Advertisement

Panasonic LUMIX S5 Lab Test – Rolling Shutter, Dynamic Range and Latitude

September 23rd, 2020 Jump to Comment Section 16
Panasonic LUMIX S5 Lab Test - Rolling Shutter, Dynamic Range and Latitude

We are running the LUMIX S5, the newest addition to the LUMIX S line, through our lab test. Curious to see how it fares? Read on!

Since my colleague Johnnie has brought the Panasonic LUMIX S5 from Japan I was eager to run it through our lab test.

By the way, please have a look at the nice portrait of Ai Kishimoto that Johnnie shot with the Panasonic S5 in Japan here.

If you are not familiar how we test, please head over here.

Rolling Shutter of the Panasonic LUMIX S5

As usual, we are using our strobe light to generate pairs of white and black bars (they appear due to the nature of CMOS sensor readout).

Panasonic LUMIX S5 Lab Test
Panasonic LUMIX S5 rolling shutter measurement in full frame 25p UHD mode: a quite average 21 ms are obtained.

Counting the bars in full frame 25p mode 21ms of rolling shutter are measured. This result is OK but nothing special for a full frame sensor. The Canon R5, for example, has a really good 15.5ms value (in DCI 17:9 full frame mode).

Panasonic LUMIX S5 Lab Test
Panasonic LUMIX S5 rolling shutter measurement in APS-C 25p and 50p UHD mode: very good 10.5 ms are obtained.

In APS-C mode the rolling shutter exactly halves, 10.5ms are measured for 25p and 50p. This is a very good valve for APS-C.

Dynamic Range of the Panasonic LUMIX S5 at ISO 640 in V-Log

The LUMIX S5 features a dual native ISO sensor, with ISO 640 and ISO 4000 being the two native values.

At ISO 640 in V-Log (sharpness and noise reduction set to minimum – “0”) a good 12.1 stops at signal to noise ratio (SNR) of 2 is measured. That is a very solid result.

In the waveform also a 13th and 14th stop is visible, albeit buried in the noise floor.

Panasonic LUMIX S5 Lab Test
Waveform of the Xyla21 stepchart for the Panasonic LUMIX S5 at ISO 640 V-Log in UHD 25p mode. Around 12 stops can be identified above the noise floor.
Panasonic LUMIX S5 Lab Test
IMATEST dynamic range result for the Panasonic LUMIX S5 at ISO 640 V-Log in UHD 25p mode. 12.1 stops are calculated at a signal to noise ratio of 2.

Dynamic Range of the Panasonic LUMIX S5 at ISO 4000 in V-Log

At the second native ISO setting of 4000, a very similar result for the dynamic range is obtained, 11.9 stops at SNR = 2. Also here the waveform shows a 13th and 14th stop within the noise floor.

Panasonic LUMIX S5 Lab Test
Waveform of the Xyla21 stepchart for the Panasonic LUMIX S5 at ISO 4000 V-Log in UHD 25p mode. Around 12 stops can be identified above the noise floor.
Panasonic LUMIX S5 Lab Test
IMATEST dynamic range result for the Panasonic LUMIX S5 at ISO 4000 V-Log in UHD 25p mode. 11.9 stops are calculated at a signal to noise ratio of 2.

Underexposure Test of the Panasonic LUMIX S5

As usual, for our underexposure test, we arrange the scene in such a way that my colleague Johnnie’s face is exposed at a maximum luma value (using a waveform monitor) of 60%. This gives a base exposure of F4 at a 360° shutter angle at ISO 640 using V-Log (sharpness and noise reduction set to “0”).

Then we successively reduce the shutter value to 180, 90, 45, 22.5 and 11.25° giving 5 stops of underexposure.

Panasonic LUMIX S5 Lab Test
LUMIX S5 in UHD, V-Log, 25p, ISO 640 at the zero base exposure level.

In DaVinci Resolve (16.2.5) using the waveform result of the zero base exposure scene, we push up the underexposed images by the respective number of stops using the lift, gamma, and gain controls.

So, let’s jump ahead to the 3 stops underexposed scene, pushed back to zero:

Panasonic LUMIX S5 Lab Test
LUMIX S5 in UHD V-Log, 25p, ISO 640 three stops underexposed and pushed back. The shape of the iris of the SIGMA 45mm 1:2.8 is visible in the image.

Strangely, the shape of the iris of the used SIGMA 45mm 1:2.8 DG DN is visible in the outer areas of the shot (more greenish).

Panasonic LUMIX S5 Lab Test
Shape of the iris of the SIGMA 45mm 1:2.8 DG DN at an f/4. Image credit: CineD

In-camera, the vignetting compensation was turned to “off”. I have no solid explanation for this phenomenon – which is effectively ruining the shot.

Apart from that, the 3 stops under image is quite usable, noise is finely dispersed and can be removed very effectively with noise reduction (if you feel you need it).

Panasonic LUMIX S5 Lab Test
LUMIX S5 in UHD, V-Log, 25p, ISO 640 four stops underexposed and pushed back. Getting noisy, but the noise is very finely dispersed and can be effectively removed by noise reduction.
Panasonic LUMIX S5 Lab Test
LUMIX S5 in UHD, V-Log, 25p, ISO 640 four stops underexposed and pushed back, noise reduction applied (3 frames temporal, threshold = 10). Again the shape of the iris of the SIGMA 45mm 1:2.8 is visible.

At four stops under and pushed back, if it were not for the visible iris the image is still quite usable – remarkable, see above. The noise is finely dispersed and can be easily removed by noise reduction, using 3 frames temporal (threshold = 10) in DaVinci Resolve.

Panasonic LUMIX S5 Lab Test
LUMIX S5 in UHD, V-Log, 25p ISO 640 five stops underexposed and pushed back. The image falls apart but even without noise reduction it still looks surprisingly good. Horizontal stripes appear in the image, however.
LUMIX S5 in UHD, V-Log, 25p, ISO 640 5 stops underexposed and pushed back, noise reduction applied (3 frames temporal, threshold 20, spatial threshold 2). Horizontal stripes appear in the image. But it still looks surprisingly good.

At 5 stops under, pushed back horizontal stripes start to appear in the image, and the point of no return is reached. However, noise reduction again quite effectively removes the noise and you could get away with this image.

Summary

All in all, Panasonic has provided a very, very capable package with the LUMIX S5 – rolling shutter values are OK (full frame) to very good (APS-C), dynamic range results for both native ISO’s are a solid ~12 stops.

And finally, our torture test, underexposing the image using the internal H264 10bit 4:2:2 long GOP codec, produces very good results even at 4 stops underexposure, pushed back to zero. The encoder holds up well, and the noise is finely dispersed such that noise reduction works very effectively.

The only strange phenomenon is the appearance of the iris shape in the images – having the potential to ruin your shot already at 3 stops underexposure. If any of our dear readers have a good explanation for this I would be very interested to hear it.

What do you think about the Panasonic LUMIX S5? Will you consider it to be your next camera? Please share with us your thoughts at the comment section below!

16 Comments

Subscribe
Notify of

Filter:
all
Sort by:
latest
Filter:
all
Sort by:
latest

Take part in the CineD community experience