Simple colour grading > Camera colour science (Why you should learn to colour grade)

[TomTheDP]

11 hours ago, kye said:

Interesting.  It did a pretty good job then.

The things that I noticed first was the skin tones.  This is what I always look at, and is most important.  The one on the left has those nutty tan and brown hues and the one on the right has 'porange' (pink&orange) and red tones.  The hues of the skin tones are the most important thing for me in images.

I also noticed the differences in the green/cyan area top-right, which has greens on the left but none on the right.  BUT, the skin tones are so important that they're the entire ballgame if you're cutting an interview between two angles.  

A really quick and dirty way to match skin tones when they're so far apart is to just rotate the hue of the entire image.  This sounds like a brutal thing to do to an image, but it will never break the image and we are so sensitive to skin tones that what is a huge change in skin tones is imperceptible on everything else in the image.  If there is anything that goes off in the background then you can easily do a Hue-vs curve adjustment on it, which is much more prone to inaccuracies and stressing the image so it's far preferable to rotate the hue of the whole image to suit the skin tones and then do the Hue curves on the background rather than the other way around.

This assumes that you've correctly white-balanced and exposed the cameras, and are doing colour management and transforms properly of course.

Sounds like a good method. 

I didn't use the color chart in a precise way, it was held closer to the camera, probably a 6ft difference. Plus I shot the color chart about 20 minutes later. That said I'd still imagine it being closer than it was. It was a room with Cyan walls, so there is a lot of tinted light going on. I am going to try it again today and make sure I do it properly and see if it makes any difference. 

A normal shooting situation would involve more pure lighting. 

Skitones are a large reason I choose to shoot with an Alexa or at least RED. Every time I have done a mixed camera situation the colorist comments on how nice the Alexa/Red skin tones look compared to 10 bit DSLRs. 
 

[kye]

10 hours ago, TomTheDP said:

Skitones are a large reason I choose to shoot with an Alexa or at least RED. Every time I have done a mixed camera situation the colorist comments on how nice the Alexa/Red skin tones look compared to 10 bit DSLRs. 

LOL.  That can be taken two ways..   I'd suggest it's a subtle hint rather than a comment on the cameras!

In discussions with colourists where the subject comes up, opinions I've seen range from "results are limited by the quality of the footage" through to "these cameras shouldn't ever have existed"!!

[TomTheDP]

14 hours ago, kye said:

LOL.  That can be taken two ways..   I'd suggest it's a subtle hint rather than a comment on the cameras!

In discussions with colourists where the subject comes up, opinions I've seen range from "results are limited by the quality of the footage" through to "these cameras shouldn't ever have existed"!!

It is definitely a better option to shoot on 3 of the same exact model camera or shoot high res and just punch in. Though it's fairly common now to use a cinema camera and then use a cheaper camera for a B & C cam. I'll use whatever I am given lol. 

Every video I have watched comparing a cinema camera to a lower end camera they say something like they can be matched in post, but then when the person attempts to do it they can't. This doesn't bode well with productions that won't have a color house doing the color grade in the post process. 

The whole post production scene that I usually deal with is a nightmare. Now it still stands that it probably wouldn't matter what camera I shot with in terms of the directors/audience noticing the final product. I am super tempted to try to start just using an FX30 on everything despite how much I dislike the SOC look. Would make my life easier. 😅

[kye]

5 hours ago, TomTheDP said:

It is definitely a better option to shoot on 3 of the same exact model camera or shoot high res and just punch in. Though it's fairly common now to use a cinema camera and then use a cheaper camera for a B & C cam. I'll use whatever I am given lol. 

Every video I have watched comparing a cinema camera to a lower end camera they say something like they can be matched in post, but then when the person attempts to do it they can't. This doesn't bode well with productions that won't have a color house doing the color grade in the post process. 

The whole post production scene that I usually deal with is a nightmare. Now it still stands that it probably wouldn't matter what camera I shot with in terms of the directors/audience noticing the final product. I am super tempted to try to start just using an FX30 on everything despite how much I dislike the SOC look. Would make my life easier. 😅

HA!  you don't hear the colourists saying they can match the lesser cameras to the Alexa in post!

The first thing they will say is about managing your expectations.  The second thing will be about managing your expectations.  The third thing might not be about that, but also might still be....

I have no idea what your situation is, but it might not be a bad strategy to just buy a set of matching cameras and then insist on using them on every job - you'd likely lose work because you're putting conditions on things but the final quality of the projects you shoot would go up.  Then you could focus on optimising your craft rather than struggling with random camera after random camera on each job, making your work increase over time and making you more desirable...

