Geofn wrote:
Looks to me like when the mirror started up it caught an image of the sun, previously outside of the frame but included in the lens' projected circular image, and reflected it into the viewfinder (hence the flash) and the reflection of the sun off the bottom of the pentaprism then traced the arc-shaped flare onto the sensor through the 50% transmissive mirror. Just a guess, but it's the best explanation I can come up with. All the angles would have to be just right (wrong?) for this to happen, so my guess is that you'll likely never see it again....Show more →
I think you're right about the flash OP saw in the viewfinder, but aren't the partially silvered parts of the mirror only in front of secondary mirrors that direct light to AF sensors? The mirror can send odd reflections through the finder as it moves, but not to the sensor. Besides, the shutter wouldn't open until the mirror is out of the way. The artifacts in the file are more likely related to the extreme brightness within the frame and the demosaic effects discussed in the link above.
I was shooting a wedding using film, and had a bright light in the viewfinder. Ruined the frame (overexposed) and never did figure out what happened. I thought it was somebody's flash, but I was the only one in the area with a camera, and it was dark. Really really dark.
The camera never did it before that, and never did it after that.
When I have had images that were not what I expected, I was able to solve the problem by taking off the uv filter that I had on for protection. I once had a issue using my Canon 85L f1.2. I no longer use UV filters except as lens caps.
omarlyn wrote:
+1 ...The internal reflections would undoubtedly be different and could (potentionally) cause flare that would not otherwise be seen with the mirror down. I wonder if you might be able to repeat this phenomenon by setting up a similar shot and switching back & forth from live-view to see if you create the same type of flare.
HTH,
Omar
I think this is correct. Only to add to this idea, more correctly, I think it was the mirror that eliminated the sunlight (flare) that was seen in the photo.
When light is reflected off the mirror, some of the intensity is lost (I think it's a half-silvered mirror, that means it's 1/2 the intensity). Also, dependent on the grains on the mirror, light of certain direction will be more affected and not viewable by your eyes. This fraction is still available to the sensor later of course. When the camera takes the picture, the mirror is flipped up, there is no other optical component that is in the way to the sensor. Thus, the sensor would see the full intensity coming in from the lens.
I think it is a coincidence that you're in the right place with the right angle.
I think it has been said, but the flare looks like a coincidence of timing and angle to the sun and the grid pattern is most likely the result of the demosaicing algorithm. Process the file with a different program as recommended and the pattern will likely disappear. I have a few images taken recently with a CV 15mm heliar on an E-P1. I converted to DNG...viewed in Lightroom, I see the same exact pattern as in your image. Viewed in Photoshop/ACR (which had already been updated to officially support the E-P1), it was fine.
