Sony's New a7RII Camera Delivers World's First Back-Illuminated FF Sensor

Jan 21, 2015
262
148
neuroanatomist said:
dilbert said:
What Does It Mean in the Real World?

Like a lot of laboratory testing, probably not a lot. Adapters couldn't all stink or people wouldn't use them. Like a lot of tests, you can detect a very real difference in the lab that doesn't make much difference at all in the real world.

...a concept that applies to things other than adapters.
Wait.. what do you mean? Something like.. DR? Nooo... no way. :eek: :eek:
 
Upvote 0
dilbert said:
neuroanatomist said:
LetTheRightLensIn said:
But at some point, eventually, some start giving up or at least going for mixed solutions where instead of adding a new Canon body to replace their old one they add a Sony to their old Canon.

Many who are thirsting for more DR are only just now starting to do that.

But at some point, eventually, some start giving up or at least going for mixed solutions where instead of adding a new Sony body to replace their old one they add a Canon to their old Sony.

Many who are thirsting for better native lens selection, better ergonomics, or better service are only just now starting to do that.

I'm sure both are correct...but how many are 'some' and 'many'?

Canon has more customers to lose than Sony.

Or to put it differently, if Sony gets 1% of Canon owners to switch and Canon gets 1% of Sony owners to switch, Sony has the net gain in absolute terms.

Of course it needn't be a zero sum game. Someone buying a Sony body and Canon lenses benefits both companies.
 
Upvote 0
dilbert said:
Hillsilly said:
...
Sure, unless you are interested in image quality. My travel kit is a Fuji X100, an X-E1 and a 14, 35 and 60mm lens (or sometimes a 55-230mm). It takes up a small fraction of my carry-on luggage. I feel that the image quality from the Fuji cameras is better than Canon crop sensors. But that's probably largely to do with have no AA filter on the X-E1 and the quality of the Fuji lenses. To get anything similar in the Canon (or Tamron) line, you're talking FF camera with bigger lenses. And all of a sudden, you're considering which camera bag is going to be small enough to use as a carry on bag... Unless I was going to photograph a sporting event, Canon can't really compete.
...

There are several things wrong here.

First, as you know, removing the AA filter is bad. This forum is filled with those who will tell you that once the AA filter is removed (as other manufacturers have done) that moire destroys your images. Doesn't matter about lens quality. Just look at comments around the 5DsR. No AA filter = bad. Why haven't you learnt that yet?!?!?!?! Aren't you reading enough comments from know-it-alls that have a hand picked line of experts to back them up?

Second, as you should know, that smaller camera with the smaller sensor, well, high ISO isn't going to work well at all well with that and without top quality high ISO there is no need to own a camera because then you can't take pictures of birds to put next to the pictures of cats. Now if only the cat pictures could come to life...

Ok, let me be serious for a second..

You're absolutely spot on but what you represent with that is something different and something different means change and people are afraid of change so they're afraid of what you've said and what it represents. It's also different to what they've bought and that challenges their purchasing decision making. There's a lot of fear of the new and unknown here and that is best represented in the rejections you see here of other companies and technologies...

Some people are afraid of change. Some people only care about novelty. Most people are somewhere in between. It's sensible to be cautious with new technologies. It's hard to know in advance what will ultimately take off, work better, and become the new normal.
 
Upvote 0
fragilesi said:
privatebydesign said:
Social media, yes, blowhard forums like this where any valuable advice is drowned in tidal waves of irrelevant tantrums and bickering by unimaginative, anonymous, misguided, over opinionated and under experienced wannabe experts, no.

Just watch the webinar from B&H on the 5DS/R release with a true master, Gregory Heisler, to understand how much DR, or the lack of it, is an issue to actual master photographers and it makes almost any participation here as pointless as King Canute commanding the tide to stop, actually Canute made a very good point, maybe somebody did here once.

Hard to believe otherwise though at times. Lots of low post count people turning up whose sole interest seems to be bigging up Sony etc. In one case claiming to be cancelling a previously unmentioned 5ds order in favour of this model . . . there must be some kind of element of that going on.

Wasn't there an empirical study done a while back that indicated most scathing reviews by people who don't own a product aren't instigated by the producers, but by individual brand loyalists who take it upon themselves to defend/attack online?
 
Upvote 0
canonvoir said:
Orangutan said:
canonvoir said:
I just purchased an a7ii and quite a few native lenses. I am really impressed with the image quality of the Sony. So much so, I have not used my 5Diii since making the purchase. It also helps I like to hike and the weight reduction was a nice bonus.

