NIK in all its glory....incidentally easy controllable HDR

See, the difference between you and me is that I want to know how much of an image quality degradation to expect and you just keep on saying that there is no proof of degradation.

The difference is that in over a year of asking and ranting about image degradation from heat, NO ON ANYWHERE with experience and testing has agreed to your notion nor have you ever provided any examples or proof of image degredation!

Great excuse not to buy or use a camera.

Meanwhile the rest of us photographers ARE making images without excuse.

Just one agonizing more paper trail where not one person took you seriously:

Any real measurement data on mirrorless sensor heating?

Edited October 19, 2020 by digitaldog

nor have you ever provided any examples or proof of image degredation!

I guess the camera manufacturers are all idiots.

I guess the camera manufacturers are all idiots.

You guess, assume, speculate. You have no data. The actual data provided was the opposite of your assumptions which you cannot accept.

You were told countless times the facts about this from actual users, a camera testing house and as you always do when you don't hear what you want to hear, you dismissed all the data/facts. You were shown an example from one mirrorless camera manufacturer about a warning whereby IF a sensor got to a heat level that could affect image data, the camera WOUDL shut down. You ignored that of course. Because you're sure, without proof and to the contrary, from Dx0 and actual users who did actual testing, this is a non issue. You have a flat earth theory that has zero basis in data. It keeps you from supposedly buying a camera and certainly ever showing anyone you can take let alone make an image. No one on the LuLa forums took you seriously. Long after examples of this non issue where provided by someone actually doing sound testing. You can guess the camera manufacturers are all idiots for providing warnings when their cameras will automatically shut down but you'd be guessing who the wrong idiot is.

when their cameras will automatically shut down

Wrong again. The sensor starts heating up from the moment you switch your camera on, worse so in live view or video mode. Warmer sensor = more sensor noise = less dynamic range and potentially color and other issues; that's not my opinion, that's straight our of user manuals. Camera shut-off only occurs when the temperature gets that hot that physical damage to the sensor and other electronics may occur, it doesn't happen because of the possibility of image quality degradation. That's why manufacturers warn of image quality degradation, because it occurs before shut-down. If it didn't occur before shut-down, they wouldn't have to warn you about it.

Wrong again. The sensor starts heating up from the moment you switch your camera on, worse so in live view or video mode. Warmer sensor = more sensor noise = less dynamic range and potentially color and other issues; that's not my opinion.

You are of course entitled to your uninformed opinions on this subject. As I am entitled to my fact based data points.

Do you have any data, do you have ANY hard data Frans? You do not. By your own admission, and questions.

Edited October 19, 2020 by digitaldog

Trouble with the eyes, Andrew? "

that's not my opinion, that's straight out of user manuals.

 

my fact based data points

And what are those?

Trouble with the eyes, Andrew? "

And what are those?

Trouble with your eyes, memory and Reading comprehension? We went over this days ago and you've ignored or forgotten or never understood in the first place:

 

I own many digital cameras, one IS a Sony Mirrorless. Here's exactly their warning in their manual:

High-quality movie recording and high-speed continuous shooting require large amounts of power. Therefore, if you continue to shoot, the temperature inside the camera will rise, especially that of the image sensor. In such cases, the camera will turn off automatically because the surface of the camera is heated to a high temperature or the high temperature affects the quality of the images or the internalmechanism of the camera.

 

There is absolutely nothing that talks of a reduction of DR. In fact, the camera will in such a rare case, a case that isn't a recommended usage, as I said earlier, this camera (my camera) WILL shut off AUTOMATICALLY.

You have no data, hard, measured or otherwise. Without data, you're just a person with an opinion. An opinion that further, no one who has talked to on the topic has ever agreed with you on.

I'm sorry that reality and facts continue to ruin your life.I’m sorry I offended you with my common sense and that of others who have answered your question about sensor noise and dismissed your unproven ideas. Adiós

We went over this days ago and you've ignored or forgotten or never understood in the first place:

I own many digital cameras, one IS a Sony Mirrorless. Here's exactly their warning in their manual:

High-quality movie recording and high-speed continuous shooting require large amounts of power. Therefore, if you continue to shoot, the temperature inside the camera will rise, especially that of the image sensor. In such cases, the camera will turn off automatically because the surface of the camera is heated to a high temperature or the high temperature affects the quality of the images or the internalmechanism of the camera.

