Should I or not?

[voidshatter]

I have to admit I am just an amateur landscape photographer. I am not sure you can't do panning with tech cam. I do it all the time and it's as easy as the DSLR.

The height of your pano stitch is too short. If you try to output an image with 4:3 aspect ratio or 3:2 aspect ratio and want to be wide, then you can see that you have either:

a) very soft corners if you force a rectilinear stitch out of a pano;

b) if you do a cylindrical stitch out of a pano, you get curved lines that should otherwise be straight horizontal lines .

 

[Landscapelover]

My head hurts
I am leaving now to the mountain with my old IQ180, Cambo and a couple of lenses. I hope I can get some good pictures.

 

[Guy Mancuso]

Lol I hope so too. Go have fun

My head is hurting and I'm just going to prepare for my famous flank steak dinner tonight.

I know a few workshops folks just might jump on a jet and crash the house. Lol

 

[jerome_m]

Many cameras have a built in intervalometer but they don't have timer. So you can only use the built in intervaometer to 30 seconds which in most stacking situations for night time is not long enough. Most exposures will be between 1 minute and 2 min 30 seconds in the ISO 400 to 800 range. Not being familar with Hasselblad I don't know it has the timer also. Nikon and Canon and Somy and Fuji all have intervlometer functions but no timer for bulb. However Magic Lantern on Canon adds this.

You can set the exposure time between 1/800s and 128s on H cameras, which should be sufficient combined with the built-in intervalometer.

If you can't do that on another camera, there are remote cords with a built-in intervalometer function (as I already said). You put the camera in B-mode and they do the rest. You can adjust exposure up to 99 hours.

 

[jerome_m]

You get sharp corners by shifting if you have a wide angle lens with sharp corners. You will never have sharp corners by panning if you try to stitch in 4:3 for very wide angle of view due to mathematical pixel resampling and stretching.

You did not understand what I wrote. Let me try differently: for all rectilinear wide angle lenses, the resolution decreases continuously between center and corners. This is a consequence of the projection, as optically implemented in the lens. The corners are stretched by the projection compared to the center.

 

[Paul2660]

You can set the exposure time between 1/800s and 256s on H cameras, which should be sufficient combined with the built-in intervalometer.

If you can't do that on another camera, there are remote cords with a built-in intervalometer function (as I already said). You put the camera in B-mode and they do the rest. You can adjust exposure up to 99 hours.

You missed my point, yes, I understand there are cords, that allow it on other cameras, DSLR"s, Nikon Canon both make one, I have used both extensively. Fuji does not, but you can find non Fuji brands that do the trick. Phase One/Mamiya do not make one and there are no knock off brands that I am aware of that give you this feature.

Cords are nice, however in winter they get stiff, are long, (so around water you have to wind them around your tripod as none come in a coiled cord). They work as you state as I use all of them in my work.

My point was that, if you are going to put an intervalometer in a camera, then please add a timer that lets you use it past the 30" set shutter speed. Or why put one in there at all. NOTE to Nikon and Canon and the rest. ask the user community who are doing such photography as to how often they need to be set past 30 seconds. Sony has no cable style and due to their proprietary plug, you can't find one on the web that works. I would friggin love to be able to use the A7 for my night work, but there is no intervalometer solution for it that takes you past 30".

It's good to know that Hasselblad does have such a timer, that's a well though out design.

Paul

 

[voidshatter]

You did not understand what I wrote. Let me try differently: for all rectilinear wide angle lenses, the resolution decreases continuously between center and corners. This is a consequence of the projection, as optically implemented in the lens. The corners are stretched by the projection compared to the center.

I know this, but if your aim is to get a wide and rectilinear picture with 4:3 or 3:2 aspect ratio, you still get better corner sharpness if you shoot with a rectilinear lens that has decent sharp corners (e.g. the Rodenstock Digaron HR lenses), than you stitch by pano with a lens of longer focal length.

For example, to output an image at 4:3 aspect ratio with an angle of view of 14mm in the 35mm format, if you stitch with a Rodenstock 32HR lens, you would get better corner sharpness than a stitch with a Rodenstock 90HR lens pano. You would need perhaps a Nikon 800mm f5.6 for the corners to offset the loss of pixel stretch.

