Can't do anything with my R5 video

The 10 bit 4:2:2 files you can say that they are for grading not viewing.

I have I7 model 8 and 32GB RAM and GPU with 12 GB RAM those 10-bit4:2:2 files they don't play without stutter in any videoplayer.

I use Davinci Resolve Studio and it is probably one of the best editing program I know of (atleast for PC).

If you're not into video or want to make HDR videos stay away from CLOG or HDR PQ
I’m totally into videos, I’ve just never been into making them 🙂 And HDR vids, heck yes !
and try the 8-bit 4K HQ 30p mode it is very good. If you get hooked on video you have those grading modes available.
I will 👍
I like to use HDR PQ mode and deliver HDR videos of nature or even vacation videos in Davinci Resolve because the result is amazing on my 77" Oled TV.
That sounds awesome. Ours is only a little “70”er 😀

TY.
 
I still think it’s kind of crazy that Canon would make a camera that shoots such crazy high quality video, that almost nobody can work with it “easily”.
Time to be blunt about it - you are starting to have strong opinions about things you understand very little and are just starting out with. Sorry 🙂

The highest quality this camera shoots is 8K RAW, and it's fairly easy to work with using the right software. Now, you are probably thinking - the software should come free? But why would anyone bundle software for free that normally costs hundreds or sometimes even thousands of dollars. And why would a camera company develop its own NLE when clearly that would need them to build a completely new set of expertise. They can probably bump up the price of camera a few hundred dollars and bundle resolve studio, or a premiere subscription or avid or whatever, but then which one should they bundle? What if you use the other one? The pros already have their workflows in place

Don't believe me - find out which NLE Arri Alexa ships with? It's a $100,000 camera...
 
this is just incorrect though. you can easily just expose for clog to get more DR and then apply the canon supplied luts to bring whatever video file back to normal. just drag and drop in shutter encoder or apply the lut in your editing software.
You are repeating this often, so I have to ask - assuming you are saying it is simple because all you apply is the Canon LUT - what benefit do you get shooting log in this case. This will be the same as non log footage. The higher DR you capture only helps if you can choose which parts of that range you choose to keep in your final outcome, and that's editing. If it's just applying a predefined Canon LUT, you may as well skip log and expose accordingly - the outcome would be same. If it was a custom lut, I can understand that in camera processing won't replicate it - but then, that breaks the "its easy" barrier, because one needs to know what they are doing
 
this is just incorrect though. you can easily just expose for clog to get more DR and then apply the canon supplied luts to bring whatever video file back to normal. just drag and drop in shutter encoder or apply the lut in your editing software.
You are repeating this often, so I have to ask - assuming you are saying it is simple because all you apply is the Canon LUT - what benefit do you get shooting log in this case. This will be the same as non log footage. The higher DR you capture only helps if you can choose which parts of that range you choose to keep in your final outcome, and that's editing. If it's just applying a predefined Canon LUT, you may as well skip log and expose accordingly - the outcome would be same. If it was a custom lut, I can understand that in camera processing won't replicate it - but then, that breaks the "its easy" barrier, because one needs to know what they are doing
the curve from clog is still going to be superior for DR compared to contrast -4 with HTP

the real advantage you get from using log though is you get 10bit 4:2:2 compared to 8bit 4:2:0. big advantage if you want to do any image manipulation in post. only reason to use non-log currently is if you're shooting home video or vlogging where you want the file basically straight from camera where you have zero worry about posterization or artifacts. if you're doing any editing with any effects applied, 10bit 4:2:2 will be far superior regardless of DR.
 
Last edited:
Well, we will see. Give me a bit to play with the free programs, and minimize my video settings...
 
this is just incorrect though. you can easily just expose for clog to get more DR and then apply the canon supplied luts to bring whatever video file back to normal. just drag and drop in shutter encoder or apply the lut in your editing software.
You are repeating this often, so I have to ask - assuming you are saying it is simple because all you apply is the Canon LUT - what benefit do you get shooting log in this case. This will be the same as non log footage. The higher DR you capture only helps if you can choose which parts of that range you choose to keep in your final outcome, and that's editing. If it's just applying a predefined Canon LUT, you may as well skip log and expose accordingly - the outcome would be same. If it was a custom lut, I can understand that in camera processing won't replicate it - but then, that breaks the "its easy" barrier, because one needs to know what they are doing
the curve from clog is still going to be superior for DR compared to contrast -4 with HTP
Yes, "IF" you are DR limited, clog will caputure more "editable" headroom than non-log footage. Assuming you are ready to utilize that headroom in post. Simply applying the same LUT always doesn't guarantee that
the real advantage you get from using log though is you get 10bit 4:2:2 compared to 8bit 4:2:0. big advantage if you want to do any image manipulation in post.
Yes, but also why I asked the question. If all you are going to do is apply Canon LUT and export to 8bit 4:2:0, you will be back to what camera does. Then in the process of applying that lut, if you crush your shadows, you are no better than shooting non-log. To get the benefit, you need some processing, that depends on the scene. The lut only gives a good starting point. So bottom line, if one isn't willing to do any further post or doesn't know how to process log, all bets are off
only reason to use non-log currently is if you're shooting home video or vlogging where you want the file basically straight from camera where you have zero worry about posterization or artifacts. if you're doing any editing with any effects applied, 10bit 4:2:2 will be far superior regardless of DR.
Agree. Your other posts seemed to imply just apply lut and render. That is not the editing and effects that you are saying here