Not ignored, not forgotten and certainly not misunderstood. Let's apply some critical thinking to the warning in the Sony manual. Especially the sensor will heat up in certain shooting modes. Higher sensor temperatures means more noise; more noise means less dynamic range. Other issues like color shifts, fogging, stuck pixels, etc. may occur, although Sony doesn't mention those. Whatever combination of increased noise, color shifts, fogging, stuck pixels, etc. occurs, the image quality will degrade and this degradation will get worse as the temperature rises until the camera may finally shut down at some point. So, between starting to use those modes and the point at which the camera shuts down you have image degradation; the longer you use those modes the worse the degradation.

I want to see measurement that quantify the image quality and/or dynamic range degradation, but sofar no luck.

 

BTW, since when are high-quality movie recording and high-speed continuous shooting not recommended usage?

Exposure is best illustrated by the characteristic curve, found in film specification sheets, but applicable to digital imaging as well.

 

 

The horizontal axis (abscissa) reflects the amount of light reaching the film/sensor (input). The vertical axis (ordinate) represents the effect on the image (output), which is closer to photographic intent. The dynamic range of capture is measured on the abscissa, bounded by the ends of the curve where the slope approaches zero (or some arbitrary value).

 

The conflict in the thread regarding "exposure" is found in which axis constitutes "exposure" in the photographic sense. A light meter measures the amount of light reflected by the subject. The meter will then suggest a combination of aperture and shutter speeds, modified by the ISO setting and choices by the photographer (e.g., high key, low key). In the photographic, rather than scientific, vernacular, exposure is judged on the effect on the image. There is no perfect exposure in this sense, alhough one might argue that detail (conrtast or slope)) is retained in the regions of interest.

 

If we wish to describe a photograph as well-exposed, the ordinate is the controlling aspect. If we wish to see if workplace illumination meets local and federal standards, then the abscissa is the only value of interest.

In the photographic, rather than scientific, vernacular, exposure is judged on the effect on the image.

Right, much of the argument here has been like someone using the term “sunrise” and someone else with a penchant for being didactic and abstruse insisting that the sun does not “rise.” Vernacular has been a key all along.

The effect of temperature, short of component failure, on image quality is mainly reflected in increased noise in the shadow areas. This has little effect on the measured dynamic range, but it may affect how useful the image may be in terms of shadow detail. The dynamic range of monitors and printers is far less than that of the digital sensor, so the effect may not be visible at all on those media.

 

Very little heat is generated in the sensor itself, since the electronic current in that process is exceedingly small. The sensor output is measured as a voltage, which is amplified by the very high impedance of the sensing cell. The temperature increases due to the power consumed by image processing, data storage, and the internal resistance of the battery.

 

By moving the battery and data storage outside of the camera, the heat buldup is minimal. I have experienced only one thermal shutdown shooting video with a hybrid camera, after two hours of shooting in a closed room at 93 deg F. The overload setting was on "Normal", and there was no noticeable affect on the image quality in the interrim. I changed the limit to "High", and conntinued to use the camera. That might, according to Sony, make the camera uncomfortable to use against. your face, but not on a tripod.

increased noise in the shadow areas. This has little effect on the measured dynamic range

Eh, come again? Dynamic range is the ratio of maximum signal over the noise level; more noise = less dynamic range.

Just an aside: Data are are not "facts" beyond their own miniscule, usually mistaken, and commonly irrelevant reference points in continua.

Not ignored, not forgotten and certainly not misunderstood.

Sure appears that way after your had to reply again, from the religious school of Waterlander (currently one convert).

Let's apply some critical thinking to the warning in the Sony manual.

When you do, I'll agree to your critical thinking. What you've outlined is further assumptions, speculations, generalizations and as we've all seen in the past, done without a lick of proof, data, science, examples, proof of concept.

Especially the sensor will heat up in certain shooting modes.

