Is 4k60 video different from 1/2.3" CMOS vs 1"CMOS?

[YuKay]

So what is the difference between 4k60 video made from EVO's Sony 1/2.3" CMOS and the H Plus's Sony 1" CMOS? The pixel dimensions are the same. The bit depth is presumably the same. The file sizes are presumably the same. But something has to be different because on Youtube EVO 4k60 video plays fine on my rig while C23 4k60 always stutters and drops frames.

 

[floridajames04]

The Typhoon has a larger sensor and its effective MP is 20.0 also it films in the new .265 codec at 4K 30FPS and .264 at 4K 60FPS to my knowledge. YouTube is a whole nother ball game.

 

[rdonson]

@YuKay it can be a bit complicated depending on how you want to look at it.

Caution: great simplification of how things work to follow.

4K = 3840 x 2160 pixels or 4096 x 2160 pixels

The EVO sensor is 12 MP which is typically 4000 x 3000 pixels

The 1" sensor is 20 MP which is 5472 x 3648 pixels

More pixels doesn't mean better pixels.

The sensors are likely recent Sony designs/manufacture. Typically the 1" sensor would have larger pixels and that means more sensitivity to light which is usually seen in better low light photos and videos with less noise.

Now comes the technical challenges. How do you go from 4000 x 3000 to 3840 x 2160 or from 5472 x 3648 to 3840 x 2160? Do you simply only "look" at the 3840 x2160 subset of pixels or do you sample all and downsize to the desired 3840 x 2160.

The magic that happens with the choices made in getting to 3840 x 2160 and by the image processor used. With great choices like Autel seems to have made, in good light situations you may not see much difference between 4K60p from a 1" sensor and a 1/2.3" sensor. In low level light the physics of pixel size will definitely favor the larger pixels in the 1" sensor. There's only so much magic that the image processor can do.

If you're taking photos instead of video the differences are magnified by the greater resolution of the 1" sensor.

P.S. never under estimate the value of a good glass lens

P.P.S. the stuttering *might" be attributable to how the video from the camera is handled in post processing.

I hope this makes sense.

 

[Eagle's Eye Video]

The other factor that has a huge influence on the image, is the dynamic range... the CGO3+ is unfortunately deficient in that area... and it compromises both photos and video equally. There are enough floating out there on YouTube... compare sunset footage with the CGO3+ and the C23.

 

[YuKay]

@YuKay it can be a bit complicated depending on how you want to look at it.

Caution: great simplification of how things work to follow.

4K = 3840 x 2160 pixels or 4096 x 2160 pixels

The EVO sensor is 12 MP which is typically 4000 x 3000 pixels

The 1" sensor is 20 MP which is 5472 x 3648 pixels

More pixels doesn't mean better pixels.

The sensors are likely recent Sony designs/manufacture. Typically the 1" sensor would have larger pixels and that means more sensitivity to light which is usually seen in better low light photos and videos with less noise.

Now comes the technical challenges. How do you go from 4000 x 3000 to 3840 x 2160 or from 5472 x 3648 to 3840 x 2160? Do you simply only "look" at the 3840 x2160 subset of pixels or do you sample all and downsize to the desired 3840 x 2160.

The magic that happens with the choices made in getting to 3840 x 2160 and by the image processor used. With great choices like Autel seems to have made, in good light situations you may not see much difference between 4K60p from a 1" sensor and a 1/2.3" sensor. In low level light the physics of pixel size will definitely favor the larger pixels in the 1" sensor. There's only so much magic that the image processor can do.

If you're taking photos instead of video the differences are magnified by the greater resolution of the 1" sensor.

P.S. never under estimate the value of a good glass lens

P.P.S. the stuttering *might" be attributable to how the video from the camera is handled in post processing.

I hope this makes sense.

Thanks for that, Ron. I get it that both systems end up with a 3840 x 2160 pixels video via different routes. What I don't understand is the difference between those videos once they are uploaded to Youtube. They must be different because the 4k60 video from the EVO plays smoothly for me (100% of the time from various uploads using different editors) on Youtube while the same size video from the H Plus never does play smoothly no matter who or what has edited/processed it.

