Distortion and Glare in the Swarovski 8x32 EL Swarovision (2 Viewers)

brocknroller · Jul 23, 2015

Pinewood said:
Hello James,

Yes, some binocular owners believe that their favourite glass deserves reverence, even obeisance, as their binocular becomes a fetish.

Whether if five or 20% suffer from rolling ball, it is enough that bird watchers should know about that Austrian binocular's shortcomings. There comes a point where a design compromise becomes a design fault for some. If the manufacturer corrects the problem, the consumer benefits. It is of importance to many, who might consider rolling ball, a "deal breaker."

Some folks seem to identify far too closely with their possessions.

Happy bird watching,
Arthur :hi:

Amen!

To quote from my old friend Kofi:

Knowledge is power. Information is liberating. Education is the premise of progress, in every society, in every optics forum. --Kofi Annan

<B>

typo · Jul 23, 2015

henry link said:
I hate to ask you to do more, but just as a check of the phone camera's distortion you might try making a couple of photos of a circle centered in both the binocular and the camera fields where there should be no distortion of the shape. Make one photo with the camera axis parallel to the binocular and the other with the camera tilted the same way it's tilted for these photos. That should tell us whether tilting the camera affects distortion.

Henry

Henry,

The previous edge shots had been cropped a little to the right, that is outside the binocular field of view so the very edge of the camera was cut. In these two shots the circle with the strip of tape was taken to be the centre of the binocular FOV and the first is with the camera centred and the second tilted to use the left edge of the camera view.

David

henry link · Jul 23, 2015

David,

Thanks, I've been puzzling over your photos. Obviously there's no problem with the three circle centered image, but that one appears to be so closely cropped that camera distortion would probably be out of the frame. The tilted four circle image appears to include the whole frame horizontally (is that right?), so it would include camera distortion. It appears to maybe show the effects of some pincushion distortion in the camera on the left circle. I can't tell for sure whether the camera tilt is doing anything, too many possible sources of distortion (binocular, camera, tilt) working at the same time.

In my set-up (Nikon D-40 with the kit lens set at 55mm) there is no measurable distortion within the area subtended by the five target circles (about 60% of the frame horizontally) and tilting the camera the amount that I have to tilt it doesn't change anything. Could be I'm just lucky that my stuff happens to work.

Henry

Binastro · Jul 23, 2015

David's camera lens may not be perfectly centred. Most aren't.

regards, Binastro

typo · Jul 23, 2015

Henry,

I'm not sure I quite follow what you are saying but I'll try to spell out what I've done.

The Prime has a 6.5* view. The circles are 68mm at a distance of about 9.5m and that should give 15.8 circles across the field of view and I estimated 15.5 from the images so that's pretty close. With the phone camera across the short axis I get about just a little more than the centre to the edge so something like a 35-40* view. The 4 circle shot in#62 actually covered from just left of centre to just right of the edge but cropped only from the right to show just the circles and the true left edge of the frame. Because of the distortion in the first photos I originally posted, I angled subsequent shots further out by approximately a circle width and then cropped to the edge of the field of view. That last photo in #62 has only been cropped from the right. The camera distortion in #29 should be minimal. Unfortunately hand holding the phone means there is probably vertical and well as horizontal tilt in the shot.

I had hoped to use my micro 4/3 camera with the kit lens at 14mm which should have given a 75% view, more than would be needed for the Prime. Unfortunately I was unable to position it at the exit pupil to view the edges even when zoomed to 42mm.

The D40 kit lens at 55mm I believe gives a 29* view or a bit under half the AFOV of the Swaro. That would cover about nine circles out of a 20 circle full FOV. Is that right?

David

absolut_beethoven · Jul 24, 2015

A belated thanks of gratitude Henry.

I always learn some new and useful info from you. Plus you handle the few naysayers politely with grace. BF should pay you for your time and effort that most of us here gratefully appreciate!!

henry link · Jul 25, 2015

typo said:
Henry,

I'm not sure I quite follow what you are saying but I'll try to spell out what I've done.