Heat up from what temp to what temp, with what cameras Frans? You can't answer that, you never have, you can only generalize. In my Sony camera, the manual was quite clear, but you found the need to make up stuff that Sony never spoke of seen below. All from the religious school of Waterlander (currently one convert).

Higher sensor temperatures means more noise; more noise means less dynamic range.

Exactly how much DR Frans, under what temp limit from ambient? You don't know. You have no data. You assume some camera at some temperature produces some degree of less DR and sure, that's indeed possible. It's also not possible in many conditions you don't know about. I can guarantee that when you don't expose optimally your raw data (assuming you shot raw data, or anything digitally, zero proof of that), you reduce DR/add noise. All that can be shown, proven using actual images, RawDigger etc. That idea, that suboptimal under exposure causes noise is easy to show and prove. Something you've failed to do now for nearly a year with this so called horror from sensor heating noise.

Whatever combination of increased noise, color shifts, fogging, stuck pixels, etc. occurs, the image quality will degrade and this degradation will get worse as the temperature rises until the camera may finally shut down at some point.

Zero proof of increased noise, color shifts, fogging, stuck pixels, etc occurring, only assumptions they will at some unknown and defined point, with some unknown and defined camera to some unknown and defined, unproven image. All from the religious school of Waterlander (currently one convert).

So, between starting to use those modes and the point at which the camera shuts down you have image degradation; the longer you use those modes the worse the degradation.

What point exactly Frans, with exactly what visible or measurable degradation? You can't and have never answered that because your religious ideas about this are, well based on religion not facts. No matter how long you say "The Earth is flat" it isn't flat.

I want to see measurement that quantify the image quality and/or dynamic range degradation, but sofar no luck.

To back up an unproven assumption. An assumption made without measurement data or from you, even an example from an image (with degradation or otherwise, seems you may have never made an image, still waiting on that proof of concept). ;)

BTW, since when are high-quality movie recording and high-speed continuous shooting not recommended usage?

Never. That's more of your made up assumptions. The Sony manual discussed the POSSIBILITY of a heating issue in undefined conditions with high-speed continuous shooting in which case, the people at Sony, not you, shut the camera down automatically.

 

What you could ask, but never have, is how many users have had their camera systems automatically shut down, what were the temp conditions prior to this automatic shutdown, on which camera models and, could said users supply images taken just before the shutdown. But science, testing and facts are not your agenda here. Your agenda is to get attention to a question who's multiple and universal answers you refuse accept, to make assumptions about all cameras, heat, sensors and degradation while avoiding just buying and using a camera for photography. OR proving your point about your specific camera showing degradation due to sensor heat. It's all an excuse based on a religion of one.

 

Meanwhile, the rest of the photographers who will not subscribe to the religious school of Waterlander are making images. Maybe with some noise, maybe not but images. You should try that sometime; it's far more gratifying than making up far fetched, flat earth ideas about (oh, the horror), some possible, some minor noise or reduction of DR. Your lack of making images shows a total lack of noise happy? :eek:

Edited October 19, 2020 by digitaldog

Just an aside: Data are are not "facts" beyond their own miniscule, usually mistaken, and commonly irrelevant reference points in continua.

“When the Facts Change, I Change My Mind. What Do You Do, Sir?” John Maynard Keynes

Eh, come again? Dynamic range is the ratio of maximum signal over the noise level; more noise = less dynamic range.

Eh, come again? So how do YOU measure and report maximum signal over the noise level Frans? You don't above, you simply make more generalizations, so let's dig deeper into this idea with specifics!

 

Dynamic Range | imatest

Dynamic Range (DR) is the range of exposure, i.e., scene brightness, over which a camera responds with good contrast and good Signal-to-Noise Ratio (SNR).

 

Noise is a random stochastic process. When noise is high in relation to signal level (low SNR), is can have a strong effect on measurement consistency. In particular, the slope-based dynamic range has been discussed in several places above. A key problem with this measurement is that SNR is frequently very low (−10 dB or worse) at its darker limit. (It can also be affected by tone mapping and flare light.) For this reason we don’t recommend its use.

Getting back to this idea of 'single frame HDR" and DR/noise:

 

http://dougkerr.net/Pumpkin/articles/ISO_Dynamic_range.pdf

The dynamic range of a digital camera can be simplistically defined as the ratio of the maximum and minimum luminance that a camera can “capture” in a single exposure.