This sports season I will still be using my 1DX but I will carry a Sony with a 55mm as well to see how it does for end zone and after game photos.

I will be picking up the a7rii and may even sell off my 5Diii. I hate to say it, but I believe I have purchased my last Canon body. Image quality is too impressive for me to look over. The biggest drawback is battery life. If they could double the battery life (not a big deal until you decide to do a really long time lapse and admittedly a battery grip for time lapse would solve this), it would be a no brainier for all of those minus sports pros.
This is a reasonable personal statement. Thank you for not crossing the Swedish Line to imply that those who have not reached the same conclusion are ______.

Question: are you seeing any drawbacks with the mirrorless design? Focus speed/accuracy? Since you have a 1DX I presume you'll use the Sony mostly for slow-/non-moving subjects, right?

Sorry you feel I crossed some line. I was simply stating my experience.

Please re-read: I wrote that you had "not" crossed a line. That was sincere, not sarcastic.
 
Upvote 0

Sporgon

5% of gear used 95% of the time
CR Pro
Nov 11, 2012
4,719
1,537
Yorkshire, England
Aglet said:
Sporgon said:
>However what I don't like is that the raw data is cooked: the whites are too white, the
>blacks too black. The blue was far to intense and is still inaccurate, the woman's jersey
>was a more powdery pastel blue, very different to the guy's blue coat that he has tied
>around his waist.

I'm not sure what body you've been shooting with (Sony?) but you could profile it for more accurate color response.
yes, those 2 blues are rendered quite similarly and not like how you've described.
There's good tonal info on the dog and the guy's vest tho flat overcast light like that means a low DR scene once again, except for the bright areas along horizon line where there may not have been much to keep. This looks like a totally Canon-able shot and the exmor would show little advantage other than a cleaner dog and black vest w-o color blotches or plaid FPN. A 6D, or even my old 60D, could have comfortably handled that scene.

>When you really force the issue and lift deep shadow where the Canon
>hasn't seen information, the exmor is actually adding to the information, and creating
>detail that isn't there. It is not 'honest' data like the Canon raw, it's giving unwanted
>interfering.

I'm really not sure what you're telling us here. Those of us familiar with with using exmor files aren't finding fake information, we just have an option to decide how much of the deep shadow reality we might want to render into the final result.

>However I can see how people might prefer it, but in no way can you say that at low ISO the
>exmor data is better than the Canon - unless you lift deep shadow by three stops or more.

You produce large prints so you realize that there's more tonal compression happening in the printing process to squeeze that electronic file DR into the smaller DR of printed media, there's not just the shadow lift you might do to create a certain on-screen look while editing. Having that clean info available down low means you can maintain more fidelity not only in the editing push but also in the final hardcopy output which applies some extra push too.
I'll always prefer having that optional range available and that's why I use ABC gear for challenging DR scenes where files may have to get tortured a bit in post to create what I want for final output. I'm not always a fan of crushed black used to cover up sensor shortcomings or to replicate a DR limitation some cameras are set to use as a default rendering. I want final output the way _I_ want it, not necessarily how an OEM tone-curve delivers it. Exmor/ABC provides that option for me better than anything else at the moment.

You are absolutely right, I shouldn't have or be making comments about how the camera is handling the data vs the Canon. I'm not in a position to do that given that my only intention is to find out how the 'extra two stops of dr' help me.

And so far it's "two stops, what two stops ?"
 
Upvote 0
Jan 29, 2011
10,675
6,121
dilbert said:
privatebydesign said:
...
Roger at Lens Rentals has tested the use of glassless adapters, and his conclusions were exactly what yours were. For lenses that are being used on the same sized sensor as the conversion, ie FF-FF, the center didn't do too badly but the corners lost lots of resolution.

Basically the squareness of the lens mount is so critical nowadays that the 'simple' engineering used for adapters just isn't up to the task.

http://www.lensrentals.com/blog/2013/09/there-is-no-free-lunch-episode-763-lens-adapters

And let me pull the appropriate line from that blog entry:

What Does It Mean in the Real World?

Like a lot of laboratory testing, probably not a lot. Adapters couldn't all stink or people wouldn't use them. Like a lot of tests, you can detect a very real difference in the lab that doesn't make much difference at all in the real world.

I posted the link in response to a user that was evidencing the exact same results. From that you could draw a few conclusions:-
1/ The poster is wrong or lying.
2/ Something else could be causing the exact same IQ issues.
3/ The predicted results are actually visible for some combinations of some gear.
 
Upvote 0
Jul 21, 2010
31,088
12,851
dilbert said:
neuroanatomist said:
dilbert said:
What Does It Mean in the Real World?

