Ugly Hedgehog - Photography Forum
Home Active Topics Newest Pictures Search Login Register
Main Photography Discussion
Question for high MP (36mp and up) camera users
Page 1 of 4 next> last>>
Mar 31, 2019 10:24:24   #
Rich1939 Loc: Pike County Penna.
 
This question is purely to satisfy my curiosity.
The larger pixel count cameras are universally praised for their ability to capture fine detail because of that pixel count. Does that high count also contribute to better tone graduations?

Reply
Mar 31, 2019 10:28:19   #
cameraf4 Loc: Delaware
 
Rich1939 wrote:
This question is purely to satisfy my curiosity.
The larger pixel count cameras are universally praised for their ability to capture fine detail because of that pixel count. Does that high count also contribute to better tone graduations?


I'm thinking that it depends on the Dynamic Range of the camera. Some earlier DSLRs native ISO only went down to 200. My D850 native ISO goes down to 64. I'm thinking that the increased DR helps tonal graduations.

Reply
Mar 31, 2019 10:33:59   #
Rich1939 Loc: Pike County Penna.
 
cameraf4 wrote:
I'm thinking that it depends on the Dynamic Range of the camera. Some earlier DSLRs native ISO only went down to 200. My D850 native ISO goes down to 64. I'm thinking that the increased DR helps tonal graduations.


On the other hand to my knowledge there weren't any hi-pixel count cameras early on.
My reason for asking is, if each pixel capture a "block" of light then wouldn't more but smaller captures give a finer graduation of tones?

Reply
 
 
Mar 31, 2019 13:28:01   #
speters Loc: Grangeville/Idaho
 
Rich1939 wrote:
This question is purely to satisfy my curiosity.
The larger pixel count cameras are universally praised for their ability to capture fine detail because of that pixel count. Does that high count also contribute to better tone graduations?


Better tonal graduations are usually associated with sensor/film size, so can be found in MF cameras and larger, the pixel count really does not make that much of a difference. The original 12mp Canon 5D had a better/smoother tonal range than the 21mp 5D MII!

Reply
Mar 31, 2019 13:40:13   #
User ID
 
Within the limits of a sensor's DR,
tonal transition is governed by the
firmware interpolating the colored
filter pattern into an image file. If
shooting jpeg it all happens inside
the camera. With raw, it happens
partly in-camera and partly when
processing the raw file on the PC.

Due to the required interpolation,
pixel count is not much influence.

.

Reply
Mar 31, 2019 13:41:06   #
JD750 Loc: SoCal
 
Rich1939 wrote:
This question is purely to satisfy my curiosity.
The larger pixel count cameras are universally praised for their ability to capture fine detail because of that pixel count. Does that high count also contribute to better tone graduations?


Tone graduation is dependent on the number of usable digital levels. A 12-bit conversion has 4096 discrete levels. But how many are usable? Maybe only 10 because of noise. All other things being the same, smaller pixels tend to generate more noise than larger pixels. A higher pixel count sensor will have smaller pixels.

It's more complex because camera manufacturers add noise reduction into the firmware, but I would not expect a higher pixel count camera to necessarily translate to better tone graduation.

Reply
Mar 31, 2019 13:53:30   #
Rich1939 Loc: Pike County Penna.
 
JD750 wrote:
Tone graduation is dependent on the number of usable digital levels. A 12-bit conversion has 4096 discrete levels. But how many are usable? Maybe only 10 because of noise. All other things being the same, smaller pixels tend to generate more noise than larger pixels. A higher pixel count sensor will have smaller pixels.

It's more complex because camera manufacturers add noise reduction into the firmware, but I would not expect a higher pixel count camera to necessarily translate to better tone graduation.
Tone graduation is dependent on the number of usab... (show quote)


Since a higher pixel count sensor can give better detailed images because it breaks that image into smaller pieces and since each pixel’s output is only one data point per ‘scan’ why shouldn’t the tonal data, like the detail data, be better?
Back in the dark ages, that’s B&W film days, I shot the same basic scene on tri-x using a 35mm, a 2 ¼ x 2 ¼ and a 4 X 5. As the negative size went up so did the quality of the tones. I felt that was because each larger negative broke the projected image into smaller pieces as the same grain size film was used. In the case of digital without access to MF or LF digital, using the same projected image size we break that image down to smaller pieces by breaking the sensor into smaller pieces, essentially attempting to reach the grain to negative ratios

Reply
 
 
Mar 31, 2019 14:26:58   #
JD750 Loc: SoCal
 
Rich1939 wrote:
Since a higher pixel count sensor can give better detailed images because it breaks that image into smaller pieces and since each pixel’s output is only one data point per ‘scan’ why shouldn’t the tonal data, like the detail data, be better?


Think of a two-tone image. There are only black and white dots. Even with a fine resolution, see 1st pic below, with tiny dots, it is still only two tones. You don't achieve greater tonal variation with smaller dots. You achieve greater tonal variation with more values of grey per dot. See 2nd pic below.

