Pixel-size, noise and DR.

Started May 12, 2009 | Discussions
Graystar Veteran Member • Posts: 8,373
Re: Free lunch?

richardplondon wrote:

But if it hits one molecule, then all the other molecules'
uncertainty has been collapsed.

No, it hasn't. The statistical uncertainty at any given pixel is unaffected by the events at any other pixel.

richardplondon
richardplondon Forum Pro • Posts: 10,880
Re: Free lunch?

WahTech2 wrote:

In DPreview forums, there seems to be a notion that single pixel DR
is meaningless and it has no place in sensor design.

It is just that we can't put that meaning into a useful context, unless we know what the context is. And the context for a pixel, is an image - how many pixels are there in that image, and is that a sensible number of pixels to divide it into. The context for a photosite, is a sensor - what is the total size of the sensor, and is this photosite size a sensible way to divide it up.

Of course these issues are all numerically linked - sensor size / megapixels = photosite size. If we only know only one of these, say the photosite size, we cannot deduce the others from that. Once we have TWO of these, though, we can easily deduce the third. Then we can determine image resolution, and many other related matters, which otherwise we could not even discuss.

That is why "per pixel" measurement of anything is - not meaningless - but, not enough.

Look at it another way, lets say all the photons is just hitting one
spot, at find where the spots is, resolution (the number pixels) is
very important, the higher the resolution the more precisely you know
where the stop of light is. Once you found the spots on your sensor,
you likely to want to know how strong the light is. To do that you
must know how much light the pixel can handle (it’s DR), so you know
where it’ll clip. So single pixel DR is far from meaningless, it gets
more important as you place more importance to resolution.

The argument is that you can accumulate DR from a lot of small photosites, and get approximately the same result as the DR from a large photosite. The two situations are of course physically different - and the realities will depart from the theory in separate ways, each with different advantages and disadvantages, in the two cases. But the logic is sound AFAICT.

Indeed you get far more data out of 100 pixels sensor then the 1
pixel sensor. But accuracy will depend on your view. Yes, if want to
print the image, then the 100 pixel sensor will be more accurate (and
more useful). But if just want to measure the number of photons (a
light meter) then the 1 pixel sensor is better, is it not?

I doubt that any camera's (live view) exposure metering reads just a single pixel; it will certainly average across an area of many pixels. But even if it did take a single one, this just reports an averaged value of what is happening across its own surface. We could equally well have divided that into smaller subpixels, averaged their values, and got the same answer - in principle. The reality is of course more complex, and there are practical extremes where the equivalence breaks down. But it's a reasonable enough approximation, IMO, for normal purposes.

RP

 richardplondon's gear list:richardplondon's gear list
Pentax K-5 Sigma 10-20mm F4-5.6 EX DC HSM Pentax smc DA 21mm F3.2 AL Limited Pentax smc DA 70mm F2.4 AL Limited Pentax smc DA 35mm F2.8 Macro Limited +6 more
bobn2
bobn2 Forum Pro • Posts: 61,123
Sorry, too late, we've eaten it (nt)
-- hide signature --

Bob

richardplondon
richardplondon Forum Pro • Posts: 10,880
Re: Free lunch?

Graystar wrote:

richardplondon wrote:

But if it hits one molecule, then all the other molecules'
uncertainty has been collapsed.

No, it hasn't.

I meant, their uncertainty about that same photon. Sorry if I was unclear.

If it lands on A, it therefore does not land on B, C etc - a causal link. Probability in action.

My intention was to move attention onto the actual events - the light. Each new photon will hit at a given (uncertain) coordinate regardless of large or small subdivisions of the surface. When we divide up the surface, the uncertainty whether this coordinate is inside each square, must be pro rata with the area of the square. The "shot noise" always needs to add up, because it is an artefact of the subdivision itself. The light is still just doing whatever it was going to do. If particularly fine subdivision adds in some extra read noise, for technical reasons, then that is a separate matter - an unreliability in the measurement, not an extra irregularity in the light (shot noise).

The statistical uncertainty at any given pixel is
unaffected by the events at any other pixel.

When the NEXT photon approaches, you are right of course, the probabilities have another separate opportunity to play themselves out. That was my point too.

