The 4 things even pro colorists get wrong

[@CameraTim]

I’m a simple man, Cullen posts a video, I watch it.

[@sqsyea]

When I discovered him I literally went back to his first video watched every single one and I literally re coloured an entire project in two weeks and was 100% better

[@JimRobinson-colors]

Some of my favorite subject here - but as per usual - I have a slight different take on some of the details. It's common for people to associate the way the exposure on the HDR tool works as "Photometric" like a camera exposure works, but that is not how I look at it. The idea that if you set a camera up at the default settings for maximum results according to your Gamut'/ Color space etc. The room without light is dark and the blacks are black. As you turn up the light - the light that doesn't hit the dark areas remain black. And where the light hits the image becomes brighter. But at no point does the area in the frame does the actual black become grey ( brighter ). But using Camera exposure - it does. So the offset actually does work like a camera exposure but not how physical light works. But the exposure on the HDR tool is closer to taking a light dimmer and increasing the light. Which by definition as I just explained it is the most organic way because it simulates adding light to the image - all other exposure is adding light to the sensor. So it keeps your darkest part of your image black and lowering the exposure reduces from top down. So thinking the slider as a pseudo dimmer of light on the set, should give the best results, for every reason you explained. The last one I don't agree with, because it kind of depends how you work and which actual functionality that you might be doing as a Trim. It sounds logical what you're saying but in reality Resolve has a 32-bit float in that a lot of things can be done early and reversed later without any actual harm. It is not like oil painting where you are layering on paint. Resolve will let you do a lot of things and then pull back or increase later on the node tree. So if you work from bottom up and ripple changes ( macro ) of the primary nodes - actual trim of certain footage might be required on a trim pass to match all the other clips. So keeping everything in place means that you won't accidentally ripple the changes over that edit. So I personally don't work that way. Pushing the key forward from a clean qualification is the way I do it also. What wasn't mentioned is the key will change depending on what is done on the nodes before it. So doing it the incorrect way ( after the ODT ) means that you can't really edit anything on the node tree before the qualification or it could easily break the key ( trust me, I have done this before ) but grabbing it with the orphan nodes as you showed - the key will never break when dragged forward on the dotted blue key line. And you can obviously use the same key to many nodes and inverse them if you wish.

[@daniel.lopresti]

But if you're modifying your in-camera exposure by opening the iris or adding ND, that is by definition how physical light works... If there's an area which is in complete darkness, then increasing the exposure by any light-modifying method (iris/ND) should not increase those shadows. I'm curious now whether this amounts to a difference in how ISO works, which adds gain to the signal level - though if I'm not mistaken, taking the analogy from analogue gain in audio (which is more my field of expertise), gain acts as a multiplier, not an addition, so any amount of gain times zero is still zero.

[@JimRobinson-colors]

@@daniel.lopresti Yes, the audio analogy is correct. An ND filter is going to make everything darker - more like how offset works to me. Take a greyscale step ramp - with the offset the whole thing will appear brighter as you raise the level - but with the exposure in the HDR slowly rises each stop will appear with the black staying the same. Anyway the analogy is that Exposure /HDR tool feels like you are controlling the onset lights and offset is more like controlling the light to the sensor.

[@thatcherfreeman]

@JimRobinson-colors I don't think I agree on your point regarding exposure. I have found in tests where I conduct an exposure bracket in camera and match them in post, linear gain outperforms log offset. With some cameras (IE those that hit their specifked log curves), they match incredibly accurately, with only noise levels being the difference after correcting the exposure change. In the case of your analogy, I think your description of light in a room applies only if you allow a single light to bounce a single time. In practice, a room with a single light will have all the shadows be filled in by later bounces (light bounces off the floor and then into the shadows). Doubling the brightness of that single light results in every part of the room getting a stop brighter. If the room already has some sort of separate fill light that contributes to the majority of the light reaching the shadows, then linear gain as well as an exposure change implies that both the fill and the key light are gained. The aspect where I'd give you credit is that lenses and filtration will typically have some further reflections, serving as a flare or a positive linear offset, even if it's very small. This can give the impression that raising the exposure can result in milky shadows, but it's more that you exposed the sensor to light with some amount of flare, in the first place. Even in this case though, linear gain matches differing exposures taken of this scene (bracketed via shutter speed) Tldr: They're basically the same, which is linear gain

