Ricoh
Well-known
I'm attempting to calibrate a Sekonic L-758DR to my Leica M240, but have run into a bit of a problem that I could do with some help please.
Since the M240 takes a guess at aperture, the EXIF data recorded in the resulting JPEG image (required by Sekonic's Data Transfer Software (DTS) ) is incorrect for the actual exposure settings of the camera. The JPEG test images captured of the profile target (I was using the Exposure Profile Target II) were at 1/60 and f8, nominal, and two more at +/- 3 stops respectfully, but since the M240 guesses the aperture, the JPEG EXIF shows f9 in all three shots, which are incorrect (I varied the shutter speed for the three shots). When I import the images into DTS the software pre-populates the EXIF data based on what was written incorrectly by the camera, and the DTS software won't allow me to delete or overwrite with the actual correct settings. (I'm using Windows 8.1 incidentally and I've tried the latest version of DTS, ie ver 5.0).
Anyone else have the same or a similar problem when calibrating the L-758 to their M240, or any other digital M for that matter? And and how did you overcome the problem, ie how did you overwrite the EXIF data within DTS. Or did you have to edit the JPEG data externally prior to uploading into DTS, and if so how did you do this?
Many thanks
Steve
Since the M240 takes a guess at aperture, the EXIF data recorded in the resulting JPEG image (required by Sekonic's Data Transfer Software (DTS) ) is incorrect for the actual exposure settings of the camera. The JPEG test images captured of the profile target (I was using the Exposure Profile Target II) were at 1/60 and f8, nominal, and two more at +/- 3 stops respectfully, but since the M240 guesses the aperture, the JPEG EXIF shows f9 in all three shots, which are incorrect (I varied the shutter speed for the three shots). When I import the images into DTS the software pre-populates the EXIF data based on what was written incorrectly by the camera, and the DTS software won't allow me to delete or overwrite with the actual correct settings. (I'm using Windows 8.1 incidentally and I've tried the latest version of DTS, ie ver 5.0).
Anyone else have the same or a similar problem when calibrating the L-758 to their M240, or any other digital M for that matter? And and how did you overcome the problem, ie how did you overwrite the EXIF data within DTS. Or did you have to edit the JPEG data externally prior to uploading into DTS, and if so how did you do this?
Many thanks
Steve
Godfrey
somewhat colored
I haven't used DTS, but the simple solution is to record your apertures when you make photos and update the aperture setting value in the files with EXIFtool before you bring them into DTS.
From the command line:
> exiftool -m -Aperture="{correct value}" {filename}
You can make it easy on yourself by doing a set run of N exposures at f/2.8, f/4, f/5.6, etc. That way you can update the aperture values according to the file names in groups using wildcard characters in the filename value.
G
From the command line:
> exiftool -m -Aperture="{correct value}" {filename}
You can make it easy on yourself by doing a set run of N exposures at f/2.8, f/4, f/5.6, etc. That way you can update the aperture values according to the file names in groups using wildcard characters in the filename value.
G
Ricoh
Well-known
Thanks Godfrey.
I'm not familiar with the EXIFtool or with using th command line. The command you've written looks a little like Linux - but this just shows how unfamiliar I am, sorry.
Is it a downloadable tool and is it available for Windows, I'm running 8.1
Thanks again for your help.
Steve
I'm not familiar with the EXIFtool or with using th command line. The command you've written looks a little like Linux - but this just shows how unfamiliar I am, sorry.
Is it a downloadable tool and is it available for Windows, I'm running 8.1
Thanks again for your help.
Steve
Spanik
Well-known
Could you please explain why you would like to calibrate a perfectly good lightmeter to a camera that (according to your post) "takes a guess at aperture"? I can see it the other way round and have used my 758 to check suspected lightmeters but I fail to see any possible useful result of calibrating a lightmeter to an unreliable reference.
Godfrey
somewhat colored
EXIFtool is available for Windows, MacOS, and Linux from
http://owl.phy.queensu.ca/~phil/exiftool/
There are many other EXIF manipulation tools available that utilize it which might be easier for you. A Google search for "Windows EXIF utility" or similar would likely list a few of them. Basically, you just need something that can change the EXIF Aperture values to the correct ones for the calibration software.
G
Godfrey
somewhat colored
The only "guess at aperture" the digital Ms make is at what setting the user made. This is because Leica M lenses have no mechanism to explicitly tell the body what aperture is in use... the M software has to derive it by comparing what the main sensor sees to what its external ambient light sensor sees.
