But.. but.. it was watching your channel that convinced me to move to a colour managed workflow in the first place! Plus switching to gamma 2.2? I wasn't emotionally prepared for this episode 😅
Cullen, I have seen your resistance to the HDR tool and if using panels it is not as easy as switching with the mouse. But a lot of the stuff that you are implementing by using composite mode and or changing the colorspace of nodes and turning off channels etc. is how some things work in the HDR tool without modification - so the ease of use or set-up is absolutely faster. For instance - On the HDR tool the contrast textboxes for instance doesn't affect the saturation when using it - so adding contrast will only affect the luminosity. And if used with the pivot will move the middle grey point - raising will give you more stops below and lowering will give you more stops above. Which will raise or lower the exposure as well. The contrast differences and how it is used is really evident if you push a really negative or push it positive and especially in the negative on the HDR tool see the color remain while the luminosity is lessening.
@@shueibdahir the primaries pushes the saturation with the curve - you might like the HDR contrast when you then adjust the saturation on a different node to your taste. If you use a grey ramp and compare the actual contrast it is pretty much identical - it just keep the colors the same and doesn't pump more sat into it.
May be is it coz of the switiching between hdr and primaries,like if u have 100 shots, u may find urself resting on hdr but u wanted to adjust in the primaries.
In previous videos from a while ago, he switches the exposure node to linear. With this new process it looks like the gamma on the Prim node is not set to linear? Is that correct?
I didn't know I could set the compositing mode to luminosity! That's a game changer. Coming from a long photography career that is a key thing to do in Photoshop. Didn't know it was an option in Davinci.
Dear Cullen, can you please also provide the Link to the macOS Viewing Transform you talk about? Thank you in advance! Edit: Thank you for adding the link Cullen
It‘s literally the first link in the description of this video. Just click the very first link, fill in your name and email address and it‘ll get sent to you. It ain’t that hard.
Thorough and clear explanation as per usual. i do have a couple of doubts; If there is only one camera used in a project and did not want to make groups, where would you put the CST nodes? Also is there a difference in applying the macOS VT as you did in this video, compared to doing in a node as you showed in a previous video?
I saw some image-breaking artifacts with this new approach. 7:54 - speculars are too colorful, have obvious shapes. "Before" was much more aesthetically pleasing as well. 16:50 - visible clipping of the lamp. It was in the original image, but something in the pipeline had to be done to resolve this issue.
Thanks Fedor! To be fair, these artifacts are image-inherent and not caused by anything I’m doing. Though certainly the type of details you’d want to address before delivering!
Hey Cullen, your videos have taught me so much! Thanks a ton for making them! Two questions on this one: 1. Since you've combined your exposure and ratio node, and your new primary node's composite mode is set to luminosity, when you change exposure with offset or gain, don't you get weird colors? 2. On the balance node with linear gamma, would the results be more accurate with the composite mode set to color? Would that preserve the tones better?
you could save time on the first node by having a layer mixer set to luminosity and just have the extra node disabled by default, turning it on and off is a lot quicker than going through the right click menu each time
Your MacOS viewer transforms is the best thing ever happen on internet for startup like myself. I got almost 1:1 luminance reproduction on all my mobile devices.
@@zweiche This what I get from my own testing and it works across all social media including QuickTime player for this time being. 1) On node basis, you'll always want to set your gamma to 2.2 and add "CKC Apple viewer LUT" on color viewer. 2) On export always set you gamma to 2.4. This specific hack is to make sure that QuickTime will always read your footage as gamma 2.4 and social media will always take the 2.2 input instead including TikTok, Instagram, Facebook, Telegram, iMessage. 3) I will avoid rec709-A whenever possible but I will ONLY bake my export with rec709-A when uploading to Whatsapp status as all the methods above will not work. Idk if you yourself use all the social media or messaging apps I've mentioned above but you can try it yourself. I've never tried it on TH-cam. Maybe it work slightly different. I can't tell why all of this is working but you can read on reddit where they discuss about NLC tag, ICC standard and stuff. This workaround is only for an Apple devices as for other OS I don't really know.
Could you do a Part 2 and show us what is happening with the wheels? I'm getting lost in a few parts. I know this is a more advanced episode but would still be nice to see.
Thank you so much for sharing this content. You make it super easy to understand these things and my images are on yet another level! Feels like bringing a dream to life with the sat node.
Hi Culle, thanks again for your knowledge. I would like to know why you have switched from the contrast slider to lift, gama and gain? Would be nice to heard from you, what you think about contrast. I feel you have a differently than your previews videos.
