Fuji Am I seeing one of the side effects of the X-Trans sensor?

[kae1]

I have no particular brand loyalty (any more 🙄 ) and have been trying out various second hand kits over the last couple of years mainly to get one that handles "just right" plus also seeing what they are like. I'd not tried Fuji so thought I'd give it a go and recently purchased an XH1 and as 18-55 lens. I'm enjoying the kit and took it out to test it last weekend. I was shooting various test shots in B&W JPEG and RAW with blown highlights, dark shadows etc. etc. This was one based on the RAW file processed through DXO PL8.

Join to see EXIF info for this image (if available)

However on processing this image I noticed (only at 100%) some "interesting" geometric patterns in the image.

I tried processing the RAW file through ACR and also Fuji's RAW file converter but the patterns are still there. It is not the end of the world as most of my other images were fine, but I was just wondering whether I'm seeing a side effect of the X-Trans sensor, or if there is a problem with my camera?
Any thoughts, gratefully received.

Just for further info - a 100% screen shot from the JPEG file

... and the RAW file

 

[bastibe]

The magenta tint is a hint that one or two color channels are selectively blown out. With sunlight, that's typically the blue or red channel. But, this normally results in a uniform color, not this weird aliasing pattern. Typical X-Trans aliasing patterns look like "worms", a sort of irregular false-color smearing. The whole point is to avoid regular grids like this.

However, the size if this glare makes me wonder if you shot through some kind of filter. It's unusually strong for Fuji glass (if it's clean!). Did you shoot with a filter, through a window, or through some sort if grating (mosquito net, curtains, car windshield, CPL...)?

I've seen another instance of a similar flicker-effect when shooting a propeller with the electronic shutter on the second-gen X-Trans sensor. The sensor apparently reads every eighth line in eight batches over a few hundred milliseconds, which might correspond to the pattern you are seeing if the light source was obscured during the capture.

If you'd like to share the raw file, I'd love to have a look.

 

[kae1]

However, the size if this glare makes me wonder if you shot through some kind of filter. It's unusually strong for Fuji glass (if it's clean!). Did you shoot with a filter, through a window, or through some sort if grating (mosquito net, curtains, car windshield, CPL...)?

Thanks, @bastibe for your suggestions. The lens doesn't have a filter on it, and the image wasn't taken through any window/grating. The used lens was cheaper than normal and described as "good but with small scratches on the lens" but I wouldn't have expected that to result in a specific pattern - but maybe this is how the sensor (under extreme circumstances) has dealt with them?

I've uploaded the RAW file to dropbox here.

 

[fredlong]

I think it’s a reflection of the sensor bouncing back and creating an interference pattern. I downloaded your raw file and looked at it in the raw converter in Affinity Photo. I looked at the individual color channels. I’ve never seen anything quite like it. I think it’s the result of very unique circumstances that would be hard to replicate.

So it’s an intuitive leap on my part, but that’s my hypothesis.

 

[bastibe]

Interesting hypothesis!

 

[Darmok N Jalad]

I can't say I've witnessed this before with the thousands of shots I've taken, and I shoot into the light quite often. Granted, I'm on the next generation sensor. Are the firmwares on camera and lens up to date?

 

[gryphon1911]

I think it is a mixture of a lot of what has aleady been said. It can be sensor reflection, and blown out color channels. I'll also throw in there that some of it also looks like it is an incorrectly adjusted rendering by the camera software for the PDAF pixels. There are auto focusing points on the sensor itself that a camera like the XH1 needs to determine the image data that should be there. With very bright scenes like this, it can happen rarely. All camera that use PDAF on sensor can suffer from this, so not something that is unique to Fuji.

 

[bastibe]

IIRC, the PDAF pixels on the X-Trans 2 sensors are more sparse. They look more like a grid of points, not a matrix of lines I think.

 

[camprl1]

I think it is a mixture of a lot of what has alead ybeen said. It can be sensor reflection, and blown out color channels. I'll alos throw in there that some of it also looks like it is an incorrectly adjusted rendering by the camera software for the PDAF pixels. There are auto focusing points on the sensor itself that a camera like the XH1 needs to determine the image data that should be there. With very bright scenes like this, it can happen rarely. All camera that use PDAF on sensor can suffer from this, so not something that is unique to Fuji.

I'd have to go searching, but I believe that this has been described before, specifically for Fujifilm cameras. Certain generations of X-trans sensors (I think it started with this generation) were described as having this sort of sensor reflection hitting the back element of the lens when shooting into a light source.

 

[camprl1]

In fact a quick search produced some articles suggesting that it is not a problem specific to Fujifilm along with this article that suggests it is an X-trans-specific issue:

X-Trans: A Deeper Look at the Purple Flare/Grid Artifact

When FujiFilm’s X-Trans III sensor was introduced in the X-Pro2, many users began noticing a strange new artifact in their backlit…

medium.com

 

[kae1]

Thank you all for your ideas and suggestions, especially the link to the issue { @camprl1 👍). I suppose it is reassuring to know that it does appear to be an issue with the sensor design rather than a camera fault.
Thank you, again.