Monday, January 9, 2012

Fuji X10 – Review Part 26 – Versus D7000 Sharpness and Details test …

I’m pretty tired of comparing the X10 to itself. L versus M and JPEG versus RAW and DR100 versus DR400 and on and on and on … Fuji engineers should be taken out behind the barn and …. never mind.

I’ve pretty much exhausted that line of testing and have concluded that JPEG is fine but RAW is better … that DR100 ISO100 in L size is very nice in RAW if you don’t mind managing highlights yourself … that L is sharper than M in all cases …

But back to the real world … how sharp in the X10 up against a consumer dSLR? It is the match people think? Well … let’s have a look …

I walked in the woods today for about 45 minutes – my hands were blocks of ice by the time I got back to the car – and I came upon this interesting smoke stack (chimney) at the Wild Bird Center. I t was getting pretty dark, but I was still able to get 1/500s at 100 ISO with the X10 at f/3.2. To match the DOF, I had to shoot the D7000 and 70-300VR at f/8, which necessitated 640 ISO. The meter gave me the same shutter speed.

Of course, as per usual, the Fuji had to be forced to shoot that way at –1EV, whereas the Nikon meter nailed the exposure at 0EV. This is an annoying property of the X series that I have noticed during testing … they are way too aggressive on the exposure. At M size especially, since they are 2EV hotter than L size in JPEG mode.

Anyway … here are two shots from these cameras, with the Fuji cropped to 3:2 to match the D7000.

Fuji X10  f/3.2  1/500s  100ISO  -1EV

D7000 + 70-300VR  f/8  1/500s  640ISO  0EV

The X10 is a bit flatter, which I happen to like. But before anyone gets their knickers in a knot over this victory, it is all Photoshop. I dropped contrast for the Fuji because it was hotter than the D7000, which benefitted from having the excellent neutral profile in ACR.

The Fuji bokeh even looks a tad softer in parts … but you’ll see in a moment that this is not a silver lining.

It is easy to declare them about equal at WEB sizes and for small prints. No problem.

But … what about cropping and larger prints? Well, we need to go closer to see how that would work out.

You need to click through to see what is really there.

Some observations:

  • Both have excellent detail. No problems there.
  • The X10 details are coarser however, and that would show up in a large print.
  • The X10 has halos, which I’ve noted even in the JPEGs in other tests. These halos actually hurt edge integrity somewhat.
  • The X10 has the strangest busy bokeh … each small branch appears to have an echo. There was no wind at all, so this is just plain old nervous bokeh, and it is not pretty.

I expect that enlargements of moderate size will be ok. But as you crop or go larger, these edge issues and nervous bokeh are going to show up.

I don’t know is this camera really has the chops to be a landscape camera. Architecture, maybe … but not wide enough and the edge integrity issues are a PITA. Of course, the ORBs are not exactly friendly to night time architecture shooting.

But let’s take a closer look at these halos … how bad is it, really?

Again, slick through to see what this looks like at 400%. Huge halos when compared with the D7000. Edges much thicker and details much coarser.

The X10 is definitely not an obvious choice for printing big …it will take some real skills to do it. (Or the ability to hold your nose.)

I think Fuji have several things to address. These last few days have exposed serious issues with JPEG exposures, sharpness issues, and weird edge integrity issues. They need to work on their firmware and JPEG engine, and they need to cooperate with Adobe for better demosaicing and sharpening plus they need to create some nice Adobe compatible profiles, including lens profiles to address the CA that is hurting the sharpness.