Exposing for the RAW format

[Dominic Cochran]

Threads about this subject are pretty scattered on this forum and on the web in general. Lets attempt to start a comprehensive discussion on this new sub-forum. Links to good articles, etc. would be helpful for all I'm sure.

My background is 15 years in 35/16mm film although we currently own a Sony F350 XDCam (which we've been very happy with so far btw). From what I've read my experience in film will be helpful in thinking about exposure with the Red and RAW, especially as compared to a 3 chip video system.

Looking forward to using my lightmeter again for more than measuring contrast ratios!

[Dominic Cochran]

Decent start I suppose:

http://en.wikipedia.org/wiki/Raw_image_format

[Rudi Herbert]

Dominic,

To simplify things, which is not as bad as would appear :-), here's a bit of my thinking when it comes to RAW, which I've been shooting now for about 6 years with DSLR.
- It is better to slightly under expose than over expose with it. Recovering highlights from a seemingly "dark" image is far better than dealing with blown whites.
- If you'll under expose, then it is paramount to use some fill light on the areas that will inevitable go dark. Then again, using bounce cards and boards is far easier, cheaper and more controllable than dealing with high intensity HMI's...
- To emphasise the above point, a little fill light goes a LONG wait in achieving a balanced, acceptable look when dealing with slightly under exposed RAW images.
- RAW allows you to manipulate the looks ad nauseum in post, however, there is no substitute for certain feels and moods. If you want the warm, magical glow of magic hour, then shoot at that time and you'll end up with an image that's already half way there. You can only accomplish so much in post, don't forget that.
- RAW images are, by default, low contrast and low saturation, and that's good. Now, the less you have to push them the better they'll hold up. So, if you dress your actors in bright clothes, choose saturated props, overdo the make up a bit, and in general have a richer than usual color palette before the camera, the less you'll have to degrade the image afterwards trying to achieve the same.
- Try and achieve "dimensionality" or depth separation in your images. In other words, RAW images look a LOT more filmic when the actual physical separation between the spatial layers in the frame is more pronounced. Meaning, put more distance between foreground and background, and everything in between, and you'll end up with far more clear edges that will require less sharpening in post.

and there's plenty more of course, just to get the ball bouncing :-)

[number6]

- RAW images are, by default, low contrast and low saturation, and that's good. Now, the less you have to push them the better they'll hold up. So, if you dress your actors in bright clothes, choose saturated props, overdo the make up a bit, and in general have a richer than usual color palette before the camera, the less you'll have to degrade the image afterwards trying to achieve the same.
- Try and achieve "dimensionality" or depth separation in your images. In other words, RAW images look a LOT more filmic when the actual physical separation between the spatial layers in the frame is more pronounced. Meaning, put more distance between foreground and background, and everything in between, and you'll end up with far more clear edges that will require less sharpening in post.

and there's plenty more of course, just to get the ball bouncing :-)

Interesting idea, shooting high saturation reality. I'm curious how this would work in a broun grass location and with actors wearing clothing whose color was highly saturated? Does the color contrast look phony?

[Dominic Cochran]

- It is better to slightly under expose than over expose with it. Recovering highlights from a seemingly "dark" image is far better than dealing with blown whites.


This makes sense considering my limited stills experience with DSLRs as well as the thread that mentions that the Dynamic Range of the Red is skewed more toward the shadows than say a 200T film stock.





- RAW images are, by default, low contrast and low saturation, and that's good. Now, the less you have to push them the better they'll hold up. So, if you dress your actors in bright clothes, choose saturated props, overdo the make up a bit, and in general have a richer than usual color palette before the camera, the less you'll have to degrade the image afterwards trying to achieve the same.


Good point, and something I've always tried to achieve In-Camera when shooting film, especially in a commercial workflow with a fast turn-around. I'm glad that we'll be able to push things a little more on set without having to worry about the latitude of our 3 chip camera so much.





- Try and achieve "dimensionality" or depth separation in your images. In other words, RAW images look a LOT more filmic when the actual physical separation between the spatial layers in the frame is more pronounced. Meaning, put more distance between foreground and background, and everything in between, and you'll end up with far more clear edges that will require less sharpening in post.


Good advice for all, film OR video! When space on set allows of course...

Nice post Rudi

[Dominic Cochran]

Interesting idea, shooting high saturation reality. I'm curious how this would work in a broun grass location and with actors wearing clothing whose color was highly saturated? Does the color contrast look phony?

It will be good to hear from someone with specific RAW format experience regarding this; but I've always tried to work with the wardrobe and makeup departments to achieve as much of this look as possible in camera. I assume that this will be the best approach whether shooting Film OR Raw.

As far as looking phony, I haven't had that problem with film (occasionally with 3 chip video though), unless we're going for a hyperreal look where we might embrace it of course. It will be interesting to see more tests from people shoting Red in these high contrast situations.

[Jeremy Hughes]

I thought you adjusted gain in RedCine... Are you talking about shooting at a target ISO then exposing through appeture and ND filters correctly?

[Dominic Cochran]

Yes, we're discusssing proper in camera exposure, and how to get there.

And yes, you can adjust gain, curves, and saturation levels etc. in RedCine, but setting the right stop is crucial.

[Mark Crabtree]

Yes, of course, nobody wants blown highlights. Under exposing, or exposing to the left, however, does generate noise and who wants more noise than is absolutely necessary? Many photogs use a technique called exposing to the right, over exposing according to the display yet still guarding against blown highlights. The reason you can over expose, according to the monitor, and still not blow out the highlights, is that the monitoring system is based not on the RAW file but on a converted encoded file which does not have the dynamic range of RAW. So, in other words, the monitor clips before the sensor does and the only way to really know where that clip point is is to shoot tests at increasing exposures or ISOs. The results can only be viewed in RED Cine or RED Alert. Exposing to the right is, yes, dangerous because you are shooting on the edge of clipping the highlights, however, you acheive a picture with lower noise. Playing it safe and using ISO 320 will probably keep you out of trouble regarding clipping but you will not get the best from the digital sensor. Sometimes exposing to the left means shooting a scene at more than a full stop under exposed. If you increase the fill to compensate for things going dark, you can end up with a flat noisey image. The goal for us shooting with the Red camera is the same goal as for the photogs who shoot with DSLRs, to make great low noise images. And to do that we must test and know the limits of the camera so we can exploit them.

