YouTube on color grading and LAB


Dan Margulis
 



On Apr 15, 2021, at 11:31 AM, Fabrizio Cacopardo <fabrizio.cacopardo@...> wrote:

First thing first, I'd like to thank Dan for being so generous and always dedicating his time to our work, commenting every single image. A demonstration from him is worth more than a hundred tutorials on Youtube. 

I agree with Fabrizio with respect to our field, but I don’t want to knock YouTube generally. A scad of instructional videos is available on just about any non-graphics topic. I don’t do household repairs any more without checking the videos that show how to proceed. My knowledge of classical music wasn’t bad at the start of the pandemic but it’s more impressive now, what with a dozen recordings available of every major work, and two or three dozen of the standards, not to mention worthwhile works of some lesser-known composers. And I certainly wish that YouTube had been able to prep me when I was teaching in countries where I didn’t speak the language. Anyway, here’s my impression of YouTube as it relates to our field.

Earlier this year we had a thread about the definition of “color grading”, for which I surfed YouTube in search of a consensus. I didn’t find one, but I found at least a hundred videos trying to explain what the speaker thought it was and how to do it. Several of these had a huge number of views, as in, ten times or more the rate of viewing of some of the more popular Photoshop tutorials.

After going through these in January I wondered to myself whether anybody was posting videos about the use of LAB, whether for single photos or in video production. I needn’t have wondered: there were more than a hundred in English alone. Having watched a few, I realized that similar tutorials were available in six other languages in which I know most of the words pertaining to color and so could search for them. So, between the graphic demonstrations and Google Auto Translate’s valiant attempts to add English subtitles, I was able to follow them, although after one experience each with Turkish and Vietnamese I decided to limit myself to the other seven.

Having logged more than 200 possibilities (because I logged titles that looked like they might have LAB content even if it wasn’t stated explicitly) I narrowed it down further after viewing the first dozen or so. It became clear then that in the last two or three years the technical quality of the videos (as opposed to the instructor’s presentation) had increased dramatically. So I decided to only look at stuff that was either 2018 or later, or had an extremely high number of views. I learned a lot about the current state of education in our field, not always for the better.

Back in the Photoshop book era, a huge percentage of the valuable information came out of the United States. That made sense. It hasn’t been possible to make much money on the books themselves for more than ten years, but they could also be used to develop a following for classes or conferences. 

Not so any longer. How does one make a living as an instructor? Evidently the new model is, open up a YouTube channel, release a gang of free videos, get people to subscribe, and ask for donations. Knowledge has become shareware.

The good consequence is that more content is now available in languages other than English. In fact, the majority of the English-language content appears to come from individuals who did not learn their English in the United States.

The first bad consequence are that to keep the subscribers happy and contributing the instructor has to keep releasing a stream of videos, many of which will necessarily focus on topics they know little about.

The second is that showmanship and faked enthusiasm becomes more important than content, meaning that guys like me would never make it in such a setting. No matter how stupid the technique is, it must be described as either 1) Mind-blowing! 2) Awesome! 3) Killer! 4) Incredible! or 5) Fantastic!

I’m not going to rate any videos or their hosts, as some of them are my friends. If you must have one name, I’d say that if you know who Андрей Журавлев is, his videos are worth seeing, but they aren’t in English. In fact, the very language that he speaks suggests that it might be worth watching.

I watched a total of 139 videos, generally either on the topic of color grading or because I thought they might have something to do with LAB. As it turned out, 104 were mostly about LAB and 35 had little or nothing. I rated each as follows:

Green rating: Good information, presenter shows originality and makes it understandable.
19, of which 2 have no LAB. Of the 17 LAB videos:
6 Russian, 3 Italian, 3 U.S. English, 2 Portuguese, 2 non-U.S. English, 1 Spanish.