--
PicPocket
http://photography.ashishpandey.com
 
Last edited:
this is just incorrect though. you can easily just expose for clog to get more DR and then apply the canon supplied luts to bring whatever video file back to normal. just drag and drop in shutter encoder or apply the lut in your editing software.
You are repeating this often, so I have to ask - assuming you are saying it is simple because all you apply is the Canon LUT - what benefit do you get shooting log in this case. This will be the same as non log footage. The higher DR you capture only helps if you can choose which parts of that range you choose to keep in your final outcome, and that's editing. If it's just applying a predefined Canon LUT, you may as well skip log and expose accordingly - the outcome would be same. If it was a custom lut, I can understand that in camera processing won't replicate it - but then, that breaks the "its easy" barrier, because one needs to know what they are doing
the curve from clog is still going to be superior for DR compared to contrast -4 with HTP
Yes, "IF" you are DR limited, clog will caputure more "editable" headroom than non-log footage. Assuming you are ready to utilize that headroom in post. Simply applying the same LUT always doesn't guarantee that
the real advantage you get from using log though is you get 10bit 4:2:2 compared to 8bit 4:2:0. big advantage if you want to do any image manipulation in post.
Yes, but also why I asked the question. If all you are going to do is apply Canon LUT and export to 8bit 4:2:0, you will be back to what camera does. Then in the process of applying that lut, if you crush your shadows, you are no better than shooting non-log. To get the benefit, you need some processing, that depends on the scene. The lut only gives a good starting point. So bottom line, if one isn't willing to do any further post or doesn't know how to process log, all bets are off
only reason to use non-log currently is if you're shooting home video or vlogging where you want the file basically straight from camera where you have zero worry about posterization or artifacts. if you're doing any editing with any effects applied, 10bit 4:2:2 will be far superior regardless of DR.
Agree. Your other posts seemed to imply just apply lut and render. That is not the editing and effects that you are saying here
the point i'm trying to make is that shooting in log is not hard to edit. OP wants to edit not just have files from camera. people are saying basically not to use log because it's too hard. i'm arguing that it's not hard because you can just transcode to prores and use a canon lut.

now you're inventing a new scenario where if they did want files straight from camera, is there still an argument for log?

in general, it's probably not as efficient because just by the nature of the work, if it's in a realm with no editing, it's likely to not be important enough. in terms of quality though, it still will be a higher quality final video comparatively. it's not "the same" as you seem to think.

4k 340mbps h265 > 1080p 90mbps h264

4k downscaled > 1080p from camera

10bit 4:2:2 > 8bit 4:2:0

without even getting into gamut choice, clog, or luts, you have to know you're going to end up with better quality files even after conversion. it's not like once the file gets converted it's the same. the source file significantly impacts the final compressed file.

so in theory, if i really valued quality but didn't want to edit, i could shoot 4k clog and batch process all the files to smaller downscaled 1080p h265 files w/ canon lut and they would still be better than if i just shot 1080p h264 no log from camera.
 
Last edited:
...
I like to use HDR PQ mode and deliver HDR videos of nature or even vacation videos in Davinci Resolve because the result is amazing on my 77" Oled TV.
May I ask what brand your TV is?

I really love HDR content on my Samsung (2018) & Panasonic (2017) TVs, unfortunately they are not able to play back the HDR PQ content from the R5.

I have no video editing tools or competences, but would be willing to get a program to do the necessary file conversions so my TVs could play back the files. With what codec are you exprting your HDR PQ files so your TV can read them?

