Fairly a while after the discharge of the LUMIX GH5 in early 2017, the successor GH5 II has now arrived at our lab. It claims an enhanced efficiency due to its new sensor structure and up to date Venus engine. Curious to see the way it performs? Then learn on…
Every time I hear a couple of new launch of a Micro 4 Thirds (MFT) sensor-based digital camera, I get actually excited. As I discussed in some earlier articles, my love for bigger sensor cameras began with the LUMIX GH1 within the 12 months 2009. MFT sensor-based cameras are tremendous versatile by way of lens adaptation – they sometimes present low rolling shutter values and fare fairly OK within the dynamic vary division.
Therefore, since 2009 there is no such thing as a (gear) excuse anymore as all of those MFT-based cameras shoot lovely footage 😉 Simply take a look at Johnnie’s assessment and mini-documentary he shot with the LUMIX GH5 II right here.
After I examined the LUMIX GH5 in 2018, I used to be a bit disenchanted because it confirmed solely 9.7 stops of dynamic vary at a sign to noise ratio of two, and 10.8 stops at SNR = 1. Different MFT sensor-based cameras just like the GH5S, or the Z CAM E2 exhibit a couple of cease extra, and the lately examined LUMIX BGH1 displayed even 11.6 stops at SNR = 2.
So let’s see what the revamped GH5 II has to supply.
Rolling Shutter of the LUMIX GH5 II
As common, we apply our 300Hz strobe mild to create pairs of black and white stripes from prime to backside of the recorded picture, an artifact that’s typical for CMOS sensors and permits the measurement of rolling shutter values.
A quite good 13.9ms of rolling shutter is obtained for 25 and 50p, as could be seen under for 4K DCI mode (7% much less image top than UHD):
13.9ms is a tad worse than the LUMIX BGH1 (11.9 in UHD) and the Z CAM E2 (11.1ms in 4K DCI) however general actually good.
Dynamic Vary at ISO400 and ISO2500 of the LUMIX GH5 II
In case you are not conscious how we check dynamic vary, please take a look on the documentation of our procedures right here.
At ISO400, utilizing 4096×2160 25p mode in V-Log L, the next waveform plot is obtained when taking pictures our Xyla21 chart:
As could be seen with V-Log L, clipping happens already under the 80% luma worth. About 11 stops could be recognized above the noise ground, and a weak twelfth and even weaker thirteenth cease are seen contained in the noise ground.
Imatest confirms this discovering, 10.5 stops are calculated for SNR = 2 and 11.5 stops at SNR = 1 for ISO 400, V-Log L:
For one, these outcomes are 0.8 stops higher than for the unique GH5, and second, about 2 extra stops above the blue SNR = 1 line (11.5) within the center graph are recognized. If the codec of the internally recorded H264 10bit 4:2:2 recordsdata is robust sufficient to seize additionally particulars throughout the noise ground, these extra 2 stops could be “excavated” utilizing superior noise discount methods in postproduction. Spoiler alert! Sure, the codec permits digging into the shadows, as will likely be revealed within the latitude part under.
I additionally checked out greater ISO values, and what I discovered is that the dynamic vary values on the SNR = 2 and SNR = 1 thresholds turn into decrease for every improve of ISO. For instance at ISO2500 the next waveform is obtained:
Photos turn into noisier (which was anticipated), and about 10 stops could be seen above the noise ground. IMATEST confirms this, verify the end result under. Dynamic vary values drop to 9.25 stops at SNR = 2 and 10.6 stops at SNR = 1:
Subsequently, as a result of this sensor solely options one native ISO worth at ISO400, watch out in decrease mild conditions as greater ISO values affect the dynamic vary fairly a bit.
Latitude Take a look at of the Panasonic LUMIX GH5 II
As written in earlier lab reviews, latitude describes the potential of a digital camera to retain colours and particulars when over or underexposed and normalized again to a zero baseline publicity stage in put up.
In our case, we’ve a normal studio scene the place the bottom publicity is (arbitrarily) set to 60% luma worth on the waveform for the brow of our expertise, on this case, my expensive colleague Johnnie. Digital camera settings had been once more ISO400, V-Log L in 4K DCI mode:
As could be seen within the waveform plot of the Xyla21 chart within the earlier part, this setting is already fairly sizzling for V-Log L, as solely 2 extra stops can be found for overexposing earlier than clipping happens:
The purple channel at Johnnie’s brow is near the clipping worth of 768 within the 10bit house (V-Log L). On the left aspect, some patches of the Datacolor Spyder are already clipping.
From there we underexpose the scene by stopping down the iris of the lens from T1.5 to T8, after which onwards by decreasing the shutter angle from 360° to 180°, 90°, and 45° levels – giving us 8 stops of publicity variation. Let’s see if the digital camera can address that. For reference, the shadows on our customary studio scene are 5 stops under the topic’s face.
Wanting on the 3 stops under base publicity, pushed again picture already noise is current within the picture, and the 4 stops under base is quite noisy:
Nevertheless, colours are retained effectively, and the noise exhibits a tremendous construction with out bigger blotches of chroma noise, therefore the picture could be cleaned up very properly with noise discount:
At 5 stops below base publicity, pushed again the picture is on the cusp of breaking apart. It exhibits a pinkish tone, primarily as a result of heavy chroma noise which is in all places:
Nevertheless, the noise is once more finely dispersed, there aren’t any artifacts seen like horizontal or vertical stripes, therefore noise discount can nonetheless clear this picture up moderately effectively and with that additionally the pinkish tone (from the chroma noise) could be tamed. Nevertheless, transitions into deeper shadows nonetheless have this pinkish tone – which appears fairly unnatural.
So, at this level the boundaries of usability are undoubtedly reached.
The 6 stops below base, pushed again picture appears like this:
And right here, even heavy noise discount can not reserve it – there are these bigger blotches of luma & chroma noise which might be simpler to determine within the transferring photographs than within the static, downscaled ones we present right here, and general colours are getting bizarre. That’s then “recreation over”:
Nonetheless, the outcomes are fairly superb for an MFT sensor, to be trustworthy. Publicity latitude for the GH5 II in our customary CineD studio scene is 7 stops (2 over to five below) – lifting it to an identical stage just like the lately examined full body Sony a7S III – having a 4 instances greater sensor (see our check right here)! The lately examined LUMIX BGH1 digital camera for instance was hampered by robust horizontal strains within the underexposed photographs, therefore solely 6 stops of publicity latitude had been measured right here.
And that’s primarily as a result of truth, that there aren’t any seen artifacts like horizontal or vertical strains, and the interior codec manages to encode the tremendous noise in closely underexposed photographs such that noise discount in postproduction can save the pictures.
For reference, current full-frame cameras just like the Panasonic S1H, S1, and S5 (lab check right here) or Sony A1 (lab check right here) handle 8 stops of publicity latitude and sometimes present about 1.5 extra stops of dynamic vary. And top-tier cinema cameras just like the lately examined Arri ALEXA Mini LF handle 10 stops of publicity latitude (lab check right here), and about 3 stops higher dynamic vary.
The Panasonic Lumix GH5 II undoubtedly is a significant enchancment over its predecessor, the LUMIX GH5. It fares fairly effectively in all departments, be it a low rolling shutter, OK dynamic vary outcomes, and good latitude capabilities.
Once more, Panasonic exhibits the capabilities of contemporary MFT cameras paired with stable inner recording leading to a really sturdy, natural image that may be pushed round exhausting in put up with out breaking apart.
Have you ever shot with the LUMIX GH5 II already? What are your experiences? Share your ideas within the feedback part under.