I love this node setup overall! However, I was initially getting a weird artifact going on where if my blacks were at all crushed, I would get crazy color artifacting, when I pulled the blacks down weird color ghosting would happen. I eventually figured out that this was the result of the SAT node being in HSV mode. I switched it to HSL and the artifacting went away. I have no idea why this would happen because only channel 2 was turned on, so theoretically there should be no difference, but in case anyone else encounters this issue just switch your SAT node to HSL mode
Remember: without a videocard in macos everything that is color managed goes through colorsync. That means if your viewers is using display profiles it means what you're seeing is a p3/gamma 2.2 image. Anything you export will be in relation to that. If you export a rec.709/gamma 2.4 clip, it will be lifeless (less sat and lifted shadows). That's because rec.709 is a smaller colorspace and gamma 2.4 will look incorrect since there is a gamma mismatch between the viewer and the exported video. Resolve will render a gamma 2.4 image inside a 2.2 container even if the color tags suggest its 2.4. To bypass all this bull crap use a videocard. There will be no dominant display colorspace that has everything going through it. There is only what you only output. So if you output rec.709/gamma 2.4 what you export won't be in relation to the display colorspace, it will simply export and display rec.709/gamma 2.4
Some good info here! Best approach is always to use a dedicated output card (as I’m doing here for my reference monitor), but the macOS viewing transform accounts for the factors you’re calling out to give a best possible reproduction on the GUI!
Amazing information here as always. Quick question though, how come you never mention rec 709 A instead of gamma 2.2 or 2.4? I have always set it to rec 709 A and never had any issues as far viewing the same color when the video is exported out.
Wondering this as well , i think rec709-a is only valid if you grading at mac monitor , but he grades on his calibrated gamma 2.4 monitor , i might be wrong
'calibrated gamma 2.4 monitor' make a lot of sense then. so then I guess that choice is dependent on what monitor you're using and what its calibrated method is@@zweiche
@@amadoucissefit indeed , and he says this year he ll be grading for gamma 2.2 since most of content is either on phone or tablet or pc monitors , even my lg oled i am using at 2.2 , bcs 2.4 is too dark for my everyday use
Are there any more needs for spli toning, after you set your balance to certain mood? Like in the last still example, after your set your balance to cool tone
Love me split toning, but that’s something I almost always incorporate into my overall look, rather than something I do while grading. Thats the case here - look already has split toning built in that I feel like works well!
Just a heads up, your input and output node information was covered by your talking head and a lot of new watchers might be lost without that information.
If I remember correctly , CK's earlier video (How to Color Manage using Nodes in DaVinci Resolve) from about 3 months ago covers some of that using "groups" in resolve
Hi Cullen, thanks for the update. 🎉 Are there still cases where you consider Linear Gain for Exposure within your Nodetree or do you get the same result by working LGG and Offset?
Linear Gain is good when you want to work with stops -> like 2x, 4x, the light values. Its getting harder when you want to increase by 1/3 stops for example. Then the HDR Tool might be your best option. But you would get 100% the same result with LGG and Offset. Its just a different way to work the values.
I actually feel like I am bit ahead of you! I adapted your earlier node graph to a version of what you are using now some time ago because the modification seemed to help with adjusting saturation. (I actually used a basic node graph of four: EXP, CON, BAL and SAT, but I can see that combining exposure and contrast in one node also works). Once again, some great insights and helpful advice. Thank you.
I am interested into curves in the secondary lane. If the process is non-destructive - what difference do it make, when it's better to have it up, when in the 2nd-ary lane?
If you place secondaries in a normal corrector node after primaries, and then go back to adjust primaries later, that can affect the secondaries because you’re changing their starting point. Whereas if you place secondaries in the bottom stack of the parallel mixer, the secondaries are sidestepping the primary corrections and working from the pre-graded shot. Hope that helps!
@@conortychowski yes but in context of tonal-curves (or lifting shadows curve) shouldn't this be based on the corrected exposure instead of the original one?
Ah I see, I’d imagine that’s just for organisational purposes to keep any secondary style corrections at the bottom. I don’t believe there’d be any notable difference on the image if it was done at the top
So overall this workflow makes a lot of sense and I'm setting up my timeline for it. Only snag I've run into is that Blending Modes are expecting a 0-1 range for their operations. So for example if I incorporate motion graphics or film grain / grunge effects and I use a CST to get the rec709 footage to DWG before using one of the Voyager LUTs I bought from you (awesome job by the way!) then the blending modes no longer work correctly. All the values go wonky. So I wouldn't be able to do a timeline-wide multi camera grade like you've shown here, I'd have to use CSTs for each camera then add in a Voyager lut, then convert to Rec709 with the CST all in the Group-Post section. Which isn't ideal because if I wanted to make any timeline level grade changes I'd have to do it for each separate group. So does anyone have any ideas on how to use blending modes while keeping in this workflow? Thanks all!
I am not fully grasping how to approach the saturation node, here the adjustments seems similar to those made when working with contrast ratio. Does the composite mode make all the difference here? Does Cullen have another video covering this?
it's not the composite mode, but it is the node's color space. Right-click the node, go to color space, select HSV. Then right-click the node again and go to channels and disable channels 1 and 3.
Is it a good rule of thumb to stick with outputting 709-a while on an iMac? I was trying out the Viewing Transform Lut and outputting 2.2 but getting poor results.
Same here; when I watch my video in QuickTime (Gamma 2.2), it looks completely different from what I see in DaVinci. It’s way brighter and really washed out. Did you end up finding a good solution?
Hi Cullen! Thanks for the best grading content on YT. The link for the viewing transform doesn't work. There are no fields to input my email as it says. Would love this viewing transform.
Two questions, how do I update just the photo of my template node graph I've previously saved - or will it always be on the picture of the current project? Second, why not leave the first node on luminosity? Always great info!!
