Thanks so much. Super happy to hear that 🥰 About the emojis 😬 I have not verified that they are actually usable on TH-cam 😁 But I will of course gladly take it in written form... Let's see if it actually works 🤞 ✨🙏2⃣0⃣2⃣4⃣🙏✨
Thank you Emad. Yes, I do so too. If you look at Cycles for example, you know what a really optimized engine looks like. I really hope they can do the same step by step with Octane. They made good progress with the ram usage. But it could go even a lot further 🙌 I am trying to push them towards this direction!
Hey hey. Ohhh, that is amazing. I can't remember the question you asked on Patreon 😇 hopefully I somewhat answered it already 😬 But great to hear that you now also have the answer from the video 🙌
Octane need to upgrade their gallery on their main site. Why haven't they in years? Came to you believe it or not originally from your Gameboy animation
Hey Jose, thanks for your great comment and your answer. Yes, you can indeed denoise a shadow pass AOV. With the Open Image Denoiser you can denoise every AOV available. You have to go through the comp context of Octane to do that though. Layer your shadow AOV in layer one and above that add an Open Image Denoise. If you are in C4D. It should look something like this: th-cam.com/video/bBqdE9Md8uk/w-d-xo.html I am using the beauty here, but you can use whatever channel you want. If you have further questions, just ask 🙌
Thanks for the great video again! Are you actually a frequent user of compositing? I never really used it. And very recently on an episode on the Corridor Channel somebody made a great argument. He said that if you want photorealistic results it's best to not play around with all these compositing layers, because the render engine is programmed to simulate actual light bouncing around. And when you're tweaking all these render layers in post our eyes can pick it up that for example certain reflections are much brighter than they would be in real life. I thought it was an interesting take, what do you think about it as someone who specializes in photorealism?
Hey Sleezy, thank you very much for this great comment! I really appreciate it! I personally use compositing to have freedom of choice after the render, where I e.g. play around with the strength of the scene lights. This is doable in comp and does not break physics! When it comes to the corridor statement about comp I think there is truth in the statement. But it really depends what is done to the shot in comp. As I just said, you can use comp to increase / decrease your lights just as you would have in your render. You can also use compositing to "cheat" and try to add effects in there that were not there in the original render. As e.g. fog. But you could also use compositing to emulate stuff that makes a photo look like a photo because it was shot through a lens on a piece of film for example. In my example that would be emulating lens flares from a lens and glow / halation from a film. This would in my opinion enhance the render and make it more realistic, and not degrade it. This is why I am pro EXR and 32bit compositing. Because it gives us the realistic simulated light. You can pipe this data to yet another realistic sim in 2D space in comp e.g. to reproduce glow. Also keep in mind that the sensor / film usually has it's own look as well. Especially if you think about the film era. You could have film that increases the contrast of the image, or film that has a certain tint. e.g. shooting outside with indoor film makes everything look very blueish. Last thing to think about is, that with films from the 2000s there are really good examples where there was a lot of comp work done, also because back then the compute was not as great, but the results really look realistic and believable. I really think it's time constraints and the experience of the compositing artist that can make or break a shot! If he tweaks it to just look nice, then you might end up in CGI land. Whereas if the comper has a good understanding of film and light physics, then it can look superior then the raw render in terms of realism. Oh well, I think this answer escalated a bit ha ha ha. So welcome to my Ted talk!
Would be nice to also have a free tier for C4D I think. Probably this is mostly because Blender is free also. Also the free version is limited to one GPU and does not have network rendering capability.
Thanks for your message 🙌 All builds I used in a tutorial are declared on the top header of the video 😇 I am guessing you are referring to Blender Octane, which was: Blender Octane 29.14.1 In this built I can't find the "Denoised Albedo" or "Denoised Normal" within the "Render AOVs". But I might be blind and have overlooked them 😅
@@TheThebeep Coming from C4D and finding the "Denoise Albedo" and "Denoised Normal" in the Denoise section of the AOVs I was kind of biased to expect the same in Blender 😇
Hey Maximilian, I am happy to try to shine you a light. Instead of using Photoshop, After Effect, Fusion Studio, Nuke, you can do your comp directly in the renderer. The first advantage here is that you have the power of post work right in Octane, so you do not have to pay for / use another package. Second: You can render once and have multiple looks to your render and save it all out at once. Third: Though a bit limited, you can be more time efficient by rendering out the main rendering once and then attach e.g. multiple labels to your product. I made a video about this. With a bit or creativity here, the sky's the limit: th-cam.com/video/L-dypgeAXq4/w-d-xo.html Last: I personally think it has most of its benefit in pipelines where you have to e.g. do 300 renders with variations. If you nail down the look in Octane AOV Nodes, you can hit render and the output will be the final result. With existing pipelines you would have to render out your image, go into After Effects / Photoshop and do your post processing there. Which is OK for one image, or a couple. But get's rather tedious with doing it with 300. Also less moving parts = less breaking points.
