Noise (Page 2): Aside from all the very helpful explanations, it basically comes down to "whats visable". If you would look at a "No Noise" 8X10 print with a magnifiyng glass you would...

boberic Loc: Quiet Corner, Connecticut. Ex long Islander

Aside from all the very helpful explanations, it basically comes down to "whats visable". If you would look at a "No Noise" 8X10 print with a magnifiyng glass you would see no noise. Blow that print up to wall size, the closer you get to the print the more noise will be visable. Will there ever be a no noise photo? Not unless the laws of physics are repealed. But the search for perfection is a fools errand.

Largobob

tdekany wrote:

What iso do you normally shoot at?

But to answer your question directly, google the Sony A7sII. King of high iso.

Wow.....that surely didn't answer the question!!!

Tronjo Loc: Canada, BC

johnnycamra wrote:

I was wondering what causes noise with higher iso levels and with today's technology why can't someone make a camera that can produce photos without noise at any iso level? Just like cd's did for lp's. Maybe that is something that will happen in the future? Thanks.

Read this: https://photographylife.com/what-is-noise-in-photography

bclaff Loc: Sherborn, MA (18mi SW of Boston)

TriX wrote:

If you really want to understand the mechanism that various components play in the noise of an image, the manner and which ISO adjustments in-Camera are made, and the effects of underexposure or ETTR, let me suggest the following read: http://theory.uchicago.edu/~ejm/pix/20d/tests/noise/noise-p2.html

That is a great article but the original now has many broken links.
Emil gave me permission to put up a copy at PhotonsToPhotos where I repaired all the problems:

Noise, Dynamic Range and Bit Depth in Digital SLRs by Emil Martinec

TriX Loc: Raleigh, NC

bclaff wrote:

That is a great article but the original now has many broken links.
Emil gave me permission to put up a copy at PhotonsToPhotos where I repaired all the problems:

Noise, Dynamic Range and Bit Depth in Digital SLRs by Emil Martinec

Thanks SO much for doing that - it is one of the best I’ve come across while trying to understand noise in digital photography.

Steve Perry Loc: Sylvania, Ohio

DirtFarmer wrote:

Noise is a natural process. There is noise in all photos, but when there's plenty of light the noise is just too small to notice. There is noise in the audio from a CD also, but it's too small for most people to notice. Here's the deal for photos. Your camera takes in light and converts it to an electrical signal. Electrical signals are made up of electrons. When there is lots of light, it generates lots of electrons. When the light is dim, it generates few electrons. Since electrons are discrete individual particles, they are not infinitely divisible. And when you are counting individual particles there is a natural variability. For a random process (like converting light to electrons) the variability is basically given by the square root of the number of particles counted. So if your light generates 100 electrons, the variability is 10 electrons, or 10%. That's noise. When you have lots of light, you might get a million electrons so the variability is 1000 electrons, or 0.1%. That's noise too, but it's much smaller than what you get in dim light.

When you have low light levels you get few electrons on your sensor. In order to convert that into a useful image you have to multiply the number of electrons you count so you have an image that is bright enough to see. The multiplication factor is related to the ISO level in your camera. Low light, few electrons, high ISO, large multiplication factor. But just multiplying the number of electrons also multiplies the variability in the count by the same factor so if your 100 electrons are multiplied by 10,000 to get it to a million, the variability of 10 electrons is also multiplied up to 100,000 so your noise level is still 10% of the signal.

The square root relationship between signal and noise is only an approximation for large signals. For very small signals there is still variability in the number of electrons counted. So if you have no light at all, there may still be countable electrons in your sensor. These are generated by thermal motions of the atoms in the sensor. They add to the counting noise described above. So there is inherent noise in the image for very dim light levels. Raising the ISO (multiplying the electrons by larger numbers) also multiplies the thermal noise.

Noise is a natural process. There is noise in all ... (show quote)

Excellent explanation.

Ron Dial Loc: Cuenca, Ecuador

Noise is caused by heat across the camera sensor. The higher the ISO, the greater the heat, and the greater the noise. Camera Manufacturers have software that tries to deal with noise but it is an ongoing problem

burkphoto Loc: High Point, NC

johnnycamra wrote:

I was wondering what causes noise with higher iso levels and with today's technology why can't someone make a camera that can produce photos without noise at any iso level? Just like cd's did for lp's. Maybe that is something that will happen in the future? Thanks.

When you turn up the gain (sensitivity or volume) of any electronic circuit that uses analog sensors, whether it's a radio receiver, a digital camera, or microphone, you inevitably pick up a mixture of random background radiation that is present throughout the universe. Mixed with that is interference from terrestrial radio and TV broadcasts, microwave transmissions, cell phones, power lines, electric motors, etc. Another problem is noise generated by or in the circuit itself. HEAT, or infrared radiation, is a major cause of sensor noise. (Anybody remember the first version of the Kodak 14n dSLR? After 15-20 exposures, it got hot. ISO 400 was so noisy, you couldn't use it. Worse, the noise varied in color and intensity across the frame! It was magenta on one side, green on the other... That camera was a $5000 piece of hot... Kodacrap.)

