DR testing results

[Richard Butler]

i will change to uncompressed even though compressed is suppose to be 14 stops on the a7iv. im a bit confussed as bit depth has no correlation to DR.

If you encode in a linear manner, bit-depth defines the maximum DR that you can encode in the file:

https://www.dpreview.com/articles/4...e-not-the-number-of-colors-you-get-to-capture

This isn't true for anything where you've applied a curve to the data first, but in linear Raw, it can act as a limit on how much you can store.

Richard - DPReview.com

 

[A74Me]

i will change to uncompressed even though compressed is suppose to be 14 stops on the a7iv. im a bit confussed as bit depth has no correlation to DR.

If you encode in a linear manner, bit-depth defines the maximum DR that you can encode in the file:

https://www.dpreview.com/articles/4...e-not-the-number-of-colors-you-get-to-capture

This isn't true for anything where you've applied a curve to the data first, but in linear Raw, it can act as a limit on how much you can store.

Richard - DPReview.com

After my more accurate testing this morning im going to give that one to you the 14 bit can definatly play a big roll ,expecially the colours in the mids when lifting a 7 stop under exposed image not as much in the highlights at 3 stop over exposed though. ive learnt somthing today pushing images to the extreme. Thanks for your info in this thread.

 

[JimKasson]

i will change to uncompressed even though compressed is suppose to be 14 stops on the a7iv. im a bit confussed as bit depth has no correlation to DR.

If you encode in a linear manner, bit-depth defines the maximum DR that you can encode in the file:

I prefer to call it precision, rather than it-depth. Long before we had digital cameras, we were using the word precision to define the concept.

 

[A74Me]

i will change to uncompressed even though compressed is suppose to be 14 stops on the a7iv. im a bit confussed as bit depth has no correlation to DR.

If you encode in a linear manner, bit-depth defines the maximum DR that you can encode in the file:

I prefer to call it precision, rather than it-depth. Long before we had digital cameras, we were using the word precision to define the concept.

Agree, precission is a good discription.

 

[A74Me]

i have included 3 files 3 stops over exposed 7stop under exposed and my studio work starting point for all my shoots.

10 stop

 

[JimKasson]

i have included 3 files 3 stops over exposed 7stop under exposed and my studio work starting point for all my shoots.

10 stop

Requires a Microsoft OneDrive account.

 

[bclaff]

i will change to uncompressed even though compressed is suppose to be 14 stops on the a7iv. im a bit confussed as bit depth has no correlation to DR.

If you encode in a linear manner, bit-depth defines the maximum DR that you can encode in the file:

https://www.dpreview.com/articles/4...e-not-the-number-of-colors-you-get-to-capture

This isn't true for anything where you've applied a curve to the data first, but in linear Raw, it can act as a limit on how much you can store.

Richard - DPReview.com

FWIW, Jack Hogan and I independently showed that (engineering) dynamic range is limited by quantum noise to about 1/2 stop higher than bit depth. So, for example, 14.5 stops for a 14-bit raw.

 

[A74Me]

i have included 3 files 3 stops over exposed 7stop under exposed and my studio work starting point for all my shoots.

10 stop

Requires a Microsoft OneDrive account.

drop box

https://www.dropbox.com/scl/fi/rx4w...ey=mohv0jvmjre25nxby2199yvtp&st=cy2y18do&dl=0

Dropbox

www.dropbox.com

 

[A74Me]

i have included 3 files 3 stops over exposed 7stop under exposed and my studio work starting point for all my shoots.

10 stop

Requires a Microsoft OneDrive account.

drop box

https://www.dropbox.com/scl/fi/rx4w...ey=mohv0jvmjre25nxby2199yvtp&st=cy2y18do&dl=0

https://www.dropbox.com/scl/fi/yabh...ey=n56wtetml3ivzfvephpqdtmqz&st=ivjydizz&dl=0

this is halarious, dropbox has done a better job of the thumbnail exposure in a single pass than ACR has done over 2 passes

 

[A74Me]

I love fun facts, so i have been useing my a6700 side by side to my a7iv for my pro shoots, the a6700 is really good and very close to the image quality which is great as i now use a 2 camera lens combo, but after my agonising discussion points and multipul testing i have discovered that using lossless compressed raw on the a6700 has brought the image quality even closer to the a7iv useing compressed raw. compressed raw (33meg) on the a7iv has a file size of half that of uncompressed raw (70meg) yet uncompressed raw offers practically no advantage in shadow noise on the a7iv and 1/2 stop on the highlights. but with the a6700 losses compressed raw at 28 meg and compressed at 25 meg the lossless offers a good 1 stop better noise perfomance im stoked, im going to do some more tests with some selfies now just to confirm the results.

 

[A74Me]

i knew the a6700 was good but just did some lifted portrait shadows vers the excelent a7iv and after viewing the same images at 2000% i can say that the noise is exactlly the same , i knew there was some pixy dust there and a certain person told me about that 26meg sensor that the fx30 shared with the a6700.

now im really going to push the sensor on skin tone shadows.