Edit: I'm probably wrong with this part. Need to figure this out.

 

[voidshatter]

You missed my point, yes, I understand there are cord, that allow it on other cameras, DSLR"s, Nikon Canon both make one, I have used both extensively. Fuji does not, but you can find non Fuji brands that do the trick. Phase One/Mamiya do not make one and there are no knock off brands that I am aware of that give you this feature.

Cords are nice, however in winter they get stiff, are long, (so around water you have to wind them around your tripod as none come in a coiled cord). They work as you state as I use all of them in my work.

My point was that, if you are going to put an intervalometer in a camera, then please add a timer that lets you use it past the 30" set shutter speed. Or why put one in there at all. NOTE to Nikon and Canon and the rest. ask the user community who are doing such photography as to how often they need to be set past 30 seconds. Sony has no cable style and due to their proprietary plug, you can't find one on the web that works. I would friggin love to be able to use the A7 for my night work, but there is no intervalometer solution for it that takes you past 30".

It's good to know that Hasselblad does have such a timer, that's a well though out design.

Paul

I guess you just need either the Alpa FPS or the Nikon D810A

 

[Shashin]

I know this, but if your aim is to get a wide and rectilinear picture with 4:3 or 3:2 aspect ratio, you still get better corner sharpness if you shoot with a rectilinear lens that has decent sharp corners (e.g. the Rodenstock Digaron HR lenses), than you stitch by pano with a lens of longer focal length.

For example, to output an image at 4:3 aspect ratio with an angle of view of 14mm in the 35mm format, if you stitch with a Rodenstock 32HR lens, you would get better corner sharpness than a stitch with a Rodenstock 90HR lens pano. You would need perhaps a Nikon 800mm f5.6 for the corners to offset the loss of pixel stretch.

I would love to see an example of this claim. You must have a 32mm 90mm comparison to show us.

 

[voidshatter]

I would love to see an example of this claim. You must have a 32mm 90mm comparison to show us.

I don't have a 32 or 90 by my hand. I can only show you the difference between a cylindrical stitch from a pano and a rectilinear stitch from a pano.

Below are the two stitching methods from the same set of pano shots.

Cylindrical: note the curved lines that should otherwise be straight horizontal lines.

Rectilinear: straight lines are straight, but corners are soft.

 

[Shashin]

The histogram cannot be trusted. It was described here.

ISO 35 and 40 seconds would give you much cleaner shadow than ISO 100 and 15 seconds, but ISO 100 and 15 seconds would give you 2.5 stops more highlight details.

This sort of indicates you are a little confused between exposure and DR. DR does not increase with ISO. Also, less exposure reduces signal. By changing exposure, you are simply changing where the DR of the sensor fall on the luminance range of the scene. I don't find your statement saying anything.

 

[voidshatter]

This sort of indicates you are a little confused between exposure and DR. DR does not increase with ISO. Also, less exposure reduces signal. By changing exposure, you are simply changing where the DR of the sensor fall on the luminance range of the scene. I don't find your statement saying anything.

The DR stays the same for ISO 35 and ISO 100 (the orange part on the line I plotted).

After ISO 100, if you increase the ISO, you start to lose DR (the blue part on the line I plotted).

 

[jerome_m]

Sony has no cable style and due to their proprietary plug, you can't find one on the web that works. I would friggin love to be able to use the A7 for my night work, but there is no intervalometer solution for it that takes you past 30".

You may be interested in that hack, then: Sony Alpha NEX Cameras and E Mount Lenses: Sony A7 Wired Intervalometer Hack

 

[Shashin]

The DR stays the same for ISO 35 and ISO 100 (the orange part on the line I plotted).

After ISO 100, if you increase the ISO, you start to lose DR (the blue part on the line I plotted).

You mean the plot of manufacturer ISO to measured ISO? That is an ISO plot. Don't you need a DR axis?

But, anyway. if DR stays the same, then you are not actually getting more DR at ISO 100, you are just under exposing and reducing signal. So, ISO 35 is still better. There is no benefit to ISO 100 in terms of image quality.

 

[jerome_m]

I don't have a 32 or 90 by my hand. I can only show you the difference between a cylindrical stitch from a pano and a rectilinear stitch from a pano.