Rendering can take a long time, especially for animations. AOVs can create masks for any object or texture so you can adjust it afterwards on After-effects or Fusion for example. AOVs can separate the base colour from the reflections, refractions, blurs, etc so you can adjust and manipulate them, add effects to them, e.g. you could adjust the colour of the glow or light over time to suit your animation without re-rendering. Depth of field AOV for adding blur like a camera. There are other ones which you can manipulate for creative intent. There is nothing wrong with just rendering out a flat image if you like it and don’t want to really do anything after, but AOVs allow surgical manipulation within your pipeline, especially h in a linear workflow with 32 float images.
I’d consider myself fairly experienced in C4D Octane but I can’t see any practical reason to use the AOV nodes as opposed to rendering passes/cryptomatte and doing the rest in a dedicated compositing app.
@@Oldyellowbrick but isn’t it the same thing? Using the mode system is just a way of building up and organising your passes for output. I think it has some advantage in that you can combine parts together also. Am I missing something.
I am decoding a 2024 out of there. Thank you very much 🎉 Actually I have not checked if TH-cam allows for those emojis. And it seems it does not 😅 I should always check! Only Mac_Daffy managed to get it do display and I have no idea how he did it!
Thanks for the video, but this update is a complete disgrace, OTOY from year to year promise a very big roadmap with a lot of cool features that are really necessary for the engine to stay afloat, but what we got this year is a real spit in the face of people who use octane and pay for it, from update to update there are no significant changes, even appear thoughts to go to redshift, which in this regard is much more progressive and promising. Shame on OTOY
Thanks for your comment. I wish we could have gotten more features aside from the comp system. Though I think the engine has a great track record giving us features that we, the users, asked for. I love all those AOVs that came in that are so useful! If you feel RS is doing better, I encourage you giving it a try. It's also a very good engine and maybe the better one for you. While I am loving Octane. I am totally open for everyone to use what he deems best 🙌
I realize that for some people this functionality may be useful, but it's nowhere near as useful as the unreal engine decals we were promised, meshlet shading, texture streaming, neural rendering improved Lens FX, improved BRDF models, Headless rendering with multi-render and multi-engine live linking, Temporal denoiser (not open image denoiser, the open denoiser that everyone added 2+ years ago), etc. etc., the list of how OTOY deceived and continue to deceive their users can go on for a long time. I absolutely love octane and wish OTOY would at least not make false promises. Special thanks to you for not ignoring such posts like moderators do on forums.
@@_surf_8905 I do realize that those delayed and unkept promises are a problem. I was really looking forward to Meshlets as well. And while a lot of other stuff like Neural Rendering is on the direct roadmap to Octane 2024.2, there was no mention about Meshlets in there. I think they bit off a bigger piece then what they could chew unfortunately. I find "deceive" a rather strong word, because that implies to spread misinformation purposefully. And I genuinely don't think they do that. They get enthusiastic about features, but in the background they are also a rather small team that has to tackle a lot that's on the table. And while I would definitely agree that the feature announcements do need management and go into the direction of promising stuff that might never come. At the point it was given, they are developing it. or at least schedule the development of it. That said I totally can see the frustration that can come with those broken and delayed promises. It's not nice to see that happen again and again. Therefore the principle I live by is to always buy software for the features that are included and never for those that are promised. Because you never know if they will hold up to your expectations or come at all. As I am working with Redshift as well, there are users also that are complaining about delayed features, bad performance and saying they'd switch to Octane for a better experience. I always feel that it's about the expectations you have for a certain software that makes or breaks it for you. That or of there is a drastic change in direction the software is taking. Well, that's my two cents 😊 Cheers and a good start into this week to you!
this. I have been an octane user since 2019, my licence ran out mid 2023 and i still have yet to renew. these days alot of the work i do has been in ue because of cinema 4d's predatory pricing model and how poorly it handles large scale scenes. I was hoping to renew this year when the promised features was supposed to make their way in but far from them being implemented they refuse to even acknowledge that roadmap and every time it is brought up they refuse to speak on it or even silence it in the forum. very anti consumer. dont get me wrong otoy do some amazing stuff like all of the stuff that is package into a subscription but then they go and increase the base price of a subscription and remove the option for smaller artists like myself who used to purchase the 2gpu package. all the access to additional applications is great but what is the option for those who have no use for them and simply want to just use the renderer? i personally never used any off the other applications aside from dabbling in world creator here and there and would much rather have the option to be able to purchase at the older subscription model price if it meant i only had access to ocatne render itself. if they are not going to be able to deliver on their promises the least they can do is acknowledge it or make a reasonable pricing model for their customers. again dont get my wrong they arent as bad as other companys that sell plugins but for someone like myself who used to purchase the 2 gpu option the current model is almost 50% more expensive. then i have to purchase a licence for c4d which is also more expensive with every coming year at this point its just unreasonable sorry for the rant haha.
