Best video on this topic. Short and full of truw information about resolution, gamma curve, bitrate and Auto Black Balance (did not know nothing about that). So, for anyone who wants to do Netflix Original, this is the video you need. Now, all you have to do after watching this video is set up the setrings the right way and "only to be creative". Thanks for this video and for the rest of r5c videos.
Very detailed information! Much appreciated! Question: I am using the Rode Wireless Pro and renting the Canon R5C. We're in the development stage of a documentary (targeting Netflix) and will be interviewing the main character. The information from the interview will be used to develop a Pitch Deck, scripting writing, story development, etc., and possibly some footage will be in the documentary. With these things in mind, how should I set up the audio portion of the Canon R5C and the Rode Wireless Pro (I already have the video setup based on the provided Netflix info)? Thanks in advance.
Sorry, but audio is something that I'm not even remotely qualified to give advice on. I barely know what I'm doing for my own videos, never mind trying to tell someone how they should setup their audio.
I'd love to, but... I have barely more than 0 clue what I'm doing when it comes to audio stuff. I can record the audio for these videos, I think well enough, but beyond that I'm barely past knowing which way to point the microphone. I'll put it on my list of things to look into, because, hey you never know I might learn something, but I can't make any promises that it'll happen.
I just built a brand new video editing PC with an i9 14th gen (quick sync) plus a 4090. I'm wondering if you have a suggestion for proxies? I can't wait for the video on HEVC versus XF AVC. I'm still trying to decide shooting everything all I XAVC or HEVC. I did notice a small quality drop between the two. Not much difference in timeline performance.
If you want small files, use 4:2:0 HEVC or AVC (either will be accelerated by your GPU). If you want more accuracy use ProRes Proxy or DNxHR LB. Neither ProRes or DNx will be hardware accelerated, but they should run smoothly at 4K on an i9-14900K (4K ProRes ran fine on my i7-8700K from 5 years ago, and the coded was designed for systems that were much less powerful than that). I'm betting the problem you're seeing with switching from XF-AVC to HEVC is that you're still shooting 4:2:2 chroma sub-sampling. As far as I know, Premiere Pro won't use Intel's QuickSync decode hardware to accelerate that (regardless of whether it's AVC, HEVC, or AV1). And no Nvidia or AMD GPU supports hardware decode of 4:2:2 chroma sub-sampled content. In DaVinci Resolve, I see full hardware supported playback of 4:2:2 files on my Intel IGP while my Nvidia card handles the 4:2:0 stuff.
I am shooting a lot on raw 8K and 12 bits and a love the image coming from the R5C (I am finishing to 4K but the master for DaVinci is as high as I can) the end product looks better that shooting in regular 4K
To be honest, that's one of the things that I found interesting about Netflix's guide. They basically don't promote/permit shooting in raw. I know the bitrate on the R5C is lower than the bit rate on other Canon cine cameras at least on a per pixel basis, but if you're only targeting 4K output, the raw files should probably be better than XF-AVC. Raw performance is actually something I'm trying to investigate, but it's near impossible to make objective comparisons between similar but not identical videos. You can't use the normal tools like SSIM, PSNR, or VMAF. And I'm not sure I'm smart enough to figure out a way that would actually work. And subjective is problematic, because of differences in perception and biology. That is I might not notice something that someone else does either because I'm not looking for it, or because I physically can't see the difference.
@@PointsInFocuswell is something like a star wars idea “Use the force” is what your trained eyes see, is not about tests in front of lens charts and color checkers, is about shooting interesting terrain, nature, lots of depth in plains and compare it, you will be amazed about what you see
So I am guessing if I am shooting in slow and fast motion mode I am getting lower quality. Should I be just slowing down the footage in post? That's what I always used to do for my slow mo in 4K. Interesting...
Slowing down footage in post requires the computer to either hold frames for longer than they should be held (e.g. reduces your effective frame rate) or create new frames based on some algorithm. The AI frame generation can work, but it can also result in serious visual errors as well - though usually it works fine. I wouldn't recommend slowing things down in post at all. Pretty much ever. The motion blur will be wrong, and whatever gains you think you're getting in image qualitiy for not using S&F motion at >60 FPS is going to largely be offset by the excessive emotion blur and the reliance on computer generated frames to fill in the gaps.
One thing is bothering me before getting this camera. I am a fan of non ibis cameras in cinema line up. But does Canon's electronic stabilization with stabilized lens like RF 24-70 2.8 give bad/good/very good performance in handheld recording while doing slight panning? Thanks
I don't think I've seen too many issues with panning handheld with optical IS system enabled on that lens and my R5C. I want to say I have some a few instances of the IS system jumping while panning, but I can't remember if they were handheld or on a tripod where I forgot to turn IS off. The problem, as always, is that it's not deterministic. What happens in one take won't necessarily happen in the next. I've always tended to avoid digital IS, and almost never recommend it. But in thinking about it the R5C may be an exception to this, at least if you're outputting at 2K or 4K. The camera should still be down sampling, just not from the full 8K frame. I'll have to stick that on my list of things to look into.
One one hand, you can save all of the menu settings to a file on a SD card. On the other hand, that doesn't work quite as well as the custom user modes on the regular EOS cameras.
For now, that's a question you'd have to ask the folks at Netflix. I'm hoping to dig into the image quality situation with raw on the R5C in the coming months, but try as I might I haven't gotten there yet.
Best video on this topic. Short and full of truw information about resolution, gamma curve, bitrate and Auto Black Balance (did not know nothing about that). So, for anyone who wants to do Netflix Original, this is the video you need.
Now, all you have to do after watching this video is set up the setrings the right way and "only to be creative".
Thanks for this video and for the rest of r5c videos.
