Jump to content

Long exposures problems with typ 246


Giulio Zanni

Recommended Posts

Advertisement (gone after registration)

Giulio - 

 
As I have written above, the two jpeg images (exposed at 9sec and 1/500sec, respectively) have been converted from the DNG files using different parameters. The 9sec image shows vignetting because the lens profile has not been applied. This difference alone might possibly account for the clipping of the dark part in the corner.The picture is also darker because there was exposure correction applied which specified a darker rendering.
 
The water in the right half of the picture roughly at the lower half is a dark grey. It becomes steadily darker until it is completely black at about one third. It can then, of course, not become any darker. There will be a disconuity from the black area to the surrounding water only if you raise the contrast of the image. Otherwise, the transition from dark grey to black is very smooth and linear.
 
The water in the bottom left corner is lighter by a smallish amount. However, the difference to the bottom right corner is just sufficient for some details still to show, even after darkening it by the same amount as the other corner.
 
As I said above, if you edit the 1/500sec jpeg in such a manner that the rocks and the building in the foreground and the land in the background looks the same as in the 9sec picture, you will have the pitch black corner of the water in the bottom right corner as well.
 
I don't write this to spite you. All the evidence I use to reach this conclusion is readily available, and I have documented the more important parts in my previous posts.
 
My recommendation: convert the DNG to JPEG again, taking great care to use the same parameters for the conversion as you used for the 1/500 frame.
  • Like 1
Link to post
Share on other sites

If you read through the thread you would have noticed that the problem is not limited to this image,but it happens also in images with no water. I just regret I didn't keep all those bad images but I will be out again during the week-end and I am sure that the problem will come up again as it has happened before.

 

"If you read..........", well a 'thank you for trying' would have been the polite way to disregard somebody who has read all the way through your tedious thread. 

 

There is a natural vignetting from the lens evident in each photograph, forget that, take that out of the equation, what is left is one of the few things that can cause one corner to be darker than another, and that is polarised light. It works as a reflection or refracted in exactly the same way a polarising filter on the lens will often darken one corner of the sky more than the other. Notwithstanding pop making great efforts to explain the technical side of lens profiles etc. to you and my conclusion is that while you regret not keeping all your bad images, I regret you keeping any.

 

 

Steve

Edited by 250swb
Link to post
Share on other sites

my conclusion is that while you regret not keeping all your bad images, I regret you keeping any.

 

Steve

Well apparently it was worthwile keeping some as they won several international awards, which I doubt it has ever happened to you. I am into photography for 40 years so I know what I am talking about. As you find this thread tedious you are very wellcome to go pontificating somewhere else.

 

Giulio

Edited by Giulio Zanni
Link to post
Share on other sites

Here is a series of 3 images, all unedited jpegs from unedited DNGs (DNGs and JPEGs look the same). All shot at the same time with the same lens and the long exposures with the same filter. The first is a normal exposure, the second is a bad long exposure, the third is a good long exposure. Clearly the longer the exposure, the more the camera is prone to the problem. An underexposed shadow does not justify such a reaction, especially considering that the problem does not apply to the whole shadow but just some bits. 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Edited by Giulio Zanni
Link to post
Share on other sites

Funky stuff

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Edited by Giulio Zanni
Link to post
Share on other sites

Good long exposure

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

  • Like 1
Link to post
Share on other sites

Are all these problems with water shots?

 

 

No, it happens also in shots without water. It is unpredictable. It goes fine for half a day and suddenly it starts freaking out...It seems that it has difficulties in dealing with shadows in very long exposures.  

Edited by Giulio Zanni
  • Like 1
Link to post
Share on other sites

Wow, this is the strangest thing I've seen on this forum. Very interesting. Something seems to make the shadow response fairly suddenly drop in sensitivity after a certain level is reached. It is not without transition though. Sensor heat? I wonder what kind of heat transfer system there is, and whether it is failing. Some surfaces separating sometimes? No explanation seems to fully make sense.

Link to post
Share on other sites

Funky stuff

This new set of pictures seems to be a bit difficult to compare as they differ in both brightness and contrast.

 

These are the original pictures; left (or above): the good exposure, right (or below) the bad exposure:

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

 

I picked a number of points in the good exposure and compared their values in the bad one.

 

good/bad

254 / 245

239 / 214

221 / 186

195 / 129

123 / 054

055 / 000

038 / 000

 

This yields the following curve:

 

I then applied the curve to the good picture to make it look like the bad one. To the left (or above) is the adjusted good picture; to the right as reference the unmodified bad picture again.

 

Clearly, the modified good picture looks much more similar to the bad one, with some differences remaining. In order to make those differences more readily apparent, I then "posterized" both to 16 levels of grey:

 

The overall appearance of the two images is very similar, with two notable expections: The lower right corner in the bad picture is notably darker and the top middle part (the sky) is brighter in the bad picture.

 

Discussion.