[Hans von Sonntag]

Hi,

IMHO I never found digital pictures phoney. Some how different to film for sure, but not phoney. They might look phoney before grading due to the flat gamma. But graded properly skin tone and such come out nicely. When shooting digital I never use extra colourful make up or costumes. I even do not change lighting that much. I do care about higlights as I do care about shadows when shooting negative. BTW flat telecined negative does look "phoney" as well.

My two cents,

Hans

[number6]

Hi,

IMHO I never found digital pictures phoney. Some how different to film for sure, but not phoney. They might look phoney before grading due to the flat gamma. But graded properly skin tone and such come out nicely. When shooting digital I never use extra colourful make up or costumes. I even do not change lighting that much. I do care about higlights as I do care about shadows when shooting negative. BTW flat telecined negative does look "phoney" as well.

My two cents,

Hans

Hans, substitute "contrived" in place of phoney in my post.

Hope that makes it more palatable.:whistling:

[Hans von Sonntag]

Hans, substitute "contrived" in place of phoney in my post.

Hope that makes it more palatable.:whistling:

Hi

Indeed, it does. :)

Hans

[Hans von Sonntag]

Just to put some thoughts on this:

1. Protecting highlights in a sequence can be impossible. Just imagine a pan from shadow to a backlit sunny area. Highlights will burn out. With negative you have a smooth, pleasing transition from light grey to pure, overexposed white whereas with digital a more harsh, edgy, sometimes over saturated, less pleasing transition will occur. With the latest Sony F900R / F23 they manage this quite well with build in circuits that emulate a negative/film like shoulder. Not really there, but a big improvement though.

2. Shooting RAW with RED just means shooting digital with a high range (11 Stops/66db) barely processed with a third of data required compared to YUV/RGB. Hence you can store picture data with a high fidelity on a comparable small device. This is pretty clever. Downside: you cannot use the data instantly - bad for the ENG folks.

3. REDCIDE does what a traditional video camera is doing inside its body with a lot of circuits. The downside of this conventional workflow: Treating every shot individually is is hardly to achieve, ending up shooting as flat as possible (gamma wise, not lighting wise). Positive for a lot of folks: YSWYG, instant access to the footage , etc...

4. It will be interesting to see how REDCINE can manipulate blown out highlights and make them looking pleasing.

Shooting RAW seems to me nothing special in terms of costumes, make up, etc. I find it cumbersomely to fight possible artificiality with contrived costumes, make up etc. IMHO not the way to make a picture looking less contrived.

Slightly underexposing seems the way to go but this is common sense with any other high end digital camera. Like telecine in film REDCINE or any other CC tool that can read RAW is the bottleneck where all has to go through: To the better or the worse.

Hans

[Dominic Cochran]

Shooting RAW seems to me nothing special in terms of costumes, make up, etc. I find it cumbersomely to fight possible artificiality with contrived costumes, make up etc. IMHO not the way to make a picture looking less contrived.



Hans

Depends on what you mean by nothing special. I think it's pretty crucial to get much of the look on set in terms of production design, makeup, wardrobe, (and obviously lighting). Most professionals will agree. You really can't underestimate the importance of these departments.

[Hans von Sonntag]

Depends on what you mean by nothing special. I think it's pretty crucial to get much of the look on set in terms of production design, makeup, wardrobe, (and obviously lighting). Most professionals will agree. You really can't underestimate the importance of these departments.

Hi Dominic,

no, I am not underestimating any art department. Crucial for story, style, etc. All I am saying is that shooting RAW does not mean that the work of these departments has to be modified in some way just because one shoots RAW.

Regards,

Hans

[Billy Summers]

Lighting and Exposure are pretty IMPORTANT...

[Emmanuel Cambier]

Just to put some thoughts on this:


4. It will be interesting to see how REDCINE can manipulate blown out highlights and make them looking pleasing.



Excuse me Hans, but I Wouldn't expect RedCine to perform miracles in this domain.
Blown Highlights in digital are just that… blown highlights.

The only miracle here is how low noise the camera is, and that it records Raw.
From there you can underexose to protect highlights and then raise your levels until noise creeps in.

Emmanuel

[Floris Liesker]

Excuse me Hans, but I Wouldn't expect RedCine to perform miracles in this domain.
Blown Highlights in digital are just that… blown highlights.

The only miracle here is how low noise the camera is, and that it records Raw.
From there you can underexose to protect highlights and then raise your levels until noise creeps in.

Emmanuel

Yes indeed, it's far better to avoid blown out highlights than to try and fix them in post. Redcine does have a slider for that called 'highlight' which does what the DRX did in Red Alert; it will look at non-clipped channels and compose some info out of that. I find it hard to get it really good but it's something.

I am working on a noise reduction workspace within Combustion to be able to use more of the low end of the sensor without seeing too much noise. It works in the same manner as noise reduction in telecine does, making noise lower and slower-moving without affecting resolution. Maybe there's already a plug in that does exactly that but I dont know it.
I try to make it work in FCP2 too but I'm not very familiar with those yet.
Maybe Graeme already developed such a plug-in?

[Steve Sanacore]

Just to put some thoughts on this:

Slightly underexposing seems the way to go but this is common sense with any other high end digital camera. Like telecine in film REDCINE or any other CC tool that can read RAW is the bottleneck where all has to go through: To the better or the worse.

Hans


Underexposing is the most common mistake digital photographers seem to make when starting out. There are a few great books out there on shooting raw and processing the files for DSLRs. There is very specific information on the raw image data and for best dynamic range and color you must expose to the right without over exposing. If you play it safe by underexposing you will give up a lot of color, dynamic range and have more noise in all the shadows when you try and recover shadow detail. According to the experts who write the software that converts raw files in Photoshop, "50% of the image data lies in the brightest F stop captured" (hope I my memory serves me right). So by underexposing only 1 stop - you may give up half of your potential image data, dynamic range and colors.

An accurate histogram is the best way to judge exposure. It is the only way to see exactly where the highlights and shadows will clip. On most DSLR's we can view this in Luminance and for each RGB color. Highlights can usually be recovered even if only one color channel has detail. I just hope that RED's raw converter is on par with what Adobe offers for most pro DSLR's - it is amazing what we do can with raw files. Once you learn how to process them, you will be astounded by the results.

http://www.luminous-landscape.com/tutorials/expose-right.shtml

Here is a link to one of my favorite web sites on the subject of shooting RAW.

I am curious as to how Redcine will work. Can we optimize the raw processing when we make our QT files for editing? And does it retain that information in a sidecar file for each clip when we return to the raw files to output a high res version for the colorist to finish?

