Canon Inc. Boss Wants to See More Innovation

Jul 21, 2010
31,100
12,864
angrykarl said:
neuroanatomist said:
3kramd5 said:
Even if you review two photos at sensor dimensions, i.e. no enlargement (the digital equivalent of a contact print), your noise performance will be better with the larger sensor, given the same exposure time and t-stop with a format-appropriate image circle (focal plane exposure) and the same composition. Why? You sampled the scene with more photons, thus generating more charge and requiring less amplification when digitized to a standardized output signal.

How do differently-sized image constitute a standardized output?

The image is not differently-sized. You've got the same shutter speed, t-stop, composition, resolution,
equivalent focal length, just different size of sensors. The resulting images are equal sized.

Please read the part highlighted in red above, which clearly states that images of different sizes are being compared.

He wrote "standardized output signal". Signal is electric charge from collected photons in one photosite. Which results in one pixel brightness. Raw collected signal is smaller, but according to ISO the brightness must equal, therefore there must be amplification.

We’ve already been down the rabbit hole of comparing pictures (i.e. the image from a whole sensor) versus comparing pixels. They are not the same.
 
Upvote 0
Jul 28, 2015
3,368
570
angrykarl said:
I never disputed that there is the same amount of light captured from the same area. I am simply telling that distributing the same amount of anything into more buckets means there is less in each.
And the amount of lighting each pixel (water in each bucket) is irrelevant to the image as a whole because the amount of signal being transmitted to the final processor is the same whether it is a bit of light from lots of pixels or a lot of light from a few pixels. The difference is that with a higher number of smaller pixels you are also transmitting more noise which raises the noise floor and reduces dynamic range
As neuro says, comparing what is happening at pixel level is completely different to looking at the whole image

angrykarl said:
5D3 has max native ISO 25600, 5DSR only 6400... I wonder why is that.
Because that is the limit Canon put on it. No other reason. Canon decide on behalf of the photographer what is a usable quality and will design the cameras accordingly and with the higher pixel count the noise becomes intrusive and Canon decide they will not let the ISO of the 5DSR go higher.
Why can Sony claim 3million ISo for their FF cameras? Are you saying they have a single humungous pixel? Nope. Canon could if they wanted put 3 million ISO on their camera but the image would be total shit.


angrykarl said:
So you finally agree that APS-C collect less light per pixel? And according to you the pixels are less bright, but when viewing the whole picture the downsampled pixels are more bright? What about histogram -- which shows distribution of bright and dark pixels -- does that also magically change when I zoom the picture? Or do you think histograms from these FF and APS-C images differ? And what if I create a large resolution uniform gray picture in Photoshop and downsample it, would I also get a more white picture?

I never denied the APS-C collects less light. I was pointing out that basing your argument on pixel size was flawed. It has nothing to do with pixels but the fact both sensors how long the shutter is open at a specific aperture and the light per unit area is the same but there is not as much area to collect the light.
The histogram of the whole picture never changes because the histogram shows a percentage distribution of luminosity for the whole picture so if the framing is the same the histogram is the same. If you crop the image the histogram changes for this very reason.

angrykarl said:
And would that be a problem? You want to have exactly exposed image at ISO400 as from other cameras. That's the point of ISO. How the camera handles it is just a technical detail.
You do get the same exposures - exposure of the same unit area of sensor. ISO is defined by the ability to fill the wells on the pixel, not by the output viewed on screen. This is where your reference to pixels falls down.
 
Upvote 0
Mar 2, 2012
3,187
543
neuroanatomist said:
3kramd5 said:
Even if you review two photos at sensor dimensions, i.e. no enlargement (the digital equivalent of a contact print), your noise performance will be better with the larger sensor, given the same exposure time and t-stop with a format-appropriate image circle (focal plane exposure) and the same composition. Why? You sampled the scene with more photons, thus generating more charge and requiring less amplification when digitized to a standardized output signal.

How do differently-sized image constitute a standardized output?

It’s an I/O-based designation. The expected output signal is a function of known input levels and “ISO speed”. ISO speed is typically defined including a noise basis.

Per the standard:
“numerical value calculated from the exposure provided at the focal plane of a DSC to produce specified camera output signal characteristics using the methods described in this International Standard

NOTE The ISO speed is usually the highest exposure index value that still provides peak image quality for normal scenes. However, a DSC does not necessarily use the ISO speed value as the exposure index value when capturing images.”

Unless I’m mistaken (might be), enlarging a digital file doesn’t affect its luminance (histogram stays the same) nor amount of noise present. Enlarging it more can make noise more apparent when viewed, but what is inherent to the file is baked in at capture, i.e. in camera, which is what ISO speed is concerned with. Less total light gathered -> more gain required before the analog is digitized, all else being equal.
 