The dynamic range of a digital camera is often (very simplistically) defined as the ratio between the maximum and minimum luminance which, in a single image, can be successfully “captured”.

Frans very simplistic definition without metric.

 

The role of noise

Another matter we must consider before settling on a definition of dynamic range is that of noise, the random variation in the reported luminance of pixels compared to their actual luminance values. We recognize that in fact the image, for areas below a certain base luminance, may be so “corrupted” by noise that we cannot honestly say that detail carried by small variations of luminance about that base luminance is “captured”, at least not in a way that is usable.

 

International standard ISO15739-2003, which covers the definition and measurement of the noise performance of digital still cameras, also provides a definition for dynamic range and an associated measurement procedure, based on the noise outlook we discussed above.

The concept behind the dynamic range definition given by ISO 15739 is based on the ratio of the maximum luminance that receives a unique coded representation (the “saturation” luminance) to the lowest luminance for which the signal to noise ratio (SNR) is at least 1.0. This is based on the very arbitrary assumption that detail recorded with a SNR of 1.0 or above is useful and that recorded with an SNR less than 1.0 is not.

 

We measure the noise at a certain arbitrary low luminance (one that is still high enough that the “truncation” problem mentioned just above would be negligible, since only a very tiny fraction of the occurrences would now have negative values). This is called the “reference black” luminance. (In the ISO standard, this luminance is in fact 1/100 of the maximum recordable luminance.)

Congratulations !

 

You have successfully addressed two popularly-appreciated photographic phenomena using an exquisitely non-photographic frame of reference..

 

They can also be addressed using a jazz frame of reference. Think about clarinet and sax in a Duke Ellington composition. .

Wish I could congratulate you for actually confirming that there IS such a thing as a single frame HDR rather than a misunderstanding and use of a BS marketing term, but you can't do so; so congratulations are not in order. :(

Maybe you have specific examples of sensor heat image degradation measurments for Frans and then congratulations are in order? :D

Edited October 21, 2020 by digitaldog

Dynamic range for photographic purposes is generally measured based on the characteristic curve, which varies with the ISO/gain of the medium. Digital has hard limits for dark (0%) and light (100%) in the ordinate, which cannot be exceeded. If the response curve were perfectly linear, there would be a sharp break at each end, where further decreasing the exposure (abscissa) has no effect. In practice there is curvature at the low end due to random noise and in-sensor image processing, which can also impose a curve at the light end. Dynamic range is determined by the ordinate value between two arbitrary points on the curve, for example where the slope of the curve falls below a certain value. The slope of the curve describes the contrast (gamma) at that point, hence the distinction of the boundary between white and black patches.

Dynamic range for photographic purposes is generally measured based on the characteristic curve, which varies with the ISO/gain of the medium. Digital has hard limits for dark (0%) and light (100%) in the ordinate, which cannot be exceeded. If the response curve were perfectly linear, there would be a sharp break at each end, where further decreasing the exposure (abscissa) has no effect.

Fortunately, raw data is perfectly linear (TRC 1.0). No H&D curve like film.

https://wwwimages2.adobe.com/content/dam/acom/en/products/photoshop/pdfs/linear_gamma.pdf

Fortunately, raw data is perfectly linear

As Ed indicated, the noise floor causes an output value when the input is zero (a true black part of a scene). So, apart from possible sensor and reading circuitry minor non-linearities above the noise floor, the noise itself causes non-linearity at the shadow end of the response curve. So, while as a first approximation one can say that digital cameras have a linear response, that's not really the case when you take a closer, detailed look.

As Ed indicated, the noise floor causes an output value when the input is zero (a true black part of a scene). So, apart from possible sensor and reading circuitry minor non-linearities above the noise floor, the noise itself causes non-linearity at the shadow end of the response curve. So, while as a first approximation one can say that digital cameras have a linear response, that's not really the case when you take a closer, detailed look.

