Monday, June 03, 2013

Tandoori Lens Summit Part Two: Working Methods

Robbie on the K-5

In the first article I gave some context for our lens summit. In this article I'll provide some rationale for our lens choice, testing method, and processing, along with our motivation.

Lens Choice
The initial incentive was to compare our fastest lenses, after we realised that between us we have the three available f/1.2 lenses for Pentax K mount. Then, because neither of us are fixated on shooting that wide open, we expanded the comparison to other more-or-less "normal" lenses.

I gave the six lenses in the test single-letter codes for convenience.

C = Cosina 55mm 1:1.2 MC
K = SMC Pentax 50mm 1:1.2
A = SMC Pentax-A 50mm 1:1.2
Z = Carl Zeiss Makro-Planar 50mm F2 T* ZK
L = Pentax SMC-FA 43mm f/1.9 Limited
O = Olympus M.Zuiko 45mm 1:1.8

Unfortunately I left two lenses in a drawer, the PEN F 40mm and the Leica Macro-Elmarit 60mm. I've looked at both extensively on this site in the past. Do a search in the side panel if you are interested. Sorry for the omissions. I blame my oversight on the desire to finish cooking dinner!

We shot on two different camera systems with similar sensor sizes. Robbie did the APS-C shoot on a Pentax K-5. I used the Olympus E-PL2, a Micro Four Thirds (MFT) unit. The Pentax lenses were attached to the PEN using a K-mount to MFT mount adapter. Actually, it's a tilt adapter, since it's the most solid of those I own. But of course we did not use the tilt function!

I should note that the Zeiss lens is the variant in K mount, which is no longer made. So it worked just fine off the same adapter.

The Olympus M.Zuiko, being an MFT native lens, did not need an adapter on the PEN. But we could not use this on the Pentax. I threw it into the comparison at the last minute as an extra data point.

The use of the PEN camera raises a question: Why did I shoot lenses on a sensor size they were not made for (M.Zuiko excepted)? First, consider that none of these lenses are actually built for APS-C. The first five all cover the so-called "full frame" of 135 film. So in fact none of them provide a normal field of view even on the Pentax K-5.

The main reason is that I can mount lenses from dozens of systems on the PEN series bodies. They make a handy back-end for these sorts of comparisons, whereas the large flange distance of the K-mount means it's restricted to a much smaller subset of glass. In addition, I predicted that it would be easier to get focus and consistent exposure on the PEN, making the rather tedious process somewhat easier.

Testing Method
One of the disadvantages of an SLR is that accurate focus through the viewfinder at apertures as wide as f/1.2 is impossible An enhanced focus screen makes this somewhat easier, and Robbie has the best installed on his K-5. But still he used "Live View", which means the mirror is up and you work directly from the back LCD. I too used the LCD on the mirrorless Olympus. We both took advantage of the maximum x10 magnification. This is right at the limit of acceptable image noise and grain. More than this and these artefacts would be a significant barrier to getting tight focus. (But expect technology to improve in this area.)

The subject was a still life, set up inside the living room, door closed, with no wind or disturbance. Light was completely diffuse and consistent, since we had a typically Irish overcast day. I have blinds on either side of the room to adjust for glare control. Once set I didn't touch these. There may be slight (1/3EV) highlight burns in the extreme edges of the frame from patches of window light. Otherwise you should see no overexposure.

All lenses had hoods, sometimes originals and sometimes after-market varieties. No filters were used on the front elements.

We used a Feisol tripod with minimal extension (no centre pole) to ensure a solid mount for our cameras. There was slight frame movement as lenses were changed. There is no practical way to avoid that and neither does it matter much, since every shot was manually focused.

As subject we chose a vase of fresh cut flowers. Flowers have areas of texture, fine detail, and subtle tonal gradations. They have enough structural complexity to evaluate foreground and background elements, including bokeh. They have places that are easy to pinpoint for consistent focusing. They sit still.

You will notice when we come to the shots that our focal point was the stamen of the central flower, the top left anther to be precise. DOF was so thin that sometimes the whole anther was not in focus at the same time! This location is just below the centre of the frame, so all optical qualities should be at their best.

Distance to the subject was about 70cm, which is on the close side, but nowhere near macro range.

Since our focus was on fast lenses, we did not look at edge-to-edge sharpness. Besides, I have found it almost impossible to align in parallel planes a flat subject and a lens. Short of test lab quality laser sightings hard mounted to the tripod platform, the likelihood is that slight tilt or skew of the camera plane will occur. Besides which, walls are rarely flat or perfectly vertical to the ground. It's all a nightmare.

Naturally we shot in RAW format at the lowest ISO the camera provided. After the shoot, the contents of the memory cards were transferred to the computer, consistently named, sorted, and backed up. I pulled the images into Lightroom and accepted a default pre-processing that included two steps: sharpening with amount 25, radius 1.0, and detail 25 plus a negligible colour noise reduction of 25.

No further processing was carried out, even though this is quite unrealistic given my usual workflow. In all cases you can get nicer images than these with a little work. But what we have here will do for comparison.

Surely our methodologies are not fool-proof. We wanted to have fun and not get too bogged down in minutiae. At the same time, any job doing is worth doing with a modicum of consistency and care. Hopefully this article has been useful in describing factors you have to consider when trying to be consistent in lens comparisons.


We had several goals in mind.

1. To have fun with our tools.

2. To learn more about the qualities of different lenses, so we can choose one over the other depending on circumstance and intention.

3. To help determine which lenses to sell and which to keep.

4. To learn from the responses we get back from other photographers.

5. To challenge our own expectations.

If you are like us, you'll want to stay tuned for part three.


1 comment:

robin said...

Part three is here:

Post a Comment