[Rob Lohman]

An accurate histogram is the best way to judge exposure. It is the only way to see exactly where the highlights and shadows will clip

Make sure you have RGB histograms (so you can see individual channel clipping). This is definitely not the only way and it can't always show you where the clipping is happening exactly. Our camera also comes with a false color exposure meter which can tell you where exactly something is clipping.


I am curious as to how Redcine will work. Can we optimize the raw processing when we make our QT files for editing? And does it retain that information in a sidecar file for each clip when we return to the raw files to output a high res version for the colorist to finish?

I'm not sure I'm following 100%. REDCINE (as does RED ALERT & QT codec) work directly with the RAW data. Changes parameters will work directly on this data, not on a smaller range (through the QT codec this is only possible at the moment by changing settings in RED ALERT and then re-generating the QuickTime files).

REDCINE can save / load sessions that will retain whatever parameters you set for the clips in that session.

[Philip Powell]

Is it true, as was stated earlier on the thread that the Raw format (and the Red by implication) reallly records low saturation?

As I thought I understood from Grame, it only appears to be low sat. and that it's actually recording a wide range of saturation levels.

Because I would think, if not, if you were boosting low sat levels in post, it would make it very noisy.

[David Mullen ASC]

Most or all digital cameras shoot with a somewhat muted color level until it is boosted in post or by the color processor in the camera, as an HD camera does when recording a REC709 image. The REC709 color specs involve boosting some of the color channels, such as red, so that it looks fairly saturated on an HD monitor. Turn REC709 off in an HD camera and you'll get a more muted image, color-wise. But the color range is not being limited just because it looks more pastel. In fact, the REC709 color space, though it looks more saturated, involves throwing away some color information to fit within the color space that a monitor can handle.

[I Bloom]

Is it true, as was stated earlier on the thread that the Raw format (and the Red by implication) reallly records low saturation?

As I thought I understood from Grame, it only appears to be low sat. and that it's actually recording a wide range of saturation levels.

Because I would think, if not, if you were boosting low sat levels in post, it would make it very noisy.

Well in RGB, saturation is determined by the ratio values between red green and blue. Low contrast images are by nature low saturation. So simply by adding contrast to a RED image, you are adding saturation overall. You don't neccessarily need to increase the actual saturation value in REDcine. Though you might, and noise isn't a problem when adding a small amount of saturation.

IBloom

[Bill Anderson]

Sir, if you shot a subject that had low color saturation and you wished to boost that in post, then you might take a little quality hit (negligible even). But if the subject was vivid, then all you're basically doing is re constituting that value from the RAW data.
And I think these are two fairly different things. I've never had a RAW equivalent that failed to meet the saturation level of the subject. I think I've had more problems with over saturation and luminosity extremes than less. Of course much of this depends on to what lengths I go to reconstitute a value- but let me just say, there's no better way than RAW. (Yet)

When you shoot for a file that is not RAW, then its info is converted
to a fairly limited set of values as far as allowing you to reinterpret the color and luminosity, etc.. Just like a transparency, everything is baked in so that any bold re-interpretations will come at a fairly high price as far as quality is considered.
The RAW file resembles a film neg in that when you hold the neg up for viewing it really doesn't resemble the scene at all. But once an appropriate paper and developer (RAW conversion method) have been deployed, you will see all the saturation and luminosity that you've so carefully planned for. Not only that but what you see will still be quite open to interpretation- (within limits).
When representing an image: Tiffs are color swatches- RAW is the cans of paint. More or less.

[Bill Anderson]

Of course, there's no free lunch and all of the conversion and workspaces are to some degree destructive, fitting square pegs into round holes. However, once we've carefully sifted through our square peg pile and found the best fit:RAW, there's little point in bitching and theorising any farther. It's time to start hammering. At least until someone like Graeme Nattress invents another peg, or hole.

[Gregor Hagey]

Shooting RAW is new to me and it opens up a lot of creative and technical opportunities. I shot some stouffer step wedges at different T Stops and processed them at different ISOs (in Red Alert). My conclusion is that processing the .r3d in Red Alert at 320 ISO with -.40 exposure gives you the most latitude. You can apply a slight inverted 'S' curve to pull more out of the shadows and highlights, but the resulting image is very low contrast which is great for exporting for colour grading, but not a good look for editorial. When applying a slight inverted ‘S’ curve by eye I can read 24 steps, but by a waveform monitor 27 steps are definitely visible which equals 9 stops of dynamic range.

Upon analysing the steps you will notice that they don’t change on a logarithmic scale. The steps should be in 1/3 stop increments, but at the knee the first two steps are expanded and cover an entire stop each and at the toe the steps are compressed and cover only 1/10 stop each. Just like other video cameras there is plenty of information in the shadows, but it’s important to protect your highlights. When I read the Stouffer on a waveform I see there are actually only 10 steps (3 1/3 stops) above 50% and a whopping 17 steps (5 2/3 stops) below 50%. For the purpose of capturing the most dynamic range placing middle grey at 30-35 (instead of 50-55) on your zebras is the best way to go. When I set the camera to 320 ISO (archaically called ASA in the menu) and expose for 30-35% zebras on the viewfinder and then check the T Stop against my incident meter I actually end up with 320 ISO. It seems Red has already built this concept into the ISO settings on the camera. If I want to expose middle grey for 50-55% zebras I need to overexpose by 2 stops. This results in a cleaner image, but you risk clipping the highlights and reducing your overall dynamic range.

You can view the files here:

http://www.dghagey.com/Director_of_Photography/Red.html

[Steve Sanacore]

Gregor;

Have you ever tested Canon or Nikon Pro DSLR's in the same manor as this test? I would love to know the differences in the results. Some of your results here seem to conflict with my experiences with Canon raw files.

I think we are going to need someone from RED to decipher this information and explain the details to us. If RED placed more image information in the bottom half of the exposure then I think this is directly opposite of what Canon and Nikon have done.

I just want to make sure I know the best way to set exposure. I have not yet used Red Alert or Red Cine, so maybe this isn't as complicated as it sounds?

Thanks for the great tests and explanations Gregor.

[Gregor Hagey]

Gregor;

Have you ever tested Canon or Nikon Pro DSLR's in the same manor as this test? I would love to know the differences in the results. Some of your results here seem to conflict with my experiences with Canon raw files.

No I haven't tested DSLRs at all. This is the testing I would run for HD, but I don't know if it's appropriate to use a waveform/vectorscope to analyse stills work because a print is your finished format.

