Be brutal

[TerryToast]

I think you have some very nice videos
I have a real short attention span for 1-2 min after that it hard to keep focus
You may want to do one location per video
I think we all have the same problem getting the perfect transition that every one will love
Keep us posted would love to see the next video

I appreciate your honesty and most viewers are the same. We abandon videos quickly if we lose interest.

Below is a graph of Audience Retention which I took out of YouTube Studio for this video. As per the comment from YouTube (included below) the audience retention for this video is pretty typical.

 

[TerryToast]

@TerryToast I might have missed it, but what are you flying? Also, what is your post rendered color space and codec?
Thanks.

I’m flying a Mavic Air 2S. I film in 4K 25P D-Log.

For editing, I first apply a Buttery LUT (Natural Rec709) to bring most of the colour back to the flat D-Log video, then I do some minor colour grading.

 

[DiscoverSpain]

I've been skim reading the comments on this post. Without pretending to be the smartest in the class, I think that almost everyone is wrong and so are you in assuming that you have done something wrong in your video so that it doesn't have the same number of views as the previous ones.

In my humble opinion, the number of views a video gets on YouTube has nothing to do with the quality of its content, narrative, ...etc. It has to do with how attractive your video is to YouTube's algorithm, so that it shows it to potential viewers and they decide to watch it. I don't know exactly what makes a video attractive to YouTube's algorithm, but I suspect it has a lot to do with the title of the video. For the second step, getting potential viewers to watch it, the video title and thumbnail come into play again.

Regarding your video technically, in my opinion, you should try it for the future:

- Don't burn the highs, like for example at 0:41, 1:04, 1:34, ...etc. and if you do that, discard the scene for the video.

- Make the whites always white, which is not the case throughout your video: I can see bluish, greenish water foam, or greenish, bluish, yellowish sailboats (and it's not because of the sunlight) ...etc.

- In general, I miss consistency in the colours and exposure throughout the video. An easy example, the lighthouse in 3:03 is pink (by the way, which is incomprehensible because we are seeing the part that does not get sunlight, nor has light reflections). In the next scene, at 3:09, it is almost white, a little greenish.

- Also, in general, I miss real colours. I haven't been there, I don't know what the colours are like, but it doesn't seem real to me a sunset where the shadows are red and the sunlight is almost white. Example in 3:16, ...etc.

- And another in general, you tend to oversaturate the colours, but again, without consistency throughout the video.

- Control the shadows better and also avoid having colour: throughout the video I see a lot of bluish, reddish, greenish, ...etc. shadows.

I hope you find my comments helpful

 

[chiangmaicharlies]

Some excellent footage there, music was fine, but I'd have liked a title and some credits.

 

[TerryToast]

I've been skim reading the comments on this post. Without pretending to be the smartest in the class, I think that almost everyone is wrong and so are you in assuming that you have done something wrong in your video so that it doesn't have the same number of views as the previous ones.

In my humble opinion, the number of views a video gets on YouTube has nothing to do with the quality of its content, narrative, ...etc. It has to do with how attractive your video is to YouTube's algorithm, so that it shows it to potential viewers and they decide to watch it. I don't know exactly what makes a video attractive to YouTube's algorithm, but I suspect it has a lot to do with the title of the video. For the second step, getting potential viewers to watch it, the video title and thumbnail come into play again.

Regarding your video technically, in my opinion, you should try it for the future:

- Don't burn the highs, like for example at 0:41, 1:04, 1:34, ...etc. and if you do that, discard the scene for the video.

- Make the whites always white, which is not the case throughout your video: I can see bluish, greenish water foam, or greenish, bluish, yellowish sailboats (and it's not because of the sunlight) ...etc.

- In general, I miss consistency in the colours and exposure throughout the video. An easy example, the lighthouse in 3:03 is pink (by the way, which is incomprehensible because we are seeing the part that does not get sunlight, nor has light reflections). In the next scene, at 3:09, it is almost white, a little greenish.

- Also, in general, I miss real colours. I haven't been there, I don't know what the colours are like, but it doesn't seem real to me a sunset where the shadows are red and the sunlight is almost white. Example in 3:16, ...etc.

- And another in general, you tend to oversaturate the colours, but again, without consistency throughout the video.

- Control the shadows better and also avoid having colour: throughout the video I see a lot of bluish, reddish, greenish, ...etc. shadows.

I hope you find my comments helpful

Thanks Discover.