Thanks alot in advance!
 
the point i'm trying to make is that shooting in log is not hard to edit. OP wants to edit not just have files from camera. people are saying basically not to use log because it's too hard. i'm arguing that it's not hard because you can just transcode to prores and use a canon lut.
It's hard precisely because of that extra step, the extra software needed and the fact that you need to learn the editing workflow for log. Anyone who is a beginner and doesn't know what log does won't have a clue on how to recover the extra latitude from log footage without learning about it. It can be avoided altogether if all one is doing is starting out. There are plenty of other things to focus on as a beginner than worry about log at this point
now you're inventing a new scenario where if they did want files straight from camera, is there still an argument for log?
Relevant scenario. The suggestions here are in context of user who is just starting out with video. It will not be the same with someone who has had some experience, even OP after he has spent some time. Context is king.
in general, it's probably not as efficient because just by the nature of the work, if it's in a realm with no editing, it's likely to not be important enough.
Precisely, and hence the suggestions in this thread. Remember, the OP is just starting out
in terms of quality though, it still will be a higher quality final video comparatively. it's not "the same" as you seem to think.
In what sense? Quality can mean a lot of things. Log does a specific thing
4k 340mbps h265 > 1080p 90mbps h264
Resolution, bit rate, codex. The last one doesn't imply quality by itself
4k downscaled > 1080p from camera
Nothing to do with log
10bit 4:2:2 > 8bit 4:2:0
Only matters during editing, or if rendering for HDR
without even getting into gamut choice, clog, or luts, you have to know you're going to end up with better quality files even after conversion.
It's precisely knowing about those things is how you make use of the quality. That takes effort
it's not like once the file gets converted it's the same. the source file significantly impacts the final compressed file.
Only if you transform it in a way where where you are making sense of that extra information. Remember all footage starts with RAW in camera. The quality you are talking about is purely a function of where you leave the final output. It takes some skill to make those extra choices
so in theory, if i really valued quality but didn't want to edit, i could shoot 4k clog and batch process all the files to smaller downscaled 1080p h265 files w/ canon lut and they would still be better than if i just shot 1080p h264 no log from camera.
Not in practice though. You will just end up with same rendering. Depending on clips, some may end up with squashed highlights or shadows, or a more linier curve than you could achieve with properly graded log footage - all the things for which people shoot log in the first place. If your theory was correct, the camera could always do the exact same thing and that would be the default. Applying a lut is a very cheap operation and can be done is almost real time. The down scaling isn't relevant to lut discussion
 
Video Response: https://streamable.com/pimphb
It's hard precisely because of that extra step, the extra software needed and the fact that you need to learn the editing workflow for log. Anyone who is a beginner and doesn't know what log does won't have a clue on how to recover the extra latitude from log footage without learning about it. It can be avoided altogether if all one is doing is starting out. There are plenty of other things to focus on as a beginner than worry about log at this point
that's like saying if you're new to photography don't shoot raw for photos because you need lightroom to edit and using the sliders is too hard.
Relevant scenario. The suggestions here are in context of user who is just starting out with video. It will not be the same with someone who has had some experience, even OP after he has spent some time. Context is king.
when i first started with video i had to learn how to do this too. it was one of the first things i learned in editing
in terms of quality though, it still will be a higher quality final video comparatively. it's not "the same" as you seem to think.
In what sense? Quality can mean a lot of things. Log does a specific thing
yeah idk if you missed it, but just enabling clog regardless if you care about using log will also automatically apply the advantages i list below. clog is not an isolated feature.
4k 340mbps h265 > 1080p 90mbps h264
Resolution, bit rate, codex. The last one doesn't imply quality by itself
h265 will be better than h264 for the same bitrate or filesize. higher bitrate with a better codec will encode better than a lower bitrate with worse codec even if they are both ending up as a 10mbps vp9 file on youtube.
4k downscaled > 1080p from camera
Nothing to do with log
yeah but you were suggesting 1080p in-camera in another post which isn't the same because in-camera it uses line-skipping to downscale compared to something like spline on your pc. also you gain color resolution and higher bitrate comparatively.
10bit 4:2:2 > 8bit 4:2:0
Only matters during editing, or if rendering for HDR
op wants to edit and 10bit 4:2:2 will still encode better. in-camera just throws away extra data where on the computer, it can avg or interpolate data to make a better 8bit file.
without even getting into gamut choice, clog, or luts, you have to know you're going to end up with better quality files even after conversion.
It's precisely knowing about those things is how you make use of the quality. That takes effort
it takes no more effort editing a raw file in lightroom. honestly everything you're saying here can be applied to photos.
it's not like once the file gets converted it's the same. the source file significantly impacts the final compressed file.
Only if you transform it in a way where where you are making sense of that extra information. Remember all footage starts with RAW in camera. The quality you are talking about is purely a function of where you leave the final output. It takes some skill to make those extra choices
ok so you think it's pointless to shoot raw photos? because it's the same thing. shooting raw then compressing to jpeg will still result in a higher quality final image compared to just shooting in-camera jpeg. higher bitrate and better codec is already undeniable and you don't have to do anything extra.
so in theory, if i really valued quality but didn't want to edit, i could shoot 4k clog and batch process all the files to smaller downscaled 1080p h265 files w/ canon lut and they would still be better than if i just shot 1080p h264 no log from camera.
Not in practice though. You will just end up with same rendering. Depending on clips, some may end up with squashed highlights or shadows, or a more linier curve than you could achieve with properly graded log footage - all the things for which people shoot log in the first place. If your theory was correct, the camera could always do the exact same thing and that would be the default. Applying a lut is a very cheap operation and can be done is almost real time. The down scaling isn't relevant to lut discussion
you will not end up with the same rendering that is just wrong. if what you were saying is true, then over the past 10 years of using RAW video with magic lantern, we would see no difference between a 1080p h264 in-camera video vs a 1080p raw processed video on youtube. even if you intentionally make them as similar as possible, you see massive differences in resolution/detail, color, DR, etc.

