White Balance

The results of our automatic white balance test mirrored closely our experience of shooting with the camera, excellent results in natural light (almost always perfect grays) but disappointing under artificial light, the strong pink cast on the incandescent test shot was quite similar to that of the Canon EOS 20D (and we weren't impressed in that review either). It's a pity that on such an advanced digital camera they can't nail automatic white balance. You can of course always select a preset white balance setting or take a manual reading.

Outdoor - Auto WB
Red: -0.7%, Blue: 0.1%

Fluorescent - Auto WB
Red: -3.1%, Blue: -2.6%
Incandescent - Auto WB
Red: 8.3%, Blue: -14.0%

Long Exposure noise reduction / Night shots

The 7D provides for timed long exposures of up to 30 seconds, beyond this you can use the Bulb shutter mode (shutter stays open for as long as you hold the shutter release), obviously this is best executed with the optional remote release cord.

The camera's optional noise reduction employs the typical 'dark frame' system we're used to, a second frame is taken with the mirror down and shutter closed, this is used to map 'hot pixels' which are then removed from the image. In our quick test the 7D did exhibit some stuck pixels without noise reduction, with it these were removed but replaced with 'black pits' which is unfortunate, most cameras which use dark frame subtraction replace the hot pixel with the color of surrounding pixels.

Noise reduction Off Noise reduction On
30 sec, F9 30 sec, F9


The 7D's built-in flash worked well, the first 'portrait' test shot below is pretty much perfect with good flash power metering which wasn't fooled by the white background. Our color patches test shot came out slightly over-exposed (this could have been to do with distance to subject) and you could have compensated for this with a little negative flash exposure compensation. Overall good.

Image thumbnail Histogram

Program mode, ISO 100, 1/125 sec, F4.5

Program mode, ISO 100, 1/125 sec, F5.6


As described earlier in this review Konica Minolta's Anti-Shake system works by moving the camera's CCD via horizontal (x) and vertical (y) actuators to counteract sensed camera body movement. It is advertised as providing "up to 3 stops longer shutter speeds while maintaining image sharpness". That's a fairly bold statement, that's the difference between 1/60 sec and 1/8 sec. Field testing the 7D we found that the AS system worked at moderately slow shutter speeds (with a wide angle lens) but there was a definite drop-off in performance at much slower shutter speeds.

DISCLAIMER: We wanted to test KM's Anti-Shake but aren't completely satisfied that the test in this section of the review can be seen as a 100% representation of the capabilities of the system. The Anti-Shake On / Anti-Shake Off test below was repeated several times to ensure we were getting similar results each time. The comparison test (as you will see) was performed as five shots in sequence to perhaps find some kind of 'mean' result.

Testing Anti-Shake

Test Scene

This is perhaps the most difficult aspect to test, I'll make no claims as to how scientific our results are. To test AS we constructed a simple scene which was placed at head height about 1 m in front of the photographer (me). We lit the room at a constant moderate 7.3 EV of ambient light. A series of shots was taken of the test scene at gradually slower shutter speeds (this was done by increasing the aperture with each shot), each shot was taken twice, once with AS on and once with it off.

Test notes

  • Lighting was simulated daylight measured as 7.3 EV (at ISO 100)
  • Lens used was Minolta 50 mm F1.4 (75 mm equiv. FOV)
  • Camera was hand held by photographer approx. 1 m away from scene
  • Shots taken in sequence
  • Crops below are at 100%
Anti-Shake On Anti-Shake Off
50 mm, 1/60 sec, F3.5
50 mm, 1/30 sec, F5
50 mm, 1/15 sec, F7.1
50 mm, 1/8 sec, F10
50 mm, 1/4 sec, F14

As you can see up to around 1/8 sec (at this focal length, camera held by me) the Anti-Shake system does offer noticeable improvement, I would say that it provides approximately two stops longer shutter speeds than I could hand-hold (your mileage may vary). While it's clear that AS works we weren't totally convinced that it's the complete answer (as in applicable to all situations), or that it's going to be as good as optical (lens element) image stabilization at longer focal lengths.

Some theory about why AS might not be the total solution

  • First of all it's clear that the capability of Anti-Shake to stabilize a shot is related to body movement, lens focal length and shutter speed (we're confident of these three factors playing upon the final result).
  • At telephoto focal lengths the amount of movement will be more for a single fixed point than at wide angle focal lengths. We assume therefore that the CCD movement is some function of body movement multiplied by the magnification of the lens (or thereabouts).
  • KM's CCD Anti-Shake is limited by two factors; firstly the speed at which it can react (how quickly it can move that large, sensitive CCD), and secondly the latitude of movement available to the sensor. Even at telephoto focal lengths a lens element proving 'optical' stabilization only has to move a small amount to produce a big movement at the imaging plane, where as AS will always be trying to 'keep up' with the movement of the body multiplied by the magnification of the lens.

Head to head with Nikon's VR (1/8 sec @ 75 mm)

There was only one way to find out, attempt a like for like comparison. Now this is difficult so I'll place a disclaimer here right away. These test shots were carried out by me in the same stance, the same brace, the same distance from the subject, I tried my hardest to hold both cameras as steady as I could, but I can't say that this is representative of what you may be able to achieve. In this test we chose longer focal lengths and a fairly punishing shutter speed. To be fair to both systems we took the same shot FIVE times in succession.

Test notes

  • Lighting was simulated daylight measured as 7.3 EV (at ISO 100)
  • Camera was hand held by photographer approx. 1 m away from scene
  • Konica Minolta 7D: AS On, 28 - 75 mm Lens @ 75 mm (~113 mm equiv.)
  • Nikon D70: 24 - 120 mm VR Lens (VR On) @ 75 mm (~113 mm equiv.)
  • Exposure (both cameras): ISO 200, 1/8 sec, F10
  • Shots taken in sequence
  • Crops below are at 100%
Konica Minolta 7D (Anti-Shake) Nikon D70 + Vibration Reduction lens

I'll get the obvious out of the way first, the Konica Minolta achieved its stabilization without the use of a special lens which would have potentially cost more. The results are mixed, Nikon's VR appears to be more consistent, with less movement in the blurred shots. The KM's AS system has a lower 'hit rate' but when it does work it produces an acceptably blur-free and perfectly usable result.

Overall Image Quality / Specifics

The Maxxum 7D delivered consistently attractive photographs with good contrast, vibrant color and an almost 'film like' quality which encompasses good dynamic range and also smooth tone in gradients. Resolution appears to be about where we would expect for this particular sensor. It appears to have a fairly strong anti-alias filter (a bit like the Nikon D100) this means that out-of-the-camera processed JPEG's can look slightly softer per-pixel than some other digital SLR's. More resolution and detail are available by shooting RAW and converting through Adobe Camera RAW or other third party RAW conversion tools (I was a little disappointed with the results from DiMAGE Viewer and DiMAGE Master).

Noise at ISO 100 is a little higher than some other digital SLR's but not so much as to be an issue, at higher sensitivities noise (in JPEG's) is controlled by the in-camera noise reduction system which maintains clean flat areas but also appears to reduce sharpness.

Color clipping (sRGB Natural+)

We did detect one minor issue which showed up a sequence of sunset shots I took (this was the only time we experienced this effect). It appears that the camera's Natural+ setting can lead to hard-edged color clipping and artifacting in strong colors. In the example below this can be seen as the ragged pure yellow 'glow' effect around the sun. Converting RAW to JPEG doesn't produce these artifacts (so we know they're not JPEG artifacts).

ISO 100, 1/400 sec, F6.3, sRGB Natural+ ISO 100, 1/160 sec, F6.3, sRGB Natural+