Concerning your second question, he showed in one of the examples that the luminosity composite mode altered the saturation too much. So I guess it's just more about feeling than really pure technical aspect to choose to use that composite mode or not.
You can put a big Text into the timeline, create a compund node, apply your power grade and then save it as power grade again. The photo of that new powergrade will now be the Text or anything you want really
That's a good question! When we flip the color space into HSV, our RGB channels no longer represent red, green, and blue. Because we're in HSV, red is now hue, green is now saturation, and blue is now value (or brightness). And since we're looking to only perform a saturation adjustment, we only want to affect the 2nd channel, or green channel -- which in HSV is our saturation channel.
I feel like ive learned so much from this channel. its funny most of the changes that you're making in this video i also implemented prior. Just using your "general way" of doing things i made these adjustments my own because they work for me. Thanks for not teaching us which knob to turn but also the reasoning and thought process.
Dear Cullen, i love your videos and your approach. Given that you (mostly) work with RAW footage, i believe that most of us are working with let's say more basic cameras (personaly i own an R6m2), can you show some tips on how you edit videos using such 10bit footage? Some of us struggling when color grading happens and it's not clear whether we should use CST, Resolve color magement or any LUT for the basic corrections.
Cullen, I absolutely appreciate you taking the time to create this content, but more often than not, the things you mention will be linked below, never are. Just the same 4 links you leave on every single video.
Hi @CullenKelly. I am learning about these new duct's by monoxides, and some by pixel tools. Lot of these actually providing easy to use sliders for saturation and density. I saw your videos on saturation, that show the HSV method to work with saturation in a more "cinematic" way. However, these dicta's are also providing same kind of granular control on individual colors like red cyan etc. I have wondered how would you go about doing that without these dctls. It would be amazing if that is something you could throw some light on. Either way, do you use those dctls kind of dctls in your workflow when grading professionally. Please advise.
I really enjoy looking at other colorist's node trees and their strengths and weaknesses. No approach is absolute. I do have a question about the underexposed lady with the greenery background. You attempted to recover the highlights peaking in the background by making a curve adjustment in the Prim node. Would you consider using the other Parallel node line you have (unaffected by the exposure alterations) to bring that back down? When I've graded footage with great contrast differences like that shot, I would add a preliminary node using HDR tools to bring down the exposure in those highlights.
I think Cullen didn't try to recover, more like softening the top and highlights while controlling saturation. But yes. HDR Tools could help here or you could pull a luma-key for the upper 1/3 of the luma range and soften it down to the 2/3 part of the luma range. Then you could use gamma, gain or offset to bring back those near blown-out parts.
Gamma 2.2 gets tagged as 1-4-1 by default .. However many codecs encoded at gamma 2.2, and being uploaded to platforms like TH-cam, can benefit by tagging as 1-1-1 to prevent the platform from applying an additional OOTF from 2.2 to 2.4. I’d say if you are encoding Gamma 2.2 and you upload to TH-cam.. If you see a shift in the online rendering .. Tag gamma on the delivery page as REC709… this will overwrite the tags as 1-1-1 .
Dear Cullen, As we all know, the order of nodes affects the output, but what is the order of processing when the nodes are processed separately at the timeline level and at the group level, apart from the clip level, as you are doing here? Assuming the processing described here is correct, does that mean that the order is "group > clip > timeline"? Thanks in advance.
Would love to know about the tradeoffs between HSV channel 2 vs a Denisity or SubSat DCTL? I would also love to know your thoughts on HGL, would you ever see yourself switching to HDR deliverables for web content? What would be the deciding factors? I personally didn't want to deal with Mac Gamma shift (although that would also be slightly mitigated by targeting gamma 2.2), and prefer to target HGL (Rec2100), as youtube can handle that, should be backwards compatible with Rec709, opens up more on better displays (iPhones, iPads, etc). I guess I am just curious on your thoughts on HDR in general?
Are you using the Lift Gamma Gain knobs on the upper left of the Resolve panel? Or using the wheels? I’m using the Micro Panel. I have to hit the offset button to make the rightmost wheel work as offset.
hey hey, i ve just fallen into this amazing video, whats the difference between your lut and setting the timeline to rec709A for mac os people on apple screens or imacs? been using gamma 2.4 as i learnd from you, but, the content has been washing out on social media browsers, this thing about 1-2-1 and 1-1-1 .. should we make gamma 2.2 or rec709 A. confused , thanks again
just use the CST and set it from Apple Log to your working colorspace at the beginning of your nodetree and CST from your working colorspace to your output colorspace at the end of your nodetree. Everything else you need to do, you can use Cullens nodetree for it.
question for you mr kelly.. how to fix , colores being washed out after uploading my videos on youtube , they looks tottaly diffrent then they were in davinci , any fix ?working on windows pc btw
dont fix whats not broken. forget this lut and gamma 2.2 bs. just stick with rec709-a workflow, dont forget to check "use mac color profiles" in settings. any luts must be made for SPECIFIC mac display as they are not the same across devices
You had already uploaded a video of your CKC Macos lut. Should we follow the entire process from the last video? Or do we only activate this lut in the Color Viewer section?
okay, I found little hint in the official manual: "Color Space: Debayering .braw data requires choosing a color space to convert the raw signal into. Bear in mind that the color space you choose is merely a starting point for further correction. There is no requirement that you choose one or the other color space for any given workflow, and all settings will yield high-quality image data suitable for further color correction. You should choose the color space that provides the most pleasing starting point for your particular project."