Upvote 0
neuroanatomist said:
angrykarl said:
neuroanatomist said:
3kramd5 said:
Even if you review two photos at sensor dimensions, i.e. no enlargement (the digital equivalent of a contact print), your noise performance will be better with the larger sensor, given the same exposure time and t-stop with a format-appropriate image circle (focal plane exposure) and the same composition. Why? You sampled the scene with more photons, thus generating more charge and requiring less amplification when digitized to a standardized output signal.

How do differently-sized image constitute a standardized output?

The image is not differently-sized. You've got the same shutter speed, t-stop, composition, resolution,
equivalent focal length, just different size of sensors. The resulting images are equal sized.

Please read the part highlighted in red above, which clearly states that images of different sizes are being compared.

Ok. What if they are the same?
 
Upvote 0
Jul 21, 2010
31,100
12,864
3kramd5 said:
neuroanatomist said:
3kramd5 said:
Even if you review two photos at sensor dimensions, i.e. no enlargement (the digital equivalent of a contact print), your noise performance will be better with the larger sensor, given the same exposure time and t-stop with a format-appropriate image circle (focal plane exposure) and the same composition. Why? You sampled the scene with more photons, thus generating more charge and requiring less amplification when digitized to a standardized output signal.

How do differently-sized image constitute a standardized output?

It’s an I/O-based designation. The expected output signal is a function of known input levels and “ISO speed”. ISO speed is typically defined including a noise basis.

Per the standard:
“numerical value calculated from the exposure provided at the focal plane of a DSC to produce specified camera output signal characteristics using the methods described in this International Standard

NOTE The ISO speed is usually the highest exposure index value that still provides peak image quality for normal scenes. However, a DSC does not necessarily use the ISO speed value as the exposure index value when capturing images.”

Unless I’m mistaken (might be), enlarging a digital file doesn’t affect its luminance (histogram stays the same) nor amount of noise present. Enlarging it more can make noise more apparent when viewed, but what is inherent to the file is baked in at capture, i.e. in camera, which is what ISO speed is concerned with. Less total light gathered -> more gain required before the analog is digitized, all else being equal.

All else isn't equal if you're comparing images of different sizes.

That aside, you seem to be suggesting that more analog gain is applied to signals from a smaller sensor. Consider...if you capture an image with a FF and an APS-C camera at the same focal length, aperture, ISO setting, and subject distance, then crop the FF image so it matches the FoV of the APS-C camera (and output both to the same size, but since the sensor areas are equal, that doesn't matter in this case), you'll get two images that are equivalent – same framing, DoF, noise, brightness, etc. They should be the same – they represent the same physical sensor area being sampled. If more analog gain was applied to the APS-C sensor signal, why isn't the resulting image brighter or noisier than the equivalent image cropped from the FF sensor?
 
Upvote 0
Mar 2, 2012
3,187
543
neuroanatomist said:
3kramd5 said:
neuroanatomist said:
3kramd5 said:
Even if you review two photos at sensor dimensions, i.e. no enlargement (the digital equivalent of a contact print), your noise performance will be better with the larger sensor, given the same exposure time and t-stop with a format-appropriate image circle (focal plane exposure) and the same composition. Why? You sampled the scene with more photons, thus generating more charge and requiring less amplification when digitized to a standardized output signal.

How do differently-sized image constitute a standardized output?

It’s an I/O-based designation. The expected output signal is a function of known input levels and “ISO speed”. ISO speed is typically defined including a noise basis.

Per the standard:
“numerical value calculated from the exposure provided at the focal plane of a DSC to produce specified camera output signal characteristics using the methods described in this International Standard

NOTE The ISO speed is usually the highest exposure index value that still provides peak image quality for normal scenes. However, a DSC does not necessarily use the ISO speed value as the exposure index value when capturing images.”

Unless I’m mistaken (might be), enlarging a digital file doesn’t affect its luminance (histogram stays the same) nor amount of noise present. Enlarging it more can make noise more apparent when viewed, but what is inherent to the file is baked in at capture, i.e. in camera, which is what ISO speed is concerned with. Less total light gathered -> more gain required before the analog is digitized, all else being equal.

All else isn't equal if you're comparing images of different sizes.

I don’t think images of size are being compared. I think signal levels are being considered.


neuroanatomist said:
That aside, you seem to be suggesting that more analog gain is applied to signals from a smaller sensor. Consider...if you capture an image with a FF and an APS-C camera at the same focal length, aperture, ISO setting, and subject distance, then crop the FF image so it matches the FoV of the APS-C camera (and output both to the same size, but since the sensor areas are equal, that doesn't matter in this case), you'll get two images that are equivalent – same framing, DoF, noise, brightness, etc.

Is that so? I’ve never seen two sensors which are the same in every way excepting area with which to verify that noise levels are equal.

Brightness indeed would be the same as that’s implicit in the ISO speed formulation.