RP

 richardplondon's gear list:richardplondon's gear list
Pentax K-5 Sigma 10-20mm F4-5.6 EX DC HSM Pentax smc DA 21mm F3.2 AL Limited Pentax smc DA 70mm F2.4 AL Limited Pentax smc DA 35mm F2.8 Macro Limited +6 more
ejmartin Veteran Member • Posts: 6,274
Re: Free lunch?

richardplondon wrote:

Graystar wrote:

richardplondon wrote:

But if it hits one molecule, then all the other molecules'
uncertainty has been collapsed.

No, it hasn't.

I meant, their uncertainty about that same photon. Sorry if I was
unclear.

If it lands on A, it therefore does not land on B, C etc - a causal
link. Probability in action.

My intention was to move attention onto the actual events - the
light. Each new photon will hit at a given (uncertain) coordinate
regardless of large or small subdivisions of the surface. When we
divide up the surface, the uncertainty whether this coordinate is
inside each square, must be pro rata with the area of the square.
The "shot noise" always needs to add up, because it is an artefact of
the subdivision itself. The light is still just doing whatever it was
going to do. If particularly fine subdivision adds in some extra read
noise, for technical reasons, then that is a separate matter - an
unreliability in the measurement, not an extra irregularity in the
light (shot noise).

The statistical uncertainty at any given pixel is
unaffected by the events at any other pixel.

When the NEXT photon approaches, you are right of course, the
probabilities have another separate opportunity to play themselves
out. That was my point too.

Not quite; but don't feel bad, you're in good company -- Einstein was unable to give up the notion of objective reality of photons (ie that there is always a photon present, and if it's seen in one place then it's not in another place, etc). This notion of objective reality of quanta was subjected to some analysis in the '60s by the physicist John Bell, who showed that such an assumption is inconsistent with quantum mechanics.

The electromagnetic (EM) field intensity describes the probability per unit time of recording a photon. There is no stream of photons waiting to be observed -- there is no this photon, and the next photon, and the next; there is simply a probability amplitude for an interaction between the EM field and the detector, which results in a quantum of energy being transferred from the field to the detector, which we think of as a photon being absorbed. But the photon wasn't in that particular place; the field gives an amplitude for photon observation spread over the whole region where it is nonzero. And the observation of a photon in one location is largely independent of the observation of a photon in a neighboring location.

Finally, the photons do not have labels, they are indistinguishable particles. All one can say is that a photon was recorded in some location at some particular time. They don't carry identification tags, so a statement such as "that same photon" cannot be given meaning.

It's not that there is a fixed number of photons, and it's just that their arrival times are spaced out according to Poisson statistics (classical objects, like raindrop arrivals, obey Poisson statistics in this way). Rather, the number of photons itself is uncertain and in a sense isn't determined until the measurement is made. And then it is just one of a number of possible outcomes whose probabilities are governed by the rules of quantum mechanics. It is the probability of the outcome that a given number of photons is registered that obeys the Poisson distribution.

-- hide signature --
richardplondon
richardplondon Forum Pro • Posts: 10,880
the mealtime continuum

ejmartin wrote:

Not quite; but don't feel bad, you're in good company -- Einstein was
unable to give up the notion of objective reality of photons (ie that
there is always a photon present, and if it's seen in one place then
it's not in another place, etc). This notion of objective reality
of quanta was subjected to some analysis in the '60s by the physicist
John Bell, who showed that such an assumption is inconsistent with
quantum mechanics.

I had heard, despite not getting the memo directly; I try not to think about this kind of thing too much, it makes me fretful ;-(

I did use the word "simplify" - also the words "thought experiment", and "logic". My interest has been in logical rather than purely technical aspects. I simply lack the technical background to wrangle the really useful quantitative concepts here, but enjoy unpicking bad logic, wherever it is found. I do welcome any comments and corrections, of course. Or a gentle suggestion that I shut up, that would work too.

The electromagnetic (EM) field intensity describes the probability
per unit time of recording a photon. There is no stream of photons
waiting to be observed -- there is no this photon, and the next
photon, and the next; there is simply a probability amplitude for an
interaction between the EM field and the detector, which results in a
quantum of energy being transferred from the field to the detector,
which we think of as a photon being absorbed. But the photon wasn't
in that particular place; the field gives an amplitude for photon
observation spread over the whole region where it is nonzero. And
the observation of a photon in one location is largely independent of
the observation of a photon in a neighboring location.