[@JimRobinson-colors]

@@thatcherfreemanActually that wasn't my point - I wasn't including linear gain in my analogy - just offset in primaries as compared to the exposure slider on the HDR tool . It raises the blacks. Breaking it down to a single light wasn't part of it either. The point was that the darkest of the dark if black will still be black using linear gain. With the offset the black will become grey.

[@thatcherfreeman]

@@JimRobinson-colors Well the HDR panel exposure slider is linear gain.

[@HimanshuSingh-cl2yk]

Thanks cullen , I'm a beginner here....and i love your explanation. I would request you to reply the comments in a separate video .... because I can't understand what people are saying and i wanna learn about it.

[@_burd]

The great thing with workflow choices is that you can always test the other method and compare their results. There are a lot of 'adaptive' techniques that become the go-to method because of circumstantial things like the current state of the software or colourspace, that later require re-learning if you wish to use more refined tools. But it's not a lossy learning method as you begin to understand /why/ you are doing x,y and z steps, not just /what/ those steps should be.

[@Chemabal]

Best color grading channel. So helpful always!

[@sofetybilali349]

This guy is my all time color grading hero. I will watch every single video he post ❤

[@patryk8509]

Hi :) thanks for your videos. I can take advantage from nearly every single clip on your channel. But there's one question, which bothers me quite a lot. It may sound stupid, but maybe it's worth for some people to dive into that topic: how do you take care of your eyes? What are you doing to keep them relaxed, or maybe kinda "reset' them during color grading? Some say it's good to look a few times on a neutral thing like a sheet of white paper, or take look of the green lawn outside of your window, while making small breaks during the workflow, etc. I think, our eyes often get used to a certain look after some time of working on a project and sometimes we don't realise, that our perception is a little bit off. Maybe it's a good subject for a next video of yours. I appreciate your contribution to teaching us color grading, it helps me to elevate the quality of my work. Cheers from Poland :)

[@dlaimer]

Second that.

[@alfp1904]

outside screen breaks

[@JohnnoWaldmann]

It’s also useful to not linger on any one image for very. Adjust one thing, compare before /after. Maybe tweak. Before /after. --then move on to the next image. And repeat. The less time you spend looking at one image, the less locked into an image your eyes become and the less your eyes normalise whatever garbage is in an image. Eyes /brain will normalise an extremely weird white point image if you give them time to do so. But if you do not dwell on a single image, you do not get locked on to a random weird normal -to the same extent. So grade and move on. But rest breaks are still important. I take many breaks because my body objects to screen time due to old injury. Others use egg timers to remind them to take breaks. Then coffee, or mailbox -a short walk in the sun or rain time. Inevitably I will return with a better idea I need to apply to the last image I touched. Those breaks improve my grades and improve my overall speed. To prevent OLED pixel burn in, I set the current clip to loop while I am away from the desk. Sometimes the casual glance as I sit reveals an insight that I would have missed with closer focus. I think of this as soft attention as opposed to the hard focus of regular grading.

[@NathanHoang]

another great and informative video Cullen, the 4th one i'm definitely guilty of all the time :) thanks for the reminder

[@bobtronic73]

Coming from a VFX background I always found the Offset stuff weird :) linear exposure all the way. Great teaching, as usual!

[@thatcherfreeman]

Question: I think Mistake 1 and Mistake 2 are somewhat at odds with each other. Your choice of primaries makes a difference in terms of how your RGB linear gain balance adjustment feels. Simultaneously, your choice of primaries dictates what will be clipped off in the footage if you were to node cache (say, caching after noise reduction at the start of the pipeline, which imo is a great use-case for a pre-IDT adjustment). The former encourages you to choose a set of primaries that results in the best feeling balance adjustments. The latter encourages you to choose a set of primaries that encompasses all the cameras used in your project, which can be quite hard if you have RED footage. I would pose that these two goals are in conflict and you can't solve both at once. What do you think?

