Although the holistic approach to understanding a film simulation is probably the best, there's some value in a clinical approach. Especially for folks like me who aren't really good at making sense of the simulation results based solely on actual photographs.
Here's the DPReview studio scene in Provia:
Let's focus in on the color-checker. Here's a comparison of the Provia results compared with what X-Rite says the sRGB values should be. The full-size squares are the official values, the smaller ones inside them are Provia:
Across the bottom we can see that Provia gives higher contrast with darker shadows and brighter highlights, except at the brightest highlights where Fujifilm has indulged its obsession about avoiding blown highlights by rolling off the response curve.
[Methodology note: to compensate for the DPReview exposure, I brought the brightness of the standard down to match the Provia results in the middle-gray patch, then brought the combined brightness back up again to match ColorChecker standards. As a result, the Provia patches are brighter than in the DPReview studio scene.]
The Provia colors have noticeably more "pop." Much of that is simply from luminosity:
But obviously not all of the difference is from luminosity. Reds are much more saturated in Provia, and of course related colors such as oranges and magentas are also affected. Blues are also more saturated, and greens a little bit more saturated.
I've said "saturated" but the image I've shown is for both saturation and hue. That's because I know this: there is virtually no hue difference. In terms of hue, Provia is virtually bang-on the standard values:
If you look closely, you'll see that Provia can't quite keep the purple from turning a bit magenta. Purples are really challenging colors for digital sensors, and X-Trans II does come up just a little short.
And now for something that many of you will be surprised at. Let's look at what happens if we shoot in Adobe RGB color space.
You might've thought that the only difference when you chose a color space was the color space. All of these colors are in-gamut for sRGB, so converting from Adobe RGB (with absolute colorimetric intent) won't affect the colors. But the green patch is noticeably darker from the Adobe RGB shot, and there are small differences in the other colors, too.
Make of it what you will. I did this test because my experience with Canon DSLRs is that Canon's results are quite different between sRGB and Adobe RGB. Canon apparently figures that anyone using Adobe RGB wants accurate colors, while sRGB users like more "punch." That doesn't seem to be what's going on here, though.
The open-source LightZone Project: http://lightzoneproject.org/
|Post (hide subjects)||Posted by||When|
|10 months ago||19|
|10 months ago|
|10 months ago|
|10 months ago||7|
|10 months ago|
|9 months ago|
|9 months ago||2|
|Christine by JP Zanotti|
from Car wreck
|Fangorn Forest by cand1d|
|Yosemite Falls with Moonbow by Jonathan Shapiro|
from Best Landscape of the Week 4