The Prime has a 6.5* view. The circles are 68mm at a distance of about 9.5m and that should give 15.8 circles across the field of view and I estimated 15.5 from the images so that's pretty close. With the phone camera across the short axis I get about just a little more than the centre to the edge so something like a 35-40* view. The 4 circle shot in#62 actually covered from just left of centre to just right of the edge but cropped only from the right to show just the circles and the true left edge of the frame. Because of the distortion in the first photos I originally posted, I angled subsequent shots further out by approximately a circle width and then cropped to the edge of the field of view. That last photo in #62 has only been cropped from the right. The camera distortion in #29 should be minimal. Unfortunately hand holding the phone means there is probably vertical and well as horizontal tilt in the shot.

I had hoped to use my micro 4/3 camera with the kit lens at 14mm which should have given a 75% view, more than would be needed for the Prime. Unfortunately I was unable to position it at the exit pupil to view the edges even when zoomed to 42mm.

The D40 kit lens at 55mm I believe gives a 29* view or a bit under half the AFOV of the Swaro. That would cover about nine circles out of a 20 circle full FOV. Is that right?

David

David,

Sorry I haven't replied sooner. I made more photos that might indicate a small amount of horizontal stretching in my set-up when the camera is tilted, but I haven't had time to put them on the computer and follow through. For comparative purposes it won't matter and even for absolute measurements it looks like it will only introduce a tiny amount of shape distortion. I'll try to post something in the next few days. And yes, nine out of 20 circles is about right.

Thanks absolut_beethoven.

Henry

Holger Merlitz · Jul 26, 2015

brocknroller said:
I agree, now if you can just convince Jerry, who keeps quoting Dobler's lower estimate, even though it's not clear where Dobler got his 5-10% figures from or if they represent all who see RB or only those who see it and can't adapt to it, which was the group Zeiss was interested in contacting when designing the SF.

I hope to get answers to those questions. If the numbers are from a study, and it's been published online, I will send you the link or forward it to you if it's sent to me via email.

What leads you to estimate 30% see RB?

You make an interesting point about not knowing whether trying newer sample SV ELs and finding less RB is due to Swaro tweaking the formula or if you are getting used to the "mustache effect." I wondered about that, too, that is, once someone is exposed to the PC/AMD pattern in the SV EL or some other low distortion bin, if months later, they were exposed to it again, if the brain "remembers" that pattern and then compensates for it? Sort of like immunization therapy where an allergist keeps exposing you to the allergens until your body adjusts and is no longer sensitive to it.

If Swaro has tweaked the SV EL's distortion formula, it should be easy to tell by doing what Henry did with the circle photos and comparing the original SV ELs to the latest ones made.

Brock

Hello Brock,

Apparently, they have not carried out systematic studies about the globe effect, just a couple of rather informal gatherings which took place long time ago when Herr Dobler was still working with Swarovski. He has informed me that he doesn't have any data.

My 30% guesstimate is not fully waterproof either, and based on some assumptions. Look at the lowermost figure in

http://www.holgermerlitz.de/globe/test_distortion.html

This histogram summarizes the result of the checker board test. Those observers who have chosen 'A' are likely to have a very low amount of distortion of their visual system. Those who chose 'B' seem to have a minor barrel distortion, that is why they chose checker board 'B' with a slight pincushion distortion (which compensates for their barrel distortion) as their personal 'distortion-free' candidate, and so forth.

If a binocular that is free of distortion were panned, then the A-guys would be fine, while the B-guys would have to deal with a minor residual barrel distortion (the one of their own perception). Would that be sufficient to generate a globe-effect? Probably not, as my computer animations would indicate: The B-guys would still perceive an almost perfectly fine panning behavior. Somewhere between C and D, the residual barrel distortion would become obvious enough to enable the perception of the globe effect. I am quite sure that practically all individuals who have chosen D, E of F would clearly perceive the globe effect - look at Fig. 7 in

http://www.holgermerlitz.de/globe/distortion.html

These animations are generated while assuming a visual distortion of 0.6, corresponding to those guys who would choose 'D' in the checker board experiment. The lowermost row is calculated for distortion-free binoculars. Those guys would see quite a significant globe effect. Even with a distortion of 0.7 (checker board C), the globe effect is already quite pronounced, if the binocular is free of distortion (I had simulations for that, years ago, which are not shown here).

