Light Meters and the GH1

[thisisapocalypse]

Hi everyone,

After reading the threads posted by the creators of "Rejouer" I, like a number of you were intrigued about their comments in regards to setting one's light meter to ISO 640 when shooting the GH1 at ISO 100.

I'm using a Sekonic Digi Master L-718 and I tested it out over the weekend, and just wanted to confirm that indeed setting an incident meter like this one to ISO 640, when metering for ISO 100 on the GH1 is as close to right as you can get.

I found that the GH1 meter would read a couple of bars towards underexposure when setting it this way, but as many of you have noted, the GH1 meter seems to want things to be a bit brighter than what many of us would call "correct." I think depending on your scene preferences, you can take the light meter reading and go 1/2 a stop in either direction, but I just wanted to confirm that setting the meter this way does indeed seem to get as close to proper exposure as I've seen.

How useful is this information given that you can see the image on the LCD and just set exposure visually?

It really depends on what you're doing - The main reason you'd want an outboard light meter (incident) is for setting up your lighting and when doing location checks so that you know what kind of light you'll be working with once you start shooting...how fast a lens you'll need, etc. It's really invaluable for that, and definitely worth having one in your kit.

 

[pix2pixels]

I am very confused about this. I used my Minolta meter, and found hardly any difference between the GH1 and the meter's reading for the Macbeth color chart for a balanced histogram with no clipped blacks or highlights.

 

[thisisapocalypse]

I am very confused about this. I used my Minolta meter, and found hardly any difference between the GH1 and the meter's reading for the Macbeth color chart for a balanced histogram with no clipped blacks or highlights.

It's strange - if I used the meter's readings for ISO 100, the image on the GH1 was incredibly overexposed.

When I set the meter for ISO 640 (as the guys from Rejouer indicated) my settings worked. This was in Creative Movie Mode, full manual control, with Canon FD lenses, Smooth setting, 1080, anything that could be auto, turned off.

What was your metering technique? I had the meter in the shot aimed at the camera lens.

 

[pix2pixels]

Exactly, the same here. Same settings, same lenses, different meter.

I use this meter all the time, it's a close match both for (continuous light and flash) to the Canon EOS 1D3 that I use every day, I used it for a feature film shot with RED ONE, and there was a difference there: instead of ISO 320 as stated by RED for their camera, I got the best histogram and overall image with the incident meter set at ISO 200, pretty consistent with the experience of other users.

I have quite a collection of meters from all ages (Gossen, Spectra, Sekonic, Pentax, Minolta) and I know there are differences between them. I set each one at different settings, assessing the best image I see on the screen for whatever lab setting (push/pull processing, printing lights, best rendition of skin tone against loss/gain of highlight detail, or noise in the blacks, best possible histogram or wave form.

In order to eliminate any subjective assessments of the image, depending on the subject, I prefer to test the sensitivity of the system using the chart. I have a number of RAW conversion presets for various DSLRs I use, where I deliberately underexpose and adjust the settings for post-processing if I feel I need to preserve highlight detail and get the best image quality.

I am a bit reluctant to go this route with GH1, because we start in 4:2:0, with a high compression algorithm, and the IQ could degrade substantially even for a minor tweak in post. I aim to get the best image from the camera with as much data possible and add the extra contrast, saturation in post.

I just finished shooting a music video over, using Canon FD primes, the camera in full manual mode at 400 > 500 ISO, daylight balanced fluoro banks and lots of LED fixtures, mainly locked in primary colour channels. I have some stills (not frame grabs) I could share at this stage, and they were shot with a Canon 1D3, with similar or close exposure/ISO settings as I had on the GH1.

Am I doing something wrong? Is my camera dodgy? Does Panasonic know something that Canon, Red don't?

 

[dvbrother]

I have a Sekonic L-398A. I tested mine using a DSC Labs Chart. With mine, if I set the GH1 to 100 ISO, it agreed with my meter set to 200 ISO. In fact, across all ISO settings, they matched when the Sekonic was rated one stop hotter than the GH1. This was on Smooth setting. In Dynamic setting, I think I had to set the meter one and 1/2 stops hotter. All this seems to point to discrepancies in the sensitivity of GH1 sensors, discrepancies between different light meters, or both.
I suppose each person needs to find what makes sense for them and stick with it. When Illya Friedman first mentioned that 400 ISO equaled 2500 ISO on his meter, I thought he was crazy. Apparently not. But for me and my Sekonic, 400 ISO on the GH1 equals 800 ISO on my meter.

 

[Appleman]

I concur ie. agree with dvbrother. I've got the Sekonic 608 and over the weekend tested the numerous ISO settings that will match the GH1's filming ISO settings. I was only able to do about 1 hour or so of filming before the sun went down but my conclusion is that if i set my GH1 to ISO100 - my sekonic should be set to ISO 200 to give the most matching realistic to the scene image. THe GH1 has to be at the full manual filming mode - i think that's the creative filming mode, right - for these type of manual exposure usage. I didn't have time to test out the other GH1 ISO settings - i'll do that when i'm able to. I believe if i had set it to ISO 640 for the sekonic it's gonna be severely under exposed.

My next tests will be for the GH1 at ISO 400, 800 and 1,000. I think i will start by doubling the ISO for the sekonic for the match testing ie. GH1 ISO 400, Sekonic ISO 800- let's hope the sensor is linearly sensitive without any funny spike anywhere in its sensitivity ratings. That'll save a whole heap of time!

I believe the lightmeter will still give a superior light reading if one has the time, oppurtunity and effort to do it for the scene. There is a lot of leeway either side of over/under exposure when using the camera's histogram. It is very difficult to 'nail' the exposure with fiming just by using the histogram. It's ok to rely on the histogram when you're shooting still as you can adjust the exposure further with the raw files with lightroom/camera raw.