BMPCC4K : False Color and Exposure Tools Work Correctly! No Warnings Necessary! 

I'm with you, expose so you have as much information as possible in your recorded file ! I wish they added control the brightness and contrast before the lut. This would make the lut more useful on camera.

this is what atomos ninja v does and it's wrong

Not a problem and I agree. I think that we've got a lot of people with good intentions out there but sometimes they get ahead of themselves. Like a reporter fighting to get that story before everybody else, you're likely to hit first but have the most errors. I think it is better to take your time and try and be as accurate as possible when reporting things like this but I know that doesn't always bode well for clicks.

I don't really care about your personal experience to be frank and I'm not going to tit for tat about what dps you supposedly know personally do or do not do opposed to dps I know personally either. I'm telling you how the tool is designed and it in fact works exactly like it is supposed to. You want to expose for a lut, go ahead and do that. The information from that lut is in fact wrong regarding what the sensor is capturing and there is no dispute about this. Good luck with your general store.

@@TonyDae its not personal experience.... always read false color on LUT... false color on log is funny...

@@TonyDae dont spread wrong info

@@scoopoutclub4677 nothing I said is false information. Metering off a lut is literally reading false information. If you have a problem with that, that's your business, but don't pretend I'm wrong. You want to read information that isn't accurate then go ahead, but you should at least first monitor in log to know exactly what's clipped and what isn't because the lut won't tell you that. If you insist on disregarding logic and require an authority to tell your what to do, read the manual for Small HD, a widely used monitor brand for professional cinema, and they say exactly what I do, to meter off log and not the lut for the exact same reason I state.

@@TonyDaeno comments....why do i wast my time .....have a good day

My Suggestion is that metering off of values expressed by a LUT will result in generally underexposing your shot, because the values are incorrect. This is 100 percent valid and repeatable in testing. Even Small HD agrees with me in regards to metering based on the actual record values instead of the LUT values which will make shadows appear crushed or highlights appear clipped when they are in fact, not. I also suggest over exposing (obviously not clipping or running your exposure all over the place in one scene so don't misunderstand) to get cleaner signal from the camera, which works perfectly. My personal system for using a cheap camera like this, and avoiding denoising entirely, is fast and easy and requires very little post production work at all since the luts or setting I use in RCM affects all shots the same way. Any DP or Director or Producer or combination will have their own workflows and you will have to do whatever they say. Sometimes what they say to do is nonsensical and doesn't work properly or yield the best results, but they just "like it" that way, and force somebody to sit there and run denoising on footage because the DP wanted to underexpose for no particular reason other than they wanted to. As far as getting everything right always in camera when it is shot, you can do this easily by having proper LUTs that compensate for how you expose. If you expose +1 stop, you have a workflow in place for it. If you go under a stop to protect highlights, you create a lut for it or a workflow for it. As long a scene is consistent in the way it is shot, it will look fine. The point I make here is that workarounds are not necessary for a tool that works as it should. Having a toggle for people who want to meter off of incorrect readings would be a nice to have for those individuals, but is far from necessary.

@@TonyDae The point is that if you expose +1 stop and have a monitor LUT and post workflow for that, then it would be completely understandable to want to monitor that +1 stop compensation LUT so that you're simply exposing properly according to false color and zebras, rather than having to expose +1 stop by the tools and only use the LUT as visual reference of ranges.

@@THIRTYONEFILMS You can do it however you want. My points are that metering based on incorrect values (lut values) is not a good way to meter since the values are not actually correct. Values will look clipped or crushed when they in fact are not. You can get perfect out of camera images either way if you tweak how you work, however in my experience and the experience of others widely recognized as at least capable artists in this field, intentionally using incorrect meters is not optimal. Metering via +1 or -1 luts also give you false readings when in fact you could be clipping highlights or crushing shadows but the lut values do not show this. You can show this yourself by creating luts bringing clipped highlights into range and if you meter based on those values, it will appear as if those highlights are not clipped when in fact they are. You can do whatever you want for your own workflow, and if it includes metering based on inaccurate values, I will continue to say that its not a good way to manage exposure as the values are not accurate. I can understand WHY people do this, and I think there is a longer discussion thread in this video's comments section about it, but I still disagree with its use as I have always achieved better results by exposing with the real meter values shown in camera.

