Just how 'standard' is ISO from camera to camera

dinners

dinners

In Memoriam
Messages
15,745
Name
Phil
Edit My Images
Yes
I didn't want to harp back to the old film ASA 'standards' - this is more of a general question regarding ISO and DSLRs......

How 'standard' is ISO from one DSLR compared to another ?

I'm not talking about how well one camera performs compared to another at a given ISO but whether ISO is constant from one camera to another.

Example: Shooting birds in aperture priority mode.....I know I want to use 400mm, I know I want to be at f6.3 for the depth of field I require and (for arguments sake) I know I want to keep the sutter speed to 1/400

Camera 1

With the focal length and aperture set in stone, I'll adjust the ISO till my shutter speed is 1/400. It may result in ISO 1600

Camera 2

With the focal length and aperture set in stone, I'll adjust the ISO till my shutter speed is 1/400. Will it also result in ISO 1600 or might it give me ISO 800 or ISO 3200

I appreciate there are some differences in metering from camera to camera but asuming everything is set the same, same lens, same shot, same light, same time, same everything. - would I expect one camera to throw me a different resulting ISO to another ?
 
Last edited:
No! ISO means that it's a standard to which cameras (in this case) should comply. I dare say that there are plenty of examples of cameras that are adrift but in theory, ISO ### means just that.
 
Which was my thinking however I often chop and change cameras and find myself perplexed regarding the resulting ISO.
 
Last edited:
dinners said:
Which was my thinking however I often chop and change cameras and find myself perplexed regarding the resulting ISO.
Click to expand...
I think that's probably because the situations aren't exactly comparable. Or because the two cameras perform evaluative metering differently.

Two cameras metering the same subject (using spot metering) in the same lighting will recommend the same ISO values.
 
Natural light is constantly changing but the auto aperture we usually view it through is pretty damn good at its job! In exactly the same light, shooting exactly the same subject, the exposure should be exactly the same. Some cameras may choose lower ISOs because their high ISO performance isn't deemed good enough (note the may - not played to check!)
 
DPReview used to measure 'real' ISO levels for a given exposure and there were some discrepancies. Not sure they do it now though.
 
Cheers folks

So given the cameras I use (1D3, 1D4, 40D) I may find variation in terms of the ISO figure the camera gives me when all other variables are constant.
 
dinners said:
Cheers folks

So given the cameras I use (1D3, 1D4, 40D) I may find variation in terms of the ISO figure the camera gives me when all other variables are constant.
Click to expand...
That's certainly possible Phil and there's another inconsistency that remains hidden. The camera does not necessarily use the ISO that it shows and subsequently reports in the EXIF. Some lenses (ie,EF 85/1.2) cause an unreported ISO offset to be applied so that the exposure parameters appear to remain linear.

Bob
 
Canon Bob said:
That's certainly possible Phil and there's another inconsistency that remains hidden. The camera does not necessarily use the ISO that it shows and subsequently reports in the EXIF. Some lenses (ie,EF 85/1.2) cause an unreported ISO offset to be applied so that the exposure parameters appear to remain linear.

Bob
Click to expand...

Interesting Bob - I didn't know that!
 
They do sometimes vary, but not by much - maybe a third of a stop max. Fuji X-Trans comes to mind for a bit of over-stating ISO.

It's not easy to test that small amount of difference. Must be same subject, same lighting, same lens, same aperture and same post-processing - everything identical. If you're seeing more than a third of a stop difference, there's probably another cause. A common one is lenses that have a different T/stop to the marked f/number, and mechanically inaccurate aperture settings.
 
dxo tests very clearly suggest that manufacturers apply pretty much any ratings they like to their marketing advantage. It is always nice to boast about the highest ISO, when it isn't quite there, and as a consequence "ISO100" can be as low 60-70. APS-C bodies may be the most "uprated" to make them look more competitive. Copies of the same model should be pretty consistent.
 
Canon Bob said:
That's certainly possible Phil and there's another inconsistency that remains hidden. The camera does not necessarily use the ISO that it shows and subsequently reports in the EXIF. Some lenses (ie,EF 85/1.2) cause an unreported ISO offset to be applied so that the exposure parameters appear to remain linear.

Bob
Click to expand...

isn't 85mm more like f/1.3 (and the 2.8s are sometimes 3.0-3.1+, etc) with some heavy vignetting?
 
daugirdas said:
isn't 85mm more like f/1.3 (and the 2.8s are sometimes 3.0-3.1+, etc) with some heavy vignetting?
Click to expand...
It's not quite so simple Daugirdas. The efficiency of the sensor's micro-lenses becomes "unlinear" with wider apertures. An example....The 1Ds3 was about 2/3 of a stop with the 85L whereas the lower pixel density and larger micro-lenses means it loses just under 1/2 stop on the 1Dx.

Bob
 
Thanks for the feedback folks

I'm sure I could test this out and I may well give it a go if only to satisfy my own curiosity.

I'm not really one for conducting 'controlled' tests so it's nice to quickly get an opinion from those who are more clued up on the technical side of things.
 
dinners said:
Thanks for the feedback folks

I'm sure I could test this out and I may well give it a go if only to satisfy my own curiosity.

I'm not really one for conducting 'controlled' tests so it's nice to quickly get an opinion from those who are more clued up on the technical side of things.
Click to expand...

If you don't do highly controlled tests, you're much more likely to see some other variable affecting things rather than ISO discrepancy.

I forgot to mention vignetting - optical darkening of the image towards the edges and corners. It's the number-one culprit, so comparisons should only be made in the dead centre of the frame. Super-wide zooms are worst, and often fast aperture primes are not much better, and on full-frame exposure can be as much as three stops darker in the extreme corners at lowest f/number. Some lenses don't get decently even across the frame until f/8.

Example: Canon 24-105L at 24mm http://www.the-digital-picture.com/...LensComp=355&CameraComp=9&FLIComp=0&APIComp=2
 
HoppyUK said:
If you don't do highly controlled tests, you're much more likely to see some other variable affecting things rather than ISO discrepancy.
Click to expand...

Fully accept that Richard

Of course there will always be variables at work in real world shooting but it's amazing how you get a feel for what to expect.

Whilst I accept it would require controlled tests to confirm the theory, it's interesting to know (all things being equal) it's possible.
 
You must log in or register to reply here.
Back
Top