 

Both brightness and contrast are markedly different in the two pictures. I can not account for this difference.

 

The difference in brightness and contrast readily explains the clipped shadows (the dark parts) of the boat.

 

This overall difference does clearly not explain the bottom right corner. 

 

However, the brightness distribution of the bottom right corner reminds me of some picture I have taken with a variable ND filter at its strongest position. The variable ND filter is in fact a pair of polarizing filters where one is rotated against the other one. In the middle range the effect will be distributed quite evenly over the whole frame. Only when they are at a right angle to each other, the pair will act as a IR pass filter with an uneven distribution of the brightness.

 

Is it possible that there has been a variable ND filter in use, or that there have been several ND filters stacked onto each other? If so, a few experiments (such as shooting a uniformly lit wall) might assist in either finding or disproving another possible source for the problem.

  • Like 1
Link to post
Share on other sites

This new set of pictures seems to be a bit difficult to compare as they differ in both brightness and contrast.

 

These are the original pictures; left (or above): the good exposure, right (or below) the bad exposure:

attachicon.gif00-good_long.pngattachicon.gif03-bad_long.png

 

I picked a number of points in the good exposure and compared their values in the bad one.

attachicon.gif01-good_long (Kopie).png

 

good/bad

254 / 245

239 / 214

221 / 186

195 / 129

123 / 054

055 / 000

038 / 000

 

This yields the following curve:

attachicon.gif01-curve.png

 

I then applied the curve to the good picture to make it look like the bad one. To the left (or above) is the adjusted good picture; to the right as reference the unmodified bad picture again.

attachicon.gif02-test_good_long.pngattachicon.gif03-bad_long.png

 

Clearly, the modified good picture looks much more similar to the bad one, with some differences remaining. In order to make those differences more readily apparent, I then "posterized" both to 16 levels of grey:

attachicon.gif04-test_good_long-16.pngattachicon.gif05-bad_long.-16.png

 

The overall appearance of the two images is very similar, with two notable expections: The lower right corner in the bad picture is notably darker and the top middle part (the sky) is brighter in the bad picture.

 

Discussion.

 

Both brightness and contrast are markedly different in the two pictures. I can not account for this difference.

 

The difference in brightness and contrast readily explains the clipped shadows (the dark parts) of the boat.

 

This overall difference does clearly not explain the bottom right corner.

 

However, the brightness distribution of the bottom right corner reminds me of some picture I have taken with a variable ND filter at its strongest position. The variable ND filter is in fact a pair of polarizing filters where one is rotated against the other one. In the middle range the effect will be distributed quite evenly over the whole frame. Only when they are at a right angle to each other, the pair will act as a IR pass filter with an uneven distribution of the brightness.

 

Is it possible that there has been a variable ND filter in use, or that there have been several ND filters stacked onto each other? If so, a few experiments (such as shooting a uniformly lit wall) might assist in either finding or disproving another possible source for the problem.

They were all shot with 50/1.4 asph and the long exposures with heliopan 10 stops 46mm normal ND filter. I don't use variable NDs as they are very problematic

 

Sent from my HUAWEI GRA-L09 using Tapatalk

Link to post
Share on other sites

I noticed in the exif info that the Bad was 41s and the Good was 4s.

Since I'm assuming the blackframe's are also the same amount of time as the exposure, could you maybe have a little light leak around the shutter in one corner? Not exactly sure what sort of subtraction is going on with the blackframe, but if it's seeing a little lighter corner it may darken up the final image to compensate for what it thinks is noise. Thinking this effect would be much more pronounced at 41 vs 4 seconds.

 

Could you try another long exposure and then try covering the camera with a jacket or something opaque during the black frame exposure?

Edited by pechelman
Link to post
Share on other sites

I noticed in the exif info that the Bad was 41s and the Good was 4s.

Since I'm assuming the blackframe's are also the same amount of time as the exposure, could you maybe have a little light leak around the shutter in one corner? Not exactly sure what sort of subtraction is going on with the blackframe, but if it's seeing a little lighter corner it may darken up the final image to compensate for what it thinks is noise. Thinking this effect would be much more pronounced at 41 vs 4 seconds.

 

Could you try another long exposure and then try covering the camera with a jacket or something opaque during the black frame exposure?

There are light leaks around the bayonet, that's why I use a hairs elastic band around the bayonet for long exposures. I tried to replicate the problem in the house but everything was fine. What is running me crazy is that the problem is not consistent.

 

Sent from my HUAWEI GRA-L09 using Tapatalk

Link to post
Share on other sites

Interesting and neat tip for the elastic band for the bayonet. Guess I have to steal one from my wife to keep in the bag :)

 

Bayonet aside though, I was wondering if it was a leak past the actual shutter itself, so any light coming in from the lens or anywhere else could change the black frame exposure.

With the 2 examples I've seen here, they seem to occur in relatively bright ambient levels. Being inside the house would likely mask this or make its effect much much less. Maybe you could try this again inside the house but then shine a bright flashlight all over the camera and into the lens for a little bit?

