M43RawAnalyzer: First Results

Thanks

First of all I want to thank everyone who has downloaded the program and shared his data!

Reading TIFF Tags

Now I want to present you the first results of this project. Those are the most obvious results I have found so far, but I have also some assumptions which I want to investigate further. But my biggest problem at the moment is the deciphering of the Panasonic RAW format. The current version of the M43RawAnalyzer is just crudely reading the values from predefined adresses inside the RAW files. But to access more data and to make the reading more robust and error prone it needs to understand TIFF tags. Those TIFF tags are also used in many other RAW formats. I have tried using this C# LibTiff library, but it does not work, because the RAW files are no TIFF, albeit they are using TIFF tags and are quite similar to TIFF files. So I will try to write a simple TIFF tag parser to be able to read more data from the RAW files.

My Assumptions

My main motivation is to get the aperture value and lens firmware version. I am not sure if distortion corrections values are independent of the aperture. I have seen one example where the correction values were different, but the camera and focal length were identical. And now I don’t know if the reason for this is a different aperture or a different lens firmware. Also I am more comfortable in interpreting my findings, when I am able to understand the RAW format better.

Olympus Distortion Correction Data

Additionally I am trying to find the location of the distortion correction data inside of the Olympus RAW files. I have three positions which look promising, but they are not constant. And I am suspecting this is because the data is inside of a TIFF tag.

First Results

I have first results in three areas and the results are derived from the data of the Results Spreadsheet:

• Camera Constants
• Prime Lenses and if they need corrections
• Zoom Lenses and if they need corrections

Camera Constants

As described in detail by Andrew, one (the value 12) of the 16 values related to the distortion correction could be linked to the camera. I suggest it could be linked to the sensor, because for all 12 Megapixel sensors its 2500, for the 16 Megapixel G3, G5 and GH2 it’s 2860. The 16 Megapixel GX1 has the similar value 2886. Unfortunately, it appears that the latest model GH3 (also 16 Megapixel) has no constant value. But also all of my current samples are taken from the internet and the values could be related to a beta firmware.

Prime Lenses and Corrections

The following table shows the prime lenses for which I have information about the distortion correction data.

Zoom Lenses and Corrections

The following table shows the zoom lenses for which I have information about the distortion correction data.

The Future

My plan is to update the M43RawAnalyzer with TIFF tag support and then trying to collect more samples. Also I am searching for the location of the distortion correction information inside the Olympus RAW files. Unfortunately, I have to divide my time between this project and the improvements/completion of MirrorlessDB. So I cannot guarantee you lighting-fast progress.