1/3,1/2 or full stops?

I

ian-83

Messages
1,567
Edit My Images
Yes
When setting up my camera for f stops, shutter speed and ISO is it best to use a full stop,1/2 or 1/3 setting?
 
The problem with increments of full ISO stops is that they're not really 'real' increments, just adjusted under/over exposed full stops that are dealt with in-camera. These half and third stop increments often produce more noise than the full stops because of this.

Otherwise, I don't have an issue with third/half stops in aperture and shutter speed because these are physical increments that are real. I find increments of aperture stops useful for fine tuning when using flash.
 
The important thing is that you know how it's set-up for easy adjustment/compensation by "counting clicks".

I usually have ISO set to full stops, shutter speed and aperture set to 1/3rd stops.
 
specialman said:
The problem with increments of full ISO stops is that they're not really 'real' increments, just adjusted under/over exposed full stops that are dealt with in-camera. These half and third stop increments often produce more noise than the full stops because of this.
Click to expand...

AFAIK it's only the lower end Canons that do that, Nikon uses "real" 1/3 stops.

There's not really much on the net about it, and there's mention of it in my D700 manual.
 
I find it easier to work in half stops particularly when it comes to shutter speeds. I do like to work in full stops for aperture and iso though. Something to do with OCD I think!
 
I much prefer using whole stop increments for aperture, shutter, and ISO and luckily my D700 supports this as I seem to remember my old D70s only allowed 1/2 & 1/3 stop increments.
I mainly tend to follow the "sunny 16" rule when assessing exposure and working with whole stops makes this quite easy - f/16 for sunny, f/11 bright overcast, f/8 overcast etc.
The D700 has a "native" iso of 200 so I will mainly use this setting with 1/250 shutter for well lit outdoor work as sunny 16 works best using a shutter speed which most closely matches the iso.
 
Think I light try using full stops as currently my camera is set on 1/3 stops.
 
redddraggon said:
AFAIK it's only the lower end Canons that do that, Nikon uses "real" 1/3 stops.

There's not really much on the net about it, and there's mention of it in my D700 manual.
Click to expand...

Read somewhere that its only the 1D/1Ds series that has native third stop ISOs. Could be wrong... :)

Think I need to do some tests tomorrow in the studio and see how adjustment affects noise.... :)
 
Last edited:
1/3 stops for everthing; Shutter and Aperture set for 1/3 increments because there are some shutter speeds I like for particular situations such as 1/640 sec for rugby

1/3 for ISO; my first dslr didn't have the option to set ISO in whole stops, so I got used to 1/3 stop increments, and back in pre digital days, film had 1/3 stop increments eg Kodachrome 64 rated at ISO80.
 
I use 1/2 stops as the 1/6th of a stop difference there is between 1/3 and 1/2 isn't worth worrying about and 2 clicks per stop is easy to work out.
 
specialman said:
The problem with increments of full ISO stops is that they're not really 'real' increments, just adjusted under/over exposed full stops that are dealt with in-camera. These half and third stop increments often produce more noise than the full stops because of this.

Otherwise, I don't have an issue with third/half stops in aperture and shutter speed because these are physical increments that are real. I find increments of aperture stops useful for fine tuning when using flash.
Click to expand...

Pat, Do you have a reference for this statement ?...

I always felt that the 1/3 and 1/2 stop adjustments on my Canon were noisier than full stops adjustments but I couldn't find reference to it anywhere on the net. Sorry for jumping in.
 
Tugster said:
Pat, Do you have a reference for this statement ?....
Click to expand...

I have to admit that I don't - I read it on a many occasions and the logic behind it made a sort of sense.... if that logic is actually founded on a sound basis of knowing how the camera manufacturers make their native ISOs. That's the crux.

As said, I'm gonna do some tests on my Nikon to see if there's an actual truth behind it, although any results will solely be based on findings from D2x, quite an old camera. Personally, I've always thought my third-stop ISO increments seemed noisier but that could have just been down to overall exposure issues at the image-capture stage and not because it is inferrer to a standard ISO stop (100/20p0/400 etc).

Would be interesting to see how newer models like the D7000 work at these 'non-native' ISOs (if they are non-native) and how these ISO increments react to pushing the exposure in PP.

It could well be a long-standing internet myth and one that has fooled suckers like moi... :)
 
Last edited:
I find 1/3 increments very useful on digital for eeking out the best compromise in difficult situations. Very rarely will I switch ISO to 1/3 increments though unless it's really tricky.

On film it matters not, My Zuiko lenses work only in full stops for aperture and manual shutter speeds on every film camera almost are always full stops. I can have fully stepless spees in aperture priority mode if I want ultimate control though, although the display is in 1/3rds. My pentaxes have half stops on the aperture but full stops only for shutter speed (ME, MX etc)) and full stops only in manual 1/3 in Aperture priority (LX)
 
I use full stops for ISO and half stops for shutter speed and aperture. For me it is much more valuable to be able to quickly change settings than it is to have really fine control. In particular partial ISO stops are a gimmick in my opinion.

