ISO Assignment... please help

[HeyFrankie]

Hi All,
I am new to this forum, but hoping you can help me out a little

I am currently studying a diploma in photography and am doing an assignment on ISO, so I had to take a series of photos outside with good tonal range same shot at different ISO, doubling with every shot until I reached cameras max. Then repeat inside in low light photographing something low contrast, then I have to discuss what happens to the images.

In the first set of photos my main observations are that as the ISO increases, contrast seems to increase (shadows become slightly darker and more noticeable) up until the second max ISO for my camera when contrast decreases dramatically and the photo starts looking a bit washed out.
Also as the ISO goes up the image noise level increases, introducing a pixelated/grainy look, with random colour pixels in the shadows.
... does this make sense? I am a bit confused about the relationship between ISO and contrast.

Then for the inside photos I have to say how the increase in ISO differs to the outside shots.
So for those the decrease in contrast and the image noise and randomly coloured pixels seem to be more noticeable at a lower ISO then of the images from outside.

Can someone tell me, am I on the right track here?

 

[Krisstiffer]

In a word - yes!

Standard thing is that on any given camera higher ISO = higher levels of noise. Maybe one day camera manufacturers will invent a way of stopping this but as of today what you've seen is standard stuff.

As to the contrast thing - the differences are probably down to the different lighting you describe. Again in general as the ISO goes up the dynamic range decreases. What you're probably seeing (but happy to be corrected) is that outside, with high tonality, the noise is giving you the impression of more contrast until you get to a threshold where the lower dynamic range of high ISO images takes over. The indoor images are unlikely to have the initial dynamic range in the first place so as the ISO goes up the noise is more likely to appear just as noise......

I think........

 

[Phil V]

As Chris said, your findings are right. The only issue I have is the assumption people have that high ISO's should maybe one day be as good as low ISO's.

The whole assignment strikes me as a convoluted way to teach you that you should always shoot at the lowest ISO that'll get the shot.

 

[ernesto]

It is a safe assumption that high ISOs will be as good as current low ISOs. However, as sensors improve they may improve at 100 as well as 12800.
Technology is bound to get to the point where sensors are redesigned and can work with less light with no detectable degradation in quality, just a matter of time.

 

[Phil V]

It is a safe assumption that high ISOs will be as good as current low ISOs. However, as sensors improve they may improve at 100 as well as 12800.
Technology is bound to get to the point where sensors are redesigned and can work with less light with no detectable degradation in quality, just a matter of time.

I don't understand the above, those two statements are contradictory.

Do you believe that as sensors improve, they do so across the range - or do you believe that sensors will one day be as good at high ISO's as they will be at low ISO's?

Whilst we can lessen the impacts of physical limitations, we can't rule them out completely. Therefore more light = cleaner images, because amplifying the light will add noise, we'll get better at amplifying the light cleaner, but that works across the ISO range. Why would any camera manufacturer give us high ISO's as good as their low ISO's (they could do this at any point), when their competitors can easily promote their low ISO files as being better.

The manufacturers have to produce a camera that works as well as they can make it across the range.

 

[ernesto]

Whilst we can lessen the impacts of physical limitations, we can't rule them out completely.

we can with a complete redesign of the whole sensor concept. You are limiting your thinking by basing it on the methods/technology used today

I can see why the comments looked contradictory, because they were
The first was backing up anothers assumption and included 'may', the second is what I think myself.

 

[Phil V]

we can with a complete redesign of the whole sensor concept. You are limiting your thinking by basing it on the methods/technology used today

I can see why the comments looked contradictory, because they were
The first was backing up anothers assumption and included 'may', the second is what I think myself.

It's not design, its Physics.

It doesn't matter what technology we may invent to capture light, it's difficult to believe how more light isn't easier to capture.

Thinking out of the box is one thing, but as photographers we have to remember all we're doing is manipulating light. And light behaves in ways so fixed they're defined as laws.
The only chance we have of changing that is discovering an energy source (from the dark) that we can capture with the same tools we use to capture light.

(edit)
Forget that, if it's not light we're capturing it's no longer photography.

 

[Pookeyhead]

I think the point ernesto is trying to make is that while you are right, physics is physics... we may reach a point where the losses, while measurable, are so low that they are not visually perceptible. Take Hi-Fi as an example... back at the turn of the 20th century, a recording of a violinist on a Bakelite 78rpm record was awful, yet was the height of technology. Noise levels were high, dynamic range was poor, frequency response was limited and not flat. Today, all those parameters are still measurable.. we can measure the difference between reality and the recording, but they are so low that they are just numbers on a spec sheet.

If you asked someone 20 years ago whether it would be possible to build a home computer that could calculate Pi to 1 million decimal places in 5 seconds they'd have laughed in your face, but now it's perfectly possible.

 

[swanseamale47]

Back to the original question.
Your own findings sond about right, I would ask thogh how sure are you that the contrast changes you see in the outside photos are due to the ISO changes and not changing lighting conditions?

 

[ernesto]

I think the point ernesto is trying to make is that while you are right, physics is physics... we may reach a point where the losses, while measurable, are so low that they are not visually perceptible.

Yes, this is my point exactly. The ways and means used to capture images in 50 years time will be incomparable from those today.

 

[GHP]

Whenever folks talk about ISO, the conversation always seems to concentrate on the high ISO range, which I fully understand is important in the real world and why good high ISO performance is so desirable.
But what happens to noise levels when you drop the ISO to BELOW the "native" ISO ?
My D700 will shoot at 100, although I've never tried it. Does noise increase when shooting at the Low1 setting?
Just curious really..