@@TonyDae I'm not suggesting to only meter that way, but to have the option to stack however fits your workflow would be ideal. Bouncing between the two as reference would be my process. See the overall exposure with LOG and false color, then switch to LUT and see where I need to modify light to get the falloff I want within the contrast curve I'll be using.

You can meter based on luts with some other monitors such as the SmallHD monitors, but even Small HD recommends bypassing the lut and metering based on what is actually being recorded unless you are baking a look to the file. The luts that you create or use should be made relative to a particular exposure style, color science, gamma, and color temperature anyways so all you have to do is be consistent with how you go about exposing your shots and everything should be good. What I have seen often is that people tend to underexpose on purpose when working with lut metering and especially for lower end cameras with lower DR, this is a sure way to get underexposed shadows and add noise when it is not necessary. If anything, in my experience it is better to slightly over expose when shooting raw (and 10 and 12 bit prores to a degree) than underexpose unless the highlights are important and are going to be blown otherwise. "Correct" exposure in my experience is great if you know you will not make any adjustments in post to shadows and plan to crush them, but as we tend to see with users of a camera like this, shadows WILL likely be pushed up in effort to maximize the dynamic range and noise WILL be exposed. I see this ALL OVER the forums and the solution is to overexpose! In the case of saving highlights, it is still better in my experience to use the meters showing the actual data being captured as luts will almost always push the highlights well out of range. By using the meters the way they are design in the pocket 4K, you can know much more accurately IF the highlights will truly be blown (red) or are at risk, especially red/blue channels (yellow), instead of guessing based on a lut. Blown highlights in unimportant areas aren't even THAT bad if they are rolled off nicely. Red IPP2 resolve color managed workflow makes this STUPIDLY easy to achieve. I know that some people in Hollywood will say to always just expose dead on and make it look right in camera, but I hate to break to Hollywood types, people end up having to fix shots often because someone wanted it a particular way and was thinking about what it looked like on set and not what it will look like in the cut. They see the cut, and want it changed. I worked in VFX production on high end features for years and changes to the exposure, contrast, color ALWAYS happens in post production, and done so to the lamentation of cinematographers everywhere.

@Jesse Timm Miller NP! Feel free to join the Discord for gear/skill/whatever chat! discord.gg/tMWnnR6m

My attitude was a little over the top on this one. I just get annoyed when someone says a tool is broken when they just don't know how to use it.

I'd be all for having the option to toggle between so the user can choose. I still think that if you are going to meter off a lut that you should meter for your clip points in log to make absolutely sure you know what is or isn't clipping.

@@TonyDae What about having both False Color and a waveform showing on the same screen of the external monitor? The False Color reading off the LUT and the waveform reading off the LOG signal. If you know you are 100% using the same LUT in the edit, and that by applying it certain areas will definitely clip - why would you want to expose that way anyway as a result of monitoring from the LOG signal?

Or rather if I want to expose so brighter areas of the face are showing pink in False Color - why do that off the LOG signal when applying the LUT in the edit will make those areas brighter?

Understand that in my way of doing things, a camera is a data collector. If you are exposing a scene and do not want clipped highlights or noisy shadows, meaning you want there to be data in these areas of the image, you will meter to be sure you have information there. If you choose to use a lut for monitoring purposes in matching a look across a scene, you should do that IN ADDITION to monitoring actual data. If you don't care these that shadows or highlights are clipped, you should at least know what will be clipped so you aren't surprised on post production. The way I generally do things is monitor data from the sensor with log and monitor look through a lut on an external monitor so that we have something to look. This provides me with information to know what adjustments I may need to do in post production as well as know what I am actually capturing. You get ALL relevant information this way. Keep in mind that the context for this video is more of a response to this thing going around at the time that the false color feature worked incorrectly because it shows what data you're actually getting instead of processing through a lut first, which is totally wrong. It works exactly as intended.

