AndreyBlanco
Active member
The picture says it all...
C200: Why RAW? C200 12bits vs 8bits
- Thread starter AndreyBlanco
- Start date
puredrifting
Major Contributor
Nice comparison. We are hoping that a White Paper will be forthcoming from Canon, explaining, from an engineering and DP perspective, what exactly Cinema RAW Light is. Most of us are waiting with baited breath to see if there is some way that CRD or the plug-ins for Premiere, FCP X or if Resolve can allow you to manipulate the look parameters beyond White Balance and Gamut. It appears from what BMD is saying that Resolve 14 Beta 8 allows you to manipulate all of the Resolve CC parameters without conversion until output. Not sure about Premiere, FCP X or AVID?
Last edited:
Eric Coughlin
Veteran
I recently compared some footage I shot on the Epic-W Helium in 8K compared to Amira footage shot at the same location with the same actress in 3.2K. My conclusion was the Helium looked sharper if you like to view your footage in 200% crop. Otherwise, the Amira looked better.
In other words, showing the difference between 200% crops is a faulty way of reviewing footage, as in most cases the end viewer is not going to be viewing the footage at a 200% crop.
Let's see the 100% crop samples for a better comparison.
In other words, showing the difference between 200% crops is a faulty way of reviewing footage, as in most cases the end viewer is not going to be viewing the footage at a 200% crop.
Let's see the 100% crop samples for a better comparison.
Can you share the full frame and crops you are referring to in order to help illustrate your point?
(I don't disagree or anything, I'm just curious.)
I agree with the others. The only thing this tells me is that the RAW looks better at 200%, which is only "better" in an academic way, not a practical way.
Also it looks like the MP4 may have been shot with one of the baked in looks, and who knows what that does to the image.
Also it looks like the MP4 may have been shot with one of the baked in looks, and who knows what that does to the image.
Eric Coughlin
Veteran
Amira and Red are ungraded aside from a LUT which was already applied when I brought it into Premiere. For the C300 Mark II (shot in Canon Log 3 Cinema Gamut), the second shot of just the guy, I applied a Cinema Gamut C-Log 3 to Rec 709 LUT, along with 60 sharpness in Premiere, and for the first shot, that same LUT did not work (looked really bad, why does the LUT not work for all shots if it works for one?), so I graded it manually with Colorista color wheels.
If you right click on the images, copy the url, you should be able to view them at full resolution in a new tab in your browser.
Arri Amira 3.2K, Canon 35mm T/1.5 CN-e
Arri Amira 3.2K @ 200%, Canon 35mm T/1.5 CN-e
Red Epic-W Helium 8K, Canon 15.5-47 T/2.8
Red Epic-W Helium 8K @ 200%, Canon 15.5-47 T/2.8
Arri Amira 3.2K, Canon 35mm T/1.5 CN-e
Arri Amira 3.2K @ 200%, Canon 35mm T/1.5 CN-e
Red Epic-W Helium 8K, Canon 15.5-47 T/2.8
Red Epic-W Helium 8K @ 200%, Canon 15.5-47 T/2.8
C300 Mark II 4K, Canon 15.5-47 T/2.8
C300 Mark II 4K @ 200%, Canon 15.5-47 T/2.8
C300 Mark II 4K, Canon 15.5-47 T/2.8
C300 Mark II 4K @ 200%, Canon 15.5-47 T/2.8
Arri Amira 3.2K, Canon 15.5-47 T/2.8
Arri Amira 3.2K @ 200%, Canon 15.5-47 T/2.8
Last edited:
Really stupid answer, sorry. An end user know nothing about raw, LUT, color correction, etc.
When I'm studying a codec quality I'm resizing picture size from 100% to 400% as I need, it depends on what I want view (noise, compression artifacts, etc). Both picture above give clear answer about technical quality of codec, not about art or feeling from picture. Something else would be to judge the distribution type of compression after all processing.
Last edited:
Eric Coughlin
Veteran
An end user may not know what those are, but an end user can always tell (consciously or subconsciously) the quality of an image presented, which is affected by many factors - lighting, codec, camera, color correction, etc. Some things are more noticeable to the end user than others. If they're watching the end video at a 100% crop, then what is only visible during a 200% crop will be irrelevant to them.
I'm not saying you shouldn't pixel peep to judge image quality. I'm saying that you have to keep what the end viewer is going to see in mind. Imagine if you're deciding between a C100 or C200, or C200 RAW or C200 MP4, with the cost difference being substantial, you do all these 200% tests to confirm it's worth the extra cost, and then your clients only ever view it on their phones since you rented the camera for a mobile ap video you were producing, and the end client benefited very little from the 200% crop improved image quality.
I'm saying one has to be careful not to get too caught up in viewing a camera's footage in a way that the end user will never benefit from, and then falsely thinking the end viewer wiill benefit from that, and spending money unessarily on extra stoarge space, CFast cards, etc., because of that.
I'm not saying RAW is bad. All I really said to begin with is that a 100% crop would perhaps be a more accurate example of the image quality we're getting since a 100% crop is most likely what the end viewer is going to see.