Finally, the photons do not have labels, they are indistinguishable
particles. All one can say is that a photon was recorded in some
location at some particular time. They don't carry identification
tags, so a statement such as "that same photon" cannot be given
meaning.

OK, but it seems to me that we still have some distinct events to observe: detectors reporting that they have been triggered. To reword my example, we have an array of detectors assigned to a given area of the sensor, and an uncertainty about which one of these we will FIRST hear to ping (and we will assume that this triggering was influenced by photon-related... whatever we want to call it). The silence from the other detectors, and the ping from the triggered one, can be said to all reflect a shared probability context - is that any better? Still nonsense? I know that this is all highly artificial, I was trying to make a point, thinking as I went along.

I propose to disregard any relativistic observer timing issues in working out which ping was first...

So, that granted, we also have an equivalent uncertainty about which detector will be triggered second, and about when a given one will be next triggered, etc... let's forget actual photons by all means. Let's talk vaguely about some disorderly and problematic "lightiness" with severe domestic problems, due to an unsettled family background. And let's, please, leave out of consideration the space curvature caused by the gravity well of my own huge mass of ignorance

best regards, RP

 richardplondon's gear list:richardplondon's gear list
Pentax K-5 Sigma 10-20mm F4-5.6 EX DC HSM Pentax smc DA 21mm F3.2 AL Limited Pentax smc DA 70mm F2.4 AL Limited Pentax smc DA 35mm F2.8 Macro Limited +6 more
bobn2
bobn2 Forum Pro • Posts: 61,123
Re: Free lunch?

and if you connect your camera to a really hot cup of tea....

ejmartin wrote:

richardplondon wrote:

Graystar wrote:

richardplondon wrote:

But if it hits one molecule, then all the other molecules'
uncertainty has been collapsed.

No, it hasn't.

I meant, their uncertainty about that same photon. Sorry if I was
unclear.

If it lands on A, it therefore does not land on B, C etc - a causal
link. Probability in action.

My intention was to move attention onto the actual events - the
light. Each new photon will hit at a given (uncertain) coordinate
regardless of large or small subdivisions of the surface. When we
divide up the surface, the uncertainty whether this coordinate is
inside each square, must be pro rata with the area of the square.
The "shot noise" always needs to add up, because it is an artefact of
the subdivision itself. The light is still just doing whatever it was
going to do. If particularly fine subdivision adds in some extra read
noise, for technical reasons, then that is a separate matter - an
unreliability in the measurement, not an extra irregularity in the
light (shot noise).

The statistical uncertainty at any given pixel is
unaffected by the events at any other pixel.

When the NEXT photon approaches, you are right of course, the
probabilities have another separate opportunity to play themselves
out. That was my point too.

Not quite; but don't feel bad, you're in good company -- Einstein was
unable to give up the notion of objective reality of photons (ie that
there is always a photon present, and if it's seen in one place then
it's not in another place, etc). This notion of objective reality
of quanta was subjected to some analysis in the '60s by the physicist
John Bell, who showed that such an assumption is inconsistent with
quantum mechanics.

The electromagnetic (EM) field intensity describes the probability
per unit time of recording a photon. There is no stream of photons
waiting to be observed -- there is no this photon, and the next
photon, and the next; there is simply a probability amplitude for an
interaction between the EM field and the detector, which results in a
quantum of energy being transferred from the field to the detector,
which we think of as a photon being absorbed. But the photon wasn't
in that particular place; the field gives an amplitude for photon
observation spread over the whole region where it is nonzero. And
the observation of a photon in one location is largely independent of
the observation of a photon in a neighboring location.

Finally, the photons do not have labels, they are indistinguishable
particles. All one can say is that a photon was recorded in some
location at some particular time. They don't carry identification
tags, so a statement such as "that same photon" cannot be given
meaning.

It's not that there is a fixed number of photons, and it's just that
their arrival times are spaced out according to Poisson statistics
(classical objects, like raindrop arrivals, obey Poisson statistics
in this way). Rather, the number of photons itself is uncertain and
in a sense isn't determined until the measurement is made. And then
it is just one of a number of possible outcomes whose probabilities
are governed by the rules of quantum mechanics. It is the
probability of the outcome that a given number of photons is
registered that obeys the Poisson distribution.