Again, I’m not an expert and claim to not be. I’m merely someone who has read and tried to interpret the standard. I’m happy to be educated ;)
 
Upvote 0
neuroanatomist said:
Consider...if you capture an image with a FF and an APS-C camera at the same focal length, aperture, ISO setting, and subject distance, then crop the FF image so it matches the FoV of the APS-C camera (and output both to the same size, but since the sensor areas are equal, that doesn't matter in this case), you'll get two images that are equivalent – same framing, DoF, noise, brightness, etc. They should be the same – they represent the same physical sensor area being sampled. If more analog gain was applied to the APS-C sensor signal, why isn't the resulting image brighter or noisier than the equivalent image cropped from the FF sensor?

Great. Now have both sensors the same resolution and use equivalent focal length on the crop so that you cannot blame the enlarging of image. Do they have the same noise? But you never compare this, because that wouldn't suit you.

I am leaving this hopeless discussion. Have a nice day.
 
Upvote 0
Jul 21, 2010
31,100
12,864
angrykarl said:
neuroanatomist said:
Consider...if you capture an image with a FF and an APS-C camera at the same focal length, aperture, ISO setting, and subject distance, then crop the FF image so it matches the FoV of the APS-C camera (and output both to the same size, but since the sensor areas are equal, that doesn't matter in this case), you'll get two images that are equivalent – same framing, DoF, noise, brightness, etc. They should be the same – they represent the same physical sensor area being sampled. If more analog gain was applied to the APS-C sensor signal, why isn't the resulting image brighter or noisier than the equivalent image cropped from the FF sensor?

Great. Now have both sensors the same resolution and use equivalent focal length on the crop so that you cannot blame the enlarging of image. Do they have the same noise? But you never compare this, because that wouldn't suit you.

I am leaving this hopeless discussion. Have a nice day.

You're right. If you lack the ability to comprehend, then further discussion is indeed hopeless.

neuroanatomist said:
The enlargement comes from the way we (appropriately) compare images, which is at the same output size. ... The smaller the sensor, the more you enlarge it when comparing output.

In your scenario —same framing on APS-C vs FF (i.e., different focal lengths or different subject distance), same aperture, same ISO, same output size— the APS-C image will be noisier, becuase you enlarged it more. Incidentally, the DoF of the APS-C image will be deeper...for the same reason – more enlargement from the smaller sensor.
 
Upvote 0
Jul 21, 2010
31,100
12,864
3kramd5 said:
I don’t think images of size are being compared. I think signal levels are being considered.

That's the rabbit hole of pictures vs. pixels, again. If you are talking about pixels, things like pixel diameter and well depth matter. When comparing pictures, noise is determined primarily by total light gathered. A bigger sensor gathers more light. Simple as that.


3kramd5 said:
neuroanatomist said:
That aside, you seem to be suggesting that more analog gain is applied to signals from a smaller sensor. Consider...if you capture an image with a FF and an APS-C camera at the same focal length, aperture, ISO setting, and subject distance, then crop the FF image so it matches the FoV of the APS-C camera (and output both to the same size, but since the sensor areas are equal, that doesn't matter in this case), you'll get two images that are equivalent – same framing, DoF, noise, brightness, etc.

Is that so? I’ve never seen two sensors which are the same in every way excepting area with which to verify that noise levels are equal.

Brightness indeed would be the same as that’s implicit in the ISO speed formulation.

Again, I’m not an expert and claim to not be. I’m merely someone who has read and tried to interpret the standard. I’m happy to be educated ;)

I'm not the biggest fan of DPR, but credit where credit is due, they did a reasonably good job with this.

https://www.dpreview.com/articles/5365920428/the-effect-of-pixel-and-sensor-sizes-on-noise
 
Upvote 0
Mar 2, 2012
3,187
543
neuroanatomist said:
3kramd5 said:
I don’t think images of size are being compared. I think signal levels are being considered.

That's the rabbit hole of pictures vs. pixels, again. If you are talking about pixels, things like pixel diameter and well depth matter.

Of course those things matter, but they’re behind the scenes. ISO doesn’t look at sensor size, nor pixel size, nor well capacity. It looks at how to assign an ISO speed to relate a known focal plane exposure (which has an effective f number component) to an expected signal level.


neuroanatomist said:
When comparing pictures, noise is determined primarily by total light gathered. A bigger sensor gathers more light. Simple as that.

I’m neither comparing pixels nor pictures. I’m reading a document about how ISO speed is defined, and speculating that all else being equal (same t-stop, same framing, same exposure time) since less light hits a smaller sensor, that more gain must be used to result in the appropriate signal strength.

Given a controlled input signal, how do I have to tune the electronics to result in a file with expected levels, and what ISO speed do I assign based on the characteristics?
 