Back with the C100 and external recorders people were posting their 200% crop images as well. You could tell a difference at 200%. At 100%, the images looked pretty identical to me. So, people spent money on external recorders, media, had bulkier and more unwieldy ergonomics on their camera, etc., in part inspired by these 200% improved images, but later upon finding there wasn't necessarily much improved real world difference in image quality, went back to doing just the internal recordings.
RAW vs MP4 is a bigger leap than the leap on the C100 when going external. But I'm just saying, let's keep things in perspective here.
Last edited:
Eric Coughlin
Veteran
I'll have to try that out. There was a topic here recently about people going back to using Canon Log 2 after using 3 for awhile, so maybe I'll try Canon Log 2 again. I mostly shoot a flavor of Wide DR, but for more important projects I've been doing log lately. I guess you get an extra stop of dynamic range out of log 2, reportedly. So, I'm still figuring out the best way to grade log footage.
I've had bad experience with LUTs; sometimes they work, sometimes they don't (as in, after applying the LUT the image will be way too constrasty, saturated, or not enough, or just far off from a proper image in general). That's one thing I like about the Amira; the post is so simple. The LUT is already applied on the Amira as metadata recognized in Premiere so no need to research which LUTs to apply, find out where to download them, then have them not work half the time like in my experience with the C300 Mark II. The Amira is also primarily shot in Log-C, so you don't have to research and test out various custom profiles to see which works best.
I was a bit surprised to need to add 60 sharpness in Premiere to the C300 Mark II footage to get a level of sharpness I liked. The Amira footage didn't take post sharpness well, and the Red footage could take some, but not a whole lot. So, I guess C-log 3, even after applying a LUT, is still desharpened (for lack of a better term) a lot, so can be sharpened in post a lot more than the other cameras. Wide DR has a much sharper look out of the gate.
I didn't really touch color on this particular image aside from applying the LUT. I'd say your correction matches less closely to the colors from the Amira of the same actor. Not saying it's bad; I don't profess to be that proficient with color grading.
Amira-user
Active member
i totally agree about your assessment about the amira, i've used it many times, and the color always seems to 'pop' for me, and its pro res codec is beefier in many ways compare to canon. With the new arri look library there are a ton of LUTS to choose from, def very user friendly on the color grading stage!
btw eric, when you shoot outdoors, do you ever go up to 10 stops of internal ND on your C300ii? i mean on the amira 7 stops seems plenty enough for me, just curious if you ever encounter a scenario shooting in florida weather that you find 10 stops necessary?
btw eric, when you shoot outdoors, do you ever go up to 10 stops of internal ND on your C300ii? i mean on the amira 7 stops seems plenty enough for me, just curious if you ever encounter a scenario shooting in florida weather that you find 10 stops necessary?
Eric Coughlin
Veteran
I didn't realize the Amira was up to 7 stops; I'd thought it was 6. I guess 7 is a bit better.
Yeah, I go up to 8 or 10 stops outdoors on occasion. I dunno, maybe once or twice a month, 10-15%% of my shoots. A lot of my shoots are indoors, though. I haven't paid too much attention as going to 10 stops is such a simple and thoughtless process. It's useful if I want to shoot shallow depth of field outdoors. I don't always want to be at f/5.6-f/8. On the C300 Mark II you can go down to ISO 160 I believe it is, while on the previous C-series cameras ISO could only go down to ISO 320, so they suffer even more from not having as much internal ND available, and the Amira goes down to ISO 200.
On the flip side, ISOs below 800 loose dynamic range in the highlights and gain it in the shadows, and when I'm in bright sunlight, more often I want more dynamic range in the highlights, so I may want to stay at ISO 800, in which case, again, the extra ND stops can be helpful to stay at ISO 800.
So on my C100s with six stops of ND sometimes I'm stuck needing to stay at say f/11 or so just to get proper exposure, assuming I don't want to bump the shutter up. Shooting on the beach with the white sand reflecting back light is one scenario where things are extra bright.
I'm curious if what Sony has done with the Fs5's electronic variable ND will become standard in cameras. From what I hear it doesn't affect image quality, and being continuously variable is quite nice. I see the Fs5 is up to seven stops; since it's electronic I wonder why it can't easily do more than seven. It's not uncommon that I find myself at less ideal ISOs (1250 and 1600) because maybe I just wanted say one stop of ND instead of two, so had to tweek ISO to get the exposure I want, so a variable ND would be quite nice.
With the Amira I'd typically have a matte box setup with cinema lenses, so adding an external ND filter isn't quite as much of a pain as with a smaller shoot using a C100 or C300 if they're not configured for a cinema setup. I own a couple matte boxes; I actually don't own external ND filters as my cameras have always had internal ones. I may have some cheap still lens NDs which I've never used which came as a free add-on in the kit with my 24-70.
