Foveon sensor clipping - some data to crunch

Started Jun 15, 2011 | Discussions thread
Flat view
AndyNewton Forum Member • Posts: 51
Foveon sensor clipping - some data to crunch

The subject of blown highlights and sensor clipping is mentioned occasionally in these forums. For examples try a search for "sensor clipping X3F".

There seems to be a shortage of hard data on these issues. So here is some real data to tear into. In the two images below are experimental results of two tests I have run using a Sigma DP2s.

What do the test results mean? All interpretations are welcome... please!

Since performing these tests I use only ISO 200 or higher for RAW shooting, based on my own interpretation of the results.

The first test: A quick and dirty test.

Massive overexposure intentionally created in a single shot at ISO 50, then repeated at ISO 100, etc to ISO 3200.

Each RAW file developed in SigmaPhotoPro with clipped highlights pulled down into rendered range of output JPG.

In SPPro review window the SPPro histogram was captured directly off the PC screen, and all seven histograms assembled in PShop, upsized for readability, grid overlaid to help comparison, labelled and text added....

The second test: A much longer task.

A total of 154 images of a blank white wall recorded. Each single image gives just three numbers - the data values of R and G and B that are plotted as points on the chart below. All 462 data values together, keyed into an Excel spreadsheet, produce the chart shown below.....

 AndyNewton's gear list:AndyNewton's gear list
Nikon Coolpix 800 Sigma DP1x Sigma DP2s Canon ELPH 100 HS Canon ELPH 115 IS +1 more
Flat view
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow