Night-time photography - image sharpness issues

[Stagecoach]

Manfred,

I'll throw in another. If a tripod foot is on hard packed snow is there a possibility that the temperature of the tripod foot caused the snow/ice under it to partially melt even during these shortish exposures?

Grahame

[blakemcguire]

Personally, I carry a flashlight and use it to light up whatever I've chosen to focus on. I point the flashlight, then focus with magnified live view.

[Glenn NK]

Manfred,

I'll throw in another. If a tripod foot is on hard packed snow is there a possibility that the temperature of the tripod foot caused the snow/ice under it to partially melt even during these shortish exposures?

Grahame

Grahame:

Right, and that's what I said rather cryptically in posts 10 and 14 ("you know why I asked"). Manfred and I have both spent a considerable part of our lives in the Great White North (a euphemism for Canada originating from the colour of snow - every part of the country experiences snow).

The ice under the blade of an ice skater liquifies from the high contact pressure, and reduces friction. (that's why ice skaters can glide so far). As soon as the skate is lifted, the water freezes again. Outdoors in very low air temperatures, (-35 C), the ice doesn't melt as easily, and skating is harder - I've personally experienced that when I lived in Saskatchewan.

A steel tripod point could very well melt the snow a bit and cause it to sink very slightly - it's my best guess as to the problem.

As DanK pointed out, with a lens at 24 mm, DOF is very large at f/4.5, so if shutter shake has been eliminated (MLU), and a remote was used, tripod movement is a likely culprit. It's not necessarily THE culprit, but the first one I would try to eliminate.

G

[NorthernFocus]

Manfred, I tried to read the whole thread but may have missed something. It seems the biggest issue to resolve is whether is is a focus issue or camera shake. If it is camera shake the degree of blurring should be uniform. Have you inspected the entire image carefully to see if any part is sharp at 100 percent?

Second, I'm loathe too mention this because you know your own equipment. With several lenses that I've owned, infinity focus i.e. hard on the stops, is overfocused. I've always wondered at this due to hyperfocal distance etc. but there it is. I overcome this when shooting objects when DOF is not a problem by manually focusing just under infinity as there is plenty of DOF so accurate focus is not an issue.

As improbable as it seems, the underground work you mention could be the issue. A good test is to go to live view and zoom in to max magnification. Then see if you can perceive and vibration in the image. I've done this before and have been amazed at how little wind etc. it takes to make things move.

Good luck with it.

[chauncey]

see if you can perceive and vibration in the image

I had that floating image on my LCD screen as well @ 10X magnification...was caused by Canon's image stabilization feature that was supposed to be deactivated by sensing a tripod.

[ajohnw]

You might do better to bump up the ISO and cut down the exposure time Manfred. This for instance 1/6 sec at 3200 leaves no problems that couldn't be taken care of. It's a dpreview shot.

https://s3.amazonaws.com/masters.gal...G%2Fs06vEXs%3D

Live view boost on an EM-5 would allow focusing on the building without any problem other than having to use a rather extreme magnified view to get a match between sensor pixels and the view pixels. Maybe the D800 has a similar facility or would account for low light levels when the ISO is pumped up Magnified view. - eg 7x with a magnified view shown via 1.25mp ok, 10x dead certain on a 12mp sensor. (pen) Sacrilege maybe to mention an EM-5 but going on some shots taken in caves in Turkey the EM-5 can see better in the dark than I can. The max size an image can finish up at is in part determined by the magnification used to focus. You might find contrast touch the screen focusing on one of the street lights works as well but again I don't know what facilities the D800 has.

Must admit I would be surprised if there was camera shake on your set up. I've managed 2min plus exposures on film with much lighter gear and I do mean much much lighter. In one instance a Parisian bent down and looked into the camera, took one look at me and quickly cleared off. The shot was fine though.

John
-

[Manfred M]

This is a bit puzzling. DOF is huge at that focal length, even when open to f/4.5. If the lens was actually focused anywhere close to infinity, it is hard to see how the buildings could be out of focus. From the first image, it looks like the very close areas are sharp, although it is hard to see without having that area blown up. If the close areas are sharp, then it is clearly a focusing problem. If there is no region, near to far, that is sharp, then it is not a focusing problem. So, my first step would be to blow the whole image up to 100% and then start with near detail and work back systematically.

Did that right off the bat, but no areas are completely sharp, hence my initial comments about my suspecting it is both a focus and movement issue.

Another possibility that occurred to me, if it is not a focusing problem, is that the weight actually hurt you--that the tripod leg on the hard packed snow actually sank a tiny bit while you were taking that part of the pano.

I do a little night photography, and I generally keep the aperture smaller because focus is so hard to manage in the dark.