-- hide signature --

Bob

kwik Forum Member • Posts: 84
What an elegant description...

ejmartin wrote:

Not quite; but don't feel bad, you're in good company -- Einstein was
unable to give up the notion of objective reality of photons (ie that
there is always a photon present, and if it's seen in one place then
it's not in another place, etc). This notion of objective reality
of quanta was subjected to some analysis in the '60s by the physicist
John Bell, who showed that such an assumption is inconsistent with
quantum mechanics.

The electromagnetic (EM) field intensity describes the probability
per unit time of recording a photon. There is no stream of photons
waiting to be observed -- there is no this photon, and the next
photon, and the next; there is simply a probability amplitude for an
interaction between the EM field and the detector, which results in a
quantum of energy being transferred from the field to the detector,
which we think of as a photon being absorbed. But the photon wasn't
in that particular place; the field gives an amplitude for photon
observation spread over the whole region where it is nonzero. And
the observation of a photon in one location is largely independent of
the observation of a photon in a neighboring location.

Finally, the photons do not have labels, they are indistinguishable
particles. All one can say is that a photon was recorded in some
location at some particular time. They don't carry identification
tags, so a statement such as "that same photon" cannot be given
meaning.

It's not that there is a fixed number of photons, and it's just that
their arrival times are spaced out according to Poisson statistics
(classical objects, like raindrop arrivals, obey Poisson statistics
in this way). Rather, the number of photons itself is uncertain and
in a sense isn't determined until the measurement is made. And then
it is just one of a number of possible outcomes whose probabilities
are governed by the rules of quantum mechanics. It is the
probability of the outcome that a given number of photons is
registered that obeys the Poisson distribution.

...of the distinction between "reality" and "measurement".

jeffkrol Veteran Member • Posts: 6,225
2 interesting papers

based on simulations:

http://www.imageval.com/public/Papers/EI%205678-01%20Peter%20Catrysse.pdf
http://www.imageval.com/public/Papers/ResolutionSensitivityTradeoff_SPIE06

-- hide signature --

360 minutes from the prime meridian. (-5375min, 3.55sec) 1093' above sea level.

'The exposure meter is calibrated to some clearly defined standards and the user needs to adjust his working method and his subject matter to these values. It does not help to suppose all kinds of assumptions that do not exist.'
Erwin Puts

WahTech2 New Member • Posts: 15
Re: Free lunch?

richardplondon wrote:

That is why "per pixel" measurement of anything is - not meaningless

  • but, not enough.

I don’t know enough to know what else to measure DR in, other then at per pixel level.

The argument is that you can accumulate DR from a lot of small
photosites, and get approximately the same result as the DR from a
large photosite. The two situations are of course physically
different - and the realities will depart from the theory in separate
ways, each with different advantages and disadvantages, in the two
cases. But the logic is sound AFAICT.

Yes, you are right. You can simply sum up a number of pixels to give you a “larger” pixel. I think it's pixel binning. As far as I know it doesn’t have much effect on DR because it needs to underexpose the sensor to avoid clipping.

But the Fuji EXR sensor have very clever way expanding DR, its basically two sensors interlaced together, to expand DR it simply underexpose one of the sensor, so at lest half of the sensor isn’t clipped.
So summing can work, but you do loss resolution.

I doubt that any camera's (live view) exposure metering reads just a
single pixel; it will certainly average across an area of many
pixels. But even if it did take a single one, this just reports an
averaged value of what is happening across its own surface. We could
equally well have divided that into smaller subpixels, averaged their
values, and got the same answer - in principle. The reality is of
course more complex, and there are practical extremes where the
equivalence breaks down. But it's a reasonable enough approximation,
IMO, for normal purposes.

