Panasonic G9 Image Quality


Color

Saturation & Hue Accuracy
Typical mean saturation levels, with good hue accuracy.

In the diagram above, the squares show the original color, and the circles show the color that the camera captured at base ISO. More saturated colors are located toward the periphery of the graph. Hue changes as you travel around the center. Thus, hue-accurate, highly saturated colors appear as lines radiating from the center. Click for a larger image.

Saturation. The Panasonic G9 produced a default mean saturation of about 109.3% (9.3% oversaturated) at the base ISO of 200, which is fairly typical. The Lumix G9 pumps dark blues and dark red moderately, some other colors mildly, but undersaturates yellow a bit more than we'd like. Still, overall default saturation levels are quite pleasing and keep in mind saturation can be adjusted. Most consumer digital cameras produce color that's more highly saturated (more intense) than found in the original subjects. This is simply because most people like their color a bit brighter than life.

Skin tones. The Panasonic G9 produced natural-looking Caucasian skin tones with a slight boost to pinks with either Auto or Manual white balance in simulated daylight. A good job here. Where oversaturation is most problematic is on Caucasian skin tones, as it's very easy for these "memory colors" to be seen as too bright, too pink, too yellow, etc.

Hue. The Panasonic G9 shifted orange toward yellow moderately and yellow toward green mildly, while cyan was shifted toward blue by quite a bit (for better-looking skies), but other hue shifts were quite minor. The typical Panasonic yellow to green shift and desaturation still exists producing slightly dingy yellows, but it's not as pronounced as some earlier models. The G9's mean "delta-C" color error after correction for saturation was 4.67 at the base ISO of 200 (100 is an extended ISO). That's slightly better than average hue accuracy these days. Hue is "what color" the color is.

See full set of test images with explanations
See thumbnails of all test and gallery images

Sensor

Exposure and White Balance

Indoors, incandescent lighting
Warm color casts with default Auto and Incandescent white balance settings. Very good color balance with the Manual setting. Average amount of exposure compensation required.

Auto White Balance (AWB)
+0.3 EV
Auto White Balance (AWBc)
+0.3 EV
Incandescent White Balance
+0.3 EV
Manual White Balance
+0.3 EV

Indoors, under normal incandescent lighting, color balance was warm with the default Auto white balance (AWB) setting, however when AWBc was selected, colors were more neutral though with a slight magenta tint. Results with the Incandescent setting were much too warm, with a strong orange-yellow cast. The Manual (Custom) white balance setting produced very neutral results. The Panasonic G9 required +0.3 EV exposure compensation here, which is about average for this shot. (Our test lighting for this shot is a mixture of 60 and 100 watt household incandescent bulbs, a pretty yellow light source, but a very common one in typical home settings here in the U.S.)

Outdoors, daylight
Natural colors overall, with a tendency toward high contrast under harsh lighting. About average exposure accuracy.

Auto White Balance,
+0.7 EV

In simulated daylight, the Panasonic G9 performed quite well. Skin tones were realistic in our "Sunlit" Portrait shot, with a healthy-looking pinkish tint. Exposure accuracy was about average, as the camera required +0.7 EV compensation to keep facial skin tones reasonably bright. That's typical for this shot. Despite the bright appearance of the mannequin's white shirt, very few highlights are actually blown which is quite good, though there are some very deep shadows that contain very good detail, but are discolored with a greenish tint.

Native Resolution
~ 2,700 to 2,750 lines of strong detail.

Strong detail to
~2,700 lines horizontal
Camera JPEG
Strong detail to
~2,750 lines vertical
Camera JPEG
Strong detail to
~2,700 lines horizontal
ACR converted raw
Strong detail to
~2,750 lines vertical
ACR converted raw

In a best-quality camera JPEG, our laboratory resolution chart reveals sharp, distinct line patterns to just over 2,700 lines per picture height horizontally, and to about 2,750 lines in the vertical direction. (Some might argue for higher, but lines begin to merge and aliasing artifacts begin to interfere at these limits.) Complete extinction of the pattern occurred between 3,500 and 3,600 lines per picture height in both directions. We weren't able to extract significantly more resolution by processing the Panasonic G9's RW2 file using Adobe Camera Raw, and the ACR conversion also shows much higher color moiré which is practically nonexistent in the camera JPEG.