Thanks for this round trip on the new features!!! Good to update on things!!!
You always inspire me to load up C4D explore around push what I have to the next level.
That's amazing to hear that.
One of the best things that can happen. Inspire people to explore and be creative 🙏🙌😊
@@SilverwingVFX we also can have 48 Analytical Lights now
@@Clemyfourtwenty Ha, thanks a lot about that addition. I learned something today. Thank you good sir ✨🎩✨
Thanks for the overview of the updates. You are a treasure for Octane users. 2024 (I couldn't find the emojis)
Thanks so much. Super happy to hear that 🥰
About the emojis 😬 I have not verified that they are actually usable on TH-cam 😁 But I will of course gladly take it in written form... Let's see if it actually works 🤞 ✨🙏2⃣0⃣2⃣4⃣🙏✨
Babe wake up, silverwing uploaded new video
Ha ha ha. Love it 🙌🙌
The RAM and render time optimizes are really appreciated. Hope they focus even more on optimizing the engine. Thanks a lot Raphael!
Thank you Emad.
Yes, I do so too. If you look at Cycles for example, you know what a really optimized engine looks like. I really hope they can do the same step by step with Octane.
They made good progress with the ram usage. But it could go even a lot further 🙌
I am trying to push them towards this direction!
Thank you Raphael, the compositing nodes section answers my DM on Patreon perfectly!
Hey hey. Ohhh, that is amazing. I can't remember the question you asked on Patreon 😇 hopefully I somewhat answered it already 😬
But great to hear that you now also have the answer from the video 🙌
This AOV muti-pass system is becoming a very strong feature in Octane. U can really do things around there.
Great video as always thank you Raphael. 🙏
Thanks so much. You are very welcome 🙌
Thanks for doing Blender Octane as well🙏
Thank you very much. In this case it was actually necessary to show off some of the features 😇
Thank you so much for Sharing the love with octane blender🙏🙏🙏🙏🙏🙏
My pleasure! And thank you very much for the comment 🙏✨🙏✨🙏
The coffee tastes so much better now.
That's fantastic to hear 😊
Such a pleasure being part of your coffee ritual and managing to somehow make it taste better ha ha ha.
Octane need to upgrade their gallery on their main site. Why haven't they in years?
Came to you believe it or not originally from your Gameboy animation
Good point. I am not sure. Probably no one was assigned to that job.
If I speak to someone at Otoy, I will bring it up!
Thank you so much, master, amazing as always! Is there a way to denoise a shadow AOV pass?
Hey Jose,
thanks for your great comment and your answer.
Yes, you can indeed denoise a shadow pass AOV.
With the Open Image Denoiser you can denoise every AOV available.
You have to go through the comp context of Octane to do that though.
Layer your shadow AOV in layer one and above that add an Open Image Denoise. If you are in C4D. It should look something like this:
th-cam.com/video/bBqdE9Md8uk/w-d-xo.html
I am using the beauty here, but you can use whatever channel you want.
If you have further questions, just ask 🙌
Yeah!!! I was hoping for this one!
Thaaank youuu! I hope you like it 🙏🙌
Thanks for the great video again! Are you actually a frequent user of compositing? I never really used it. And very recently on an episode on the Corridor Channel somebody made a great argument. He said that if you want photorealistic results it's best to not play around with all these compositing layers, because the render engine is programmed to simulate actual light bouncing around. And when you're tweaking all these render layers in post our eyes can pick it up that for example certain reflections are much brighter than they would be in real life. I thought it was an interesting take, what do you think about it as someone who specializes in photorealism?
Hey Sleezy,
thank you very much for this great comment! I really appreciate it!
I personally use compositing to have freedom of choice after the render, where I e.g. play around with the strength of the scene lights. This is doable in comp and does not break physics!
When it comes to the corridor statement about comp
I think there is truth in the statement. But it really depends what is done to the shot in comp. As I just said, you can use comp to increase / decrease your lights just as you would have in your render.
