Surprisingly narrow "Depth-Of-Field"

[Roamer105]

The other day I shot some video from take-off without specifically focusing the camera. I was surprised at this footage shot in a way that demonstrates how narrow a depth-of-field this particular lens has. I don't remember what the aperture setting was but I am sure it was less than 5 - most likely at or close to wide open. Nevertheless, clearly, focusing correctly is quite critical with this lens.

 

[CanadaDrone]

That is like 2 feet in front of the drone haha - of course it's going to look like that Given the extreme situation, the background is also surprisingly clear - on a larger sensor that background would probably be indistinguishable at the same aperture.

Most people are going to be happiest locking the drone into infinity focus unless they are flying extremely close to their subject. The last thing you want is the drone changing focus mid-flight, especially with the M2P which does not have usable continuous autofocus.

Depth of field is also the same regardless of what focal length is used, assuming a constant aperture and subject size.

There are DOF calculators online if you're curious - you can calculate exactly what it is.

Further explanation here:

Best camera settings for Mavic 2 Pro (guide)

 

[Meta4]

I was surprised at this footage shot in a way that demonstrates how narrow a depth-of-field this particular lens has. ...
I don't remember what the aperture setting was but I am sure it was less than 5 - most likely at or close to wide open. Nevertheless, clearly, focusing correctly is quite critical with this lens.

Now look at anything you've shot from up in the air where your subject is more than 10 metres away from the lens (like almost every photo ever taken with a drone) and you see that the lens has more depth of field than you can use.

What you've observed is the small zone of focus when the lens is at the extreme close end of its focus range.
I would expect that the number of photos taken with a drone within 2 feet of the subject to be something like 1 in a million.

 

[Roamer105]

Now look at anything you've shot from up in the air where your subject is more than 10 metres away from the lens (like almost every photo ever taken with a drone) and you see that the lens has more depth of field than you can use.

What you've observed is the small zone of focus when the lens is at the extreme close end of its focus range.
I would expect that the number of photos taken with a drone within 2 feet of the subject to be something like 1 in a million.

I don’t want to start another controversy, but needless to say... but I still have to: All you said is also true for a 4K DSLR. Point is, aside from infinity which probably starts at around 30 feet or so, the depth of field is surprisingly narrow - and not what I would have expected from a micro/mini cam. The same is certainly not true (at least not to that extent) for the MP lens.

 

[DanMan32]

Aperture affects depth of field. The smaller the aperture, the more the camera acts like a pinhole camera. Pinhole cameras have infinite depth of field but at the cost of less light.

 

[Meta4]

I don’t want to start another controversy, but needless to say... but I still have to: All you said is also true for a 4K DSLR. Point is, aside from infinity which probably starts at around 30 feet or so, the depth of field is surprisingly narrow - and not what I would have expected from a micro/mini cam. The same is certainly not true (at least not to that extent) for the MP lens.

You have it focused at it's extreme closest distance.
Obviously only a narrow zone is in focus, that's just what happens when you focus any lens to its closest.
Try the same test with it focused at 20 ft and see how much is in focus.
The lens behaves in a similar way to what's on the Mavic pro
The lens does not have a narrow depth of field and focusing is nowhere near as critical as you imagine.

 

[Mattanous]

Everyone here waiting to correct you.. lol I'm pretty sure he was just stating how great and impressive it looks.

 

[Johnmcl7]

I don’t want to start another controversy, but needless to say... but I still have to: All you said is also true for a 4K DSLR. Point is, aside from infinity which probably starts at around 30 feet or so, the depth of field is surprisingly narrow - and not what I would have expected from a micro/mini cam. The same is certainly not true (at least not to that extent) for the MP lens.

No, you have it the opposite way round - a 4K DSLR would have much less depth of field than the Mavic 2's 1in sensor as for the same field of view the DSLR will have a longer focal length (depth of field works on actual focal length, not equivalent focal length). The person you are quoting is correct, it's the short focus distance which creates the shallow depth of field in this picture - even a tiny 1/2.3in sensor will have very shallow depth of field at close focus distances.

Even wide open at F2.8 at a focus distance of 10 feet, you can see the depth of field is huge as you can see in the attached image. People frequently think DoF is primarily down to aperture but it's a combination of focus distance, actual focal length and aperture as well. Larger sensors have more depth of field control because their actual focal length is much longer.

 

[Hedonist222]

View attachment 50667 Everyone here waiting to correct you.. lol I'm pretty sure he was just stating how great and impressive it looks.

Yah, most responses here are like that.

I'm very glad of the narrow DoF.

I really want to capture myself riding my motorcycle or buggy with myself in focus and everything out of focus (aka bokeh).

I'm hoping to figure this out by setting aperture to wide open ( f/2.8) and then observing the furthest distance before everything starts becoming focused.

Then I'll activate Active Track in profile mode at that distance.

I think the footage will be lovely.

 

[parkgt]

Yah, most responses here are like that.

I'm very glad of the narrow DoF.

I really want to capture myself riding my motorcycle or buggy with myself in focus and everything out of focus (aka bokeh).