Upvote 0
Jul 28, 2015
3,368
570
angrykarl said:
I am leaving this hopeless discussion. Have a nice day.

You ask a question - a genuine and very interesting question.
In that question you posit an idea and ask if that is correct.
Somewhere along the line that idea turns in your mind from an idea to a fact
People disagree with that 'fact' and refuse to agree with you
So you call the discussion hopeless

Interesting.
 
Upvote 0
Jul 21, 2010
31,100
12,864
privatebydesign said:
For anybody actually interested in understanding this stuff then I always link to this article.

http://www.josephjamesphotography.com/equivalence/

It explains everything we see when we use different sensors.

Indeed, it's an excellent reference.

From it:

Given four cameras, one with...
  • ...an mFT (4/3) sensor,
  • ...another with a 1.6x sensor,
  • ...another with a 1.5x sensor,
  • ...and another with a FF sensor...
...and...
  • ...a photo of a scene from the same position with the same focal point and the same settings (e.g. 25mm f/1.4 1/200 ISO 400) with all cameras,
  • ...the photos cropped to the same framing as the photo from the mFT (4/3) camera,
  • ...and the photos are displayed at the same size...
...then the resulting photos will be Equivalent. In addition, if...
  • ...all the sensors are equally efficient, then all the photos will also have the same noise,
  • ...the pixels are all the same size, the AA filter the same strength, and the lens is the same sharpness, then all the photos will also have the same detail,
  • ...the exact same lens is used and the sensors are of the exact same design with the exact same size pixels, AA filter, CFA, and processing...
...then the photos will not merely be Equivalent, but be identical.

Which was my point:

neuroanatomist said:
Consider...if you capture an image with a FF and an APS-C camera at the same focal length, aperture, ISO setting, and subject distance, then crop the FF image so it matches the FoV of the APS-C camera (and output both to the same size, but since the sensor areas are equal, that doesn't matter in this case), you'll get two images that are equivalent – same framing, DoF, noise, brightness, etc. They should be the same – they represent the same physical sensor area being sampled.
 
Upvote 0

Don Haines

Beware of cats with laser eyes!
Jun 4, 2012
8,246
1,939
Canada
neuroanatomist said:
privatebydesign said:
For anybody actually interested in understanding this stuff then I always link to this article.

http://www.josephjamesphotography.com/equivalence/

It explains everything we see when we use different sensors.

Indeed, it's an excellent reference.

From it:

Given four cameras, one with...
  • ...an mFT (4/3) sensor,
  • ...another with a 1.6x sensor,
  • ...another with a 1.5x sensor,
  • ...and another with a FF sensor...
...and...
  • ...a photo of a scene from the same position with the same focal point and the same settings (e.g. 25mm f/1.4 1/200 ISO 400) with all cameras,
  • ...the photos cropped to the same framing as the photo from the mFT (4/3) camera,
  • ...and the photos are displayed at the same size...
...then the resulting photos will be Equivalent. In addition, if...
  • ...all the sensors are equally efficient, then all the photos will also have the same noise,
  • ...the pixels are all the same size, the AA filter the same strength, and the lens is the same sharpness, then all the photos will also have the same detail,
  • ...the exact same lens is used and the sensors are of the exact same design with the exact same size pixels, AA filter, CFA, and processing...
...then the photos will not merely be Equivalent, but be identical.

Which was my point:

neuroanatomist said:
Consider...if you capture an image with a FF and an APS-C camera at the same focal length, aperture, ISO setting, and subject distance, then crop the FF image so it matches the FoV of the APS-C camera (and output both to the same size, but since the sensor areas are equal, that doesn't matter in this case), you'll get two images that are equivalent – same framing, DoF, noise, brightness, etc. They should be the same – they represent the same physical sensor area being sampled.
No.

You are wrong.

I use magic lenses that change brightness, fstop, and focal length when I move them between crop and FF cameras.
 
Upvote 0
Mar 2, 2012
3,187
543
I tried a little more to understand the input (had to go to CIPA rather than ISO) which can either be measured focal plane exposure or measured scene luminance. My assumption was that the test chart was shot to fill the camera frame, but that doesn’t appear to be the case. In the test setup the camera is to be placed or zoomed to take up the full area within targets, but the whole test area need not fill the camera FOV. It stands to reason then that it is evaluated at the pixel level, as neuroanatomist suggested, not based on the total light gathering area being used.
 
Upvote 0

RGF

How you relate to the issue, is the issue.
Jul 13, 2012
2,820
39
It seems to me that since the problems with the 1D M3 Canon has become much more conservative. Too bad. Their cutting edge products, 5Ds/R, are tepid. It seems they are afraid to push the envelop for fear of what? Eating their own lunch, someone is going to eat it, they don't, Nikon or Sony will? Of failing and having egg on their faces. They will need to handle the problems than they did with 1D3.
 
Upvote 0