You can also use compositing to "cheat" and try to add effects in there that were not there in the original render. As e.g. fog. But you could also use compositing to emulate stuff that makes a photo look like a photo because it was shot through a lens on a piece of film for example.
In my example that would be emulating lens flares from a lens and glow / halation from a film. This would in my opinion enhance the render and make it more realistic, and not degrade it. This is why I am pro EXR and 32bit compositing. Because it gives us the realistic simulated light. You can pipe this data to yet another realistic sim in 2D space in comp e.g. to reproduce glow.
Also keep in mind that the sensor / film usually has it's own look as well. Especially if you think about the film era. You could have film that increases the contrast of the image, or film that has a certain tint. e.g. shooting outside with indoor film makes everything look very blueish.
Last thing to think about is, that with films from the 2000s there are really good examples where there was a lot of comp work done, also because back then the compute was not as great, but the results really look realistic and believable.
I really think it's time constraints and the experience of the compositing artist that can make or break a shot!
If he tweaks it to just look nice, then you might end up in CGI land. Whereas if the comper has a good understanding of film and light physics, then it can look superior then the raw render in terms of realism.
Oh well, I think this answer escalated a bit ha ha ha.
So welcome to my Ted talk!
I don't understand why octane is free for Blender but pay for cinema 4d
Would be nice to also have a free tier for C4D I think.
Probably this is mostly because Blender is free also. Also the free version is limited to one GPU and does not have network rendering capability.
I can assume that Otoy needs to follow the « free to use » mindset pushed by the whole blender community
🔥🔥🔥
🙏🙏🙏
U r the goat
Ohhh, thank you so much 🥰
Albedo and Normal passes for Open Image Denoiser inputs are available in latest builds. What Builds do you use?
Thanks for your message 🙌
All builds I used in a tutorial are declared on the top header of the video 😇
I am guessing you are referring to Blender Octane, which was: Blender Octane 29.14.1
In this built I can't find the "Denoised Albedo" or "Denoised Normal" within the "Render AOVs". But I might be blind and have overlooked them 😅
Seems like I was blind. I had another look 😇 They are not in the denoised section. Where I was searching mostly 😅
@@SilverwingVFX actually they are not denoised passes but enhanced diffuse filter pass and normal smooth pass with added reflections.
@@TheThebeep Coming from C4D and finding the "Denoise Albedo" and "Denoised Normal" in the Denoise section of the AOVs I was kind of biased to expect the same in Blender 😇
@@SilverwingVFX in Standalone it is in same section as in Blender so I guess in C4D it should be moved over to proper section in coming releases.
I am still a bit confused about why AOV Nodes are cool, can somebody more knowledgable enlighten me?
Hey Maximilian, I am happy to try to shine you a light.
Instead of using Photoshop, After Effect, Fusion Studio, Nuke, you can do your comp directly in the renderer.
The first advantage here is that you have the power of post work right in Octane, so you do not have to pay for / use another package.
Second: You can render once and have multiple looks to your render and save it all out at once.
Third: Though a bit limited, you can be more time efficient by rendering out the main rendering once and then attach e.g. multiple labels to your product.
I made a video about this. With a bit or creativity here, the sky's the limit:
th-cam.com/video/L-dypgeAXq4/w-d-xo.html
Last: I personally think it has most of its benefit in pipelines where you have to e.g. do 300 renders with variations.
If you nail down the look in Octane AOV Nodes, you can hit render and the output will be the final result.
With existing pipelines you would have to render out your image, go into After Effects / Photoshop and do your post processing there.
Which is OK for one image, or a couple. But get's rather tedious with doing it with 300.
Also less moving parts = less breaking points.
@@SilverwingVFX Thank you so much for the in depth reply! I totally forgor about the possibility to reduce the comp stack thanks to that :D
Rendering can take a long time, especially for animations. AOVs can create masks for any object or texture so you can adjust it afterwards on After-effects or Fusion for example. AOVs can separate the base colour from the reflections, refractions, blurs, etc so you can adjust and manipulate them, add effects to them, e.g. you could adjust the colour of the glow or light over time to suit your animation without re-rendering. Depth of field AOV for adding blur like a camera. There are other ones which you can manipulate for creative intent.
There is nothing wrong with just rendering out a flat image if you like it and don’t want to really do anything after, but AOVs allow surgical manipulation within your pipeline, especially h in a linear workflow with 32 float images.
I’d consider myself fairly experienced in C4D Octane but I can’t see any practical reason to use the AOV nodes as opposed to rendering passes/cryptomatte and doing the rest in a dedicated compositing app.
