High ISO experiment

[jaapv]

Advertisement (gone after registration)

Encouraged by an earlier thread (too lazy to use search), I did the following:

1. M8 at staright ISO 2500

2. M8 at straight ISO 2500 100% crop

3. M8 at ISO 160, exposed for ISO 5000 and recovered in C1

3 The same 100% crop.

 

In both cases I used the noise surpression in C1.

 

My conclusion: standard ISO 2500 is not bad, but the " recovery trick" produces better files at ISO 5000 with repectable detail and surprisingly little noise. Colour is better too.

 

 

 

 

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[Skippy Sanchez]

try:

cleaning your sensor

cleaning your wall

cleaning my screen

 

seriously,

 

i see little difference, except that @ iso 5000 you get an extra stop of exposure, right?

[jaapv]

The wall I fear.

Look at the speakers under the screen. The detail is far better at 5000. It probably would be even better at ISO 2500 exposed at 160 ISO.

[luigi bertolotti]

REALLY EXPOSED AT 5000 ? Woah ... the residual noise is "dirtier" for the colors (see the crops of the plastic cases) but sharpness is even better than at 2500 (or were you handeld with half the exp. time ?)

[Guest Walt]

Is there not both much more chrominance noise and much more detail in the C-1 recovered shot? Was the non-C1 shot out of focus (for example, on the monitor perforations)?

[Skippy Sanchez]

wow! thanks for pointing that out. i'll be rethinking my exposure procedures from here on for a few tests of my own.

 

this is where a hand-held meter really comes in handy. today i had an assignment at my local airport (attatched) and shot a decidedly underexposed pick of a guy picking up his luggage @ 1250 iso. the shadows on the luggage (lower right) was very grainy. this trick would have probably helped

[jaapv]

Advertisement (gone after registration)

Was the non-C1 shot out of focus (for example, on the monitor perforations)?

 

I don't think so, when I look at the way DOF runs. They were both developed in C1, the same f-stop, one with the camera at ISO 2500, one with the camera at ISO 160, but exposure time half ( first 2.0/ 125, second one 2.0/ 250) = pushed to ISO 5000

[Jamie Roberts]

Three qualifications for this approach:

 

1) when you need highlight detail (these are all greys, don't forget) then shoot higher ISO

 

2) when you can't handhold the shutter, then shoot higher ISO

 

3) always remember that when you shoot 1250 or 2500, you are losing a stop (or more) of exposure latitude due to noise. So lower ISOs will definitely push up shadow details more satisfactorily (with NR etc...). But see 1: if you way underexpose highlight details at ISO 160 then push them not only will you not get the same look as the ISO 1250 stuff you'll be hard pressed to regain detail.

 

Dunno why exactly 1 & 3 work, but they do...

[jaapv]

Three qualifications for this approach:

 

1) when you need highlight detail (these are all greys, don't forget) then shoot higher ISO

 

2) when you can't handhold the shutter, then shoot higher ISO

 

3) always remember that when you shoot 1250 or 2500, you are losing a stop (or more) of exposure latitude due to noise. So lower ISOs will definitely push up shadow details more satisfactorily (with NR etc...). But see 1: if you way underexpose highlight details at ISO 160 then push them not only will you not get the same look as the ISO 1250 stuff you'll be hard pressed to regain detail.

 

Dunno why exactly 1 & 3 work, but they do...

 

1 I agree. One is pushing the limit of the dynamic range of the camera. I tried this on a sunset. In one word: no.

2. In general yes, or get a tripod, but the point is, shoot at low ISO and underexpose.

3.Much to my surprise the detail was superior in the forced low-ISO shot.