Then I'll activate Active Track in profile mode at that distance.

At the required distance you might want to wear your helmet!

 

[Hedonist222]

At the required distance you might want to wear your helmet!

At the required distance you might want to wear your helmet!

I always do.
Even in my buggy that has a roof.

But why?

The last firmware update mandated that the drone come no closer than 8 or 10 meters from the subject in active track.

 

[Kilrah]

I really want to capture myself riding my motorcycle or buggy with myself in focus and everything out of focus (aka bokeh).

The last firmware update mandated that the drone come no closer than 8 or 10 meters from the subject in active track.

That's the point, it's impossible. Given the camera characteristics, and regardless of what the OP thinks his photo is showing... If you wanted to have bokeh on a shot with your M2 it would have to be so close that the props would basically have sliced your head already.

 

[FiShBuRn]

... M2P which does not have usable continuous autofocus....

I have no problems with AFC and i also use Peak Focus to check if the focus is correct...

 

[kilomikebravo]

and i also use Peak Focus to check if the focus is correct.

This feature is very useful IMHO. It gives the pilot instant information along with peace of mind and especially so when one's near vision isn't quite as crisp as it once was (without cheaters.)

 

[parkgt]

I always do.
Even in my buggy that has a roof.

But why?

The last firmware update mandated that the drone come no closer than 8 or 10 meters from the subject in active track.

Because at 8 to 10 meters you are not going to get the shallow depth of field you are looking for. Active track can do some weird things like have the M2 fly into you or you run into it.

 

[CanadaDrone]

I have no problems with AFC and i also use Peak Focus to check if the focus is correct...

The M2P uses contrast detect autofocus, which is a glorified guess & check algorithm that eventually arrives at correct focus on stationary subjects. This is a horrible method of continuous autofocus, but it's how most cheap cameras do it that offer the feature. Phase detect autofocus, on the other hand, can provide instructions to the camera on both the direction it needs to move the focusing elements, and also how far to move those elements (if you've ever used a DSLR this is how they snap to focus so quickly, and how they can track a moving subject without hunting). AFC on the M2P is going to hunt for focus at times, which ruins the footage. There have been several posts about ruined footage already from people not locking it to infinity focus. Use whatever works for you, but just make sure you understand how CDAF works and why it is the worst possible autofocus method to use for a moving subject (or moving camera in this case).

Those 1" sensors are available with PDAF pixels on them from Sony but DJI did not opt for that. The Mavic 2 Zoom on the other hand has PDAF.

 

[FiShBuRn]

The M2P uses contrast detect autofocus, which is a glorified guess & check algorithm that eventually arrives at correct focus on stationary subjects. This is a horrible method of continuous autofocus, but it's how most cheap cameras do it that offer the feature. Phase detect autofocus, on the other hand, can provide instructions to the camera on both the direction it needs to move the focusing elements, and also how far to move those elements (if you've ever used a DSLR this is how they snap to focus so quickly, and how they can track a moving subject without hunting). AFC on the M2P is going to hunt for focus at times, which ruins the footage. There have been several posts about ruined footage already from people not locking it to infinity focus. Use whatever works for you, but just make sure you understand how CDAF works and why it is the worst possible autofocus method to use for a moving subject (or moving camera in this case).

Those 1" sensors are available with PDAF pixels on them from Sony but DJI did not opt for that. The Mavic 2 Zoom on the other hand has PDAF.

Once the recording starts AFC stops working, what is in focus stays in focus thats why i use peak focus check.

 

[CanadaDrone]

Once the recording starts AFC stops working, what is in focus stays in focus thats why i use peak focus check.

If that is the case then you might be OK provided you confirm it is in focus before the recording starts (and if starting the recording subsequently locks AF indefinitely). Sounds like you are doing this with the peaking. AF-C is unnecessary if that is how it's being used though, as tapping to focus right before starting the recording and locking AF would achieve the same thing without any risk of AF-C making unwanted adjustments.

For most people I think it's better to just lock infinity focus unless you are flying close to your subject, which will eliminate any chance of anything being out of focus and you don't have to think about anything.

Just to clarify I wasn't trying to tell you what to use, so I apologize if it came off that way. Just make sure you understand how CDAF works if you ever change your usage of it because it's really quite bad if you are expecting it to track anything or accurately change focus as the landscape/subject changes below the drone. It's unfortunate they did not incorporate PDAF like the M2Zoom, as that would open up some interesting possibilities with close-distance subject tracking.

 

[Roamer105]

Mavic 2 Zoom on the other hand has PDAF.[/QUOTE" data-source="" class="bbCodeBlock bbCodeBlock--expandable bbCodeBlock--quote js-expandWatch">

Where did you get the information about the focus method difference of the ZOOM (PDAF) vs the M2P (CDAF).

I was under the impression PDAF really needs a mirror to work well.

 

[Roamer105]

Regarding "infinity focus" - how accurate is the calibration of the "infinity focus" setting? I remember there had been issues in that arena on the Mavic 1.