Unreliable Exposure Value using Entry Level Digital Cameras

[sharpy]

I was curious about Exposure Value (EV) accuracy from modestly priced digital and film equipment. The equipment tested was:
Pentax ME Super film SLR
Miranda Sensorex II film SLR
Minolta Auto Spot 1 deg Spotmeter
Sekonic L158 hand held reflection meter
Pentax K-1 DSLR
Canon Powershot SX160 digital camera

I shot a white wall with the above equipment. All were set to ISO 100. Of course, only the Sekonic and Minolta meters read directly in EV. The camera reported shutter speed and aperture was converted by formula EV = log2(aperture)^2/speed)
Pentax ME Super 8 EV
Miranda 8.6 EV
Minolta meter 8.5 EV
Sekonic meter 8 EV
Pentax K-1 7 EV
Canon SX160 6 EV

It appears that modestly priced digital cameras don't seem to meet any tight EV standards criteria. The film cameras fared much better. This seems to mean you can't count on trading exposure info between different low cost digital cameras and expect consistency. I would be curious to know if professional digital cameras are held to a higher standard of inter-camera interoperability. Can someone with pro digital gear and a high end stand alone meter post some data on EV consistency?

 

[sharpy]

correction. The Pentax DSLR is not K-1, it's a K-x.

 

[Agulliver]

Oh yes....I've used a small number of compact digital cameras and while they have all taken decent photos the aperture/shutter speed read out don't come close to matching what I get with any film SLR or even DSLR. So I've given up using the readout of the compact digital camera as any sort of guide for exposure with film. I have found that smartphone light meter apps give a decent enough idea, to get in the ballpark.

Just yesterday I was out with a little Praktica branded compact digital which does a fair job....but set to 800ISO it was reported f7.8 and 1/200 second. I had 400ISO film loaded in a Zeiss-Ikon folder and my light meter app suggested f5.6 and 1/50th which is two stops more exposure than the digital suggested. And yet the shots on the little digital were a little over exposed.

Maybe the whole ISO thing is just not sufficiently similar between digital compacts, DSLR and film cameras?

 

[grat]

The current ISO standard (as of 2006) for digital cameras allows the manufactures to choose how their REI (Relative Exposure Index) is interpreted. In theory, a photo taken at ISO 100, f/5.6 and 1/30th of a second should be visually similar in terms of light/dark, to an image captured with film with a speed of ISO 100, shot at f/5.6 and 1/30th of a second.

In reality, as I say, it's up to the manufacturers these days.

See https://en.wikipedia.org/wiki/Film_speed#ISO_12232 for the cleaned up summary of the brain-melting exercise that is attempting to decipher what "ISO" means in the land of digital.

 

[Wallendo]

I suspect the real issue is the the digital camera meters are optimized for digital sensors and programmed to maximize quality on a sensor. This may mean that the sensor is exposed to higher or lower exposures and digitally manipulated. With higher-level digital gear, it may be assumed that many users will use manual modes or assisted manual modes and correlation with traditional ISO measurements becomes important.

It is as supposition on my part, but I suspect that lower-level cameras use digital manipulation to overcome issues with cheaper sensors and meters.

 

[Chan Tran]

What metering mode do you set on the K-x? If you set it on evaluative/matrix it could increase exposure due to the wall is white.

 

[George Mann]

How about a digital slr that underexposes all daylight shots because of unusably low dynamic range?

I am referring to the Nikon D100, the first consumer lever offering of its type.

The exposure values of a digital camera is directly tied to the performance envelope of its sensor.