examples:





 
Last edited:
Anyone who is a beginner and doesn't know what log does won't have a clue on how to recover the extra latitude from log footage without learning about it.
Do you use Raw for your photos? It's a pretty daunting task to edit a Raw photo for the first time, but most learn it quickly.

I'm a super beginner with video-editing, and have never used Davinchi Resolve or any other video editor before I got my R5 a couple of months ago. Well, I've added music to some clips in Windows Video editor:)

I started learning the basics first, and then I tried CLog3. The LUTs which Canon provides get you a long way to a pleasing looking video, and just by watching a few short tutorials on YT on how to grade Canon Log-footage to your taste, it's pretty easy to get a nice looking video with little effort. It's not all about recover the last bit of DR.
 
Anyone who is a beginner and doesn't know what log does won't have a clue on how to recover the extra latitude from log footage without learning about it.
Do you use Raw for your photos? It's a pretty daunting task to edit a Raw photo for the first time, but most learn it quickly.

I'm a super beginner with video-editing, and have never used Davinchi Resolve or any other video editor before I got my R5 a couple of months ago. Well, I've added music to some clips in Windows Video editor:)

I started learning the basics first, and then I tried CLog3. The LUTs which Canon provides get you a long way to a pleasing looking video, and just by watching a few short tutorials on YT on how to grade Canon Log-footage to your taste, it's pretty easy to get a nice looking video with little effort. It's not all about recover the last bit of DR.
yup! exactly my thoughts :-)
 
Can't even open it, let alone edit it. Not that its super important to me, as I never shot video in the past anyway. I just kind of felt like, with all the talk about how nice the video can be from this camera, I might give it a shot.

But here's the thing, "IF" I'm going to shoot video with this camera, I want it to be high quality video, or else why bother ? That said, I've been really stubborn about not lowering my video quality.

So my settings have been at 4K U, 59 fps, IPB, using CLOG3. I figure I could probably drop all of this stuff down to the lowest quality, and maybe be able to open and edit video from this camera.... But if I had to do that, I'd rather just skip video altogether.

My PC is an Intel(R) Core(TM) i7-7700K CPU @ 4.20GHz 4.20 GHz, with 32GB of RAM. Its about 2 1/2 years old. Should this be enough ?

I'm not beyond using proxies, if that's what's needed, as I don't see myself doing a lot of video anyway.
I feel your pain. I fortunately have a Mac that comes with video editing software called iMovie. Its not the best for heavy pro usage but fine for guys like us. There are a few free programs out there, easy google search. Also most all paid programs will give you 30 days to try before buying. That might be a good option for you, play around with video over the course of a month and at least it will give you an idea if thats something you want to explore more in depth and worth paying for Video.

But as far as included video software, if Canon included something you probably wouldn't want it. Dpp that comes with the camera for stills is just barely OK. Its slow, archaic and just clumsy to use, but it works and gets the job done for very basic editing.
 
Anyone who is a beginner and doesn't know what log does won't have a clue on how to recover the extra latitude from log footage without learning about it.
Do you use Raw for your photos? It's a pretty daunting task to edit a Raw photo for the first time, but most learn it quickly.

I'm a super beginner with video-editing, and have never used Davinchi Resolve or any other video editor before I got my R5 a couple of months ago. Well, I've added music to some clips in Windows Video editor:)

I started learning the basics first, and then I tried CLog3. The LUTs which Canon provides get you a long way to a pleasing looking video, and just by watching a few short tutorials on YT on how to grade Canon Log-footage to your taste, it's pretty easy to get a nice looking video with little effort. It's not all about recover the last bit of DR.
Hey Tambopata

What tools have you decided to use to edit your videos? I'm sure plenty who are new to video or just purchased an R5 or similar would be greatful to know.
 
