Noise: ISO push vs. PP push

[Stosh]

Yes, a JPEG has a tone curve applied, it is not a simple truncation of the linear data. There is, therefore, no fixed relationship between the scene brightness range that a JPEG represents and the 8-bit nature of a JPEG.

Best,
Helen

OK, you're right that any random 8-bit JPG image does not necessarily correspond to any real-world dynamic range, but that's because we as image processors make those 8 bits represent whatever we want.

Conversely, the camera outputs a RAW file that DOES correspond to a real-world dynamic range. This we can't change that because it's written into the firmware of our cameras. My whole point is that this RAW file has more dynamic range than we're used to seeing in our images. When our 8-bit calibrated monitors try to display a RAW file, they look extremely dull from low contrast because we're squeezing 10, 12, or even 14 bits of dynamic range into 8.

Yes, I now realize and agree that curves are applied to the data, so it's not nearly as simple as truncating bits, but why can't we select good real-world data from the top, middle, or bottom of this wide dynamic capture?

 

[Stosh]

I'm telling you, Stosh... you can look at sample shots all day long... but the conceptual under-pinnings are all you really need to know to understand the answer to your question.

Well that's an angle I never heard before. Most people say, "we can talk theory all day, but when it comes down to a real world example.........."

Thus, it's most definitely not "multiplying the same internal data". The two data sources are much different. One is the original signal produced by light striking your sensor, the other is an under-exposed, low-data, digital record of that signal.

Well now we're getting into semantics. Both instances do come from the exact same signal if go back far enough in the process (the signal coming directly from the sensor before the amplifier).

I do get what you and others are saying - apparently the amplified analog signal can be more accurate because the digital signal is very low bit depth at the dark end of the spectrum. I think the moral of this whole thing is it made little difference at 2 stops, but significant difference at 4 stops and above. I'd like to do some examples myself, but I fear I have exhausted everyone's interest in this subject lol.

 

[JG_Coleman]

Well that's an angle I never heard before. Most people say, "we can talk theory all day, but when it comes down to a real world example.........."

Lol... no doubt that real-world investigation can be a good thing. It's just that, in this case, the known way in which a camera handles signal and the known way in which exposure is boosted in PP will produce the same predictable results every time.

 

[LokiZ]

Here are the three things I know...

- With my camera the higher the ISO the more noise.

- With my camera the less light available the more noise.

With me so far?

- Thirdly with my camera the degree of difference noise wise with my ISO settings (100 through 1600) is minute compared that of the difference noise wise when comparing my max metered under exposure to my max metered over exposure.

My test went about it a little differently:

I took a base image correctly exposed at ISO 400. I under exposed a 100 ISO image by -2 on the in camera meter and over exposed the 1600 ISO by +2. In PP I was able to acceptably lower the 1600 ISO image and match my 400 ISO image but adjusting the 100 ISO up was unacceptable and noise wise they did not match.

Flipping the test around I under exposed a 1600 ISO image by -2 on the in camera meter and over exposed the 100 ISO by +2. In PP I was able to more then acceptably lower the 100 ISO image and match my 400 ISO image but adjusting the 1600 ISO up was very unacceptable and noise wise they did not match.

So... In conclusion the effect of exposure has a much greater consequence on noise in an image then that of the the ISO setting on a camera. That in turn makes it possible for images to be made where higher ISO value images have less noise then that of lower ISO images (after PP adjustments have been made.)

Basically if more light comes in on a higher ISO shot then does on a lower ISO shot it is then possible to have less noise in the high ISO shot through the use of PP to decrease the exposure.


Edit: line 1 two changed to three.

 

[Garbz]

but I'm still interested in the why.

What about my noise reduction discussion? Doesn't the camera automatically apply more noise reduction when you use a higher ISO? And did you apply that same increased noise reduction to the PP pushed image to make it a little more fair?

Noise reduction can only come into it if I shot in JPEG. I didn't. The noise reduction was equal on all parts of the image due to camera corrections not being applied in RAW.

The why is still simple. ISO is increasing the sensitivity of your light capturing device before conversion to digital. Increasing brightness is a linear increase on the data after it has been converted to digital. The former has very complicated non-linear relationships between noise and increased photo conversion. The latter has a linear relationship. Double the brightness, double the noise.

Look at the word "Mathematics". Notice how much sharper it is in the ISO 100 shots? Both methods of pushing ISO 100 were sharper than the ISO 400 shot. My guess is back to what I said about noise reduction. The camera (or some PP) applied more noise reduction to the ISO 400 shot automatically. The noise reduction was not applied to the ISO 100 shots. So although the noise reduction certainly did its job of reducing noise in the dark areas, it significantly wounded the brighter part of the image by reducing sharpness.

So back to what I suggested before - if you were to somehow find the amount of noise reduction that got applied to the ISO 400 shot, then also applied it to the pushed ISO 100 shot, would it look similar?

No sorry, the real explanation is far more sinister. The camera was not on a tripod, and the focus was not locked, the camera attempted to refocus every time it takes a picture. In some cases it pushed it out. Notice how the sharpness effect is exactly the opposite in the ISO1600 shots as the ISO400 shots? Well... my bad

 

[Stosh]

The why is still simple. ISO is increasing the sensitivity of your light capturing device before conversion to digital. Increasing brightness is a linear increase on the data after it has been converted to digital. The former has very complicated non-linear relationships between noise and increased photo conversion. The latter has a linear relationship. Double the brightness, double the noise.

Thanks Garbz. This is what I've surmised over the previous few days discussion, but you've summarized it nicely in a couple of sentences.

No sorry, the real explanation is far more sinister. The camera was not on a tripod, and the focus was not locked, the camera attempted to refocus every time it takes a picture. In some cases it pushed it out. Notice how the sharpness effect is exactly the opposite in the ISO1600 shots as the ISO400 shots? Well... my bad

OK, that actually makes much more sense.