 

anyway, will keep following this. Sorry for your troubles and I hope you figure this out soon.

Have really been enjoying your long exposure photos here and in the 246 thread.

Link to post
Share on other sites

This new set of pictures seems to be a bit difficult to compare as they differ in both brightness and contrast.

 

These are the original pictures; left (or above): the good exposure, right (or below) the bad exposure...

I have to say that I'm impressed with how much work you put into this, Pop.

Link to post
Share on other sites

I have to say that I'm impressed with how much work you put into this, Pop.

Thanks, Eoin. However, it's not that much work and I have done it mostly because I wanted to learn using some techniques I have been thinking about for a while.

  • Like 1
Link to post
Share on other sites

Next round. Some might say that I have too much time on my hands. :D

 

This time I compare the good short exposure with the bad long one. First the original picture:

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

 

Then the brightness values of the good short one adjusted so that the appearance is closer to the bad dark one, applying a simple curve:

 

Then posterized to 16 levels of grey. As the ripples in the water in the short exposure make the gradations hard to see, I first applied a liberal amount of gaussian blur:

 

As can be seen, the gradation of the good short exposure is not all that different from that of the bad long exposure. The dark splotches in the foreground are present in the scene. They become nearly featureless blobs by raising the contrast of the image, thus lowering the brightness value of that already dark part of the image.

 

Indeed, upon close inspection of the dark lower right corner of the bad long exposure, it becomes apparent that it's not in fact a featureless blob. Parts of the corner still exhibit some kind of texture, even in the processed jpeg image.

 

I still tend to believe that the annoying dark corner in the bottom right corner of this scene is produced by two different causes:

This part of the water already has a dark appearance within the scene or, perhaps, after the application of a polarizing filter. This effect has been explained above by another member.

The darkness is confounded in PP when parts that are already dark are still more darkened until they are clipped. This appears not only to be the case in the lower right corner of the bad long exposure but also in parts of the boat where massive clipping is evident.

 

I still suspect that the DNG might contain features where the values have been clipped in the conversion to the 8-bit colour space.

 

Whether this explanation would apply to other photographs in other scenarios, I am unable to tell.

Link to post
Share on other sites

Next round. Some might say that I have too much time on my hands. :D

 

This time I compare the good short exposure with the bad long one. First the original picture:

attachicon.gifgood_short.pngattachicon.gif03-bad_long.png

 

Then the brightness values of the good short one adjusted so that the appearance is closer to the bad dark one, applying a simple curve:

attachicon.giftest_good_short.pngattachicon.gif03-bad_long.png

 

Then posterized to 16 levels of grey. As the ripples in the water in the short exposure make the gradations hard to see, I first applied a liberal amount of gaussian blur:

attachicon.giftest_good_short-16.pngattachicon.gif05-bad_long.-16.png

 

As can be seen, the gradation of the good short exposure is not all that different from that of the bad long exposure. The dark splotches in the foreground are present in the scene. They become nearly featureless blobs by raising the contrast of the image, thus lowering the brightness value of that already dark part of the image.

 

Indeed, upon close inspection of the dark lower right corner of the bad long exposure, it becomes apparent that it's not in fact a featureless blob. Parts of the corner still exhibit some kind of texture, even in the processed jpeg image.

attachicon.gifbad_long_lower_right.png

 

I still tend to believe that the annoying dark corner in the bottom right corner of this scene is produced by two different causes:

This part of the water already has a dark appearance within the scene or, perhaps, after the application of a polarizing filter. This effect has been explained above by another member.

The darkness is confounded in PP when parts that are already dark are still more darkened until they are clipped. This appears not only to be the case in the lower right corner of the bad long exposure but also in parts of the boat where massive clipping is evident.

 

I still suspect that the DNG might contain features where the values have been clipped in the conversion to the 8-bit colour space.

 

Whether this explanation would apply to other photographs in other scenarios, I am unable to tell.

Thank you Pop for you efforts in trying to identify the problem. No polarizer filter was used and I can assure you that there is no visible difference between the dng and the jpeg. I have been doing long exposures for many years with different camera systems and this problem never happened to me before. Whether this is a problem with my two 246s or is a general problem I don't know, as I guess that there are not many people out there doing long exposures with a Leica M. One might argue that the M is not the best solution for doing long exposures but I would expect the camera to perform within the specs. After several back and forth e-mails with Leica they concluded that likely it was a problem due to the combination of adapter + filter, meaning they did not even bother to look at the exif data and realize that these latest images were shot with the 50 lux and not with R 28 PC as the previous ones, therefore no adapter involved. Anyway, I have now bought a different camera system for doing long exposures, so I will not bother the forum anymore with this issue.

 

Giulio

 

Sent from my HUAWEI GRA-L09 using Tapatalk

Edited by Giulio Zanni
  • Like 2
Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...