[biz-engineer]

They realy make over 100 megapixels sensors, just they average nearby pixels to make noise harder to notice.

Agreed. I had the impression that no matter the test charts, images with more pixel density (same sensor size), look smoother than images taken with larger pixels (same sensor size), while SNR test charts indicate more noise for the higher pixel density sensor.

Now, with regards to dynamic range, the charts tell me values around 13ev - 14ev, I've never been able to exploit more than 8 stops of exposure latitude.

Considering true SNR measurements again for any full frame CMOS sensor released in the last five years, they all deliver max SNR compatible with 8bits JPEG: all sensors deliver between 42dB and 49dB of SNR max, and no sensor deliver more than 49.9 dB => fortunately, to avoid banding. There is so little difference between sensors of any brand, that it won't make any practical difference. So for me now, choice of camera only comes down to camera ergonomics, camera software features, stabilization, burst rate, autofocus, availability of lenses, flash and eventually video.

[TedH42]

I need another shipment of "equivalence" popcorn.

[biz-engineer]

I need another shipment of "equivalence" popcorn.

Does anyone know if a cup of small popcorns gets eaten faster or slower than a cup of larger popcorns?

[Wasp]

I have taken a shine to dark shadows lately. More dynamic range is not always what is needed. I actually tweaked this picture for darker shadows. i.e. less shadow detail, in Darktable.

[Dartmoor Dave]

Upsampling = extrapolated data (mathematical best guess for what was never captured) Downsampling = judicious pixel removal (mathematical best guess for retaining the gist of what was captured) For fun and games, one can do a 2X upsample followed by a 0.5X downsample and compare the end result with the original image of the same pixel dimensions. Then do the reverse for complete amusement.

This not intended as an argumentative reply, because I know you know your stuff. But what is turning 50x50 pixels into 400x400 pixels, as in the OP's examples, if it isn't upsampling? And since even a 4K monitor is only about 8.5 megapixels, how can we view photos taken with more megapixels than that without downsampling?

As I say, this is a genuine request for clarification, not an argument (which you would win, because I fully accept that you know more than me).

[stevebrot]

As I say, this is a genuine request for clarification, not an argument (which you would win, because I fully accept that you know more than me).

Sorry...I should have been explicit that we are in agreement.


Steve

[jbinpg]

@wasp - You are my hero

[Kunzite]

There's some truth to what you say.

Never underestimate humanity's willingness to pay 100% more for a 1% improvement that lets them shout "my gear is top-ranked!" The measurebators and "rank" amateurs only want the best-of-the-best. I'd like to say "let them have their fun" if only they weren't so noisy about it.

Never underestimate humanity's willingness to pay 100% more for a 1% improvement they don't need, and a 10% worsening of a characteristic that is actually of some importance to them.

[Rondec]

This makes sense. The question I suppose is how much one is willing to pay for 1 EV difference in dynamic range at base iso. The thing is that full frame cameras are now priced as cheaply as top end APS-C cameras used to be priced and therefore it is no longer a situation (like medium format is now) where you might want to try it out, but the cost is astronomical to do so. A thousand dollars is still a lot of money, but it is better than where full frame cameras used to be priced.

I do own a K3 and a K-01 along with a couple of K-1 cameras and regardless of the explanation, the difference in performance at low iso (the amount I can bump the shadows before detail goes away and noise becomes too bad) and high iso (both with regard to noise and dynamic range in these images) is significant. I could (and did) get along with APS-C cameras for a long time, but there are differences and they are visible in real life situations.

[Dartmoor Dave]

Sorry...I should have been explicit that we are in agreement.


Steve

Thanks Steve. Apologies for misinterpreting your reply. Makes me realise that I'm a bit tired and cranky after some real world stuff that was going on today, so I'm definitely better off avoiding threads like this one.

[photoptimist]

I need another shipment of "equivalence" popcorn.