Like a lot of laboratory testing, probably not a lot. Adapters couldn't all stink or people wouldn't use them. Like a lot of tests, you can detect a very real difference in the lab that doesn't make much difference at all in the real world.

...a concept that applies to things other than adapters.

Unfortunately noisy sensors that lack DR hinder real world work as much as lab work.

Ahhhh, that explains why noisy Canon sensors lacking DR can't produce good images. At least in the hands of some.

Why haven't you switched to Sony yet?
 
Upvote 0

Sporgon

5% of gear used 95% of the time
CR Pro
Nov 11, 2012
4,719
1,537
Yorkshire, England
dilbert said:
neuroanatomist said:
dilbert said:
neuroanatomist said:
dilbert said:
What Does It Mean in the Real World?

Like a lot of laboratory testing, probably not a lot. Adapters couldn't all stink or people wouldn't use them. Like a lot of tests, you can detect a very real difference in the lab that doesn't make much difference at all in the real world.

...a concept that applies to things other than adapters.

Unfortunately noisy sensors that lack DR hinder real world work as much as lab work.

Ahhhh, that explains why noisy Canon sensors lacking DR can't produce good images. At least in the hands of some.

Why haven't you switched to Sony yet?

What camera I own/use when is none of yours or anyone else's business.

Eh ? So how's it your business to keep telling us our Canon cameras are crap ?
 
Upvote 0
Jul 21, 2010
31,088
12,851
dilbert said:
What camera I own/use when is none of yours or anyone else's business.

In that case, you should probably stop babbling about it publicly.


dilbert said:
If Canon are going to be so complacent and consider us to be caged animals that they've already caught then I've got news for them because whilst I haven't seriously considered switching brands before, I am now.

In 12 months time either the Canon DSLR that I own will have substantially better IQ or I won't be using Canon any more.

dilbert said:
4. Would not buy a 5DS
5. Next camera might be a Sony.

Yup, looking that way for me too.

dilbert said:
And to round this out, I'll re-iterate that whilst my first 5 cameras I bought were all Canon, the next one wasn't and it is looking likely that the one after that won't be either.

::)
 
Upvote 0
dilbert said:
neuroanatomist said:
dilbert said:
neuroanatomist said:
dilbert said:
What Does It Mean in the Real World?

Like a lot of laboratory testing, probably not a lot. Adapters couldn't all stink or people wouldn't use them. Like a lot of tests, you can detect a very real difference in the lab that doesn't make much difference at all in the real world.

...a concept that applies to things other than adapters.

Unfortunately noisy sensors that lack DR hinder real world work as much as lab work.

Ahhhh, that explains why noisy Canon sensors lacking DR can't produce good images. At least in the hands of some.

Why haven't you switched to Sony yet?

What camera I own/use when is none of yours or anyone else's business.

Really? You've racked up nearly 4,000 public posts commenting about Canon gear on a Canon gear forum ... and don't wish to share that bit of gear info? It's just THAT private?
 
Upvote 0
dilbert said:
Sounds like you suffer from this:
https://en.wikipedia.org/wiki/Post-purchase_rationalization
Actually Dilbert it sounds like YOU suffer from post purchase rationalisation. You are the one trying to convince everyone to buy Sony. Plenty of people here are perfectly happy with their gear.
 
Upvote 0
I think there is more than plenty of post purchase rationalization to go around on these forums. There is no point in trying to convince anyone to buy different gear. People clearly have their preferences and their reasons for having them. These brand wars persist because no one can let an argument go...

I walked away from this thread...what, days ago...I thought it was about done and dead back then...seems these threads never die. Buy what you want. If your happy with brand lockin, great. If you want something Canon doesn't offer, know there are plenty of great options elsewhere. Bodies come and go, not a lot of harm adding a highly compatible body like the A7r II to your kit for a time until Canon gets around to delivering what you need.

Beyond that...I really agree with Don:

Don Haines said:
Dilbert, Neuro,

This public hate-on you two have for each other demeans both of you. Walk away! Take the high road!
 
Upvote 0
raptor3x said:
Neutral said:
raptor3x said:
Sporgon said:
jrista said:
You CAN preserve the highlights, and still have better shadow tonality, than with a Canon camera. I mean, we are talking about total tonality of around 2100-2400 tones on a Canon, and anywhere from 7300 to 8100 tones or more on Exmor-based cameras. The entire tonal range of a Canon camera can fit within the shadow quarter of the signal on an Exmor...I mean, think about it: 8000/2000...if you consider the bottom quarter of the signal to be "the shadows", you could fit an entire Canon exposure in the shadows of an Exmor, and have the same tonality. Earlier highlight clipping? Saturation falloff? That's a total misnomer. You have GOBS more tonality in an Exmor signal than a Canon has in it's entirety, and you have as much tonality just in the shadows as a Canon has in it's entirety. There is no such thing as early highlight clipping or blue saturation falloff with an Exmor...

