Pinhole color gradients: controllable ??

[daveleo]

These gradients must be light bending around the edge of the hole?? I thought it was dust, but it's not.
What would cause this? and what can I do do reduce this effect?
😕
Thank you for your comments.
.

 

[Larry H-L]

If it is a picture from a digital camera, most likely from light striking the sensor at an angle.

 

[daveleo]

I have a theory

I have a theory

How severe this effect is depends on the flange-to-film or flange-to-sensor distance. The shorter the distance, the stronger the color gradients. Based on my pinhole photos from a Nikon DSLR (I saw no such gradients) and photos from my mirrorless Fuji XA1 (the picture posted above).
I am also now aware that lots of pinhole pictures posted on the internet are cropped square, which may be how people get this out of their pictures. The effect is most strong out at the edges of the frame.
That's my theory.
Workaround .... shoot B&W for now.

 

[daveleo]

If it is a picture from a digital camera, most likely from light striking the sensor at an angle.

Maybe, but see my post above here. I don't think what you're saying disagrees with what I'm saying. In fact now it seems we are both correct . . . the angle induced by a short mirrorless distance has to be a lot wider than from a DSLR (same sensor size), so the effect you talk about would be stronger, as I noted.

EDIT. so to reduce this effect on a mirrorless camera.... by a pinhole disk for a DSLR camera and buy a DSLR-mirrorless adapter (Nikon-Fuji, or whatever).

 

[segedi]

This looks like the same effect as when you use a wide angle lens with a deep exit point on digital as well. Like when using a Zeiss 21mm M mount lens on the M digitals or mirrorless.

I had this even with the Voigtlander 35mm f/2.5 on the M 240 - but the effect an really be lessened if not entirely eliminated by coding the lens as a Leica lens. For me, a 28mm lens code worked well. You may be able to manually change your camera's settings to get rid of magenta edges. I know the Ricoh GXR allows for that type of thing.

The only surefire way to eliminate colour casts is to either convert to B&W or use film 🙂

 

[jesse1dog]

I can see the effect but wonder about the pinhole itself.
Did you construct it or is it a commercial variety?
Have you tried several different pinholes both size and 'make'?
Is it the camera? I haven't noticed this effect using my Panasonic GF1 with several different pinholes.
Head scratching - hope you have some hair left!

 

[daveleo]

John,

It's not the pinhole. I blew it clean (bulb blower) and switched it out with another - same effect.
Is it the camera? Have to say "no" - except it's probably the sensor and light angles, per the above.
It is a quality brand - Skink, so not a cheapo.

 

[JoeV]

Keep in mind that with glass optics, the light exiting the rear element onto the sensor does so using most of the diameter of that rear element, so that the angle of incidence upon the sensor, near the edges and corners, is relatively mild.

But with a pinhole, the light exits the pinhole aperture near the center of the sensor, and so the angle of incidence near the edges is much more severe.

It's not due to the quality of the pinhole but the fact that it is a pinhole. If you make the focal length much longer (and thus the angle of view much narrower), you can minimize this effect. Or convert the files to monochrome.

~Joe

 

[rbiemer]

Dave,
You may be getting to the "diffraction limit" of your pinhole. A larger pinhole would probably reduce the effect. At the expense of apparent sharpness.
Rob

 

[daveleo]

For the record, I got a note that the program CornerFix can correct this. Reading up on it, it needs a DNG file of the image.

I am going to see if I can attach the Skink pinhole to my Minolta-Fuji adapter. That will push it away from the sensor quite a way.
Probably have to tape it on 😛
Let's see what happens.

 

[daveleo]

Dave,
You may be getting to the "diffraction limit" of your pinhole. A larger pinhole would probably reduce the effect. At the expense of apparent sharpness.
Rob

I tried a few different plates and all the same.

 

[daveleo]

Keep in mind that with glass optics, the light exiting the rear element onto the sensor does so using most of the diameter of that rear element, so that the angle of incidence upon the sensor, near the edges and corners, is relatively mild.

But with a pinhole, the light exits the pinhole aperture near the center of the sensor, and so the angle of incidence near the edges is much more severe.

It's not due to the quality of the pinhole but the fact that it is a pinhole. If you make the focal length much longer (and thus the angle of view much narrower), you can minimize this effect. Or convert the files to monochrome.

~Joe

Yeh, that's pretty much how I understand it now.

 

[daveleo]

Okay, I couldn't wait. I mounted the minolta-Fuji lens adapter on the XA1. This pushed the lens plane out more than an inch (?). I just held the Skink plate against the outer edge of the adapter and handheld this shot through a window.
Ta-da !! . . . colors gradients are gone. So . . . all the comments above were pointing in the right direction. Angles at the image edges and digital sensors, etc.
The implication here is that using pinhole lenses on mirrorless digital cameras requires some re-thinking and workarounds.
.