Now, the point is that the manufacturers like Swarovski are already moving away from the (almost) zero-distortion paradigm and are adding at least a little bit of pincushion. That reduces the fraction of people who could possibly perceive the globe effect. On the other hand, some of the real life binoculars do not at all have monotonous distortion curves, but curves which drop off near the edge. That does amplify the perception of the globe effect - actually, it is a local barrel distortion which is built into the optical formula. The combination of the pincushion distortion in the central regions of the field, then the inflection point, then the barrel distortion near the edges, leads to the mustache effect, which is not even included in my simulations and tests. Here I have simulated the particular example of the BPO

http://www.holgermerlitz.de/bpo7x30/bpo7x30.html

and how it does behave during panning.

In order to know how strong the globe effect is perceived, we need both distortion curves, the one of the individual observer and the one of the particular binocular he is using. Regarding the observers, we at least have some statistics, but regarding the instrument, the manufacturers should publish their distortion curves, something they refuse to do. That is why I have that disclaimer in

http://www.holgermerlitz.de/newk/newk.html

in which I note: Another problem arises from the fact that the perception of the globe effect is not solely related to the k-factor, but also depending on the detailed shape of the distortion curve (from the center toward the edge), and these curves are not published by the manufacturers.

It is hard to judge the panning behavior of a binocular from the k-number alone, as long as there exist binoculars with mustache distortion of unknown intensity. Here, Henry's contributions are really useful, they allow us to get at least a couple of hints.

I finally had to give up any further studies on this subject, because no manufacturer was willing to supply the necessary data, i.e. the distortion curves of their binoculars.

Cheers,
Holger

ronh · Jul 28, 2015

Henry,

I don't know how I missed this but am pleased to discover it. There is a lot of discussion that I will read when I get a chance. For now I will just say I think your tests are very useful, if one only can add the essential next step of what the eye does. That is tricky of course, and variable with user and conditions, which is the very reasons that overall performance reviews are so subjective and sometimes seem flaky. You have done well what you intended. A+ for blocking the exit pupil to simulate dawn and dusk viewing!

Ron

Binastro · Jul 28, 2015

I think I'm bothered more than most by the globe effect in some binoculars, because I mainly pan along the horizon looking for things.
The change in magnification is slightly bothersome.
My reasoning is, why should I bother with a slightly weird binocular, when I can use one which is very restful?

I also get this to a lesser extent with the Nikon Action VII binocular, which has some aspherics in the eyepiece.

NDhunter · Jul 29, 2015

Holger Merlitz said:
Hello Brock,

Apparently, they have not carried out systematic studies about the globe effect, just a couple of rather informal gatherings which took place long time ago when Herr Dobler was still working with Swarovski. He has informed me that he doesn't have any data.

My 30% guesstimate is not fully waterproof either, and based on some assumptions. Look at the lowermost figure in

http://www.holgermerlitz.de/globe/test_distortion.html

This histogram summarizes the result of the checker board test. Those observers who have chosen 'A' are likely to have a very low amount of distortion of their visual system. Those who chose 'B' seem to have a minor barrel distortion, that is why they chose checker board 'B' with a slight pincushion distortion (which compensates for their barrel distortion) as their personal 'distortion-free' candidate, and so forth.

If a binocular that is free of distortion were panned, then the A-guys would be fine, while the B-guys would have to deal with a minor residual barrel distortion (the one of their own perception). Would that be sufficient to generate a globe-effect? Probably not, as my computer animations would indicate: The B-guys would still perceive an almost perfectly fine panning behavior. Somewhere between C and D, the residual barrel distortion would become obvious enough to enable the perception of the globe effect. I am quite sure that practically all individuals who have chosen D, E of F would clearly perceive the globe effect - look at Fig. 7 in

http://www.holgermerlitz.de/globe/distortion.html

These animations are generated while assuming a visual distortion of 0.6, corresponding to those guys who would choose 'D' in the checker board experiment. The lowermost row is calculated for distortion-free binoculars. Those guys would see quite a significant globe effect. Even with a distortion of 0.7 (checker board C), the globe effect is already quite pronounced, if the binocular is free of distortion (I had simulations for that, years ago, which are not shown here).