I also get all the stuff about the 1" sensor capturing more light and therefore producing better video in low-light conditions; and the fact that depth of field is different. But these are qualitative measures and would not affect playback. I'm not really asking whether one of these sensors can produce better video than the other. My question is why one of these 4k60 videos plays smoothly for me while the other doesn't, after numerous tests.

I'm hoping that some EVO 4k60 source footage will give me some clues but I haven't been able to find any yet.

 

[Barton]

I'm hoping that some EVO 4k60 source footage will give me some clues but I haven't been able to find any yet.

@YuKay My EVO should arrive on Tuesday. I'll attempt to get some EVO 4K/60 videos for you ASAP. I think I can provide them through Dropbox so we can eliminate YouTube manipulation. I'll also provide an H+ 4K/60 source for comparison.

 

[YuKay]

On the quality front (which is not my question), I think this is a pretty good explanation of why a 1/2.3" sensor might produce better daylight video with fewer artifacts than a 1" sensor:

One difference that wasn’t mentioned in other answers is frame downsampling. With a large sensor you have to downsample the image data read from the sensor to the final frame size – for example 1920✕1080 is just around 2 megapixels, whereas the full sensor output is easily ten times as much with recent full-frame cameras.

The problem with downsampling is that the camera is usually not able to do it fast enough while keeping the quality high, it needs to cut corners. As an example, this is a quote from a review of Sony RX-10, a camera with a small 1" sensor:

One of the big selling points on the Cyber-shot DSC-RX10 is its movie mode. You wouldn't know it from looking at the already impressive spec sheet, but the way the RX10 creates its videos allows for much greater quality than most other cameras.
The reason for the higher quality is how video is sampled. Since the sensors on the vast majority of cameras cannot be 'read out' quickly enough, they don't read the whole sensor when recording video. Instead they skip lines of pixels, which means that the amount of data used to generate is frame is reduced. The gaps between these lines reduce the resolution they can capture, and can introduce moiré.
The RX10, on the other hand, samples the entire 1"-type sensor, so no lines are skipped. The results are noticeable both in our test scene and real world samples.​

In other words, smaller sensor can under some circumstances mean less downsampling artifacts and higher image quality.

Which sensor is better for video, cropped or full frame?

 

[YuKay]

@YuKay My EVO should arrive on Tuesday. I'll attempt to get some EVO 4K/60 videos for you ASAP. I think I can provide them through Dropbox so we can eliminate YouTube manipulation. I'll also provide an H+ 4K/60 source for comparison.

Thank you, Barton - you're a star! Just a 1 minute clip (or two) is all I need. I will also be looking out for your review in the other forum.

 

[rdonson]

On the quality front (which is not my question), I think this is a pretty good explanation of why a 1/2.3" sensor might produce better daylight video with fewer artifacts than a 1" sensor:


Which sensor is better for video, cropped or full frame?

If you’re referencing the 3rd Q&A that’s what I meant when I mentioned the image processor. Down sampling is made in all photo cameras these days that all shoot good 4K video. As an example, the latest Fuji X-T3 (that I’ve started to lust after) has a quad core processor. It writes 4K60 at speeds up to 400 Mbps on UHS-II V60 cards that can handle it.

 

[YuKay]

Thank you, Barton - you're a star! Just a 1 minute clip (or two) is all I need. I will also be looking out for your review in the other forum.

@YuKay My EVO should arrive on Tuesday. I'll attempt to get some EVO 4K/60 videos for you ASAP. I think I can provide them through Dropbox so we can eliminate YouTube manipulation. I'll also provide an H+ 4K/60 source for comparison.

Barton, if I may impose further, could you possibly also include a RAW still image as well as an in-camera JPEG? (I note that the EVO allows still photos to be taken at the push of a button during a video shoot, which is pretty neat.)

 

[Barton]

Barton, if I may impose further, could you possibly also include a RAW still image as well as an in-camera JPEG? (I note that the EVO allows still photos to be taken at the push of a button during a video shoot, which is pretty neat.)

@YuKay I'll try to get my act together Tuesday afternoon. FedEX surprised me today by delivering the EVO a day early. Spent the afternoon charging batteries (3), updating FW, and attaching the Marco Polo tracking device. Will spend more time this evening reading the EVO explorer app documentation.

