June 21, 2018, 03:10:21 AM

Author Topic: EOS 7D Mark II Spec List Surfaces [CR1]  (Read 127681 times)

Pi

  • EOS 5DS R
  • ******
  • Posts: 912
    • Math and Photography
Re: EOS 7D Mark II Spec List Surfaces [CR1]
« Reply #210 on: October 30, 2013, 07:46:08 PM »
Full well capacity is really more a function of pixel area than pixel Q.E. Area is the defining factor when it comes to how much charge a pixel can hold. Increasing Q.E. might improve sensitivity, which is the rate at which photons convert to charge, but it doesn't do anything to increase the maximum charge (FWC) a pixel can contain. That is why cameras with larger sensors generally perform better than smaller sensors...they have historically had larger pixels.

No, larger sensors perform better (for a given exposure) because they are, well, larger. Pixel size is a secondary, mostly irrelevant factor.

canon rumors FORUM

Re: EOS 7D Mark II Spec List Surfaces [CR1]
« Reply #210 on: October 30, 2013, 07:46:08 PM »

Don Haines

  • Canon EF 800mm f/5.6L IS
  • ***********
  • Posts: 6859
  • posting cat pictures on the internet since 1986
Re: EOS 7D Mark II Spec List Surfaces [CR1]
« Reply #211 on: October 30, 2013, 07:59:01 PM »
Full well capacity is really more a function of pixel area than pixel Q.E. Area is the defining factor when it comes to how much charge a pixel can hold. Increasing Q.E. might improve sensitivity, which is the rate at which photons convert to charge, but it doesn't do anything to increase the maximum charge (FWC) a pixel can contain. That is why cameras with larger sensors generally perform better than smaller sensors...they have historically had larger pixels.

No, larger sensors perform better (for a given exposure) because they are, well, larger. Pixel size is a secondary, mostly irrelevant factor.
Fairly simple reason for this..... assuming the same technology for reading the charge, if you make the photosite with twice the area, it will collect twice the charge, and the accuracy of the reading will be twice as accurate.....but this is only true until saturation is reached.... and that's why FF is so much better in poor light with less of an advantage in good light.
The best camera is the one in your hands

Don Haines

  • Canon EF 800mm f/5.6L IS
  • ***********
  • Posts: 6859
  • posting cat pictures on the internet since 1986
Re: EOS 7D Mark II Spec List Surfaces [CR1]
« Reply #212 on: October 30, 2013, 08:12:25 PM »
If all the activity here is any indication of the pent-up demand for a 7DII, it is going to fly off the shelves when it gets introduced.... Come on Canon.... introduce it now! A 7DII would fit into my Christmas stocking :)
The best camera is the one in your hands

Pi

  • EOS 5DS R
  • ******
  • Posts: 912
    • Math and Photography
Re: EOS 7D Mark II Spec List Surfaces [CR1]
« Reply #213 on: October 30, 2013, 08:15:17 PM »
Fairly simple reason for this..... assuming the same technology for reading the charge, if you make the photosite with twice the area, it will collect twice the charge, and the accuracy of the reading will be twice as accurate.....but this is only true until saturation is reached.... and that's why FF is so much better in poor light with less of an advantage in good light.

Wrong, the D800 is as good as the D600. Pixel size is an almost irrelevant factor, sensor size is the main factor (for the same exposure). You are confusing what happens at pixel level with what happens at image level.

Don Haines

  • Canon EF 800mm f/5.6L IS
  • ***********
  • Posts: 6859
  • posting cat pictures on the internet since 1986
Re: EOS 7D Mark II Spec List Surfaces [CR1]
« Reply #214 on: October 30, 2013, 08:47:43 PM »
Fairly simple reason for this..... assuming the same technology for reading the charge, if you make the photosite with twice the area, it will collect twice the charge, and the accuracy of the reading will be twice as accurate.....but this is only true until saturation is reached.... and that's why FF is so much better in poor light with less of an advantage in good light.

