Pages: 1 ... 4 5 [6] 7 8 ... 10   Go Down

Author Topic: larger sensors  (Read 191407 times)

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
larger sensors
« Reply #100 on: January 03, 2007, 11:00:27 pm »

Quote
Simply put, with a higher resolution, your captured blur will more closely resemble the analog blur.  A lower resolution would confuse things further, either a tiny amount or a lot.
[a href=\"index.php?act=findpost&pid=93568\"][{POST_SNAPBACK}][/a]

I agree with John that there's no advantage in lower resolution. Four small pixels covering the same area as one large pixel, could ideally produce the same (or close to) dynamic range, at the same resolution or print size, as the one large pixel. But the four smaller pixels have the advantage, lens quality permitting, of producing higher resolution, with good light conditions.
Logged

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3387
larger sensors
« Reply #101 on: January 04, 2007, 07:05:00 am »

Quote
Of course, it almost goes without saying, if you want increased dynamic range in a system that is mostly limited by photonic noise, it has to be through increased exposure. You can't have increased dynamic range as well as faster shutter speeds. No matter how many pixels are on the sensor, the sensor as a whole receives the same amount of light for a given exposure at a given f stop.

[{POST_SNAPBACK}][/a]

No, this is a common misconception and is explained at length by Roger Clark on his website:

[a href=\"http://www.clarkvision.com/photoinfo/f-ratio_myth/index.html]http://www.clarkvision.com/photoinfo/f-ratio_myth/index.html[/url]

Other things being equal, for a given f/stop and exposure time, the number of photons per second per square micron arriving in the focal plane will be the same, but the  camera with larger pixels will collect more photons because of its larger area.

Bill
Logged

John Sheehy

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 838
larger sensors
« Reply #102 on: January 04, 2007, 09:36:01 am »

Quote
Why don't you read Roger Clark's article?
[a href=\"index.php?act=findpost&pid=93399\"][{POST_SNAPBACK}][/a]

I don't think anyone is contesting that the statistics for shot noise favor bigger pixels, at the pixel level.  What your interpretation of Roger's data doesn't consider is the fact that the pixel does not determine image quality.  *All* the pixels in the image, collectively, do.  Noise at a microscopic level, no matter how intense, averages out to no noise, if you're not looking through a microscope.  Noise statistics, as you are considering them, are only measurements in the z axis, and ignore the x and y axii.
Logged

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3387
larger sensors
« Reply #103 on: January 04, 2007, 11:02:19 am »

Quote
I don't think anyone is contesting that the statistics for shot noise favor bigger pixels, at the pixel level.  What your interpretation of Roger's data doesn't consider is the fact that the pixel does not determine image quality.  *All* the pixels in the image, collectively, do.  Noise at a microscopic level, no matter how intense, averages out to no noise, if you're not looking through a microscope.  Noise statistics, as you are considering them, are only measurements in the z axis, and ignore the x and y axii.
[{POST_SNAPBACK}][/a]

By your reasoning the 8MP EOS 1D (8.2 um pixels) and the 7.1MP S70 (2.3 um) should have similar noise characteristics. They do is you are shooting at ISO 100 with the S70 and ISO 1600 with the 1DM2 as [a href=\"http://www.clarkvision.com/imagedetail//does.pixel.size.matter2/]Roger[/url] shows.

Of course, if you took a 1DM2 sensor and increased the pixel count such that the pixel size were 2.3 um, the noise characteristics might be similar at the same print size. However, the primary reason for more MP is to have the ability to print at larger sizes.

Bill
Logged

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
larger sensors
« Reply #104 on: January 04, 2007, 05:08:13 pm »

Quote
By your reasoning the 8MP EOS 1D (8.2 um pixels) and the 7.1MP S70 (2.3 um) should have similar noise characteristics. They do is you are shooting at ISO 100 with the S70 and ISO 1600 with the 1DM2 as Roger shows.

Of course, if you took a 1DM2 sensor and increased the pixel count such that the pixel size were 2.3 um, the noise characteristics might be similar at the same print size. However, the primary reason for more MP is to have the ability to print at larger sizes.