Use these numbers to compare with other cameras of similar resolution, or use them to see just what higher resolution can mean in terms of potential detail.

Sharpness & Detail
Very good sharpness overall, with only minor edge-enhancement artifacts on high-contrast subjects. Signs of mild noise suppression visible in the shadows at base ISO.

Very good definition of high-contrast
elements, with just slightly visible
sharpening artifacts.
Subtle detail: Hair
Noise suppression tends to blur
detail in areas of subtle contrast.

Sharpness. The Panasonic G9 captures sharp, detailed images with a sharp lens. Some minor edge enhancement artifacts are visible on high-contrast subjects such as the sharpening "haloes" along the lines and text in the crop above left, but default sharpening appears to be well judged and not too overdone. Edge enhancement creates the illusion of sharpness by enhancing colors and tones right at the edge of a rapid transition in color or tone.

Detail. The crop above right shows relatively mild noise suppression artifacts such as the smudging of individual strands together in the darker areas of the mannequin's hair, as well as in areas with low local contrast. This is very good noise versus detail processing performance for a 20-megapixel Micro Four Thirds model, leaving plenty of detail intact instead of blurring much of it away in an attempt to suppress noise. Do however notice the aliasing artifacts ("jaggies") and demosaicing errors in the hair. This is unfortunately quite common, especially from cameras without an optical low-pass filter like the G9. The G9's context-sensitive anti-noise processing can also produce somewhat unnatural looking artifacts in some cases (as mentioned below), however overall we think it does a great job at holding onto detail while keeping noise in check. Noise-suppression systems in digital cameras tend to flatten-out detail in areas of subtle contrast. The effects can often be seen in shots of human hair, where the individual strands are lost and an almost "watercolor" look appears.

RAW vs In-Camera JPEGs
As noted above, the Panasonic G9 produces sharp in-camera JPEGs with very good detail. As is almost always the case, better detail can be obtained from carefully processing RAW files than can be seen in the in-camera JPEGs, with fewer sharpening artifacts to boot. Take a look below, to see what we mean:

Base ISO (200)
Camera JPEG, defaults
RAW via Adobe Camera Raw

In the table above, we compare an in-camera JPEG taken at base ISO using default noise reduction and sharpening (on the left) to a matching RAW file converted with Adobe Camera Raw 9.1 via DNG Converter 10.4, using default noise reduction with some fairly strong but tight unsharp masking applied in Photoshop (300%, radius of 0.3 pixels, and a threshold of 0).

As you can see, the in-camera JPEG contains very good detail (among the highest we've seen from a MFT camera thus far) with better contrast, however as is usually the case, ACR extracted additional detail, particularly in the red-leaf fabric where it managed to resolve more of the fine thread pattern. (The camera's context-sensitive noise reduction blurred some areas within the fabric more than others producing a less consistent, slightly unnatural-looking rendering.) However, the ACR conversion also contains much higher luminance noise after sharpening, especially noticeable in flatter areas with little detail. You can always turn up the luminance noise reduction (default of zero was used here), though, or process the files in your favorite noise reduction program or plugin if you find the noise objectionable.

Bottom line, though, while the G9's JPEG engine does a very good job and Panasonics are getting better with each iteration, shooting in raw mode and converting them yourself can still yield better detail and provides more control than in-camera JPEGs.

ISO & Noise Performance
Excellent high ISO performance for a Micro Four Thirds model.