Wrong, the D800 is as good as the D600. Pixel size is an almost irrelevant factor, sensor size is the main factor (for the same exposure). You are confusing what happens at pixel level with what happens at image level.

Could you explain so that I can understand your logic? I am not being sarcastic or trolling..... just looking for the explanation so that I can understand....
The best camera is the one in your hands

Pi

  • EOS 5DS R
  • ******
  • Posts: 912
    • Math and Photography
Re: EOS 7D Mark II Spec List Surfaces [CR1]
« Reply #215 on: October 30, 2013, 09:02:05 PM »
Fairly simple reason for this..... assuming the same technology for reading the charge, if you make the photosite with twice the area, it will collect twice the charge, and the accuracy of the reading will be twice as accurate.....but this is only true until saturation is reached.... and that's why FF is so much better in poor light with less of an advantage in good light.

Wrong, the D800 is as good as the D600. Pixel size is an almost irrelevant factor, sensor size is the main factor (for the same exposure). You are confusing what happens at pixel level with what happens at image level.

Could you explain so that I can understand your logic? I am not being sarcastic or trolling..... just looking for the explanation so that I can understand....

A pixel twice as large will have double the well capacity but it has to collect double the light. No gain, no loss. As far as accuracy goes (read noise), it tends to get higher with the maximal charge (see sensorgen) so it all depends on the technology. The photon noise is part of the image an no trickery can reduce it.

You can just forget about the pixels. You have a sensor of a given size, illuminated somehow. If the full well capacity depends mainly on the area of the pixel, then the total capacity of the sensor remains constant. In any case, this is relevant at base ISO only. It does not matter much how you split the sensor into pixels in general - smaller pixels just record more information but with a very noisy image, there is no much information beyond some point anyway.

Don Haines

  • Canon EF 800mm f/5.6L IS
  • ***********
  • Posts: 6859
  • posting cat pictures on the internet since 1986
Re: EOS 7D Mark II Spec List Surfaces [CR1]
« Reply #216 on: October 30, 2013, 09:10:18 PM »
Fairly simple reason for this..... assuming the same technology for reading the charge, if you make the photosite with twice the area, it will collect twice the charge, and the accuracy of the reading will be twice as accurate.....but this is only true until saturation is reached.... and that's why FF is so much better in poor light with less of an advantage in good light.

Wrong, the D800 is as good as the D600. Pixel size is an almost irrelevant factor, sensor size is the main factor (for the same exposure). You are confusing what happens at pixel level with what happens at image level.

Here is my logic......
Let's say that you have two cameras, one with pixels twice the size of the other. We assume a 16 bit a/d converter. We assume that all the circuitry is of the same technology.

You take a picture of the same scene with both cameras. In the first camera there are 0 units of charge in a dark pixel and 500 units of charge in a bright white pixel... Gain is applied (multiply by 128, either digital or analog) and you end up with a 16 bit number where the last 8 bits are noise.
In the second camera there are 0 units of charge in a dark pixel and 1000 units of charge in a bright white pixel... Gain is applied (multiply by 64, either digital or analog) and you end up with a 16 bit number where the last 7 bits are noise.

The best camera is the one in your hands

canon rumors FORUM

Re: EOS 7D Mark II Spec List Surfaces [CR1]
« Reply #216 on: October 30, 2013, 09:10:18 PM »

Pi

  • EOS 5DS R
  • ******
  • Posts: 912
    • Math and Photography
Re: EOS 7D Mark II Spec List Surfaces [CR1]
« Reply #217 on: October 30, 2013, 09:23:14 PM »
Fairly simple reason for this..... assuming the same technology for reading the charge, if you make the photosite with twice the area, it will collect twice the charge, and the accuracy of the reading will be twice as accurate.....but this is only true until saturation is reached.... and that's why FF is so much better in poor light with less of an advantage in good light.