I had similar thoughts as you when I bought my R5.

I just love doing family video with it because it looks so good straight out of camera. Mostly I use 4KHQ IPB light, 8K IPB light or 4k120P. I downloaded Davinci Resolve Free version and started to learn basic editing: clipping, colors, transitions between clips, animated text intros. Plenty of tutorials on YT.

Clog poked my interest, and I quickly found out that Davinci free version won't open canon Clog, so I used Shutter Encoder to convert my Clog videos. After a few tutorials my Clog3 videos looked really nice to my eyes. I ended up with buying the studio version because I found it really bothersome to do the workaround. I already pay monthly for Photoshop, and DVR is at least a one time buy with free updates. I don't do any paid work at all, but I don't like to limit my options. Next up is to do some more testing of HDR PQ videos.
 
Hey Tambopata

What tools have you decided to use to edit your videos? I'm sure plenty who are new to video or just purchased an R5 or similar would be greatful to know.
I started out with the Free version of Davinci Resolve, and just found tutorials on YT. Davinci Resolve will cover anything you need (except opening Canon Clog/HDR PQ, you need the paid version for that), even though it seems awfully complex in the beginning.

I've found Casey Faris videos to be very good and informative, and he's really explaining everything for the beginners:
RESOLVE 17 CRASH COURSE - Davinci Resolve 17 Walkthrough [BEGINNER] - YouTube
 
Last edited:
The issue for playback (and editing) is that the Canon files are 10bit 422 (both H264 and H265 compression). Most other cameras are 8bit and 420, which most CPU's and software players can play.

So, the problem is the codec (which is higher quality than produced by most cameras). The log *gamma* is just a curve. You need to convert log to REC709 (another curve) for the clips to look good. But, that has nothing to do with whether the video can be decoded (played), which is a codec issue.

If you want high video quality, you have to invest in leaning and in software and hardware.
 
The issue for playback (and editing) is that the Canon files are 10bit 422 (both H264 and H265 compression). Most other cameras are 8bit and 420, which most CPU's and software players can play.

So, the problem is the codec (which is higher quality than produced by most cameras). The log *gamma* is just a curve. You need to convert log to REC709 (another curve) for the clips to look good. But, that has nothing to do with whether the video can be decoded (played), which is a codec issue.

If you want high video quality, you have to invest in leaning and in software and hardware.
all one needs to do is batch process their h265 files to prores files and then you're good to go. you don't need any special hardware. also shutter encoder is free.
 
The issue for playback (and editing) is that the Canon files are 10bit 422 (both H264 and H265 compression). Most other cameras are 8bit and 420, which most CPU's and software players can play.

So, the problem is the codec (which is higher quality than produced by most cameras). The log *gamma* is just a curve. You need to convert log to REC709 (another curve) for the clips to look good. But, that has nothing to do with whether the video can be decoded (played), which is a codec issue.

If you want high video quality, you have to invest in leaning and in software and hardware.
all one needs to do is batch process their h265 files to prores files and then you're good to go. you don't need any special hardware. also shutter encoder is free.
You need software that reads H265 10bit 422 clips and renders prores files - does Shutter Encoder do that? Are you sure? The point is that the H265 clips from the R5 are not the usual clips that most software can handle.

"Investment" does not mean money outlay, it also includes time spent figuring out what works and what does not, and why.
 
Last edited:
The issue for playback (and editing) is that the Canon files are 10bit 422 (both H264 and H265 compression). Most other cameras are 8bit and 420, which most CPU's and software players can play.

So, the problem is the codec (which is higher quality than produced by most cameras). The log *gamma* is just a curve. You need to convert log to REC709 (another curve) for the clips to look good. But, that has nothing to do with whether the video can be decoded (played), which is a codec issue.

If you want high video quality, you have to invest in leaning and in software and hardware.
all one needs to do is batch process their h265 files to prores files and then you're good to go. you don't need any special hardware. also shutter encoder is free.
You need software that reads H265 10bit 422 clips and renders prores files - does Shutter Encoder do that? Are you sure? The point is that the H265 clips from the R5 are not the usual clips that most software can handle.

"Investment" does not mean money outlay, it also includes time spent figuring out what works and what does not, and why.
yes that's exactly what shutter encoder does. you drag and drop your h265 files into shutter encoder, select prores, and render them. it's easy and free. all popular editors will open the prores files and everything will run smooth as butter since prores was designed to be edited.
 

Keyboard shortcuts

Back
Top