In general I set Red to 320 and meter at 320 ISO, but if I have complete control of highlights I keep Red set to 320 ISO and set my meter to 160 or 100 ISO to get a cleaner image.

[Esmaile Neissi]

Wooow

[Benjamin Rowland]

Would like to hear someone from the red team weigh in on Gregor's test results. Interesting stuff...

[Dj Joofa]

Well in RGB, saturation is determined by the ratio values between red green and blue. Low contrast images are by nature low saturation. So simply by adding contrast to a RED image, you are adding saturation overall. You don't neccessarily need to increase the actual saturation value in REDcine. Though you might, and noise isn't a problem when adding a small amount of saturation.

IBloom

Saturation is not directly the ratio of combinations of R,G, B to each other. It is measured as the length of "outwardness" of the chroma vector. In (device-dependent) RGB space, fully saturated R would be [1,0,0], and a blue green would be [0, 1, 1], hence, it is not directly a ratio.

[lordtangent]

My experience form shooting DSLR RAW for quite some time now is that you want to expose as far to the right of the histogram as possible. Even though digital cameras are much more linear in the shadow areas compared to film, the shadow area is where the noise is. Or rather, the noise is more obvious there. It's the random base noise of the sensor but since the "signal" (you image) part is weak compared to the "noise" in the dark areas what you get when you boost the dark areas is more visible noise. So really you want to push the "meat" part of your exposure as high above the noise as possible to get the best picture. (i.e. "expose to the right")

The problem is that doesn't leave you a lot of room to protect for highlights. It's really not practical to underexpose enough to protect for spec hits and direct light hits anyway. Film can only handle gross overexposure to a limited extant as well. But at least over exposed areas don't look as ugly on negative film as digital due to the more organic roll-off of film.

One solution would be if Red switched to a LOG CMOS for the image sensor. There is a technology called LinLog that combines traditional LOG CMOS with programmable roll-off and an instantaneous shutter. It would fix two problems at once! (Rolling shutter and hard clipping highlights) What log CMOS does is add a little analog amp (more like a valve) to each pixel. The voltage of the pixel can be fed into the "valve" and be used to create a analog fall-off on the pixel. The curve is programmable on LINLOG, so part of the exposure can be linear (the meat) and the top can be programmed to roll off. The coolest part is it all happens before the A to D step, so the digital circuits don't even need to know anything about it. (THough it'd probably be a good idea to keep some meta data about what roll-off curve you recorded with) I wish I could get this on my DSLR. It solves what I consider the last remaining curse of digital acquisition.

Here is some LinLog info:

http://www.photonfocus.com/html/eng/cmos/linlog.php

[combatentropy]

Veterans of the 35mm negative know to protect the shadows. This is because after a certain point of darkness no emulsion is left on the negative -- i.e., no texture. Meanwhile, there is a great degree of latitude in the highlights.

With digital, you need to drop that habit, er, like a bad habit.

A key word of the medium should tell you that its rules may not apply: negative. That's right, while film is a negative reflection of what you expose, an electronic sensor is a positive. And as it turns out, the rules for exposure are the opposite.

After a certain point of overexposure, an electronic image clips. And just like when no emulsion is left on a negative for the dark areas in a scene, no variation is left in the electronic signal for the bright areas of the scene.


According to the experts who write the software that converts raw files in Photoshop, "50% of the image data lies in the brightest F stop captured"

That is technically true, yes. (http://www.normankoren.com/digital_tonality.html) But gamma and the ultimate conversion to 8-bit from 12-bit help a great deal in evening it out.

The fact is, digital images hold up quite well to boosting the shadows:

http://www.clarkvision.com/imagedetail/dynamicrange2/

[Dj Joofa]

( But gamma and the ultimate conversion to 8-bit from 12-bit help a great deal in evening it out.

The fact is, digital images hold up quite well to boosting the shadows:



Actually, the process here is a bit destructive for a sensor that is matched in bit-depth to number of stops it can offer. Firstly, it will *not* create more information in the darker areas (shadows if you will), which it should have ideally, and will only adjust whatever information it has available to the right gamma level. For e.g., Red One has a 12-bit ADC and about 11 stops to offer, and hence, about over 70% percent information is thrown away in the range from standard reference black to standard reference white, under certain assumptions.

Therefore, the situation is a little bit tricky in shadows. On the other hand, it is fine to throw away extra information in the brighter stops as along as you don't intend to do tone mapping later, say during color correction.

Some readers may be perplexed at my figure of 70% information being thrown away when the digital camera images they acquire look very good. Well the reason is that most of the discarded (not captured) information is in the darker stops, where as a lot of time people are not operating in that range, say shooting outdoors during the day, or reasonably-lit interiors. In the range of mid tones to whites the information loss is less perceptive, as the distribution of discarded information is less dense. However, under low illumination the loss in the darker stops shows up as posterization, which needs to be reversed.

[lordtangent]

The fact is, digital images hold up quite well to boosting the shadows:


That's true... but as I mentioned even when underexposing it's tough to protect for gross over exposure anyway (spec hits and other stuff that tends to look "clippy" when shooting digital) Putting your exposure to the left of the histogram to gain headroom is a bad trade-off unless you are willing to accept the noise you'll get as a result. The Red One sensor is pretty clean, so it might be possible to trade off some noise to get more head room and still get an acceptable result. But I doubt you will ever be able to get enough headroom to make any difference on grossly over exposed stuff anyway. (i.e. the things you ARE NOT exposing for. Not the picture but stuff like spec hits etc) I know I've never been able to really emulate film-like roll off that way with my Canon 20D though. (Sensor is not clean enough to "under expose" enough to leave headroom for that stuff)

Digital exposure is essentially linear. Reducing exposure by one or two stop in digital only buys you one two stops of headroom (not much at all when talking abut stuff like spec hits and other grossly over exposed areas). Noise is linear also but PICTURE, is gamma corrected. A one or two stop under-exposure usually increases the apparent noise considerably. the reason is because when you make "picture" from that linear data you are pushing up the noise by a power function in the process.

IMHO, the best approach is to try to put the "meat" of the exposure as far to the right as possible. The spec hits are gonna clip anyway. Don't worry about shadows at all. With digital the toe is very linear so you will get detail there. As long as you don't push it up too much the noise will stay buried down there and you wont see it too much. But your goal should be to get the cleanest mid range exposure ( without over exposing of course). You want as much "signal" as possible over the noise of the sensor.