 

[Eagle's Eye Video]

and attaching the Marco Polo tracking device.

so when using that device who starts by saying "Marco" ?

 

[Jagerbomb52]

Barton, if I may impose further, could you possibly also include a RAW still image as well as an in-camera JPEG? (I note that the EVO allows still photos to be taken at the push of a button during a video shoot, which is pretty neat.)

The EVO does take a photo while videoing but it is only a screen capture from the video file.

 

[Barton]

Thank you, Barton - you're a star! Just a 1 minute clip (or two) is all I need. I will also be looking out for your review in the other forum.

@YuKay Here is the link to the .mp4 EVO video: Dropbox - Public Videos - Simplify your life
Note that this 'just random video' in the sense that I was just trying to get the bird in the air and some quick footage. If there is something more specific you would like, please let me know. Also note that this file is close to 3 GB in size (about 1 GB per minute). It appears that the video will attempt to play fairly quickly BUT AT A LOW RESOLUTION. If you want to see at 4K/60 I suspect you will need to let most (or all of the file to download before viewing - this is just a guess as I have never done this before).

Also in this folder is a jpg. Unfortunately, I haven't figured out all the 'fiddly bits' for the EVO yet so didn't get the photo mode set for RAW and jpg. Hopefully I'll figure this out soon ("life" happened today that got in the way of me being technically on top of all the EVO settings ).

ASIDE (my first impression of the EVO):

• First, let me say that I'm presently happy with the EVO. It feels physically very solid and well made. It has a 'quality' build feeling to it.
• Performance, from a flying perspective, seems to be very good. I waited until the battery voltage kicked in the return to home feature just to make sure it worked. The EVO NAILED the return to home, landing on the launch pad within 6 inches of its takeoff position.
• I was a little surprised by its speed. After launch and doing a little yawing etc, I pushed the 'forward' stick full forward and YIKES, off it went like a bat out of... I had gotten spoiled by the J knob on the H+ to continuously control speed from turtle to rabbit. Will need to do a little investigating on how to control the speed on the EVO - are there options other than just the joystick?
• On the positive side, the YAW control already has the ability for the user to change the expo rate of the yaw stick. Very easy to do so it made the yaw control of the EVO nicely controllable.
• video: I am very pleased with the quick 4K/60 fps video I recorded today. While the results are not 'professional', I think that as I gain knowledge of the EVO, I will be able to control the color and use features like 'histogram' that is displayed on the control display to do real-time adjustments. For sure, the 4K/60 will meet my present needs. I hesitate to say this but had I purchased the EVO before the H+, I would be hard pressed to argue that I needed the H+. That said, I will wallow in the luxury of having both since each has unique features.
• Which UAV will I probably use more? My bet, even after my first 24 minute flight, is that the EVO will be my primary bird, mostly because of the portability and its video output exceeds my personal requirements. As the saying goes, the best camera you have is the one that is with you. I can foresee the EVO living in the car, while the H+ in its travel backpack is just too big... (without doing an actual measurement, I would say the H+ bag volume is about 6 times the volume of the EVO bag). If we go on a dedicated 'UAV video shoot', both systems will go with us. On random drives with other purposes, probably only the EVO will be tagging along...

OK, enough random thoughts for now.

 

[YuKay]

@YuKay Here is the link to the .mp4 EVO video: Dropbox - Public Videos - Simplify your life

Barton, thanks a bunch for this - you are an officer and a gentleman. Downloading now…

 

[YuKay]

@YuKay Here is the link to the .mp4 EVO video: Dropbox - Public Videos - Simplify your life
Note that this 'just random video' in the sense that I was just trying to get the bird in the air and some quick footage. If there is something more specific you would like, please let me know. Also note that this file is close to 3 GB in size (about 1 GB per minute). It appears that the video will attempt to play fairly quickly BUT AT A LOW RESOLUTION. If you want to see at 4K/60 I suspect you will need to let most (or all of the file to download before viewing - this is just a guess as I have never done this before).