It's likely because the media you are recording to is not fast enough for the data rate. Check out my storage video.

@@TonyDae Ye I just founded the answer for that, but there is something else, there is like red dots of pixels or I dunno what they are, they're all around the items when recording. I've got one more question, please let me know what is the best lens for this camera, cos I tried two different lenses but they both seem to have some problems with working to this camera, my canon lens doesn't focus from close or far distance, and the other lens is just not right for this camera. so please let me know what could be the best lens for this camera

​@@E2VProduction I may create a video with a response to your questions here, but I will answer here on this thread as well. Regarding red dots or pixels, there is really no way for anybody to answer that question without an example of what you are talking about. If you are shooting in CDNG with old software, noise like that is prominent in underexposed areas and is color noise. If the pixels are fixed as discolored pixels, that would indicate stuck or dead pixels. Use the pixel remapping function in camera to remove these if there are only a couple. If you have rows or huge amounts then the camera should go back because the sensor got dinged or something. If the red pixels are something else then I am not sure would it would be, you would need to provide examples. Regarding lenses, it is not possible for me or anybody else to tell you what the "best" lens or lenses might be for any camera. All I can do is suggest options to you based on whatever information you have provided. Since there are no details on what you are likely to shoot, I can only provide general advice or tips here if you are shooting with the bmpcc4k. If you are shooting general production and want a native mount all purpose lens I suggest the Olympus 12-100 f4 or the Olympus 12-40 f2.8 Pro. The 12-100 f4 is a fantastic lens with very good OIS and the 12-40 is also a great lens but a little brighter. If you want what I would call the best 1-2 punch cost effective lenses that exist today for s35, it is easily the Sigma 18-35 and 50-100 f1.8 sister lenses, but you will need a lens support for the 50-100. Use Metabones adapters, NOT Viltrox, and you can use either the speed booster to get a s35'ish FOV or a normal adapter depending on your needs. I do not imagine you will need a wider lens but if so, the Tokina 11-16 or 11-20 are both great. I suggest Canon EF mount, not Nikon F.

Tony Dae thanks a lot for all your support mate, I really appreciated, somehow I managed to keep settling until I fixed all the issues, thank you again

1: The actual exposure is what the false color shows you in the camera, NOT what the LUT shows. 2: The reference LUT exposure you are talking about is an interpretation of data fed from the camera, which is incorrect data as it can only be HD/10bit from camera and is interpreted by a LUT. Because the data fed from the camera is incomplete and interpolated, it cannot ever be actual exposure data.

Please feel free to join the discord. Did be in the about page

Which Slog are you talking about specifically? Slog2 is notorious for example for being difficult to expose properly for, requiring over exposure to get skin tones "right". If Sony's own log curves cannot provide a proper middle gray point for exposure, that sounds like the reason why so many people complain about Slog in the first place. I again, as I have previously, cite SmallHD in that they specifically point out that ignoring the lut for metering is preferable unless you plan to bake a look into the shot or do not plan to do much color grading in post. It is for all of the reasons that I have already explained many times in the comments section of this video: That the lut may in fact mislead you to what is actually being captured and may push you to expose improperly. As I have said in many explanations here, luts in general will not be giving you an accurate reading of what is actually clipping or being captured, which can and will likely cause you to adjust exposure, erring on the side of under exposing, again for no real reason. This is especially true if you are using your own look luts and not technical luts from a manufacturer that may show clipping when there isn't anything of the sort taking place. This is even worse if you have created a look that rolls off highlights in a manner that actually shows clipped highlights within the dynamic range, when they are actually being clipped. You will end up with clipped highlights in post and scratch your head as to how that happened. Ultimately, you can do whatever you want. The point of this video is to illustrate that the tools in camera work correctly as designed and is not at all an error by the manufacturer or a bug in the software as certain youtubers had claimed.

