5D Mark II - ISO one-stop increments?

I have a 5D Mark II on order and was wondering if I should set the ISO to one-stop increments instead of 1/3 stop increments. That's what I do on my 50D. Is the 5D's ISO any different in that regard?

You can set the ISO increments to 1/3, 1/2 or 1 stop. Just look at CF I-1 and I-2 and customize to your liking.

Does the 5DM2 'push' the in-between settings like the 50D does? Or are the 1/3 stop 'real' ISO settings. I'm probably not wording this very well.

I don't know if the 5D II is the same, but with my 20D and 5D (classic) the 1/3 stop ISOs are, very slightly, noisier than the next up 'normal' ISO (according to published tests). This is due to them being adjusted differently in processing apparently.

So, if you need a little above 800 ISO, you are better going straight to 1600 as it is slightly cleaner than the intervening thirds, as well as giving you a little more for shutter or aperture.

I doubt there is a particularly noticeable IQ loss on 'the thirds', away from lab testing, but there is no benefit to using them compared to the next up ISO.

Unlike shutter speeds for example, where the 'thirds' can be useful. For example, shooting helicopters with the 100-400L (usually at around 400mm) I use 1/80th second as any slower increases the odds of softness due to camera shake / motion blur, any faster and rotor blur is reduced. 1/80th is just right to give reasonable rotor blur and a fair percentage of keepers.

So, I set my 5D for 1/3 stops with shutter and aperture, but full stops with ISO.

Personal preference plays a big part though, which is why you can customise the camera to your choice.

abigailandsam wrote in post #9275246
Does the 5DM2 'push' the in-between settings like the 50D does? Or are the 1/3 stop 'real' ISO settings. I'm probably not wording this very well.

Worded fine to me. That is what I referred to in my answer above, the pushing of the thirds. I don't know if the mk II does it or not, but I would assume so until shown otherwise, as it seems to be the norm on previous bodies.

abigailandsam wrote in post #9275246
Does the 5DM2 'push' the in-between settings like the 50D does? Or are the 1/3 stop 'real' ISO settings. I'm probably not wording this very well.

They are software-generated. Only 1-series cameras have native 1/3-stop ISO increments.

toxic wrote in post #9275860
They are software-generated. Only 1-series cameras have native 1/3-stop ISO increments.

Can anyone verify this? I've heard that the 5d/5d2 1/3 ISO's are native.

As far as I know, everything but 100 ISO is software generated in some way as sensors are not film.

toxic wrote in post #9275860
They are software-generated. Only 1-series cameras have native 1/3-stop ISO increments.

Thanks for that info. I suspected that was the case but wasn't sure.

Sandpiper wrote: "I don't know if the 5D II is the same, but with my 20D and 5D (classic) the 1/3 stop ISOs are, very slightly, noisier than the next up 'normal' ISO (according to published tests)."

That sounds very unlikely - though I'm not saying you are wrong. Where are the tests? I would be very interested to know, or better still see the evidence.

Nigel

HankScorpio wrote in post #9276132
As far as I know, everything but 100 ISO is software generated in some way as sensors are not film.

Well, no, the full stop increments are hardware amplified, not software generated.

puttick wrote in post #9282319
Sandpiper wrote: "I don't know if the 5D II is the same, but with my 20D and 5D (classic) the 1/3 stop ISOs are, very slightly, noisier than the next up 'normal' ISO (according to published tests)."

That sounds very unlikely - though I'm not saying you are wrong. Where are the tests? I would be very interested to know, or better still see the evidence.

Nigel

I've seen personal tests shared on other forums. It's something that you probably would never see outside of a pixel-peeping test, but what the tests also prove is that there's no benefit to the 1/3 intervals, either, except to clutter up the workflow.

Film often came in odd ISOs because film manufacturers wanted to market every bit of film speed they could--an ISO 125 film looked better in the advertisements than an ISO 100 film. But in actual use, it made little difference (and discriminating photographers ran tests to determine their own exposure indices anyway). I consider 1/3 stop ISO increments merely to be a holdover from what people were accustomed to with film.

sandpiper wrote in post #9275256
with my 20D and 5D (classic) the 1/3 stop ISOs are, very slightly, noisier than the next up 'normal' ISO

The 20D doesn't do 1/3 stops. Only full stops.

And from my experience don't see more noise than the next step up using my 40D but I do find the ability to control the shutter speed/Aperture a little more with the ISO very useful.

RDKirk wrote in post #9282691
Well, no, the full stop increments are hardware amplified, not software generated.

Well, yes but anything but the native ISO of a sensor is in some way faked which is what I mean.

HankScorpio wrote in post #9282969
Well, yes but anything but the native ISO of a sensor is in some way faked which is what I mean.

Even ISO 100 is amplified. The difference between ISO 100, 200, 400, 800 and so forth is the amount of amplification. They are all hardware generated. 1/3 stop ISOs have the additional step of either pushing or pulling the exposure via software.

And does anyone have a source for the 1D's having 1/3 stop hardware amplification? I've read the Northernlights page that seems to indicato otherwise.
http://www.northlight-images.co.uk …ras/canon_1ds3_​noise.html