Wrong, the D800 is as good as the D600. Pixel size is an almost irrelevant factor, sensor size is the main factor (for the same exposure). You are confusing what happens at pixel level with what happens at image level.

Here is my logic......
Let's say that you have two cameras, one with pixels twice the size of the other. We assume a 16 bit a/d converter. We assume that all the circuitry is of the same technology.

You take a picture of the same scene with both cameras. In the first camera there are 0 units of charge in a dark pixel and 500 units of charge in a bright white pixel... Gain is applied (multiply by 128, either digital or analog) and you end up with a 16 bit number where the last 8 bits are noise.
In the second camera there are 0 units of charge in a dark pixel and 1000 units of charge in a bright white pixel... Gain is applied (multiply by 64, either digital or analog) and you end up with a 16 bit number where the last 7 bits are noise.

That is noise per pixel, not per area. The signal in the first case will have weight 1/2 of that in the second in the final image. So you multiply by 128 but when you resize to the final output image, you divide that by 2.

jrista

  • Canon EF 400mm f/2.8L IS II
  • *********
  • Posts: 5319
  • EOL
    • Nature Photography
Re: EOS 7D Mark II Spec List Surfaces [CR1]
« Reply #218 on: October 30, 2013, 09:52:23 PM »
Fairly simple reason for this..... assuming the same technology for reading the charge, if you make the photosite with twice the area, it will collect twice the charge, and the accuracy of the reading will be twice as accurate.....but this is only true until saturation is reached.... and that's why FF is so much better in poor light with less of an advantage in good light.

Wrong, the D800 is as good as the D600. Pixel size is an almost irrelevant factor, sensor size is the main factor (for the same exposure). You are confusing what happens at pixel level with what happens at image level.

Here is my logic......
Let's say that you have two cameras, one with pixels twice the size of the other. We assume a 16 bit a/d converter. We assume that all the circuitry is of the same technology.

You take a picture of the same scene with both cameras. In the first camera there are 0 units of charge in a dark pixel and 500 units of charge in a bright white pixel... Gain is applied (multiply by 128, either digital or analog) and you end up with a 16 bit number where the last 8 bits are noise.
In the second camera there are 0 units of charge in a dark pixel and 1000 units of charge in a bright white pixel... Gain is applied (multiply by 64, either digital or analog) and you end up with a 16 bit number where the last 7 bits are noise.

That is noise per pixel, not per area. The signal in the first case will have weight 1/2 of that in the second in the final image. So you multiply by 128 but when you resize to the final output image, you divide that by 2.

Assuming you resize.

@Don: Pi is referring to Equivalence. He has brought it up in prior threads. The source of the theory is actually very good. Read this: http://josephjamesphotography.com/equivalence/index.htm. It might clear some things up.

Pi

  • EOS 5DS R
  • ******
  • Posts: 912
    • Math and Photography
Re: EOS 7D Mark II Spec List Surfaces [CR1]
« Reply #219 on: October 30, 2013, 09:59:11 PM »
Assuming you resize.

From 24x36mm? I always do.

This has nothing to do with equivalence. It is enough to understand the fallacy for a sensor of a fixed size.

jrista

  • Canon EF 400mm f/2.8L IS II
  • *********
  • Posts: 5319
  • EOL
    • Nature Photography
Re: EOS 7D Mark II Spec List Surfaces [CR1]
« Reply #220 on: October 30, 2013, 10:05:34 PM »
Full well capacity is really more a function of pixel area than pixel Q.E. Area is the defining factor when it comes to how much charge a pixel can hold. Increasing Q.E. might improve sensitivity, which is the rate at which photons convert to charge, but it doesn't do anything to increase the maximum charge (FWC) a pixel can contain. That is why cameras with larger sensors generally perform better than smaller sensors...they have historically had larger pixels.

No, larger sensors perform better (for a given exposure) because they are, well, larger. Pixel size is a secondary, mostly irrelevant factor.