[TomTheDP]

1 hour ago, kye said:

HA!  you don't hear the colourists saying they can match the lesser cameras to the Alexa in post!

The first thing they will say is about managing your expectations.  The second thing will be about managing your expectations.  The third thing might not be about that, but also might still be....

I have no idea what your situation is, but it might not be a bad strategy to just buy a set of matching cameras and then insist on using them on every job - you'd likely lose work because you're putting conditions on things but the final quality of the projects you shoot would go up.  Then you could focus on optimising your craft rather than struggling with random camera after random camera on each job, making your work increase over time and making you more desirable...

I don't do enough multi cam stuff for it to be worthwhile. I prefer to shoot 1 cam for narrative which is 99% of what I do at this point. 

If I were to go back a year though when I was mostly doing multi cam corporate it definitely would make more sense. It looks like this weekend I'll be using the Alexa(requested by director), a Fuji XH2S (directors personal camera), and maybe an FX30 or Pocket 4k as a C cam. We'll see how the color chart does 😅

[kye]

19 minutes ago, TomTheDP said:

I don't do enough multi cam stuff for it to be worthwhile. I prefer to shoot 1 cam for narrative which is 99% of what I do at this point. 

If I were to go back a year though when I was mostly doing multi cam corporate it definitely would make more sense. It looks like this weekend I'll be using the Alexa(requested by director), a Fuji XH2S (directors personal camera), and maybe an FX30 or Pocket 4k as a C cam. We'll see how the color chart does 😅

Actually, the XH2S and P4K are great options when combined with the LUTs/powergrades from Juan Melara:

https://juanmelara.com.au/products/bmpcc-4k-to-alexa-powergrade-and-luts
https://juanmelara.com.au/products/x-h2sfuji-to-alexa-powergrade-and-luts

Juan is a working DP and colourist, and is one of the most knowledgeable colourists online, especially when it comes to matching complex transformations.  I've bought from him before and had good experiences.  The A/B results speak for themselves too.

Here's the Fuji A/B shots, and the P4K ones match even better.

 

[TomTheDP]

16 hours ago, kye said:

Actually, the XH2S and P4K are great options when combined with the LUTs/powergrades from Juan Melara:

https://juanmelara.com.au/products/bmpcc-4k-to-alexa-powergrade-and-luts
https://juanmelara.com.au/products/x-h2sfuji-to-alexa-powergrade-and-luts

Juan is a working DP and colourist, and is one of the most knowledgeable colourists online, especially when it comes to matching complex transformations.  I've bought from him before and had good experiences.  The A/B results speak for themselves too.

Here's the Fuji A/B shots, and the P4K ones match even better.

Yeah I purchased his luts, as well as emotive color(probably more precise), and the trendy Phantom luts.

[PannySVHS]

13 hours ago, TomTheDP said:

Yeah I purchased his luts, as well as emotive color(probably more precise), and the trendy Phantom luts.

What are your takes on these? cheers

[TomTheDP]

9 hours ago, PannySVHS said:

What are your takes on these? cheers

You have to go further then simply exposing the cameras the same and having them set to the same WB. For example having say the Alexa set to 5600k and the FX30 the same in a daytime situation will yield very different results. 

[PannySVHS]

Seems like my temptation to buy any Luts is still lower than my urge to save money, even moreso after your message is rather calming any desire to purchase anything for even the most little amount of pennies. 😂 @TomTheDP

[TomTheDP]

@PannySVHS

Here is the Sony with the Alexa. I just adjusted the Hue on the sony and applied the Emotive Color ARRI Daylight lut. 

Now this is a really bad lighting situation. Sunlight, combined with Tungsten, and also bright green walls in the room lol. 

here is the raw video from both cameras as well as PNG stills of the LOG image if anyone wants to play with them. 

https://drive.google.com/drive/folders/1fYuQyBHEq9zvrIB_pQ-GBMXWOKxIhP8W?usp=sharing

[TomTheDP]

@kye

I used your hue adjustment technique as a starting point. I think that improved it a lot, no more pink/magenta from the Sony. Might have went too far though. I still have hope for the FX3 as an A cam. 😅

[kye]

6 hours ago, TomTheDP said:

You have to go further then simply exposing the cameras the same and having them set to the same WB. For example having say the Alexa set to 5600k and the FX30 the same in a daytime situation will yield very different results. 

I recommend doing a proper WB.

I've seen a bunch of side-by-side tests where they set both cameras to the same Kelvin WB and then nothing matched.  I have no idea why the cameras wouldn't all match the WB, but it looks like they never do.

I'd suggest trying a custom WB on a grey card and then trying the comparison again?

[TomTheDP]

I'll do one with a grey card next time. 

