Studio Tests

The Pentax smc DA* 55mm F1.4 SDM is naturally somewhat soft wide open, but improves rapidly on stopping down, giving excellent results right across the frame at normal working apertures. It's softer at F1.4 than the Sigma 50mm F1.4 EX DG HSM (or indeed the Nikon AF-S Nikkor 50mm F1.4G), however it is still noticeably sharper than the older smc FA 50mm F1.4. The differences between all of these lenses become rather small on stopping down beyond F2.

Our tests revealed some asymmetry at large apertures, with the the checkerboard crops across the 'arms' of the chart showing different sharpness and colors (due to longitudinal chromatic aberration), indicating a slight tilt in the focus plane. This occurred (to a greater or lesser extent) with three different samples of the lens, perhaps hinting at a barrel design or construction problem.

Sharpness Sharpness isn't wonderful wide open, but stop down even slightly and the centre improves markedly. The lens is at its sharpest in the centre between F3.3 and F5; the corners lag behind slightly, with the most consistent results across the entire frame at around F5.6-F8.
Chromatic Aberration At large apertures we see blue color blur in the centre, a symptom of color-specific spherical aberration. This disappears on stopping down to F2, from which point onwards the lens shows a measurable, but practically unimportant degree of lateral CA.
Falloff We consider falloff to become perceptible when the corner illumination falls to more than 1 stop less than the centre. There's nothing to worry about here; just a stop at the extreme corners wide open, which clears on stopping down.
Distortion Distortion is very low - we measure just 0.5% barrel, which is unlikely to have any impact on normal use.

Macro Focus

Maximum magnification is 0.19x, at a measured closest focus distance of 44cm, which gives a working distance of 34cm from the front of the lens to the subject.

As expected image quality isn't great at larger apertures, but the centre sharpens up nicely at F2.8, with the corners a little behind. By F5.6 our flat test-chart image is sharp right across the frame. There's a little barrel distortion, and some barely-visible chromatic aberration.

Overall a creditable result for a non-macro lens.
Macro - 125 x 89 mm coverage
Distortion: Mild barrel
Corner softness: low
Focal length: 55mm (82.5mm equivalent)

Full frame coverage

Not something that's of much practical relevance for Pentax users at the moment, but the fact that the optical unit is much the same size as the older DA 50mm F1.4 made us curious about the coverage of the 55mm F1.4's image circle. A couple of simple tests using a Pentax film SLR confirm that in terms of illumination at least, the lens does (despite its DA* designation) cover the full 35mm frame.


One key new feature of the smc DA* 55mm F1.4 SDM is its built-in ultrasonic-type autofocus motor, which according to Pentax gives fast and silent focusing. However the lens lacks Pentax's traditional screw-drive focus coupling, and the AF motor is powered by 2 contacts inside the lens mount, which are absent from bodies older than the K10D of late 2006. Essentially this means that the lens won't autofocus on the *ist series DSLRs, or the K100D and K110D. The SDM autofocus motor is certainly very quiet indeed, but in terms of speed, the 55mm is in fact (like Nikon's AF-S Nikkor 50mm F1.4G) noticeably slower than its screw-driven predecessor.

Autofocus speed

To measure the difference in focus speed between the smc DA* 55mm F1.4 SDM and the older screw-drive smc FA 50mm F1.4, we conducted some basic focus speed tests (slightly refined from the method we used for the Nikon AF-S Nikkor 50mm F1.4G). We used the Pentax K-7 as the test body, on the basis that this latest model should have the best-optimised focus system.

In the first test, we pointed the camera at a featureless white target and measured the time the lens took to drive from infinity to closest focus and back (using a sensitive microphone to record focus motor noise) - this gives a very basic measure of focus motor speed. In the second test, the camera was placed 1 meter from a high contrast focus target, the lens set to infinity and the time required to achieve focus and initiate shutter release measured (defined as the time from the focus motor starting to move to the mirror flipping up for exposure) - this provides a measurement much closer to an 'actual use' scenario. In both cases, 'lag' represents the time between pressing the shutter button and the focus motor starting to move. The light level was approximately 10 EV, and all times reported are the average of five measurements.

