gvarner Loc: Central Oregon Coast

What determines the dynamic range in a sensor? Is it just the number of pixels and the sensor dimensions? I don’t recall seeing dynamic range info featured in camera specs, e.g. +- 5 stops. Thank you for you thoughts.

rmalarz Loc: Tempe, Arizona

Dynamic range is the ratio between the maximum output signal level and the noise floor at minimum signal amplification (noise floor which is the RMS (root mean square) noise level in a black image). The noise floor of the camera contains sensor readout noise, camera processing noise and the dark current shot noise.
--Bob

Fotoartist Loc: Detroit, Michigan

A big factor in dynamic range is the size of the pixels. A larger pixel will have the capability of having a larger dynamic range. From that you could reason that higher megapixel cameras where the pixels are made smaller to pack more in the sensor will have lower dynamic range.

From this I reason that there is a sweet spot for overall digital camera performance of around the 24 megapixel point.

RWR Loc: La Mesa, CA

I will not pretend to know the answer to your question, but I equate dynamic range to exposure latitude - how much overexposure will blow out highlights, and how much underexposure will result in lost shadow detail. I know it varies with the type of film and how it is processed, and with digital probably varies depending upon how much it is fussed with in post processing. Perhaps that’s why you don’t see it in the specs.

camerapapi Loc: Miami, Fl.

I am no tech guru, as a matter of fact I usually quote what I learn on line and I can verify. I had a Nikon D2H, most probably the camera with the largest pixels of all the cameras Nikon has made. Its dynamic range was pretty poor in my opinion.
All modern cameras have wide dynamic ranges, usually in the neighborhood of 10 stops and that is considerably more than b&w film that has about 7 stops. Many cameras, like my Nikon D610 goes to 12 stops.

I do not know what determines dynamic range. Old sensors went up to 5 stops as it was for slide film so there must be something done different to new sensors for such an extraordinary long dynamic range.
As I said, I am no tech guru.

burkphoto Loc: High Point, NC

Soooo many factors.

> Sensel size (the size of each photosensitive site on the sensor)

> Sensel technology (lenses, tri-color filters, anti-aliasing technology, IR/UV filtration, etc.

> Physical properties not related to sensel size (distance from next point in circuit, heat dissipation, shielding, etc.)

> Amplifier design (sensel output is voltage, and is usually amplified at certain ISOs)

> Analog-to-Digital converter configuration (pre- or post-amplification, shielding, etc.)

> Processing scheme (Bayer algorithms that merge adjacent digitized sensel voltage values into pixel values)

That's just a small sampling.

Most cameras can record raw files that contain the digitized values from each sensel, so you can process the data on a computer. However, each raw file ALSO contains a JPEG processed in the camera for use as a preview image by your computer operating system. That JPEG also contains an EXIF table of metadata that records exposure data, hardware configuration, time, date, and camera menu settings.

The processing algorithms used to develop raw data into images CHANGE OVER TIME. They get better! So a modern algorithm in 2021 software might provide cleaner, more accurate images than software used in 2005.

Dynamic range is typically measured by dxomark.com, for most popular camera bodies. It is an IMPORTANT measure of sensor performance, but hardly the most important in all situations. If you head over there and check out some of the graphs, you can see a trend:

> Full frame cameras tend to record 14 to 15 stops of dynamic range.

> APS-C cameras tend to record 13-14 stops of dynamic range.

> Micro 4/3 cameras tend to record 12-13 stops of dynamic range.

There are exceptions, of course, but this is close to state of the art.

THAT SAID, to get the full potential of that range, post-processing raw files into images generally is required.

In-camera JPEG processors produce images with only around 5.5 to 6 stops of dynamic range. Silver Halide photo paper can reflect about 5 to 5.5 stops of that dynamic range. Monitors (typical 8-bit monitors) can display maybe six stops of dynamic range properly.

And while you CAN manipulate camera menu features on most models to extend highlight and shadow range conversion to JPEG, most people don't do that. It's easier to record raw data and post-process it to recover some of the remaining details the sensor recorded.

In post, you can COMPRESS the tonal range to fit the output medium of choice, giving viewers the illusion that they are seeing a better approximation of reality.

Rongnongno Loc: FL

Answers are too complicated...

The dynamic range is simply the exposure latitude you can use before you blow one extreme, highlight or dark. The range is a number between 10/16 at the moment. +-5/+-8. The greater the number the better.

The dynamic range is related to luminosity, not color rendition*.

