Jump to content




Photo

Lightmeter and calibration


  • Please log in to reply
12 replies to this topic

#1 Daniel Meier

Daniel Meier
  • Basic Members
  • PipPip
  • 69 posts
  • Cinematographer
  • Leipzig, Germany

Posted 04 January 2016 - 06:30 AM

I got me a Sekonic L-308 DC to get started with lightmeters. The manual says it's calibrated coming out of the box.

But for both my cameras (Sony FS100 and Canon 7D) I had to push the calibration by +0.6 EV in order to get proper exposure.

 

Is there a way to check, if this thing is really calibrated when set to 0,0 EV correction? Maybe by measuring a standardized light source?

I thought about incident metering the sunlight (with the LumiDisc on, instead of Lumisphere) and compare the lux/fc value to the ones in the books. But since there are reflections of buildings and whatsoever, clouds, different angles ... wouldn't a tungsten source be better?

 

Also I can't really get use of the spot meter. Since it has an angle of 40°. If I get to close on a surface, to tighten the viewing angle, the meter of course casts a shadow on it.

Any ideas on how to use it properly?


  • 0




#2 Mark Dunn

Mark Dunn
  • Basic Members
  • PipPipPipPip
  • 2182 posts
  • Other
  • London

Posted 04 January 2016 - 06:56 AM

Assuming you bought new, the meter is correct. It has been factory calibrated. The cameras are wrong.


  • 0

#3 David Landau

David Landau
  • Basic Members
  • PipPip
  • 87 posts
  • Cinematographer
  • New Jersey & New York

Posted 04 January 2016 - 09:21 AM

Spot meters are reflected meters for getting readings far away. If you're shooting with a 250mm lens, instead of sending an AC running down to the subject, you use the spot meter on the face of the subject. The center circle is what the meter is reading, not the entire image you see.  It'll give you a reading based on the subject being 18% grey.  So now, depending on the subjects skin tone, you close or open up the iris. 


  • 0

#4 Brian Drysdale

Brian Drysdale
  • Basic Members
  • PipPipPipPip
  • 4745 posts
  • Cinematographer

Posted 04 January 2016 - 09:42 AM

The ISO on video cameras can vary depending on how it's set up. Assuming you've got your meter calibrated, you should test the ISO rating of your camera, not take what the camera says as gospel. Although with modern lenses it's not large, you can lose light though your lens and unless it has T stops, you'll find a variation. 

 

You can use an incident outdoors. For fc/lux readings you point the meter at the light source, just the same as with any other light. However, I just use the dome outside.and just point it at the camera from the subject or face the subject in the same direction as the camera and take a reading over my shoulder (unless the subject is in different lighting). This works 99% of the time.


  • 0

#5 Stuart Allman

Stuart Allman
  • Basic Members
  • PipPipPip
  • 151 posts
  • Cinematographer
  • San Diego, CA

Posted 04 January 2016 - 09:59 AM

Daniel,

 

There are a couple things that could be going on here.

 

1.  You're using DSLR lenses that measure aperture in f-stops instead of t-stops

2.  The DSLR could reference ANSI standard 12.5% gray as middle gray instead of 18% (0.5 stop difference).

 

The way to tell is to get a Kodak R-27 chart off Amazon and use it to measure both incident and reflected exposure.  The Sekonic meter should show the exact same readings off of an 18% gray chart.  If not, then it's out of calibration (doubt it).  My L-758DR shows the exact same exposure numbers off an 18% gray chart in both measurement modes.

 

Here are two links I recently helped publish on the subject.

 

http://illuma.blogsp...d-exposure.html

 

I've found that some manufacturers put 18% gray at 44 IRE and some at 50 IRE (and there are other variations too).  It really depends on your camera manufacturer.  Finding out where the camera believes middle gray should be is usually just a matter allowing the camera to auto-expose a uniformly illuminated 18% gray chart and looking at a luma scope.  Non-cinema lenses also vary all over the place.  We're going to do more tutorials on this subject on the Vimeo blog in the near future.  Hopefully that will address some of your woes. 

 

Stuart Allman

--------------------------------------

illuma.blogspot.com


  • 1

#6 Daniel Meier

Daniel Meier
  • Basic Members
  • PipPip
  • 69 posts
  • Cinematographer
  • Leipzig, Germany

Posted 04 January 2016 - 11:16 AM

Thanks for the help!

