FS7: Cine EI LUTS and Custom Scene Files

Doug Jensen

Doug Jensen

Carbonite Member
I'm doing some tests of custom Scene Files vs. Baking in LUTs in Cine EI. The idea is to find settings that will provide a "ready-for-broadcast" look right out of the camera with virtually no grading. Obvsioulsy, if you're going to grade anyway, why burn in a LUT, right? So the premise is that no grading will be done to the footage. WYSIWYG.

Before I can do the comparisons properly, I would like to get a consensus (if that is possible) on what the correct exposure should be for the LUT I have posted below. It doesn't matter right now WHAT the LUT is, I just need some opinions on which of the four exposures looks best . . . if any. I will then use that exposure in the rest of my testing. Why? Because I don't want to have to shoot a bunch of test shots at different exposures, or have someone say the tests are no good because I didn't expose the LUT right. So which one of these do you like best? Now is the chance to chime in and help. If you had to choose one of the four options as your FINAL look, which one would it be?


****** EDIT ********

I HAVE DELETED THE FIRST VIDEO THAT WAS POSTED HERE.
PLEASE SEE POST #25 IN THIS THREAD
http://www.dvxuser.com/V6/showthread.php?335147-Cine-EI-LUTS-and-Custom-Scene-Files/page3

************



They are all the same LUT, just exposed differently.

Please let me know your thoughts.

Doug
 
Last edited:
It's difficult to judge your exposure when my brain is also judging the original background pictures exposure! A real scene would be more effective for judging I think.

If I had to choose one of your levels, LT03, but my own preference would sit somewhere between LT02-03. I do tend to channel Gordon Willis though :)
 
Thanks.
FYI, I chose the chart because it is a well-known DSC chart that many professionals can identify with and they already know what it should look like.
 
Great timing. I was testing this evening...trying to set up my go-to burned-in LUT or scene profile for a shoot on Monday and will likely keep tweaking. Agree with the early consensus. But - If I had to choose one of the four options as FINAL look...
LT02
 
I've also been trying to work out a similar workflow, whilst getting my head around using Zacuto's standard FS7 LUTs in the Gratical at the same time. I would agree that LT02 looks the closest to the ideal, albeit slightly under ('exposed'). Look forward to seeing further results. Cheers, Jason
 
Thanks for the input so far, but I want to ask you all to look one more time to make sure you aren't choosing a shot that is underexposed.
I know it can be hard to judge what looks best on a chart.

FYI, the white strip is an official Kodak 90% white card and the exposures are based on that white card:

LT01 = 60%
LT02 = 70%
LT03 = 80%
LT04 = 90%

So, those of you who said LT02 looked best, are you really going to expose whites at 70%? And those of you who said that something between LT02 and LT03 would look best, that would be around 75%. Are you sure? Is that really what you would expose for? With that settting, nothing except for specular hightlights or really super whites are going to exceed 70% or 75% and skin tones will be well below 50%. That does not seem right for a final image. Personally, I think something between LT03 and LT04 would be correct, let's call it 85%.

I'm just asking one more time before I move onto the next stage of my test shots (won't be charts next time) because I feel LT02 and LT03 are both underexposed. Anyone care to change their mind or tell me why they would purposely expose white at 75% or less?
 
Last edited:
Doug Jensen said:
Thanks for the input so far, but I want to ask you all to look one more time to make sure you aren't choosing a shot that is underexposed.
I know it can be hard to judge what looks best on a chart.

FYI, the white strip is an official Kodak 90% white card and the exposures are based on that white card:

LT01 = 60%
LT02 = 70%
LT03 = 80%
LT04 = 90%

So, those of you who said LT02 looked best, are you really going to expose whites at 70%? And those of you who said that something between LT02 and LT03 would look best, that would be around 75%. Are you sure? Is that really what you would expose for? With that settting, nothing except for specular hightlights or really super whites are going to exceed 70% or 75% and skin tones will be well below 50%. That does not seem right for a final image. Personally, I think something between LT03 and LT04 would be correct, let's call it 85%.

I'm just asking one more time before I move onto the next stage of my test shots (won't be charts next time) because I feel LT02 and LT03 are both underexposed. Anyone care to change their mind or tell me why they would purposely expose white at 75% or less?
Click to expand...

Maybe the question should be, "Are you viewing on a color corrected monitor in a proper room?".

Hope you folks have a good weekend,
Robert
 
A final image should look good on many devices, from a 80" TV to an iPad or iPhone. We generally have no control over where and how someone watches our stuff. A final image should look pretty good universally.
 
