#108. OM-D! An IBIS in my hotel room!

Now that don’t happen often, does it? Unless you live in Egypt or Australia (or …), ibises don’t fly that often into hotel rooms. So imagine my surprise when a 5 stop IBIS landed right in the palm of my hand in the very center of Paris?

Gone potty? Not really, as I’ve always had screws loose. But the ibis in question is the internal stabilization system inside my lill’Olympus OM-D E-M5 camera and it is pure magic (as befits the name, since Thoth is the Egyptian God of Magic and is often represented Ibis-headed).

The test scene. ISO200, f/2.5, 1/80th. Easy.

Nuff with the silliness 😉 The little thingy inside the OM-D is neither legend or mythology. But it is pure technological goodness that works little miracles of its own and, when the sky turns deep blue and the clouds all purple, and your hand-held pictures still look sharp, you can be excused for feeling awe and magic in the air.

Technological explanation

It works along 5 axes, which is better than four, better than three and probably better than two, as well.

In the good old days of 2-axes IS, you had your back covered if you suffered from that exotic plane-tremor, a.k.a 2D-jiggy. Then came 3-axes IS, which by some sprinkling of fairy powder induced evolution allowed you to tremble the human way. Now, wait for it, we have 5 axes, which mean you can shiver on a boat pitching in a tourmented sea like a rapper on steroids and still bring back sharp memories for the family ! A May Zing.

A bit later. NO IBIS! 1/13th => Unsharp.

Enough boring techno babble 😉

The proof

Thing is, Thoth inside (R) or not, my first tests left me with a rather luke-warm impression of the system. Using the 45mm 1/8 Olympus (gem of a) lens, I managed 3-4 stop gains with IBIS on compared to IBIS off but found that the OM-D + 45mm combo required higher speed than an equivalent focal length on the Sony NEX-5n, for instance. So the actual gain was more like 1.5 – 2 stops.

Worse still, my recent Street of Paris by night with the OM-D post, in spite of showing the first ever real photo of a ghost (2 in fact), attracted vehement criticism because my IBIS-demonstrating shots with the 45 were no slower than 1/100th of a second.

Later still. IBIS ON. ISO200, f/2.5, 0.4s. Perfectly sharp. Incredible performance.

So I decided to conduct a little test before the impartial jury of the setting sun, using the ISO-27003-standard sharpness test target : ugly brick walls. No one can accuse me of acting like a photographer inserting street photography images in a blog post dedicated to street photography. Not this time. Ugly & boring is my motto of the day. Pixel peepers can wet their pants over this one.

For this test :

• I used the most powerful lighting system in the world : the SUN!
• I was not satisfied with a lousy studio reproduction of right angle and radial lines, but had a whole set of buildings with various textures assembled before my lens.
• I made sure no pretty flowers or artistic feeling came anywhere near the frame!
• I repeated my shots when unsure of focus accuracy!
• I took notes with a graphite compound rod on pressed and dried cellulose pulp
• I left my tripod at home 540 miles away, so that its stabilizing influence would be less than at 500, 400, 300, 200 or 100 miles away, for the least artificially supported shots science can provide.

What you are seeing in this post, then, is a series of shots taken in the decreasing light of a sunset and the challenge was to determine how far into the night and long-exposure territory I could maintain it.

The first picture is the test scene in good light. It is close to as sharp as the camera + lens (Panasonic Lumix 14mm f/2.5) will produce hand-held. I recommend reading the picture captions for tech info and an uninterrupted sequence.

Even later. This time I upped the ISO to 500. F/5.0, 1/4th. Perfectly sharp and in focus. Great image quality all around.

A 100% view of the previous image. It’s past 10PM and almost completely dark, as you can tell from the window lights. Impressive quality.

All along the post are successive exposures as night fell. 3 hours in the lab I remained, foodless, companyless and boozeless in the name of photographic science. I hope this clears my Karma of all debt and that all artistic sins are forgiven. Pfew !

My interpretation of results

As I found previously, without IBIS I am unable to take consistently sharp pictures below a speed of 1/f. In this test, the 1/13th exposure is blurred.

After this set, I repeated my experiment with the 45. I shot series of 3 images at several speeds with and without IBIS. At 1/50s 2 out of 3 were good. At 1/30s, none were really tack sharp. This is better than my previous results, but not as good as I previously experienced with my NEX-5n (routinely using 1/15th with the 25mm Biogon). With IBIS on and adjusted to 50mm lenses, I got sharp pictures at 1/5s and could probably have gone on. Here’s a full size sample. It’s slightly blurred because of the heay diffraction at f/13, but it is rock stable. That’s over 3 stops better than without IBIS.

The 45 parenthesis. 1/5s (click for 100% view)

As I estimated before, IBIS brings at least 3-4 stops of stability, which is enormous! Given my greater difficulties with stability using the OM-D compared to the NEX-5n (stiffer shutter release ? Clunkier shutter ? Unsupportive grip ? …) I’d guess the advantage over the NEX-5n is around 2 stops. Meaning ISO3200 on the NEX-5n should be compared to ISO800 on the OM-D.

Pushing it too far. ISO1000, f/2.5,0,6s. Noise is high but only present in luminance. Almost stable but not quite. Focus not quite perfect either. This is pushing too far. Although the picture doesn’t show it, it’s totally dark.

The same frame at 100%. Not good 😉

Leaving aside technical considerations, what I love about these pictures is the subtlety and fullness of colours. I have altered the slightly wavering white balance in some pictures but saturation and individual colours are untouched. Compared to many other cameras, the OM-D E-M5 really is a prodigious colorist in the noblest sense. No over saturated exaggeration, but a real balanced and subtlety that cannot fail to please. With other cameras in the past, I could come home and look at my pictures on the laptop and wonder why I had made them in the first place. This doesn’t happen with Lill’Ollie. See “Looking Down” (a couple of pics below), for instance. Make that picture grey and off-balance, and it looses all its meaning. It’s simply wonderful here.

One last picture

A bothersome aspect of high-ISO performance and internal stabilisation is that they always seem to serve get-the-picture-at-all-costs purposes. It’s alright if you don’t get that shot. In fact it’s better than getting a noisy and slightly blurred one that will stay for ever on a hard drive or on a Facebook page for all to revile.

30 seconds on a window frame

What I’m getting at is that technology doesn’t replace careful shooting. For all its marvel, IBIS doesn’t replace a tripod. Here’s the same scene with the camera (perilously) placed on the window frame and exposed 30 seconds at f/4.5 and ISO200. The full aperture high-ISO IBIS-enabled version does not come even close to this one. What do you think ?

The same view at 100%

Just for fun, here are more stabilized images with long exposures. For some, I used available hand, elbow or shoulder support, but still, results are impressive.

Looking down. ISO 500 f/2.5 0.4s

Looking West. Perfect exposure. ISO500, f/2.5, 1/6s. Perfectly sharp.

At the limit. ISO1000, f/2.5, 0.6s. Stable but the autofocus couldn’t focus. This seems to be the low-light limit of full auto on this camera. Impressive!