Default Noise Reduction
ISO 100 ISO 200 ISO 400
ISO 800 ISO 1600 ISO 3200
ISO 6400 ISO 12,800 ISO 25,600

The Panasonic G9's images are very detailed and clean at ISOs 100 (extended) through 400, with only minor luminance and chrominance noise detectable in the shadows. ISO 800 shows slightly higher luminance noise and stronger noise reduction, blurring some very fine detail in the process, though overall detail remains quite strong. ISO 1600 is of course a little noisier and softer, but fine detail is still very good. ISO 3200 shows increased blurring and loss of detail from stronger noise reduction which starts to produce a bit of a digital-looking "crystalline" effect in flatter areas. However chroma noise remains well-controlled. At ISO 6400 and above, noise and noise reduction artifacts become progressively stronger while fine detail is reduced to the point where very little fine detail is left in the hair at the maximum ISO of 25,600. Chroma noise remains well-controlled up to ISO 6400, but some chroma blotching is visible at higher ISOs. There's also a peppered effect at ISO 12,800 and especially ISO 25,600 from darker noise pixels.

Still, high ISO performance overall appears to be excellent for a Micro Four Thirds camera and significantly improved over its predecessor, the G85, despite the smaller pixels. We're of course pixel-peeping to an extraordinary extent here, since 1:1 images on an LCD screen have little to do with how those same images will appear when printed. See the Print Quality section below for our evaluation of maximum print sizes at each ISO setting.

Extremes: Sunlit, dynamic range and low light tests
Somewhat high default contrast but with very good dynamic range. Excellent low-light focus performance.

+0.3 EV +0.7 EV +1.0 EV

Sunlight. The Panasonic G9 performed well with the deliberately harsh lighting of this test. Contrast is a little high at its default setting, but dynamic range is quite good in JPEGs. We felt the +0.7 EV exposure is the best compromise here. Although shadows around the eyes are a bit dark, we prefer it to the +1.0 EV exposure overall, as the latter is a bit too bright (although surprisingly few highlights are blown even at +1.0 EV). It's really the photographer's choice here as to which direction to go in. For those Panasonic G9 owners that are going to want to just print an image with little or no tweaking, the +1.0 image would probably produce a better-looking face uncorrected. The bottom line though, is that the Panasonic G9 performed well with the wide dynamic range of this shot even without the use of Intelligent D-Range or Highlight/Shadow tweaks (see below), especially for a Four Thirds sensor.

Because digital cameras are more like slide film than negative film (in that they tend to have a more limited tonal range), we test them in the harshest situations to see how they handle scenes with bright highlights and dark shadows, as well as what kind of sensitivity they have in low light. The shot above is designed to mimic the very harsh, contrasty effect of direct noonday sunlight, a very tough challenge for most digital cameras. (You can read details of this test here.)

Face Detection
Off at 0 EV
Aperture priority, f/8
On at 0 EV
Aperture priority, f/8
Full Auto
f/2

Face Detection. Like most cameras these days, the Panasonic G9 has the ability to detect faces, and adjust exposure and focus accordingly. As you can see from the examples above, it worked well, as the center image with face detection enabled is much better exposed for the face without having to use exposure compensation, though some highlights were blown. The Full Auto setting chose Portrait scene mode which selected a larger aperture (f/1.6) for much less depth-of-field and better subject isolation. It too was much brighter than Aperture Priority without face detection, though again, quite a few highlights are blown.

Intelligent D-Range Comparison

Panasonic's Intelligent Dynamic Range
The above shots are examples of Panasonic's Intelligent Dynamic Range Control (or iD-Range) at work, with no exposure compensation. Note that the camera does not take multiple shots and merge them as HDR mode does (see below). It's a system that adjusts local contrast and exposure more akin to Nikon's Active D-lighting, Canon's Automatic Lighting Optimization or Sony's Dynamic Range Optimization.

