Help Identifying a Particular Sort of Image Defect

[SCoombs]

I had originally posted this in a different thread, but that thread was running its course so I don't know how much notice it got and this is a slightly different question from the overall theme of that thread.

The short version: I've been testing and comparing lenses and stopped at a pond I frequent a few times to do this. Most times recently I experienced what I had thought was relatively bad thermal distortion making it very hard to evaluate the lenses themselves. Then yesterday I came across something I've not seen before testing this lens. From a 10 fps burst here is a heavily cropped shot that could be sharper but was mostly acceptable to me:

You can only see EXIF info for this image if you are logged in.

From two shots later in the burst here is a photo that has much, much lower image quality:

You can only see EXIF info for this image if you are logged in.

You can see what looks like ghosting around the left side of the bird and especially in the catch light and in the pupil.

I am now wondering:

1) What could be the cause of this? Can atmospherics (thermal distortion, water vapor in the air, etc.) cause this? Can this happen with simple missed focus? Is it more likely a misaligned element? If it is, why do some shots exhibit it and others don't, even a few tenths of a second apart when nothing has changed about the physical orientation of anything in the lens?

2) If this is not likely the result of thermal distortion/atmospherics, how likely is it that many of the other poor shots I attributed to these atmospherics are actually caused by whatever the problem was here?

Another example. First, here is a very good representation of the effect I was seeing an awful lot yesterday as well as the other recent trips:

You can only see EXIF info for this image if you are logged in.

I'd blame this on motion blur if the shutter speed wasn't 1/6400!

Now we see a MUCH better shot from the same burst:

You can only see EXIF info for this image if you are logged in.

I don't see the same obvious ghosting here as with the gull, so is it the same effect? A different effect?

A goose which is not nearly as bad as that mallard but still clearly not right. This one is only 1/400 so maybe it's motion blur, but this goose really was sitting still and I don't typically find the same kinds of problems with jerky movement from geese as from songbirds for instance. However, and this is very important, I post this goose photo only as an example which looks very much like what I am seeing an awful lot even at shutter speeds which should preclude motion blur. I'm just trying to demonstrate the two major sorts of effect I am seeing, the mallard being an example of when it's really bad and the goose of when it's more mild but still a problem, so even if the goose were motion blur in this case (which to be clear I don't think it is), just take the overall appearence as an example of what I see a lot when the SS is 1/3200 or 1/4000 or greater.

You can only see EXIF info for this image if you are logged in.

The bottom line: are all of these effects - especially the ghosting - consistent with atmospherics or do they indicate a potential hardware problem?

 

[bleirer]

This is something that's been pretty widely reported on the Canon R7, only when shot at the high fps, where unsharp images pop up here and there in the same burst. Who knows why, but suspects might be the VR, maybe resonating somehow at that frame rate.

 

[SCoombs]

This is something that's been pretty widely reported on the Canon R7, only when shot at the high fps, where unsharp images pop up here and there in the same burst. Who knows why, but suspects might be the VR, maybe resonating somehow at that frame rate.

To be clear it's not been here and there, but pretty frequent. Also, any shots at the 1/3200 or 1/6400 or in between - or any above 1/2000 at least - do not have VR turned on because I DID discover that the VR is drastically reducing sharpness at more than say 1/800 or so (I went into this in more detail in the other thread I mentioned.

 

[DRwyoming]

It just looks like the focus missed to me. The thing with any continuous AF (AF-C in Nikon world) is that the focus is continually seeking the best solution and sometimes it overshoots in one direction or the other especially during fast burst shooting. Sure in a perfect world the continuous AF would track perfectly and we'd never have a focus miss but sometimes focus just misses a bit. Things like atmospheric turbulence or just shooting longer distances with a lot of dust and moisture in the air can further confuse the AF tracking but sometimes focus just misses a bit.

I don't see anything that would suggest lens alignment problems as these tend to cause consistent problems in every shot including static target shots where your problem though it may seem frequent isn't in every image.

 

[SCoombs]

It just looks like the focus missed to me. The thing with any continuous AF (AF-C in Nikon world) is that the focus is continually seeking the best solution and sometimes it overshoots in one direction or the other especially during fast burst shooting. Sure in a perfect world the continuous AF would track perfectly and we'd never have a focus miss but sometimes focus just misses a bit. Things like atmospheric turbulence or just shooting longer distances with a lot of dust and moisture in the air can further confuse the AF tracking but sometimes focus just misses a bit.