You’re right on the YouTube Search Engine Optimisation (SEO), title, thumbnail, etc. This has a major impact on the number of YouTube views. I think is certainly a factor.

Good points on the white balance. Most of the video was shot using Auto settings, which I’m starting to move away from. The auto white balance particularly struggles when filming ocean shots at dawn and a breaking wave suddenly introduces a ton of white. To correct this problem, I have recently started using Manual white balance in some of my dawn ocean shoots, where breaking waves are seen.

I appreciate that my colours may appear over saturated, but in honestly they are simply D-Log with a Buttery natural Rec709 LUT applied and some very minor colour grading (where necessary). In a past video (my Northern Beaches video) I did oversaturate the colours and pushed them far too hard, but certainly not in this film.

The lighthouse shot at 3.03 was genuinely pink due to the dawn light reflecting off the clouds which filled the sky. I have attached the original file below, which you are welcome to download and apply a D-Log Buttery Natural Rec709 LUT to, so you can see I am being genuine.

2022-09-19_01.MP4

drive.google.com

The lighthouse shot at 3.09 was taken exactly 9 mins after the prior lighthouse shot (mentioned above). You can see the clouds in the sky are the same, but the magic hour “pink lighting” had faded during this 9 minute interval. No colour grading was done to either shot (it wasn’t needed), just the Buttery LUT applied.

The lighthouse shot at 3.16 was taken exactly 1 minute after the 3.09 shot. The lighting looks very different purely because the sun is now behind the drone/camera rather than in front of the drone/camera. Very minor colour grading has been done to this shot. The red dawn light hitting the trees and cliff has not been enhanced, that is just the colour of the dawn light reflecting off of them. Noting that if I filmed exactly the same scene an hour later, the trees would be a dull dark green and the cliffs would be dull rocks. That’s the beauty of “magic hour” and why I wake up at 4:30am.

Talking about all three lighthouse shots together, it maybe possible to balance out the colours in all three shots, by using Manual (not Auto) camera settings. I just don’t have that skill. It would be interesting to hear from a professional videographer about how much better he/she could have done that shot, if they had used manual setting. I just don’t know.

 

[TerryToast]

Some excellent footage there, music was fine, but I'd have liked a title and some credits.

Thanks Charles. Music title and credits can be found in the YouTube description.

 

[JavaJack]

Can I please get some brutally honest feedback on my latest video? This video has done very poorly on YouTube compared to my prior videos, which are very similar.

I understand that YouTube views is a poor gauge of quality, but if there is something wrong (or less engaging) about this particular video, I’d love to understand that mistake so I don’t repeat it.

 

[JavaJack]

Can I please get some brutally honest feedback on my latest video? This video has done very poorly on YouTube compared to my prior videos, which are very similar.

I understand that YouTube views is a poor gauge of quality, but if there is something wrong (or less engaging) about this particular video, I’d love to understand that mistake so I don’t repeat it.

Looks good to me. I can’t fault anything technically. I think you did a great job.

 

[TerryToast]

Looks good to me. I can’t fault anything technically. I think you did a great job.

Thanks Jack, much appreciated.

Ignoring technically, was there anything you didn’t like as a regular viewer? Did you get bored quickly, etc?

 

[DiscoverSpain]

Thanks Discover.

You’re right on the YouTube Search Engine Optimisation (SEO), title, thumbnail, etc. This has a major impact on the number of YouTube views. I think is certainly a factor.

Good points on the white balance. Most of the video was shot using Auto settings, which I’m starting to move away from. The auto white balance particularly struggles when filming ocean shots at dawn and a breaking wave suddenly introduces a ton of white. To correct this problem, I have recently started using Manual white balance in some of my dawn ocean shoots, where breaking waves are seen.

I appreciate that my colours may appear over saturated, but in honestly they are simply D-Log with a Buttery natural Rec709 LUT applied and some very minor colour grading (where necessary). In a past video (my Northern Beaches video) I did oversaturate the colours and pushed them far too hard, but certainly not in this film.

The lighthouse shot at 3.03 was genuinely pink due to the dawn light reflecting off the clouds which filled the sky. I have attached the original file below, which you are welcome to download and apply a D-Log Buttery Natural Rec709 LUT to, so you can see I am being genuine.

2022-09-19_01.MP4

drive.google.com

The lighthouse shot at 3.09 was taken exactly 9 mins after the prior lighthouse shot (mentioned above). You can see the clouds in the sky are the same, but the magic hour “pink lighting” had faded during this 9 minute interval. No colour grading was done to either shot (it wasn’t needed), just the Buttery LUT applied.

