White Balance

As you can see the *ist D produced good automatic white balance results in natural light but faired less well in incandescent and fluorescent light. That said the preset setting for outdoor and incandescent light is set well, however we were not able to find a fluorescent setting which produced good results with our fluorescent lamp. As expected manual preset white balance delivered the best result.

Settings: ISO 200, FA 50 mm F1.4 @ F5, Small/

Outdoors, Auto Outdoors, Cloudy (or Sunny) Outdoors, Manual
Incandescent, Auto Incandescent, Incandescent Incandescent, Manual
Fluorescent, Auto Fluorescent, Fluor 1 (2 or 3) Fluorescent, Manual


Overall flash performance was good, using exposure compensation with the flash up doesn't modify the exposure but instead modifies flash output (effectively flash exposure compensation). As you can see from the samples below the camera's TTL flash system was fooled by the light background behind the subject, adding some exposure compensation solved this. The shot of our color patches was well exposed with no color cast and a nice tonal balance.

Settings: ISO 200, FA 50 mm F1.4 @ F5, Large/

Av, 1/60 sec, F7.1
Underexposed - fooled by light background
Av, 1/60 sec, F7.1, +0.7 EV
Better with some exposure compensation
P, 1/60 sec, F2.5
Well metered, no color cast, good exposure

Night exposures

The *ist D uses dark frame subtraction noise reduction method for long exposures. It performs this by taking a second exposure immediately after the long exposure and using the noise from that image to subtract from the first. In our tests the *ist D delivered clean long exposure images (at ISO 200), that said they did have the same soft appearance as other *ist D images.

Settings: ISO 200, FA 28-105 mm, Large/

ISO 200, 10 sec, F16
ISO 200, 6 sec, F9


Like all digital SLR's (and several prosumer level digital cameras) Pentax offers a RAW format option on the *ist D. RAW simply means data direct from the sensor (12 bits per pixel) which hasn't been processed in any way. Additionally the current camera settings (such as parameters, exposure etc.) are recorded in the header of the RAW file. Pentax uses the .PEF extension for RAW files which appear to use some kind of lossless data compression (probably similar to Zip) as the actual size of the file varies. Interestingly the *ist D's RAW files are very large (12+ MB), approximately 16-bits per pixel location which is a shame because this will have a storage impact for users.

The *ist D is supplied with Pentax Photo Browser and Pentax Photo Laboratory, the later of which is a RAW converted with a relatively good range of flexibility. The biggest problem with Photo Laboratory is that you can't work on an image in a large window (or magnified).

RAW vs. JPEG resolution

Below are two 100% crops taken from images shot within seconds of each other. The first crop is from a Large/ JPEG, the second from a TIFF created from a RAW (.PEF) file using Photo Laboratory 1.0. While there isn't a noticeable resolution advantage shooting RAW the final image output is certainly sharper than JPEG from the camera. This is interesting as one thing we have observed with *ist D JPEG's straight from the camera is their softness.

JPEG RAW (converted to TIFF then cropped)

RAW vs. JPEG sharpness

Below are a set of crops from the same shot taken in JPEG and RAW modes, as you can see the RAW image appears to be sharper (although equally has some edge artifacts around black detail), so either in-camera sharpening isn't as strong as that carried out in Photo Laboratory or there is a problem with the interpolation algorithm used in-camera.

JPEG RAW (converted to TIFF then cropped)

Photo Laboratory RAW artifacts

One thing we observed on our resolution chart were strange artifacts which look like Bayer interpolation artifacts. These appear as uneven colored blocks along the edge of curve detail. It looks to me as though Photo Laboratory could do with a little tweaking.

RAW latitude (digital exposure compensation)

Shooting RAW on the *ist D appears to provide you with somewhere between three quarters and one stop (0.7 - 1.0 EV) of additional latitude above the visible clipping point (255,255,255) of the original image. In the example below you can see that some detail is recovered from the top of the clock and more noticeably in the gradiation of the crayon tips.