ping duck

On 2013-05-24 19:28:01 -0700, PeterN said:

On 5/23/2013 11:55 PM, Tony Cooper wrote:
On Thu, 23 May 2013 20:50:32 -0700, Savageduck
wrote:

On 2013-05-23 20:02:59 -0700, Tony Cooper said:

On Thu, 23 May 2013 17:12:52 -0700, Savageduck
wrote:

First I suspect the problem lies with the unnecessary use of ISO 1600
and 1/4000 shutter speed. The sky noise is excessive and could have
easily been avoided by dialing down to ISO 400 or 200, A shutter speed
of 1/500 @ f/8 would have been more than adequate to capture a gliding
Oystercatcher.


I tend to bristle when someone says I shot something at the wrong ISO
or speed.

...and what if you were wrong?
All this is, is constructive criticism. Something we might not enjoy
hearing, but made in good faith in the hope that the recipient might
just consider what has been said. We do this with the SI, and sometimes
we like to defend our work if a particular comment hits too close to
home, but I always consider there might be a lesson I need to pay heed
to.

Often, the shot is one of those grab shots where there was
no opportunity to adjust the settings. The camera had been set for
something else, and something new popped into the picture I turned and
fired.

That is something which happens to all of us from time to time.
Sometimes I might screw up more than a single shot and I only have
myself to blame. If I am lucky I might have the "Oh crap!" realization
while I am still shooting rather than discovering it as I upload to my
computer.

Dunno if this is the case with Peter's shot, though.

He pretty much explained his reasons for making the exposure settings
he used on the shoot which included the image under discussion. All I
have pointed out was with the lens he used, the light conditions and
his subject matter, was that he might have achieved better results by
using more appropriate settings.

Aside from his abstract creations, Peter's recent shooting and results
have been, in my opinion greatly effected by his overuse of extremely
high ISO settings, even when conditions and circumstances do not call
for them. For more than a year now many of his shots have been
excessively noisy when they shouldn't be, and it isn't artistic grain,
it is noise. Many times it has been exacerbated by some Post
adjustment, or "different" sharpening technique he is working on.
I have made my position on this clear in my SI comments and in
discussion regarding individual shots he has shared here from time to
time.
Personally, I think Peter is one of those guys who likes to push the
envelope when it comes to his camera equipment, and post processing.
However, to my eye there are too many times lately when he pushes
things right out of the safety zone of that envelope.

What did surprise me was Peter's SmugMug pages. I backtracked and
looked at some other shots, and several are far better than anything
he's ever submitted to the SI. I suspect he's holding out on us.

I believe Peter's work in the past was much better that the stuff we
have been shown in recent months. He is going through his high ISO
period.

Gee, you'd think that Peter shoots for his own amusement and not ours.
It's all about him. Wait until Wolfgang hears about it.


Please don't tell him. I am begging you. Anything but that!

Beware the Wolfgang vortex!

--
Regards,

Savageduck

On 5/24/2013 5:58 PM, Me wrote:


SNIP


The "original size" seems to be 2065 x 1739 - so is this cropped or scaled?
The OP's shot also had Nikon's ADL turned on, which under-exposes to
save highlights, then applies a tone curve (in some raw converters) to
lift shadows and results in more noise.
Then comparing 24mp to 36mp at 1:1 pixel view is also not a very
reasonable way to gauge high iso performance.

THANKS, i WILLHAVETO EXPERIMENT WITH adl TURNED OFF.

--
PeterN

Me wrote:

The "original size" seems to be 2065 x 1739 - so is this cropped or scaled?

Cropped, of course.

The OP's shot also had Nikon's ADL turned on, which under-exposes to
save highlights, then applies a tone curve (in some raw converters) to
lift shadows and results in more noise.

Do you want another one underexposed by a stop and pushed?
Oh, wait, that's nearly the same as shooting at a higher ISO
(ok, the latter is less likely to have banding).

Then comparing 24mp to 36mp at 1:1 pixel view is also not a very
reasonable way to gauge high iso performance.

Scale up my example or scale down the OP. And it's 22, not
24 MPix.

Still, I'm suprised how easily the sky can become mottled
with the D800.

-Wolfgang

Savageduck wrote:
On 2013-05-24 06:44:54 -0700, Wolfgang Weisselberg
said:

Le Snip

Here's a series of (fairly strong cropped) photos of the
sky & a crane with different ISO settings, as 'straight out
of AfterShot Pro, no denoising', 'with a (personal) default
setting of sharpening', and 'said sharpening + an ISO matching
(personal) default setting of denoising':
http://www.smugmug.com/gallery/29597388_7KjVfB

Feel free to look at the (O)riginal size, download and zoom
in, ...

An interesting exercise which punctuates my argument.
The lighting condition for your test shots is good, as was the light
for Peter's shot, and certainly does not warrant using high ISO.

I found that I'd rather have some noise due to higher ISO
than loss of sharpness due to camera or subject movement.
The former one can correct much easier ...