Yes I bought the lightmeter new.

 

Might be my rather cheap graycard and the f.stop inaccuracy then.

 

Or it is due to REC 709.

I pointed the camera towards an evenly lit 18% gray card, then exposed and checked my waveform monitor to be at 50% IRE.

The FS 100 was set to PictureProfile off. So I assume it was set at REC709. Hence I only woudl've had to expose the gray card to 42% instead of 50%. Maybe that's the fault.


  • 0

#7 John E Clark

John E Clark
  • Basic Members
  • PipPipPipPip
  • 780 posts
  • Other
  • San Diego

Posted 04 January 2016 - 01:04 PM

Thanks for the help!

Yes I bought the lightmeter new.

 

Might be my rather cheap graycard and the f.stop inaccuracy then.

 

Or it is due to REC 709.

I pointed the camera towards an evenly lit 18% gray card, then exposed and checked my waveform monitor to be at 50% IRE.

The FS 100 was set to PictureProfile off. So I assume it was set at REC709. Hence I only woudl've had to expose the gray card to 42% instead of 50%. Maybe that's the fault.

 

I'm pretty sure, absent any manufacturing defect, the Sekonic is 'calibrated'... the question is 'to what'...

 

I believe Sekonic uses a 12% reference for it's meters.

 

That and whether or no you are using f-stop based aperture or t-stop...

 

What one needs to do is find one's own ISO/EV value for one's own cameras.

 

For Rec. 709 and using a the 18% grey card (Kodak is a popular one...) then the resulting 'grey' value as determined by using the Waveform display in many NLE's, is the value should yield a IRE reading of 45-50%.

 

For 'Log' type representations, there are a number of other values, often 30-40%.

 

For my Blackmagic Pocket camera the recommended value is 38.4% IRE for the 'BMDFilm' mode, which is a 'log-ish' type recording mode...

 

As a note, in the case of Rec 709, 50% may be 'too high' since that may only give 1.5-2 stops (which is what my Lumix GH-1 gives...) over that value for 'high lights', and so one may 'blow' highlights if one slavishly uses 50%.


  • 0

#8 Stuart Allman

Stuart Allman
  • Basic Members
  • PipPipPip
  • 151 posts
  • Cinematographer
  • San Diego, CA

Posted 04 January 2016 - 01:30 PM

Daniel,

 

If you purchase the same 18% gray card linked on my blog post (illuma.blogspot.com) then you should have an accurate gray card.  I have no association with Amazon or the card vendor.

 

The long and short of it is to pick a specific camera picture profile on your particular camera, light an accurate 18% gray card uniformly, set auto-exposure on your camera, and record the aperture setting, then offset your light meter to that aperture.  That should work OK with your FS-100 since it should base middle gray off 18% gray.  The exactly IRE will then be irrelevant.  This method takes into account both the camera and lens.  I don't know if you can do this with the Canon DSLR since I'm not familiar with it's gray target (ANSI 12.5% versus 18%).

 

Middle gray may also depend on the picture profile you use.  As I recall the FS-100 has one "cine-gamma" like picture style.  My experience with Sony started with the FS-700 so I can't help too much.

 

Stuart Allman

-----------------------------

illuma.blogspot.com


  • 0

#9 Maximilian Motel

Maximilian Motel
  • Basic Members
  • PipPip
  • 29 posts
  • Other
  • Berlin, Germany

Posted 24 January 2016 - 01:23 PM

Might just be your lenses, actually. It sounds like you will be using regular photography lenses, and their f-stops tend to not be what they say they are (at least concerning the amount of light that hits the sensor) unlike t-stops. So that might just be the case.

 

I can't remember where that was, but I saw a list of popular canon lenses online with their f-stops and their (real) t-stops somewhere a while ago.


  • 0

#10 Stuart Allman

Stuart Allman
  • Basic Members
  • PipPipPip
  • 151 posts
  • Cinematographer
  • San Diego, CA

Posted 24 January 2016 - 05:16 PM

DXOMark.com has the t-stop ratings of still photography lenses.  I recently did a part two tutorial of the video I posted earlier, which pretty well agrees with what you're finding.