I don't see anything that would suggest lens alignment problems as these tend to cause consistent problems in every shot including static target shots where your problem though it may seem frequent isn't in every image.

Hmmm. Is ghosting a common effect when focus is missed?

The thing is that if this was missed focus, it seems to me to call a lot more into question because this wasn't just the odd shot here and there. That very blurry mallard shot was the norm, with the better-but-still-bad goose shot being what I saw second most often and the in focus shots representing a tiny minority of shots. We're talking about a hit rate of like 10% that can be called sharp and maybe one or two in the ~3000 photos I shot which you might call critically sharp.

It would be like the worst "stereotypes" of the Z system's autofocus times a million.

It wasn't just focus being off by a bit, either, and it wasn't even just the focus missing and then taking a bit to get back to the right place resulting in a stretch of missed focus, but it would have been focus fluctuating wildly. In the shot below, the focus is on the snow about 2 or 3 inches in front of the duck's eye, but this was one badly out of focus shot in between two shots focused well on the eye in a 10fps burst. That means the AF was correct, then .10 seconds later went to somewhere completely off the mark, then .10 seconds after that went back to the right spot again. This wasn't a case of an AF-area box giving an issue with closest subject priority, either. The subject tracking was on the eye full time here and didn't waver.

You can only see EXIF info for this image if you are logged in.

 

[BillW]

Shooting a Z9 & Z8 these days.

I do find at times that some shots in a burst are not fully sharp. This happens a bit more with a moving subject. Not enough to be a problem for me.

I also find that atmospheric distortion is not always constant. When I am worried about atmospheric distortion but want a photo of the subject, I often shoot a burst. Sometimes there will be a few sharp shots in the burst, notwithstanding the general atmospheric distortion.

 

[SCoombs]

Shooting a Z9 & Z8 these days.

I do find at times that some shots in a burst are not fully sharp. This happens a bit more with a moving subject. Not enough to be a problem for me.

"Some" would not be a problem. As you said, it' isn't enough to be a problem for you.

"All" is a problem, unless it is a case of atmospherics. If it's atmospherics, sure, you might get "all" being poor

To try to be clear, I'm in particular trying to sort of what could be the cause of the ghosting seen in the example. Could atmospheric distortion cause that? If so, well, that explains it. As I said, "all" being out of focus is just what happens sometimes with poor conditions.

But if atmospherics wouldn't be expected to cause ghosting like that, well then that raises other questions.

 

[Cameron T]

From my experience, nothing being particularly sharp in the frame is due to atmospherics. Even missed focus should have something more sharp in the frame somewhere.

I'd recommend setting up a test target (on a tripod and stable head) to see if you can induce this behavior when you're not outside. Test with vr on/off/etc. If you can't, well, we know the problem isn't the camera.

 

[Mal Paso]

It looks like motion blur (VR) rather than a focus miss. Atmospherics won't change completely in a fraction of a second.

 

[SCoombs]

From my experience, nothing being particularly sharp in the frame is due to atmospherics. Even missed focus should have something more sharp in the frame somewhere.

I'd recommend setting up a test target (on a tripod and stable head) to see if you can induce this behavior when you're not outside. Test with vr on/off/etc. If you can't, well, we know the problem isn't the camera.

I have set up a test target and it is generally fine. I also get results which are more or less fine in the backyard. It's when going to the pond I've had problems. This would seem to suggest atmospherics as the culprit!

But it's not so simple, because everything I have done at home and in the backyard has been - by the limitations of the space available - 10m or closer in distance. Every time I've gone to the pond so far I've had very poor results in a variety of temperature/atmospheric conditions. There's an unanswered question, then: is the cause bad atmospherics, or is it rather that the lens has issues at distances >10-15 meters? Sometimes the latter is the case. There have even been very well regarded lenses which were known to be amazing at close distances and borderline unusable at greater distances.

What has prompted this thread is, again, very specifically the ghosting - something that so far nobody has actually commented on. I have encountered bad atmospherics with other lenses before but never seen ghosting of this sort as a result. I am trying to specifically find out what might cause that kind of effect and if atmospherics is one of those things so that I can try to better evaluate whether I'm just getting back luck with conditions the last few weeks or whether I am seeing a lens which has distance problems.