That sounds so impressive.

You'll be able to see that 8000 / 2000 difference here then.

Just to reiterate, because you may not have caught it, jrista was confusing dynamic range with tonal range. You can't just take 2^(# stop DR) and say that's the number of tones the camera can represent. Dynamic range represents the ratio between the lowest and highest tone that can be represented, but the actual number of tones that can be represented within that range is dependent on the quantization of the signal into discrete levels, which is in turn dependent on the standard deviation of the signal as a function of intensity. Just as an example, no current 35mm camera is anywhere close to being able to represent 8000 levels of grey in a single shot. The D810 would be closest with up to 910 tonal levels at ISO 64. To compare that with the 1DX, the 1DX has up to 648 tonal level at ISO 100 (the D810 has 792 at ISO 100). Comparing the 1DX and A7S at ISO 12800, the 1DX has a potential of 77 tones while the A7S has up to 84 tones; certainly an improvement but not the revolution jrista implies (at least not in terms of tonality). The real strength of the A7S is how amazingly well it preserves color and detail at high ISO, much better than the 1DX once you get above ISO 25600.

Could you please clarify your calculations on numbers of possible tones values?
This does not seem correct to me.
Here some basic math from theory of signal detection:
In general possible number of detected tones has limit of number of quantization levels if signal noise is going to zero.
Each quantizatin level is a decision slot for assigning digital value to the received analog signal at the input of ADC. For 14 bits ADC there are 16384 possible representatin of input analog signal. In theoretically ideal situation (with zero input analog signal noise) there are 16384 possible tonal values that could be assigned to received input analog signal.
Now when we come to real systems with noise (regardless of the noise origin) we have fundamental thing which is called SNR which affects precision of the signal detection - in our case to which tonal slot signal will be assigned. More signal noise more probability that signal will be assigned wrong digital value. Roughly if 99% of the signal energy is within particular decision slot ( in the center of it) then there is possibility that there is 99 percent probability that signal will be assigned correct value and 1% that that will be assigned value from adjucent decision slot. For image sensor this will result in 1% variations in image tonality signal with given noise level and noise distribution pattern. If signal value is on boundary of decision slot with the same conditions as above than there will be 50/50 distribution for output value assignements. This is actually why possible tonal values are less than ADC quantization levels.
This is actual limitation of one dimention signal detector when only signal amplitude modulated with noise is taken into account.
So overall all depends on number of signal detector decision slots and intensity and distribution pattern of the signal noise and actual signal level at the input.
If majority of signal noise power spectrum width becomes wider than width of the decision slot than this is where we would see that number of the possible correct tonal numbers would be reduced.
Also errors in signal values assignmets would be more frequent for lower level signals - this is just signal detector SNR function for two input noise varàibles - read noise and photon noise in our case.
I do not think we need to go more deep into that. These are just basics.

So according to all said above Jrista calculations seem correct to me.
If you can actually prove that this is different and Jrista is not correct somewhere I àm really interested to see that.

Sorry for the delay, I was away all week. You're both missing that the width of the quantization for non-overlapping levels is limited by the STD of the signal as a function of signal level. You have to actually perform the integration, or more realistically the summation, over the entire signal range to get the number of tonal ranges. You don't need to take my word for it though, DxO does the exact same thing; that's actually where I took the numbers from.

It is nice to have some interesting and useful discussions.

What you were mentioning and what is illustrated in DXO tonal range chart is basically the same what I was talking about signal detection in presence of noise and having ADC to assigning digital values (combination of analog and digital circuits).
Exactly this relates to the part where I was talking about signal detection decision slot width and width of the signal spectrum containing most of the signal energy (analog tone strip width). Signal detection decision slot width is basically ADC quantization step for ideal (theoretical) signal without noise and for real signal with noise the other limitation is tonal strip width which is STD dependent. Both together puts limit at number of possible tonal values at both ends of dynamic range at the output of ADC.
And DXO tonal range charts bring us to something interesting which might not be actually very obvious.
Even I was not paying attention to this earlier.

