to Chroma Key Rotating Image? Because when i tried to rotate, the image, It does.. nor Not all the Images but just the ones created whit Z Offeset... hope you Respond... because you just show us how to make all that.. but not to render the 3d image... or make it to rotate
Well the 3D to 2D rendering is done in a Render3D node. I assume I added one in the example file but if not, just add after the merge3d. Then you have your image in 2D. From there you can transform it with a transform node if you like. In Fusion Studio you can save it to disk through a saver node, in DaVinci Resolve you would add a MediaOut node to bring your forward to Color and Deliver Page.
@@VFXstudy il try.. thank you so much! im barely new to Resolve.. and thanks to you.. Fusion Doesnt look Like Quantic Work..... i love davinci resolve.. i want to make 3d Images rotating, to Make it Green screen orChroma key. or diferent Models in 3D. Anyway... Have a nice Day
So Fusion is, in contrast to Blender, incapable to give depth to 2D images (with a single fader)?
Not sure about blender, but I assume if there is a simple fader it would first require a vector graphic so there are some vertices to extend from. Except for the roto splines Fusions 2D stuff is mostly based on raster image. I suppose that's why having that simple slider may be more difficult.
Of course the Text3d tool also has that extrusion slider.
thanks you are beautiful
Thank you for this video! I have an issue that I can't, for the image I have, have just the image and not also the black background that the 2D image is on. Is there a way to strip out just my image from the black background? Hope that makes sense...
Try the luma keyer?
@@VFXstudy thank you!!
I cannot move around the object with my mouse in the preview like you do. When I move my mouse it moves the object and not the view. How can I do it like you do please ? thks !
Ctrl / alt + third mouse button. If the object is moving you probably have it selected with the left mouse button
Having replaced Loader & media-out, the output is visible on fusion page but the fusion composition does not play in edit page timeline
Loader is instead of MediaIn. You always need the mediaOut node to send the image back to the edit.
@@VFXstudy Actually I wanted to connect media out without using saver node.
How did you add the texture to the white background?
You can attach an image texture directly to the green texture input of any shape3d node. It gets wrapped around the shape. Alternatively you can add shaders from the existing templates or create your own using the material nodes like blinn, cook-torrance etc. For more on that you might want to watch my materials in Fusion tutorial
@@VFXstudy hey, thank you so much. Will definitely watch that tutorial! Fusion is so much fun to learn
спасибо!
Came here wishing there was something like a "solidify modifier" :(
That said. For a simple scene like this, even if you never used Blender before, it will probably be a lot easier to get a great result. (Resolve will be easier to get an ok result like this, but I can't imagine how much work it'd take to get close to the default Principled BSDF of Blender + some free HDRi).
Yeah the full rendering with raytracing and image based lighting g etc.is not something for Fusion. If you have 3d meshes then this here is also not needed and you can load fbx meshes directly. But sometimes having a simple way for some added titles or simple graphics without going to Blender or similar is quite nice.
Hey Bernd, seems like your link is not working. Just so you know.
In the video description? Seems fine to me, just opened the page and able to download as well... could you check again?
@@VFXstudy Ah, it worked now. Maybe was a temporary problem there since I tried another of your links and also didn't work and my internet was working fine in other pages. Anyway, problem solved.
Oh, ok, your name is Bernd. Meiner auch ;)
or u just can use krokodove plugin
For which part? The best true 3d option would be to convert the image into a vector via some edge tracing and then convert that into a 3D mesh. But I don't think krokodove has such capability - what do you have in mind?
@@VFXstudy i mean crocodove can do perfect 3d models from any 2d imgs. But u should use Studio version of davinci for this
Or u can create 3d logo in 3dbuilder. Standart windows 10 application and import in davinci. But krokodove is much easily
Are you german?
Ja, bin ich.
very detailed, but I don't understand why explain for 5 min what not to do? Just come to the point... I watch videos becaus what I want to see, not what I don't need to do ;)
Thanks. I'll do my best. Generally I make videos so people understand the background, not just copy a single effect without understanding. Sometimes that includes additional advice 😉