 

[SCoombs]

It looks like motion blur (VR) rather than a focus miss. Atmospherics won't change completely in a fraction of a second.

For the duck shot which shows drastically shifting focus, VR was off and SS was at 1/4000.

For the gull and mallard shots, VR was off and SS was 1/3200 and 1/6400.

I cannot see how it is remotely possible that any kind of motion is at fault here.

 

[Cameron T]

I'm not sure I see 'ghosting' in your images. It just looks soft due to light bending in bad ways from temp differences to me.

 

[Cameron T]

It looks like motion blur (VR) rather than a focus miss. Atmospherics won't change completely in a fraction of a second.

Atmospherics absolutely can change. It's why astro imaging sometimes relies on 'lucky imaging'. Using video to record thousands of frames to pick out the sharpest (And least impacted by atmospherics) ones to stack and process them.

 

[SCoombs]

I'm not sure I see 'ghosting' in your images. It just looks soft due to light bending in bad ways from temp differences to me.

You don't see the ghosting here (and in other places)?

You can only see EXIF info for this image if you are logged in.

 

[Cameron T]

You don't see the ghosting here (and in other places)? View attachment 85046

that isn't the image in your op here, so no, I didn't see that. I have no idea what could cause that except weird VR issues as one possibility.

 

[SCoombs]

that isn't the image in your op here, so no, I didn't see that. I have no idea what could cause that except weird VR issues as one possibility.

Other than the marks I made to help point the ghosting out, this is the very same image from the OP. As noted a few other times already, VR was off for all of the photos posted here except the case of the goose.

Perhaps you got confused because the OP has two extremely similar photos from a few tenths of a second apart in a burst. One was sharp, and that is posted above as well for comparison, while the other from .20 seconds later has the ghosting.

 

[Steve]

FWIW, I lean towards atmospherics. A little breeze can carry a lot of temps that are messing you up and then instantly it's clear. I've seen it happen at 20FPS. In these photos, I see water, snow, and sun in what looks like midday light. It's a recipe for atmospheric induced trauma

Also, I took my 180-600 out the other day and one thing I noticed is that AF is a bit inconsistent at times, at least compared to my primes. I tend to shoot a little heavier (more frames) with it for that reason. However, I'm not sure that's what I'm seeing here. Maybe in some, but not all (the female in the snow was missed focus for sure).

Also, keep in mind that atmospherics can mess with AF distance calculations too since the AF points aren't getting a clear picture and can't quite get things into phase.

The gull with the slight ghosting does look like movement, although at 1/3200th that seems unlikely for that species. I almost would suspect VR if it was on. Still, if it was rapidly turning its head and that's a frame filler, then it's possible that it's motion blur. I'm also not convinced that it's not distortion via the air. Temperature differentials do weird things to light rays.

The goose is motion blur for sure It probably moved its head as you shot. You'd think 1/400th would be enough, but you'd be surprised how easy it is for that speed to show motion.

FWIW, I've shot in conditions like that and, despite the fact that I tech this stuff, I still question my sanity and ability when I get results like this

 

[Ralph]

FWIW, I lean towards atmospherics. A little breeze can carry a lot of temps that are messing you up and then instantly it's clear. I've seen it happen at 20FPS. In these photos, I see water, snow, and sun in what looks like midday light. It's a recipe for atmospheric induced trauma

Also, I took my 180-600 out the other day and one thing I noticed is that AF is a bit inconsistent at times, at least compared to my primes. I tend to shoot a little heavier (more frames) with it for that reason. However, I'm not sure that's what I'm seeing here. Maybe in some, but not all (the female in the snow was missed focus for sure).

Also, keep in mind that atmospherics can mess with AF distance calculations too since the AF points aren't getting a clear picture and can't quite get things into phase.

The gull with the slight ghosting does look like movement, although at 1/3200th that seems unlikely for that species. I almost would suspect VR if it was on. Still, if it was rapidly turning its head and that's a frame filler, then it's possible that it's motion blur. I'm also not convinced that it's not distortion via the air. Temperature differentials do weird things to light rays.

Cking The goose is motion blur for sure It probably moved its head as you shot. You'd think 1/400th would be enough, but you'd be surprised how easy it is for that speed to show motion.

