[kenspo]

I have a K-1 *and* a Sony A7, Kenneth.

Sony is one of the companies trying to highjack me

[clackers]

Sony is one of the companies trying to highjack me

Don't do it, Kenspo, you are a rocker and have a soul!

I'm joking. I remember you saying that the Norwegian distributor, not Ricoh itself, is paying at the moment.

You are a professional and must go with wherever the work is, of course, for your career and family.

But do us a favour if you leave. When you do a Gary Fong style YouTube advertisement, touch an ear whenever you want to tell us the script is not true.

[kenspo]

Yes, that is correct! But no worries hehe..Im so happy at the moment, that it would be too expensive for anyone else to get me to switch!

[Nicolas06]

(Laughs). You would just try to change the subject from noise to something else over which there is no disagreement, @Nicolas, like post processing or resolution.

yep the point of Snr is only just to spend long summer evening on Pentax forum... D we use print or screen. The offcial is print, everybody use print... But let's use screen.

That using screen may have no actual link with actual practical results is of no importance We speak theory refuse the mathetical proof in the middle that explain normalization and also explain that practical matters are completely unrelated.

And DxO's 'normalization'? Your post 110 showed you did not understand it. I had to explain it to you in Post 115!

We don't have the same understanding of it. And the DxO expect neither... The practice is conclusive with the print and using normallization and also the DxO number used for their scores... But still you would continue on forever on why screen is the only reality and that you can't expect the result to have any link with actual practical performance...

[clackers]

yep the point of Snr is only just to spend long summer evening on Pentax forum... D we use print or screen....

It has been explained to you, @Nicolas06, that you are clinging to the superstition of the Print Screen and its made up numbers.

No other testing site does this, just DxOMark in its ecccentric way.

That you are dazzled by DxO and their high profile is clear. As well as your fantasy of 'Total Light', an old wives tale plenty of other camera owners also believe in.

The Screen tab is real measurements, they never measured anything else.

If you keep talking about 'practical', go back to Post 169 and again, see for yourself the SNR is the same.

[Gimbal]

Since we are referring to old posts, I might as well remind you all about post #138 (made by myself).
To save some time I quote it here.

Give it up Nicolas06, clackers doesn't want to or isn't able to understand the print tab. His examples with the phone sensor or one million pixels one inch sensor show's he doesn't have a clue, even though it has been explained numerical times for him.

[Nicolas06]

Since we are referring to old posts, I might as well remind you all about post #138 (made by myself).
To save some time I quote it here.

100% agree. I given up... It's a lost cause.

[clackers]

@Gimbal and @Nicolas, you've some extraordinarily incorrect beliefs and I've marked them for future use for everyone to see. :-)

[Nicolas06]

@Gimbal and @Nicolas, you've some extraordinarily incorrect beliefs and I've marked them for future use for everyone to see. :-)

We think the same for you... So what ? Each of us sure to be right anyway. We are no reason to be more impressed by your position than you are by our own. Sure we have the benefit of having DxO behind us, most reviewer and most people here but this doesn't mean we are right... Or that you are right neither .

Reality doesn't have to be what you or us believe.

[Nicolas06]

@Gimbal and @Nicolas, you've some extraordinarily incorrect beliefs and I've marked them for future use for everyone to see. :-)

We think the same for you... So what ? Each of us sure to be right anyway. We have no reason to be more impressed by your position than you are by our own. Sure we have the benefit of having DxO behind us, most reviewer and most people here but this doesn't mean we are right... Or that you are right neither .

Reality doesn't have to be what you or us believe.

[clackers]

Sure we have the benefit of having DxO behind us

DxO are not behind you, @Nicolas, I had to explain to you that sensor size is not part of their 'normalization.'


Any time you want me to be a guest speaker at your 'Total Light Believers Fan Club' meetings, I'm happy to be there.

[Gimbal]

Neither is the size of a pixel, or the technology used.

Instead they look at the result, but as it turns out the size of a pixel does matter for the result, and the technology used as well. They also look at the number of pixels in their formulas when calculating the print graph.

So good pixels (large and with latest technology) and plenty of them is a winning concept, and guess what, that means a larger sensor has an advantage due to size, EVEN though size is not a direct input.

[crussellsprout]

Actually, both the sensor size and the pixel size ARE part of the DxO normalization, since the quantity they use to normalize by, N=number of pixels, is almost exactly the ratio of sensor area to pixel area (moreso for recent sensors than those from the early days when there would be more wasted space around each pixel -- now the wasted space is quite negligible)! (I made a similar comment before, here: https://www.pentaxforums.com/forums/169-pentax-full-frame/315763-k-1-so-what-...ml#post3577426 )

The intention of the print score is to be able to compare sensors fairly. As DxO says "Normalizing by pixel resolution makes it possible to compare actual sensor performance, similar to printing two images on the same printer." This is much more useful than simply comparing per-pixel performance.