[@thatcherfreeman]

For what it's worth, my take is that on multi-camera projects, it makes sense to choose working primaries that "feel" the best and is natively supported by your tools and luts. Upstream of the IDT, you would put those low performance spatial adjustments (noise reduction, fake lens effects, scatter, etc) that you need to node cache. Somewhere probably towards the beginning of your look, you'll roll off all the colors that landed outside of the working primaries and map them to smoothly land within the working primaries.

[@JimRobinson-colors]

I agree with this one Thatcher

[@ozzmanzz]

Thanks for posting this. With your linear gain suggestion, are we meant to also change luma mix to zero? In your explanation here at 10.00min, luma mix was still at 100.

[@thatcherfreeman]

@@ozzmanzz Definitely set Lum mix to zero when using linear gain.

[@thatcherfreeman]

@@ozzmanzz Lum mix should be zero when working with linear images (and imo at all other times too)

[@hanspeterwagner5881]

... and another absolutely helpful input ... and .. reminder! thanks a lot, Cullen!

[@pdou6796]

Cullen, great job with this. On point the entire time, and I've struggled with these issues as I'm learning from scratch. Offset vs Global, outstanding explanation. Linear for the Balance using Gain, another point! Since I'm NOT a pro, but trying to produce consistent good quality work, the struggle has been to avoid the push-pull method which is unmanageable for the 300-500 clips I'm dealing with in a project. Love the point about replacing two moves with one. I don't understand the available tools well enough yet to sense when I'm fighting my own inputs. For Node stacking, from a newbie perspective, there has been a phase where I'm trying to discipline myself to put separate adjustments into separate nodes, but I'm still thinking in a one-node mindset. Often I fall back on the habit of reaching for another adjustment without selecting that specific node. This causes confusion later, when trying to find an adjustment buried in the wrong place. Overall, since subscribing and applying what I've learned here, I can absolutely see the benefits you describe. My throughput & overall production quality have risen dramatically. It has made this creative space so much more satisfying. Thank you for that.

[@ksharky888]

Totally hear you with the node stacking, have you considered using a fixed node tree?

[@pdou6796]

@@ksharky888 I do use a node tree, usually a slimmed down version of one I saw on Darren Mostyn's channel, but it's a fairly new habit. When you're adjusting balance, for instance, and feel like it needs a touch more contrast, it's too easy to mess with contrast in the balance node. Rather than switch back to the proper node for each tweak. It's a muscle memory thing, and partly because the micro color panel makes it so easy to jump between dials that I forget to change nodes. It has taken a couple full projects to break the old habits.

[@ksharky888]

@ yup I still do that from time to time and kick myself for it later down the line haha

[@diegochak]

Great video! thanks for this content. I like the idea of ​​keeping it clean or with few nodes. I would like if you could make a video using the nodes at the bottom, for secondary adjustments, like when a client starts asking for specific adjustments and you need to have more nodes. Thank you!

[@HspncAt7heDisco]

Super helpful. Thank you Cullen!!

[@inknpaintCW]

Even when I don't agree with(not the case here) I love being put in a position to consider another perspective. Forced growth even if incremental and slow is a great thing.

[@foxyvisionsvideographer]

I liked to have the look that I get with a high black point with ofset. But soon noticed that if I need edit my footage recorded in the dark, and correct it HDR global works so much better. What I also did is reduce the HDR global -0.99 and then pulled up the offset, gave me a nice flat image to do some soft looking image.

[@tomaskonvicka4135]

Hey, Cullen. I have noticed that some practices are subject to trends and can be misleading. What practices do you see as rock solid and what practices do you see as short term and we should be wary of? That would make an entire episode.