The lighthouse shot at 3.16 was taken exactly 1 minute after the 3.09 shot. The lighting looks very different purely because the sun is now behind the drone/camera rather than in front of the drone/camera. Very minor colour grading has been done to this shot. The red dawn light hitting the trees and cliff has not been enhanced, that is just the colour of the dawn light reflecting off of them. Noting that if I filmed exactly the same scene an hour later, the trees would be a dull dark green and the cliffs would be dull rocks. That’s the beauty of “magic hour” and why I wake up at 4:30am.

Talking about all three lighthouse shots together, it maybe possible to balance out the colours in all three shots, by using Manual (not Auto) camera settings. I just don’t have that skill. It would be interesting to hear from a professional videographer about how much better he/she could have done that shot, if they had used manual setting. I just don’t know.

The clip you sent me has a big problem and that is that the white balance varies throughout the scene: it is much warmer when the drone is on the ground than when the drone is on top. By the way, the headlight is white from the middle of the scene onwards. If you want to avoid this happening to you, as well as exposure variation when you make a turn, etc., use the camera in full manual mode and use the histogram to expose correctly. If you've never done it before, it's a bit difficult at first but it's a matter of time to get the hang of it.

On the other hand, I see that you limit yourself, because if the whites are not white after applying a LUT, they stay that way, or if the colours are oversaturated after applying a LUT, they also stay that way. LUTs don't do magic and I don't think you will find a LUT that correctly corrects all the scenes in a video unless you have exposed them + white balance, perfect, which is very difficult to do and even more so when recording with a drone. Try to do it yourself.

On the other hand, DJI drones are not exactly known for their colour fidelity, so don't take for good what comes out of the camera.

 

[TerryToast]

The clip you sent me has a big problem and that is that the white balance varies throughout the scene: it is much warmer when the drone is on the ground than when the drone is on top. By the way, the headlight is white from the middle of the scene onwards. If you want to avoid this happening to you, as well as exposure variation when you make a turn, etc., use the camera in full manual mode and use the histogram to expose correctly. If you've never done it before, it's a bit difficult at first but it's a matter of time to get the hang of it.

On the other hand, I see that you limit yourself, because if the whites are not white after applying a LUT, they stay that way, or if the colours are oversaturated after applying a LUT, they also stay that way. LUTs don't do magic and I don't think you will find a LUT that correctly corrects all the scenes in a video unless you have exposed them + white balance, perfect, which is very difficult to do and even more so when recording with a drone. Try to do it yourself.

On the other hand, DJI drones are not exactly known for their colour fidelity, so don't take for good what comes out of the camera.

Agreed, I need to start setting my white balance manually and locking it. For many clips like the lighthouse, the average viewer probably wouldn’t even notice, but it’s good to fix.

It’s true that LUTs aren’t magic, but I find that they’re certainly a good starting point for my colour grading. Without some LUT applied, I wouldn’t be able to colour grade the footage back to “Normal”. My ability allows me to make small colour grading adjustments, but I don’t mess with the footage too much. If the colours aren’t “right” after the LUT, I’m probably not seeing the error anyway, which makes it hard to fix. This is another good reason why I should be manually setting the white balance and fixing these issues in camera.

 

[DiscoverSpain]

Agreed, I need to start setting my white balance manually and locking it. For many clips like the lighthouse, the average viewer probably wouldn’t even notice, but it’s good to fix.

It’s true that LUTs aren’t magic, but I find that they’re certainly a good starting point for my colour grading. Without some LUT applied, I wouldn’t be able to colour grade the footage back to “Normal”. My ability allows me to make small colour grading adjustments, but I don’t mess with the footage too much. If the colours aren’t “right” after the LUT, I’m probably not seeing the error anyway, which makes it hard to fix. This is another good reason why I should be manually setting the white balance and fixing these issues in camera.

I don't understand why you ask for a brutal critique of your video when later on, you refuse to try to correct the mistakes in the video. I don't think any of us who edit videos were born learning how to do it. We have learned by trying and educating our eyes.

Unfortunately, you're never going to get perfect exposure and perfect colours from a drone so you're going to have to correct it in post/edit.

If you are not able to see if a scene is wrong in colour, look at the small details that you know, in the case of your video, for example, the white of the sailboats, the white of the foam on the water, the colour in the shadows, ... etc. and above all, try to ensure that the whole video has coherence of colours and exposure.