Being a 9-image pano that was taken over a 10 minute period (plus having to wait for about 20 minutes completely set up and ready to go, waiting for another group of photographers to finish), this seems unlikely. The issue can be found across all 9 images.

[Manfred M]

Manfred,

I'll throw in another. If a tripod foot is on hard packed snow is there a possibility that the temperature of the tripod foot caused the snow/ice under it to partially melt even during these shortish exposures?

Grahame

Equipment was set up and stationary for about 20 minutes before I started to shoot, and there is a 10 minute gap between the first and last image. The tripod has large plastic pads, two of which were definitely on bare concrete and the third one was definitely on a snowy area.

[Manfred M]

Personally, I carry a flashlight and use it to light up whatever I've chosen to focus on. I point the flashlight, then focus with magnified live view.

I would have needed a very powerful flashlight, Blake; the main subject was over 30m / 100ft away. I did focus on the spot-lighted area of the monument as my "infinity" point, so it should have at least been sharp.

[Manfred M]

Grahame:

Right, and that's what I said rather cryptically in posts 10 and 14 ("you know why I asked"). Manfred and I have both spent a considerable part of our lives in the Great White North (a euphemism for Canada originating from the colour of snow - every part of the country experiences snow).

The ice under the blade of an ice skater liquifies from the high contact pressure, and reduces friction. (that's why ice skaters can glide so far). As soon as the skate is lifted, the water freezes again. Outdoors in very low air temperatures, (-35 C), the ice doesn't melt as easily, and skating is harder - I've personally experienced that when I lived in Saskatchewan.

A steel tripod point could very well melt the snow a bit and cause it to sink very slightly - it's my best guess as to the problem.

As DanK pointed out, with a lens at 24 mm, DOF is very large at f/4.5, so if shutter shake has been eliminated (MLU), and a remote was used, tripod movement is a likely culprit. It's not necessarily THE culprit, but the first one I would try to eliminate.

G

No spikes on the tripod, but the whole setup sat in one place for about 20 minutes while another group of photographers was working in front of me. I can confirm that three feet were definitely on the concrete and one foot was on hard-packed snow. No spikes were in use, so the weight distribution was a bit wider. While things could have shifted between shots, Photoshop registered the 9 images perfectly, not even a hint of overlapping / missing pixels.

[Manfred M]

Manfred, I tried to read the whole thread but may have missed something. It seems the biggest issue to resolve is whether is is a focus issue or camera shake. If it is camera shake the degree of blurring should be uniform. Have you inspected the entire image carefully to see if any part is sharp at 100 percent?

That's the first thing I checked for, and while some areas are softer than others, there is no 100% sharp area in the image. This mix of some areas being sharper than others, while nothing is perfectly sharp is why I initially stated that I thought it might be a combination of sharpness and tripod movement.

Second, I'm loathe too mention this because you know your own equipment. With several lenses that I've owned, infinity focus i.e. hard on the stops, is overfocused. I've always wondered at this due to hyperfocal distance etc. but there it is. I overcome this when shooting objects when DOF is not a problem by manually focusing just under infinity as there is plenty of DOF so accurate focus is not an issue.

This is definitely the case with this lens. It is one of the modern lenses that left the factory without ever seeing an optical bench to set a proper hard stop at infinity; so your observation is bang on. That being said, with the focus point that I used at 24mm, I would have expected pretty well everything to be in focus.[/QUOTE]

As improbable as it seems, the underground work you mention could be the issue. A good test is to go to live view and zoom in to max magnification. Then see if you can perceive and vibration in the image. I've done this before and have been amazed at how little wind etc. it takes to make things move.

Good luck with it.

I'm planning to try a number of the suggestions and was getting ready to head out the door when the snow flurries hit. Hopefully I'll get out there tomorrow or the day after. The Christmas lights stay on until Jan 7, so I have close to two weeks to get the shot again.

[Manfred M]

I had that floating image on my LCD screen as well @ 10X magnification...was caused by Canon's image stabilization feature that was supposed to be deactivated by sensing a tripod.

The lens is not stabilized, so this is definitely not the issue for sure. As I recall, the Canon f/2.8 24 - 70mm is not stabilized either.

[William W]

Hi Manfred.

I am assuming it is a stitched panorama comprising two or more shots?

The 100% view appears to me that there is movement.
There appears to be a movement blur in the STATIC vertical lines.
I wouldn't pay any attention to the lights on the trees, they themselves could be moving.

Taking the original - have a close look at the RHS Vertical Lines - I cannot detect the same movement in the original sample that you posted. .