[@movie-trailer19]

Thanks for the advice on HDR. Somehow I overlooked it when setting up the exposure Fine

[@S.P.I.L.L.]

Amazing tips! Thank you!

[@songsbydaniel]

Please tackle GoPro Native WB in one of your videos. It's the absolute worst to match with other cameras and I'd love to get your take on that.

[@bukenyarosco201]

i always put my noise reduction node and blur in pre idt is that wrong? and talk about printer lights can they be activated in the linear mode or they work only with offset?

[@AllThingsFilm1]

Great tips. So glad I discovered this channel. Thank you!!

[@ricardobirnbaum]

So much greatness in this video

[@elliotyip9844]

Cullen, many things happen in my balancing node. I'm going to use your Linear Gain method to adjust moving forward, but I also use the node for hue vs hue adjustment, colour saturation adjustment using colour slice, colour warping and so on. By changing the node to linear does all the other adjustment still function the same or do you suggest to create other nodes for those and reserve the balancing node purely for Linear Gain colour temp adjustments?

[@VisualMarkus]

Great video and cool shirt!

[@ewaldhentze1428]

Hi Cullen, thanks for these great advices, but shouldn´t LUM MIX set to 0 when using linear Gamma?

[@CullenKelly]

Yes, great catch! I posted a dedicated video on this after realizing it had been set incorrectly for a few videos I’d already recorded: kzbin.info/www/bejne/d4LNfIqNi850Y9ksi=O8Tbq8zbCCLvcycf

[@NOIRGRADE]

About ACEScct in the second half ~14:00 Could you elaborate on the part about clipping colors because of the size. Is that something from the times when grading software was still 10bit or 12bit? Because in 32bit float Resolve I have never experienced any clamping with IDTs (expect for maybe the inverse ODTs for R709 etc.)

[@rpmusick]

In ACES it's really hard to clip anything, or even get it to the maximum value, and sometimes you want that. So you have to work outside of the ACES to get that and that's the only thing that I've found that I can't do inside that color working space, but cullen can elaborate more on this.

[@christykail3314]

This was a known issue with neons on the ALEV3 Alexas, because they could reproduce colours outside of of what a human could see. ACES 1.3 kind of fixed this by rolling-off out-of-gamut neons with the reference gamut compression, but the Alexa 35s AWG4 completely fixed it on Arri's side. I've not seen the issue on any other cameras.

[@NOIRGRADE]

@@christykail3314 out of gamut are neither a problem specific to Arri cameras nor are they clamping signal in ACES per se. The data gets clamped after or during applying the ODT but up until this point all the data is still there (given 32bit float). I've seen those with all sorts of cameras. In fact I am currently in the making of a video about out of gamut colors that'll be released in a couple of days.

[@christykail3314]

@@NOIRGRADE I'm talking about Gamut clipping into ACES AP0, which I've only ever seen on the ALEV3 Alexas.

[@NOIRGRADE]

Hi Cullen, I'm gonna be honest. I have to slightly disagree about Offset. Surely it is not a "perfect" tool for exposure and whitebalance, but it's behavior really depends on the working color space of your choice. For example in LogC4 it is extremely inaccurate, while in ACEScc it is mathematically the same thing as Linear Gain. ACEScct is so close to ACEScc that the fogging in the shadows get's close to neglectable. So why do I disagree. Talking about speed and efficiency a tool like Offset that can do exposure and whitebalance at the same time in one go but is 1% inacurrate is imo sometimes the better choise than using two nodes with two different tools. To be honest using Linear Gain because the HDR Global Wheel doesn't do a great job for WB seems like a compromise again. Let me know your thoughts on this. Given that we usually don't have to do big exposure changes of +2 stops or more (yes, then I would use a more accurate tool as well) but rather shot match in the ballpark of ~1/2 of a stop.

[@elcasanelles5806]

Maybe I did not understand correctly, but if he is doing both exposure and balance in linear you could still only use one node for both.

