Panasonic LUMIX S1II Lab Test – Rolling Shutter, Dynamic Range, and Exposure Latitude

Lab Test

by Gunther MachuMay 13^th, 2025

Panasonic has just released the successor to their popular LUMIX S1 full-frame camera – the LUMIX S1II. Time to run it through our standard CineD Lab Test to verify the bold claim of 14+ stops of dynamic range in their announcement. However, in the course of trying to validate that statement, we found a bug in the current firmware that needs to be fixed by the company. So, please note that this Lab Test is incomplete for now. Curious to see the results? Then read on …

The announcement of the LUMIX S1II piqued my interest, as I’m a long-term LUMIX S1 shooter myself (I began using it in October 2019, and have written several articles on CineD about my wildlife adventures using the LUMIX S1, which you can read here and here). I always wondered whether Panasonic would implement the “Dynamic Range Boost” feature previously seen in the LUMIX GH6, G9II, and now GH7 in their line of full-frame cameras. Well, the day has finally arrived!

And as always, I must give a huge shout-out to my colleague Florian, who helped shoot and analyze the test results. Please head over here for the announcement and the impressive list of specs.

Rolling shutter of the LUMIX S1II

As usual, we used our strobe light to create the sequences of black-and-white bars, which are characteristic of the read-out nature of rolling shutter sensors.

Now, typically with Panasonic cameras, there is a big (big) list of shooting modes available: full frame, APS-C crop, and pixel-to-pixel modes in all the different resolutions and framerates (up to 4K/120p).

All these modes have a different rolling shutter readout. Also, the rolling shutter changes depending on whether “Dynamic Range Boost” mode is “ON” or “OFF”. Hence, the combinations are almost endless.

However, what struck us is the following:

We started the analysis with 12-bit internal 5.8K ProRes RAW—only available in pixel-by-pixel mode—resulting in a slight crop to 5760 horizontal pixels from the full 6000×4000 sensor. The rolling shutter we were getting was the same fast readout speed with DR Boost “ON” or DR Boost “Off”.

Then we switched to 5.8K as well as 4K Long GOP (H265) and ProRes HQ full frame modes, and were getting more than double the rolling shutter with DR Boost ON: 27.5ms, than with DR Boost OFF: 12.7ms:

Note: This actually led us to reach out to Panasonic and ask what is going on here. And they confirmed that we indeed found a bug: DR Boost ON is not properly activated in 5.8K ProRes RAW mode, explaining the low rolling shutter in that mode. Panasonic has recognized the issue with the sample units, and this matter is going to be resolved in the mass production versions. Hence, this also means that our latitude test (see below) was actually done without the DR Boost ON function properly activated!

In 4K/25p ProRes RAW (again pixel-by-pixel mode, DR Boost “ON”) – hence, effectively APS-C – the rolling shutter is 18.7ms, as well for long GOP mode (e.g. H265) in APS-C.

In summary, the DR Boost OFF rolling shutter values are quite good in full frame and APS-C modes, but the DR Boost ON function comes at the cost of severely increasing the rolling shutter.

Please head over to the CineD Camera Database to see all other sensor modes and DRB ON / OFF combinations.

Dynamic range of the LUMIX S1II

As ProRes RAW 5.8K on the sample unit doesn’t work properly for now, let’s start with a waveform of 4K ProRes RAW 25p (pixel-by-pixel hence actually APS-C), with DR Boost “OFF” (with a base ISO640), developed to V-Log in Final Cut Pro to understand what DR Boost is actually doing, have a look on the left side below:

With DR Boost “OFF” (on the left), we can see 12, if not 13 stops above the noise floor, including a 14th and 15th stop, the base ISO is 640, as with the previous generation LUMIX S1. Clipping occurs slightly above 896 (or around 90% luma value). Now, if we turn “ON” the Dynamic Range Boost, the base ISO moves to ISO1000, and we get the waveform shown on the right side above, with all the code values moving up, and clipping occurs around a code value of 960. Hence, we gain close to a stop in the highlights with DR Boost ON.

Let’s have a look at IMATEST scores: with DR Boost OFF in 4K ProRes RAW 25p, we get 11.5 stops at a signal-to-noise ratio of 2 (SNR), and 12.8 stops at SNR = 1. With DR Boost ON, we now get a superb 13.1 stops at SNR = 2 and 14.3 stops at SNR = 1, on the right side below!!

You can also see in the lower left-hand diagram of IMATEST (Pixel noise normalized) that the image is actually less noisy with DR Boost ON.

Now, let’s have a look at some of the compressed codecs. 4K full frame ProRes HQ DR Boost ON mode subsampled from the 6K sensor should give the best dynamic range result – and we are seeing 12.9 / 14.1 stops at SNR 2 / 1 – wow!

That’s close, but not quite the result from ProRes RAW 4K crop. Interesting to see that ProRes RAW with DR Boost ON is better than the compressed codecs – typically compressed codecs have more internal noise processing, hence IMATEST results are better! That means (in the future) 5.8K full frame ProRes RAW should yield similar results – to be tested at a later stage.

Now looking at 4K APS-C ProResHQ DRB ON we are getting 12.7 / 13.9 stops at SNR = 2 / 1 the same as in 5.8K ProRes HQ full-frame mode where we get 12.6 / 13.9 stops at SNR = 2 / 1 (as is to be expected – pixel by pixel the dynamic range should stay the same).