What am I doing wrong, there is absolutely no difference when I put the Transform LUT in the Color viewer lookup table slot that Cullen uses, but there IS a difference when I put it in the Video monitor lookup table slot... I have all other settings identical and I use an Apple display like Cullens...
Please, if I set up CM in nodes, what will timelime colour space set in DWG vs Rec.709 do exactly? I did some tests and if I set timelime colour space to DWG and put a node in between CST nodes, HDR tools colour temp will look different if I turn off those CST nodes and leave HDR tool node on. But if I leave CST nodes on with HDR tool node also on and change the timeline colour space, there is no visual change. Thank you very much.
Hey Adam! I think what you're referring to is the switch from gamma 2.4 to 2.2? Davinci color management is something I still use, and definitely a viable solution to color manage your projects. The change I'm making is simply the gamma adjustment to better target deliveries for web. Hope that helps!
Hi Cullen, i was actually referring to what you were saying at around 9:35 in your video. Color managing in nodes, and using color space transform, instead of setting the input and output color space in the project settings. @@CullenKelly
Really wish I could see you working the controls when you're manipulating these images and narrating. Am I missing something? Otherwise, stoked to check out the rest of your content.
Hello Cullen! I have never heard you talking about changing the gamma to linear when setting your balance. I have always thought you used offset when balancing an image. So whats the benefit of having the gain in linear for balancing?
I'd recommend giving it a try! There a quite a few operations that can work better when done in linear. I find it to be a cleaner approach than using the offset wheel. Throw a node into linear and try playing with gain. You might also try using curves in linear to balance your image. Compare it against a balance adjustment using offset. I think you'll like the result.
@@CullenKellyThanks a lot Cullen for answering. And thanks for putting a lot of effort into this channel. Would you mind one more question? In an earlier video, you are talking about decreasing saturation using the HSV method with the gain wheel. But in this video you are using the gamma wheel. Curious to know the difference. All the best!
You are digging yourself in all these complications, by using Davinci Wide Gamut and the wheels. You are trying to do a simple gain or lift operation, and there's NO WAY to achieve that in DWG without the curves. Thus, all the mumbo jumbo.
But.. but.. it was watching your channel that convinced me to move to a colour managed workflow in the first place! Plus switching to gamma 2.2?
I wasn't emotionally prepared for this episode 😅
You have changed my colour grading life.
I've watched so many videos from different people too but they don't compare
Cullen, I have seen your resistance to the HDR tool and if using panels it is not as easy as switching with the mouse.
But a lot of the stuff that you are implementing by using composite mode and or changing the colorspace of nodes and turning off channels etc. is how some things work in the HDR tool without modification - so the ease of use or set-up is absolutely faster.
For instance -
On the HDR tool the contrast textboxes for instance doesn't affect the saturation when using it - so adding contrast will only affect the luminosity. And if used with the pivot will move the middle grey point - raising will give you more stops below and lowering will give you more stops above. Which will raise or lower the exposure as well.
The contrast differences and how it is used is really evident if you push a really negative or push it positive and especially in the negative on the HDR tool see the color remain while the luminosity is lessening.
I was wondering why my image looked lifeless using the contrast in hdr tools. I prefer the primaries
@@shueibdahir the primaries pushes the saturation with the curve - you might like the HDR contrast when you then adjust the saturation on a different node to your taste. If you use a grey ramp and compare the actual contrast it is pretty much identical - it just keep the colors the same and doesn't pump more sat into it.
but he did mention on some clips he preferred the sat that's added in the offset, and i think he prefers the control that he can just turn it on/off
May be is it coz of the switiching between hdr and primaries,like if u have 100 shots, u may find urself resting on hdr but u wanted to adjust in the primaries.
In previous videos from a while ago, he switches the exposure node to linear. With this new process it looks like the gamma on the Prim node is not set to linear? Is that correct?
I didn't know I could set the compositing mode to luminosity! That's a game changer. Coming from a long photography career that is a key thing to do in Photoshop. Didn't know it was an option in Davinci.
Dear Cullen,
can you please also provide the Link to the macOS Viewing Transform you talk about?
Thank you in advance!
Edit: Thank you for adding the link Cullen
It looks like he put the link in the video description.
@@JGphotoyt anyone else having problems downloading it there is nowhere I see to download
It‘s literally the first link in the description of this video. Just click the very first link, fill in your name and email address and it‘ll get sent to you. It ain’t that hard.
@@SebastianFriedrichVisuals Cullen provided the link a few hours after releasing the video.
@@SebastianFriedrichVisuals I did but the link didn’t work for me there was nowhere to leave my email for some reason.
any chance you'll upload a video of what you did in the wheels as well as the outcome?
Yea, I have no idea what is happening with the wheels in the primary node. Also, how is he adjusting saturation in each Lift / Gamma
Glad your starting anew, I will be following closely trying to move my color grading to another Level UP
Thanks for sharing this new workflow! I love how you explained your reasoning behind each node.
