ISO 160 vs. 100

[Marsu42]

On the 1D X there's really not much difference at any ISO from 100 - 400.

Same on 6d, "thanks" to Canon's read noise problem there's hardly any reason to shoot at base iso unless the shutter speed has been maxed out.

I did some test images on my 5D2 by taking a black frame with the shutter cap on at each ISO and then increasing them by the same amount to more easily see the noise. I always shoot at increments of 160.

Interesting, esp 640/800 ... but I'd say it might be a bit different on newer cameras. Personally, I'd be hesitant as it's not only about noise with lens cap shots, but about color fidelity and post-processing elasticity which afaik are better at lower base iso settings even on full frame sensors.

 

[ejenner]

I thought this craze had died an appropriate death, but anyway, OK, true read noise is lower at ISO160. But I for one don't tend to take many shots of my lens cap.

Instead of asking the question and getting a bunch of 'facts' that may or may not be relevant to your actual shooting, why not test it yourself with the type of photography you do?

If you can't tell the difference, then don't worry about it. If you can, use what you think is best.

BTW the lens cap shots above show that ISO 320 has lower read noise than ISO 100. Now if you can't tell the difference between those in real photos, then just shoot at whatever ISO you want.

 

[adventureous]

It seems popular in video. Here Vincent Laforet shows the setup for 5dII and 5dIII.

http://blog.vincentlaforet.com/2012/03/12/setting-up-your-canon-5d-mkii-and-mkiii/

 

[martti]

Am I misuderstanding or does that curve mean that on 5DIII you are OK up to 800 ISO after which the noise starts to grow? That would certainly simplify things...

Now, I do not see any standard deviations here nor 'n' so I take it this is the data of one single camera of each brand. The slight curvatures may not be significant at all.

 

[Marsu42]

Am I misuderstanding or does that curve mean that on 5DIII you are OK up to 800 ISO after which the noise starts to grow? That would certainly simplify things...

Not quite, lookie here for the data: http://www.sensorgen.info/CanonEOS-5D-Mark-III.html

But for practical purposes and non-lens-cap shots, on recent ff Canon cameras I'd say the "It's really a wash" area is from 100-400, that's why I'd rather shoot with 400 and higher shutter speed for anything that moves. You can extend this area to 800 if you accept a little drop in color fidelity. Just my experience, mind you.

 

[martti]

Of course that makes sense.
Still, I have a bit of a hard time looking at a curve where there is no mention of the methods how it was got.
Is it theoretical or is it measured. If it is measured, how many cameras were tested? What about the statistics? Was the value achieved at 160 ISO statistically different (p<0.05) from the value at 400 ISO?

And if it was, so what?
Personally, I have noticed that a monopod makes the pictures sharper no matter what lense I am using when shooting in available light inside. Which sort of lines up with your experience, doesn't it.

 

[Marsu42]

Still, I have a bit of a hard time looking at a curve where there is no mention of the methods how it was got.

It's data from DxO, and general opinion is that their pure *measurements* are just fine, other than the *aggregated* variables and the interpretations of them (which favor Sonikon).

Personally, I have noticed that a monopod makes the pictures sharper no matter what lense I am using when shooting in available light inside. Which sort of lines up with your experience, doesn't it.

Yes, imho the shutter speed or other means of camera shake reduction is *hugely* underestimated. IS helps to get "good" pictures, but it cannot compensate fast enough if you re-frame so you get more sharpness at 100% crop with other means like a clumsy monopod.