[@rockindude3001]

@@elcasanelles5806 Which, in ACEScc(!), is the offset-wheel for exposure AND whitebalance. The only problem with ACEScc arises with out of gamut colors, or negative values. For those values offset doesn't work properly. So ACEScc+Offset has limitations, however at least I think that for Exp+WB ACEScc is a viable choice when managed correctly.

[@NOIRGRADE]

@@elcasanelles5806 that's totally true. however doing exposure with the linear gain approach comes with the annoying fact, that the wheels response is way different if you drop exposure than if you boost exposure.

[@mrshaheedmalik]

I switched to HDR Global awhile back. I just wish i could map Printer Lights to HDR Global.

[@AladinBenfatna]

Hey Cullen, I've seen in the last episode of Grade School that you've got a new version (1.4) of your macOS Viewing Transform for Apple Studio Display (both for gamma 2.2 and gamma 2.4 now!), will you update the link to the lut with this version instead of the currently available 1.3 one soon? :)

[@ozzmanzz]

Thanks for posting this. With your linear gain suggestion, aren’t we meant to also change luma mix to zero? In your explanation here at 10.00min, luma mix was still at 100. Is there ever a reason when we should leave it at 100 when using linear gain instead of zero?

[@thurialb]

This is exactly what I want to see, all these tips are awesome and make me rethink my way of working, I love it! I have a question for the linear gamma balance node. I usually start my balance by checking and if necessary adjusting my shadow with my Printer Light Hotkey, but they really act differently in linear gamma, any advice or tips ? Again thanks for another great video

[@foxyvisionsvideographer]

very good point on mistake 3

[@HoneyB_kyoto]

Great video :: SOLID knowledge, thanks

[@sirdelek2629]

I've watched many of your previous videos, but I'm a bit confused as a beginner. So is there any difference for exposure when using HDR Global vs Gain in linear gamma or they are doing exactly the same? And just for balancing the image (aesthetics), should I be using offset in default or offset in linear? On your earlier videos I haven't noticed that offset balancing in linear so it's a bit unclear for me :) I know it's preference but I'm asking for best practice :)

[@balamurali1784]

Is there any source to learn theoretical part of colours and colour correction, grading everything regarding colours for editing purpose. Cuz I don't know what to do when I open a video to edit. Now i didn't understand this white balance though you explained it good . What should I do... How can I know that I can change or add particular thing, is there any book or something else, please 🙏 let me know sir

[@jonathanmaccc]

Hi Cullen. I'd love to know that if we do dodge and burn on some specific spot , should we use gain or just exposure (hdr) to brighten or darken things up and down ?

[@baruchratz512]

Hey, I'm pretty new to color grading, been doing it on and off for a couple of years, mostly in Lightroom when I was still a photographer, and now for about 2 years of the 3 I've been a cinematographer and editor, in premier pro, and I'm slowly transitioning to resolve (I have a few technical problems because of computer stuff, not important rn). Can you please make a technical resolve color grading 101? Just so I'd know the basics of all the tools? There is so much going on here, and while I do know the basics of wheels and curves, there is all this node stuff, and other things that just confuse me, and I probably need to use in order to get the results I wanna get (saying this because for now every project where I needed to grade I edited in premier, I just can't seem to get good grades I resolve rn). Thank you, and keep up the great informative videos!

[@grahamf695]

I am a bit confused by your guidance on colour balancing. Only a month or two ago, you presented the HDR Global wheel as being equivalent to changing a node to linear and using Offset. I wrote this note after watching your previous video “The HDR Global wheel operates in a linear way, which is more photometric. It is often better for controlling exposure and colour balancing.”

[@elliotyip9844]

No, I think he said changing the gamma to linear and using Gain, not Offset, to adjust exposure. It is indeed equivalent to using HDR global.

[@grahamf695]

@@elliotyip9844 sorry, I stand corrected. He did say Gain, not Offset.

[@marcelolutter-paz3805]

Changing a node to linear and using GAIN, not OFFSET !