 

I don't think so. The curve--which in this case is a straight line--represents the sensor's response to light input. Noise is superimposed on that. I would expect the system to be homoskedastic--that is, randomly equivalent distributions of noise at any point on the ordinate. This doesn't make the relationship nonlinear.

As Ed indicated, the noise floor causes an output value when the input is zero (a true black part of a scene). So, apart from possible sensor and reading circuitry minor non-linearities above the noise floor, the noise itself causes non-linearity at the shadow end of the response curve. So, while as a first approximation one can say that digital cameras have a linear response, that's not really the case when you take a closer, detailed look.

So how do YOU measure and report maximum signal over the noise level Frans (he has to ask again)?

Fortunately, raw data is still perfectly linear. Now some camera manufacturers can muck around with their raws but that doesn't change what Bruce wrote about how sensors 'count' photons: They just count photons in a linear fashion.

And if you wish to use the full DR of your camera (in a single capture since that's a fixed attribute) without the worry about excessive heat before auto shutdown, consider optimal exposure for raw:

How to Use the Full Photographical Dynamic Range of Your Camera

That is if, you have a digital camera, use it, and it provides you raw data Frans. ;)

Fortunately, raw data is still perfectly linear.

Quote from Characterization of CMOS Image Sensor, Master of Science Thesis, For the degree of Master of Science in Microelectronics at Delft University of Technology, Utsav Jain August 31, 2016:

"4.2.2.Non-Linearity Ideally, a CMOS image sensor should have a linear behaviour i.e. it should respond linearly to the incident light (photons) but due to nonlinear devices in the pixel and in the signal processing unit, the sensor deviates from a perfectly linear response. The major source of non-linearity in the image sensor comes from the source follower MOSFET in the 4T pixel design, as it is used as trans-impedance amplifier and its gain depends on the source resistance which induces the non-linearity [25]. Other transistors are just used as a switch and thus do not contribute much to the non-linearity of the image sensor. Other sources of non-linearity are [26]: 1. Image lag in the PPD. 2. The non-linearity of the photodiode or of the floating diffusion. 3. Non-linearity’s in the further downstream analog processing and multiplexing. The non-linearity can be classified into Integral non-linearity (INL) and differential nonlinearity (DNL). INL is the measure of maximum deviation or error from the ideal response and DNL quantifies the deviation of two consecutive output values corresponding to ideal output values of the image sensor. In the case of an image sensor, only INL is calculated, as there are no DACs used in the imager for signal processing. So the photo response tells the response of the sensor for incident optical power and ideally it should be linear, and INL is evaluated from the actual response to the ideal response of the sensor [27]. INL[%] = Emax FS × 100 (4.2-1) Where, INL = Integral non-linearity, Emax = maximum error from best-fit straight line [Volts/DN], FS = full-scale value or maximum output of sensor [Volts/DN]."

Quote from Characterization of CMOS Image Sensor, Master of Science Thesis, For the degree of Master of Science in Microelectronics at Delft University of Technology, Utsav Jain August 31, 2016:

"4.2.2.Non-Linearity Ideally, a CMOS image sensor should have a linear behaviour i.e. it should respond linearly to the incident light (photons) but due to nonlinear devices in the pixel and in the signal processing unit, the sensor deviates from a perfectly linear response.

 

Wow Frans, not only can you not answer my question asked twice, you don't seem to understand the other points made, some of which you blindly copied and pasted: Now some camera manufacturers can muck around with their raws but that doesn't change what Bruce wrote about how sensors 'count' photons: They just count photons in a linear fashion.

Do attempt to read and comprehend the actual info YOU just posted beginning with:

"Ideally, a CMOS image sensor should have a linear behaviour i.e. it should respond linearly to the incident light (photons)"

Exactly what Bruce and I wrote.

Then you missed:

...but due to nonlinear devices in the pixel and in the signal processing unit...

And can't connect the dots to what was written:

"Now some camera manufacturers can muck around with their raws but that doesn't change what Bruce wrote about how sensors 'count' photons: They just count photons in a linear fashion."

For a guy posting on multiple forums about hard measurement data, you seem to really struggle when asked to provide any:

So how do YOU measure and report maximum signal over the noise level Frans (I have to ask for the third time)?

 

Edited October 21, 2020 by digitaldog