Image Quality

Our test scene is designed to simulate a variety of textures, colors and detail types you'll encounter in the real world. It also has two illumination modes to see the effect of different lighting conditions.

The GH5S's video-centric sensor is, unsurprisingly, less impressive when compared to cameras focused more on stills. For example, the Panasonic G9 captures far greater levels of detail, even if you compare the images at the same size..

Color rendition looks like it's continuing Panasonic's trend of improvement. Looking at the Raw files, the noise levels seem to be a touch lower than the GH5's, even when compared at a common size: suggesting that the dual gain design, plus the likely use of BSI technology, gives the GH5S a bit of an edge (larger pixels in an of themselves would only be expected to make any differences in very light limited/high ISO images where photons are scarce).

Video stills comparison

Our studio scene is useful for evaluating resolution and image artifacts, though keep in mind that it doesn't give us information about things like motion blur, rolling shutter or how well the footage stands up to color grading, all of which are best evaluated by watching real world footage and working with sample videos (which we do further down the page).

In its UHD mode, you can see the slight difference between near-native capture of data (with an anti-aliasing filter to stop it capturing higher frequencies and producing moiré), and oversampling on the GH5 (which means correctly capturing some of those high frequencies that would cause moiré for the 5S, then blurring them away, to avoid it).

The DCI mode looks pretty similar to the UHD output.

Electronic IS, which is only available when the camera is paired with an OIS lens, lowers resolution capture very slightly. The 1.1x crop may result in a fractionally higher noise level, as well as a loss of field-of-view.

Video quality

Having looked at the performance of the GH5S in low light, here's how it looks in good light. The footage is primarily UHD/23.97p V-Log L footage shot in 10-bit 400Mbps mode. It also includes a UHD/59.94p V-Log L clip, which meant shooting in 8-bit, 150Mbps capture, with exactly the posterization we'd expect from trying to fit lots of dynamic range into just 160 values.

A still grab from 8-bit V-Log L footage. Note the significant posterization in the sky at the top right (you'll probably need to view fullscreen to see).

Panasonic's own V-Log to Rec709 LUT was used for most shots, with additional contrast added in places. For the most part the results look good, with reasonable amount of highlight capture allowing relatively gentle roll-off in the scenes that do feature highlight clipping. However, the final shot, which was shot in both V-Log L and HLG seems to reinforce the idea that you can probably get better results by using HLG, even if you have SDR display as your intended output.

Click here to download a ProRes 422 version of this video to avoid YouTube compression. (6.2GB)

In this instance we used (Panasonic Ambassador) Nick Driftwood's HLG to Rec709 LUT and, on this evidence, would probably prefer this as a starting point for further grading. Interestingly, although the camera will let you shoot VLog L in 8-bit mode, it won't allow HLG capture, which would have necessitated an external recorder for shooting the UHD/60p clip, had I wanted to continue shooting HLG.

Comparison with a7S II

We've already seen how the Sony compares with the GH5S in low light: the Sony is better if you can tolerate shallower depth-of-field, but falls behind if you need to maintain a depth-of-field that the Panasonic can offer. But how do they perform in good light?

Alongside the Panasonic, I also shot the Sony a7S II in its S-Log2 mode, putting middle grey around 40 IRE as most users recommend. Click here to see the waveforms of the two clips in their unprocessed states (S-Log 2 on the left, HLG conformed to the REC709 colorspace on the right).

The Sony's results were a little disappointing, compared with the GH5S, with the 100Mbps codec and 8-bit capture of the Sony resulting in strange color errors creeping into the waves (suggesting an insufficient bitrate to encode the complex motion) and blotchy color errors in the skies at the top-right suggest the limits of the 8-bit capture are also being reached. With better processing technique these problems can probably be made less obvious but there are no such problems with the similarly-exposed Panasonic footage.

Note the slight red, green and blue artefacts appearing in the grey/green waves in the middle distance and also the grey and blue botches at the top right of the left-hand (Sony) image. Again, you'll probably need to view fullscreen to see the difference.

So, while the Sony's larger sensor has an advantage over the Panasonic, this advantage is squandered by the camera being limited to relatively low bitrate 8-bit capture. The Panasonic's files may be large, but we found them to produce final footage with fewer artifacts.

And, although we didn't need to use the camera's second gain mode (which is used from ISO 1600 upwards in both Log modes,) our other experiences and testing suggest that the GH5S will be able to offer good results across a wider range of circumstances than its stabilized sister model.