Before going there in more details I need to clarify some things related to relations between signal levels , SNR and STD as I feel that this is kind of confusion for most of the people here.
What is important here is what I mentioned before - signal detection decision slot size/width (ADC quantization step) and width of the pure sin(x) signal spectrum with added noise (basically width of the tonal strip before ADC).
Here some simple as possible explanations, by simple steps - simple math and outcomes of that:
1. As a starting point:
STD of ideal signal (e.g. F(x)= A * sin(x) ) without noise is always zero regardless of the signal amplitude. Changes of signal amplitude A do not change STD - it is always zero.
Also number of possible different values between value M and value N is always infinity even if difference between M and N is decreasing to zero (N-M->0).
2. Signal-to-noise ratio is inversely proportional to the relative standard deviation of the signal amplitude
3. STD for signal with the fixed noise level is decreasing if you increase pure signal level and keep noise level at constant value.
What this actually means is that more SNR results in less STD and more narrow is spectrum strip containing 99% of the signal energy.
4. STD for given SNR is constant if you are increasing value of the (signal+noise) not changing SNR - actually doing amplification of the real signal with noise - amplification does not change SNR and as result STD is also not changing. Outcome of this is that noise spectrum width is increasing with the increased level of amplifications. In other words tonal strip width is increasing with amplification and it takes width of more and more signal detection decisions slots.
5. If you decreasing SNR for given pure signal level (adding more noise to the fixed signal) this results in increasing STD of the signal mix with noise . Result of this is that noise spectrum width (or tonal strip) becomes wider. This width eats up more and more ADC signal detection decision slots.
Result of this is that number of distinguishable tonal strips across system dynamic range (before ADC) is also decreasing.
Results for sensor - less tonal values.
6. Worst case scenario when you amplifying signal while decreasing SNR - this results in both increase of STD and non liner but rather exponential increase of noise amplitude spectrum width.
So for sensor this results in even less possible tonal strips across system DR range before ADC and even less values after ADC conversion.
This is actually what is happening when increasing camera ISO settings to compensate for lower input light entering camera - doing more amplification for the signal with less and less SNR to cover full ANC input DR .

All above is illustrates of what we see on the DXO tonal chart and basically this give something which I find interesting for overall systems understanding and understanding better their physical limits.
It seems that the reason why tonal numbers are so close for sensors with different DR and different sensor read noise is the noise factor which is independent from the sensor.
This is actually seems to be photon noise which is reducing tonal range differences between different systems.
Photon noise is basically photons 4D jitter (time+ x,y,z variations).
Result of this that is for given integrating time (exposure ) there will be variations of the signal levels of the optical detector .

It seems that for sensors with low read noise photon noise starts adding more to the overall system noise compared to the read noise. It looks from DXO chart that photon noise is about the same order as read noise for existing Canon sensors (may be somewhat smaller) and more than read noise of best Exmor sensors and as result tonal strip width is bigger than ADC quantization step
As result overall noise (read noise +photon noise) amplitude spectrum width (which is basically visual representation of STD) is very close for all existing systems so we see that that number of possible tonal values (tonal strips) is close for different systems even when system has MF with 16bit ADC compared to DSLR with 14 bit ADC (e.g. look at medium format - e.g. IQ180 chart compared to 1Dx )

Seems that there are not two many ways to have some improvements in this area:
1. Have lower possible in camera native ISO (ability to have longer light (photon noise) integration time and longer read noise integration time)
2. Have higher photocell well capacity (increasing DR by increasing max number of photos received before reaching saturation point). Here is where BSI sensor could be useful. Also still reducing read noise.
3. Both 1 &2 above would reduce tonal strip width and as result increase tonal range at low ISOs
4. Exposure blending - this is basically results in increasing normalized SNR (by reducing STD) - this works across whole ISO range.

All that is not very exiting , may be DXO measurements are not correct somewhere ?
Maybe I am missing something ?
 
Upvote 0
dilbert said:
neuroanatomist said:
dilbert said:
neuroanatomist said:
dilbert said:
What Does It Mean in the Real World?

Like a lot of laboratory testing, probably not a lot. Adapters couldn't all stink or people wouldn't use them. Like a lot of tests, you can detect a very real difference in the lab that doesn't make much difference at all in the real world.

...a concept that applies to things other than adapters.

Unfortunately noisy sensors that lack DR hinder real world work as much as lab work.

Ahhhh, that explains why noisy Canon sensors lacking DR can't produce good images. At least in the hands of some.

Why haven't you switched to Sony yet?

What camera I own/use when is none of yours or anyone else's business.

