The GX100 has just four white balance presets in addition to the default automatic mode and a manual (custom) option.
In normal shooting conditions we found the auto white balance to be fairly reliable (if a little on the warm side in bright daylight), though - as as the examples below show - if you want neutral colors under artificial light you're better off switching to manual.
|Incandescent - Auto WB
Red 4.1%, Blue -4.6%
|Incandescent - Incandescent preset WB
Red 0.1%, Blue 0.6%
|Fluorescent - Auto WB
Red 5.0%, Blue -9.7%
|Fluorescent - Fluorescent preset WB
Red 2.3%, Blue -4.3%
Like the GR-D we tested recently the GX100 has a slight tendency to overexpose flash shots if you're not careful, though for normal 'friends in the bar' type shots it's rarely a serious problem (update: just as we were finishing this review Ricoh released a new firmware version that addresses this issue - flash exposures are now fine). The inclusion of a flash hot shoe expands the GX100's shooting versatility considerably, though you'll need to shoot raw because, inexplicably, there's no flash white balance preset.
Slight warm color, good exposure
|Color chart (cropped)
Very slight warm tone
|Macro - 22 x 16 mm coverage
164 px/mm (4155 px/in)
Corner softness: High
Equiv. focal length: 24 mm
|Macro - 20 x 15 mm coverage
180 px/mm (4559 px/in)
Corner softness: Moderate to high
Equiv. focal length: 72 mm
Ricoh has always offered excellent macro capabilities on its digital compacts, and the GX100 is no exception, allowing you to focus down to 1cm at the wide end, capturing an area just over 2 cm across. Even more impressive is that the camera can focus down to 4cm at the long (72mm) end of the zoom, again capturing an area of around 2cm across - and with a less corner softness or distortion.
Barrel and Pincushion Distortion
|Barrel distortion - 1.4 %
Equiv. focal length: 24 mm
|Barrel distortion - 0.3 %
Equiv. focal length: 72 mm
Inevitably with such a wide lens there is some barrel distortion - around 1.4% at the 24mm setting, dropping to barely visible 0.3% by the time you get to the other end of the zoom.
The GX100 features CCD-shift Vibration Reduction, though with a 24-72mm equiv. lens it's far from essential. Our experiences with the system were fairly inconclusive - a reflection of the fact that to really need it at the wide end of the zoom you have to be shooting at 1/30 second or lower, and there are no stabilization systems that come even close to 100% effective at such slow speeds. That said we found the design of the camera itself made camera shake more prevalent than we'd expect (in other words it's hard to hold very steady), and at longer focal lengths there is a small advantage to using it.
The stabilization test
In this simplified version of our SLR IS test, four hand-held shots were taken of a static scene with the stabilization off and on. The shutter speed was decreased and repeated (from 1/125 sec to 1/4 sec). The zoom was set to its maximum position (72mm equiv.), the test target was 2.0m away from the camera. The test was repeated three times and an average taken.
The resulting images were then inspected and given a blur score -
- Sharp (no visible blurring at 100%)
- Mild Blur (the kind of camera shake that is tolerable at normal 'postcard' print sizes)
- Heavy Blur (blur visible even at small print sizes)
- Very Heavy Blur (totally unusable due to camera shake - little if any detail visible).
As the charts below show the VR system offers maybe a 1.5 stop advantage, at best (though how much you get from it will depend on how steady your hand is in the first place). As mentioned above the reason it's no greater is simple; the zoom isn't long enough for most users to need VR at anything under 1/250 sec and we've yet to see a system that can reliably stabilize speeds of 1/8th sec or slower, so there's a fairly narrow band of shutter speeds for the VR to work on.
Where the VR system does help is on the 'borderline' shutter speeds such as 1/125-1/60 at the long (72mm) end of the zoom, where it pretty much guarantees a totally sharp result. Once you get more than one stop below the recommended minimum speed (using the good old reciprocal focal length rule) the VR system increases your chances of getting a sharp - or almost sharp - shot, but it is far from foolproof.
Hand-held (held in one hand), stabilization off (72mm equiv.)
We had no problem getting 100% sharp shots at the long end of the zoom at anything over 1/125 second. Once we dropped below 1/60 sec we couldn't get a totally sharp shot at all. The vast majority of shots below 1/30 sec are totally unusable.
Hand-held (held in one hand), stabilization on (72mm equiv.)
With stabilization on we had no problems at all getting consistently sharp shots at 1/60th second, and even down to 1/8th (3 stops below the recommended minimum) we were getting 1 in 4 totally sharp, with only 1 in 4 shots totally unusable. Basically if you leave the VR turned on and take a few safety shots (and of course hold the camera as steady as you can) you should be able to get usable results right down to 1/8 of a second - or even lower if you've not overdosed on coffee that day.
Specific Image Quality Issues
As with the GR-D my overall impressions of the GX100's image quality are, to be honest, mixed, though for slightly different reasons. Color is vivid (sometimes a little too vivid, though this is something you can tone down with in-camera parameters), focus is generally very accurate and edge to edge detail surprisingly good for such a wide lens (there is some corner softness but this is only really an issue when shooting macro or copying documents at the wide end of the zoom). There is very little purple fringing or chromatic aberration (you'll see a small amount at the 24mm setting but you'll need to look very hard).
One the downside the JPEG output is a bit soft (particularly at apertures over f5.6, when diffraction kicks in) and looks a bit over sharpened (in fact at 100% it looks over-processed full stop), but for standard print sizes this - and the noise/noise reduction artefacts (slight mushing of fine low contrast detail) - isn't going to be a big issue. Much more problematic is the higher than average highlight (and color) clipping in bright conditions, something that's not helped by the rather erratic exposure system that is far too happy to overexpose for our liking. You have to be very careful indeed with exposure, pushing the AE compensation down by as much as a stop to preserve some highlight detail, though even then there's only so much you can do with such limited dynamic range (an issue that would seem to support our suspicion that the GX100's 10MP sensor is even less sensitive than that used by its competitors).
The only way to deal with the issue of clipping (and the occasional white balance issues) is to be very careful with exposure, shoot raw and be prepared for some fairly hefty post processing, but of course this brings a whole new set of problems, not least the extra 5.5 second you'll need to wait between shots.
Highlight clipping / channel clipping / over exposure
The GX100's JPEG dynamic range appears to be very limited indeed, and the camera struggles to capture highlight and shadow detail on bright days, something that's not helped by the rather steep default contrast curve or the fact that the exposure system tends to overexpose contrasty scenes. As well as straightforward highlight clipping we saw quite a lot of posterization of primary colors on bright days (channel clipping). We'd suggest trying the lowest contrast setting and using negative AE-C to retain highlights if you're shooting JPEGs in bright weather (though of course this also means pretty much every shot is going to need post-processing), or to get a little more dynamic range out of the sensor switch to shooting raw.
|100% crop||72mm equiv, F6.3, default exposure|
|100% crop||72mm equiv, F7.1, -0.7 EV|
|100% crop||72mm, F5.6|
|24mm equiv., F5.1||24mm equiv., F3.2|