[@elieskateur]

Hi, to compare shot to shot differents colorgrade I see that you always use the grabstill fonction but it would be more efficient if you create a new version with cmd y and then switching with cmd b or n, to switch easily and quickly color grade to another in full screen and clearly see the differences on all the images.

[@csound55]

Good tips. Any advice to streamline grading always useful.

[@elcasanelles5806]

Great tips. Regarding balancing on linear I saw Waqaz Qazi saying that in order for it to work correctly you had to pull Lumix mix to 0. If the did explain why I did not understand it. Seeing that you don't, and being this the channel where one understands things and nut just learns recipes, I was wondering why yo do not think it is necessary and what might he be trying to do by doing so.

[@shifteleven]

He (CK) has had a video where he did pull that down to 0 because otherwise highly saturated things can look "hot". Maybe one sees it as a good baseline to keep it to 0, the other will turn it down when it's a problem.

[@CullenKelly]

Hey great questions, as @shifteleven mentioned I've got a video discussing the lum mix thing here: kzbin.info/www/bejne/d4LNfIqNi850Y9k You can also watch my original linear balance video from a few weeks ago if you'd like: kzbin.info/www/bejne/aXqYaYGqoryCo6c Hope those help!

[@elidaughtry4289]

Quick question: If I know in advance that all the clips on a given project were captured, for example, in REDWideGamut/Log3G10, should I still work in an intermediate color space like DWG, or can I work from the much larger RWG? In my mind, it seems like an unnecessary step when all the clips are the same, but I could also see there being potential issues going from SUCH a large color space down to Rec.709 or even DCI-P3. Any insight?

[@elidaughtry4289]

Or is it more of a preference thing? Like if you grade mostly in DWG, you know more precisely how clips in that color space react to certain tools?

[@Kodzegura]

Thank you so much!

[@Mrbenbarbie]

So did I understand correctly that the HDR WHEEL is linear gamma within the working color space for exposure, but color shifting is not linear within the same tool?

[@Patrick-mg7vv]

I would like clarification on that also. Perhaps something to do with color space aware tools drop down

[@chebosludo]

And this linear trick just work with gain ? What happens if you use lift or gamma in linear gamma ?

[@moklooking3485]

hi Cullen, thanks for your content rich video. I remember at the early stage you teaching us using project base colour management which at the output CST is set to rec709 gamma 2.4 earlier this year you start using node base colour management and the output CST has changed to rec709 gamma 2.2. now I m confused, 2.4 or 2.2 please help., thanks.

[@TheRealBarkinMadd]

What about staying in the primaries for mistake 1 and using linear gain? Isn't that equivalent to the HDR palette's global exposure, Cullen?

[@RallyAddict]

Interesting with the hdr global exp. For a while I've been testing hdr global vs offset on my slog3 footage and I tend to get opposite behavior. My footage from A7S3, mostly exposed to the right, tend to get really low contrast when treated with hdr global. So I have to compensate my hdr global settings with some correction in shadows e.i. contrast curve or lift. The overexposed example you showed is really contrasty for a start, but my overexposed slog3 footage behaves like this: hdr global -> exp down -> I get my prefered exposure, but the shadows stay waaay lifted. And the offset wheel seems to get those shadows to more "normal" levels quicker. It's not like global exp doesn't lower the shadows, but they are falling down really slow, so you get a decent shadow range only when like -3 or -4 exp on global ;). I'm using color management with CST's and working in DWG.

[@VictorRiley]

How exactly do you setup your workspace when you color manage using nodes? In order for the HDR palette to work correctly, you need to have the timeline color space set to your working color space, or have the HDR tools set to the corresponding color space/transfer function manually. For example: I mostly grade in Arri LogC, with S-Log3 material, because most of my looks are built for LogC in-out. So, my setup looks like this: Project settings: timeline color space - Arri WG/LogC Output color space - Rec709-A (for approximately correct viewer gamma, I'm on a Mac) Input node - S-Log3/S-Gamut3.Cine to Arri LogC/Arri WG Output node - Arri to Rec709/Gamma 2.4 This lets the HDR palette work in Arri LogC. If I adjust the exposure wheel by +1 or -1, I get exactly one stop of exposure increase/decrease (checked with a tonemap in Arri LogC). It's almost photometrically accurate - when I compare an image that's exposed "correctly" using a grey card with the same image overexposed by one stop, then brought down one stop using the HDR tools, they almost look identical. There is some slight change to contrast, but that's also to be expected when changing ISO or aperture in-camera. Also, the HDR tool's exposure knob produces virtually the same result that an adjustment in linear gain would.