Well yes and no. I would argue that correct exposure is whatever you want it to be in general terms, but under exposing intentionally makes no sense to me for the most part. Some people expose really dark because the end result is going to look like that, ignoring the general rules about skin exposure. But for me, when it comes to video, the point is to get the best possible image you can. Most of the times this means, in my experience, getting as little noise as possible and getting your lighting ratios the way you want them. If your goal is to minimize noise, generally this means not underexposing your shadows. Some people think that putting zero light on your shadows is OK and I don't believe that. I believe that you have to expose your shadow areas properly in accordance to your ratios if you want detail there and you should never have to necessarily "recover" shadows in post. If you capture detail in the shadows properly, there's no recovery necessary. Generally I find that exposing in accordance to the log profile (NOT the LUT) for skin tones (read up on proper IRE for different skin tones) and checking your lighting ratios does the job correctly... you should not be clipping shadow areas and trying to recover them in post. For noise, generally I have found that overexposing by a stop or two, depending on the camera and codec which requires thorough testing and analysis, can help with this tremendously in this regard with having little to no negative effect on the picture after grading. The more the camera captures in the shadow area, the less noise will be apparent at the same ISO after grading. I have found that I can over expose by about 1 and a half to 2 stops (against the LUT) on the pocket 4k and have no problems pulling the exposure down in post with prores and get absolutely clean shadows with no apparent negative effect on mid tones.

Because false color should be calibrated closer to the camera than a light meter would be, I would think that relying on false color more than the light meter when using this camera and many others in lower price ranges would yield more consistent results. If you have access to additional monitors with professional meters, I would consider using RGB parade and waveform to check exposure as well since the false colors will generate an average of the RGB channels and the parade for example should make it more obvious when a channel is getting clipped. Many cameras today do not adhere to any strict ISO rules regarding how they are calibrated, which is very different from film which had to adhere to certain guidelines to be considered say, 400 or 800 ASA or ISO. Back in the day, the film itself held the exposure data and was the same regardless of what camera it was in. The camera just pulled the film through and had mechanical shutter so it was a bit easier to measure with a light meter. I know Arri cameras follow certain guidelines so they can operate similar to film cameras, but I am not sure about cheaper cameras such as this BMPCC4K, and I have not checked to be honest. My hunch is that it does not. If you do exposure comparisons between many different cameras, in lower price ranges especially, you will usually see some variance in exposure with the same settings, so this has lead me to use digital monitors reading from the camera instead of light meters. Today, cameras are processing different sensors by different companies all using their own metrics so it can be a crap-shoot to know if a light meter will help more or not regarding overall exposure. This is why comparing noise levels between different cameras at different ISOs can make one camera look under or over exposed compared to the other one. Then when you add LOG curves, the exposure can be vastly different depending on how the curve is done. If you do use a light meter with a digital camera today, I would recommend doing a bit of practice so you know exactly how the camera is going to expose for the measurements read off the light meter in any given circumstance as it could vary from camera to camera. I will admit that I have all but abandoned light meters since waveforms, false colors, and zebras have come into my life, kind of like how I have abandoned the tape measure as well for checking focus. However, the light meter would be very useful for lighting ratios more than anything. If you want to use a light meter, I would recommend using false color to get your overall exposure and use the light meter for checking lighting ratios.

@@TonyDae Thank you for taking the time to answer my question! Many thanks Sr. Subbed!

Yes because its fed the lut exposure information, which is technically not what the sensor is seeing... its an interpretation of the exposure values, which can lead to highlights appearing blown or shadows appearing to be below the noise floor when technically they are not.

​@@TonyDae so technically the false color in the BMPCC reads what the sensor is seeing and not the exposure on it screen

The camera is showing its full dynamic range displayed based on its film log curve, not a limited dynamic range or contrast curve that the LUT would be showing. This is assuming that you are not intentionally baking anything into the image.

@@TonyDae What if you were monitoring exposure with an external monitor with the BMPCC?

The values fed to the external monitor are going to be truncated because the camera can only send a 10bit HD signal. If you are sending a LUT out to the monitor, it will also have actual exposure values truncated. The most accurate real information regarding your exposure is going to be the in camera monitor.