But yes, in that light ISO 100-200 would have worked. Good
that many cameras have an auto-ISO settings for when the time
is too short.

(I
understand that it was an exercise).

And thus f/32.

You have provided a good argument
that his image might have had a lot less noise if he had used ISO 100
or ISO 200.

I more wonder how comes that a camera as reputedly good as
the D800 can produce such mottled, blotchy sky at ISO 1600
(or effectively 3200).

-Wolfgang

Wolfgang Weisselberg wrote:
Savageduck wrote:

You have provided a good argument
that his image might have had a lot less noise if he had used ISO 100
or ISO 200.

That hasn't been demonstrated yet...

I more wonder how comes that a camera as reputedly good as
the D800 can produce such mottled, blotchy sky at ISO 1600
(or effectively 3200).

Shot noise has nothing to do with the camera, as such. Any
half decent camera will capture it.

--
Floyd L. Davidson http://www.apaflo.com/
Ukpeagvik (Barrow, Alaska)

Floyd L. Davidson wrote:
Wolfgang Weisselberg wrote:
Savageduck wrote:

You have provided a good argument
that his image might have had a lot less noise if he had used ISO 100
or ISO 200.

That hasn't been demonstrated yet...

What part of 'might' didn't you get?

And did the fact that the D800 is regarded as a rather noise
free camera --- routinely doing *much* better --- pass you by?

I more wonder how comes that a camera as reputedly good as
the D800 can produce such mottled, blotchy sky at ISO 1600
(or effectively 3200).

Shot noise has nothing to do with the camera, as such. Any
half decent camera will capture it.

s/Shot/Photon/

However, there is also amplifier noise, A/D-converter noise,
dark current noise, ... which has *a lot* to do with the camera.

You really have to try *hard* for such mottled skies.

-Wolfgang

Wolfgang Weisselberg wrote:
Floyd L. Davidson wrote:
Wolfgang Weisselberg wrote:
Savageduck wrote:

You have provided a good argument
that his image might have had a lot less noise if he had used ISO 100
or ISO 200.

That hasn't been demonstrated yet...

What part of 'might' didn't you get?

And did the fact that the D800 is regarded as a rather noise
free camera --- routinely doing *much* better --- pass you by?

It's an insignificant difference. Sorry that you don't
realize that, but I'm not at all surprised...

I more wonder how comes that a camera as reputedly good as
the D800 can produce such mottled, blotchy sky at ISO 1600
(or effectively 3200).

Shot noise has nothing to do with the camera, as such. Any
half decent camera will capture it.

s/Shot/Photon/

It is properly called either Shot Noise or Photon Noise. In
fact it can also be correctly called Poisson Noise. For that
matter it can also be called Impulse Noise.

Did you have a point, or just need the education?

However, there is also amplifier noise, A/D-converter noise,
dark current noise, ... which has *a lot* to do with the camera.

And has nothing at all to do with this discussion.

You really have to try *hard* for such mottled skies.

Yeah, right! Shoot a bright blue sky... and crop out a very
small section. Boy is that hard!

--
Floyd L. Davidson http://www.apaflo.com/
Ukpeagvik (Barrow, Alaska)

PeterN wrote:
On 5/23/2013 11:02 PM, Tony Cooper wrote:
On Thu, 23 May 2013 17:12:52 -0700, Savageduck
wrote:

First I suspect the problem lies with the unnecessary use of ISO 1600
and 1/4000 shutter speed. The sky noise is excessive and could have
easily been avoided by dialing down to ISO 400 or 200, A shutter speed
of 1/500 @ f/8 would have been more than adequate to capture a gliding
Oystercatcher.


I tend to bristle when someone says I shot something at the wrong ISO
or speed. Often, the shot is one of those grab shots where there was
no opportunity to adjust the settings. The camera had been set for
something else, and something new popped into the picture I turned and
fired.

Dunno if this is the case with Peter's shot, though.

What did surprise me was Peter's SmugMug pages. I backtracked and
looked at some other shots, and several are far better than anything
he's ever submitted to the SI. I suspect he's holding out on us.


Thank you for the comment. someof the Smugmug images have been in the SI
I think.
But, I try to fit a mandate. Or possibly my o\photography is getting worse.

Sometimes it has to get worse before it gets better. The best way of
finding out where the edges of the envelope are is by repeatedly
stepping over them. I'm sure many of us have had a high ISO period, an
oversharpening period, a wide aperture period. I'm currently
recovering from my wide angle period :-)

--
Chris Malcolm

Floyd L. Davidson wrote:
Wolfgang Weisselberg wrote:
Floyd L. Davidson wrote:
Wolfgang Weisselberg wrote:
Savageduck wrote:

You have provided a good argument
that his image might have had a lot less noise if he had used ISO 100
or ISO 200.

That hasn't been demonstrated yet...

It has. ISO 100 shows a *lot* *lot* *lot* less mottled sky.