I was thinking about a hand held light meter and yes all it does is to give you an average. I guess you can add up a number of small pixels to give you the same value but, why? if all I want is an average value:)

Lee Jay Forum Pro • Posts: 54,144
You are right

bobn2 wrote:

Part of the problem is that Roger says things that are actually
profoundly misleading. Two quotes from that page:
'Because good digital cameras are photon noise limited, the larger
pixels will always have higher signal-to-noise ratios unless someone
finds a way around the laws of physics, which is highly unlikely.'
That's where all these spurious quotes of the 'laws of physics' come
from! Well, it's true that larger pixels have higher signal to noise
ratios, but that doesn't mean that images produced from them have
higher SNR's at any given spatial frequency. You'd think, as a
physicist, he'd understand that.

'Image detail can be blurred by diffraction. Diffraction is more of
an issue with smaller pixels, so again cameras with larger pixels
will perform better, giving sharper images.'
Which is just plain wrong.

Right. Here's a little test I did. ISO 1600, f5.6, 1/3s, 20mm. Pixel sizes are drastically different as shown on the image (1.5 micros versus 8.2 microns). Both images are down-sized from their original size. Unfortunately, processing is different because raw isn't available in one of them yet, so take this with a little grain of salt.

-- hide signature --

Lee Jay
(see profile for equipment)

 Lee Jay's gear list:Lee Jay's gear list
Canon IXUS 310 HS Canon PowerShot SX50 HS Canon EOS 7D Mark II Canon EF 70-200mm F2.8L IS II USM Canon EF 8-15mm f/4L Fisheye USM +23 more
WahTech2 New Member • Posts: 15
Re: Free lunch?

ejmartin wrote:

Not quite; but don't feel bad, you're in good company -- Einstein was
unable to give up the notion of objective reality of photons (ie that
there is always a photon present, and if it's seen in one place then
it's not in another place, etc). This notion of objective reality
of quanta was subjected to some analysis in the '60s by the physicist
John Bell, who showed that such an assumption is inconsistent with
quantum mechanics.

The electromagnetic (EM) field intensity describes the probability
per unit time of recording a photon. There is no stream of photons
waiting to be observed -- there is no this photon, and the next
photon, and the next; there is simply a probability amplitude for an
interaction between the EM field and the detector, which results in a
quantum of energy being transferred from the field to the detector,
which we think of as a photon being absorbed. But the photon wasn't
in that particular place; the field gives an amplitude for photon
observation spread over the whole region where it is nonzero. And
the observation of a photon in one location is largely independent of
the observation of a photon in a neighboring location.

Finally, the photons do not have labels, they are indistinguishable
particles. All one can say is that a photon was recorded in some
location at some particular time. They don't carry identification
tags, so a statement such as "that same photon" cannot be given
meaning.

It's not that there is a fixed number of photons, and it's just that
their arrival times are spaced out according to Poisson statistics
(classical objects, like raindrop arrivals, obey Poisson statistics
in this way). Rather, the number of photons itself is uncertain and
in a sense isn't determined until the measurement is made. And then
it is just one of a number of possible outcomes whose probabilities
are governed by the rules of quantum mechanics. It is the
probability of the outcome that a given number of photons is
registered that obeys the Poisson distribution.

This is new to me, I don’t understand quantum mechanics, all that randomness does my head in!

From what I understand of photons, is that, they are well behaved and well understood, it can be described as a particle or a wave. If you send one photon, you’ll receive one photon nothing more, nothing less, this is what “quantum communication” and “quantum cryptography” is relying on, to create a single photon, encode it and send. The receiver picks up that photon and decodes it. That’s how I understood it, but I could be wrong, the science behind it is way beyond me.

Graystar Veteran Member • Posts: 8,373
Re: 2 interesting papers

jeffkrol wrote:

http://www.imageval.com/public/Papers/EI%205678-01%20Peter%20Catrysse.pdf
http://www.imageval.com/public/Papers/ResolutionSensitivityTradeoff_SPIE06

Yeah...seen those a while ago. Been waiting for them to perform their "human psychophysical measurements" but haven't seen anything on it.

Graystar Veteran Member • Posts: 8,373
Re: You are right

ljfinger wrote:

Here's a little test I did.

Did you ever get a chance to take that "control" image?

Lee Jay Forum Pro • Posts: 54,144
Re: You are right

Graystar wrote:

ljfinger wrote:

Here's a little test I did.

Did you ever get a chance to take that "control" image?

Not really possible. I was away from home when I did this test and I don't have the object here.