Inf - 0.45m - inf
Infinity - 1m
Pentax smc DA* 55mm F1.4 SDM
0.05 sec
1.6 sec
0.05 sec
0.41 sec
Pentax smc FA 50mm F1.4
0.05 sec
0.6 sec
0.05 sec
0.27 sec

The conclusion is that the SDM focus motor of the 55mm F1.4 is substantially slower than the camera's screw-drive system is with the 50mm F1.4, taking a whole second longer to move the lens from infinity to closest focus and back. This also translates to a longer focus time in a more realistic test of actual use, although the proportional increase is much smaller. However it's also worth bearing in mind that in normal use, the distance differences when refocusing tend to be rather smaller than that which we've used here, which will decrease the differences still further; and of course if you're in the habit of prefocusing every shot, AF speed differences will be largely irrelevant. So the SDM lens's slower focusing is only likely to be a practical problem for quick grab shots or when using continuous focus mode.

Autofocus accuracy

Sadly, the moment we started testing this lens we ran into problems with autofocus accuracy and consistency. We tested three samples of the lens, and quite simply none could focus accurately enough to use at F1.4 without applying AF adjustment (using test bodies which had no such problem with the smc FA 50mm F1.4). On our K20D the lens back-focused considerably, and required the maximum correction of +10 to function properly. On the K2000 the lens front-focused, and due to the lack of AF adjustment we were unable to achieve consistently accurate focus by any means. We got the best results on Pentax's latest model, the K-7, but the lens still front-focused slightly and, while many pictures turned out acceptably, it needed an adjustment of -3 for best results.

To put numbers to these observations, we ran a series of tests measuring the accuracy and consistency of various focusing methods, with the K-7 and K20D as the test bodies. Using our standard lens test chart, we measured the central sharpness achieved from 10 independent focus attempts, with the lens set to F1.4. The methods used were manual focus via the magnified live view display, contrast detect AF in live view, phase detect AF without adjustment, and phase detect AF with the optimum adjustment applied; the autofocus target was a single vertical black/white edge. The results from the K-7 tests are shown below, with the sharpness expressed as a percentage of the single best result (mouse over the appropriate boxes to see the ten individual results from each method). The K20D's AF results with adjustment (+10) applied were similar to the K-7's; without adjustment, the images were simply too out-of-focus for our software to analyze.

Focus methods comparison
Live View Manual Focus
Live View AF Contrast Detect
Phase Detect AF, no adjust
Phase Detect AF, -3 adjust

Perhaps surprisingly (given the K-7's primitive implementation of magnified live view) we were able to get sufficient information from our standard manual focus targets for this method to give overall the best results, although with the occasional obvious mis-focus. Phase detect AF with +3 adjustment came in behind by a hair, well ahead of the imaging sensor-based live view contrast detect AF; but phase detect AF with no adjustment only managed to consistently deliver around 70% of the optimum sharpness (and in these tests the overall average is dragged down by one serious misfocus).

In normal use, we found autofocus consistency to be lower than in our studio tests (few perfect black and white targets exist in nature), with an higher proportion of shots than we'd expect unusable due to mis-focusing. The problem is compounded by the fact that any errors simply aren't visible in the viewfinder, and the only way to spot them is to check every single image for critical focus while shooting (this is because all reasonably large APS-C viewfinders designed to give a bright image when using F5.6 kit lenses give a visual indication of the depth of field you'd achieve at about F2.8, and Pentax's are no exception). Consistency again seemed to be camera-dependent, with the K-7 doing best.

Now it's worth pointing out that any misfocusing will often be masked by the additional depth of field when shooting at smaller apertures or longer subject distances, which means that generally acceptable results can be obtained even without AF adjustment. However we'd argue that a lens which is specifically being marketed as ideal for portrait shooting (and therefore by implication selective focus and shallow depth of field techniques) should be functional at all apertures.