Also in this folder is a jpg. Unfortunately, I haven't figured out all the 'fiddly bits' for the EVO yet so didn't get the photo mode set for RAW and jpg. Hopefully I'll figure this out soon ("life" happened today that got in the way of me being technically on top of all the EVO settings ).

ASIDE (my first impression of the EVO):

• First, let me say that I'm presently happy with the EVO. It feels physically very solid and well made. It has a 'quality' build feeling to it.
• Performance, from a flying perspective, seems to be very good. I waited until the battery voltage kicked in the return to home feature just to make sure it worked. The EVO NAILED the return to home, landing on the launch pad within 6 inches of its takeoff position.
• I was a little surprised by its speed. After launch and doing a little yawing etc, I pushed the 'forward' stick full forward and YIKES, off it went like a bat out of... I had gotten spoiled by the J knob on the H+ to continuously control speed from turtle to rabbit. Will need to do a little investigating on how to control the speed on the EVO - are there options other than just the joystick?
• On the positive side, the YAW control already has the ability for the user to change the expo rate of the yaw stick. Very easy to do so it made the yaw control of the EVO nicely controllable.
• video: I am very pleased with the quick 4K/60 fps video I recorded today. While the results are not 'professional', I think that as I gain knowledge of the EVO, I will be able to control the color and use features like 'histogram' that is displayed on the control display to do real-time adjustments. For sure, the 4K/60 will meet my present needs. I hesitate to say this but had I purchased the EVO before the H+, I would be hard pressed to argue that I needed the H+. That said, I will wallow in the luxury of having both since each has unique features.
• Which UAV will I probably use more? My bet, even after my first 24 minute flight, is that the EVO will be my primary bird, mostly because of the portability and its video output exceeds my personal requirements. As the saying goes, the best camera you have is the one that is with you. I can foresee the EVO living in the car, while the H+ in its travel backpack is just too big... (without doing an actual measurement, I would say the H+ bag volume is about 6 times the volume of the EVO bag). If we go on a dedicated 'UAV video shoot', both systems will go with us. On random drives with other purposes, probably only the EVO will be tagging along...

OK, enough random thoughts for now.

Very helpful thank you, Barton - although I'm no nearer (yet) to solving my Youtube playback mystery.

Congratulations to you on producing an excellent first EVO video within hours of unboxing. A lot of things impressed me, including the clarity and the very smooth/liquid pans and tilts (apart from the jittery sequence from 2:48--3:05 - which might have been you getting used to the controls).

Perhaps even more impressive was the scarcity of artifacts. Remarkably, I could see little or no antialiasing in the tree foliage or the short grass, or even the car park surface. There was some in the foreground ploughed field at the beginning but I don't know if any prosumer drone could have managed better.

The colour was pleasing too: to my eye, much superior to the washed-out look of most GoPro type cameras and only bested by the richer colours produced by the H Plus's 1" sensor. What settings did you use?

Overall, it's as smooth and distraction-free viewing experience as I've seen from a drone camera (albeit my universe is limited to those which can be operated without a mobile device) …although I did wonder if that was your hat or your lunch on the table.

I did wonder if maybe the EVO 4k/60 file size would somehow be smaller than that from the H Plus but a rough comparison suggests not. Your 2.9GB file works out at 12.5MB per second of footage while my beloved H Plus Spanish mountains/rusty tower 4k/60 clip equates to 12.6MB per second. And they are both identical in respect of pixel dimensions/ratio, codec, bit rate and colour depth. So I need to look deeper to discover why the EVO Youtube footage plays smoothly for me while the H Plus 4k/60 stutters and drops a lot of frames.

 

[YuKay]

@YuKay Here is the link to the .mp4 EVO video: Dropbox - Public Videos - Simplify your life
Note that this 'just random video' in the sense that I was just trying to get the bird in the air and some quick footage…
…Also in this folder is a jpg. Unfortunately, I haven't figured out all the 'fiddly bits' for the EVO yet so didn't get the photo mode set for RAW and jpg. Hopefully I'll figure this out soon ("life" happened today that got in the way of me being technically on top of all the EVO settings ).

The JPEG is over-sharpened but nonetheless looks a lot better than the (surely faulty) H Plus in-camera JPEG.