 

[Detail Man]

... I independently showed that (engineering) dynamic range is limited by quantum noise to about 1/2 stop higher than bit depth. So, for example, 14.5 stops for a 14-bit raw.

W. R. Bennett of Bell Labs analyzed the spectrum of quantization noise in his 1948 paper.

 

[JimKasson]

i have included 3 files 3 stops over exposed 7stop under exposed and my studio work starting point for all my shoots.

10 stop

Not sure wha the point is, but here are the raw histos.











--

the last word the last word - Photography meets digital computer technology. Photography wins -- most of the time.

Photography meets digital computer technology. Photography wins -- most of the time.

blog.kasson.com

 

[A74Me]

i have included 3 files 3 stops over exposed 7stop under exposed and my studio work starting point for all my shoots.

10 stop

Not sure wha the point is, but here are the raw histos.

the exercise was to demonstrait the 10 stop range of recoverable DR for my work flow, which also basically matched the same test by just under and over exposing neutral grey, skintone, green grass , ect. the claimed 14/15 stops

 

[JimKasson]

i have included 3 files 3 stops over exposed 7stop under exposed and my studio work starting point for all my shoots.

10 stop

Not sure wha the point is, but here are the raw histos.

the exercise was to demonstrait the 10 stop range of recoverable DR for my work flow, which also basically matched the same test by just under and over exposing neutral grey, skintone, green grass , ect. the claimed 14/15 stops

I don't see how those shots produce quantitative results.

--

the last word the last word - Photography meets digital computer technology. Photography wins -- most of the time.

Photography meets digital computer technology. Photography wins -- most of the time.

blog.kasson.com

 

[A74Me]

Woke up this morning with the puzzel complete. i know where the 14 stops of DR come from. as my 22 yo daughter (science/bioligy, teacher) said to me the other day after a uni class studying medical health science/masters in natropothy , the Lecturers have no idea how to teach complex topics to the students because the are not qualified teachers, so she asked if she could teach the specfic class which the lecturer agreed on in amazment at her ability to comunicate to all levels of academic mindsets. over the years I have noticed no one has actually writen or discussed where the 14 stops of DR are measured from, only after reading a video camera spec sheet from facebook last night it all became apparent while i was sleeping.. Talk about black and white turning into grey literally at least i got there in the end from the maze of posts which didnt lead to the finish line

 

[A74Me]

Woke up this morning with the puzzel complete. i know where the 14 stops of DR come from. as my 22 yo daughter (science/bioligy, teacher) said to me the other day after a uni class studying medical health science/masters in natropothy , the Lecturers have no idea how to teach/comunicate complex topics to the students because the are not qualified teachers, so she asked if she could teach the specfic class which the lecturer agreed on in amazment at her ability to comunicate to all levels of academic mindsets. over the years I have noticed no one has actually writen or discussed where the 14 stops of DR are measured from, only after reading a video camera spec sheet from facebook last night it all became apparent while i was sleeping.. Talk about black and white turning into grey literally at least i got there in the end from the maze of posts which didnt lead to the finish line

Marketing lets just say, non disclosure works, if they dont ask dont tell them

lets see if anyone can sum up 14 stops DR measument in a single line post

 

[JimKasson]

lets see if anyone can sum up 14 stops DR measument in a single line post

log2(fullscale/readNoise) at base ISO.

Fujifilm GFX 100S II EDR - the last word

Lensrentals has loaned me a Fuji GFX 100S II for testing. Today I'm reporting on engineering dynamic range (EDR). In single shot mode at putative 14-bit precision: Same as the GFX 100 II, including the odd results at ISO 80, which in that camera was the result of gratuitous in-camera black point...

blog.kasson.com

 

[A74Me]

lets see if anyone can sum up 14 stops DR measument in a single line post

log2(fullscale/readNoise) at base ISO.

https://blog.kasson.com/gfx-100s-ii/fujifilm-gfx-100s-ii-edr/

slog3 is shorter

AI: It gives you more dynamic range which makes post processing easier because it's less likely for highlights to be blown out or shadows to be crushed.

no formulas allowed

your answer was compressed to 2 bit jpeg

AI answer.

The expression "log2(full scale / read noise)" at base ISO is a key parameter in understanding and comparing camera noise and dynamic range. It represents the number of "stops" of dynamic range a camera can capture before its image signal is dominated by noise. Essentially, it quantifies how much "clean" signal you have before noise becomes a significant factor.

 

[JimKasson]

lets see if anyone can sum up 14 stops DR measument in a single line post

log2(fullscale/readNoise) at base ISO.

https://blog.kasson.com/gfx-100s-ii/fujifilm-gfx-100s-ii-edr/

slog3 is shorter

AI: It gives you more dynamic range which makes post processing easier because it's less likely for highlights to be blown out or shadows to be crushed.

no formulas allowed

Be serious. This is PS&T. Keep up.

your answer was compressed to 2 bit jpeg

AI answer.

The expression "log2(full scale / read noise)" at base ISO is a key parameter in understanding and comparing camera noise and dynamic range. It represents the number of "stops" of dynamic range a camera can capture before its image signal is dominated by noise. Essentially, it quantifies how much "clean" signal you have before noise becomes a significant factor.