 

I also spoke with the folks at Sekonic.  They explained that camera manufacturers put their 18% gray point at camera vendor specific IRE levels.  So Sekonic expects you to calibrate your light meter to your camera and lenses. The best they can do is calibrate their meter to the ANSI standard since they don't know which camera and lenses you're using.  Add on top of that the f-stop to t-stop compensation factor and you'll need to do the calibration to get things working properly.  I found that with a Sony FS-7 and my group of Zeiss ZF.2 primes I had to use a compensation factor of anywhere between 0.6 and 0.8 stop (lens dependent).  So your finding of 0.6 stop isn't too far off from what I would expect.

 

When I first bought a light meter I thought it was plug-n-play too...until my images started coming back too dark.

 

Stuart Allman

------------------------------

illuma.blogspot.com


  • 1

#11 Alexandros Angelopoulos Apostolos

Alexandros Angelopoulos Apostolos
  • Basic Members
  • PipPipPipPip
  • 346 posts
  • Student
  • Athens

Posted 25 January 2016 - 02:26 AM

Would this help you on how to calibrate it?

 

If it would, please do explain it to me once you're done.

 

I wish someone wrote a transcript of this or transformed it into some simple step-by-step guide on how to calibrate a light meter.

 


  • 0

#12 Daniel Meier

Daniel Meier
  • Basic Members
  • PipPip
  • 69 posts
  • Cinematographer
  • Leipzig, Germany

Posted 27 January 2016 - 09:10 AM

Would this help you on how to calibrate it?

 

If it would, please do explain it to me once you're done.

 

I wish someone wrote a transcript of this or transformed it into some simple step-by-step guide on how to calibrate a light meter.

 

 

He basically did as follows:

- Take an 18% graycard an light it evenly (without any hotspots) ideally with a tungsten source

- take an incident reading of the incoming light in front of the graycard

- match the f-stop value, ISO and Shutterspeed in your camera

- look at the resulting image, the camera is producing

- if the graycard on the image is lower than 50% (in this case a histogram-value of 128) you need to calibrate your meter to a positive value (e.g. +0,5 EV)

- if the graycard appears to bright dial the calibration down (e.g. -0.5 EV)


  • 0

#13 Stuart Allman

Stuart Allman
  • Basic Members
  • PipPipPip
  • 151 posts
  • Cinematographer
  • San Diego, CA

Posted 27 January 2016 - 10:23 AM

Daniel,

 

I wish it was as simple as that.  The 18% gray point is dependent on the camera vendor and picture style you use.  So you need to do your homework and find out where YOUR camera vendor puts 18% gray with YOUR picture settings.

 

For instance, if you use Sony's hypergammas then your 18% gray IRE value will be at 33 or 40 IRE.  When I set the FS-7 to S-log3 and use the rec.709 800% monitor viewing LUT 18% gray comes out at 44 IRE (the mathematically correct value by my spreadsheet calculations).  If you're using S-log2 or C-log then 18% gray is recommended to be at 32 IRE.  A Red camera might have completely different 18% gray targets...so use these tutorials as a starting point, but don't assume 50 IRE is correct or you might find your images all over exposed.

 

The reason we concentrated on one specific camera and one specific picture style in the log gamma tutorial is because Sony explicitly recommends 41 IRE for 18% gray exposure. It's well documented.  You just have to find that documentation for your camera and the picture style you use on it - then walk through the same process we present in the video. 

 

An easier way to do this might be to just set your camera in auto exposure mode in front of the gray card and find out where it sets 18% gray with your picture settings.  The compensation factor is going to be lens dependent, so you'll have to test every lens in your arsenal.

 

Stuart Allman

---------------------------

illuma.blogspot.com


  • 0


CineLab

Technodolly

Abel Cine

The Slider

Paralinx LLC

Pro 8mm

CineTape

Broadcast Solutions Inc

Zylight

Ritter Battery

Metropolis Post

Visual Products

Aerial Filmworks

Tai Audio

Glidecam

Rig Wheels Passport

Willys Widgets

rebotnix Technologies

Abel Cine

CineTape

Metropolis Post

CineLab

Willys Widgets

Pro 8mm

Broadcast Solutions Inc

The Slider

Tai Audio

Rig Wheels Passport

rebotnix Technologies

Glidecam

Ritter Battery

Visual Products

Zylight

Paralinx LLC

Technodolly

Aerial Filmworks