Thorough and clear explanation as per usual. i do have a couple of doubts; If there is only one camera used in a project and did not want to make groups, where would you put the CST nodes? Also is there a difference in applying the macOS VT as you did in this video, compared to doing in a node as you showed in a previous video?
I saw some image-breaking artifacts with this new approach.
7:54 - speculars are too colorful, have obvious shapes. "Before" was much more aesthetically pleasing as well.
16:50 - visible clipping of the lamp. It was in the original image, but something in the pipeline had to be done to resolve this issue.
Thanks Fedor! To be fair, these artifacts are image-inherent and not caused by anything I’m doing. Though certainly the type of details you’d want to address before delivering!
What is the difference in using gamma vs gain for saturation in this method?
Hey Cullen, your videos have taught me so much! Thanks a ton for making them!
Two questions on this one:
1. Since you've combined your exposure and ratio node, and your new primary node's composite mode is set to luminosity, when you change exposure with offset or gain, don't you get weird colors?
2. On the balance node with linear gamma, would the results be more accurate with the composite mode set to color? Would that preserve the tones better?
wow, this is very similar to how I work! Thanks for the luminosity composite mode idea, I love that.
you could save time on the first node by having a layer mixer set to luminosity and just have the extra node disabled by default, turning it on and off is a lot quicker than going through the right click menu each time
True! As long as you don’t mind the extra space this occupies in the node graph 😉
Your MacOS viewer transforms is the best thing ever happen on internet for startup like myself. I got almost 1:1 luminance reproduction on all my mobile devices.
is there a guide on this ? Do we still use rec709-a as output on project settings and gamma 2.2 on cst output node ? i am lost
@@zweiche This what I get from my own testing and it works across all social media including QuickTime player for this time being.
1) On node basis, you'll always want to set your gamma to 2.2 and add "CKC Apple viewer LUT" on color viewer.
2) On export always set you gamma to 2.4. This specific hack is to make sure that QuickTime will always read your footage as gamma 2.4 and social media will always take the 2.2 input instead including TikTok, Instagram, Facebook, Telegram, iMessage.
3) I will avoid rec709-A whenever possible but I will ONLY bake my export with
rec709-A when uploading to Whatsapp status as all the methods above will not work.
Idk if you yourself use all the social media or messaging apps I've mentioned above but you can try it yourself. I've never tried it on TH-cam. Maybe it work slightly different.
I can't tell why all of this is working but you can read on reddit where they discuss about NLC tag, ICC standard and stuff.
This workaround is only for an Apple devices as for other OS I don't really know.
@@zweiche Yeah, I think we need Cullen to clear this up a bit.
@@VictorRiley Agree! I'm using rec709-a…
@@zweiche You are not alone, I'm so lost too 😅
Really helpful thank you! God bless
I just wish I could see the changes you were making as you make them
Doing graduation curves via luminosity in Photoshop for years but never knew how to do it in davinci👍
Could you do a Part 2 and show us what is happening with the wheels? I'm getting lost in a few parts. I know this is a more advanced episode but would still be nice to see.
Thank you so much for sharing this content. You make it super easy to understand these things and my images are on yet another level! Feels like bringing a dream to life with the sat node.
Hi Culle, thanks again for your knowledge. I would like to know why you have switched from the contrast slider to lift, gama and gain? Would be nice to heard from you, what you think about contrast. I feel you have a differently than your previews videos.
I love this node setup overall! However, I was initially getting a weird artifact going on where if my blacks were at all crushed, I would get crazy color artifacting, when I pulled the blacks down weird color ghosting would happen. I eventually figured out that this was the result of the SAT node being in HSV mode. I switched it to HSL and the artifacting went away. I have no idea why this would happen because only channel 2 was turned on, so theoretically there should be no difference, but in case anyone else encounters this issue just switch your SAT node to HSL mode
hey cullen can you shed some light on the importance of changing your sat node to HSV & channel 2
Remember: without a videocard in macos everything that is color managed goes through colorsync. That means if your viewers is using display profiles it means what you're seeing is a p3/gamma 2.2 image. Anything you export will be in relation to that.
If you export a rec.709/gamma 2.4 clip, it will be lifeless (less sat and lifted shadows). That's because rec.709 is a smaller colorspace and gamma 2.4 will look incorrect since there is a gamma mismatch between the viewer and the exported video. Resolve will render a gamma 2.4 image inside a 2.2 container even if the color tags suggest its 2.4.
To bypass all this bull crap use a videocard. There will be no dominant display colorspace that has everything going through it. There is only what you only output. So if you output rec.709/gamma 2.4 what you export won't be in relation to the display colorspace, it will simply export and display rec.709/gamma 2.4
Some good info here! Best approach is always to use a dedicated output card (as I’m doing here for my reference monitor), but the macOS viewing transform accounts for the factors you’re calling out to give a best possible reproduction on the GUI!
Amazing information here as always. Quick question though, how come you never mention rec 709 A instead of gamma 2.2 or 2.4? I have always set it to rec 709 A and never had any issues as far viewing the same color when the video is exported out.