To exploit the full DR of a camera one needs to use raw. A JPG out of a camera will benefit from the camera native dynamic range but will not offer the potential edit that the raw format does.

There is a relation between the sensor element size (density) but this is not of any interest here.

---------------
* Some will dispute that. A correctly exposed capture colors will not suffer from the lack of dynamic range. The color simply will not have the same shade variations.

PHRubin Loc: Nashville TN USA

Long but informative. The range shown is for current technology. The state of the art has been improving steadily. How long it will continue is anyone's guess. And, NO, Dynamic Range is not usually listed in specifications. They do list autofocus sensitivity range, but that doesn't tell you how much range is in the photo.

gvarner Loc: Central Oregon Coast

Good answer. Non-technical. Sounds right to me.

BebuLamar

For short answer Bob Malarz already had it. It's the signal to noise ratio of the sensor. But what affect the signal to noise ratio of the sensor then your list is correct.

burkphoto Loc: High Point, NC

There is little point to considering "sensor dynamic range" in isolation. We record images with whole cameras. A camera is a system, and every system I've ever worked with is only as good as its weakest point. In photography, with recent cameras, that weak point is seldom the sensor. It's usually the image processing.

Since part of the photography system can be outside of the camera (computer, monitor, calibrator, software, printers, profiles, etc.), it makes sense to focus on that, too, when you're striving to maximize *apparent* dynamic range.

Silver halide photographic paper reflects about 90% of the light falling on it, under the best of circumstances. Out of an 8-bits per channel image, a range of values somewhere between 12-242 and 18-236 is all we see reflected from most papers. That's around 5 f/stops. But with 12-15 stops of range possibly recorded in raw data, what do we do with the other 7-9 stops?

The answer is found in the various sliders in post-processing software. Detail that is "burned out" or "plugged up" in an out-of-camera JPEG may be there in a raw file of the same image. When that is the case, much of it can be "recovered" (tonally compressed to the point we can see it within the range of brightness that the paper or screen can reflect or transmit). Of course, it is possible to make an awful mess of an otherwise good image, by over-applying adjustments. And unless the monitor is properly capable, calibrated, and profiled, adjusting images at all may well do more harm than good.

So take dynamic range analysis of sensors with a grain of salt. In the final analysis, most PEOPLE don't give a rat's patoot what camera a photographer used, or whether it was film or digital, or whether it was full frame or smaller. They care about the art, or the communications value, or the emotional impact of the image. If it speaks to them in a way that has the photographer's intended effect, who cares?

Most of the spec wars don't amount to anything tangible unless we're making HUGE prints on 8-14 color inkjet printers, on really exotic, archival, museum grade papers. That can be important in the world of landscape photography, or point-of-purchase advertising, or copying artist's renderings from paper or canvas. But for the rest of us making 16x20 and smaller prints, or just viewing our images on screens, it's a fairly moot point.

Wallen Loc: Middle Earth

No simple answer as although you are asking only about the sensor, there are many other component and settings that work together in and out of a camera.

User ID

Amen. So simple, but Hogsters do seem to thrive on long winded pointlessness.

:::::::::::::::::::::::::::::::::::::::

There’s several usernames I skip right over, hardly ever read a word. Not saying I disagree with them. I just don’t need endless looong pages of non-nutritional filler on my plate. But you can see by the replies that many Hogsters just eat that stuff right up !

Strodav Loc: Houston, Tx

You got some good answers on what it is, but what it means is more important. Here's a graph of the Dynamic Range of a Nikon D850 from DxOMark.com. The human eye can capture a difference between light and dark of about 20 stops. That's 2^20 or we can perceive something about 1,000,000 times darker than the brightest object in a scene. Higher end cameras can capture around 14 to 15 stops as base ISO, less than the human visual system, but the graph shows that ability drops very quickly a stop or two above base. At the same time the noise (green to red bar on the right) starts to go up. It's a good idea to understand this curve before bumping up ISO and then wondering why the image looks flat and washed out or the highlights are blown out or the shadows are buried. If you are getting blinkies, one option might be to lower the ISO at the expense of shutter speed and/or aperture. At the same time, the number of bits of color a sensor can capture also goes down with increasing ISO.

Good topic OP

(Download)

repleo Loc: Boston

It is not what you asked, but if you want to compare the DR of any two (three ?) cameras go to DXOMark and check their camera database. Sort the results by 'Landscape' for order of descent of DR.