My dream is a digital sensor with some sort of roll-off to help take the "curse" off of over exposed areas. As I've mentioned several times LinLog would be a solution but no camera manufacturer seems interested in putting it in a production camera.

[Dj Joofa]

LordTangent, I think you are mixing up a few things. Firstly, when you say putting up "the exposure as far to the right as possible", there are three gross ways of doing it:

(1) Increase the size of the aperture
(2) Increase the exposure time
(3) Boost up analog gain

Under additive noise model, (3) will always increase the noise (say your camera increased the ISO rating up by boosting analog gain), and hence, the assertion "the noise will stay buried down there", shall not be correct.

For (1) and (2), if the noise statistics show a tendency to converge to a fixed number given sufficient time, then it can help, and you can move your exposure to the right using (1) or (2). However, (1) is normally fixed by dof and other requirements, and (2) is bounded by the frame rate and the amount of motion blur you want to tolerate.

In addition, in reference to your comment on LinLog, non-linear CMOSes are being investigated for some time. They have both advantages and disadvantages compared to a linear CMOS.

[mikeburton]

Here is an interesting article that stu maschwitz wrote a few days ago. Interesting read and some good information in this write up,
http://prolost.blogspot.com/2008/02/digital-cinema-dynamic-range.html

[Dj Joofa]

Here is an interesting article that stu maschwitz wrote a few days ago. Interesting read and some good information in this write up,
http://prolost.blogspot.com/2008/02/digital-cinema-dynamic-range.html

Stu's article is very interesting and informative. I could be wrong but this is my take on that issue: The last time when I read it appeared to me that Stu was taking the notion of the implementation of a particular manufacturer's implementation of Rec 709 space and how they treated the white point as a guide to expose for the log space. Graeme Nattress has already pointed out that the implementation of white point for the Red Camera is done by a 0.4 EV boost up, and hence, to "see" that extra highlight in Rec 709 that one sees in Red Log one should use a -0.4 EV value in software.

[Mitch Deoudes]

(1) Increase the size of the aperture
(2) Increase the exposure time
(3) Boost up analog gain


What about (4) Add lights / remove ND? As you noted, 1 & 3 aren't really options in this case, and changing the shutter angle isn't the first thing to try.

[Dj Joofa]

What about (4) Add lights / remove ND? As you noted, 1 & 3 aren't really options in this case, and changing the shutter angle isn't the first thing to try.

Yes, you are right about (4). I was not thinking about it. I thought LorgTangent was aiming for a fixed lighting and only going for an in-camera adjustment.

But, again, your point is well-taken, (4) should be there.

[lordtangent]

I mean "expose to the right" And that implies #4 given there is no play left in factors 1 and 2. Use more light or less ND if that's what it takes. That's what you would do with slow film in the same situation. Boosting the gain (#3) is exactly NOT what I'm talking about.

The "LinLog" type of nonlinear CMOS specifically fixes the main problems of the older style "log" CMOS (the fixed pattern noise and rolling shutter being the two worst problems) And the roll-off is programmable. You could even use it as a normal linear sensor if you wanted to.

The beauty of LinLog is that it protects the highlights without requiring that you under expose to the point that the "meat" of the image is down in the noise. And the highlights will have a nice "logy" rolloff. The images produced should be very filmic as a result.

Of course, fabricating a 12 megapixel LinLog would require a LOT of transistors. But the total is still less than a modern CPU. It should be do-able. I'm sure there are fabs that could do it.

[Dj Joofa]

Sorry LordTangent, I did not read you correctly.

I read the information on LinLog page that you mentioned. It seems interesting with the programmable option of switching between the linear and log region.

[donatello b]

another interesting post by Stu on exposing to right or to left

http://prolost.blogspot.com/2008/03/exposing-to-left-vs-exposing-to-right.html

seems to come down to experience , knowing your camera, knowing what you want , adding in light/fill or whatever it takes or not adding anything and YOU deciding which exposure is best for the shot -
to the right or to the left = you decide ...

[lordtangent]

Stu is completely right. But (and I'm biased here so bear with me) he seems to be espousing the expose to the right philosophy without realizing it.

If the goal is to get the cleanest shot possible without over exposing it implies "exposing to the right". (Even when the histogram doesn't show a distinct bump at the right. )

If you don't care at all about noise then put the exposure anywhere you want.

[Tai Wah Lim]

While Stu summary is valid, what about this logic on exposing to the right for maximum S/N ratio - http://www.luminous-landscape.com/tutorials/expose-right.shtml

[Dj Joofa]

While Stu summary is valid, what about this logic on exposing to the right for maximum S/N ratio - http://www.luminous-landscape.com/tutorials/expose-right.shtml

Stu does not talk about the statistics of the noise!

Typically with digital sensors the S/N improves by "exposing to the right" because the statistics of temporal noise make it converge to a fixed number given sufficient time.

Hence, noise which appears random in shadows, "smoothes" out with in increase in exposure value (EV), whether, brought by an increased aperture size, or by larger integration time.

[Gregor Hagey]

Great post by Stu. In the Autumn I came to some similar conclusions about getting the most dynamic range out of Red. By setting Red to 320 ISO and setting middle grey at 30-35% (instead of 45-50%) on the waveform and Caucasian skin tone to 40-45% (instead of 60-65%) you'll capture the best dynamic range.

I realize now though that if the scene has low contrast you should overexpose by up to 2 stops to get the least amount of noise. Essentially I agree that exposing to right is the way to go. Move your histogram as far to the right as you can without clipping and you can create the 'mood' later when the .r3d's are processed (with the cleanest image possible). The main problem I find is that the REC709 monitoring is quite poor. I've desperately tried to get a DSC ChromaDuMonde chart to line up with the Vectorscope boxes from the 720p HD SDI out and couldn't. The HD SDI out is deficient.

[David Mullen ASC]

Great post by Stu. In the Autumn I came to some similar conclusions about getting the most dynamic range out of Red. By setting Red to 320 ISO and setting middle grey at 30-35% (instead of 45-50%) on the waveform and Caucasian skin tone to 40-45% (instead of 60-65%) you'll capture the best dynamic range.

Isn't that the same thing as rating the camera at 640 ASA?

[Steve Sanacore]

I am getting confused.... I have been shooting with Canon and Nikon DSLRs for years, exposing perfectly (most of the time ;).

What I don't understand is how you can rate the ISO from 100 to whatever, but it doesn't really change the sensitivity? So this means you are grossly underexposing and then compensating in the post software?