Yellow rating: Has some of the characteristics of green and might be useful to certain people.
34, of which 5 have no LAB. Of the 29 LAB videos:
7 Spanish, 6 Russian, 4 Italian, 4 U.S. English, 3 German, 3 non-U.S. English, 1 Portuguese, 1 Vietnamese

Red rating: Not recommended. Unfortunately I did not break down why. Some of the reasons might be: nothing of interest; poor quality of presentation; totally superficial; complete lack of originality; presenter doesn’t know WTF he’s talking about.
75 total, 28 of which have no LAB. Of the 47 LAB videos:
27 English (all flavors), 8 Spanish, 6 Russian, 4 German, 1 French, 1 Turkish.

The overall inferiority of the English-language instruction is conspicuous. Some other points I noted:

*There was much more emphasis on facial retouching using LAB than I expected.

*Every instructor who was making overall correction in LAB used some kind of landscape for it, no surprise.

*More than half the Spanish videos appear to be made by Spaniards; I would have expected more from the Western Hemisphere.

*Not much has been made in French or German in the past few years.

*The Russian stuff was head and shoulders above other languages in quality. Despite what Fabrizio says, the Italian content was also good.

*When I was doing the first edition of CC, I made sure that half of the beta-reading team had no prior experience with LAB. One of these, the late Les De Moss, became so taken with the power of the techniques he was learning that he constructed a color graphic with eight spokes, showing how LAB works. When he showed it to me I immediately axed the graphic that I had planned, and substituted his, with his permission of course, as Figure 4.9. It was so successful that I repeated it in the second edition, and stole it for PP5E as well. Apparently others felt the same way, because Les’s graphic appears (unchanged over the years) in around 20 of these instructional videos!

The Russian videos showed some interesting AB channel blends to affect fleshtones. They also show some interesting variations on what one instructor named the “Saturation Clock”. They use half-curves in LAB, often in combination with Blend If. The Russians also often do not content themselves with declaring that their method is the best; they attempt to achieve the same results in RGB or with other commands and they show why it doesn’t work as well.

They are also pretty scrupulous about crediting where their ideas come from, whether from each other or from the foreigner they call Дядя Дэн. One instructive example: apparently there is enormous interest in copying the kind of color grading effects found in certain movies.  So, one of the big English-language Photoshop instructional channels came up with a video entitled How to Steal the Color Grading Secrets from a Certain Movie, or words to that effect. Why anyone would wish to do so, since it’s unlikely that images unrelated to that movie would respond well to the same color trick, is unclear. But, the instructor showed an Awesome! a Killer! method of doing so. It was not, shall we say, the method that I would recommend. But, the guy got a million and a half views.

Not too surprisingly, about a month later one of the other big players did a video on the same topic, showing the exact same Incredible! Mind-Blowing! dubious technique, and also getting a million views or so. I don’t know which of these two guys borrowed the trick from the other, but neither gave the other credit. When the same method was picked up for a Russian video, they did give the first guy credit.

Not that the credit is always accurate. The Spanish videos often showed some interesting LAB maneuvering, but of course they all started with basic straight-line steepening of the AB curves of the type found in our Color Boost action. The Spanish, however, call it the Martian Method (!) and speaker after speaker referred to it in those terms.

On the whole I learned more about comparative linguistics from all this watching that I did about color, although I got some ideas from watching those photographers who had good artistic judgment. For technique, I saw one that demonstrated that when painting to alter an existing mask, the dodge and burn tools are useful. This strikes me in retrospect as exceedingly obvious, and I am wondering how much time I have wasted over the last twenty years not doing it.

I also noted that the skin retouchers were making use of Color Dodge and Color Burn mode in various ways. I may experiment with this, because it looked like they were getting good results.

In short, there is some decent content but you have to waste a lot of time viewing garbage to find it. I suppose that’s what I expected, although I underestimated how much instructional material is out there. I had also thought that videos about technique in video production would be somewhat more popular than those about still photos, but I was shocked by how wide the gap was.

I do agree with Fabrizio that hands-on exercises like our case studies are better, but this may be psychological. We’ve all, I think, submitted entries here that we thought were pretty good, only to find out on Monday morning that half a dozen people did much better. That does tend to grab our attention in a way that YouTube cannot.

Dan





Frederick Yocum
 

Too true! I regret letting work get in the way of completing the series of challenges, but benefited from downloading the examples and following along with the comments.