Of course these are just tips, which you can throw away and I assure you that they will not affect the number of views your videos get.

 

[Eagle Eye 62]

First off the positive. Great shots and nice editing. I would offer the following suggestions to improve the video overall.

1) it is too long with a lot of similar shots. for example the marina appears in five or six different shots and could have been "covered" in two shots with different perspectives.

2) The dark shots seem out of place and just randomly mixed with the other daylight shots. I would have done a progression from daylight to golden hour to night in a sequence.

3) I did not like the speed ramped shot of the marina right after another shot of the marina. It felt unnecessary and out of place with the rest of the edit.

4) The music was ok, but perhaps a little slow paced at times, especially during action shots like surfing, etc.

There are just suggestions and obviously there is no right or wrong since it is just my perspective. I hope this is helpful.

 

[Cymruflyer]

Did you chop off the music in the transition between the surfers and the yacht mariner in the first part of the film?

 

[Wild Drone Pilot]

Can I please get some brutally honest feedback on my latest video? This video has done very poorly on YouTube compared to my prior videos, which are very similar.

I understand that YouTube views is a poor gauge of quality, but if there is something wrong (or less engaging) about this particular video, I’d love to understand that mistake so I don’t repeat it.

Super hard for me to focus for a 5 minute video
I could do it for this one thou
Link Bali & Java
I thought it may help you as it also has many clips
Hope this gives you some good ideas

 

[TerryToast]

Super hard for me to focus for a 5 minute video
I could do it for this one thou
Link Bali & Java
I thought it may help you as it also has many clips
Hope this gives you some good ideas

Thanks Wild, I appreciate the commitment!

I watched the Bali & Java video and I suspect it is CGI, not real drone video at all. That said, the creators are claiming it is real and I have no reason to argue, so I’m happy to concede and just assume it was heavily photoshopped. That would also explain why it took a team of people 1.5 years to create the video. Not really to my taste. That said, I do like Kold’s videos as most of the film is genuine and he only pours on the special effects in the transitions.

 

[Lister]

If this was prepared as a tourism video, intended to demonstrate the range of activities available to Sydney visitors (in other words, a committed audience) then all bets are off and you can ignore the following.

If it's for a general audience, then:

As others have said "Too long". You could lose two minutes and have a better show. "A second is an eternity in television", someone said long ago.

If you've seen one marina at sunset, you've seen 'em all. Once and done.

The beaches all look pretty similar to the casual viewer, even though some of those surfer waves were spectacular. Bondi Beach was a letdown after those. Avoid letdowns.

The paddleboard shots were similarly repetitive and not revealing enough of the participants. Too wide.

One shot of the lighthouse would do fine, thanks.

I hated the music. Distracting rather than enhancing. But then, I'm old.

Auto white balance caught me once during my first attempts with a video camera. Never again. A quick trip to auto to see what the camera thinks is good, sure. Then back to manual. Eyeball estimation is usually good. It's an aesthetic choice as much as anything.

The wide shots of the harbour are great. Only Sydney can compete with Vancouver's harbour. (my home town)

Story, story, story. If it doesn't have a sequencing component, it's empty. Even slide shows have some kind of story. Drones are not particularly good storytelling cameras unless very cleverly used. Google "intentional" : )

Evaluating your film making performance by YouTube views is not a good idea. Much better you evaluate your performance by your clients' response. Nobody's remotely interested in my drone videos except me and my friends and I made films for decades. For me, drones are just for fun.

For example, I bailed out of the Bali video for my own personal taste reasons. (I dislike oversaturated, fast moving, fly-through-tiny-spaces POV material) That wouldn't be a fair evaluation of the film for others.

SFX (especially the whales) is always a good idea, even if there's no way in heck you could mic a whale from a drone. Well done. Apparent only to those editors in the crowd.

The overhead panning shot at 4:30 of the solo surfer is as good a drone aerial as I've seen on YouTube. Superbly executed.

 

[Grandpa]

It’s true that LUTs aren’t magic, but I find that they’re certainly a good starting point for my colour grading. Without some LUT applied, I wouldn’t be able to colour grade the footage back to “Normal”. My ability allows me to make small colour grading adjustments, but I don’t mess with the footage too much. If the colours aren’t “right” after the LUT, I’m probably not seeing the error anyway, which makes it hard to fix. This is another good reason why I should be manually setting the white balance and fixing these issues in camera.