Doug Jensen said:
A final image should look good on many devices, from a 80" TV to an iPad or iPhone. We generally have no control over where and how someone watches our stuff. A final image should look pretty good universally.
Click to expand...

Really surprised at that reply. I don't necessarily disagree but there has to be a standard. When judging exposure even a notalentwannabe bottomfeeder like me knows that one should be using a proper monitor and room. If you want to measure plutonium for your reactor you don't use a bathroom scale.

Most displays are way too bright and contrasty.

Robert
PS: My room is currently not dark enough but on my corrected monitor something between 3 & 4 would look right to me.
 
Doug Jensen said:
If you're going to grade anyway, why burn in a LUT, right?
Click to expand...

You could want to 'burn' to make full (or best, or at least 'better') use of the range of the codec. For example shooting low DR scenes in Slog may not make best use of the 'data bucket' but just because you have filled 'the bucket' by burning a look/lut/scene doesnt mean you would not want to grade afterwards.
 
Last edited:
BTW I think your 'scene' is 'flawed'.

I think it is most important to have a scene that has a wide dynamic range (beyond the camera DR) which usually would include a physical light source in the shot.

When testing cameras/looks I would tend to have the usual test card, black velvet, shiny watch, models face (if available) ,etc, but also point a Dedo at the wall behind to create a hotspot that will definitely blow.

There is no way the waves in your scene are actually many stops lighter than the girls as they would actually be on a beach at midday.

Abel (for example) in my experience often make profiles that go to 'rat sh_it' when used in 'the real world' as highlight handling/roll off is not fully examined
 
Last edited:
infohou said:
Really surprised at that reply. I don't necessarily disagree but there has to be a standard. When judging exposure even a notalentwannabe bottomfeeder like me knows that one should be using a proper monitor and room. If you want to measure plutonium for your reactor you don't use a bathroom scale.

Most displays are way too bright and contrasty.

Robert
PS: My room is currently not dark enough but on my corrected monitor something between 3 & 4 would look right to me.
Click to expand...

I think what Doug meant was that the image has to be articulate enough to look good universally. Many things can be done to how you film something to make it look still reasonably good on different displays, you have no control over how people watch the content and can't expect them to always sit in a dark room. He did not say it will look the same on all displays.

In the context of this discussion, exposure LT02 may look good to most people's eyes and what many would intend the final image to look like, but is not safe as a "final" image because it does not take into consideration the fact that we have no control over how it will be viewed. You could say that choosing LT02 is not playing safe.

The more I think about this discussion I realize that it is a TV/broadcast background vs Cine background of people here that makes it harder to find a consensus. TV/broadcast people usually play it safe while cine people take more risks. As mentioned in another discussion, we live in interesting times where how the boundaries between the two are defined changes rapidly. FS7 is actually very relevant in that regard, I think Sony understands the times pretty well.
 
Last edited:
morgan_moore said:
BTW I think your 'scene' is 'flawed'.
Click to expand...

Sam, you are totally missing the point of this test. The purpose is not to show the dymanic range of the camera or anything else about the quality of the picture.

The premise is simple: With this camera, with this LUT, with this shot . . . what is the best exposure if not grading will be done? Period. I want to know how other people would expose this shot. Nothing more.

I really didn't think it would be a big deal, but apparently it is. I think I'm going to just abandon the idea of the further tests and forget the whole thing. It is clear that I will never be able to do them to someone else's satisfaction.
 
Last edited:
If this is a pre test that is great.

Im just saying that IMO some of the Abel profiles in the past have not translated well to the true DR of the outside world and think this setup risks leading you down the same path.

S
 
Hey, I don't disagree with your comments about Abe's profiles. I've never liked any of them on any camera, and the funny thing is they spend a lot of time trying to make their LUTs or profiles look like cameras that are inferior to the F5/F55/FS7. I don't need a LUT that makes my camera look like a 10 year old Varicam, thank you.

I am just trying to establish what should be the BEST exposure for this particular LUT because I don’t want to have to do a dozen different exposures of a dozen different LUTs and scene files when I do some further tests.

If someone asks me, “how should I expose your scene file called _____?” I have the answer, and that exposure advice will work for all but the most extreme scenarios. So can’t a LUT be treated the same way? If the answer is no, it depends on the circumstance, then that makes that LUT much harder to use because there is no consistency in determining the exposure. And if the answer is yes, then what the f!@%$ is the number? It should be that simple.

I don't want to be the person who says what the exposure of this LUT should be. I want other people to TELL ME how they think it should be exposed so it removes my bias from the equation.
 
You must log in or register to reply here.
Back
Top