The xxD and xxxD series cameras implement intermediate ISO's by software multiplication, as can readily be checked by observing the pattern of gaps/spikes in the raw data -- for instance, ISO 125 is obtained by multiplying the digitized raw data of ISO 100 by a factor 1.25, and so 25% of the raw levels will be unpopulated (similarly, a pulled exposure will bin adjacent raw levels together when the raw levels are divided by 1.25, resulting in spikes in the otherwise smooth raw histogram).
On the other hand, the 5D and 1 series cameras do not exhibit such spikes or gaps in the raw data at intermediate ISO's, indicating that they are implemented by analog amplification in hardware rather than software amplification in firmware. Measurements of camera read noise indicate that there are two stages of amplification, one for the main ISO's of 100-200-400-800 etc, and another auxiliary amplifier for the intermediate ISO's which boosts the signal by 1.25 or 1.6. This amplifier is sufficiently noisy that there is almost no benefit to using the intermediate ISO's (if one is shooting raw) -- very slightly lowered read noise (when referred to its equivalent in photons/photoelectrons) for ISO 125 and 160, and beyond that the camera does no better than a pushed exposure from the next lowest "main" ISO.
ShaneEngelking wrote:
Actually, it is for exposure level increments, not ISO. BTW, I agree, use only 100, 200, 400, 800, 1600, "3200" ISO's on the 5D. The rest are useless.
ejmartin wrote:
The xxD and xxxD series cameras implement intermediate ISO's by software multiplication, as can readily be checked by observing the pattern of gaps/spikes in the raw data -- for instance, ISO 125 is obtained by multiplying the digitized raw data of ISO 100 by a factor 1.25, and so 25% of the raw levels will be unpopulated (similarly, a pulled exposure will bin adjacent raw levels together when the raw levels are divided by 1.25, resulting in spikes in the otherwise smooth raw histogram).
On the other hand, the 5D and 1 series cameras do not exhibit such spikes or gaps in the raw data at intermediate ISO's, indicating that they are implemented by analog amplification in hardware rather than software amplification in firmware. Measurements of camera read noise indicate that there are two stages of amplification, one for the main ISO's of 100-200-400-800 etc, and another auxiliary amplifier for the intermediate ISO's which boosts the signal by 1.25 or 1.6. This amplifier is sufficiently noisy that there is almost no benefit to using the intermediate ISO's (if one is shooting raw) -- very slightly lowered read noise (when referred to its equivalent in photons/photoelectrons) for ISO 125 and 160, and beyond that the camera does no better than a pushed exposure from the next lowest "main" ISO.
Duncan Gibson wrote:
aero145, thanks for the confirmation. I almost never use the 1/3rd stop ISOs, so it is a pain to scroll though them. Oh well, hopefully a Canon engineer is reading this (and if they are, make the "print button" customizable to be a mirror lock up, please!).
on the 50D didn't they make it the liveview button? which is almost a mirror lockup
Ian.Dobinson wrote:
the xxxD series is full stops only isnt it?
You're right; I had simply assumed that they did since they share most hardware features. The statements I made were based on measurements of the xxD series, and still hold for them.
With 3 years of shooting well over 100,000 frames with 5D's my experience has been that iso 200 is significantly more noisy than 100 and not that much better than 400. Iso 500 is significantly better for noise than iso 800, I use 500 a lot.
Note how the photon equivalent noise plateaus for each trio of ISO's, with the intermediate ISO's being about the same as the next lowest main ISO. There is a slight improvement for ISO 125 and 160, but nothing to get excited about.
How to interpret this data? Because the electronic read noise is the same for, say, ISO 500 and 640 as it is for ISO 400, if you shoot in manual mode there is essentially no difference in image quality for fixed Av/Tv among these three since one is collecting the same number of photons (so photon noise is the same) and the electronic noise is the same. The difference is that, because the signal is amplified less, ISO 400 provides 1/3 stop more highlight headroom than ISO 500, and 2/3 more stop for ISO 640. One simply needs to apply exposure compensation in raw conversion to bring up the histogram; the noise will be no different. If instead one is shooting in aperture or shutter priority mode, the camera's metering enters the picture, and so to get the same aperture and shutter speed for ISO 400 as one has for say ISO 500, one needs to dial in -1/3 EC (again to be brought back up during raw conversion).
Since there is so little noise benefit to the intermediate ISO's, and a substantial cost in lost highlight headroom, I have disabled them on my 1D3.
I should also note that Nikon implements intermediate ISO's differently than Canon; there are no such plateaus as far as I have seen (mostly D300 data, and some D3 data) at intermediate ISO's and so they could indeed be useful.
?? on 5D