Now, the point is that the manufacturers like Swarovski are already moving away from the (almost) zero-distortion paradigm and are adding at least a little bit of pincushion. That reduces the fraction of people who could possibly perceive the globe effect. On the other hand, some of the real life binoculars do not at all have monotonous distortion curves, but curves which drop off near the edge. That does amplify the perception of the globe effect - actually, it is a local barrel distortion which is built into the optical formula. The combination of the pincushion distortion in the central regions of the field, then the inflection point, then the barrel distortion near the edges, leads to the mustache effect, which is not even included in my simulations and tests. Here I have simulated the particular example of the BPO

http://www.holgermerlitz.de/bpo7x30/bpo7x30.html

and how it does behave during panning.

In order to know how strong the globe effect is perceived, we need both distortion curves, the one of the individual observer and the one of the particular binocular he is using. Regarding the observers, we at least have some statistics, but regarding the instrument, the manufacturers should publish their distortion curves, something they refuse to do. That is why I have that disclaimer in

http://www.holgermerlitz.de/newk/newk.html

in which I note: Another problem arises from the fact that the perception of the globe effect is not solely related to the k-factor, but also depending on the detailed shape of the distortion curve (from the center toward the edge), and these curves are not published by the manufacturers.

It is hard to judge the panning behavior of a binocular from the k-number alone, as long as there exist binoculars with mustache distortion of unknown intensity. Here, Henry's contributions are really useful, they allow us to get at least a couple of hints.

I finally had to give up any further studies on this subject, because no manufacturer was willing to supply the necessary data, i.e. the distortion curves of their binoculars.

Cheers,
Holger

Holger:

I do like your study of the globe effect, and have read it thoroughly.

Tell us how you got to your 30% of the population, ?

Jerry

JabaliHunter · Jul 29, 2015

[email protected] said:
The SV 8x32 is still the best 8x32 birding binocular on the market today.

[email protected] said:
If you don't buy it [...] you are missing out on the best 8x32 birding binocular made.

[email protected] said:
overall it is one of the best 8x32 binoculars out there

Do you at least agree that it is probably a 8x32 ?

JabaliHunter · Jul 29, 2015

NDhunter said:
The important thing for me, and most everyone here, is how they perform in the field. These photos do not mean anything in daily use

[email protected] said:
The point is are any of these anomalies really affecting the view that much.

The point of this thread, as clearly stated in the OP, was to illustrate some differences that have been topics of discussion. For that Henry should be applauded.

There are hundreds of other posts, reviews and sponsored and other articles available to the casual reader which attempt to describe how binoculars perform in the field.

The point of posts such as these is to seek explanations for why they perform in certain ways in the field.

That is what sets this place apart and without it there would be precious little to discuss.

Pileatus · Jul 29, 2015

JabaliHunter said:
The point of this thread, as clearly stated in the OP, was to illustrate some differences that have been topics of discussion. For that Henry should be applauded.

There are hundreds of other posts, reviews and sponsored and other articles available to the casual reader which attempt to describe how binoculars perform in the field.

The point of posts such as these is to seek explanations for why they perform in certain ways in the field.

That is what sets this place apart and without it there would be precious little to discuss.

Excellent post and, yes, Henry should be thanked for his many illustrative contributions.

Holger Merlitz · Jul 29, 2015

NDhunter said:
Holger:

I do like your study of the globe effect, and have read it thoroughly.

Tell us how you got to your 30% of the population, ?

Jerry

Hi Jerry,

I thought I had mentioned it - let's go back to the lowermost figure in

http://www.holgermerlitz.de/globe/test_distortion.html

Those guys who selected D, E or F should perceive a significant globe effect (when using a binocular that is practically free of distortion), considering the quantity of their own visual distortion. I suppose that some, though perhaps not all of the guys who selected C would at least see a glimpse of the globe effect (since I do, and I belong to the C or the D group, somewhere in between I guess). Roughly speaking, there are easily 30%, if not more, who belong to these groups, and who might perceive the globe effect. Note: I don't claim that they would see the globe effect through any particular Swarovski 8x32 SV, or any other particular binocular, since those have already got some pincushion. I just claim that these 30% would see a globe effect if the binocular were free (or nearly free) of distortion.