FWIW, I've shot in conditions like that and, despite the fact that I tech this stuff, I still question my sanity and ability when I get results like this

I’m with Steve. If your test target are fine, it’s atmospheric. Many times the camera has a very difficult time locking on, It gets confused and starts to hunt. It’s not really obvious but I can hear the chattering in the lens. There were times when I just got very frustrated and went home.

 

[SCoombs]

FWIW, I lean towards atmospherics. A little breeze can carry a lot of temps that are messing you up and then instantly it's clear. I've seen it happen at 20FPS. In these photos, I see water, snow, and sun in what looks like midday light. It's a recipe for atmospheric induced trauma

Also, I took my 180-600 out the other day and one thing I noticed is that AF is a bit inconsistent at times, at least compared to my primes. I tend to shoot a little heavier (more frames) with it for that reason. However, I'm not sure that's what I'm seeing here. Maybe in some, but not all (the female in the snow was missed focus for sure).

Also, keep in mind that atmospherics can mess with AF distance calculations too since the AF points aren't getting a clear picture and can't quite get things into phase.

The gull with the slight ghosting does look like movement, although at 1/3200th that seems unlikely for that species. I almost would suspect VR if it was on. Still, if it was rapidly turning its head and that's a frame filler, then it's possible that it's motion blur. I'm also not convinced that it's not distortion via the air. Temperature differentials do weird things to light rays.

The goose is motion blur for sure It probably moved its head as you shot. You'd think 1/400th would be enough, but you'd be surprised how easy it is for that speed to show motion.

FWIW, I've shot in conditions like that and, despite the fact that I tech this stuff, I still question my sanity and ability when I get results like this

Thanks, this makes sense. In another place, someone reminded me that sometimes these superzooms produce the kind of ghosting effect in out of focus areas. I've seen it with my 200-500 a fair bit, so it could very well be that the focus had just shifted.

This would raise other questions for me. Just how inconsistent have you found the focus? For instance, here's another shot from the same day:

You can only see EXIF info for this image if you are logged in.

Clearly the focus is 2 or 3 inches in front of the duck - but 1/10 of a second before this and 1/10 of a second after this the focus was squarely on the duck's eye. That seems like a pretty wild swing of focus in so short a time when the Z8 reported a solid lock (straight green solidly on the eye - no yellow or red or the box moving) on the eye the entire time.

Edit: I just realized I'd already posted this photo above!

 

[Alistair]

I had originally posted this in a different thread, but that thread was running its course so I don't know how much notice it got and this is a slightly different question from the overall theme of that thread.

The short version: I've been testing and comparing lenses and stopped at a pond I frequent a few times to do this. Most times recently I experienced what I had thought was relatively bad thermal distortion making it very hard to evaluate the lenses themselves. Then yesterday I came across something I've not seen before testing this lens. From a 10 fps burst here is a heavily cropped shot that could be sharper but was mostly acceptable to me:

View attachment 85027

From two shots later in the burst here is a photo that has much, much lower image quality:

View attachment 85028

You can see what looks like ghosting around the left side of the bird and especially in the catch light and in the pupil.

I am now wondering:

1) What could be the cause of this? Can atmospherics (thermal distortion, water vapor in the air, etc.) cause this? Can this happen with simple missed focus? Is it more likely a misaligned element? If it is, why do some shots exhibit it and others don't, even a few tenths of a second apart when nothing has changed about the physical orientation of anything in the lens?

2) If this is not likely the result of thermal distortion/atmospherics, how likely is it that many of the other poor shots I attributed to these atmospherics are actually caused by whatever the problem was here?

Another example. First, here is a very good representation of the effect I was seeing an awful lot yesterday as well as the other recent trips: View attachment 85032

I'd blame this on motion blur if the shutter speed wasn't 1/6400!

Now we see a MUCH better shot from the same burst:

View attachment 85034

I don't see the same obvious ghosting here as with the gull, so is it the same effect? A different effect?