If this is the case, what good is the histogram at anything other than the cameras base ISO? If you set the ISO to 320 and 'expose to the right' then you are really like 1.5 stops under and your highlights are really way down in the middle, am I right?

[lordtangent]

That is a good question that makes me wonder: What is the histogram a histogram of? Is it the actual values form the sensor or the tone mapped values? And how is it relateded to the ISO set on the Red One? (I'm not an owner and I don't have a manual to look at)

[Hans von Sonntag]

I've posted in the beginning of this thread without the experience shooting with RED. Now I've done a project under real world circumstances and that's what I find:

1. The histogram one sees on the camera display is based upon the rec 709 LUT which is created in the camera - definately not from the actual values of the sensor.

2. 320 ASA at 5000K seems the way to go, after evaluating the footage in REDCine.

3. In doubt, closing the aperture about half a stop is not doing a harm, on the contrary, essentially exposing for 500 ASA.

4. On very flat lit scenes it may be apropriate to expose in a 250 ASA direction, although noise in such scenes is not an issue IMO.

5. Noise is visible and not really avoidable, at least if there are a lot of shadows and strong highlights (high contrast).

6. I don't find RED's noise distracting and surely less than found on a f900R for instance.

7. We never used the monitor for exposure, always a lightmeter as it were
film.

8. The monitor maybe usefull some day if one is used to it's shortcomings and knows how to interpretated the picture exposure wise. Otherwise a LUT that mimics the sensors values would be great.

Hans

[Dj Joofa]

I've posted in the beginning of this thread without the experience shooting with RED. Now I've done a project under real world circumstances and that's what I find:

1. The histogram one sees on the camera display is based upon the rec 709 LUT which is created in the camera - definately not from the actual values of the sensor.


You may not want to see actual sensor values. They don't necessarily correspond to any standard color space and need to be adjusted to any appropriate color space for viewing. Given two different digital sensors viewing the same scene the internal signals generated may be different and LUT-type mappings map these different signals to the same output signal for display on the same chosen display.

Every display has a notion of primaries and the white point and the sensor signal has to be mapped to it.



7. We never used the monitor for exposure, always a lightmeter as it were
film.


That may work for certain (perhaps the default ISO the camera is rated at). However, I do not know if there is always a 1-1 correspondence between metering in the film world and digital world, especially considering that if you are not operating at the camera default ISO, then its the software that is "simulating" the behavior of an intended ISO.

[Michael Hastings]

Isn't that the same thing as rating the camera at 640 ASA?

Seems like it to me. But then you wouldn't be able to confuse people like they are with the focal length of S35 and full frame SLR.

In other words you have done the logical thing which is rate the camera at an ISO that allows you set the middle grey like you have traditionally done.

Whereas if you leave it at 320 and change the standard way of doing things we can add one more level of confusion.

1) Measure distance to subject with metric tape measure then convert meters to feet. Set your feet marked lens. (Do the opposite if you are European.)

2) Set ISO to 320 and then expose to a non standard grey level.

3) Take a still photo of the scene with a full frame DSLR then divide by 1.6 to get the proper focal length to use on S35/RED.

This is called entropy: the tendency of the universe toward chaos and disorder.

Hopefully you all appreciate my wry sense of humor.:wink: :sarcasm::biggrin:

[dylancarter]

I had an established DP shoot with the RED last week. They really pushed the camera to see how many stops they could get out of it. The scene in question has a single light source and about the room has about 20 stops. Now they know that they can't capture the entire 20 stops, but they want to see what they can get.

Unfortunately the DP was not very familiar with the camera and put a lot of faith in the hand of the DIT that came from the camera rental house. As the DP tells it, the DIT just kept insisting that they use the FALSE COLOR mode. As long as things stayed the right shade of blue, the shot was fine.

Well the shot was not fine (thank god it was a test). The blacks were all super lifted, and the whites were totally clipped.

After transferring the shots to both REC709 and REDlog as DPX and then taking those into a Color Correct bay - it was obvious the whites were long gone. Just a huge white hole.

So the DP asked me "What do I look at on the camera to make sure this does not happen?" From the discussion on this thread, it seems like the LCD image is a very rough guide, and that the histogram is what should be trusted..the most? The DP and VFX super were both uncomfortable with not having a good sense of WYSIWYG. What else can I tell them or recommend?

[Steve Sanacore]

7. We never used the monitor for exposure, always a lightmeter as it were
film.

Hans

I have never used a lightmeter from the day I started shooting digital on my still shoots. The most accurate way to set exposure is with a histogram that works, and experience. A meter will get you close, but a correctly calibrated histogram should let you put the highlights and shadows exactly where you want them. The midtones are set in post.

I agree a monitor will not give you critical exposure info - but a histogram will, and an RGB histogram is even better. Not sure if RED has that yet.

At least that's how I've done it for many years with Nikon and Canon raw files.

[Hans von Sonntag]

At least one has to start with something. Since the picture on the monitor does not show what the sensor "feels" I can only rely on what I've learned and used before and this is a light meter. I found it working nicely.

Using other ASA ratings need probably (I said probably because I don't know it yet) quite a bit of experience with this particular sensor. One has to count in: Contrast of the scene, colour temperature (the sensor is unfortunately balanced on 5000K), oversampling rate (a given nosie is very different visible in 4K or PAL) and many more. Quite complex I find.

Hans

[Hans von Sonntag]

I have never used a lightmeter from the day I started shooting digital on my still shoots. The most accurate way to set exposure is with a histogram that works, and experience. A meter will get you close, but a correctly calibrated histogram should let you put the highlights and shadows exactly where you want them. The midtones are set in post.

At least that's how I've done it for many years with Nikon and Canon raw files.

If you get WYSWYG a histogram is the way to go probably. Since the RED's histgram is off a LUT, just an exposure friendly interpretation (with some in-build headroom for the whites) of the sensor, I find a light meter more accurate, at least now.

Hans

[David Mullen ASC]

Well, I doubt there would be a situation where a highlight would be clipped on the RAW recording but visible (not clipped) on the monitor, so if they see an important area clipping on the monitor, they should worry. I would think that lighting for the dynamic range visible on the Rec 709 monitor would mean that you have a little more range beyond that on the RAW recording.

In other words, if it looks like crap on the monitor, you should stop and think... rather than just trusting that it's all there on the RAW recording. But if it's all there on the monitor, it's definitely all there on the RAW recording...

I don't see a problem with using extreme lighting as long as you are aware of how it will turn out.