Bill
[a href=\"index.php?act=findpost&pid=93645\"][{POST_SNAPBACK}][/a]
 

If you were to take the very same 2.3 um pixels of the S70 and spread them over a 1D2 sensor, you might well have similar noise characteristics at same size prints, but any professional level camera that attempted such a high density, large sensor would contain a lot more advanced technology than the S70. It might then be reasonable to predict that noise levels in much larger prints would also be less than (or at least equal to) that from the 8 um camera.
Logged

John Sheehy

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 838
larger sensors
« Reply #105 on: January 04, 2007, 05:25:47 pm »

Quote
If you were to take the very same 2.3 um pixels of the S70 and spread them over a 1D2 sensor, you might well have similar noise characteristics at same size prints, but any professional level camera that attempted such a high density, large sensor would contain a lot more advanced technology than the S70. It might then be reasonable to predict that noise levels in much larger prints would also be less than (or at least equal to) that from the 8 um camera.
[a href=\"index.php?act=findpost&pid=93738\"][{POST_SNAPBACK}][/a]
For shot noise, there can't be any improvement unless quantum efficiency improves.  For read noise, however, all you have to do is have it increase less than the linear resolution does, and it should be less powerful in the image.  IOW, if you squeeze 9x as many pixels into the sensor, as long as the read noise does not increase by 3x, there should be less read noise at the image level.
Logged

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
larger sensors
« Reply #106 on: January 04, 2007, 06:01:59 pm »

Quote
No, this is a common misconception and is explained at length by Roger Clark on his website:

http://www.clarkvision.com/photoinfo/f-ratio_myth/index.html

Other things being equal, for a given f/stop and exposure time, the number of photons per second per square micron arriving in the focal plane will be the same, but the  camera with larger pixels will collect more photons because of its larger area.

Bill
[a href=\"index.php?act=findpost&pid=93609\"][{POST_SNAPBACK}][/a]


I knew it! I take it you didn't see my edit, Bill. I added the last comment for the very specific purpose of avoiding this red herring. I thought I'd made it clear that  I was talking about same size sensors. Clearly a small sensor, with lens at f8, does not receive as much light as a large sensor, with lens at f8 using the same shutter speed. How could it? You don't need to refer me to Roger Clark to get that point.

Here's what I wrote a couple of pages back.

Quote
Of course, it almost goes without saying, if you want increased dynamic range in a system that is mostly limited by photonic noise, it has to be through increased exposure. You can't have increased dynamic range as well as faster shutter speeds. No matter how many pixels are on the sensor, the sensor as a whole receives the same amount of light for a given exposure at a given f stop.

I'll have to edit this in case someone tries to argue that a 200mm lens at f8 lets more light pass for a given exposure than a 50mm lens at f8. I am of course referring to a situation of equal size sensors, ie. equal formats.


This post has been edited by Ray: Yesterday, 12:11 AM
Logged

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
larger sensors
« Reply #107 on: January 04, 2007, 06:31:17 pm »

Quote
For shot noise, there can't be any improvement unless quantum efficiency improves.  For read noise, however, all you have to do is have it increase less than the linear resolution does, and it should be less powerful in the image.  IOW, if you squeeze 9x as many pixels into the sensor, as long as the read noise does not increase by 3x, there should be less read noise at the image level.
[a href=\"index.php?act=findpost&pid=93742\"][{POST_SNAPBACK}][/a]

That makes sense. There are probably lots ways of making incremental improvements, whether it be in relation to quantum efficiency or read noise, which are currently being explored in the laboratories.

I wouldn't pretend to understand what's possible and what's not. What I might see as an insurmountable obstacle might be no obstacle at all. For example, my simplistic way of viewing pixels (or photodiodes) as 3-dimensional buckets that hold electrons instead of water, might lead me to suppose that 4 small pixels covering the same area as 1 large pixel could not hold the same charge unless each small pixel had the same depth. If the small pixels do have the same depth as the singe large pixel, is this an advantage or disadvantage? Perhaps there's an advantage in having a greater wall area in total, that's exposed to photons.
Logged

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3387
larger sensors
« Reply #108 on: January 04, 2007, 06:56:38 pm »