You actually own a camera? :eek:
 
Upvote 0
Neutral said:
raptor3x said:
Neutral said:
raptor3x said:
Sporgon said:
jrista said:
You CAN preserve the highlights, and still have better shadow tonality, than with a Canon camera. I mean, we are talking about total tonality of around 2100-2400 tones on a Canon, and anywhere from 7300 to 8100 tones or more on Exmor-based cameras. The entire tonal range of a Canon camera can fit within the shadow quarter of the signal on an Exmor...I mean, think about it: 8000/2000...if you consider the bottom quarter of the signal to be "the shadows", you could fit an entire Canon exposure in the shadows of an Exmor, and have the same tonality. Earlier highlight clipping? Saturation falloff? That's a total misnomer. You have GOBS more tonality in an Exmor signal than a Canon has in it's entirety, and you have as much tonality just in the shadows as a Canon has in it's entirety. There is no such thing as early highlight clipping or blue saturation falloff with an Exmor...

That sounds so impressive.

You'll be able to see that 8000 / 2000 difference here then.

Just to reiterate, because you may not have caught it, jrista was confusing dynamic range with tonal range. You can't just take 2^(# stop DR) and say that's the number of tones the camera can represent. Dynamic range represents the ratio between the lowest and highest tone that can be represented, but the actual number of tones that can be represented within that range is dependent on the quantization of the signal into discrete levels, which is in turn dependent on the standard deviation of the signal as a function of intensity. Just as an example, no current 35mm camera is anywhere close to being able to represent 8000 levels of grey in a single shot. The D810 would be closest with up to 910 tonal levels at ISO 64. To compare that with the 1DX, the 1DX has up to 648 tonal level at ISO 100 (the D810 has 792 at ISO 100). Comparing the 1DX and A7S at ISO 12800, the 1DX has a potential of 77 tones while the A7S has up to 84 tones; certainly an improvement but not the revolution jrista implies (at least not in terms of tonality). The real strength of the A7S is how amazingly well it preserves color and detail at high ISO, much better than the 1DX once you get above ISO 25600.

Could you please clarify your calculations on numbers of possible tones values?
This does not seem correct to me.
Here some basic math from theory of signal detection:
In general possible number of detected tones has limit of number of quantization levels if signal noise is going to zero.
Each quantizatin level is a decision slot for assigning digital value to the received analog signal at the input of ADC. For 14 bits ADC there are 16384 possible representatin of input analog signal. In theoretically ideal situation (with zero input analog signal noise) there are 16384 possible tonal values that could be assigned to received input analog signal.
Now when we come to real systems with noise (regardless of the noise origin) we have fundamental thing which is called SNR which affects precision of the signal detection - in our case to which tonal slot signal will be assigned. More signal noise more probability that signal will be assigned wrong digital value. Roughly if 99% of the signal energy is within particular decision slot ( in the center of it) then there is possibility that there is 99 percent probability that signal will be assigned correct value and 1% that that will be assigned value from adjucent decision slot. For image sensor this will result in 1% variations in image tonality signal with given noise level and noise distribution pattern. If signal value is on boundary of decision slot with the same conditions as above than there will be 50/50 distribution for output value assignements. This is actually why possible tonal values are less than ADC quantization levels.
This is actual limitation of one dimention signal detector when only signal amplitude modulated with noise is taken into account.
So overall all depends on number of signal detector decision slots and intensity and distribution pattern of the signal noise and actual signal level at the input.
If majority of signal noise power spectrum width becomes wider than width of the decision slot than this is where we would see that number of the possible correct tonal numbers would be reduced.
Also errors in signal values assignmets would be more frequent for lower level signals - this is just signal detector SNR function for two input noise varàibles - read noise and photon noise in our case.
I do not think we need to go more deep into that. These are just basics.

So according to all said above Jrista calculations seem correct to me.
If you can actually prove that this is different and Jrista is not correct somewhere I àm really interested to see that.

Sorry for the delay, I was away all week. You're both missing that the width of the quantization for non-overlapping levels is limited by the STD of the signal as a function of signal level. You have to actually perform the integration, or more realistically the summation, over the entire signal range to get the number of tonal ranges. You don't need to take my word for it though, DxO does the exact same thing; that's actually where I took the numbers from.

It is nice to have some interesting and useful discussions.

What you were mentioning and what is illustrated in DXO tonal range chart is basically the same what I was talking about signal detection in presence of noise and having ADC to assigning digital values (combination of analog and digital circuits).
Exactly this relates to the part where I was talking about signal detection decision slot width and width of the signal spectrum containing most of the signal energy (analog tone strip width). Signal detection decision slot width is basically ADC quantization step for ideal (theoretical) signal without noise and for real signal with noise the other limitation is tonal strip width which is STD dependent. Both together puts limit at number of possible tonal values at both ends of dynamic range at the output of ADC.
And DXO tonal range charts bring us to something interesting which might not be actually very obvious.
Even I was not paying attention to this earlier.