Wondering this as well , i think rec709-a is only valid if you grading at mac monitor , but he grades on his calibrated gamma 2.4 monitor , i might be wrong
'calibrated gamma 2.4 monitor' make a lot of sense then. so then I guess that choice is dependent on what monitor you're using and what its calibrated method is@@zweiche
@@amadoucissefit indeed , and he says this year he ll be grading for gamma 2.2 since most of content is either on phone or tablet or pc monitors , even my lg oled i am using at 2.2 , bcs 2.4 is too dark for my everyday use
Are there any more needs for spli toning, after you set your balance to certain mood? Like in the last still example, after your set your balance to cool tone
Love me split toning, but that’s something I almost always incorporate into my overall look, rather than something I do while grading. Thats the case here - look already has split toning built in that I feel like works well!
Just a heads up, your input and output node information was covered by your talking head and a lot of new watchers might be lost without that information.
If I remember correctly , CK's earlier video (How to Color Manage using Nodes in DaVinci Resolve) from about 3 months ago covers some of that using "groups" in resolve
Hi Cullen, thanks for the update. 🎉 Are there still cases where you consider Linear Gain for Exposure within your Nodetree or do you get the same result by working LGG and Offset?
I find the slider on the HDR tool below the offset to be the most organic in feel and accurate for stops.
Linear Gain is good when you want to work with stops -> like 2x, 4x, the light values. Its getting harder when you want to increase by 1/3 stops for example. Then the HDR Tool might be your best option. But you would get 100% the same result with LGG and Offset. Its just a different way to work the values.
I love that you show full screen image while doing grading. Thank you.
Thank you so much for sharing these tricks! Will defo try them out in my new projects :)
Cheers!
Fantastic like always! Thank you, Cullen!
Great video Cullen! The detail for some of the primary nodes was much needed
Hey Cullen, I’m curious why you are returning the colour space in the timeline level rather than in the group post clip?
I actually feel like I am bit ahead of you! I adapted your earlier node graph to a version of what you are using now some time ago because the modification seemed to help with adjusting saturation. (I actually used a basic node graph of four: EXP, CON, BAL and SAT, but I can see that combining exposure and contrast in one node also works). Once again, some great insights and helpful advice. Thank you.
17:08 clip it boys...seriously. thanks for the update
Great! What about Rec 709-A insted of gamma 2.4 for macbooks? is the only gamma that gives me accurate color
I am interested into curves in the secondary lane. If the process is non-destructive - what difference do it make, when it's better to have it up, when in the 2nd-ary lane?
If you place secondaries in a normal corrector node after primaries, and then go back to adjust primaries later, that can affect the secondaries because you’re changing their starting point. Whereas if you place secondaries in the bottom stack of the parallel mixer, the secondaries are sidestepping the primary corrections and working from the pre-graded shot. Hope that helps!
@@conortychowski yes but in context of tonal-curves (or lifting shadows curve) shouldn't this be based on the corrected exposure instead of the original one?
Ah I see, I’d imagine that’s just for organisational purposes to keep any secondary style corrections at the bottom. I don’t believe there’d be any notable difference on the image if it was done at the top
@@conortychowski yes thank you
So overall this workflow makes a lot of sense and I'm setting up my timeline for it. Only snag I've run into is that Blending Modes are expecting a 0-1 range for their operations. So for example if I incorporate motion graphics or film grain / grunge effects and I use a CST to get the rec709 footage to DWG before using one of the Voyager LUTs I bought from you (awesome job by the way!) then the blending modes no longer work correctly. All the values go wonky. So I wouldn't be able to do a timeline-wide multi camera grade like you've shown here, I'd have to use CSTs for each camera then add in a Voyager lut, then convert to Rec709 with the CST all in the Group-Post section. Which isn't ideal because if I wanted to make any timeline level grade changes I'd have to do it for each separate group.
So does anyone have any ideas on how to use blending modes while keeping in this workflow? Thanks all!
Sir please make a list for color graiding for movies because it is very rare on TH-cam
Big love for you ❤❤❤
I am not fully grasping how to approach the saturation node, here the adjustments seems similar to those made when working with contrast ratio. Does the composite mode make all the difference here? Does Cullen have another video covering this?
it's not the composite mode, but it is the node's color space. Right-click the node, go to color space, select HSV. Then right-click the node again and go to channels and disable channels 1 and 3.
Very helpful. Thank you.
friend is there a guide which shows you how much situration to ginve? for not going too far?
Your vectorscope can be a great guide for this!
@@CullenKelly can in cectorscope add a limit line? To go not gar?
Is it a good rule of thumb to stick with outputting 709-a while on an iMac? I was trying out the Viewing Transform Lut and outputting 2.2 but getting poor results.
Same here; when I watch my video in QuickTime (Gamma 2.2), it looks completely different from what I see in DaVinci. It’s way brighter and really washed out. Did you end up finding a good solution?
Hi Cullen!
Thanks for the best grading content on YT.
The link for the viewing transform doesn't work. There are no fields to input my email as it says. Would love this viewing transform.
Could you please provide a link to download the power grade for your node template
If you’re that lazy maybe you shouldn’t be attempting color grading…
you've helped me so much THANK YOU
Two questions, how do I update just the photo of my template node graph I've previously saved - or will it always be on the picture of the current project?