Quote
For shot noise, there can't be any improvement unless quantum efficiency improves.  For read noise, however, all you have to do is have it increase less than the linear resolution does, and it should be less powerful in the image.  IOW, if you squeeze 9x as many pixels into the sensor, as long as the read noise does not increase by 3x, there should be less read noise at the image level.
[a href=\"index.php?act=findpost&pid=93742\"][{POST_SNAPBACK}][/a]

Well, current Canon sensors have a read noise of about 3-4 electrons (and it does not vary that much with pixel size), and there is not a whole lot of room for improvement. Because of the lower gain associated with small pixels, the effect of read noise is much more prominent with small pixels. IOW, if the gain with a large pixel is 9 electrons per ADU, a noise of one electron will only cause 1/9 ADU of noise. However, if your small pixel has a gain of 1 electron per ADU, 1 electron of noise would cause a change of 1 ADU in the noise. Quantum efficiency could double or treble, but how would you increase the pixel density nine fold without significantly decreasing the gain. The capacitance of silicon is limited and well depth can increase only so much.

Bill
« Last Edit: January 06, 2007, 12:54:37 pm by bjanes »
Logged

howiesmith

  • Guest
larger sensors
« Reply #109 on: January 04, 2007, 07:52:54 pm »

Quote
Ultimately, Howard, it's up to each of us to decide what makes sense, what's meaningful, what's useful, what's right or wrong. How we decide such matters is determined by our entire world experience, our education, our genetics, our environmnet and everything that has happened to us from the time, and including the time, that we were in the womb.

[a href=\"index.php?act=findpost&pid=93551\"][{POST_SNAPBACK}][/a]

There is an interesting thing about truth Ray.  It doesn't matter one iota whether you believe it or not.  What is true is true, regardless whether it makes sense to you, is meaningful to you, or useful to you.  Or anyone for that matter.  There are likely many truths today that are being overlooked because they lack meaning or use now.  At Newton's time, relativity existed and was as true as it is today.  It just wasn't useful or meaningful.  (Who really cared then that as you approach the speed of light, your mass increases?)  How you decide such matters is determined by your entire experience, education, genetics, environmnet and everything that has happened to you.   However, that does not alter the trueth - just your perception of it.

Scientific method allows very little room for being optimistic or pessimistic.  The data must support your conclusion(s).  That is the purpose of peer review - to assure you are not too optimistic about what you are testing.  I would guess you are not a peer of any of those doing cutting edge physics.  Am I right?
Logged

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
larger sensors
« Reply #110 on: January 04, 2007, 10:08:36 pm »

Quote
There is an interesting thing about truth Ray.  It doesn't matter one iota whether you believe it or not.  What is true is true, regardless whether it makes sense to you, is meaningful to you, or useful to you.  Or anyone for that matter. 

Well, that could be a very long philosophical debate, Howie. You are describing what you apparently think is a truth about truth. My understanding is, the belief that something is true can have a profound effect on any individual, whether or not it is true in reality. It's sometimes called the placebo effect. Such beliefs can be beneficial or harmful.

I don't think it is reasonable to expect people to concern themselves with matters they consider meaningless and useless to themselves and others. But, if the point you are making is that, what one person thinks is a meaningless or useless activity, another person might not, then of course I agree.

Quote
At Newton's time, relativity existed and was as true as it is today.  It just wasn't useful or meaningful.

I don't agree. Newton would have found Einstein's theories of relativity very meaningful and very useful. He just wasn't able to think of them because the groundwork in other areas of science and mathematics had not been laid.

Quote
Scientific method allows very little room for being optimistic or pessimistic.

Little room for being optimistic or pessimistic about what? You're not making much sense. If a scientist is pessimistic about the efficacy of the scientific method, then maybe he/she should be doing something else, or improve the method if he/she can.

Quote
I would guess you are not a peer of any of those doing cutting edge physics. 

That's right. Are you?
Logged

howiesmith

  • Guest
larger sensors
« Reply #111 on: January 05, 2007, 04:26:57 am »

Quote
Well, that could be a very long philosophical debate, Howie. You are describing what you apparently think is a truth about truth. My understanding is, the belief that something is true can have a profound effect on any individual, whether or not it is true in reality. It's sometimes called the placebo effect. Such beliefs can be beneficial or harmful.