No, it would not be over exposed. Exposing to the right and pulling highlights down in post is a very common exposure technique used to prevent excessive noise in the shadows. If you starve the sensor of light without needing to, you cause noise when it is not necessary. The small red patches in the log reading show a much smaller area of at risk clipped highlights than the lut shows. The lut shows the whole table getting blown out. If we were exposing for the table, and it looks blown out with lut applied but not blown out in log, we would have to pull exposure down a few stops in camera. This would not be necessary in reality and will cause the shadow areas to be underexposed. Since we are baking the log profile in and not the lut, we will have enough flexibility for the shot to pull down the higher luminance values and preserve shadow details without adding noise in post. If we did the opposite and underexposed, we would have to push the shadows up in post instead which will add noise. You'd get the same overall picture except the shadows will be noisier in the under exposed part. In this particular example, the small patches of red in the blinds I would personally roll off because they aren't important to the shot and have exposure left as is. The stuff going on in the area of focus is what the exposure is built around for this particular example.

LUTS have a curve that is applied and dips shadows deeper and highlights raised higher than is being recorded. If you expose for the lut, you will adjust your exposure based on these curves when in reality you don't have to. It will push you to generally underexpose and that will deprive the sensor of information. Your picture quality generally improves overall when you work to capture as much data as possible by feeding the sensor with light. Obviously don't clip highlights if that's important, but if I were to expose for say the video lut to "protect" highlights, I would be grossly underexposing the footage for no reason since the log/film values captured are much higher than the lut. Expose to the right and In post, create a node to bring down highlights that sits just before the lut and you will see how effective this is. Compare to the underexpose technique and you will see the benefits.

+Tony Das the only cameras that really applies to are Sony and Panasonic. Their noise floor is so high on the ire scale. I believe Panasonic noise floor is at 30ire, where some skin tones should be and sony I forgot but they babe their iso at 2000. But with red, canon, and Blackmagic I never had to expose to the right, I just exposed. Now if you're somewhere that barely has any light of course you will get noise because it is so dark.

Eh... I was a little annoyed at some videos that kept saying there was something wrong with the camera when there isn't... I think I came off as a jerk in this unfortunately.

Tony Dae I didn’t take it that way. 70% of Camera RU-vid is varying degrees of misinformation. It’s important to set the record straight.

I don't understand. What rumors and what mess? The only mess regarding this camera as far as I know is incorrect information being given by the community. I was debating between this camera, gh5s, or fs5. I am glad I went with this option. There's a lot out there to choose from and the fs5 would definitely not be at the top of my list personally, but my needs are different than yours I'm sure.

@@TonyDae yeah i would have stuck with the gh5s... the fuji is starting to look like the only option. it was more an i'm mad at sony joke.

@@punkrachmaninoff oh okay haha. Yeah the xt3 looks really good. I shoot Fuji for stills so I was thinking about getting it but I wanted to have a true 4k sensor, internal 10 bit prores, and BMD's interface design. I don't think you'd be unhappy with either the Fuji or the BMD. I think Sony is protecting their cinema cameras at this point or can't deal with the heat dissipation to offer a true 10 bit log in their mirrorless. I mean, imagine you just bought a new fs5 and then Sony dropped a 10bit internal recording low light beast for less money. I would have been pretty pissed. I'd love to get pocket 4k type camera from Sony with the Venice color science in it.

@@TonyDae last night i was trying to figure this engineering problem out... according to the last rumors, the sony camp must be unable to reproduce the xt3's 10 bit internal... if this is the case, can we attribute the 4:2:2 10 bit as a direct result of fuji's ongoing fiasco of 4gb filesize workflow? like why can't sony do this, and are they kidding if they are going to avoid cannibalism of their fs series? different world. not asking for xlr and timecode.