The manufacturers know that, and nobody nowadays dares to make a binocular entirely without any distortion. The question is: How much is needed so that a vast majority of binocular users don't see that globe effect. I would guess that a k-value of about 0.7 would be fine, but that would assume the absence of any mustache - anomaly. Since many low-distortion binoculars nowadays come with such a mustache, I cannot give quantitative numbers for these binoculars and the strength of their globe effect, as long as I don't have their distortion curves.

Cheers,
Holger

brocknroller · Jul 29, 2015

Holger Merlitz said:
Hello Brock,

Apparently, they have not carried out systematic studies about the globe effect, just a couple of rather informal gatherings which took place long time ago when Herr Dobler was still working with Swarovski. He has informed me that he doesn't have any data.

My 30% guesstimate is not fully waterproof either, and based on some assumptions. Look at the lowermost figure in

http://www.holgermerlitz.de/globe/test_distortion.html

This histogram summarizes the result of the checker board test. Those observers who have chosen 'A' are likely to have a very low amount of distortion of their visual system. Those who chose 'B' seem to have a minor barrel distortion, that is why they chose checker board 'B' with a slight pincushion distortion (which compensates for their barrel distortion) as their personal 'distortion-free' candidate, and so forth.

If a binocular that is free of distortion were panned, then the A-guys would be fine, while the B-guys would have to deal with a minor residual barrel distortion (the one of their own perception). Would that be sufficient to generate a globe-effect? Probably not, as my computer animations would indicate: The B-guys would still perceive an almost perfectly fine panning behavior. Somewhere between C and D, the residual barrel distortion would become obvious enough to enable the perception of the globe effect. I am quite sure that practically all individuals who have chosen D, E of F would clearly perceive the globe effect - look at Fig. 7 in

http://www.holgermerlitz.de/globe/distortion.html

These animations are generated while assuming a visual distortion of 0.6, corresponding to those guys who would choose 'D' in the checker board experiment. The lowermost row is calculated for distortion-free binoculars. Those guys would see quite a significant globe effect. Even with a distortion of 0.7 (checker board C), the globe effect is already quite pronounced, if the binocular is free of distortion (I had simulations for that, years ago, which are not shown here).

Now, the point is that the manufacturers like Swarovski are already moving away from the (almost) zero-distortion paradigm and are adding at least a little bit of pincushion. That reduces the fraction of people who could possibly perceive the globe effect. On the other hand, some of the real life binoculars do not at all have monotonous distortion curves, but curves which drop off near the edge. That does amplify the perception of the globe effect - actually, it is a local barrel distortion which is built into the optical formula. The combination of the pincushion distortion in the central regions of the field, then the inflection point, then the barrel distortion near the edges, leads to the mustache effect, which is not even included in my simulations and tests. Here I have simulated the particular example of the BPO

http://www.holgermerlitz.de/bpo7x30/bpo7x30.html

and how it does behave during panning.

In order to know how strong the globe effect is perceived, we need both distortion curves, the one of the individual observer and the one of the particular binocular he is using. Regarding the observers, we at least have some statistics, but regarding the instrument, the manufacturers should publish their distortion curves, something they refuse to do. That is why I have that disclaimer in

http://www.holgermerlitz.de/newk/newk.html

in which I note: Another problem arises from the fact that the perception of the globe effect is not solely related to the k-factor, but also depending on the detailed shape of the distortion curve (from the center toward the edge), and these curves are not published by the manufacturers.

It is hard to judge the panning behavior of a binocular from the k-number alone, as long as there exist binoculars with mustache distortion of unknown intensity. Here, Henry's contributions are really useful, they allow us to get at least a couple of hints.

I finally had to give up any further studies on this subject, because no manufacturer was willing to supply the necessary data, i.e. the distortion curves of their binoculars.