LUTs are no help in matching color grading. It is pretty easy to copy color adjustments to similar clips once you have done the grading adjustments. I am not a fan of LUTS, but I find that they work better if applied after all editing and color corrections are made. Just my workflow 2 cents.

 

[TerryToast]

If this was prepared as a tourism video, intended to demonstrate the range of activities available to Sydney visitors (in other words, a committed audience) then all bets are off and you can ignore the following.

If it's for a general audience, then:

As others have said "Too long". You could lose two minutes and have a better show. "A second is an eternity in television", someone said long ago.

If you've seen one marina at sunset, you've seen 'em all. Once and done.

The beaches all look pretty similar to the casual viewer, even though some of those surfer waves were spectacular. Bondi Beach was a letdown after those. Avoid letdowns.

The paddleboard shots were similarly repetitive and not revealing enough of the participants. Too wide.

One shot of the lighthouse would do fine, thanks.

I hated the music. Distracting rather than enhancing. But then, I'm old.

Auto white balance caught me once during my first attempts with a video camera. Never again. A quick trip to auto to see what the camera thinks is good, sure. Then back to manual. Eyeball estimation is usually good. It's an aesthetic choice as much as anything.

The wide shots of the harbour are great. Only Sydney can compete with Vancouver's harbour. (my home town)

Story, story, story. If it doesn't have a sequencing component, it's empty. Even slide shows have some kind of story. Drones are not particularly good storytelling cameras unless very cleverly used. Google "intentional" : )

Evaluating your film making performance by YouTube views is not a good idea. Much better you evaluate your performance by your clients' response. Nobody's remotely interested in my drone videos except me and my friends and I made films for decades. For me, drones are just for fun.

For example, I bailed out of the Bali video for my own personal taste reasons. (I dislike oversaturated, fast moving, fly-through-tiny-spaces POV material) That wouldn't be a fair evaluation of the film for others.

SFX (especially the whales) is always a good idea, even if there's no way in heck you could mic a whale from a drone. Well done. Apparent only to those editors in the crowd.

The overhead panning shot at 4:30 of the solo surfer is as good a drone aerial as I've seen on YouTube. Superbly executed.

Thanks Lister for the very detailed response. I appreciate it.

There is certainly a consensus that the video is too long and I’ll aim for a shorter video in my next attempt. Cutting content is always easy.

Sequencing. I actually put a lot of effort into sequencing by grouping the clip for each location (~5 clips each), separating each beach locations with a marinas/lighthouse location and by also transitioning between locations on song tempo changes. This worked ok for Sydney locals who knew the locations, but clearly it failed badly for non locals (as none of you picked it up). Ah well, back to the drawing board.

Story. I also tried to introduce some story telling. For example, in Clip 1 the surfer is running down the beach, Clip 2 the surfers are paddling out duck diving waves, Clip 3 the surfers are turning in preparation to catch for a wave, Clip 4 the surfer is surfing the wave, Clip 5 is a wider shot showing more waves coming in to suggest that many more waves will be enjoyed that. That said, while some locations had semi-stories/themes, I acknowledge there was no broader “film wide” story. Clearly more work to do here too.

 

[TerryToast]

I don't understand why you ask for a brutal critique of your video when later on, you refuse to try to correct the mistakes in the video. I don't think any of us who edit videos were born learning how to do it. We have learned by trying and educating our eyes.

Unfortunately, you're never going to get perfect exposure and perfect colours from a drone so you're going to have to correct it in post/edit.

If you are not able to see if a scene is wrong in colour, look at the small details that you know, in the case of your video, for example, the white of the sailboats, the white of the foam on the water, the colour in the shadows, ... etc. and above all, try to ensure that the whole video has coherence of colours and exposure.

Of course these are just tips, which you can throw away and I assure you that they will not affect the number of views your videos get.

I asked for brutal feedback because I had no idea why the film did so poorly with regard to YouTube views. I said that openly in my original message.

I am most certainly listening to all of the feedback provided and where I agree with the feedback, I will correct as many of those mistakes as I can, in my future videos. That includes the length of the video, confusing/missing storyline, white balance, colour grading, etc. That said, this is a hobby and I won’t be able to fix all these issues, unless I turn onto Steven Spielberg overnight.

I am constantly trying to improve my drone videography skills, in all areas, all of the time. However complex issues where I have the least skill and which most viewers simply don’t care about (or even notice), such as perfect colour grading, will get a lesser amount of my time and effort, than other matters which most viewers do care about. I’m sorry if you disagree with this approach.