Every once in a while I'll shoot with a Red on a passion project narrative shoot, and like many Red owners, they're under-supplied on accessories, so we're stuck shooting outside with no NDs whatsover, on their $20-$30k camera. Then I think, "I should buy some ND filters, so when I do unpaid Red shoots the footage will look better. Wait, this is unpaid, how am I going to profit off of these ND filters?"
I've heard people claim on high end shoots they'd rather forgo the internal NDs and use the ND of their choice externally. Why is that? I'd imagine that the internal NDs on a camera would get less reflection, and also be designed to work best with that camera, such that you'd (almost) always get better results with internal NDs than external. I suppose it's possible on a low end camera the internal NDs use low quality optics, but on a high end camera I don't imagine them skimping on the NDs.
I saw on Newsshooter that Arri is coming out with some external ND filters. I wonder what the pricing will be.
Yeah, I go up to 8 or 10 stops outdoors on occasion. I dunno, maybe once or twice a month, 10-15%% of my shoots. A lot of my shoots are indoors, though. I haven't paid too much attention as going to 10 stops is such a simple and thoughtless process. It's useful if I want to shoot shallow depth of field outdoors. I don't always want to be at f/5.6-f/8. On the C300 Mark II you can go down to ISO 160 I believe it is, while on the previous C-series cameras ISO could only go down to ISO 320, so they suffer even more from not having as much internal ND available, and the Amira goes down to ISO 200.
On the flip side, ISOs below 800 loose dynamic range in the highlights and gain it in the shadows, and when I'm in bright sunlight, more often I want more dynamic range in the highlights, so I may want to stay at ISO 800, in which case, again, the extra ND stops can be helpful to stay at ISO 800.
So on my C100s with six stops of ND sometimes I'm stuck needing to stay at say f/11 or so just to get proper exposure, assuming I don't want to bump the shutter up. Shooting on the beach with the white sand reflecting back light is one scenario where things are extra bright.
I'm curious if what Sony has done with the Fs5's electronic variable ND will become standard in cameras. From what I hear it doesn't affect image quality, and being continuously variable is quite nice. I see the Fs5 is up to seven stops; since it's electronic I wonder why it can't easily do more than seven. It's not uncommon that I find myself at less ideal ISOs (1250 and 1600) because maybe I just wanted say one stop of ND instead of two, so had to tweek ISO to get the exposure I want, so a variable ND would be quite nice.
With the Amira I'd typically have a matte box setup with cinema lenses, so adding an external ND filter isn't quite as much of a pain as with a smaller shoot using a C100 or C300 if they're not configured for a cinema setup. I own a couple matte boxes; I actually don't own external ND filters as my cameras have always had internal ones. I may have some cheap still lens NDs which I've never used which came as a free add-on in the kit with my 24-70.
Every once in a while I'll shoot with a Red on a passion project narrative shoot, and like many Red owners, they're under-supplied on accessories, so we're stuck shooting outside with no NDs whatsover, on their $20-$30k camera. Then I think, "I should buy some ND filters, so when I do unpaid Red shoots the footage will look better. Wait, this is unpaid, how am I going to profit off of these ND filters?"
I've heard people claim on high end shoots they'd rather forgo the internal NDs and use the ND of their choice externally. Why is that? I'd imagine that the internal NDs on a camera would get less reflection, and also be designed to work best with that camera, such that you'd (almost) always get better results with internal NDs than external. I suppose it's possible on a low end camera the internal NDs use low quality optics, but on a high end camera I don't imagine them skimping on the NDs.
I saw on Newsshooter that Arri is coming out with some external ND filters. I wonder what the pricing will be.
Last edited:
Amira-user
Active member
yes amira goes up to 2.1 on the internal ND. i've always wondered why the japanese cameras only go up to 6 stops and arri has always been 7 stops... must be a cultural thing lol.
you can bet your ass the arri nd filters will be at least $500 a piece if not more, its arri after all! if you want a set of good nd filters i would recommend 'NISI', they are like $250 a pop more or less and come with leather pouches; but with your canon set up i guess since you already have 10 stops you are pretty much set.
i have a theory that when you use any filters above 7 stops, you will end up with color shift and sometimes IR pollution, have you experienced that with your C300 ii?
you can bet your ass the arri nd filters will be at least $500 a piece if not more, its arri after all! if you want a set of good nd filters i would recommend 'NISI', they are like $250 a pop more or less and come with leather pouches; but with your canon set up i guess since you already have 10 stops you are pretty much set.
i have a theory that when you use any filters above 7 stops, you will end up with color shift and sometimes IR pollution, have you experienced that with your C300 ii?
NorBro
Major Contributor
This is a well-known issue with cameras that's been commonly tested. For cameras without ND filters, it's been recommended to use up to .9 (3 stops) safely, maybe a bit more, without needing any additional IR cut. But anything above may start to affect the image and you'll have to counter to get back your true colors.
Cameras with built-in ND filters probably differ as you seem to be able to get away with more, but the IR filtration might also increase as you go up (I don't know).
But in general, too much of any kind of filtration should probably be monitored.
With builtin NDs, the camera manufacturer can perform corrections in-camera.
Software, as with Canon's Peripheral Illumination Correction.