@@punkrachmaninoff Fuji's is 10 bit 420 internal so that's all they have to beat in my opinion, but Sony likes their XAVC codec and doesn't use the h264/h265 used in the Fuji from what I remember. No idea what the problem might be regarding engineering other than maybe it's the full frame sensor in the a7s series making a bottleneck and the apsc versions have been plagued with overheating likely due to super sampling done in camera as well. Maybe it is their super sampling that's causing the over heating issue. I'm not an engineer so I have no solutions on that front. Protecting their own products sounds like the most likely reason to me honestly. Canon has been doing this for a long time with their cinema line so I wouldn't be surprised if Sony is following suit. Not a slight against Canon as they have the most reliable cameras I've ever used, just pointing out their business strategy.

A look up table designed by Blackmagic Design for the Pocket 4K.

@@TonyDae in camera i see 3 options for DR I know Video and Film but what is extended in terms of DR

I do not have a measurement as to what the dynamic range is. All I know is that it is essentially in between video and film curves. Film is the flattest curve, then extended video adds a contrast curve while attempting to preserve highlights, then video is like a rec709 contrast curve that blows highlights rather easily.

@@TonyDae Thx

The final image should not look any different if you adjust the contrast accordingly in post and did not clip anything. I get where you are coming from, and I get that you shoot in a specific way, however it still does not make sense to me that the metering would be based on something that is not being baked to the file. Since that is not what is being recorded, it makes no sense to me that is what you would be metering for. Perhaps for people who like to shoot in this particular way it would be good for BMD to provide a mode for this. The information actually getting recorded provides the most accurate information regarding what is actually going into the camera. Lighting ratios, as in the difference of exposure between fill and key, are going to be determined by how you light the scene and the difference of exposure of those specific areas/subjects/whatever of the scene will be maintained regardless of what the overall exposure is at, isn't that true? If the fill side of a face is meant to be say, two stops under the key for example, the lighting ratio won't change will it? It will be the same. Exposing to the right and pulling down in post can help to preserve details in shadows in the sense, and to underexpose instead while pulling up shadows to compensate increases noise in those areas. In both situations, the lighting ratios will be maintained with the exact same difference values. I get that the LUT is the target look, but like photography, getting as much as you can out of the sensor gets you all the data you can get. If there are benefits to ETTR, and no benefits to ETTL, then why ETTL? In post, you can always dip shadows down without noise, but you cannot, generally speaking, bring shadows up in post without noise. If a LUT is going to bring shadows down drastically in post, you are better off giving yourself lee-way in post to have those shadows a bit higher if it is asked for. I know I have personally had this happen where we wanted shadows brought up higher than in the original look and it's a good thing we used ETTR to do so. I am sure you know how fickle producers/directors can be in regards to post production :) If you don't shoot this way, cool beans man you do you. ANd if it works for you, it works for you. What I am saying is that the way things are metered in the camera aren't broken and make sense. If you want to ETTR or ETTL or nail it right there in the middle, that's up to you and comes down to your particular workflow. There's no right or wrong on how you want to expose, and what I am saying here is that the tools just make sense and do not require a workaround to use them properly.

I want to also be sure that I note by ETTR I mean doing it correctly, as in not clipping or over saturating important elements of the scene. Doing that would just be overexposing, not properly exposing to the right.

Perhaps it is a different train of thought, for me I couldn't care less about the log exposure values. For lighting it makes not difference, what matters is where the log values are going to be mapped to for the look. If middle grey is mapped to 40 for the log and for lut then I can use the green from the false color with my lut. But if lets say white skintones are 50 on the log and with the lut, 50% is mapped to 70% then I know I can't expose at pink even if the log is properly exposed, becuase the skintones would be over exposed based on the tone curve that I'm using. So I would rate white skin tones half a stop down to have it exposed properly for the look. The main issue is that the same lighting ratios will look different with different contrast curves. The higher the contrast of your curve, the lower the lighting ratios and vice versa for the most part. If you have a low con movie your not going to light with a 3:1 key to fill ratio, they will look to flat, but if its high con, anything over a 4:1 ratio might fall off to black on the fill side if the key is metered at key. Thats the whole point of the lut, to see the contrast and light and expose according. If my lut is clipping but the negative isn't, I'm still in trouble. There will be more work in post to get that shot to work. And if characters cross the overexposed areas then it won't be possible to fix. As far as the ETTR and bring down in post to get shadow detail, for low key looks thats not really practical unless the ISO down adjustment in post is baked into the viewing lut on set and your exposing off that. For low key cinematography, stops are too big of increments to even communicate in to decide fill levels. Its usually talking percentage points or tenths of a percentage point with the gaffer while looking at a calibrated monitor with a solid viewing lut. If I was trying to make those decisions off the log, or even a lut with the wrong contrast I would be pulling my hair out. I guess my question is are RED and Alexa then wrong in having exposure tools that change based on the gamma applied.