What part of 'might' didn't you get?

And did the fact that the D800 is regarded as a rather noise
free camera --- routinely doing *much* better --- pass you by?

It's an insignificant difference. Sorry that you don't
realize that, but I'm not at all surprised...

Does that mean you own a D800 and need to downplay the
problems or does that mean you don't own a D800 and need to
downplay the fact you have a lesser camera?


I more wonder how comes that a camera as reputedly good as
the D800 can produce such mottled, blotchy sky at ISO 1600
(or effectively 3200).

Shot noise has nothing to do with the camera, as such. Any
half decent camera will capture it.

s/Shot/Photon/

It is properly called either Shot Noise or Photon Noise. In
fact it can also be correctly called Poisson Noise. For that
matter it can also be called Impulse Noise.

Did you have a point, or just need the education?

Shot noise: Electrons.
Photon noise: Photons.
Poisson noise: a distribution.

However, it's Photon noise (sometimes called Photon Shot
Noise), the electrons are counted, and even though it's *a*
Poisson noise, it's not the only one, thermal noise is also
a Poisson noise.


However, there is also amplifier noise, A/D-converter noise,
dark current noise, ... which has *a lot* to do with the camera.

And has nothing at all to do with this discussion.

Noise sources for an image has nothing to do with a discussion
about noise in the image? Are you even more drugged than usual?


You really have to try *hard* for such mottled skies.

Yeah, right! Shoot a bright blue sky... and crop out a very
small section. Boy is that hard!

Did so already.
http://www.smugmug.com/gallery/29597388_7KjVfB
Look at the original sizes. (or do you need me to create a
150x150 pixel crop for you?)
See how the sky is not mottled in 8x8 blocks and see how the
sky is easily de-noised.

Oh, yes, sorry, you're legally blind. Get someone else to
look at the photos.

-Wolfgang

Wolfgang Weisselberg wrote:
Floyd L. Davidson wrote:
Wolfgang Weisselberg wrote:
Floyd L. Davidson wrote:
Wolfgang Weisselberg wrote:
Savageduck wrote:

You have provided a good argument
that his image might have had a lot less noise if he had used ISO 100
or ISO 200.

That hasn't been demonstrated yet...

It has. ISO 100 shows a *lot* *lot* *lot* less mottled sky.

He didn't shoot any of the shots at ISO 100, so we don't know
that to be true in this set of circumstances.

Yes you can set up a different set of circumstances that will
give that effect, but that is just that: different.

What part of 'might' didn't you get?

And did the fact that the D800 is regarded as a rather noise
free camera --- routinely doing *much* better --- pass you by?

It's an insignificant difference. Sorry that you don't
realize that, but I'm not at all surprised...

Does that mean you own a D800 and need to downplay the
problems or does that mean you don't own a D800 and need to
downplay the fact you have a lesser camera?

I own several cameras. What I've described is in particular true
because of the extremely high dynamic range of the D800.

I more wonder how comes that a camera as reputedly good as
the D800 can produce such mottled, blotchy sky at ISO 1600
(or effectively 3200).

Shot noise has nothing to do with the camera, as such. Any
half decent camera will capture it.

s/Shot/Photon/

It is properly called either Shot Noise or Photon Noise. In
fact it can also be correctly called Poisson Noise. For that
matter it can also be called Impulse Noise.

Did you have a point, or just need the education?

Shot noise: Electrons.

Shot noise: Poisson distribution

Photon noise: Photons.

Photon noise: shot noise in the arrival time of photons.

Poisson noise: a distribution.

Poisson distribution: the characteristic of Photon Noise.

Looks like you needed more education than I provided the first
time. Or perhaps it's just too much for you...

However, it's Photon noise (sometimes called Photon Shot
Noise), the electrons are counted, and even though it's *a*
Poisson noise, it's not the only one, thermal noise is also
a Poisson noise.

Photons, and thus Photon Noise, is not directly electrons.

Thermal noise in a photo detector has a Gaussian Distribution.

However, there is also amplifier noise, A/D-converter noise,
dark current noise, ... which has *a lot* to do with the camera.

And has nothing at all to do with this discussion.

Noise sources for an image has nothing to do with a discussion
about noise in the image? Are you even more drugged than usual?

The noise we are talking about is Photon Noise.

You really have to try *hard* for such mottled skies.

Yeah, right! Shoot a bright blue sky... and crop out a very
small section. Boy is that hard!

Did so already.
http://www.smugmug.com/gallery/29597388_7KjVfB
Look at the original sizes. (or do you need me to create a
150x150 pixel crop for you?)
See how the sky is not mottled in 8x8 blocks and see how the
sky is easily de-noised.

We never were talking about JPEG artifacts, so why are you?

Oh, yes, sorry, you're legally blind. Get someone else to
look at the photos.

Read the discussion you are blundering into.

--
Floyd L. Davidson http://www.apaflo.com/
Ukpeagvik (Barrow, Alaska)