Source: https://www.dreamstime.com/royalty-free-stock-photo-mariachi-cartoon-playing-trumpet-image15879825
Source:  https://www.dreamstime.com/royalty-free-s...
(Download)

https://unsplash.com/search/photos/grayscale
https://unsplash.com/search/photos/grayscale...
(Download)

Reply
Mar 31, 2019 14:31:45   #
Rich1939 Loc: Pike County Penna.
 
JD750 wrote:
Think of a two-tone image. There are only black and white dots. Even with a fine resolution, see below, with tiny dots, it is still only two tones. You don't achieve greater tonal variation with smaller dots. You achieve tonal variation with more values of grey per dot.




I believe you can only get one value per pixel per scan. More pixels then should translate to more opportunities per scan for more values

Reply
Mar 31, 2019 14:33:28   #
JD750 Loc: SoCal
 
Rich1939 wrote:
I believe you can only get one value per pixel per scan. More pixels then should translate to more opportunities per scan for more values


Ok Rich. If that's what you believe then my work is done here.

Reply
Mar 31, 2019 14:38:04   #
Rich1939 Loc: Pike County Penna.
 
JD750 wrote:
Ok Rich. If that's what you believe then my work is done here.


Am I wrong in that? I am trying to learn something here.

PS: I just dug this up


“However, each small cavity can’t distinguish how much of each color has fallen in, so the sensor can only record gray scale images. To record color pictures, a filter called a Bayer mosaic is placed over the sensor.”

That is from; https://www.lifepixel.com/infrared-photography-primer/ch2-basic-theory-how-digital-camera-sensors-record-light

Reply
 
 
Mar 31, 2019 17:39:08   #
blackest Loc: Ireland
 
Rich1939 wrote:
This question is purely to satisfy my curiosity.
The larger pixel count cameras are universally praised for their ability to capture fine detail because of that pixel count. Does that high count also contribute to better tone graduations?


Well detail really means a difference in color between neighboring pixels giving contrast. Maybe the adc makes a difference e.g 12bit or 14bit when we see steps in fine graduations that's usually in jpeg format where the tones have been pushed too far.

If contrast is boosted too much or oversharpened that's more down to post processing than the capture.
High Pixel count doesn't equal a better image in itself, if you have a tiny sensor then the enlargement is going to bring things into our visual resolution that we wouldn't see with a larger sensor magnified to the same image size but to a smaller degree of magnification.

Dynamic range is more to do with iso than number of pixels. If your camera is multiplying the values then the step size is coarser.

Reply
Mar 31, 2019 17:55:19   #
selmslie Loc: Fernandina Beach, FL, USA
 
Rich1939 wrote:
This question is purely to satisfy my curiosity.
The larger pixel count cameras are universally praised for their ability to capture fine detail because of that pixel count. Does that high count also contribute to better tone graduations?

No, but more sensor DR and higher bit counts (14- or 16- vs. 12-bit raw, 16-bit TIFF vs. 8- or 10-bit JPEG files or displays) provide smoother tone graduations.

Reply
Mar 31, 2019 18:00:26   #
Rich1939 Loc: Pike County Penna.
 
blackest wrote:
Well detail really means a difference in color between neighboring pixels giving contrast. Maybe the adc makes a difference e.g 12bit or 14bit when we see steps in fine graduations that's usually in jpeg format where the tones have been pushed too far.

If contrast is boosted too much or oversharpened that's more down to post processing than the capture.
High Pixel count doesn't equal a better image in itself, if you have a tiny sensor then the enlargement is going to bring things into our visual resolution that we wouldn't see with a larger sensor magnified to the same image size but to a smaller degree of magnification.

Dynamic range is more to do with iso than number of pixels. If your camera is multiplying the values then the step size is coarser.
Well detail really means a difference in color bet... (show quote)



I may have chosen the wrong words here but, my question isn't really about DR. How smoothly does the image move from dark to light, whether that is 3 stops or 8 stops, or from one color to another, is what I am curious about. I think having more pixels has the potential to provide a more gradual and more natural progression but, I don't know if this is a fact or not.

Reply
Mar 31, 2019 18:11:41   #
selmslie Loc: Fernandina Beach, FL, USA
 
Rich1939 wrote:
I may have chosen the wrong words here but, my question isn't really about DR. ...

You can just about see some of the steps in an 8-bit display or JPEG because there are only 256 of them from 0 to 255.

But a 10-bit display has 4x as many steps, 1024 from 0 to 1023. It's unlikely that you can see the breaks between that many steps.

But images are made from raw files. A 12-bit raw file can record 4096 steps but they are unevenly distributed. A 16-bit raw file can record 4x as many steps which will also be unevenly distributed but it gives you smooth tonality over two additional stops. That's two stop of additional useful DR.

Reply
Page 1 of 4 next> last>>
If you want to reply, then register here. Registration is free and your account is created instantly, so you can post right away.
Main Photography Discussion
UglyHedgehog.com - Forum
Copyright 2011-2024 Ugly Hedgehog, Inc.