27 minutes ago, kye said:

I recommend doing a proper WB.

I've seen a bunch of side-by-side tests where they set both cameras to the same Kelvin WB and then nothing matched.  I have no idea why the cameras wouldn't all match the WB, but it looks like they never do.

I'd suggest trying a custom WB on a grey card and then trying the comparison again?

what looks glaringly off to you in the above comparison? 

[kye]

@TomTheDP

Here's a basic match.

As I don't own the LUT packs, I have just used a CST sandwich - first node converts to Davinci, second does adjustments, third converts to 709/2.4.

The Alexa has no adjustments at all, Sony has a Hue rotation for skin and Tint adjustment pushing green for neutrals.  These were literally done just by looking at the vector scope.

Added Hue curves for the green wall and the jacket:

That would be good enough for most work I'd imagine.

7 minutes ago, TomTheDP said:

I'll do one with a grey card next time. 

what looks glaringly off to you in the above comparison? 

Skintones.  Always skin tones.  It's where the viewer will be looking, and we're sensitive to very slight differences.

[kye]

By the time you put a grade over it, those appear to be fine.  This is with a 2383 PFE LUT and I adjusted the WB to be cooler for a bit of separation and reduced the saturation a little:

[BrunoCH]

On 11/7/2023 at 10:07 PM, TomTheDP said:

@PannySVHS

Here is the Sony with the Alexa. I just adjusted the Hue on the sony and applied the Emotive Color ARRI Daylight lut. 

Now this is a really bad lighting situation. Sunlight, combined with Tungsten, and also bright green walls in the room lol. 

here is the raw video from both cameras as well as PNG stills of the LOG image if anyone wants to play with them. 

https://drive.google.com/drive/folders/1fYuQyBHEq9zvrIB_pQ-GBMXWOKxIhP8W?usp=sharing

Here's my proposal for matching the two cameras. And a grade test. No LUT was used. 

[PannySVHS]

When i graded with directors at my side almost all gave great input elevating my work in a substantial way for the narrative and the quality of work itself. But it was only fun and to the good of work when they had a understanding und emphathy for the passion and quirks i am embodying sometimes when i burn for a project. When a director is not appreciative of the process it loses all its magic and the project better be paid well enough. The shots above look good enough to not to invest too much work on them.:)

[MrSMW]

2 hours ago, PannySVHS said:

When a director is not appreciative of the process it loses all its magic and the project better be paid well enough

Same for wedding clients and probably ALL clients everywhere, every thing.

Despite doing all the filtering prior to them booking and then educating from the point of booking to their event, about 1 in 20 slip the net.

And then it's always a case of a long list of impossible questions to answer after the fact.

"No, you are right, it doesn't look anything like your sisters wedding. I've never met your sister, wasn't at her wedding, and certainly didn't shoot it".

"Yes, in different light, your bridesmaids dresses will look a slightly different colour to Pantone 317884. In the results I have provided to you, those are not dresses. That is not sky. Those are not trees, or people, or anything put millions of different coloured pixels. And then do not get me started on your monitor/device screen calibration, because unlike you, I actually have a life".

Yawn...

Given the choice...and I try to ensure this situation, I'd rather take a lower paying client who gets it over a higher paying one that does not. Every time. Unless they are paying CONSIDERABLY more.

[kye]

I am reviewing my lineup and decided to compare the OG BMPCC to the GH5, and ended up filming shot-by-shot duplicates and matching them in post.  The results were surprising.  I noticed that lots of shots had differences because the vNDs on the two setups were in different orientations and polarised the scene differently, but unless you were doing a direct comparison then that wouldn't be noticeable.

GH5 was with 14-42mm F3.5-5.6 kit lens, shooting wide-open in 150Mbps UHD / HLG.  P2K was with 12-35mm F2.8, shooting wide-open RAW.  1080p timeline as always.

I graded each P2K shot and then graded the GH5 to match.

I also shot a few of my wife as well, which matched nicely, including skin tones, but I'm not allowed to share those!

The P2K has a desirable level of sharpness (being a native 1080p sensor) that is very similar to film, and I didn't need to blur the GH5 footage to match, which I suspect is due to the kit lens not being so sharp wide-open.  

Obviously the P2K has significantly more dynamic range, but it seems that when that is not required, the GH5 can do a pretty good imitation of it.

Tools required to match shots included: contrast, saturation, colour boost, WB, exposure.. plus, I had to rotate the skintones slighty magenta, the greens slightly blue, raise the shadows slightly and tint them, and darken the saturated areas significantly.  There were other tiny changes that I noticed that I didn't bother with.  This isn't "basic" colour grading but matching GH5 to the legendary P2K isn't a basic goal either.