[@RallyAddict]

@@VictorRiley thx! My project settings are: color management: off (DaVinci YRGB) timeline color space: rec.709 (scene) output color space: rec.709 (scene) Maybe I understood it wrong, but those settings are overwriten because of CST's that are effectively doing the color management. I use CST's instead of color management in project settings because color management messes up any titles/graphics etc. even if you bypass color management on them. input node - S-Log3 / S-Gamut3.Cine to DaVinci Wide Gamut / DaVinci Intermediate, Tone Mapping set to DaVinci, Gamut Mapping set to None output node - DaVinci Wide Gamut / DaVinci Intermediate to Rec.709 / Gamma 2.2, Tone Mapping set to Luminance Mapping, Gamut Mapping set to Saturation Compression So I do all the grading in DWG inside of those two CST's.

[@RallyAddict]

But I can already see that changing timeline color space to DWG changes the grade... eh. Also changing HDR tool color space changes the grade. So the mistake is propably on my side... :) Still new to this color managment thing ;) What's even more interesting: Changing color space/gamma on the project level messes up a power window on my "defocus background" effect. I've got some weird artifacts on the mask feather. Edit: needed to fix the highlights parameter. Now's fine.

[@VictorRiley]

Yeah, as I guessed: the timeline color space is the issue with this setup. All tools that are color space aware in resolve will use the timeline color space as a reference to set up their internal pipeline. When you set the timeline color space to Rec709(Scene), the HDR palette expects the signal chain to be in Rec709(Scene), so it won't work accurately in S-Log3 or DWG.

[@CullenKelly]

Hey Rally, check out this video here to get yourself caught up on color management: kzbin.infoJpRuQQ__-YA?feature=share

[@maxkruse3717]

Hi Cullen, thank you for your video. I already implemented your advice a couple of weeks ago and adjusted my node tree accordingly regarding mistake #1 and #2. However, I'm curious about the mistake #3. With 99% of my projects, I'll color manage on clip level. Means, every single clips gets the camera specific CST treatment. That being said, my standard node tree starts with an empty node for Noise Reduction. My CST from Camera to DWG comes 2nd. I don't have any specific reason to do that other than "it feels better this way" and "it should make more sense this way". Don't you agree, that Noise Reduction should be applied before the information gets transformed into a different working space? I think thats one, if not the only application to do adjustments before the CST. Curious to know what you think. All the best and thank you for your awesome knowledge. Max =)

[@Longdancer]

Hi Max This has been my approach too and i was going to ask this exact same question. So...if not before the CST, then where? Although.....it's worth noting that i've been grading for quite some time now (bout 4 years), and i still find that i'm trying to weed out all the bad advice i've garnered by watching inexperienced influencers on KZbin. Maybe this is one of those habits that's i'm finding hard to break. Be good to know the correct way.....if there is one.

[@uehal]

Do color panels work with the HDR wheels?

[@CullenKelly]

They do!

[@brandonbural]

What about if you want to make your output space say C film log and then use dehancer or a film out to use after your CST? I know that’s a very common technique.

[@CullenKelly]

Fair play in my book, as I’d consider your film emulation to be part of your display transform, not something tacked on afterward.

[@qiyuantan]

wait so im a little confused here, do we still use the primaries (gain and midtone) for exposure changes, or should we just use the HDR wheels (global, highlight etc) instead , or using hdr just for major exposure changes?