Second, why not leave the first node on luminosity?
Always great info!!
Concerning your second question, he showed in one of the examples that the luminosity composite mode altered the saturation too much. So I guess it's just more about feeling than really pure technical aspect to choose to use that composite mode or not.
You can put a big Text into the timeline, create a compund node, apply your power grade and then save it as power grade again. The photo of that new powergrade will now be the Text or anything you want really
thank you for the update, it is very useful 🙏
What does setting saturation node to HSV + Channel 2 really do? Sorry if you've explained in other videos.
That's a good question! When we flip the color space into HSV, our RGB channels no longer represent red, green, and blue. Because we're in HSV, red is now hue, green is now saturation, and blue is now value (or brightness). And since we're looking to only perform a saturation adjustment, we only want to affect the 2nd channel, or green channel -- which in HSV is our saturation channel.
I feel like ive learned so much from this channel. its funny most of the changes that you're making in this video i also implemented prior. Just using your "general way" of doing things i made these adjustments my own because they work for me. Thanks for not teaching us which knob to turn but also the reasoning and thought process.
The balance node's gamma is set to Linear?
That's right! Give it a test in your own grades and let me know what you think!
Man, this was a good one.
Dear Cullen, i love your videos and your approach. Given that you (mostly) work with RAW footage, i believe that most of us are working with let's say more basic cameras (personaly i own an R6m2), can you show some tips on how you edit videos using such 10bit footage? Some of us struggling when color grading happens and it's not clear whether we should use CST, Resolve color magement or any LUT for the basic corrections.
You should use them according to how you think your project will unfold and think which makes it more productive with the time you have for grading it
@@Visethelegend but is it "proper" that in the same project you may have to use different approach as you suggest?
I think you missed adding the link to the Color viewer lookup table in your description.
Cullen, I absolutely appreciate you taking the time to create this content, but more often than not, the things you mention will be linked below, never are. Just the same 4 links you leave on every single video.
Hi @CullenKelly. I am learning about these new duct's by monoxides, and some by pixel tools. Lot of these actually providing easy to use sliders for saturation and density. I saw your videos on saturation, that show the HSV method to work with saturation in a more "cinematic" way. However, these dicta's are also providing same kind of granular control on individual colors like red cyan etc. I have wondered how would you go about doing that without these dctls. It would be amazing if that is something you could throw some light on. Either way, do you use those dctls kind of dctls in your workflow when grading professionally. Please advise.
I really enjoy looking at other colorist's node trees and their strengths and weaknesses. No approach is absolute.
I do have a question about the underexposed lady with the greenery background. You attempted to recover the highlights peaking in the background by making a curve adjustment in the Prim node. Would you consider using the other Parallel node line you have (unaffected by the exposure alterations) to bring that back down?
When I've graded footage with great contrast differences like that shot, I would add a preliminary node using HDR tools to bring down the exposure in those highlights.
I think Cullen didn't try to recover, more like softening the top and highlights while controlling saturation. But yes. HDR Tools could help here or you could pull a luma-key for the upper 1/3 of the luma range and soften it down to the 2/3 part of the luma range. Then you could use gamma, gain or offset to bring back those near blown-out parts.
Are you tagging your gamma 2.2 as 1-1-1 or 1-2-1 on export?
Gamma 2.2 gets tagged as 1-4-1 by default .. However many codecs encoded at gamma 2.2, and being uploaded to platforms like TH-cam, can benefit by tagging as 1-1-1 to prevent the platform from applying an additional OOTF from 2.2 to 2.4. I’d say if you are encoding Gamma 2.2 and you upload to TH-cam.. If you see a shift in the online rendering .. Tag gamma on the delivery page as REC709… this will overwrite the tags as 1-1-1 .
Hi, good Video as always. I have a doubt what is the difference between HSV and HSL because I feel that with HSL saturation is less aggressive. Thanks
If operating exposure with linear gain, how do the adjustments change to exposure and ratio combined in the primary node?
Dear Cullen,
As we all know, the order of nodes affects the output, but what is the order of processing when the nodes are processed separately at the timeline level and at the group level, apart from the clip level, as you are doing here?
Assuming the processing described here is correct, does that mean that the order is "group > clip > timeline"?
Thanks in advance.
Group Pre-Clip>Clip>Group Post-Clip which can be processed by Render Cache Color Output and then >Timeline which is outside the cache.
@@willemnagtglas4569 Thank you!
Would love to know about the tradeoffs between HSV channel 2 vs a Denisity or SubSat DCTL? I would also love to know your thoughts on HGL, would you ever see yourself switching to HDR deliverables for web content? What would be the deciding factors? I personally didn't want to deal with Mac Gamma shift (although that would also be slightly mitigated by targeting gamma 2.2), and prefer to target HGL (Rec2100), as youtube can handle that, should be backwards compatible with Rec709, opens up more on better displays (iPhones, iPads, etc). I guess I am just curious on your thoughts on HDR in general?
Man of the people!