Common sense should have also told the people doing the test that an object that is more than three-stops overexposed is likely to show clipping problems.

As for using a light meter and not a monitor, that's probably fine if you've done tests and are taking spot meter readings of bright areas when you aren't sure they will be clipped, and know the clip point from previous testing.

[Steve Sanacore]

If you get WYSWYG a histogram is the way to go probably. Since the RED's histgram is off a LUT, just an exposure friendly interpretation (with some in-build headroom for the whites) of the sensor, I find a light meter more accurate, at least now.

Hans

Seems like a lot of testing is prudent when I get my camera. I guess i'll have to send off my meters for calibration soon.....

Thanks for the info Hans. It is well appreciated.

[Harry Clark]

I enjoyed Stu's articles very much. They are very informative and remind me that I still have a LOT of testing to do with my camera to really wrap my skills around it.
I think what's important to remember is that in order to decide where to place your exposure you need to know what your display medium will likely be. Digital projection? Filmout? Downconvert to HD or SD? (ugh) traditional video projection at a show?
There seemed to be a big difference in results according to what the end product will be.
Cheers,
Harry

[Gregor Hagey]

It's important to test, test, test so you can confidently set your iris though. The ISO might be just metadata but you can't change your mind later about the T-Stop!

One confusing aspect of shooting/metering at ISO 320 is that a middle grey card actually reads at 40% on the waveform instead of 50%. You need to overexpose the card by 2/3 stop to get it to 50%.

I think the most important thing to understand though is that it's not important to get middle grey or skin tone in the right place, just expose as bright as you can without clipping white. This means each scene needs to be evaluated holistically. A simple and easy tool for this is the histogram. A lit white object should have a slope at the right of your histogram not a straight line. If you are monitoring at ISO 320 you can trust that when the histogram is clipping the sensor is clipping too. First the blue channel, then green and finally the red channel will clip. Of course don't let specular highlights throw you off (they will clip no matter what you do). Dulling spray and polarizers are the best way to deal with them or power window a blur effect on them in post.

When the .r3d's are processed you can darken your picture at this stage to set the mood you want. This is where prep is important so you can create Red Alert or Redcine presets(LUTs) so the digital lab processing the .r3ds knows what your intentions are.

Of course a meter is still indispensable for balancing lights. Also, you should stick to one exposure level for a scene or your grain/noise could change with each shot.

[Hans von Sonntag]

One confusing aspect of shooting/metering at ISO 320 is that a middle grey card actually reads at 40% on the waveform instead of 50%. You need to overexpose the card by 2/3 stop to get it to 50%.

I think the most important thing to understand though is that it's not important to get middle grey or skin tone in the right place, just expose as bright as you can without clipping white. This means each scene needs to be evaluated holistically. A simple and easy tool for this is the histogram. A lit white object should have a slope at the right of your histogram not a straight line. If you are monitoring at ISO 320 you can trust that when the histogram is clipping the sensor is clipping too. .

With the current LUT implemented whites on the sensor don't clip when they clip on the monitor. There is a 0.40 EV offset in the monitor out LUT which gives a bit headroom on the whites. The historgram shows only the values of the LUT not of the sensor. Hence for really accurate exposure the histogram is not useful, though safe (which is good, thanks Graeme).

Hans

[Hans von Sonntag]

Of course don't let specular highlights throw you off (they will clip no matter what you do). Dulling spray and polarizers are the best way to deal with them or power window a blur effect on them in post.

When the .r3d's are processed you can darken your picture at this stage to set the mood you want. This is where prep is important so you can create Red Alert or Redcine presets(LUTs) so the digital lab processing the .r3ds knows what your intentions are.


Of course the whites clip when they are off. Nothing bad with this. In a high contrast scene unavoidable. The beauty about film is the nice shoulder (roll off) it creates in the whites. It's one of the secrets of a great colour correction to mimic with digital footage this pleasing film shoulder. For this some headroom in the footage is nice to have.

The goal of RC is to create the best base to grade off. Better than using RC is grading directly off the r3d files with a dedicated grading tool.


Hans

[Dj Joofa]

The beauty about film is the nice shoulder (roll off) it creates in the whites. It's one of the secrets of a great colour correction to mimic with digital footage this pleasing film shoulder. For this some headroom in the footage is nice to have.

Hans

To get to the latitude the film is providing a digital camera must have at least 17-bit ADC, where even the least significant bit is measuring signal and not noise. When digital cameras get to that level then film's latitude and shoulder won't be missed that much.

[Mitch Deoudes]

Of course a meter is still indispensable for balancing lights.

This is one thing that always confused me about "digital" guys lighting without meters... How are they setting their contrast ratios? I've not shot as extensively as some of the folks around here, but we were taught that CR was the most important aspect of matching shots.


Also, you should stick to one exposure level for a scene or your grain/noise could change with each shot.

Another interesting point - film is probably a lot more forgiving in this area, as well. So much for not waiting for that cloud to pass before shooting...

[lordtangent]

To get to the latitude the film is providing a digital camera must have at least 17-bit ADC, where even the least significant bit is measuring signal and not noise. When digital cameras get to that level then film's latitude and shoulder won't be missed that much.'

Or just use LinLog?

like Gregorios, polarizers have been a favorite trick of mine for handling spec hits for quite some time. It really helps take the "video curse" off it.

[Gregor Hagey]

With the current LUT implemented whites on the sensor don't clip when they clip on the monitor. There is a 0.40 EV offset in the monitor out LUT which gives a bit headroom on the whites. The historgram shows only the values of the LUT not of the sensor. Hence for really accurate exposure the histogram is not useful, though safe (which is good, thanks Graeme).

Hans

Not completely true by the tests we've been running in Toronto with firmware 13. Yes the histogram will show more clipping then what is happening on the sensor at 320 ISO, but it gives the most accurate reading of what's happening on the sensor.
At 1 2/3 stops overexposure the blue channel on a lit white surface will start to clip. At 320 ISO the histogram will show this clipping (both blue and green; the green isn't actually clipping yet though), but monitoring at 250 ISO the clipping is not apparent at all. The blue channel will look like a spike, but it won't be near the right edge of the histogram and could mislead you into thinking you're safe. It's not until 2 stops overexposure (when green and blue channels of a lit white surface are clipping) that the 250 ISO monitoring will show clipping on the histogram. I would always monitor at 320 ISO for this reason.

