Dynamic Range -- what it is, what it's good for, and how much you 'need'

Started Oct 17, 2011 | Discussions thread
FrankyM Senior Member • Posts: 2,274
Re: Let's put it all together.

Great Bustard wrote:

FrankyM wrote:

If you understood what DR was, then you'd not be confused on this point. Would you like me to explain to you how it works out? Specifically, why neither sensor size nor pixel size figures into the calculation of DR?

Maybe you should explain it to Chuck Westfall, Technical Advisor in Canon USA's Pro Engineering and Solutions Division. who says


'There's a couple of things that we consider when we think about IQ: number one on this sensor is noise. It's clear the noise level is better than in the 1D Mk IV or the 1DS III. The pixel size is larger than in the 1DS III or 5D Mark II (6.95 microns, versus 6.4) and the difference is even more striking compared to the 5.7 micron pixels in the 1D Mark IV. That helps us in terms of light capturing ability and increases the signal to noise ratio. In turn, that does nothing but help the dynamic range of the camera. '

It's easy to see that pixel size is not the player, but rather sensor size and efficiency. The pixels in the 5Dc sensor have 58% more area than the pixels in the 5D2 sensor, yet the 5D2 has less noise and more DR. Why? Quite simple -- the sensors are the same size, but the 5D2 sensor is more efficient.

It's a different technology.

So far as I'm aware, it's the same tech, just on a finer scale. It's not the pixel size or sensor size for DR, but the sensor efficiency. Pixel size does play a role, but it's a secondary role, and they way it pans out is that smaller pixels, for a given sensor size and efficiency , result in greater DR per area , but the same DR per pixel .

So, the sensor in the 1Dx will have less noise and more DR due to a more efficient sensor, which, of course, one would expect after four years since the 1Ds3 (when the 1Dx is released).

Yes, and bigger pixels too. What he's saying is that for a given technology, larger pixels have a better SNR, a higher DR and thus the camera will have a higher DR.

That's not what he's saying at all. As I highlighted, he claims better per pixel noise performance , which is true for larger pixels. But that doesn't amount to better per area noise performance , which is the proper measure for the photo.

Well read what he does say -

That helps us in terms of light capturing ability and increases the signal to noise ratio. In turn, that does nothing but help the dynamic range of the camera. '

If you're talking about sensors, pixel noise performance is what it's about. If you're talking about photos then it depends also on the processes followed after image capture. However, image SNR cannot be more than pixel SNR at the same resolution.

And, as I've said elsewhere in this thread, the DR is the number of stops from the noise floor to the saturation limit. Smaller pixels have both lower noise floors and lower saturation limits, which are in proportion to the pixel area for a given efficiency, so the DR / pixel remains unchanged.

I'm not arguing the definition of DR, merely pointing out that what you say is in disagreement with Canon's spokesman.

If you don't agree with him you might want to send him an email.

Doesn't Chuck Westfall work as a marketing representative for Canon? He would be no more interested in hearing that smaller pixels have no IQ advantage than hearing that lens IS has no advantage over sensor IS.

I don't know who Chuck Westfall is but his title is given as "Technical Advisor in Canon USA's Pro Engineering and Solutions Division" by DPR. It doesn't sound like marketing to me.

In other words, if Canon could have made a 36 MP sensor that was just as efficient as the 18 MP sensor, noise performance, at the image level , would be superior to the 18 MP sensor.

The question, of course, is why they couldn't make a 36 MP sensor just as efficient. The answer is likely not that they couldn't, but that they chose not to due operational considerations. That is, a 36 MP sensor may not have been able to keep the same frame rate as an 18 MP sensor (there is evidence that is why the D3 and D3s did not have more pixels as well).

They obviously chose not to, but what 'operational' considerations?

That would be the green text above (I put it in bold) and also the green text in the paragraph below.

Then I misunderstood you. I consider that a 'technical' rather than operational consideration (normally operational refers to the actual operation of a business - so if, for example, not enough 36mp sensors could be manufactured in time for a fixed date product launch, that would be an operational reason to go with something else).

Nevertheless, that's just conjecture on your part.

So, combine that with a marketing spin based on so few understanding the difference between a pixel level measure and an image level measure, and, voila, the 18 MP FF sensor in the 1Dx being sold as having less noise in part due to larger pixels to cover the fact that Canon couldn't get the desired frame rate with more pixels.

A nice story. But is it true?

If not a design choice , then it's likely that Canon is using a new sensor tech, and they lacked the ability to make smaller pixels with it at this time. In the next generation, they will likely shrink the scale, making more pixels with it. In fact, that has been Canon's history in sensor design for some time.

Well the history with their G series is one of increasing pixels and then decreasing them. So the question is why?

Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow