Do the Fuji X Series cameras have an issue with underexposure and overinflated ISO? It's a common statement, and if true, it would be important to know how to compensate. In the next few posts I'm going to walk through a series of tests that show the ISO ratings and spot metering on the FujiFilm XE-1 are actually quite well calibrated. I'll explore the dynamic range of the camera and compare it to Ansel Adams' Zone System. I will also demonstrate why there's a good reason automatic "multi-segment metering" will tend to overexpose or underexpose in some situations. And I'll lay out the biggest reason the Fuji X Series cameras, or any professional cameras for that matter, may seem to take poor exposures when compared with today's point-and-click cameras.
Why worry abut manual exposure and ISO values? I recently switched from shooting with a Nikon D7000 to the FujiFilm XE-1. WIth my D7000 I fell into the habit of "chimping" the histogram, adjusting with the exposure compensation dial, shooting again, and then pulling amazing latitude out of the Raw files. Right out of the box I found the XE-1's multi-segment metering isn't nearly as advanced as the Nikon's - sometimes significantly underexposing due to the typical white-out skies here in Oregon. Even more frustrating was the consistency - turn the compensation dial up a stop, reframe, and the next shot would come out over exposed. Reframe again and the shot is back to underexposed.
It was clear I needed to dust up on my exposure skills a bit and start relying on the spot meter and manual exposure. And you know what? That is OK with me. Slow down a bit, think things through, tell the camera what I want, take better shots. I grew up with traditional 35mm and spot metering, I can do this!
There are a few few things I need to know about the XE-1:
- Will an incident meter reading create an accurate exposure with the XE-1?
- Does the Sunny 16 rule apply?
- Is the XE-1's spot meter accurate?
- What is the XE-1's exposure latitude?
THE SUNNY 16 TEST
Before there was digital there was the Sunny 16 rule, it went like this:
Sunny 16 - If the sun is bright set your shutter speed to your film speed (1/100 second for 100 speed film) and your aperture to f/16. If the shadows are soft, open up a stop to f/11. When cloudy with no shadows, open another stop to f/8. For dark and shady areas another stop to f/5.6. And for deep shade open to f/4.
Some say "Sunny 16" doesn't apply in the digital age, that the digital camera doesn't have the exposure latitude…. except, neither did slide film! Of course, if the ISO on the X series isn't accurate, then the Sunny 16 rules won't apply either.
To test the Sunny 16 rule I set up an 18% gray card out in the bright midday sun. My Minolta Auto Meter IVF measures f/16 + 1/3 stop (f/18), just a hair brighter than the "Sunny 16" rule.
With the 18-55 zoom lens in place, the XE-1's spot meter on the 18% gray card reveals exactly the same reading of f/18. The test consists of an exposure sequence of varying ISO's with matched shutter speeds, all at f/18. If well calibrated, an exposure of ISO 200 at 1/200 second should create exactly the same exposure as ISO 800 at 1/800 second, each exposure producing gray values of 50% (The 'L' level in the Lab color scale).
The incident and spot meter are in agreement, with the exposure coming about as close to 50% gray as could be expected.
I repeated the sequence with the 35mm lens. Resampling the light levels with the incident light meter now showing the sun to be hitting right at f/16.
(The sun faded as I reached the ISO 3200 shot so I've left that out of this sequence.)
I noticed 2 things:
- The camera's spot meter wanted to subtract 1/3 stop from the incident reading.
- The gray values are about 5% higher (+1/3 stop) than they should be according to the Minolta incident meter.
Being a prime lens, the 35mm is more efficient than the zoom lens, so this makes sense that it is brighter than the zoom. Had I used the camera's spot metering the exposure would have been spot on. Since this isn't a lab test, 1/3 stop is well within my margin of error, I'm not too worried about the difference here.
Inflated ISO? Nope. ISO, aperture, and shutter are all in perfect relationship and well calibrated. The incident meter, spot meter, and gray value all agree with one another, and match to the Sunny 16 rule as a baseline. If anything, the 35mm lens is producing pictures that are just a tad brighter than they should be, but definitely not darker as an inflated ISO value would indicate.
Would different copies of the XE-1 produce different results? I doubt it. A camera system, by design, must be highly calibrated. The CCD is solid state and is highly likely to be identically calibrated from copy to copy - an uncalibrated ISO value would throw the entire system into question. As for the other cameras in the series, the FujiFilm X-Pro1 should measure out the same way when using the same lenses. I'm not sure how this plays out on the Fuji X-100s, as the excellent fixed lens on that camera may have slightly different characteristics, though I don't expect the light loss to be any worse than the 18-55 zoom lens I tested here.
You might be wondering how the different film simulations and DR modes impact spot metering. I exercised the camera through these as well - each spot-metered exactly the same on the gray card as the tests above, producing the same gray values. (In the next post I'll explore what the DR mode is doing to the dynamic range, but all modes generate 50% gray at proper exposure.)
In the next post I'll explore the XE-1's dynamic range and compare it with the Zone System. I'll also shoot a series of portraits with problematic back-lighting to explore the impact of the XE-1's different metering modes. I'll apply the Zone System to the same difficult shot, and I'll give an answer to why point-and-shoot cameras often win when comparing out-of-camera shots in these situations.
Please leave feedback! I would enjoy hearing your thoughts on this topic.
By Adriel Henderson