Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
07-31-2015, 12:14 PM   #271
Veteran Member
Ikarus's Avatar

Join Date: Apr 2010
Location: California
Photos: Albums
Posts: 471
QuoteOriginally posted by Rondec Quote
The interesting thing is that the D7200 is measured as being slightly better than both of them, even though it has 24 megapixels. By your reasoning, it should not be able to do that.
I wouldn't say this follows from my reasoning. The individual pixel DR directly factors into the sensor DR. Just as two same-sized pixels on different sensor designs aren't equal, there must be something in the design of those D7200 pixel elements that gives them the edge regarding size-normalized DR.

07-31-2015, 08:54 PM   #272
Pentaxian
Class A's Avatar

Join Date: Aug 2008
Location: Wellington, New Zealand
Posts: 9,193
QuoteOriginally posted by Ikarus Quote
Photons hit small pixel sensor area with distribution 2/1/1/0, pixel #1 clips, result after downsampling to low resolution is 3, large pixel result is 4.
It could be argued that your scenario is not fair, as the signal is not sufficiently bandwidth limited.

You are are asking the high-pixel pitch sensor to resolve more spatially, but retain a disproportionate amount of DR per sample.
The audio analogue would be to ask a 1-bit ADC to perform the same signal jumps between samples as a multi-bit ADC using a lower sampling rate.

In other words, the high pixel pitch sensor would not have any difficulty to represent the signal as captured by the low pixel pitch sensor. Alternatively, if you bandwidth-limited the signal (e.g., blurred it a bit) so that it does not contain spatial frequencies that exceed the limits of the low pixel pitch sensor, the high pixel pitch sensor would be able to record it without clipping of pixels.

You may say that real scenes are not bandwidth limited but
  1. pixel-level clipping does have a negligible impact on image DR.
  2. your example demonstrates that one cannot consider DR without resolution.
After all, even a large pixel sensor will exhibit pixel clipping when all the energy of the image is focused on a few pixels.

In terms for "pure level DR" that ignores resolution, a single pixel just measuring the average brightness of the scene would be best. We all know that this wouldn't be ideal for photography purposes.

P.S.: Too bad falconeye left the thread.

Last edited by Class A; 07-31-2015 at 09:05 PM.
07-31-2015, 08:56 PM   #273
Pentaxian




Join Date: May 2009
Location: Ponoka Alberta Canada
Photos: Gallery
Posts: 545
QuoteOriginally posted by Ratcheteer Quote
thats more a comparison between jpeg processing than spatial resolution

even a nikon P900 scores well above 2400 lw/ph (above 105mm mark in its zoom range) and thats with a 16mp sensor the size of 1/2.3" (6.2 x 4.6mm)

anyway this thread is so off topic, its time to let it die now.
I never trust tests done using vertical and horizontal measurement on AAless cameras.
More often than not the tests are skewed because of false detail.
In the very images between the DIII and K3 you can clearly see that the AA filter of the canon is doing its job, while still giving you more detail and on top of that at a full stop shallower DOF.
08-01-2015, 01:11 AM   #274
Veteran Member
Ikarus's Avatar

Join Date: Apr 2010
Location: California
Photos: Albums
Posts: 471
QuoteOriginally posted by Class A Quote
It could be argued that your scenario is not fair, as the signal is not sufficiently bandwidth limited.
That is true, it isn't, but that's just because if it were bandwidth-limited to the resolution of the larger pixel sensor, there wouldn't be much of a point in having the fine resolution. It would be giving up on the very feature this sensor has in favor over the other.

QuoteOriginally posted by Class A Quote
You are are asking the high-pixel pitch sensor to resolve more spatially, but retain a disproportionate amount of DR per sample.
Why is it disproportionate? The smaller pixels have collectively the same capacity as the larger pixel.

QuoteOriginally posted by Class A Quote
After all, even a large pixel sensor will exhibit pixel clipping when all the energy of the image is focused on a few pixels.
Yes, even in the example I gave, some of the larger pixels will clip as well, since if a small pixel can be hit with a value of 2, some sites of the quadruple size are bound to reach a value as high as 8, thus exceeding the large pixel capacity of 4. The point is though, it will clip a bit less than the combination of the small pixels. The effect is not that large, though. The biggest difference for a symmetric distribution (e.g. Gaussian or uniform) occurs if the large pixel clips it right in the center. Then, in case of the uniform distribution, the small pixel sensor clips 0.23bit more than the large pixel one. Not really all that relevant, I would say, especially when considering that this value is only attained if the large pixel sensor clips rather badly as well.