Cheers,
Holger

Holger,

Just catching up. Thanks for your detailed reply. Very interesting to know that the 5-10% figures that were being quoted here repeatedly were not from the "Drs" at Zeiss but from informal gatherings which took place a long time ago when Herr Dobler was still working with Swarovski, and that Dobler has no data to support those numbers.

The fact that Swaro underestimated the number susceptible to RB explains the number of people reporting RB in the first generation of SV ELs and why Swarovski is now backpedaling from its initial design and adding more pincushion to the SV EL line, which makes me wonder if the biggest difference in the new ProField SV ELs is not advertized - more pincushion in all models?

After looking at the histogram, I see how you derived your 30% guesstimate. I also took that test but never sent you the results because my dad got sick, went into the hospital for 11 days, and we had home healthcare workers coming in for months afterward while he recovered, and by that time, you had completed your study. But had I sent in my results, you would have had another data point of RB susceptibility to add to the graph.

The odd thing is that when Dobler led the design team that made the Zeiss SF he still thought that RB affected only 5-10%, as he told Lee, so it surprises me that the would go back to that sample group from years back and use them as "guinea pigs" to find the best balance between PC and AMD. Here's what he said to Lee:

We found a good balance between these different things and especially in the control of the so-called rolling ball or globus effect that affects between 5 and 10% of people. We contacted some of these people and asked them to try different binoculars having different levels of field-flattening and we came up with a value that gave the best balance between field flatness and control of the globus effect. This means SF is not quite so flat-field at the very edge but it is nearly so, and from what those people told us almost everybody should be able to enjoy SF without noticing any globus effect.

I can see how Jerry thought he was speaking about a recent study since they used feedback from that group to design the SF.

But some still do see RB in the 8x42 SF, and that can be explained by the distortion pattern rather than the amount of pincushion. As you noted, it's not a simple matter of the k-value with the SV ELs because of its "mustache" distortion pattern but rather how much it dips at the edges that can create the globe effect in some individuals. Apparently, the distortion pattern in the 8x42 SF also causes some people to experience RB.

Once this is common knowledge to designers, they will take both the k-value and the distortion curve into consideration when designing binoculars they want to be "sharp to the edge" (which many companies claim but few deliver). I was quite happy with the edge sharpness on the 10x42 SLC-HD, 10x42 SE, and the 10x42 EDG. It seems that you can get good edge performance (though perhaps not to the very, very edge) w/out having to lower distortion levels so much that it causes RB or making wavy distortion patterns that also causes RB simply to eek out that last 5-10% edge sharpness that is not useful in most birding situations anyway. Aside from Edz and a few others over on Cloudy Nights, even most amateur astronomers aren't so obsessive about edge performance as to worry about the outer 5% of the edge. It surprises me that some birders do.

Thanks again.

Brock

perterra · Aug 5, 2015

Sounds like Swaro has some issues that need to be worked out before I buy one. 3

I'm sorry, I couldnt help myself. :-C

HighNorth · Aug 5, 2015

perterra said:
Sounds like Swaro has some issues that need to be worked out before I buy one. 3

I'm sorry, I couldnt help myself. :-C

:t:

Pileatus · Aug 6, 2015

perterra said:
Sounds like Swaro has some issues that need to be worked out before I buy one. 3

I'm sorry, I couldnt help myself. :-C

They're main issue is keeping up with demand. Your hesitation to purchase will surely allow someone to take a well-deserved vacation. :t:

perterra · Aug 6, 2015

Pileatus said:
They're main issue is keeping up with demand. Your hesitation to purchase will surely allow someone to take a well-deserved vacation. :t:

:-O

I'm not planning on buying a high end Swaro or a Zeiss. I really did like the Leica 8X32 ultravid HD I played around with the other day, I just didnt like it enough to spend $2,000.

Distortion and Glare in the Swarovski 8x32 EL Swarovision (2 Viewers)

porromaniac

Well-known member

Attachments

Well-known member

Well-known member

Well-known member

Well-known member

Well-known member

Well-known member

Well-known member

Well-known member

Experienced observer

Well-known member

Well-known member

"Experientia Docet”

Well-known member

porromaniac

Well-known member

Well-known member

"Experientia Docet”

Well-known member

Similar threads

Users who are viewing this thread