There are three levels of iD-Range available on the Panasonic G9: Low, Standard and High, plus Auto and Off. It's automatically invoked in iAuto and manually selectable in PASM modes. Here, you can see enabling iD-Range made a huge difference to the exposure without requiring any exposure compensation (it appears to use smarter metering and detected a face). Auto did a good job here, and darker midtones and shadows were progressively boosted as the strength was increased, although some highlights were blown in the process. As a result, though, noise is a little more visible, but that's expected.

Highlight/Shadow Adjustment
Above, we can see the G9's preset Highlight and Shadow control options at work on our "Sunlit" Portrait test shots. This feature lets you adjust the tone curve using 4 preset options (Standard, Higher Contrast, Lower Contrast and Brighten Shadows), plus it gives you 3 Custom settings which let you adjust the curve on a grid. This gives you individual control of the highlight, midtone and shadow portions of the tone curve, unlike the regular contrast adjustment. Mouse over the links to load the associated thumbnail, and click on the links to visit the full resolution image. (Apologies for the dim images, we should have applied some positive exposure compensation here.)


HDR Comparison

HDR mode
Here, you can see the Panasonic G9's High Dynamic Range mode at work. HDR mode takes three images at different exposures and combines them to increase dynamic range. Four settings are available, ranging from Auto to up to +/-3 EV of exposure difference. Mouse over the links, and click on them the view the full resolution files.

Auto and +/-2 EV produced very similar results, while +/-3 dimmed the entire image, so the feature did not work very well for these shots. (Again, we should have used some positive exposure compensation here.) Be aware that double images and ghosting are possible when elements in the scene move between frames. Also notice the angle of view is narrower in the HDR images, likely because the images have been cropped and upsized during the optional auto alignment process.

Dynamic Range Analysis (RAW mode)
While we once performed our own dynamic range measurements based on in-camera JPEGs as well as converted RAW images (when the camera was supported by Adobe Camera Raw), we've switched to using DxO Labs' results from their DxOMark website. As technology advanced, the dynamic range of modern high-end cameras in some cases exceeded the range of the Stouffer T4110 density scale that we used for our own measurements. DxO's approach based on RAW data before demosaicing is also more revealing, because it measures the fundamental dynamic range of the sensor, irrespective of whatever processing is applied to JPEGs, or to RAW data by off-the-shelf conversion software.

In the following, we use DxO's "Print" dynamic range results, which are scaled based on camera resolution. As the name suggests, this scaling corresponds to the situation in which you print at a given size, regardless of how many megapixels the camera might have. (In other words, if you've decided to make a 13x19 inch print, that's the size you're printing, whether the camera's resolution is 16 or 300 megapixels.) Also note that DxO Labs uses a signal-to-noise (SNR) threshold of 1 when defining the lower boundary of acceptable luminance noise in their dynamic range measurements, which corresponds to the "Low Quality" threshold of the Imatest software we used to use for this measurement.

Unfortunately, DxOMark has not yet tested the Panasonic G9 as of this writing, but we'll come back and update this section once they do. In the meantime, Photons to Photos has published results for the G9, and it appears to offer significantly improved dynamic range compared to the GH5, while being comparable to the Olympus E-M1 Mark II.

Low-light Autofocus
The G9's autofocus system was able to focus on our legacy low-contrast AF target down to -4.8 EV unassisted with an f/2.8 lens, and down to about -6 EV with our newer high-contrast target. This is excellent performance, especially for a camera with contrast-detect autofocus. The Panasonic G9 also has a focus-assist light which allows it to autofocus in total darkness, as long as the subject is within range and has sufficient contrast.

NOTE: This low light test is conducted with a stationary subject, and the camera mounted on a sturdy tripod. Most digital cameras will fail miserably when faced with a moving subject in dim lighting. (For example, a child's ballet recital or a holiday pageant in a gymnasium.) Thanks to their phase-detect AF systems, digital SLRs tend to do much better than point & shoots, but you still shouldn't expect a quick autofocus lock with moving subjects. The G9 uses contrast-detect autofocus, as is found in most point & shoot cameras, so its low-light focusing ability may be less than that of some cameras with phase-detect systems. That said, though, the larger, more sensitive pixels of the G9's sensor do better under dim lighting than do the tiny pixels of most point & shoots.