-- hide signature --

Lee Jay
(see profile for equipment)

 Lee Jay's gear list:Lee Jay's gear list
Canon IXUS 310 HS Canon PowerShot SX50 HS Canon EOS 7D Mark II Canon EF 70-200mm F2.8L IS II USM Canon EF 8-15mm f/4L Fisheye USM +23 more
kwik Forum Member • Posts: 84
QM vs CM

WahTech2 wrote:

This is new to me, I don’t understand quantum mechanics, all that
randomness does my head in!
From what I understand of photons, is that, they are well behaved and
well understood, it can be described as a particle or a wave. If you
send one photon, you’ll receive one photon nothing more, nothing
less, this is what “quantum communication” and “quantum cryptography”
is relying on, to create a single photon, encode it and send. The
receiver picks up that photon and decodes it. That’s how I understood
it, but I could be wrong, the science behind it is way beyond me.

The funny thing about QM (quantum mechanics) is that particles don't exist until you measure them. In fact, it's not even necessary to assume they exist even then. All you know is that you made a measurement that is consistent with the existence of a particle, but even this is not proof of existence, and the wave function simply gives the probability amplitude of a measurement yielding a particular result.

This is the difference between QM and CM (classical mechanics). Most people think of the statistical nature of QM in the same way as we might describe traffic. That is, we do not know what any given car will do at any given moment, but there really is a car, and we can predict what will happen on average.

On the other hand, QM is more like we place a sensor to detect the passing of a car, and it will tell us the probability that the sensor makes a detection. But even though the wave function is based on the idea of a car, that does not mean that such a thing as a "car" really exists, and that it has set off our detector.

I know this sounds ridiculous -- if the car didn't set off the detector, and the detector was designed with a car in mind, and the wave function was based on the idea of a car, and the detector detected cars in the manner predicted by the wave function -- then what else besides a car could it have been? Well, that's the whole thing -- there wasn't necessarily an "it" at all!

Bizarre, I know. But, there you are. This notion is what even Einstein himself couldn't come to grips with, and he's one of the people that developed QM!

WahTech2 New Member • Posts: 15
Re: QM vs CM

kwik wrote:

The funny thing about QM (quantum mechanics) is that particles don't
exist until you measure them. In fact, it's not even necessary to
assume they exist even then. All you know is that you made a
measurement that is consistent with the existence of a particle, but
even this is not proof of existence, and the wave function simply
gives the probability amplitude of a measurement yielding a
particular result.

This is the difference between QM and CM (classical mechanics). Most
people think of the statistical nature of QM in the same way as we
might describe traffic. That is, we do not know what any given car
will do at any given moment, but there really is a car, and we can
predict what will happen on average.

On the other hand, QM is more like we place a sensor to detect the
passing of a car, and it will tell us the probability that the sensor
makes a detection. But even though the wave function is based on the
idea of a car, that does not mean that such a thing as a "car" really
exists, and that it has set off our detector.

I know this sounds ridiculous -- if the car didn't set off the
detector, and the detector was designed with a car in mind, and the
wave function was based on the idea of a car, and the detector
detected cars in the manner predicted by the wave function -- then
what else besides a car could it have been? Well, that's the whole
thing -- there wasn't necessarily an "it" at all!

Bizarre, I know. But, there you are. This notion is what even
Einstein himself couldn't come to grips with, and he's one of the
people that developed QM!

Bizarre is an under statement.

I think I understand the bit about the car not being a car until you detected it.

The thing is, if I send a car down the road, will you detect only a car? and not two cars. And if my detector only detects buses will my car be comes a bus, or is it does it just not detect anything?
My head hurts.

Thnaks for the insight.

kwik Forum Member • Posts: 84
It's more bizarre than that!

WahTech2 wrote:

Bizarre is an under statement.

Don't you know it!

I think I understand the bit about the car not being a car until you
detected it.

It's even more bizarre. The only reason it is a car is because you were searching for a car. That's part of the wave-particle duality conundrum of QM. That is, many ask, for example, if an electron is a wave or particle. The common answer is "both". A more accurate response would be "neither". If you are measuring the properties of a particle, then you will detect a particle. If you are measuring the properties of a wave, then you will detect a wave. In neither instance does it mean that the actual phenomena is either a particle or a wave. It simply means that you measured the properties of a wave or particle, depending on what your experiment was designed to measure.

