More Big Megapixel Talk [CR1]

[Woody]

To add a little more to the pressure that Canon is having it's all in the numbers. Nikon and Sony are the only places Canon is feeling any pressure from in tech from but companies only care about these numbers.

Nikon 2011 profit: ~349 million USD
Sony 2011 profit: ~ (-584 million) USD
Canon 2011 profit: ~3177 million USD (1.17 billion of which is from consumer products)

As jrista said, Canon may or may not have the tech to beat Nikon/Sony but why should they go out and drop money on new stuff when selling the same stuff is working quite well making almost 10x as much as Nikon last year.

//all numbers come from the companies 2011 financial report. Nikon and Sony are numbers from the entire company

They are #1 in worldwide market shares for both interchangeable lens as well as fixed lens cameras. In the former category, their market share is ~ 40%, compared to Nikon's 30%; the remaining 30% is shared among Olympus/Panasonic/Sony/Pentax/Fujifilm. Clearly, they are dominating the charts.

As you said, this is also their problem. Why should they innovate and improve on anything when they are already doing so well? The only way to make them sit up and listen is to stop buying their products...

Unfortunately, I am still very enticed by their 6D...

[generalstuff]

As I new member  find  comments here  about DxOmark disappointing , just because  some of you do not understand them, and they produce a result you don't like, is no reason to accuse them of not being reliable. Maybe you should seek to understand what they measure before you damn it.

[Woody]

As I new member  find  comments here  about DxOmark disappointing , just because  some of you do not understand them, and they produce a result you don't like, is no reason to accuse them of not being reliable. Maybe you should seek to understand what they measure before you damn it.

It's not they are not understood. Their derived data are just plain wrong sometimes while their RAW data is fine.

[generalstuff]

As I new member  find  comments here  about DxOmark disappointing , just because  some of you do not understand them, and they produce a result you don't like, is no reason to accuse them of not being reliable. Maybe you should seek to understand what they measure before you damn it.

It's not they are not understood. Their derived data are just plain wrong sometimes while their RAW data is fine.

Explain

[neuroanatomist]

It's not they are not understood. Their derived data are just plain wrong sometimes while their RAW data is fine.

Explain

Well, the obvious example is the DR score on a sensor like the Nikon D800 - DxOMark reports that as 14.4 EV, while the sensor has a 14-bit analog-to-digital converter meaning a real, measured DR of 14.4 is electronically impossible for the sensor.  If you look at the raw data they actually measure, before their flawed method of normalization, it's 13.2 EV (which is, of course, very good...just not impossibly good). 

[jrista]

It's not they are not understood. Their derived data are just plain wrong sometimes while their RAW data is fine.

Explain

Well, the obvious example is the DR score on a sensor like the Nikon D800 - DxOMark reports that as 14.4 EV, while the sensor has a 14-bit analog-to-digital converter meaning a real, measured DR of 14.4 is electronically impossible for the sensor.  If you look at the raw data they actually measure, before their flawed method of normalization, it's 13.2 EV (which is, of course, very good...just not impossibly good).

This is exactly correct.

@GeneralStuff: If you are truly concerned about what the "camera" can do, you want to look at DxO's "Screen DR" results. Screen DR is what the sensor/ACD/image processor in the camera itself produces in the RAW image. Print DR, on the other hand, is what computer software produces after some specialized processing well past the point its been imported off the camera, and has little to do with the actual hardware capabilities.

[generalstuff]

Sorry, then you do not understand the read outs from Nikon and the Sony sensors , and  it is more than one  data reading.
Even the D3x with 12 bit ADC have more than 12 stops DR because of more than one data reading
Study the subject before pre mature conclusions

[Razor2012]

Guys, no matter what facts you have or what you say it's not going to help much, lol.

[neuroanatomist]

Guys, no matter what facts you have or what you say it's not going to help much, lol.

I know, I know.  But...the windmill is right there, just sticking straight up out of the field and begging to be tilted at and charged...

Sorry, then you do not understand the read outs from Nikon and the Sony sensors , and  it is more than one  data reading.
Even the D3x with 12 bit ADC have more than 12 stops DR because of more than one data reading
Study the subject before pre mature conclusions

Really, the D3x has a 12-bit ADC?  Quick, you'd better call Nobuyoshi Gokyu (President & CEO of Nikon, Inc., but I'm sure you know that) and tell him that the features page for the D3x, which states, "Fast 14-bit A/D conversion incorporated onto the image sensor," is wrong and needs to be corrected immediately, based on your thorough understanding of the read out from Nikon sensors.