> We’ve all, I think, submitted entries here that we thought were pretty good, only to find out on Monday morning that half a dozen people did much better. That does tend to grab our attention in a way that YouTube cannot.

regards,

Frederick Yocum


On Apr 26, 2021, at 10:18 AM, Dan Margulis via groups.io <dmargulis@...> wrote:



On Apr 15, 2021, at 11:31 AM, Fabrizio Cacopardo <fabrizio.cacopardo@...> wrote:

First thing first, I'd like to thank Dan for being so generous and always dedicating his time to our work, commenting every single image. A demonstration from him is worth more than a hundred tutorials on Youtube. 

I agree with Fabrizio with respect to our field, but I don’t want to knock YouTube generally. A scad of instructional videos is available on just about any non-graphics topic. I don’t do household repairs any more without checking the videos that show how to proceed. My knowledge of classical music wasn’t bad at the start of the pandemic but it’s more impressive now, what with a dozen recordings available of every major work, and two or three dozen of the standards, not to mention worthwhile works of some lesser-known composers. And I certainly wish that YouTube had been able to prep me when I was teaching in countries where I didn’t speak the language. Anyway, here’s my impression of YouTube as it relates to our field.

Earlier this year we had a thread about the definition of “color grading”, for which I surfed YouTube in search of a consensus. I didn’t find one, but I found at least a hundred videos trying to explain what the speaker thought it was and how to do it. Several of these had a huge number of views, as in, ten times or more the rate of viewing of some of the more popular Photoshop tutorials.

After going through these in January I wondered to myself whether anybody was posting videos about the use of LAB, whether for single photos or in video production. I needn’t have wondered: there were more than a hundred in English alone. Having watched a few, I realized that similar tutorials were available in six other languages in which I know most of the words pertaining to color and so could search for them. So, between the graphic demonstrations and Google Auto Translate’s valiant attempts to add English subtitles, I was able to follow them, although after one experience each with Turkish and Vietnamese I decided to limit myself to the other seven.

Having logged more than 200 possibilities (because I logged titles that looked like they might have LAB content even if it wasn’t stated explicitly) I narrowed it down further after viewing the first dozen or so. It became clear then that in the last two or three years the technical quality of the videos (as opposed to the instructor’s presentation) had increased dramatically. So I decided to only look at stuff that was either 2018 or later, or had an extremely high number of views. I learned a lot about the current state of education in our field, not always for the better.

Back in the Photoshop book era, a huge percentage of the valuable information came out of the United States. That made sense. It hasn’t been possible to make much money on the books themselves for more than ten years, but they could also be used to develop a following for classes or conferences. 

Not so any longer. How does one make a living as an instructor? Evidently the new model is, open up a YouTube channel, release a gang of free videos, get people to subscribe, and ask for donations. Knowledge has become shareware.

The good consequence is that more content is now available in languages other than English. In fact, the majority of the English-language content appears to come from individuals who did not learn their English in the United States.

The first bad consequence are that to keep the subscribers happy and contributing the instructor has to keep releasing a stream of videos, many of which will necessarily focus on topics they know little about.

The second is that showmanship and faked enthusiasm becomes more important than content, meaning that guys like me would never make it in such a setting. No matter how stupid the technique is, it must be described as either 1) Mind-blowing! 2) Awesome! 3) Killer! 4) Incredible! or 5) Fantastic!

I’m not going to rate any videos or their hosts, as some of them are my friends. If you must have one name, I’d say that if you know who Андрей Журавлев is, his videos are worth seeing, but they aren’t in English. In fact, the very language that he speaks suggests that it might be worth watching.

I watched a total of 139 videos, generally either on the topic of color grading or because I thought they might have something to do with LAB. As it turned out, 104 were mostly about LAB and 35 had little or nothing. I rated each as follows:

Green rating: Good information, presenter shows originality and makes it understandable.
19, of which 2 have no LAB. Of the 17 LAB videos:
6 Russian, 3 Italian, 3 U.S. English, 2 Portuguese, 2 non-U.S. English, 1 Spanish.