When raising the ISO speed, you are giving up some signal-to-noise ratio for the sake of amplifying what is left to a point where it is useful. But as you increase gain (ISO speed on a camera), you lose the ability to capture the full color range of the scene as well. Color becomes muted, contrast is reduced, and random noise is amplified to the point where it is the predominant feature in an image.

Remember the snow you saw when you tuned an old analog TV set to a distant channel? That's RF noise. Remember driving away from an AM radio station and listening to the signal fade as the noise got louder and louder? That's the AGC (automatic gain control) turning up the receiver sensitivity and amplifying the residual noise of faraway stations, lightning, car ignition systems, power line hum... Remember turning up the microphone input on a cheap tape recorder and hearing a lot of background "hum and hiss"? That is the cheap preamp picking up EMI (electromechanical interference), and RFI (radio frequency interference), in addition to room noise and its own circuit noise.

Background radiation of all sorts affects all imaging technologies — even our eyes. Our eyes are very sensitive, but at extremely low light levels, the neurons tend to fire randomly. We're using our rods, which see mostly brightness levels, because the color-sensitive cones aren't as sensitive to brightness. I don't know about you, but when I'm in a dark closet lit only by a night light in the adjacent room, my vision gets rather noisy!

It will take a quantum jump to get more than a 48db signal-to-noise ratio (about 15 stops of range) out of a reasonably priced camera system. Going to a totally new sensor technology and higher bit depth processing probably will be required. But then you get into issues of heat dissipation, processing speed, shielding, storage capacity...

I'm confident that at SOME point in the future, some camera company will have a breakthrough moment, and come up with a 72db or 96db signal-to-noise ratio, and we'll have much less noisy performance at ISOs below 51,200 or even 204,800. But that will just encourage the marketing folks to set the maximum to a much higher speed! At some point, it just gets ridiculous.

Meanwhile, I'm comfortable with what we have available now. It's way better than film!

G. Crook Loc: Linden, TX

Wow! A technical, detailed response that is readable AND understandable to an amateur such as myself. Good information. Thank you.

Fotoartist Loc: Detroit, Michigan

Thanks to Farmer for the noise explanation. Signal to noise, motion vs. resistance, energy vs. entropy. Can we eliminate reality's dichotomies, repeal physics principles? In our dreams.

Bushpilot Loc: Minnesota

I find that Lightroom's noise reduction slider is very effective if used carefully.

bclaff Loc: Sherborn, MA (18mi SW of Boston)

Ron Dial wrote:

Noise is caused by heat across the camera sensor. The higher the ISO, the greater the heat, and the greater the noise. Camera Manufacturers have software that tries to deal with noise but it is an ongoing problem

Sorry but heat is not a measurable part of the issue.
Most noise that people find objectionable/noticeable arise from not gathering enough light.

bclaff Loc: Sherborn, MA (18mi SW of Boston)

Yes, the temperature of the sensor and the immediate surroundings have an effect on dark current.
(Extra electrons that are mistaken for signal.)
This is why some astro-photographers use cooled camera systems.

DirtFarmer Loc: Escaped from the NYC area, back to MA

burkphoto wrote:

...I'm confident that at SOME point in the future, some camera company will have a breakthrough moment, and come up with a 72db or 96db signal-to-noise ratio, and we'll have much less noisy performance at ISOs below 51,200 or even 204,800. But that will just encourage the marketing folks to set the maximum to a much higher speed! At some point, it just gets ridiculous...

I believe that the high ISO noise is here to stay. We're getting pretty close to the quantum limit in sensors so at some point there are just too few photons to count in order to get a good signal to noise ratio without long exposures. A significant part of the improvement in current cameras' high ISO performance is signal processing by the onboard computer. My D5 can get reasonable performance at ISO 50K (I don't believe the ISO level is really described by more than one significant figure at that point). Having tried some tests (not rigorously scientific) up to ISO 3 million it is my opinion that ISO above 50K-100K on that camera is marketing hype. It's there for emergency purposes and bragging rights only.

I'm confident that signal processing will give us some more performance at high ISO levels but I don't believe it will go very far without making significant assumptions about the image being processed. And I'd rather have the signal processing done after the raw file is written rather than before. That will give us the option of changing the assumptions.

Ron Dial Loc: Cuenca, Ecuador

Yes, I agree. Pushing the ISO on a digital camera is like pushing film in the old days. The more you push it, the granier it became.