Studio Tests

NOTE the line marked 'Nyquist Frequency' indicates the maximum theoretical resolution of the camera body used for testing. Whenever the measured numbers exceed this value, this simply indicates that the lens out-resolves the sensor at this point - the calculated MTF values themselves become meaningless.

The Pentax 15mm F4 performs quite competently in our studio tests. However like the Sigma 10-20mm F4-5.6 we tested recently, it shows distinct curvature of field, which causes problems with our (flat-field) chart tests at the shooting distance used (approx 2m). As with the Sigma, we have chosen to present data that is technically slightly rear-focused, and therefore gives a better impressions of the edge and corner sharpness at the expense of the centre (this data is also most representative of the results obtained when using the camera's autofocus). Our test sample was also slightly decentered, giving marginally softer results on the right side of frame.

Compared to the Sigma 10-20mm F4-5.6 EX DC HSM (at 14mm), the 15mm F4 comes out on top; it's a hair sharper, and has lower distortion and falloff. Unfortunately it doesn't come out so well against the excellent Tokina AT-X PRO SD 12-24mm F4 (IF) DX, which we'd expect to perform near-identically to Pentax's own smc DA 12-24mm F4 ED AL (IF) - in this case the zoom is sharper, and has lower falloff and equally low distortion, although it does show more chromatic aberration.

Macro Focus

Wideangles aren't generally the first choice for macro work, but the 15mm F4 doesn't do too badly in our test. Maximum magnification is 0.16x, with a measured closest focus distance of 18cms, and a working distance of just 9.5cm from the subject to the lens. At such a close focus distance barrel distortion has become quite obvious, but chromatic aberration is still reasonably low with just a little red/cyan fringing. Central sharpness is high, but the corners of our test chart shot remain soft until the lens is stopped right down to F16, again due to curvature of field.

Macro - 152 x 101 mm coverage Distortion: moderate barrel Corner softness: moderate Focal length: 15mm

Specific image quality issues

Flare

Control of flare is a critical feature of a wideangle lens; with such a broad view of the world, bright light sources will find themselves in the frame on a regular basis. The 15mm F4 generally performs well in such situations, with little image degradation from flare effects.

With the sun placed in the frame, few flare patterns are visible at normal working apertures, but stop down to F11 or smaller and colored radial streaking starts to become apparent (in these situations shooting at F5.6 - F8 seems optimal). However move the sun just slightly outside of the frame area and flare problems disappear almost completely. This represents a clear advantage for the prime over most of the wideangle zooms we've tested.

F16, sun in top right corner of frame	F4, strong backlight

Chromatic aberration

Lateral chromatic aberration is visible in many real-world situations, taking the form of red/cyan fringing which is most visible on high contrast edges towards the corners of the frame. The samples below show how this appears in real-life, and how effectively it can be removed in post-processing. In this case, we've used Adobe Camera RAW v4.6, with a value of -20 applied to the 'Fix red/cyan fringe' slider in the 'Lens Corrections' tab. This has essentially eliminated the red component of the CA, leaving just moderate (and visually less disturbing) blue/yellow fringing in the extreme corners of the frame.

ACR default	ACR, -20 red/cyan CA correction
Pentax K20D, F8
100% crop, top right corner

Softness at F4

Our studio tests reveal that this lens isn't at its best wide open, and as usual we like to demonstrate this with real-life results. Here's a comparison of shots taken at F4 and F8 using the 14MP K20D (the most demanding Pentax body currently available), with 100% crops taken from the centre and the edge of the frame. It's not that the F4 shot is disastrously poor - this would probably still look OK in a 12" x 8" print, for example, especially with a little local contrast enhancement in post-processing - but the F8 version is clearly crisper, with higher contrast and much superior rendition of the fine, low contrast detail in the brickwork towards the left side of the frame.

F4	F8
Pentax K20D, RAW + ACR
100% crop, centre
100% crop, left edge of frame