PXW-FX9 Exposure Accuracy – Can I use A Light Meter?

This came up during a Facebook discussion. Can you use a light meter with the FX9 and will the exposure be correct?

When I first met the FX9 at Sony’s Pinewood Studios facility we tested and checked all sorts of different aspects of the cameras performance against various light meters and test charts. I found that the camera matched what we expected perfectly well.

But just to be sure I have just tested my own example against my trusty Sekonic light meter and once again I am happy to say that everything seems to match as expected.

In this simple setup I used a couple of different charts with middle grey and 90% white – I do find that there is some variation between charts in how reflective the 90% and 18% reflectivity areas are. So I’ve used a couple here and my main reference is the large DSC Labs white and middle grey chart.

I used the dimmers on my lights so that my metered exposure reading for 24fps 1/48th shutter came to exactly f5.6. Then I set the lens to f5.6 (Sony 24-70mm GMaster).

The result is pretty much as close to a perfect exposure as one can expect. So don’t be afraid to use a light meter with the FX9.

Screenshot-2020-07-21-at-10.11.45 PXW-FX9 Exposure Accuracy - Can I use A Light Meter?
Exposed via the light meter, screen grab from Catalyst Browse.


Screenshot-2020-07-21-at-10.14.53 PXW-FX9 Exposure Accuracy - Can I use A Light Meter?
And the levels as measured in DaVinci Resolve. All looks good and where you would expect S-Log3 to be.

12 thoughts on “PXW-FX9 Exposure Accuracy – Can I use A Light Meter?”

  1. Have you looked at the Blackmagic Ursa 12k camera and compared with price-peers like the FX9 and Canon?

    Blackmagic RAW and the new RGBW sensor look very interesting.

    I guess Blackmagic RAW will not work with an FX9 because BM put some of the RAW coding within the body of their cameras.


    1. I haven’t seen one in the flesh. On paper and in the presentations it certainly looks interesting, but without actually playing with one I have no idea how the specs will translate to real world performance given the very different sensor architecture. It would be interesting to play with one for a while.

      1. Sorry this is off light-meter topic again.

        What impressed me about the Blackmagic 12k Ursa was not just the very credible picture quality on their website presentations, but that the camera recorded its version of 12k RAW straight to a commercial USB attached SSD – nothing proprietary media wise. Then the USB drive was plugged straight into a laptop and 12k video replayed immediately without any pre-rendering.

        When I bought our first FS7 we spent nearly as much on XQD media as we did on the camera.

        1. Detach that SSD mid shot and you risk losing everything on it forever. SSD’s are not designed to be used as recording media and have absolutely no protection against unexpected power loss. While rare perhaps, is that a chance you are prepared to take?

          1. We have used external SSDs for audio for 5 or 6 years connected to a laptop. We must have been lucky – we have never lost anything. Mind you, we were mirroring on to an internal drive at the same time for safety just in case. Our Atomos Shoguns and Sumo19 also record to SSD and none has let us down yet.

            I guess it depends on how the files are written to the SSD and if the toc is fixable later if the power fails.

  2. you’re wrong.
    Try doing that at 2000 ISO base 800 ISO. Expose it the same way you did it.
    Now keep 2000 ISO but change to base 4000 ISO. Totally different results.
    The light meter works only in a few situations and only will match the exact middle gray, not the latitude.
    Even sekonic also recognizes that light meters need to use gamma curves:

    1. Uh, no, I’m not wrong.

      The camera is in the CineEI mode. SO you only have 2 actual ISO’s, 800 and 4000. If you are using 2000 EI and you set the light meter to 2000 ISO then there will be an expected exposure change of -1.3 stops with base 800 and +1 stop with base 4000. That is absolutely expected and how Log and Exposure Index works.

      Light meters do not need to use gamma curves, all they do is measure the light level, gamma is irrelevant for the meter and they don’t directly measure latitude anyway.

      1. so, the light meter it’s only usable in a perfect world, not in the real production world. Something that it’s not in your post.

        Also, you’re telling sekonic that they’re wrong about their own products.
        Nice to know.

        1. No, I am saying that I don’t think you understand how Exposure Indexing in CineEI works.

          If the exposure measurement is correct at the base ISO, then it is also guaranteed to be correct for every corresponding Exposure Index (for example using 2000 EI with 800 ISO base) as the camera ONLY records at the base ISO. Any changes in brightness you see are either:

          An expected brightness offset because that’s what changing the EI and then changing the ISO number in the light meter will lead to, different brightness of exposure depending on the combination of EI offset and base ISO.

          Or an unexpected change in brightness because you don’t understand how EI works and are using it incorrectly.

          The light meter is absolutely and totally usable and will give the correct exposure values in all sorts of real world scenarios. But you also have to understand how the exposure index system works and understand that 2000 EI is not the same thing as 2000 ISO. To be able to use a light meter correctly and accurately you must also understand how to interpret what it is telling you and then apply the settings to the camera in the correct manner.

          Light meters don’t have gamma curves and they don’t need them because they don’t need a transfer function to match the contrast range of a display device. Light meters may sometimes need a calibration offset or other table of values for the purposes of calibration, but that isn’t the same as a gamma curve.

    2. @II never say Allister is wrong as that would be considered a criminal violation of the international Digital Cinema educational standards adopted by all working professionals who understand the role and use of a light meter. I suggest you go back to school and research the proper usage and implementation of a light meter, gamma curves and how to optimize Cine EI.

  3. Nice write-up, Alister.

    Another thing to keep in mind is the difference between T-stops and F-stops.

    If the exposure result is not as expected when using a light meter with a lens using F-stops, consider that there is some light-loss passing through the lens.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.