That's going to create vigorous arguments over FF (full fryer) versus APS-C (air-popped system - corn) and how they differ in DoF (depth of fill).

But then someone will complain that they can never be equivalent because FF is multi-coated (oil + butter).

As with all things equivalence, no kernels of truth will survive.

[Serkevan]

That's going to create vigorous arguments over FF (full fryer) versus APS-C (air-popped system - corn) and how they differ in DoF (depth of fill).

But then someone will complain that they can never be equivalent because FF is multi-coated (oil + butter).

As with all things equivalence, no kernels of truth will survive.

I'm just gonna hold my M series glass next to my phone's camera and call it a day, although focusing might be complicated

---------- Post added 10-27-19 at 04:23 PM ----------

This not intended as an argumentative reply, because I know you know your stuff. But what is turning 50x50 pixels into 400x400 pixels, as in the OP's examples, if it isn't upsampling? And since even a 4K monitor is only about 8.5 megapixels, how can we view photos taken with more megapixels than that without downsampling?

As I say, this is a genuine request for clarification, not an argument (which you would win, because I fully accept that you know more than me).

If you view the image at 100% it's neither upsampled nor downsampled. You are simply viewing a crop small enough to match the resolution of your monitor.

[BigMackCam]

Any more "equivalence" blather deserves pelting with rotten apples, squishy oranges or wet aubergines until they all promise to shut up and go away.

It's not personal, just not interested, over and out.

The subject frustrates many folks, that's for sure - but others find it interesting. There's room for us all here, so long as discussions are respectful

[beholder3]

A given film base will have the same intrinsic grain properties regardless of the size of negative, but the final result will look much nicer with a large format negative.

Yes, it would have been best if my sample could be an analogue film one, because digital versus film doesn't change anything at all, but the digital example misleads too easily to think about "pixels" where those are irrelevant. It is about areas.
I had to choose a digital example on a digital forum board.

Really? Does magnification change the difference in magnitude between the darkest point and the brightess point?

Yes, absolutely. It doesn't change what is recorded on the sensor (which stays the same physical size). It hugely changes what (noise) we perceive.

Step back from the monitor 3 meters (reducing the magnification) and suddenly you will perceive the enlarged crop image from above will not be noisy any longer, but all black. Contrast up. Dynamic range up.

Or put your nose right to the screen in front of one of the bright white noise pixels (enlarging the image in your perception). Not black any longer. Contrast down. Dynamic range down.

And since even a 4K monitor is only about 8.5 megapixels, how can we view photos taken with more megapixels than that without downsampling?

Pixelcount is not of relevancy. The "magnification" only depends on the ratio between the physical size of the sensor and the physical size of the display medium.
An APSC sensor is 3,85 cm². A 27 inch 16:9 monitor is 201 cm². That means any image displayed is magnified 50x if displayed full screen fitting.

[Dartmoor Dave]

Pixelcount is not of relevancy. The "magnification" only depends on the ratio between the physical size of the sensor and the physical size of the display medium.An APSC sensor is 3,85 cm². A 27 inch 16:9 monitor is 201 cm². That means any image displayed is magnified 50x if displayed full screen fitting.

We went over this at interminable length in the last equivalence thread. The physical size of the sensor is NOT optically enlarged to the physical size of the monitor. Once the charge collected by each photodiode has been read and stored as a digital value, that data has no physical size. It's just pixel values. Then, if we want to view that data on a monitor as an image, the data is resampled to the monitor resolution (almost always downsampled and almost never upsampled).

The physical size of the monitor is irrelevant, as you can confirm by the simple experiment of looking at the graphics settings in your computer's operating system and observing that there is no way to enter physical size. Only pixel resolution.

And to repeat: there is no magnification taking place in your original examples, but only upsampling, which is exactly the opposite of what usually happens when we look at digital images. After the stage where the image projected onto the sensor by the lens has been converted to digital values, there is absolutely no resemblance between digital photography and film photography in any way at all.