Are you using the Lift Gamma Gain knobs on the upper left of the Resolve panel? Or using the wheels? I’m using the Micro Panel. I have to hit the offset button to make the rightmost wheel work as offset.
hey hey, i ve just fallen into this amazing video, whats the difference between your lut and setting the timeline to rec709A for mac os people on apple screens or imacs? been using gamma 2.4 as i learnd from you, but, the content has been washing out on social media browsers, this thing about 1-2-1 and 1-1-1 .. should we make gamma 2.2 or rec709 A. confused , thanks again
Cullen, give us a setup for IPhone 15 Apple log footage!
just use the CST and set it from Apple Log to your working colorspace at the beginning of your nodetree and CST from your working colorspace to your output colorspace at the end of your nodetree. Everything else you need to do, you can use Cullens nodetree for it.
question for you mr kelly.. how to fix , colores being washed out after uploading my videos on youtube , they looks tottaly diffrent then they were in davinci , any fix ?working on windows pc btw
When I export 2.2 it looks different in davcini than it does on TH-cam. How do I fix that? I’m used to using rec709-a
dont fix whats not broken. forget this lut and gamma 2.2 bs. just stick with rec709-a workflow, dont forget to check "use mac color profiles" in settings.
any luts must be made for SPECIFIC mac display as they are not the same across devices
You had already uploaded a video of your CKC Macos lut. Should we follow the entire process from the last video? Or do we only activate this lut in the Color Viewer section?
What about RAW footage like .braw? It's good to set DaVinci Wide Gamut and DaVinci Intermediate in Camera Raw tab instead of making CST Input node?
okay, I found little hint in the official manual: "Color Space: Debayering .braw data requires choosing a color space to convert the raw signal into. Bear in mind that the color space you choose is merely a starting point for further correction. There is no requirement that you choose one or the other color space for any given workflow, and all settings will yield high-quality image data suitable for further color correction. You should choose the color space that provides the most pleasing starting point for your particular project."
hello men, thanks for this type of videos. Do you think is better your lut and use 2.2 gamma than rec 709 A?
What software do you use to do a remote grading session?
I like to grade each frame of my movies individually in Photoshop.
can i use the voyager luts with this workflow?
Absolutely!
What am I doing wrong, there is absolutely no difference when I put the Transform LUT in the Color viewer lookup table slot that Cullen uses, but there IS a difference when I put it in the Video monitor lookup table slot... I have all other settings identical and I use an Apple display like Cullens...
Do you have Hide UI Overlays checked? This removes the LUT.
@@willemnagtglas4569 My God, that was it, thank you so much... Would have NEVER figured that out... 😂
Please, if I set up CM in nodes, what will timelime colour space set in DWG vs Rec.709 do exactly? I did some tests and if I set timelime colour space to DWG and put a node in between CST nodes, HDR tools colour temp will look different if I turn off those CST nodes and leave HDR tool node on. But if I leave CST nodes on with HDR tool node also on and change the timeline colour space, there is no visual change. Thank you very much.
Why is it better to put gamma 2.2 for web and laptop instead of 2.4 (I'd always put 2.4 before)?
Same question!
Hello! So are you suggesting that we shouldn't use davinci color management anymore or is it only for the purpose of these videos?
Hey Adam! I think what you're referring to is the switch from gamma 2.4 to 2.2? Davinci color management is something I still use, and definitely a viable solution to color manage your projects. The change I'm making is simply the gamma adjustment to better target deliveries for web. Hope that helps!
Hi Cullen, i was actually referring to what you were saying at around 9:35 in your video. Color managing in nodes, and using color space transform, instead of setting the input and output color space in the project settings. @@CullenKelly
neutral grey paint...
live in midwest, states. anyone know brand/codes/specs from lowe's, home depot, sherwin williams, or walmart? if not, amazon?
10/10 video 👏🏻
weird question, but do you use adobe podcast to clean up your vocals?
I feel like I can almost always hear when someone does lol
please make a youtube setup,,,video, and , light and color grade for your youtube channel/
Wait but all of this is happening within DWG right?
yes, CST is in the group & timeline
Really wish I could see you working the controls when you're manipulating these images and narrating. Am I missing something? Otherwise, stoked to check out the rest of your content.
Hello Cullen! I have never heard you talking about changing the gamma to linear when setting your balance. I have always thought you used offset when balancing an image. So whats the benefit of having the gain in linear for balancing?
I'd recommend giving it a try! There a quite a few operations that can work better when done in linear. I find it to be a cleaner approach than using the offset wheel. Throw a node into linear and try playing with gain. You might also try using curves in linear to balance your image. Compare it against a balance adjustment using offset. I think you'll like the result.
@@CullenKellyThanks a lot Cullen for answering. And thanks for putting a lot of effort into this channel. Would you mind one more question? In an earlier video, you are talking about decreasing saturation using the HSV method with the gain wheel. But in this video you are using the gamma wheel. Curious to know the difference. All the best!
We're using the Adobe Podcast AI aren't we?
❤❤
I don’t see what you are doing if you stay in fullscreen.
You are digging yourself in all these complications, by using Davinci Wide Gamut and the wheels. You are trying to do a simple gain or lift operation, and there's NO WAY to achieve that in DWG without the curves. Thus, all the mumbo jumbo.