I don't think it is reasonable to expect people to concern themselves with matters they consider meaningless and useless to themselves and others. But, if the point you are making is that, what one person thinks is a meaningless or useless activity, another person might not, then of course I agree.

I don't agree. Newton would have found Einstein's theories of relativity very meaningful and very useful. He just wasn't able to think of them because the groundwork in other areas of science and mathematics had not been laid.

Little room for being optimistic or pessimistic about what? You're not making much sense. If a scientist is pessimistic about the efficacy of the scientific method, then maybe he/she should be doing something else, or improve the method if he/she can.

That's right. Are you?
[a href=\"index.php?act=findpost&pid=93783\"][{POST_SNAPBACK}][/a]

I have never claimed the belief that something is true could not have a profound effect on an individual, whether or not it is true in reality.  If you take a sugar pill (placebo), believing it is aspirin, and it cures your headache, that untruth has had a profound effect, enough to cure a headache.  That does not change the reality (truth) that the pill was in fact sugar and not aspirin, no matter how firmly you believed, thought or wished it were aspirin.

I guess we will just disagree about whether Newton would have found relativity very meaningful and very useful.  I think he may have found relativity interesting.  Even today, a very many people do not find relativity meaningful or useful.  But relativity applies to savages as well as physicists.  Relativity didn't spring to life from Einstein's pen.  He merely wrote down his ideas.  He did not change the reality (truth) of how the universe works, just did a better job of describing it.

I was not expressing an opinion about being optimistic or pessimistic about scientific method. merely that the scientific method does not allow a person to insert much optimism or pessimism into the process.  Just interpret the data.  A person can be only as optimistic about a result of the scientific method as the data will support.  Scientific methods might be used to support the placebo effect dicussed above.  Expanding those results to conclude sugar is as effective as aspirin for curing headaches may be a bit too optimistic.

I have performed peer reviews but I am retired now, and no longer involved in that.
Logged

John Sheehy

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 838
larger sensors
« Reply #112 on: January 05, 2007, 08:21:46 am »

Quote
For example, my simplistic way of viewing pixels (or photodiodes) as 3-dimensional buckets that hold electrons instead of water, might lead me to suppose that 4 small pixels covering the same area as 1 large pixel could not hold the same charge unless each small pixel had the same depth. If the small pixels do have the same depth as the singe large pixel, is this an advantage or disadvantage? Perhaps there's an advantage in having a greater wall area in total, that's exposed to photons.
[a href=\"index.php?act=findpost&pid=93760\"][{POST_SNAPBACK}][/a]
I was under the impression that the wells are actually so thin that even the ones in 10MP P&S cameras were wider than they are high, but I don't know that for fact.  It would be nice to have an accurate chart of various senssors, their pixel dimensions, fill factors, QEs, and electron capacities, to see what is and isn't independent in current technology.
Logged

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
larger sensors
« Reply #113 on: January 05, 2007, 08:48:33 am »

Quote
I guess we will just disagree about whether Newton would have found relativity very meaningful and very useful.  I think he may have found relativity interesting.

Howie,
I think he would have found it profoundly interesting. A revelation, in fact; particularly the notion that the universe is expanding. The one great enigma in Newton's cosmology, where every body exterts a gravitational force on every other body, is that there was no satisfactory explanation as to why these celestial objects did not eventually crash in on each other. Just a small perturbance would upset the clockwork balance to cause everything to come tumbling down. But it didn't and doesn't. I imagine this would have worried Newton deeply. He was basically stumped and without an explanation.
Logged

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
larger sensors
« Reply #114 on: January 05, 2007, 09:58:03 am »

Quote
I was under the impression that the wells are actually so thin that even the ones in 10MP P&S cameras were wider than they are high, but I don't know that for fact.  It would be nice to have an accurate chart of various senssors, their pixel dimensions, fill factors, QEs, and electron capacities, to see what is and isn't independent in current technology.
[{POST_SNAPBACK}][/a]

John,
Having just done a search on the anatomy of a CMOS imaging device, I came across the following site which does appear to show in its diagrams that the collection area is much wider than it is deep. The article gives a fairly thorough treatment of the processes, but I notice at the foot of the article that it was last modified in July 2004.