[@Longdancer]

I think in a previous tutorial after using the 'Global' HDR wheel for exposure adjustment, Cullen switched over to the primaries (in the same node), and used Lift and Gamma to adjust the lower and midtone areas of the image. Not sure in which video he did this, but i'm pretty sure that is what he did and i have done the same, and i've found it to be a useful approch for me.

[@qiyuantan]

@@Longdancer thanks ! just what I needed as I'm working on a color session right now

[@Longdancer]

@@qiyuantan No probs. I did try the same test by comparing the HDR Global vs the Offset and I had the same results he did so….. All the best 👍

[@禄小平]

感谢 卡伦 非常的有用

[@Vvafi.1]

I saw a KZbinr Dunna Did it talking about him so I came to see 😁

[@RealColeFarlow]

This is my first video here… how is this my first video here?

[@kingkira5368]

Is it wrong if I choose to use gain + shadow combo for the exposure ?

[@imcrimson8618]

No. Nothing is wrong itll only be “wrong” if it dosent get the mood, look or feel you were going for.

[@qiyuantan]

@@imcrimson8618 do we still use the primaries (gain and midtone) for exposure changes, or should we just use the HDR wheels (global, highlight etc) instead , or using hdr just for major exposure changes?

[@imcrimson8618]

@ sorry im not sure

[@arielshpitzer]

With a client I use a LUT. we edit and do color at finish from there I can get basic guidelines and continue alone... wasting clients time and money for coloring is pointless.

[@Pala4er]

well some clients like to participate. take part in the grading process as well. it is their time and money. let them decide how to spend it.

[@novagamerv1782]

we stan cullen

[@casspc]

Cullen throwing ACES shade?

[@imcrimson8618]

I doubt. Aces has its own uses. Im not too knowledgeable on the whys here but I remember hearing the choice between choosing aces, dwg or a different working space comes down to the creative intent as well as technical. You might choose aces if you like the way it handles the footage. I think aces applies a “film like curve” where dwg is more of a neutral base, but it’s also one of the biggest working spaces.

[@otoshithekid2957]

Ok, number 3 was fucking crazy

[@posebukse]

9:18 You say that the HDR Global trackball moves into another color space than DWG, but not which one. Which space does the HDR operate in? You also use a long-ass time to say that DWG trumps ACES 😂

[@itstreamerfin]

Please build a discord server

[@CullenKelly]

It exists! Currently only open to alums of my courses, but that may change.

[@itstreamerfin]

@CullenKelly please build a public one. That's gonna be very helpful for beginners

[@ksharky888]

looking like a skater

[@factor3208]

Yeah right. This video has too much banding for a pro colorist teachings

[@hforheh5135]

but, but your background is banding crazy.😢

[@stefanocson6237]

great

[@tonymed2738]

Getting tired of cullen preaching every fkn tool in resolve is not correct. He is just capitalizing. His annoying emails prove it. Tired of his BS. stop patting yourself in the back, is just a fkn job, do it and thats it.

[@Noone-ew2wk]

I'll give you guitar lessons for a copy of contour 😂😂😂😂

[@robertruffo2134]

Mistake I see AGAIN AND AGAIN - Omitting to balance first in RAW settings (if a RAW file). It makes me crazy to see people who treat RAW video like it's baked-in 8 or 10 bit. It is not. Always get your RAW in the ballpark first - really if RAW the balance node is those RAW settings

[@CullenKelly]

I’d actually argue that balancing in RAW settings is the mistake…I discuss this here: kzbin.info/www/bejne/jKbcoKCCmrSkfcUsi=DmwL2k7XpJ0VOEj-

[@JimRobinson-colors]

There are things in the raw tab that are added after the debayer process - I think in Resolve the tools to do some of the same things are actually more flexible and accurate than a RAW tab slider. I personally don't know which are and which are not - but it seems that if any of them changes - then the file would need to be refreshed if the debayer process then baked it in.

[@steevefanny1]

top

[@orainethomas]

Amazing tips, thanks!!!

[@foxyvisionsvideographer]

very good point on mistake 3