At 5.7MB, the EVO 12 megapixel JPEG can't be directly compared to oliver's previously uploaded 15.0MB 20 megapixel H Plus JPEG but I compared them anyway. I see more colour and less noise in the EVO image and significantly fewer sharpening artifacts. Compare the ugly white halos in the two 800% zooms below where the dark landscape meets the light sky. It concerns me that Yuneec haven't fixed their obvious JPEG algorithm blunder yet.



It also appears that Autel have graded their sharpening algorithm so that the artifacts are less noticeable in the foreground, resulting in a much more usable image.

I also compared the poster frame from your EVO 4k/60 video with that of oliver's H Plus YUN_0003.MP4 4k/60 video. Both are compressed with the H.264 codec but for me, the sharpening halo is more pronounced in the H Plus image below:



The halo around the foreground balcony post in the H Plus screen grab above may be an extreme example but for my money, EVO also grades the foreground to distance sharpening algorithm in its video too, making them easier and more natural to look at. The (MP4 screen grab at 800%) example below shows EVO's sharpening halo applied at different levels to the foreground foliage and the distant tree:



That said, many people won't see the MP4 halos from either camera at 100% view unless they have a very sharp eye and are looking for them.

As things stand, for me, the EVO output beats the H Plus in some key departments, while costing less; and, for some reason, its 4k/60 video plays a lot more smoothly on Youtube. If I didn't need the more pro-oriented toolset of the H Plus, its truly standalone controller, its bigger still images and its superior in-camera colours - and its official UK availability - it would be a no-brainer!

 

[YuKay]

Barton, I have enhanced/ruined a still from your video. No pixels were sharpened in this 33-second Photoshop dash - which speaks for the clarity of the EVO footage…

 

[Barton]

Very helpful thank you, Barton - although I'm no nearer (yet) to solving my Youtube playback mystery.

Congratulations to you on producing an excellent first EVO video within hours of unboxing. A lot of things impressed me, including the clarity and the very smooth/liquid pans and tilts (apart from the jittery sequence from 2:48--3:05 - which might have been you getting used to the controls).

Perhaps even more impressive was the scarcity of artifacts. Remarkably, I could see little or no antialiasing in the tree foliage or the short grass, or even the car park surface. There was some in the foreground ploughed field at the beginning but I don't know if any prosumer drone could have managed better.

The colour was pleasing too: to my eye, much superior to the washed-out look of most GoPro type cameras and only bested by the richer colours produced by the H Plus's 1" sensor. What settings did you use?

Overall, it's as smooth and distraction-free viewing experience as I've seen from a drone camera (albeit my universe is limited to those which can be operated without a mobile device) …although I did wonder if that was your hat or your lunch on the table.

I did wonder if maybe the EVO 4k/60 file size would somehow be smaller than that from the H Plus but a rough comparison suggests not. Your 2.9GB file works out at 12.5MB per second of footage while my beloved H Plus Spanish mountains/rusty tower 4k/60 clip equates to 12.6MB per second. And they are both identical in respect of pixel dimensions/ratio, codec, bit rate and colour depth. So I need to look deeper to discover why the EVO Youtube footage plays smoothly for me while the H Plus 4k/60 stutters and drops a lot of frames.

@YuKay , Thank you. I think I'm going to enjoy flying the EVO.

Yes, the jittery sequence was me dorking with the flight controls .

Regarding settings: to speed up the process on the first flight, I left all the settings on Auto (both video and photo), except of course selecting 4K/60.

The hat: yup, just a hat. Fortunately didn't need it as I could sit in the shadows. I was using a refurbished Samsung S7 phone ($60 US) as my monitor and was pleased by how well this little < 6 " screen worked. May spring for larger tablet but for now, the phone seems very adequate. The one thing I need to get working on the phone is the app for recording the screen. I had installed it and tested it but between installation and the flight, the phone got an OS upgrade which probably tanked the recording SW. And thinking about it now, I think I'll reinstall the EVO Explorer flight control app .

 

[YuKay]

You doubtless know this but the EVO app is optimised fr either iOS 11.4 or Android 8 or later.