One word though on the compressed codecs, let’s look at the waveform for 5.8K ProRes HQ (DRB ON):

The compressed codecs like ProRes HQ show a lot of in-camera noise reduction that cannot be turned off (in V-Log, we made sure it was at zero, which is the default). Have a look at the noise floor of the waveform plot above (5.8K ProResHQ DR Boost On) – that’s way too clean and clearly too much NR for my personal taste.

Another observation we have not talked about yet is that in ProRes RAW, the second native ISO of 5000 is unavailable (you can select 5000 ISO, but in the camera menu, the second base ISO is greyed out). Hence, let’s have another look at 5.8K ProRes HQ at the now available second base ISO of 5000 (but now DR Boost ON is not available):

At the second base ISO of 5000 in 5.8K 10-bit ProRes HQ, we are getting 11 / 12.5 stops at SNR = 2 / 1 which is about half a stop less when compared to the first base ISO result at ISO640 which is 11.7 / 12.9 at SNR = 2 / 1 (comparing it to DR Boost OFF).

Latitude result of the Panasonic LUMIX S1II

As stated in previous articles, latitude is the capability of a camera to retain details and colors when over- or underexposed and pushed back to a base exposure. This test is very revealing, as it pushes the complete image pipeline of any camera to its absolute limits, not just in the highlights but mainly in the shadows.

We chose 12-bit ProRes RAW in 5.8K DR Boost ON mode for this latitude test (as mentioned above, we only found out later with Panasonic that DR Boost is not properly activated in this mode). Hence, all the results below are actually with DR Boost “OFF”!!

As usual, the latitude test is done in DaVinci Resolve, with the complication that there is no support for ProRes RAW yet. Hence, we would normally use the RAW Convertor app, which can transcode the PRR files to Cinema DNG, which can be used in DaVinci. However, this app didn’t recognize the camera, hence it didn’t transcode the files. Therefore, we developed the files in Final Cut Pro to 12-bit ProRes XQ 4444 files (and also adjusted the exposure with the exposure slider and then the ISO), and then imported them into DaVinci Resolve.

Our studio base exposure is (arbitrarily) chosen as having an (ungraded) luma value of around 60% on the forehead of our subject on the waveform monitor – in this case, my colleague Johnnie:

From here, we can overexpose by 5 stops:

The red channel on Johnnie’s forehead is at the cusp of clipping but still intact.

Now let’s underexpose by stopping down the iris of our ZEISS Compact Prime 85mm T1.5 in one-stop increments until T8, and then doubling the shutter value. All the files are then normalized back again to the base exposure levels (using the exposure slider and changing the ISO setting when developing the files in Final Cut Pro).

In general, the ProRes RAW files have fine noise, and at 3 stops underexposure, significant noise starts to creep into the image:

This type of noise is easy to remove in DaVinci Resolve. Let’s not forget we are already at 8 stops of exposure latitude, a value that is typically the limit of most consumer full-frame cameras:

As can be seen, the noise cleans up nicely, and we get a perfectly usable image. There is a slight color shift towards green, but all is fine other than that.

Now let’s move to 9 stops of exposure latitude, which is at 4 stops of underexposure, pushed back to base:

The noise is now starting to corrupt the image. Can it still be saved with noise reduction in post? Let’s have a look:

Actually, yes. The noise cleans up surprisingly well. An overall greenish cast appears in the image while some pinkish cast creeps into the shadows, but the image still looks really good! However, some significant temporal and spatial noise reduction was needed. Typically, 3 frames of temporal NR are only usable in a static image. In a moving image, you probably would notice some ghosting/smearing:

Now let’s move to 10 stops of exposure latitude, hence 5 stops underexposure, brought back to base:

Some large blotches of chroma noise are hovering around the image erratically. Plus, this can also be seen in the noise-reduced (moving) image, not in the static one displayed below, which still looks quite good:

As mentioned, the static image actually still looks rather good. But now, very distracting and large noise blotches appear in the moving image. They’re difficult to remove without applying excessive noise reduction, which leaves the entire image looking overly soft and plasticky.

Thus, game over!

In summary, a solid 9 stops of exposure latitude are exhibited by the LUMIX S1II, which is equal to the benchmark (for consumer cameras) that the Sony A9 III has set (see our Lab Test here). And this is actually with the bug that DR Boost mode is not properly activated, as we learned from Panasonic. We will definitely have to repeat this test once Panasonic has resolved the DR Boost ON bug with 5.8K ProRes RAW mode (which should be a matter of a few days only)!

Summary

The Panasonic LUMIX S1II shows really good potential and gives a great indication of the future of this camera line, as we could see with the 4K ProRes RAW pixel-by-pixel (APS-C crop) DR Boost “ON” mode, which boosts dynamic range results by 2 stops!

What should I say for now? Use APS-C crop in ProRes RAW for the best dynamic range and latitude with DR Boost ON, and APS-C and full frame modes with DR Boost OFF for best rolling shutter. Unfortunately, the drawback of DR Boost ON is the very high rolling shutter values, 27.5ms in full frame and still 18.7ms in APS-C mode.

The 5.8K ProRes RAW latitude result of 9 stops is again very good and shows the power of 12-bit RAW, even with the bug that we found (DR Boost not being activated properly in this mode). The difference to high-end cinema cameras is diminishing further, as we are already in RED RAPTOR territory with the LUMIX S1II.

In my opinion, the camera shows great potential in terms of dynamic range, but for now, you have to know which modes to use. As mentioned earlier, Panasonic has recognized the issue with the sample units, and this matter is going to be resolved in the mass production versions. And then, this camera can finally show its full potential!

What do you think about this new LUMIX S1II camera? Are you planning on purchasing it or staying with the LUMIX you already have? Let us know what you think in the comments section below…