Very useful info.. thanks
thanks for the continuing R5 C videos
Another excellent video! Thanks! 🙏🏻
Thank you so much for the education video. It's very helpfull!
Keep it up man ! Great stuff
Netflix: "Ah yes, we don't want productions to have access to raw video, that would be too easy"
Very detailed information! Much appreciated! Question: I am using the Rode Wireless Pro and renting the Canon R5C. We're in the development stage of a documentary (targeting Netflix) and will be interviewing the main character. The information from the interview will be used to develop a Pitch Deck, scripting writing, story development, etc., and possibly some footage will be in the documentary. With these things in mind, how should I set up the audio portion of the Canon R5C and the Rode Wireless Pro (I already have the video setup based on the provided Netflix info)? Thanks in advance.
Sorry, but audio is something that I'm not even remotely qualified to give advice on. I barely know what I'm doing for my own videos, never mind trying to tell someone how they should setup their audio.
Doesn't really concern my type of use but I find the information useful and thank you for the video as always
Awsome video happy someone took the time to explain this 🙏🏽🙏🏽 any chance you can do Netflix sound capture requirements
I'd love to, but... I have barely more than 0 clue what I'm doing when it comes to audio stuff. I can record the audio for these videos, I think well enough, but beyond that I'm barely past knowing which way to point the microphone.
I'll put it on my list of things to look into, because, hey you never know I might learn something, but I can't make any promises that it'll happen.
This happy dude seriously reminds me of the fly-swatting guy from Men in Black.
I just built a brand new video editing PC with an i9 14th gen (quick sync) plus a 4090. I'm wondering if you have a suggestion for proxies? I can't wait for the video on HEVC versus XF AVC. I'm still trying to decide shooting everything all I XAVC or HEVC. I did notice a small quality drop between the two. Not much difference in timeline performance.
If you want small files, use 4:2:0 HEVC or AVC (either will be accelerated by your GPU). If you want more accuracy use ProRes Proxy or DNxHR LB. Neither ProRes or DNx will be hardware accelerated, but they should run smoothly at 4K on an i9-14900K (4K ProRes ran fine on my i7-8700K from 5 years ago, and the coded was designed for systems that were much less powerful than that).
I'm betting the problem you're seeing with switching from XF-AVC to HEVC is that you're still shooting 4:2:2 chroma sub-sampling. As far as I know, Premiere Pro won't use Intel's QuickSync decode hardware to accelerate that (regardless of whether it's AVC, HEVC, or AV1). And no Nvidia or AMD GPU supports hardware decode of 4:2:2 chroma sub-sampled content.
In DaVinci Resolve, I see full hardware supported playback of 4:2:2 files on my Intel IGP while my Nvidia card handles the 4:2:0 stuff.
I am shooting a lot on raw 8K and 12 bits and a love the image coming from the R5C (I am finishing to 4K but the master for DaVinci is as high as I can) the end product looks better that shooting in regular 4K
To be honest, that's one of the things that I found interesting about Netflix's guide. They basically don't promote/permit shooting in raw. I know the bitrate on the R5C is lower than the bit rate on other Canon cine cameras at least on a per pixel basis, but if you're only targeting 4K output, the raw files should probably be better than XF-AVC.
Raw performance is actually something I'm trying to investigate, but it's near impossible to make objective comparisons between similar but not identical videos. You can't use the normal tools like SSIM, PSNR, or VMAF. And I'm not sure I'm smart enough to figure out a way that would actually work. And subjective is problematic, because of differences in perception and biology. That is I might not notice something that someone else does either because I'm not looking for it, or because I physically can't see the difference.
@@PointsInFocuswell is something like a star wars idea “Use the force” is what your trained eyes see, is not about tests in front of lens charts and color checkers, is about shooting interesting terrain, nature, lots of depth in plains and compare it, you will be amazed about what you see
So I am guessing if I am shooting in slow and fast motion mode I am getting lower quality. Should I be just slowing down the footage in post? That's what I always used to do for my slow mo in 4K. Interesting...
Slowing down footage in post requires the computer to either hold frames for longer than they should be held (e.g. reduces your effective frame rate) or create new frames based on some algorithm. The AI frame generation can work, but it can also result in serious visual errors as well - though usually it works fine.
I wouldn't recommend slowing things down in post at all. Pretty much ever. The motion blur will be wrong, and whatever gains you think you're getting in image qualitiy for not using S&F motion at >60 FPS is going to largely be offset by the excessive emotion blur and the reliance on computer generated frames to fill in the gaps.
One thing is bothering me before getting this camera. I am a fan of non ibis cameras in cinema line up. But does Canon's electronic stabilization with stabilized lens like RF 24-70 2.8 give bad/good/very good performance in handheld recording while doing slight panning? Thanks
I don't think I've seen too many issues with panning handheld with optical IS system enabled on that lens and my R5C. I want to say I have some a few instances of the IS system jumping while panning, but I can't remember if they were handheld or on a tripod where I forgot to turn IS off.
The problem, as always, is that it's not deterministic. What happens in one take won't necessarily happen in the next.
I've always tended to avoid digital IS, and almost never recommend it. But in thinking about it the R5C may be an exception to this, at least if you're outputting at 2K or 4K. The camera should still be down sampling, just not from the full 8K frame. I'll have to stick that on my list of things to look into.
Now if only we had custom modes to save different shooting profiles... Facepalm canon.... Facepalm...😤
One one hand, you can save all of the menu settings to a file on a SD card. On the other hand, that doesn't work quite as well as the custom user modes on the regular EOS cameras.
Man, why not raw?
For now, that's a question you'd have to ask the folks at Netflix. I'm hoping to dig into the image quality situation with raw on the R5C in the coming months, but try as I might I haven't gotten there yet.