The fact that the photodiodes (and photon collection areas) are considerably smaller than the pixel pitch is a limiting factor on quantum efficiency. I expect we could look forward to some improvement there.

[a href=\"http://micro.magnet.fsu.edu/primer/digitalimaging/cmosimagesensors.html]http://micro.magnet.fsu.edu/primer/digital...agesensors.html[/url]
Logged

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6600
larger sensors
« Reply #115 on: January 05, 2007, 02:12:54 pm »

How about 33x26mm? At least in the "1D" family, as opposed to the higher priced larger format 1Ds family.

Those weird numbers come from what Canon has indicated is the largest sensor size that can currently be made with standard single exposure fabrication, and so at distinctly lower cost that the multiple exposures that Canon says it needs to use to make 36x24mm format. 36x36 is too long in either direction, needing at least four exposures with a 33x26 single exposure limit, though Kodak keeps making that format.

The extra 2mm of height would probably require a new mirror and viewfinder assembly (or the radical change to EVF) but the frame fits the 35mm format image circle, so current lenses should work fine (except for a possible slight crop with some super-telephotos, if any actually has a tight 36x24 rectangular anti-flare baffle.)

Or if the extra 2mm of height costs too much in redesign, why not 33x24mm, pushing the twin current limits of camera components designed for 36x24 and fab. equipment limited to 33x26?

When choosing a sensor size that imposes a crop on lenses designed for 36x24, as in the 1D models, I see no good reason to impose an additional vertical crop just for the sake of staying with 3:2 shape. If 33x26 is the current fab. limit, 3:2 shape imposes 33x22, but you can get that just as well by cropping from 33x24 or 33x26, and the latter shapes offer a larger frame when less elongated print shapes are desired, like 10x8, 11x8.5, A4, A3, A2 ... (33x26 is almost the same shape as 10x8, 33x24 is close to the ISO A paper shapes and 7x5.)
Logged

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6600
larger sensors
« Reply #116 on: January 05, 2007, 02:29:03 pm »

Quote
the Canon 1D M2 outsells the 1DsM2 by a large margin and most users of this type of camera are not cost constrained.
[a href=\"index.php?act=findpost&pid=93089\"][{POST_SNAPBACK}][/a]
Firstly, the 1D MkII has a far higher frame rate, and this rather that high ISO performance might be the major factor why news and sports photographers prefer it. OF course, frame rate is related to pixel count due to constraints on read-out and processing speed, but these limits will likely recede with technological progress.

Also I disagree that price constraints are not a factor: evan a large news organization feels the cost difference between a large collection of $3,500 bodies and a large collection of $7,000 bodies.


And I will repeat my skepticism that moderate pixel count increases in the same sensor format produces significant worsening of visible noise levels in even-handed viewing comparisons. I have not seen a demonstration that a sensor of the same size and technology with more, smaller photo-sites gives significantly worse visible noise when one uses the same ISO, same degree of enlargement, same viewing distance. Lab. measurements of per pixel noise and 100% on-screen viewing effectively compare on the basis of a higher degree of enlargement from the higher resolution sensor, and/or by cropping the image from the higher resolution sensor to the pixel count of the lower resolution sensor. You would get an increase in visible noise by using a greater degree of enlargement on the same file!
Logged

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3387
larger sensors
« Reply #117 on: January 05, 2007, 02:42:45 pm »

Quote
And I will repeat my skepticism that moderate pixel count increases in the same sensor format produces significant worsening of visible noise levels in even-handed viewing comparisons. I have not seen a demonstration that a sensor of the same size and technology with more, smaller photo-sites gives significantly worse visible noise when one uses the same ISO, same degree of enlargement, same viewing distance. Lab. measurements of per pixel noise and 100% on-screen viewing effectively compare on the basis of a higher degree of enlargement from the higher resolution sensor, and/or by cropping the image from the higher resolution sensor to the pixel count of the lower resolution sensor. You would get an increase in visible noise by using a greater degree of enlargement on the same file!
[a href=\"index.php?act=findpost&pid=93913\"][{POST_SNAPBACK}][/a]

