Globe effect/rolling ball, there is more to it (1 Viewer)

[kimmik]

Here is a puzzle for you.

Why doesn’t camera viewfinder (dslr, mirrorless, superzoom etc) show globe effect?

The distortion is pure rectilinear, which supposedly should cause extreme globe effect.

When zoomed in, the magnification can be similar to binocular. The AFOV is pretty good too.

So in addition to:
K value
And uniformity of distortion/higher order distortion
(Distortion and globe effect in binoculars)

I have yet to see anyone describe this phenomenon, which I think is a major contributor to why some binoculars have stronger rolling ball globe effect than others.

Focus the binocular to infinity, then look at straight lines in the foreground. Then repeat the opposite.

Not sure what this is called, something similar to focus breathing which is a change in magnification, except here it is a change in distortion in the out of focus planes.

(And as you can imagine, panning will cause the background to bowl, and foreground to ball, which completes the rolling globe appearance)

@Holger Merlitz

 

[WJC]

Modern cameras have many more lenses than binoculars and photographers are willing to pay for the purity.

 

[jafritten]

View attachment 1491713

Here is a puzzle for you.

Why doesn’t camera viewfinder (dslr, mirrorless, superzoom etc) show globe effect?

I have always thought the GE was an outcome of a certain design of the eyepiece. I am curious to learn about this.

 

[exup]

Distortion and globe effect in binoculars

 

[mfunnell]

Distortion and globe effect in binoculars

Indeed. And from that article:

“The binocular generates its image inside the image space. Unlike a camera, which would record that image space directly onto its photographic sensor, the image undergoes an additional transformation when perceived by the human eye, thereby picking up a barrel distortion, or convex curvature, that eventually generates the globe effect of the panning binocular.”

I would note, also, in the case of SLR cameras, that the image space as presented on the focus screen acts in the same role. (There’s also a possible complication if there is additional processing - say, to apply software-based corrections to compensate for lens distortions - between the image at the sensor and that presented by the EVF of modern “mirrorless” cameras.)

I’ll further note that this doesn’t happen with afocal viewfinders such as those in Leica rangefinder cameras. Mine have noticeable barrel distortion (and present something of a rolling-ball effect). This is mitigated by the low magnification of the finder (.93 in the case of my M3, my highest-magnification finder) and because I’m usually more concerned with framelines and the RF patch than I am in a detailed assessment of the view (which usually happens before raising the camera to my eye).

…Mike

 

[Holger Merlitz]

View attachment 1491713

Here is a puzzle for you.

Why doesn’t camera viewfinder (dslr, mirrorless, superzoom etc) show globe effect?

The distortion is pure rectilinear, which supposedly should cause extreme globe effect.

When zoomed in, the magnification can be similar to binocular. The AFOV is pretty good too.

So in addition to:
K value
And uniformity of distortion/higher order distortion
(Distortion and globe effect in binoculars)

I have yet to see anyone describe this phenomenon, which I think is a major contributor to why some binoculars have stronger rolling ball globe effect than others.

Focus the binocular to infinity, then look at straight lines in the foreground. Then repeat the opposite.

Not sure what this is called, something similar to focus breathing which is a change in magnification, except here it is a change in distortion in the out of focus planes.

(And as you can imagine, panning will cause the background to bowl, and foreground to ball, which completes the rolling globe appearance)

@Holger Merlitz

With my Fuji camera the AFOV of the viewfinder is pretty narrow. The globe effect would be obvious only if the apparent angle is sufficiently wide and additionally the magnification sufficiently high so that the image points flow smoothly in front of the eye.

Cheers,
Holger

 

[kimmik]

I’m thinking it could be called: kappa instability.

As a function of the in-plane kappa value, 0 KI means foreground and background have the same k value as in plane.

The unit might be say, kappa per diopter.

 

[tenex]

The camera question has been addressed, but the OP is also suggesting that K isn't a single parameter, but different for background, focal plane, and foreground. So I think kimmik is wondering whether the background K when focused on a nearby object can be significantly different from the canonical K at infinity focus, and affect the RB sensation. Is that right? Of course if RB is reduced simply by having the background out of focus, it may not matter.

 

[kimmik]

K isn't a single parameter, but different for background, focal plane, and foreground.

Correct, thank you for following my logic.

The K value distortion commonly discussed, is the special case K(in-plane).

K(foreground) and K(background) can be different.

I suspect this effect is due to putting strong glass curvatures near the internal focal point, which then applies different degrees of distortion to foreground and background (will do a simulation later).

This explains why I find more globe sensation in NL than the SF, even though they have similar K(in plane). The SF uses less strong field lens based on cross section photos.

if RB is reduced simply by having the background out of focus, it may not matter.

Correct, bokeh alleviates the effect.

Strongest noticeable effect when low bokeh, ie low mag small pupil.

 

[Holger Merlitz]

This is interesting. The distortion of out-of-focus areas may indeed differ from the in-focus distortion curve. How exactly it differs would depend on the particular optical layout of the eyepiece. I fear that a ray-tracing would be needed for a quantitative evaluation of such a feature. It could obviously affect the intensity of our perception of the globe effect, since the rolling background would not necessarily be in focus. Thanks for your heads up!

Cheers,
Holger

 

[kimmik]

On my very crude simulation (and simplified without eyepiece or objective), there is interaction between the focuser and the field flattener, which results in changing distortion profile (observe the angle between rays, center vs edge).

It would be very cool to see a proper simulation and mathematical analysis someday!

 

[has530]

An interesting theory but how would you explain the fact that I perceive the same effect in the daytime and on the night sky (where there are no out of focus areas as everything is at infinity)?

 

[kimmik]

Apologies I think I misunderstood the question.

What binocular are you describing?

 

[has530]

In my Vortex Viper and my canon 15x50 IS, I perceive a strong rolling ball effect while panning for terrestrial observations and stellar observations (where all objects lie in the same focal plane so focus dependent distortion should not play a roll).

 

[kimmik]

Canon 15x50 - two thoughts. the high mag and aperture blurs out the foreground background so much that it alleviates the additional KI variable. Second thought is does it use a focusing element, or does it move the objective lens? Seems in my binos, the ones that use external focus (either objective motion or eyepiece motion) dont have much KI.

Vortex viper - this is internally focused so should have some degree of KI, but if it is higher mag and higher aperture, may alleviate the KI effect. Or perhaps it has low enough KI due to design details, that it is unnoticeable.