Before going there in more details I need to clarify some things related to relations between signal levels , SNR and STD as I feel that this is kind of confusion for most of the people here.
What is important here is what I mentioned before - signal detection decision slot size/width (ADC quantization step) and width of the pure sin(x) signal spectrum with added noise (basically width of the tonal strip before ADC).
Here some simple as possible explanations, by simple steps - simple math and outcomes of that:
1. As a starting point:
STD of ideal signal (e.g. F(x)= A * sin(x) ) without noise is always zero regardless of the signal amplitude. Changes of signal amplitude A do not change STD - it is always zero.
Also number of possible different values between value M and value N is always infinity even if difference between M and N is decreasing to zero (N-M->0).
2. Signal-to-noise ratio is inversely proportional to the relative standard deviation of the signal amplitude
3. STD for signal with the fixed noise level is decreasing if you increase pure signal level and keep noise level at constant value.
What this actually means is that more SNR results in less STD and more narrow is spectrum strip containing 99% of the signal energy.
4. STD for given SNR is constant if you are increasing value of the (signal+noise) not changing SNR - actually doing amplification of the real signal with noise - amplification does not change SNR and as result STD is also not changing. Outcome of this is that noise spectrum width is increasing with the increased level of amplifications. In other words tonal strip width is increasing with amplification and it takes width of more and more signal detection decisions slots.
5. If you decreasing SNR for given pure signal level (adding more noise to the fixed signal) this results in increasing STD of the signal mix with noise . Result of this is that noise spectrum width (or tonal strip) becomes wider. This width eats up more and more ADC signal detection decision slots.
Result of this is that number of distinguishable tonal strips across system dynamic range (before ADC) is also decreasing.
Results for sensor - less tonal values.
6. Worst case scenario when you amplifying signal while decreasing SNR - this results in both increase of STD and non liner but rather exponential increase of noise amplitude spectrum width.
So for sensor this results in even less possible tonal strips across system DR range before ADC and even less values after ADC conversion.
This is actually what is happening when increasing camera ISO settings to compensate for lower input light entering camera - doing more amplification for the signal with less and less SNR to cover full ANC input DR .

All above is illustrates of what we see on the DXO tonal chart and basically this give something which I find interesting for overall systems understanding and understanding better their physical limits.
It seems that the reason why tonal numbers are so close for sensors with different DR and different sensor read noise is the noise factor which is independent from the sensor.
This is actually seems to be photon noise which is reducing tonal range differences between different systems.
Photon noise is basically photons 4D jitter (time+ x,y,z variations).
Result of this that is for given integrating time (exposure ) there will be variations of the signal levels of the optical detector .

It seems that for sensors with low read noise photon noise starts adding more to the overall system noise compared to the read noise. It looks from DXO chart that photon noise is about the same order as read noise for existing Canon sensors (may be somewhat smaller) and more than read noise of best Exmor sensors and as result tonal strip width is bigger than ADC quantization step
As result overall noise (read noise +photon noise) amplitude spectrum width (which is basically visual representation of STD) is very close for all existing systems so we see that that number of possible tonal values (tonal strips) is close for different systems even when system has MF with 16bit ADC compared to DSLR with 14 bit ADC (e.g. look at medium format - e.g. IQ180 chart compared to 1Dx )

Seems that there are not two many ways to have some improvements in this area:
1. Have lower possible in camera native ISO (ability to have longer light (photon noise) integration time and longer read noise integration time)
2. Have higher photocell well capacity (increasing DR by increasing max number of photos received before reaching saturation point). Here is where BSI sensor could be useful. Also still reducing read noise.
3. Both 1 &2 above would reduce tonal strip width and as result increase tonal range at low ISOs
4. Exposure blending - this is basically results in increasing normalized SNR (by reducing STD) - this works across whole ISO range.

All that is not very exiting , may be DXO measurements are not correct somewhere ?
Maybe I am missing something ?