I'd like to see 100% view of a portion of the Building and Light Posts in the scene on the Camera Right Side of that centre walkway.

My guess is that the third leg of the tripod slipped a very small amount in that LHS shot.

*

IF it is one shot, then there appears to me a problem on the left, which is not on the right: but 100% crops from the original high res., can verify that.

WW

[Manfred M]

You might do better to bump up the ISO and cut down the exposure time Manfred. This for instance 1/6 sec at 3200 leaves no problems that couldn't be taken care of. It's a dpreview shot.

https://s3.amazonaws.com/masters.gal...G%2Fs06vEXs%3D

Live view boost on an EM-5 would allow focusing on the building without any problem other than having to use a rather extreme magnified view to get a match between sensor pixels and the view pixels. Maybe the D800 has a similar facility or would account for low light levels when the ISO is pumped up Magnified view. - eg 7x with a magnified view shown via 1.25mp ok, 10x dead certain on a 12mp sensor. (pen) Sacrilege maybe to mention an EM-5 but going on some shots taken in caves in Turkey the EM-5 can see better in the dark than I can. The max size an image can finish up at is in part determined by the magnification used to focus. You might find contrast touch the screen focusing on one of the street lights works as well but again I don't know what facilities the D800 has.

Must admit I would be surprised if there was camera shake on your set up. I've managed 2min plus exposures on film with much lighter gear and I do mean much much lighter. In one instance a Parisian bent down and looked into the camera, took one look at me and quickly cleared off. The shot was fine though.

John
-

I'm going to try a higher ISO / lower shutter speed for sure. There were plenty of hot spots on the image at 100 ISO, although really only about 15 or 20 were really noticeable at 100%. The other 100 or so I could have taken care of with a slight blur on the sky.

[Colin Southern]

I'm going to try a higher ISO / lower shutter speed for sure. There were plenty of hot spots on the image at 100 ISO, although really only about 15 or 20 were really noticeable at 100%. The other 100 or so I could have taken care of with a slight blur on the sky.

I wouldn't; it'll cut into your DR, and you need all the DR you can get for shots like that.

My guess is that it's still incorrect AF, but another "gotcha" when shooting around sand/snow etc is to be careful that your moving away from the camera doesn't then allow the ground to shift. Normally on unstable surfaces I'll step back a couple of feet - start the exposure and just wait it out "at ease".

[Manfred M]

Thanks for all the input - the problem ended up being a focus issue. I used the method recommeded by Colin and others and got a much cleaner set of images. I guess the focus indicator was getting fooled. I am working the new Panos and will be posting.

Thanks for everyone's help and suggestions!

[Colin Southern]

Thanks for all the input - the problem ended up being a focus issue. I used the method recommeded by Colin and others and got a much cleaner set of images. I guess the focus indicator was getting fooled. I am working the new Panos and will be posting.

Thanks for everyone's help and suggestions!

I'm reminded of the time I got up at some ungodly hour - drove 1/2 hour - waited for sunrise. Used the AF to acquire initial focus and then shifted to MF - reversed the lens hood (to cover the focus ring) and proceeded over the next hour to shoot one of the most spectacular sunrises I've ever seen (great cloud patterns).

And every - bloody - one was out of focus.

Never trusted AF in low light again. These days it's either liveview or AF plus zoom in and check (followed by MF select and reverse the hood).

[chauncey]

In addition to what Colin wrote, consider cloud movement and how it affects the light on the ground.
I shot a snow covered, wooded, golf course in the late afternoon where the changing light on the snow shifted on every shot, making the pano series unusable.

[Manfred M]

Interestingly enough; I have found another "culprit" when doing panos and that is Photoshop (CC) image merge software algorithms. The clock, being back-lit shows up this issue very quickly. A sharp image can get blurred during the process.





Initially I thought that the issue might have been too many images to merge as the 5 shots I used, two of them had the clock tower, I do remember reading that too many images can cause issues, so I relooked at what I had and rebuilt the pano with fewer images, where there is a single clock to work with. The results were the same, the clock ended up looking soft.

At this point, my guess is that Photoshop uses some blurring techniques to do the perspective correction and the issue is possibly related to this. I'm going to have to do some more testing...

In the mean time, the fix is simple in this case; just overlay a sharp image where I will manually adjust for perspective.

[DanK]

Thanks for all the input - the problem ended up being a focus issue. I used the method recommeded by Colin and others and got a much cleaner set of images. I guess the focus indicator was getting fooled. I am working the new Panos and will be posting.

Thanks for everyone's help and suggestions!

Still puzzles me. Given the DOF at that FL, it's hard to figure out where the lens could have been focused to render the entire distance range out of focus. At any rate, glad you found a solution.