Yellow rating: Has some of the characteristics of green and might be useful to certain people.
34, of which 5 have no LAB. Of the 29 LAB videos:
7 Spanish, 6 Russian, 4 Italian, 4 U.S. English, 3 German, 3 non-U.S. English, 1 Portuguese, 1 Vietnamese

Red rating: Not recommended. Unfortunately I did not break down why. Some of the reasons might be: nothing of interest; poor quality of presentation; totally superficial; complete lack of originality; presenter doesn’t know WTF he’s talking about.
75 total, 28 of which have no LAB. Of the 47 LAB videos:
27 English (all flavors), 8 Spanish, 6 Russian, 4 German, 1 French, 1 Turkish.

The overall inferiority of the English-language instruction is conspicuous. Some other points I noted:

*There was much more emphasis on facial retouching using LAB than I expected.

*Every instructor who was making overall correction in LAB used some kind of landscape for it, no surprise.

*More than half the Spanish videos appear to be made by Spaniards; I would have expected more from the Western Hemisphere.

*Not much has been made in French or German in the past few years.

*The Russian stuff was head and shoulders above other languages in quality. Despite what Fabrizio says, the Italian content was also good.

*When I was doing the first edition of CC, I made sure that half of the beta-reading team had no prior experience with LAB. One of these, the late Les De Moss, became so taken with the power of the techniques he was learning that he constructed a color graphic with eight spokes, showing how LAB works. When he showed it to me I immediately axed the graphic that I had planned, and substituted his, with his permission of course, as Figure 4.9. It was so successful that I repeated it in the second edition, and stole it for PP5E as well. Apparently others felt the same way, because Les’s graphic appears (unchanged over the years) in around 20 of these instructional videos!

The Russian videos showed some interesting AB channel blends to affect fleshtones. They also show some interesting variations on what one instructor named the “Saturation Clock”. They use half-curves in LAB, often in combination with Blend If. The Russians also often do not content themselves with declaring that their method is the best; they attempt to achieve the same results in RGB or with other commands and they show why it doesn’t work as well.

They are also pretty scrupulous about crediting where their ideas come from, whether from each other or from the foreigner they call Дядя Дэн. One instructive example: apparently there is enormous interest in copying the kind of color grading effects found in certain movies.  So, one of the big English-language Photoshop instructional channels came up with a video entitled How to Steal the Color Grading Secrets from a Certain Movie, or words to that effect. Why anyone would wish to do so, since it’s unlikely that images unrelated to that movie would respond well to the same color trick, is unclear. But, the instructor showed an Awesome! a Killer! method of doing so. It was not, shall we say, the method that I would recommend. But, the guy got a million and a half views.

Not too surprisingly, about a month later one of the other big players did a video on the same topic, showing the exact same Incredible! Mind-Blowing! dubious technique, and also getting a million views or so. I don’t know which of these two guys borrowed the trick from the other, but neither gave the other credit. When the same method was picked up for a Russian video, they did give the first guy credit.

Not that the credit is always accurate. The Spanish videos often showed some interesting LAB maneuvering, but of course they all started with basic straight-line steepening of the AB curves of the type found in our Color Boost action. The Spanish, however, call it the Martian Method (!) and speaker after speaker referred to it in those terms.

On the whole I learned more about comparative linguistics from all this watching that I did about color, although I got some ideas from watching those photographers who had good artistic judgment. For technique, I saw one that demonstrated that when painting to alter an existing mask, the dodge and burn tools are useful. This strikes me in retrospect as exceedingly obvious, and I am wondering how much time I have wasted over the last twenty years not doing it.

I also noted that the skin retouchers were making use of Color Dodge and Color Burn mode in various ways. I may experiment with this, because it looked like they were getting good results.

In short, there is some decent content but you have to waste a lot of time viewing garbage to find it. I suppose that’s what I expected, although I underestimated how much instructional material is out there. I had also thought that videos about technique in video production would be somewhat more popular than those about still photos, but I was shocked by how wide the gap was.

I do agree with Fabrizio that hands-on exercises like our case studies are better, but this may be psychological. We’ve all, I think, submitted entries here that we thought were pretty good, only to find out on Monday morning that half a dozen people did much better. That does tend to grab our attention in a way that YouTube cannot.

Dan