Found interesting article related to the discussed above:

Shot noise limits the camera resolution:
http://www.stanfordcomputeroptics.com/technology/dynamic-range/photon-noise.html
Here they are talking about tonal ranges that could be resolved:
"May be its hard to believe but in fact more bits in the A/D conversion can actually not increase the resolution of the data. There is only one chance to increase the resolution: go for higher intensity levels, i.e. longer time integration of the signal to reach higher signal to noise ratios"
 
Upvote 0
Neutral said:
Found interesting article related to the discussed above:

Shot noise limits the camera resolution:
http://www.stanfordcomputeroptics.com/technology/dynamic-range/photon-noise.html
Here they are talking about tonal ranges that could be resolved:
"May be its hard to believe but in fact more bits in the A/D conversion can actually not increase the resolution of the data. There is only one chance to increase the resolution: go for higher intensity levels, i.e. longer time integration of the signal to reach higher signal to noise ratios"

You seem to be pretty familiar with signal processing, much more so than I am, =). I'm not an EE, but I just happen to be somewhat familiar with this issue as I ran into it about 10 years ago when doing my undergraduate dissertation where I thought we were going to get a certain number of tonal values from a 10-bit CCD camera we were using, basically based on the logic jrista is using, but was resolving a much lower number of tonal levels in the actual experiment.
 
Upvote 0
from EOSHD, which I always consider a good source on the video side of things:
http://www.eoshd.com/2015/06/interesting-insights-into-the-new-sony-a7r-ii-and-rx-sensor-technology/
That 15MP by the way is a 1:1 full pixel readout from the Super 35mm crop window of the full frame sensor. No line skipping or binning at all.

The 8K sensor mode does 2×2 binning to produce full frame 4K video. Sony say in the interview that the reason 42MP was chosen and not 50MP like on the Canon 5D S was to make for a more balanced system. 8K scales to 4K better than 8.5K and also gives better low light performance. The difference between 8K and 8.5K is minimal when it comes to resolution in stills. Sony made the right decision, Canon the wrong one
This makes sense now why they picked 42. plus the referenced interview seems clear Sony selected 42MP very carefully to deliver what should be superb 4K image quality, at least in the popular super 35 format.

KM: Yeah, 45 or 50, it's very clear for people. But actually, the engineers said "if you prefer 50, you have to lose sensitivity." And also, that number doesn't fit 4K movie at all.
[Ed. Note: That is, the pixel counts don't line up well with 4K movie pixel dimensions, to provide optimal results when downsampling.]

So we could identify the imaging sensor [characteristics]; the story came from the sensitivity, 4K-suitability. From that the number of pixels was decided. We also looked at what is the best way to create such a sensor and reversed it to get lots of light and change the [metallization] material to get faster. That's the kind of process we went through.

It seems that with Sony, the engineering team won. With canon, the marketing team won. I can't wait to see them add lossless compression 14bit RAW support, or even 12 bit lossless. Then again, that is probably what the Nikon D820 is going to do so it won't be long before we can see what the most advanced full frame sensor in the world can do.

I guess the 5Ds can be basically written off as a catchup attempt in the MP race by just blowing the old ancient sensor in full frame format, and the 5D mark 4 could be canon's real move into a versatile full frame camera that can do video to 2015 standards. Can't wait to see both the 5DMk4 and the D820.
 
Upvote 0
Mar 2, 2012
3,187
542
psolberg said:
from EOSHD, which I always consider a good source on the video side of things:
http://www.eoshd.com/2015/06/interesting-insights-into-the-new-sony-a7r-ii-and-rx-sensor-technology/
That 15MP by the way is a 1:1 full pixel readout from the Super 35mm crop window of the full frame sensor. No line skipping or binning at all.

The 8K sensor mode does 2×2 binning to produce full frame 4K video. Sony say in the interview that the reason 42MP was chosen and not 50MP like on the Canon 5D S was to make for a more balanced system. 8K scales to 4K better than 8.5K and also gives better low light performance. The difference between 8K and 8.5K is minimal when it comes to resolution in stills. Sony made the right decision, Canon the wrong one
This makes sense now why they picked 42. plus the referenced interview seems clear Sony selected 42MP very carefully to deliver what should be superb 4K image quality, at least in the popular super 35 format.

KM: Yeah, 45 or 50, it's very clear for people. But actually, the engineers said "if you prefer 50, you have to lose sensitivity." And also, that number doesn't fit 4K movie at all.
[Ed. Note: That is, the pixel counts don't line up well with 4K movie pixel dimensions, to provide optimal results when downsampling.]

So we could identify the imaging sensor [characteristics]; the story came from the sensitivity, 4K-suitability. From that the number of pixels was decided. We also looked at what is the best way to create such a sensor and reversed it to get lots of light and change the [metallization] material to get faster. That's the kind of process we went through.

It seems that with Sony, the engineering team won. With canon, the marketing team won.

How do you figure? Canon clearly wasn't engineering for 4k, nor did they produce a "clear" marketing number like 45 or 50; it's 50.6.
 
Upvote 0