a7x, a9 PDAF stripe noise technical analysis, part deux

Started 9 months ago | Discussions thread
iso rivolta Contributing Member • Posts: 550
Re: test reflection hypothesis with linearly polarized light?

Horshack wrote:

iso rivolta wrote:

Horshack wrote:

iso rivolta wrote:

Horshack wrote:

The original theory was reflections, a second theory was based on presumption of G1 PDAF pixels from the magic raw. I'm now very confident the original theory was actually correct (with revision to PDAF layout), based on new evidence over the past two days and examining all previous evidence.

As we discussed previously, I think a simple experiment with a linear polarizer in front of the camera while using a strobe (like Jim Kasson did) could establish if reflections are involved.

Light polarized perpendicularly to the plane plane of incidence (s-polarization) will reflect better than light polarized in a plane parallel with the plane of incidence (p-polarization).

If it's a reflection from the masked blue pixels to the neighboring green ones, it should be possible to find a position of the linear polarizer that enhances the striping in the G2 pixels (left and right of the strobe image) while decreasing striping in G pixels rows. At 90 degrees from that orientation the opposite situation should be observed, meaning minimized G2 striping and enhanced striping in the G pixel rows (above and below the strobe image).

Maybe Rishi or Jim can take 18 pictures with the linear polarizer rotated in 10 degree increments.

I actually did this experiment soon after we discussed it on the other thread - it didn't affect the striping.

Then this suggests the light is scattered or diffused, not reflected.

The experiment does not suggest that because the materials used to half-mask the pixels is likely metal, which means it doesn't alter the polarization of light, which also means the amount of light reflected off the metal is not affected by how polarized the incident light is.

This is not quite true. Please note on how the reflectivity of tungsten, the metal that Sony most likely uses for the back aperture, varies with s- and p-polarization: refractiveindex.info/?shelf=main&book=W&page=Ordal

Furthermore, the hypothesis needs a second reflection on the microlenses above the metal mask before reaching a neighbouring pixel.

Are you sure you used a linear polarizer instead of a circular one? We need that the light that falls on the sensor be linearly polarized.

Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow