X-PRO2, Acros & SOOC...

I'm not done with the Six Days in Tokyo series just yet, but I'm sidetracking it for today. I've been intrigued by the new Acros film simulation in the new X-Pro2 ever since I got my hands on the prototype and have experimented on and off with it for months now. In fact I've based my X-Pro2 BW in-camera preset on it.

When I was in Japan I had a conversation with one of the engineers who asked me what I thought of it. I replied that my initial side by side comparisons with standard monochrome showed visible but subtle differences, but that I'd noticed some much more interesting results when using the new expanded push/pull capabilities (which I wrote about in our collective review on KAGE). I told him it seemed very intentional, as if the simulation was meant to be tweaked this way: he smiled and confirmed it was. Yesterday morning I stumbled across Donovan Bond's analysis (on Fuji vs Fuji) of the company's official post on this new film simulation and I followed the link to that article—this is a long excerpt but it's extremely interesting:

To be specific, ACROS mode has a completely different noise reduction algorithm from other modes. The “graininess” of the silver-halide films are what we see as “noise” in the digital data. For color images, they are the unwanted noise, but in the monochrome images, it becomes an important texture. Turning the noise into grain-like texture is what makes ACROS unique and different...

We developed it from the core of the image file to achieve a very complex and natural like grain expression. Optimal and different grain expressions are added to highlight and low light areas. You would not find unnatural dotted graininess in the highlight areas just like how the monochrome film behaves. In the low light area, you would see the graininess just like how it would appear with the monochrome film. There are undulating grain within the picture. And it adds depth like no other.

ACROS also changes the output of graininess depending on the sensitivity setting. As the sensitivity gets higher, stronger grain effect becomes visible, just like the film. We have seen the advancement of high S/N ratio of digital cameras, but people generally want to take photos with lowest sensitivity possible. But with ACROS, it may be a different story.

The unique grain effect, which becomes apparent at the higher ISO sensitivity. You can intentionally set the sensitivity high to enjoy the effect.
— Fujifilm-X.com

Interesting not only for the technical aspects but because it goes a long way towards explaining what I've been seeing all along: this is a film simulation that reacts differently depending on camera settings. So not only contrast adjustments but ISO as well. And as Donovan pointed out: grain is built-in AND contextual, through throttling of the NR algorithms. It reacts to the shooting conditions.

I spent most of yesterday working on a project proposal but this Acros article kept haunting me...so when I finished I did a few tests around the house. I noticed ISO 2000 seemed to be a sweet spot for this simulation, creating a visible grain that added personality without reducing sharpness or introducing anything remotely muddy into the mix (actually it scales well all the way up but 2000 felt like a good general compromise). When Jacob came in to chat around 5 PM I asked if he'd let me shoot a few images. He's a good sport — he's also getting used to having curves drawn across his face ;)

The following images are an oddity for this blog: they're all straight out of camera. Didn't do a thing to them. Zip. Plus we're talking a big 90 seconds worth of shooting time. Settings are: Acros R, highlights -1, shadows +3. No flash—I used the modeling light in one of my strobes as key, high up camera left (with a grid) along with the (low) ambient— because I purposefully shot everything at ISO 2000. Normally I'd boost exposure slightly on some of them in post, add a few local adjustments, but I wanted to show unaltered results. I have to say, I kinda seriously love these.

Below is a zoomed version of image #5. Notice the grain and texture—it has an almost graphite etching feel. Nothing like the new grain simulator that's also part of this new camera (which I'm not a fan of so far). Web can't do it justice but I assure you: on an iMac 5K it's frickin spectacular, so much so that it may indeed change how I approach certain shoots. I feel like I just found a brand new tool I'd forgotten in some hidden chest.

X-PRO2, 1/125 sec at f/3.2, ISO 2000.


Eric Kim has a post today from Dubai where he's attending GPP 2016. It's mostly about the X70, but towards the end he mentions Acros and how "funny" he finds these attempts at replicating film in digital: "why don’t we just shoot with film cameras in the first place?" he asks. It's a perfectly legitimate viewpoint—especially from someone who shoots film as seriously as Eric does—...but I don't share it. I'm not interested in shooting film on any sort of regular basis. Not interested in the manipulations, the restrictions, the lack of control or the overhead. I don't feel somehow liberated by 36 frames or being stuck at ISO 400 or not being able to see what I just captured. Don't get me wrong, I love the idea of creative constraints (I'm the guy who travels with one camera and one prime lens) but there's a certain threshold of practicality I won't cross. Simply because it annoys me; which then becomes counterproductive. But like many, I love how film looks. So as far as I'm concerned there's nothing futile about this quest of translating what is essentially the persona of film to the digital realm because yes, there is an inherent quality to film images that was for the most part lost in the switch. But it has to be more than a gimmick.

I've said it before but it bears repeating: Fujifilm's commitment to bringing their film legacy to the X Series is an issue of pride. It's where their identity can shine and how they can differentiate themselves further. I personally find the development of Acros, the research that went into its creation, very, very exciting. There's a complexity at work here that goes way beyond slapping a curve on top of a monochrome file, and this shows a thirst for exploration that could yield serious results down the line.

This may just be a glimpse of what lies ahead.

A print from Tokyo and beachcombers' bounty from PEI. Xpro2, 1/150 sec at f/2, ISO 2000. This one was processed in LR.