The thing is, if I send a car down the road, will you detect only a
car? and not two cars.

You presume that you know it is a car in the first place that you are sending down the road. In fact, you don't know what is being sent, what path it is taking, or even where it is going. You have a probability amplitude that the phenomena you are measuring will be detected in a certain spacial interval in a certain time interval.

And if my detector only detects buses will my
car be comes a bus, or is it does it just not detect anything?

This is a whole other matter. For example, the spontaneous "decomposition" of a gamma ray photon into an electron and a positron. In fact, one can make a case that there is no such "decomposition" at all. Rather, one can think instead that an electron moving backwards in time (positron) collides with the gamma ray and rebouds forward in time (electron). Inasmuch as electrons, positrons, and gamma rays are the phenomena your equipment is designed to detect, that is. ; )

My head hurts.

It is really weird stuff, that's for sure.

Thnaks for the insight.

Well, quantum physics isn't my day job (usually), but I did stay in a Holiday Inn Express the other night. : )

Ralf Ronander
Ralf Ronander Contributing Member • Posts: 796
Re: It's more bizarre than that!

If everything is a function of probability, there must be a certain probability that QM doesn´t exist

Just like the car exist only because your measuring searched for and detected it, QM exist as a platform (searched and found) in lack of better understanding.

Or, maybe the car you detected wasn´t a function of your measurement but your measurement was a function of there being a car.

Otherwise, the really bizarre thing is that if I mount my camera on a tripod and take 100 snaps of a certain scene, they all look exacytly the same!
(Yet they are far from the same if you look at the noise distribution)

How´s that for probability

bobn2
bobn2 Forum Pro • Posts: 61,123
Re: QM vs CM

kwik wrote:

WahTech2 wrote:

This is new to me, I don’t understand quantum mechanics, all that
randomness does my head in!
From what I understand of photons, is that, they are well behaved and
well understood, it can be described as a particle or a wave. If you
send one photon, you’ll receive one photon nothing more, nothing
less, this is what “quantum communication” and “quantum cryptography”
is relying on, to create a single photon, encode it and send. The
receiver picks up that photon and decodes it. That’s how I understood
it, but I could be wrong, the science behind it is way beyond me.

The funny thing about QM (quantum mechanics) is that particles don't
exist until you measure them. In fact, it's not even necessary to
assume they exist even then. All you know is that you made a
measurement that is consistent with the existence of a particle, but
even this is not proof of existence, and the wave function simply
gives the probability amplitude of a measurement yielding a
particular result.

This is the difference between QM and CM (classical mechanics). Most
people think of the statistical nature of QM in the same way as we
might describe traffic. That is, we do not know what any given car
will do at any given moment, but there really is a car, and we can
predict what will happen on average.

On the other hand, QM is more like we place a sensor to detect the
passing of a car, and it will tell us the probability that the sensor
makes a detection. But even though the wave function is based on the
idea of a car, that does not mean that such a thing as a "car" really
exists, and that it has set off our detector.

I know this sounds ridiculous -- if the car didn't set off the
detector, and the detector was designed with a car in mind, and the
wave function was based on the idea of a car, and the detector
detected cars in the manner predicted by the wave function -- then
what else besides a car could it have been? Well, that's the whole
thing -- there wasn't necessarily an "it" at all!

Bizarre, I know. But, there you are. This notion is what even
Einstein himself couldn't come to grips with, and he's one of the
people that developed QM!

There's a simple experiment that every physics undergraduate gets to do. You take two slits and send light through them, and get a diffraction pattern. You can understand the diffraction pattern in terms of the wave nature of light as two wavefronts interfering with each other, giving the classical pattern. Or you can understand it from the point of view of quantum mechanics, with that interference dictating the probability distribution of the arrival of the photons. Now the neat bit, you dim the light source so that the probability of there being more than one photon in the system at any time is vanishingly small and the diffraction pattern is still there . This completely defies a notion of photons as little bullets, since how could a bullet 'know' where the previous photons had been.

We all tend to deal in simplified metaphors when trying to understand the nature of the universe (apart from the really talented mathematicians) but you do need to know the limits of those metaphors.

-- hide signature --

Bob

Keyboard shortcuts:
FForum MMy threads