If the DxO print score was instead calculated using a normalization based around the ratio of pixel area, rather than #pixels, THEN it would be true that it would not be affected by sensor size at all. But that would be silly, as it wouldn't serve the intended purpose of their normalization. They already have the screen score which is the one that only depends on pixel area.

Further emphasis of why users should pay attention to Print sensor scores rather than Screen scores when comparing sensors (from http://www.dxomark.com/Reviews/DxOMark-Camera-Sensor2/How-fair-is-DxOMark):

"Print versus Screen modeTo compare DxOMark Camera Sensor scores between cameras with different resolutions, you need to look at the “Print” results. The overall DxOMark Camera Sensor score is “Print” level only, which is fine. For the next level of detail a viewer gets to choose between Print and Screen. This is less fine: Screen is not normally useful for end users (it can be useful for debugging your own calculations). The lowest level of data is presented in “Screen” mode only, but is not labeled as such. I would prefer to see all data to be labeled Print/Screen or –better yet– Normal/100%. Normal would stress that this is what matters. And 100% is similar to pixel peeping: here you look at the noise at the 100% crop level and loose the overview of what it means at the image level."

[zapp]

I keep hearing how superior FF is over APS-c regarding the issue of depth of field, yet no one has ever really shown me an example of this superiority in a photo... So I ran the numbers through Cambridgeincolour's Depth of field calculator to see exactly what kind of "superiority" we are talking about.


A 35mm lens on an APS-c camera with a 1.5 crop factor at f1.4 and focused at 2 feet gets you from 1.97ft to 2.03ft in focus. That's 0.06 feet.


Focused at 10 feet you get from 9.32ft to 10.78ft in focus. That's 1.46 ft.



A 50mm lens on a FF camera at f1.4 focused at 2 feet gets you 1.98ft - 2.02ft in focus. That's 0.04ft.


Focused at 10 feet you get 9.49ft - 10.57ft. That's 1.08 feet.




So how about 200mm/300mm fashion depth of field, right?


200mm lens at 2.8 focused at 10 feet gives you 9.96-10.04 feet in focus which is 0.09 feet on an APS-c 1.5x crop.


300mm lens at 2.8 focused at 10 feet gives you 9.97-10.03 feet in focus which is 0.06 feet on a FF sensor.




So clearly FF is about 30-40% better at shallow depth of field shots, on paper. I have been a professional photographer for over 30 years, I'd be hard pressed to take 2 photographs that could actually show anyone these differences.


There are many reasons to buy a FF camera, depth of field superiority really just isn't one of them in my opinion and I think the math and practical application back me up. If you can't see the difference, who cares?

Your math is not completely correct. You need to take pixel pitch into account - typically smaller for APS-C cameras so actually supporting your argument. At the same time DOF is not only a value telling you what is in focus, but also how out of focus areas are blurred. Focal length is not linearly related to depth of field but with a power of 2, so DOF really decreases quicker for larger focal length and this also includes background blur. Any 10% difference will make a difference, but that can also be lost by focus inaccuracies. Comparing medium format results to 35 film we worked with the same sensor - called film in those days - and were able to see differences. The cropt factor was around 1.6.

[Nicolas06]

Your math is not completely correct. You need to take pixel pitch into account - typically smaller for APS-C cameras so actually supporting your argument. At the same time DOF is not only a value telling you what is in focus, but also how out of focus areas are blurred. Focal length is not linearly related to depth of field but with a power of 2, so DOF really decreases quicker for larger focal length and this also includes background blur. Any 10% difference will make a difference, but that can also be lost by focus inaccuracies. Comparing medium format results to 35 film we worked with the same sensor - called film in those days - and were able to see differences. The cropt factor was around 1.6.

This is not only the dof, but also how much background blur you get, or if you prefer how much subject isolation.

Using an online too, here I compared a 35mm f/1.4 on APSC with a 50mm f/1.4 on FF: How much blur? - A visual background blur calculator

You can see that the background is much more blured with the FF and that to achieve the same blur with the FF you only need f/2. instead of f/1.4. Counting that many lenses will be much sharper at f/2 on FF than f/1.4 on APSC, this is quite practical, if the APSC subject isolation is enough.

Another way to see it if with zooms. What would require a not that common f/1.8 zoom lens on APSC doesn't only require an f/2.8 zoom on FF. And if you zoom is a bit soft, not so contrasty wide open, with the FF you can close down one stop more for the same blur.