Correct. I did read, and still remember, the article on equivalence. So, correct, assuming you are always using the full sensor area. There are situations where you don't, say cropping for detail. In which case, the part of my answer you did not quote applies...it's a tradeoff. I am not about to say pixel size is the most important factor, however I would say it is far from a "mostly irrelevant factor." It all just depends, and you make the various tradeoffs you have to in order to get what you need so service your own personal photography best.

In the end, one could often make the argument that neither pixel size nor sensor size really matters in the end anyway, as a more significant factor in final image quality is how you process, so long as the image output size is smaller than the input size (which is by and large the most common case these days with publishing occurring on the internet far more than it occurs in print in any form).

Pi

  • EOS 5DS R
  • ******
  • Posts: 912
    • Math and Photography
Re: EOS 7D Mark II Spec List Surfaces [CR1]
« Reply #221 on: October 30, 2013, 10:18:34 PM »
I am not about to say pixel size is the most important factor, however I would say it is far from a "mostly irrelevant factor."

Let us not take this out of the context in which it was said. We were talking about noise and the large pixel fallacy. I also said that smaller pixels collect more information, that there are tech related factors, etc.

jrista

  • Canon EF 400mm f/2.8L IS II
  • *********
  • Posts: 5319
  • EOL
    • Nature Photography
Re: EOS 7D Mark II Spec List Surfaces [CR1]
« Reply #222 on: October 30, 2013, 10:24:22 PM »
I am not about to say pixel size is the most important factor, however I would say it is far from a "mostly irrelevant factor."

Let us not take this out of the context in which it was said. We were talking about noise and the large pixel fallacy. I also said that smaller pixels collect more information, that there are tech related factors, etc.

Ok, fair enough.

canon rumors FORUM

Re: EOS 7D Mark II Spec List Surfaces [CR1]
« Reply #222 on: October 30, 2013, 10:24:22 PM »

Krob78

  • EOS 5DS R
  • ******
  • Posts: 1444
  • When in Doubt, Press the Shutter...
Re: EOS 7D Mark II Spec List Surfaces [CR1]
« Reply #223 on: October 31, 2013, 11:27:51 AM »
I am not about to say pixel size is the most important factor, however I would say it is far from a "mostly irrelevant factor."

Let us not take this out of the context in which it was said. We were talking about noise and the large pixel fallacy. I also said that smaller pixels collect more information, that there are tech related factors, etc.

Ok, fair enough.
Makes sense...
Ken

5D MK III, EOS M3, 100-400L II, 11-24L, 70-200 2.8L II, 24-70LIS, 17-40L, 16-35 4L, 600 EX-RT, 430EX-RT, 1.4X III, 2.0X III

pvk

  • PowerShot SX60 HS
  • **
  • Posts: 5
Re: EOS 7D Mark II Spec List Surfaces [CR1]
« Reply #224 on: October 31, 2013, 02:39:09 PM »
Lots of discussion on whether the 7DII will be better than the 1DX or 5DIII. It shows that Canon will have a formidable marketing job to do, but I think it misses the point. The 7DII may have some better features, but as a crop camera, it will never match the results of a top of the line FF camera.

I assume Canon has long decided that the main stream trend will  be FF. Maybe that is why they took so long to replace the 7D.

But there are cases where you need the crop to get close enough. That is where the 7DII comes in. As a top of the line 1.6 crop camera. Now for this to work, the image quality and the ISO results must be much better than the current 7D. The price must be below the 5DIII. No need to have the weather sealable body like the 1DX, because a press photographer will not buy it any way. Make it like the 5DIII, basically making the 7DII a crop version of the 5DIII. Add some newer features that you would expect after launching it two years after the 5DIII.

canon rumors FORUM

Re: EOS 7D Mark II Spec List Surfaces [CR1]
« Reply #224 on: October 31, 2013, 02:39:09 PM »