So, if we can agree that Nikon is correct about their own D3x specifications, and that you, despite your extensive understanding and studying, are wrong about their D3x specifications, let's just say that the D3x has a 14-bit ADC and move on...

In that case, DxO's measurements show a very decent 12.84 EV of DR for the D3x.  Their 'Landscape Score', however, is an artifically inflated 13.65 EV - still technically possible (unlike the D800), but again, artifically inflated as a direct result of a flawed method of data analysis.

Study the subject, indeed.  Oh, puuullllleeeeeeeze.  It's almost as if I didn't have a day job that includes both analyzing quantitative image analysis data and managing a large group of scientists who do the same... 

[jrista]

...

Study the subject, indeed.  Oh, puuullllleeeeeeeze.  It's almost as if I didn't have a day job that includes both analyzing quantitative image analysis data and managing a large group of scientists who do the same... 

  *clap* *clap* *clap* *clap*

[elflord]

This is exactly correct.

@GeneralStuff: If you are truly concerned about what the "camera" can do, you want to look at DxO's "Screen DR" results. Screen DR is what the sensor/ACD/image processor in the camera itself produces in the RAW image. Print DR, on the other hand, is what computer software produces after some specialized processing well past the point its been imported off the camera, and has little to do with the actual hardware capabilities.

 "The hardware" does not consist of a single pixel. "Screen" is fine if you want to evaluate the capabilities of a single pixel.

If the cameras being compared have similar resolution, the difference should be relatively minor.  (EDIT I mean both cameras should see their "dynamic range" change by a similar margin, which is the case with the D800 vs 5DIII for example -- depending on whether you believe the "screen" or "print" number, the Nikon wins by 2.5 stops, or "only" 2.2 stops.)

[elflord]

It's not they are not understood. Their derived data are just plain wrong sometimes while their RAW data is fine.

Explain

Well, the obvious example is the DR score on a sensor like the Nikon D800 - DxOMark reports that as 14.4 EV, while the sensor has a 14-bit analog-to-digital converter meaning a real, measured DR of 14.4 is electronically impossible for the sensor.  If you look at the raw data they actually measure, before their flawed method of normalization, it's 13.2 EV (which is, of course, very good...just not impossibly good).

Why is 14 a hard limit ? I understand that it's impossible to represent more than 2^14 different intensities but that's not what dynamic range is. DR is log2(saturation point) - log2( blackpoint). Why can't this be greater than the number of bits in the ADC converter ? .

[neuroanatomist]

Why is 14 a hard limit ? I understand that it's impossible to represent more than 2^14 different intensities but that's not what dynamic range is. DR is log2(saturation point) - log2( blackpoint). Why can't this be greater than the number of bits in the ADC converter ?

It could with a nonlinear ADC, except that almost all IC-based ADCs are linear.  So, while the analog DR is the delta between the full well capacity and the noise floor (in e^-), a 14-bit ADC maps signal at the noise floor to 0 and signal at full well capacity to 16,383, binning intermediate e^- values incrementally, subject to quantization error.

[elflord]

Why is 14 a hard limit ? I understand that it's impossible to represent more than 2^14 different intensities but that's not what dynamic range is. DR is log2(saturation point) - log2( blackpoint). Why can't this be greater than the number of bits in the ADC converter ?

It could with a nonlinear ADC, except that almost all IC-based ADCs are linear.  So, while the analog DR is the delta between the full well capacity and the noise floor (in e^-), a 14-bit ADC maps signal at the noise floor to 0 and signal at full well capacity to 16,383, binning intermediate e^- values incrementally, subject to quantization error.

You have the response (possibly nonlinear of the ADC). What about the response of the sensor itself to light ? Must this always be exactly linear ?

Also, if I pool four adjacent signals into one supersize pixel, how many bits do I have in my new "superpixel" ? Do I not have 56 bits ?

[generalstuff]

Guys, no matter what facts you have or what you say it's not going to help much, lol.

Well, in this case they must blaime theirs own ignorance regarding multipel read outs and not
DXO meassurements.
Go to sensorgen, there you have all data ,FWC and read out noise