dpreview iso comparison

DuncanDisorderly said:
I've uploaded the full set of test images to a gallery.
You can download full size JPEGs if that floats your boat.
http://www.wild-landscapes.co.uk/Other/2012/ISO-Tests/25573977_QL64L3
I think I've worked out how to turn off sharpening on linked images. The originals will definitely be unsharpened.

Right - Here's the crops.
One is 1/1600s, the other is 1/3s - all other settings are unchanged.
If you have been following this thread, then it should be obvious which one is which.
The images were taken in RAW and processed using LR4 with no tweaks or settings.

It's pretty clear that ISO 12800 is never going to make pixel peepers happy.
But I've printed ISO 12800 to A3+ and the prints look great, so this level of noise works for me!

As for differences between the two?
Sure the colour is different, that's what time of day does; but I don't think it is significant here.
Noise and fine detail look spookily similar and well within margins for test conditions.

No doubt my experimental technique is going to come under attack.
I'm up for it - fire away :D
Click to expand...

Excellent test Duncan.

Pretty much proves speed (bar long exposure which is another debate for another day) does not change the amount of noise.

We still daylight vs artificial light though (assuming this was an original argument).
 
Phil Young said:
Excellent test Duncan.

Pretty much proves speed (bar long exposure which is another debate for another day) does not change the amount of noise.

We still daylight vs artificial light though (assuming this was an original argument).
Click to expand...

I must admit - it did occur to me to colour correct the images and see if that made a difference!
It's late, and I suspect it's not going to do anything dramatic :help:
 
No test from me as speed has already been proven...but I did want the challenge of seeing if I would use the highest ISO setting my camera has to offer...so here's a shot at 25600 and then I applied some colour changes and selective NR. I think I managed to keep all the detail needed whilst giving the overall look of something I would expect 3-4 stops under the actual ISO. I'm sure if I spent longer than 10minutes the result would have been much better...but I'm tired and going to bed!

I'm keen to see what others can do with their highest ISO setting...

nothing_done.JPG

Retouched1.JPG
 
Years ago I did a day shooting with the highest ISO my 20D would go to and the results were perfectly ok.

One thing that I try to keep in mind is that noise at 100% is one thing but noise in a whole image or indeed in a print is yet another thing, and of course the final image size matters.

My 5D and G1 only go to 3200. I'd use the 5D at max without hesitation.

IMG_9506c.jpg


IMG_9503c-1.jpg
 
I decided that this sadness which has overtaken me is more important than sleep (even though I have a 7.15am client to train!!).

I wanted to see how close I could get to ISO 1600 and this is my best attempt tonight.

I can confirm several things...

1) The D7000 is an amazing performer at ISO 1600!
2) I can't make a 25600 ISO image look like 1600, the colour rendition is very hard to reclaim and neither can fine details.
3) 25600 is definitely usable if you are prepared to work on it in PP.

ISO 1600 (No PP just RAW conversion)
1600.jpg


ISO 25600 (PP)
256001.JPG


Not bad though!
 
Last edited:
Very good. I prefer the look of the coin in the 2nd image.
 
Phil Young said:
Pretty much. But just to be clear:

It shouldn't = more noise but the noise is more apparent.
Click to expand...

to be even more correct. Underexposed = darker = more noise AND THEN if pushed back to proper exposure in PP the noise is even more evident then if you took a properly exposed shot in the first place.

That means you can not win in some situations. Take the grumpy vicar, dark church with a splash of bright light across the dress scenario. You cant lower the dynamic range by adding more light. If you expose for the dress, the rest of the church will be underexposed, if you expose for the rest of the church, you will totally blow the dress, if you go in the middle, you partially blow the dress, and with most camera's you still get a noisy shot
 
Richard King said:
to be even more correct. Underexposed = darker = more noise AND THEN if pushed back to proper exposure in PP the noise is even more evident then if you took a properly exposed shot in the first place.

That means you can not win in some situations. Take the grumpy vicar, dark church with a splash of bright light across the dress scenario. You cant lower the dynamic range by adding more light. If you expose for the dress, the rest of the church will be underexposed, if you expose for the rest of the church, you will totally blow the dress, if you go in the middle, you partially blow the dress, and with most camera's you still get a noisy shot
Click to expand...

Very true Richard.

The above shot saw me about .3 of a stop in dynamic range!

HDR scenes will always be a nightmare for high ISO, the main thing here though was testing for 2 things;

1) does the source of light change the level of noise (ie daylight or artificial)
2) does shutter speed affect the level of noise.

For your situation I have provided a shot with my phone box (pure white) and just speaker (pitch black) and managed to expose both reasonably well applying more NR to the dark areas as per the conclusion the noise there will be more visable.

For me, the thing to take from all of this is now weather to shoot high ISO or not but hopefully; how to deal with it when you have to.
 
You must log in or register to reply here.
Back
Top