@@Oldyellowbrick but isn’t it the same thing? Using the mode system is just a way of building up and organising your passes for output. I think it has some advantage in that you can combine parts together also. Am I missing something.
🌀🌀🌀🌀
I am decoding a 2024 out of there. Thank you very much 🎉
Actually I have not checked if TH-cam allows for those emojis. And it seems it does not 😅
I should always check!
Only Mac_Daffy managed to get it do display and I have no idea how he did it!
@@SilverwingVFX I've tried several times. But this is the best I could come up with. Thanks for your great videos.
@@samoxinvik Thanks for trying 😇 As I said. I should have checked beforehand 😅
Thanks for the video, but this update is a complete disgrace, OTOY from year to year promise a very big roadmap with a lot of cool features that are really necessary for the engine to stay afloat, but what we got this year is a real spit in the face of people who use octane and pay for it, from update to update there are no significant changes, even appear thoughts to go to redshift, which in this regard is much more progressive and promising. Shame on OTOY
Thanks for your comment.
I wish we could have gotten more features aside from the comp system.
Though I think the engine has a great track record giving us features that we, the users, asked for. I love all those AOVs that came in that are so useful!
If you feel RS is doing better, I encourage you giving it a try. It's also a very good engine and maybe the better one for you. While I am loving Octane. I am totally open for everyone to use what he deems best 🙌
I realize that for some people this functionality may be useful, but it's nowhere near as useful as the unreal engine decals we were promised, meshlet shading, texture streaming, neural rendering improved Lens FX, improved BRDF models, Headless rendering with multi-render and multi-engine live linking, Temporal denoiser (not open image denoiser, the open denoiser that everyone added 2+ years ago), etc. etc., the list of how OTOY deceived and continue to deceive their users can go on for a long time. I absolutely love octane and wish OTOY would at least not make false promises. Special thanks to you for not ignoring such posts like moderators do on forums.
@@_surf_8905 I do realize that those delayed and unkept promises are a problem.
I was really looking forward to Meshlets as well. And while a lot of other stuff like Neural Rendering is on the direct roadmap to Octane 2024.2, there was no mention about Meshlets in there. I think they bit off a bigger piece then what they could chew unfortunately.
I find "deceive" a rather strong word, because that implies to spread misinformation purposefully. And I genuinely don't think they do that. They get enthusiastic about features, but in the background they are also a rather small team that has to tackle a lot that's on the table. And while I would definitely agree that the feature announcements do need management and go into the direction of promising stuff that might never come. At the point it was given, they are developing it. or at least schedule the development of it.
That said I totally can see the frustration that can come with those broken and delayed promises. It's not nice to see that happen again and again.
Therefore the principle I live by is to always buy software for the features that are included and never for those that are promised. Because you never know if they will hold up to your expectations or come at all.
As I am working with Redshift as well, there are users also that are complaining about delayed features, bad performance and saying they'd switch to Octane for a better experience.
I always feel that it's about the expectations you have for a certain software that makes or breaks it for you.
That or of there is a drastic change in direction the software is taking.
Well, that's my two cents 😊
Cheers and a good start into this week to you!
this. I have been an octane user since 2019, my licence ran out mid 2023 and i still have yet to renew. these days alot of the work i do has been in ue because of cinema 4d's predatory pricing model and how poorly it handles large scale scenes.
I was hoping to renew this year when the promised features was supposed to make their way in but far from them being implemented they refuse to even acknowledge that roadmap and every time it is brought up they refuse to speak on it or even silence it in the forum. very anti consumer.
dont get me wrong otoy do some amazing stuff like all of the stuff that is package into a subscription but then they go and increase the base price of a subscription and remove the option for smaller artists like myself who used to purchase the 2gpu package. all the access to additional applications is great but what is the option for those who have no use for them and simply want to just use the renderer? i personally never used any off the other applications aside from dabbling in world creator here and there and would much rather have the option to be able to purchase at the older subscription model price if it meant i only had access to ocatne render itself.
if they are not going to be able to deliver on their promises the least they can do is acknowledge it or make a reasonable pricing model for their customers. again dont get my wrong they arent as bad as other companys that sell plugins but for someone like myself who used to purchase the 2 gpu option the current model is almost 50% more expensive. then i have to purchase a licence for c4d which is also more expensive with every coming year at this point its just unreasonable
sorry for the rant haha.
@@SilverwingVFX So do you think Meshlets has been dropped completlely now then? Or maybe just added to the list of permanantly delayed features.
Performance increases are always welcome!
Thank you very much for your comment! And yes, Indeed they are! 🙌✨🙌