Below are the two stitching methods from the same set of pano shots.

Cylindrical: note the curved lines that should otherwise be straight horizontal lines.

Rectilinear: straight lines are straight, but corners are soft.

Once again, you are not answering the question, which was about comparing two rectilinear panos, one made by shifting and one by panning with a longer focal length.

It was not about comparing cylindrical and rectilinear projections. We already know the difference.

 

[Paul2660]

Here is what I have learned.

If you want a pan, not stitch, the closer you are to 50mm or past, the easier the construction of the file will be. If you attempt a nodal pan with a 14mm lens, you will have ton's of problems, more so if the lens is in a horizontal position. Even if you are nodal, it still won't matter. Just take a 14mm lens and pan it across a scene, you can easily see how elements in the distance change perspective as you move. This is just part of it with a ultra wide.

You also have to be level if you want anywhere near a good solution. Being level it key.

With stitching, level is not required and with a tech camera, you are moving the back, thus no parallax. issues. Again not true with a Canon solution where the movement is within the lens. However you can counter most of this by moving the camera the same amount in the opposite direction, (Jack wrote a very nice article on just this process several years ago).

If I want to nodal stitch, I tend to move the camera in the vertical as the distortions are less and most tools will come up with a better solution.

NOTE, in outdoor landscape you can get away with a ton of errors, as no one know exactly how that one rock looked, or tree or mountain in the distance. With architecture, this is not true as lines within building have to line up and features also have to line up. This is where sticking is much easier to work with. and or use a 23mm Rodie with a bit of rise or fall or both to help with straightening out the shot. Here the Canon solution will work just as well, (at least in my experience, however the 17mm TS-E will not hold the corners as well and it has a bit of coma problems, (smear on the edges).

My pref. is always stitch, to 15mm past that and you may have too much color cast crosstalk issues to contend with that you won't recover it without a ton of work (Rodie 40mm IQ260), this mainly is with blue skies, with no clouds to help break up the blue. Nothing is harder to get right is a shift across a pure blue sky with no clouds, (at least from my experience). Here a nodal pan is much easier and if I can get a level solution I will always set up the 35mm Schneider and go that way or the 28mm Rodie, which ever I happen to have with me.

The 60mm XL is a dream shifting lens, up to 25mm of shift for me and it's also very very easy to get a nodal solution on since it's such a small lens.

Paul

 

[voidshatter]

You mean the plot of manufacturer ISO to measured ISO? That is an ISO plot. Don't you need a DR axis?

But, anyway. if DR stays the same, then you are not actually getting more DR at ISO 100, you are just under exposing and reducing signal. So, ISO 35 is still better. There is no benefit to ISO 100 in terms of image quality.

Didn't have time to plot a DR axis. I'll describe: DR stays the same in the orange part, but continues to decrease in the blue part, then stays the same in the green part.

ISO 35 is nothing better than ISO 100. If you use the same shutter speed and the same aperture, you do not get better image quality if you shoot at ISO 35, when compared against shooting at ISO 100.

On the other hand, ISO 100 would be better than anything above ISO 100.

Anything below ISO 100 is just extended ISO (marketing hype).

 

[voidshatter]

Once again, you are not answering the question, which was about comparing two rectilinear panos, one made by shifting and one by panning with a longer focal length.

It was not about comparing cylindrical and rectilinear projections. We already know the difference.

Don't you already see the soft corners? With a wide rectilinear lens can you expect such softness in the corners?

 

[jerome_m]

Don't you already see the soft corners? With a wide rectilinear lens can you expect such softness in the corners?

You are still not answering the question.

Hint 1: you have soft corners in your posted sample, but would need a lens you do not have to make it by shifting.

Hint 2: the answer is a simple question of trigonometry, something like 1/ cos (theta).

 

[voidshatter]

You are still not answering the question.

Hint 1: you have soft corners in your posted sample, but would need a lens you do not have to make it by shifting.

Hint 2: the answer is a simple question of trigonometry, something like 1/ cos (theta).

I guess I know what you are talking about. I'm probably wrong with the 800mm f5.6 part, and even the 90HR part. I'll test it when I have time. Thanks for pointing out