Output Quality

Print Quality
Excellent 30 x 40 inch prints at ISO 100/200; a good 16 x 20 inch print at ISO 800, and a solid 11 x 14 at ISO 3200.

ISO 100/200 deliver superb prints at 30 x 40 inches, and even larger if your viewing distance allows given the 20-megapixel resolution. The prints provide crisp detail and impressive color rendering, with plenty of three dimensional pop to please discerning eyes. Very pleasing prints overall!

ISO 400 yields a very good 24 x 36 inch print, displaying similar characteristics to the 30 x 40 inch prints at base and extended low ISOs. The larger 30 x 40 inch prints are certainly usable here as well, but mild softening is detected in certain areas of the prints upon close scrutiny, so we'll put our seal of approval on the 24 x 36 inch prints at this gain setting.

ISO 800 prints are quite good at 16 x 20 inches, which is still quite a nice-sized print given the gain is beginning to climb. On close inspection there is a mild softening in the red channel, typical of most cameras by this ISO sensitivity and print size, and just a touch of noise in flatter areas of our test target, but still a very good all-purpose print. It's well worth noting that the 20 x 30 inch prints here are quite usable for wall display purposes and less critical applications as well.

ISO 1600 manages a good 16 x 20 inch print, displaying a touch more visible noise in some areas of our target, but not enough to miss out on our good grade. However, for your most critical printing applications at this gain setting we suggest 13 x 19 inches to be sure.

ISO 3200 sees the strain of cranking the gain increase, as the G9 is only able to manage a workable 11 x 14 inch print. This is still a fairly nice-sized print, but for fine art printing to 13 x 19 inches and higher you obviously won't be able to rely on ISO 3200 or above, as there is simply too much apparent noise at the larger print sizes to yield dependable prints.

ISO 6400 delivers a good 8 x 10 inch print, but there is a noticeable lack of "pop" and richness in color overall, and this is common for most all but full-frame and larger sensors here. For this reason we suggest limiting the G9 to either ISO 1600 or 3200 for critical printing purposes. Still, you can get a good 8 x 10 inch print at this gain setting in a pinch.

ISO 12,800 prints a decent 5 x 7, and that's not too bad given how high this ISO is for this sensor size. At this size the detail is good enough for anything but critical applications, though anything larger begins to show too much in the way of noise and some obvious smearing effects from anti-noise processing.

ISO 25,600 comes very close to allowing a good 4 x 6 inch print, perhaps closer than any MFT camera to date. There is still reasonably full color reproduction and not the typical burnt look that haunts many cameras at this sensitivity, but just not quite enough fine detail to pass our good grade. Still, and more than most crop-sensor cameras, you can get decent results at 4 x 6 inches here for less-critical applications if needed.

Overall, the Panasonic G9 turns in a solid performance for print quality. Given that this camera is marketed for the most part at still photographers, and given its competition with cameras such as the Olympus E-M1 Mark II, the G9 does a good job of holding its own against both that model and the storied Panasonic GH5. You can safely expect very good prints at large sizes up to ISO 1600, and still get useful printed results through ISO 3200. Beyond that, you're not likely to get great printed results for your most important purposes, so heeding these settings in the field will help ensure good printed results.

 

The images above were taken from our standardized test shots. For a collection of more pictorial photos, see our Panasonic Lumix DC-G9 Photo Gallery .

Not sure which camera to buy? Let your eyes be the ultimate judge! Visit our Comparometer(tm) to compare images from the Panasonic Lumix DC-G9 with those from other cameras you may be considering. The proof is in the pictures, so let your own eyes decide which you like best!



Editor's Picks