There's no other difference between calibrating an M and the Sekonic meter compared to calibrating any other camera. It's just a matter of being sure the input EXIF data is accurate. Calibrating the meter to the camera ensures that variation in a particular camera's ISO setting, shutter timing, and lens f/number settings are known by the meter.
G
Ricoh
Well-known
Straight out of the box my copy of the Sekonic L-758DR shows a difference in readings for incident and reflected light against an 18% grey target (obviously incident was measured looking towards the camera) in the same light. Regarding the M240 guessing, the camera measures using the in-built centre weighted light meter and exposure is set in the usual way, either in A mode or fully manual. EXIF data is written to the SD card along with the image data, but since there's no electrical contact between the camera and the lens, the camera guesses the value based on the other known two variables. It's just the way it works. Once the 758 is calibrated to the camera it displays the camera's DR and allows you to set conservative clipping points if you wish. Also when calibrated you can take an incident reading using the light meter and know the camera exposure will be correct given the light meter knows how the camera's sensor and associated processing 'sees' light.
shimokita
白黒
+1 for exiftool... the command line is a little tricky but useful in certain situations.
With windows the following example command line will write all exif data to a document file. The file is about 12 pages when using a Canon 5DII with a RAW (CR2) image file.
exiftool -a -u -g1 C:\folder\TestFile.jpg > C:\folder\TestFile_exifData_docu.txt
-a : (-duplicates) Allow duplicate tags to be extracted
-u : (-unknown) Extract unknown tags
-g[NUM...] : (-groupHeadings) Organize output by tag group
With windows the following example command line will write all exif data to a document file. The file is about 12 pages when using a Canon 5DII with a RAW (CR2) image file.
exiftool -a -u -g1 C:\folder\TestFile.jpg > C:\folder\TestFile_exifData_docu.txt
-a : (-duplicates) Allow duplicate tags to be extracted
-u : (-unknown) Extract unknown tags
-g[NUM...] : (-groupHeadings) Organize output by tag group
Ricoh
Well-known
They're not contacts, just black and white paint patches forming a 6 bit code read optically by the camera body to inform it what lens is attached. The 6 bit code does not relay info about aperture.
Godfrey
somewhat colored
Ricoh
Well-known
If it did, it would be witch craft for sure.
charjohncarter
Veteran
Superior thinking Godfrey!
Richard G
Veteran
I'm very impressed by you guys.
I never had a problem with my Gossen Digipro F and my metered Ms, M5 and M6. They always seem to match. My Sekonic L398A on the other hand had to be calibrated by a slightly less sophisticated method to work with my metered cameras.
(A man with two watches never knows what time it is. Applies to light meters too I reckon.)
I never had a problem with my Gossen Digipro F and my metered Ms, M5 and M6. They always seem to match. My Sekonic L398A on the other hand had to be calibrated by a slightly less sophisticated method to work with my metered cameras.
(A man with two watches never knows what time it is. Applies to light meters too I reckon.)
JoeV
Thin Air, Bright Sun
I think a better term to use is "correlate." The OP wishes to correlate his meter to the M240 files' EXIF data.
Calibration would be done at a repair facility against a known standard. Like Quality Light Metric, in Hollywood, CA, for example.
BuzzyOne
Established
Just shoot an 18% grey card and manually adjust the meter until it reads 128.128.128. No EXIF required.
http://www.frankdoorhof.com/site/2011/08/calibrating-the-light-meter-some-quick-notes/
http://www.frankdoorhof.com/site/2011/08/calibrating-the-light-meter-some-quick-notes/
Ricoh
Well-known
It talks about strobes, but I assume a bright overcast sky would do.
It also assumes Photoshop, but many use Lightroom instead, including myself.
I'll have to read the manual again, but I think I may have overlooked the thing about pressing iso1 and 2 together.
As mentioned above, correlation is a more appropriate choice of word. Teaching the meter how the camera sees light. Having metered from an 18% grey target and transferred the meter readings to the camera, just three shots are required to enable the DTS software to determine whether there is an offset between the handheld meter and the camera. But the software demands JPEGs with accurate embedded EXIF, and that's simply impossible for any digital M.
It also assumes Photoshop, but many use Lightroom instead, including myself.
I'll have to read the manual again, but I think I may have overlooked the thing about pressing iso1 and 2 together.
As mentioned above, correlation is a more appropriate choice of word. Teaching the meter how the camera sees light. Having metered from an 18% grey target and transferred the meter readings to the camera, just three shots are required to enable the DTS software to determine whether there is an offset between the handheld meter and the camera. But the software demands JPEGs with accurate embedded EXIF, and that's simply impossible for any digital M.