Camera sensor and dark images

[RobotBee]

Has anyone seen this, or noticed it for themselves?

Apparently for the same aperture, shutter speed and ISO value the MA2 images are much darker than the other Mavic drones.

I had read somewhere that the quad-bayer sensor suffered in this respect, but if the difference is this significant I'm amazed it hasn't been discussed more.

The comments on the youtube video suggest people don't really care, but it would be a really big deal to me.

 

[boblui]

This is the second time I hear the same comment about MA2. This is definitely a deal breaker for me. If the difference is that big, DJI is essentially giving false information about the product isn't it ? can DJI be sued ?

 

[RobotBee]

This is the second time I hear the same comment about MA2. This is definitely a deal breaker for me. If the difference is that big, DJI is essentially giving false information about the product isn't it ?

A deal breaker for me too. I think if I ever wanted a drone with occusync (I currently have the MA1) I'd pay extra for the M2P. The Air 2 is hardly any smaller than the M2P anyway.

 

[LivinLarge]

I would like to see this verified by others. If the lens is really f 4 rather than f 2.8, that means more diffraction as well. This problem can't be fixed in firmware.

 

[boblui]

I would like to see this verified by others. If the lens is really f 4 rather than f 2.8, that means more diffraction as well. This problem can't be fixed in firmware.

certainly it can be done, very easily in fact. Just change the label "IS200" to "ISO100" in the user interface

 

[RobotBee]

I would like to see this verified by others. If the lens is really f 4 rather than f 2.8, that means more diffraction as well. This problem can't be fixed in firmware.

Is it not more likely to do with the sensor, rather than the lens aperture? Those quad-bayer sensors just need more light?

 

[pelermon]

If you use the drone indoors
it might be a inconvenience to have to adjust it.
I will not but it until Litchi works with it myself.

 

[LivinLarge]

certainly it can be done, very easily in fact. Just change the label "IS200" to "ISO100" in the user interface

That won't change the lens aperture.

 

[boblui]

That won't change the lens aperture.

the purpose is just to avoid user complaints.

 

[LivinLarge]

Is it not more likely to do with the sensor, rather than the lens aperture? Those quad-bayer sensors just need more light?

So you are saying that Sony has lied about the sensitivity of the sensor? I see in the video that he tested it beside the Evo 2 which uses the same sensor.

 

[RobotBee]

So you are saying that Sony has lied about the sensitivity of the sensor? I see in the video that he tested it beside the Evo 2 which uses the same sensor.

No, I'm not saying that.

I'm asking the question, prompted by reading somewhere else that the sensors with quad bayer needed more light.

I didn't know he tested it against the Evo 2 which uses the same sensor. That rules that out then I guess.

 

[RobotBee]

Another interesting video.

Could the problem be just when using the 48mp mode?

 

[Ice_2k]

So you are saying that Sony has lied about the sensitivity of the sensor? I see in the video that he tested it beside the Evo 2 which uses the same sensor.

I think he had the Evo II PRO, which has a different, 1" sensor

 

[RobotBee]

I guess the key questions are:

What is the noise of the Air 2 at ISO 200 compared the the Air 1 at ISO 100
and
Is this issue only associated with the effectively smaller pixels in 48 mp mode, and there is no issue when using the full pixels in 12 mp mode?

 

[LivinLarge]

I think he had the Evo II PRO, which has a different, 1" sensor

Good point. It might be the Pro.

 

[LivinLarge]

I guess the key questions are:

What is the noise of the Air 2 at ISO 200 compared the the Air 1 at ISO 100
and
Is this issue only associated with the effectively smaller pixels in 48 mp mode, and there is no issue when using the full pixels in 12 mp mode?

I hope that someone who has both can test this for us. I would also like to see a side by side comparison of the 12 Mp Jpg files from the Air vs Air 2.

 

[hpilot]

You guys do realize that ISO is fake, and it is really just a marketing gimmick that is used by camera companies to make their cameras look good - right?

 

[boblui]

You guys do realize that ISO is fake, and it is really just a marketing gimmick that is used by camera companies to make their cameras look good - right?

Excellent info. I think it's more meaningful to compare the noise of the pictures taken with different drones when the pictures are of the same brightness.

 

[RobotBee]

You guys do realize that ISO is fake, and it is really just a marketing gimmick that is used by camera companies to make their cameras look good - right?

It's not "fake" - it's amplification of the signal from the sensor - the more the amplification, the more noise you get.

Maybe what you're referring to is companies that advertise their cameras as going up to some ridiculously high ISO number, which I agree is meaningless without knowing the quality of the image at that ISO. My SLR goes up to 204,000 or something, but I'd never use it above 12,800 or 25,600 possibly.

Assuming base ISO of 100 (i.e. no amplification) for both sensors, my point was that although by definition the MA2 image at ISO 200 would be worse (more noise) than the MA2 image at ISO 100, it may still be better than the MA1 image at ISO 100.

 

[RobotBee]

Excellent info. I think it's more meaningful to compare the noise of the pictures taken with different drones when the pictures are of the same brightness.

Exactly - hence why I suggested comparing the MA1 at ISO 100 to the MA2 at ISO 200.