You must use Canon cameras. In the Nikon line all you have to do is look at is the D70 vs the D200. The D70 has lower noise and better high ISO performance. I have observed this personally and comparisons are summarized here: URL=http://www.clarkvision.com/imagedetail/digital.sensor.performance.summary/]Clarkvision[/URL]
Logged

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6600
larger sensors
« Reply #118 on: January 05, 2007, 05:25:50 pm »

Quote
You must use Canon cameras. In the Nikon line all you have to do is look at is the D70 vs the D200. [a href=\"index.php?act=findpost&pid=93916\"][{POST_SNAPBACK}][/a]
As far as I can tell, Clark is doing per pixel S/N ratio comparisons, which I explicitly rejected as not directly relevant to image quality in prints at equal degree of enlargement.

As an extreme example, black and white film uses "pixels" in the form of silver halide crystals with atrociously low S/N ratio and DR, as each outputs either pure black or pure white. The same with B&W prints, which are an ugly scattering of pure black dots on a pure white background if viewed microscopically, "at 100% pixels".

But printing billions of these chemical pixels at high "PPI" densities and viewing from an appropriate distance produces far better performance due to the "dithering" or "half toning" effect.


(By the way, I use two Olympus cameras, a C-2040 and an E-1, and a Canon film camera.)
Logged

John Sheehy

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 838
larger sensors
« Reply #119 on: January 05, 2007, 05:59:20 pm »

Quote
Well, current Canon sensors have a read noise of about 3-4 electrons (and it does not vary that much with pixel size), and there is not a whole lot of room for improvement.

I think 1 electron read noise would be a great improvement.  0.1, even greater.

Someday, technology may count photon hits with a digital counter, and there won't be any read noise at all.  The sensors in current digital cameras aren't digital; they are the analog sensors of digital cameras.  Only the ADC stage, processing, and output, are digital.

Quote
Because of the lower gain associated with small pixels,

Smaller pixels have greater gain, AOTBE, for the same ISO.  If you're thinking of ADU/electrons as "gain"; that really stretches the definition of gain.  Gain, AFAIU, is a black box.  I don't know what voltages the ADC is looking for, so I don't know the actual gain.  What I do know is that for the same camera, the gain is proportional to the ISO setting, unless the camera design uses arithmetic to achieve some ISOs.

Quote
the effect of read noise is much more prominent with small pixels. IOW, if the gain with a large pixel is 1 electron per ADU, a noise of one electron will only cause 1 ADU of noise.  However, if your small pixel has a gain of 9 electrons per ADU,

That must be backward.  How could a small pixel have about 9*4095 electrons at saturation, and a large pixel have about 1*4095 electrons at saturation at the same ISO?

Quote
1 electron of noise would cause a change of 9 ADUs in the noise.

OK, I now see you had a typo in the previous sentence.

Quote
Quantum efficiency could double or treble, but how would you increase the pixel density nine fold without significantly decreasing the gain. The capacitance of silicon is limited and well depth can increase only so much.
[a href=\"index.php?act=findpost&pid=93763\"][{POST_SNAPBACK}][/a]

Read noise is the issue, and it is not proportional to absolute gain.  Read noise is *NOT* the amplification of existing noise in the sensor wells; it is noise *GENERATED* in the reading of the sensor.  That's why it is different as enumerated in electrons, at different ISOs in Roger's experiments, and in mine.

From some of the statements you have made, you seem to think that read noise is a quantum event, like the captured electrons.  When someone says"the read noise at ISO 100 is 30.1 electrons", this doesn't mean each pixel is off by some integer number of electrons, the standard deviation of which is 30.1 electrons for the entire image.  It means that the read noise was measured in ADUs, and then on assumed information about the relationship between ADUs and electrons for that ISO, the ADUs are translated into units of "electrons".  This figure has nothing at all to really do with sensor electrons.

P&S cameras already have nine-fold, compared to DSLRs, and they handle this small pixel thing very well, and would probably be even better with more expensive readout circuitry for a super-MP DSLR.  Look at how a Canon 10D and a Sony F707 compare with the same focal length lens (45mm), from the same distance:




A 60MP DSLR could have better microlenses and readout circuitry than a P&S, I would imagine.
Logged
Pages: 1 ... 4 5 [6] 7 8 ... 10   Go Up