[Gregor Hagey]

This is one thing that always confused me about "digital" guys lighting without meters... How are they setting their contrast ratios? I've not shot as extensively as some of the folks around here, but we were taught that CR was the most important aspect of matching shots.



Another interesting point - film is probably a lot more forgiving in this area, as well. So much for not waiting for that cloud to pass before shooting...

I think setting fill and back lights by eye is something you need to acquire with experience in a given format (and testing!). I still use my meter all the time whatever I'm shooting, but my eyes have the final word.

Don't forget there's an iris you can change to keep the S/N consistent when clouds pass or use NDs (watch out for the IR though!).

[Mitch Deoudes]

gregorios - slightly confused by your right-hand histo. If the blue channel were clipping, wouldn't the spike have a harder right-hand edge, and no detail farther to the right? (I'm seeing a small bump at the extreme right that tails off toward the edge.)

[Gregor Hagey]

The blue channel should have the same shape as the red and green 'hills'. The attached file is a histogram of an un-clipped blue channel from the same test (the picture is a model with a grey scale). Lit white is at 1 1/3 stops over.

Sorry having trouble getting file to upload.

[Gregor Hagey]

Having trouble getting the attachment to work. Here it is again.

[Mitch Deoudes]

Ah - ok. I missed the part about it being a white card. Makes sense now, thanks.

[Hans von Sonntag]

Hi Gregor,

I'm just curious: In your histogram red, blue and green don't overlap. Did you change Kelvin in camera for a test, was ist possible to get an even histo in this regard, and if so, did the blue channel clip as well?

Next thing: Did the histo in RC at 5000K look the same as the histo in camera? Did the blue clip as well? I found that although the camera is balanced to 5000 Kelvin a neutral white in a 5600 Kelivin scene will be accomplished at 6500ish in REDCine (meaning that the Kelvins in RC/RED are non-accurate numbers - probably...)

And finally: Did you light your test chart with tungsten and set the camera to 3200K?

I'm asking because as to many shooting and exposing with the RED is new to me and there is a lot to learn.

Hans

[Gregor Hagey]

Yes the colours don't align for the white patch on the right. I've attached a series of corrected histograms. The channels align when the WB is set to 9500 Kelvins and 4 Tint. I've also pulled back by 1/3 stop of exposure in Red Alert so the right edge is more clearly visible. At 1 1/3 over red, green and blue are the same, but at 1 2/3 over the blue channel is becoming slightly narrower and forming a peak. At 2 over there is obvious clipping with the blue channel compared to red and green. We shot with Daylight Kinos and I didn't note their colour temperature (I was assuming they were 5600K). I'm curious to shoot some charts with different colour temperatures to see if this changes.

In general though I find that colour temp meters don't agree at all with Red and at first I thought this was only a firmware issue. After a talk with an associate it was pointed out that colour temp meters are optimized for film sensitivity and not how silicon behaves. Maybe a colour temp meter for digital sensors will come out soon.

[Hans von Sonntag]

In general though I find that colour temp meters don't agree at all with Red and at first I thought this was only a firmware issue. After a talk with an associate it was pointed out that colour temp meters are optimized for film sensitivity and not how silicon behaves. Maybe a colour temp meter for digital sensors will come out soon.

Kelvin is Kelvin, no matter wether you measure for film or digital.

I suppose since changing Kelvins in camera does not change anything regarding the exposed RAWs the colour temperature settings are more or less just numbers. In the end the histo after WB in REDCine matters. I get over 6500K when creating an aligned histogram of a 5600K shot scene. By the way, WB does make a lot of sense, it does reduce clipping. The next clapper board we will get will contain a 18% grey section - great for WB in REDCine.

Hans

[Gregor Hagey]

Kelvin is Kelvin, no matter wether you measure for film or digital.

I think it's a little more complicated than that, but this is a subject for a new thread.

[David Mullen ASC]

He's right though -- a meter reading in Kelvin doesn't care what camera you are using, film or digital. Kelvin is Kelvin -- it's independent of the recording medium. What may need some adjusting or rethinking though is the recommended CC filters for dealing with green fluorescents, etc. due to differences in spectral response.

For example, the mix of dyes in a Tiffen FLB or FLD filter is based on film's spectral response -- otherwise you could just add some CC Magenta filters to whatever warm or cold filter needed. Just as when you compare a series of warming filters with the same MIRED shift value, whether Corals or DecaMired Yellows or 85 filters, etc. -- you can find filters that create the same shift in color temp even though they don't match each other in color (some are yellower or pinker, etc). This is because correction filters like the 85B are based on making 5500K daylight look correct for a spectral response of 3200K color film stock.

[Gregor Hagey]

Well I'm no physics major but this is how I understand it:

Tungsten and Daylight have a continuous spectrum whereas HMI, Fluorescents, LEDs, etc. are discontinuous and only simulate the visible spectrum (sometimes well, sometimes not so well). A colour meter will sample red, green and blue and give a Correlated Color Temperature for these discontinuous sources based on human perception of a black-body radiator but not necessarily how a CMOS sensor 'sees' it.

FYI A black-body radiator is a black object that is so black it absorbs all light that hits it (that's what I call a matte surface!). When this black object or 'body' is heated it first radiates Infrared and then as it gets hotter Visible light (moving from red to orange and so on). When it gets to a whopping 5, 300F (wow that's hot!) the colour it emits is what we know as Tungsten balance. In the Kelvin scale that temperature is 3, 200. The temperature of the Sun is almost 11,000F or 6,000K - which is the colour temperature of the Sun before it hits the Earth's atmosphere (and is changed). It's obvious that Fluorescents, LEDs and HMIs don't create light this way.

The practical reality is that a colour meter may give misleading data as to how the white point will look on your film stock, videotape or RAW file. Testing needs to be done to know how your camera behaves.

Maybe there's a colour scientist that can weigh in?

BTW I still think this deserves a new thread.

[David Mullen ASC]

The problem isn't that the color meter is reading the Kelvin "wrong" for digital cameras, Kelvin is Kelvin... it's just that the whole notion of Kelvin is based on a ideal black-body radiator, which discontinuous spectrum sources like LED's are far from being. So you need more information than a simple Kelvin reading to know how these lights would reproduce, whether for film or for digital.

Your point originally was that the Kelvin reading in a color temp meter was somehow optimized for film, which I don't see how that's possible. Either it reads the Kelvin of the source or it doesn't, regardless of the recording medium. How that source reproduces is another issue entirely. The only way I could see it being optimized for film is perhaps in the CC filter recommendations it makes.