@@rzemanuel All good arguments to me. I haven't had any issues at all adjusting contrast curves in post for going over a stop and getting the desired outcome. Are Arri and red wrong? The red 1mx I shot with a while ago, metered like what we're seeing in the BMPCC4K. If that's different now, then that's interesting, and I would be sure you could change that to for your needs. I certainly would not be exposing for highlight clipping by the lut as shown in the example here since clearly there is far more information there than the lut would show. Pulling exposure down would limit the dynamic range of the shot for no particular reason. If the information being recorded is not what is being monitored then to me, that would not be a correct reading at all since it is not reading what's being captured. It is quite possible that meters are changing in newer rigs to benefit a director or producer on set and not particularly for maximizing information or post production workflow. I don't at all see the point of limiting what the camera is capable of if you are shooting for the best overall image, but just because I don't understand it, I suppose doesn't make it exactly "wrong" in an emperical sense, but definitely wrong in regards to what I know in experience regarding exposure. And, I certainly don't think it requires a workaround to get the exposure right.

@@rzemanuel I'll also point out that your arguments have definitely brought me to understand why people like yourself would prefer to shoot with meters changing based on the lut. For narratives, which I haven't done in a while, I used light meters for exposure mostly. Thank you for discussing this with me as it has definitely made me understand a different mentality when it comes to shooting video and using these meters. I think our discussion here shows how cool it can be be using the web for meaningful discussion :)

This video never discussed BRAW. The video came out prior to BRAW implementation. Prores bakes in ISO/gamma adjustments. Raw never does, because it is indeed raw. It does store the information in the metadata which is used in the decoding process. The only reason to change it in raw would be to expose for those gamma values which would be indicated in the metadata. I clearly did not think I had to explain this.

Can you expand on what you don't get? The point of this video was to explain a bit about how the exposure tools are designed to meter for what you are actually baking into the file and they do not "adjust" for a LUT that is being used solely as a visual reference when recording in "log/film" or raw modes. You should be shooting to expose properly for what is being baked to the file, not for a LUT that will not be baked to the file. A video posted by another content creator advised to underexpose your files for no discernible reason. Doing that would make you push your shadows instead of pull down highlights in post, which creates more noise than the latter for no real benefit at all.

@@TonyDae what if the lut is the same lut we use in camera and in post? Wouldnt it make sense to expose with the lut turned on in camera? Rather than exposing in film mode?

No, it would not make sense to me. Log/film is retaining more information than the LUT shows. You can always pull down highlights in post without adding noise in post as long as you have not clipped the important parts of the image. You cannot do the same with shadows because you cannot artificially add information that isn't captured. Generally, cameras have an ability to capture more than the meters suggest in regards to highlights and exposing to soak as much information into the sensor as possible generally gives you better overall images after re-working in post. If your post workflow only consists of throwing down a lut and you don't adjust the contrast curves prior to putting the LUT in your pipeline, then yeah it will do exactly what the camera shows in the preview. But you are supposed to adjust contrast curves in post. If all you do is throw down a LUT, I don;t understand why you would be shooting in log/film in the first place, but everybody is different. ETTR (exposing to the right) is a common exposure method that is used in both photography and videography/cinema that is best used at the native gain and is proven to provide better results for an image after post production when compared to ETTL (exposing to the left).