A goose which is not nearly as bad as that mallard but still clearly not right. This one is only 1/400 so maybe it's motion blur, but this goose really was sitting still and I don't typically find the same kinds of problems with jerky movement from geese as from songbirds for instance. However, and this is very important, I post this goose photo only as an example which looks very much like what I am seeing an awful lot even at shutter speeds which should preclude motion blur. I'm just trying to demonstrate the two major sorts of effect I am seeing, the mallard being an example of when it's really bad and the goose of when it's more mild but still a problem, so even if the goose were motion blur in this case (which to be clear I don't think it is), just take the overall appearence as an example of what I see a lot when the SS is 1/3200 or 1/4000 or greater.

View attachment 85035

The bottom line: are all of these effects - especially the ghosting - consistent with atmospherics or do they indicate a potential hardware problem?

Firstly, congratulations on a very solid post. The examples are great and communication is crystal clear, resulting in zero ambiguity of the issue you are experiencing. We can all learn from this!

WRT the goose, the nostril looks suspiciously sharp to me making me wonder if SD has momentarily mistaken the nostril for an eye. It does happen. In-camera playback or NX Studio will give some clue of focus point location but even that is often wrong. Otherwise it is probably a shutter speed issue.

WRT to others, I see these type of issues occasionally with my 800pf but never with my 400/4.5. I don't think it is a focus issue as nothing in the frame is sharp. So I am left with it being either waves of atmospheric change or an artefact of VR. I cannot make up my mind which or whether occasionally it is both.

 

[JAJohnson]

After a recent four and a half days of shooting bursts of birds (perched and in-flight) with my 180-600 mounted on my Z9, often the first two or three on a burst were soft and then were mostly sharp afterward. I’d get the occasional OOF shot in the middle of the burst which I attributed to atmospheric disturbance, and possibly AF drift. I noticed far fewer OOF shots in a burst when I used my 70-200 with 1.4TC on my Z8.

This leads me to wonder if the AF system in the 180-600 doesn’t perform as accurately and consistently as the AF system in S lenses?

I shot handheld. Perched shots were generally 1/600. In-flight shots were 1/4000 or 1/5000. VR was set to steady (mostly).

 

[JP Onstwedder]

A lot depends on the distance to the subject. At 600mm, and a subject distance of 5 meters, a shutter speed of 1/400 is enough as long as the subject moves slower than 0.3 km/h, that is slower than 0.2 mph. Yes, zero point 2 miles per hour. Any faster, and you'll get subject blur. At 1/3200, the maximum 'no blur' speed is still only 2.5 km/h or about 1/5 mph! (Calculations courtesy of f/ Tools.) Even at 20m, 1/3200 shutter speeds freezes motion up to 10km/h or about 6.5 mph. A long focal length makes it very hard to avoid subject blur unless you're very far away. So my question is how far away you were from the birds?

 

[Len Shepherd]

do not have VR turned on because I DID discover that the VR is drastically reducing sharpness at more than say 1/800

Many do not get a drastic reduction in sharpness faster than 1/800 with VR on – especially with more recent lenses with around 5 stops VR.

I have suggested in another forum contacting Nikon as it might be a VR problem, even though VR is off.

If VR is not off when it should be, it can cause a double image effect if not vibrating correctly when intermittently coming on when it should not.

 

[Steve]

Thanks, this makes sense. In another place, someone reminded me that sometimes these superzooms produce the kind of ghosting effect in out of focus areas. I've seen it with my 200-500 a fair bit, so it could very well be that the focus had just shifted.

This would raise other questions for me. Just how inconsistent have you found the focus? For instance, here's another shot from the same day:

View attachment 85082
Clearly the focus is 2 or 3 inches in front of the duck - but 1/10 of a second before this and 1/10 of a second after this the focus was squarely on the duck's eye. That seems like a pretty wild swing of focus in so short a time when the Z8 reported a solid lock (straight green solidly on the eye - no yellow or red or the box moving) on the eye the entire time.

Edit: I just realized I'd already posted this photo above!

I've not found AF as inconsistent as in this example, but I think heat distortion can really impact it (for any lens). What I see is that the 180-600 will place critical focus slightly in front and behind the eye, for instance, more than something like the 400 4.5 or 600TC. It's not that those lenses are immune to it, it's that it happen with less frequency.

 

[Steve]

This leads me to wonder if the AF system in the 180-600 doesn’t perform as accurately and consistently as the AF system in S lenses?

That's kind of what I'm seeing as well. It's good, but I have to be a little heavier on the release than I normally would.