Last edited by Ikarus; 08-01-2015 at 01:18 AM.
08-01-2015, 01:41 AM   #275
Pentaxian
Class A's Avatar

Join Date: Aug 2008
Location: Wellington, New Zealand
Posts: 9,193
QuoteOriginally posted by Ikarus Quote
that's just because if it were bandwidth-limited to the resolution of the larger pixel sensor, there wouldn't be much of a point in having the fine resolution. It would be giving up on the very feature this sensor has in favor over the other.
Well, if the load of signal peaks can always be shared by many pixels so that none of them clip then the higher resolution can still be enjoyed without paying any DR price.

While your example was nicely chosen, it still represents a bit of a pathological case. Again, the lower pitch pixel sensor suffers from the same "problem" relatively to an even lower pitch pixel sensor and only a single pixel sensor would be free from the effect you drew attention to.

QuoteOriginally posted by Ikarus Quote
Why is it disproportionate? The smaller pixels have collectively the same capacity as the larger pixels.
Yes, but for fairness, I could ask for signal peaks at the resolution of the smaller pixel sensor to not exceed the capacity of a smaller pixel.

I think the audio analogy is useful again: When using a delta-sigma conversion scheme (consider multi-bit schemes, not just the 1-bit variety), one can trade sampling rate against bit width. With a bandwidth limited signal, a multi-bit converter can for instance reduce the bit width by one bit when the sampling rate is doubled (as there is more time to use integration to reach the same signal strength again).

What I'm asking is that you don't double the sampling rate but then still expect the same signal jumps to be captured. You should only expect to capture half the signal jumps between samples. As you can jump half the distance twice in the same amount of time as before, you can capture the same signals. Anything more than that, would be asking to deal with more bandwidth.
08-01-2015, 02:39 AM   #276
Senior Member




Join Date: Dec 2013
Posts: 223
QuoteOriginally posted by Ian Stuart Forsyth Quote
I never trust tests done using vertical and horizontal measurement on AAless cameras.
I always take test charts results with a grain of salt, since i myself do not have perfect vision and some of the idiot reviewers will claim camera a (filterless) is better than camera b (has aa filter) even though the effects of Moiré is clearly present at the point said reviewer claimed is the max usable resolution

at the end of the day the best looking photos are usually the ones that look aesthetically pleasing to the eye not the one with moiré distortion

anyway expect more AAless camera's to come out in the future since its cheaper to manfacture and one less piece to rob a little light
08-01-2015, 08:43 AM   #277
Veteran Member
Ikarus's Avatar

Join Date: Apr 2010
Location: California
Photos: Albums
Posts: 471
QuoteOriginally posted by Class A Quote
Anything more than that, would be asking to deal with more bandwidth.
Yes, but higher resolution equals higher bandwidth by definition. I really don't quite understand why we still need to belabor the point. The effect is tiny, a bit less so only when even the coarse sensor clips strongly, so essentially negligible in practice considering the scale of dynamic range we're dealing with.
Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
24x36mm, 600mm, ability, aps-c, camera, cameras, depth, depth of field?...can, equivalence jihadis, ff, frame, full-frame, image, inch, lens, lines, lot, lots, magnification, pentax, people, picture, pictures, post, sensor, sensors, stop, sunset
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Depth-of-field bracketing -- is there a way to do this? baj Pentax K-5 16 07-04-2015 02:17 PM
Depth of field and focus trouble with manual lens Craig Barber Troubleshooting and Beginner Help 10 07-09-2014 07:18 PM
diffraction vs depth of field for Q hoffo Pentax Q 6 02-07-2014 09:38 PM
K10D and depth of field rimofheaven Troubleshooting and Beginner Help 7 01-04-2010 09:21 PM
portraits and depth of field raz Pentax SLR Lens Discussion 31 06-01-2007 08:35 PM



All times are GMT -7. The time now is 08:12 AM. | See also: NikonForums.com, CanonForums.com part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top