Alex
 
Hmmm many different opinions and options seems like it's down to whatever works the best for the person holding the camera. Think I might try the full stops for ISO and maybe 1/2 for shutter and aperture.
 
Nawty said:
I use 1/2 stops as the 1/6th of a stop difference there is between 1/3 and 1/2 isn't worth worrying about and 2 clicks per stop is easy to work out.
Click to expand...


this for me. Full stops aren't quite fine enough for aperture (which I shoot in most of the time) and third stops are 50% more clicks to go through.

I normally increment my ISO in full stops when I set it manually (until I hit 2000, then I increment with smaller jumpss) but I also let the camera auto ISO as it sees fit a lot of the time and will get 1/3 stop jumps.

Thanks
Rick
 
I have been reading a lot on this matter & i am still not sure what would be best,as my 5D does not have auto ISO,unless i am mistaken on that & it is buried somewhere in the settings,but i don,t think so.
There are a lot of working pros that i have read about them using full stops,probably a lot that don,t as well.
So i think i might have a play around with it as probably a lot of us will do to work out what is best for each of us.:):thumbs:
 
'Tis easy to make your mind up in these digital days. Just point your camera at something, shoot a series of shots at different and mid ISO's... and view the results.

My GF1 and G1 are capable of auto ISO and have the mid stops and I find them ok. Certainly they're a help between 800-1600.
 
I have to admit, I'm not convinced about this ISO full stops theory, surely it is all just turning up the gain on the signal and from an amplification point of view it is all just multipliers.

I'm prepared to be proved wrong though, I'd just like some proof.
 
I think it must be down to how it's implemented. I've read reports of various mid settings being not so good in some cameras and I've also read reports that say that various intermediate settings are the best to use in other cameras.

I think it's probably best to try all the setting and then decide.
 
I used to avoid non full stops like the plague BUT over the past few months I have been using auto ISO for birds which automatically uses 1/3 stops (irrelevant of what you have the Camera (7D) set to) and have to say that I can see no evidence of the 1/3 stops causing more noise - I have now changed my mind completely over this issue. For bird photography I know rate auto ISO (with auto 1/3 stops) as the best thing since sliced bread!. Maybe over the years the in Camera processing for these 1/3rd stops has gotten a whole lot better.
 
Nawty said:
....surely it is all just turning up the gain on the signal and from an amplification point of view it is all just multipliers......
Click to expand...

You'd have thought so. I guess it's how the gain is turned up that dictates whether there has to be a software-based workaround. If the gain can only be turned up in set increments - i.e. the base gain equals ISO 100 but the next available gain 'notch' equals ISO 200 - then obviously to get ISO 160 then it's impossible to get that in reality.

Anyway, I sid I'd do a test to show how the increments look on my D2x. I've taken a test shot to show you to overall image and have then done a big drop into an area where there is shadow, plus a range of tones.

Test shot:
test_shot.jpg


ISO 100
ISO_100.jpg


ISO 125
ISO_125.jpg


ISO 160
ISO_160.jpg


ISO 200
ISO_200.jpg


ISO 250
ISO_250.jpg
 
ISO 320
ISO_320.jpg


ISO 400
ISO_400.jpg


ISO 500
ISO_500.jpg


ISO 640
ISO_640.jpg


ISO 800
ISO_800.jpg


I'm not really sure what to make of the findings. The increase in noise seems gradual as the increments go up, with no blips that standout.

I'm going to see how underexposed shots handle an exposure increase in PP, which should really show a noise increase if there is one. I'll do this later today....

BTW, these shots have no NR applied. :)
 
Last edited:
Thanks for taking the time to take and post these pictures Pat, to my eyes the noise increases in direct correlation to the ISO. That the ISO 160 looks slightly less noisy than the ISO 200 image would seem to show that the ISO isn't bumped to a full stop and then brought back down in afterwards.
 
Nawty said:
Thanks for taking the time to take and post these pictures Pat, to my eyes the noise increases in direct correlation to the ISO. That the ISO 160 looks slightly less noisy than the ISO 200 image would seem to show that the ISO isn't bumped to a full stop and then brought back down in afterwards.
Click to expand...

No probs - it's proved me wrong about my camera. Plus, it's good to do a test on your camera, as it does clear up just how far you can go with your camera before the noise gets too bad.

I didn't post my ISO 1000, 1250, 1600 and 3200 results, just to keep the number of pics down . I did take them so might post them up, just to show the overall noise increase on a D2x if anyone is interested :)

It'd be interesting to see other results from other cameras.
 
You must log in or register to reply here.
Back
Top