Which roofs are better for their 3D view? (1 Viewer)

henry link · May 11, 2018

black crow said:
Just curious if anyone knows at what distance the 3D effect of porros actually diminishes to the point where it's not noticeable?

I think it's a surprisingly long distance.

I just did some observing of overlapping fence posts through some neighbors' backyards with a tripod mounted Porro 8x and a roof 8x. I found that a fence post at a distance of 300' very obviously changed position relative to an overlapping fence post at a distance of 250' between the left and right sides of both the Porro and the roof. Those were the longest distances I had available, but given how obvious the parallax shift was even in the roof I expect it wouldn't disappear in the Porro until distances are well beyond 1000', perhaps even 2000' (for objects with the same 20% difference in distance). Maybe somebody with easy access to longer distances could make the same observations.

Henry

Binastro · May 11, 2018

3D depends on magnification and distance and the observer's acuity.
It has little to do with the distance between objectives, which varies little between binoculars, except reversed Porroprism binoculars.
Even with these 3D is easily observed because of magnification, say 6x to 8x.

Looking at a good target against distant but unmoving clouds or the sky, 3D has been observed at about 600 yards without optical aid and sometimes even further.
I easily see 3D with rested eyes at 130 yards and probably considerably further, probably 300 yards when younger.

With ship's rangefinders distances up to 15,000 yards are easily measured by young but skilled mariners.
The magnification is maybe 15x to 25x and baselines up to 6 metres, say 7 yards.
Smaller army rangefinders are from around 0.5m to 2m baseline. Also some 3m baseline.
There are different principles between British and U.S. rangefinders.
Nowadays laser rangefinders are used.

There is a lot written here about objective spacing but little about magnification, which is the dominant part of the equation with the binoculars used.

Personally, I think that in some situations, vernier acuity is also used.

So using a steady tripod mounted 8x binocular with objective spacing of say 65mm I would expect those with very good sight to perceive genuine 3D up to 5,000 yards against a suitable high contrast distant background.
With hand held binoculars the distance would be considerably reduced, perhaps half or less.

With a 20% difference in spacing this would come down to 1,000 yards or 3,000ft assuming very clear conditions and very steady seeing. This would probably need 20/12 eyesight.

Furthermore, it is possible to see 3D with one eye at close distance, which uses the pupil size.
Also with large telescopes the aperture itself can be used to see 3D.

WJC · May 11, 2018

Binastro said:
3D depends on magnification and distance and the observer's acuity.
It has little to do with the distance between objectives, which varies little between binoculars, except reversed Porroprism binoculars.
Even with these 3D is easily observed because of magnification, say 6x to 8x.

Looking at a good target against distant but unmoving clouds or the sky, 3D has been observed at about 600 yards without optical aid and sometimes even further.
I easily see 3D with rested eyes at 130 yards and probably considerably further, probably 300 yards when younger.

With ship's rangefinders distances up to 15,000 yards are easily measured by young but skilled mariners.
The magnification is maybe 15x to 25x and baselines up to 6 metres, say 7 yards.
Smaller army rangefinders are from around 0.5m to 2m baseline. Also some 3m baseline.
There are different principles between British and U.S. rangefinders.
Nowadays laser rangefinders are used.

There is a lot written here about objective spacing but little about magnification, which is the dominant part of the equation with the binoculars used.

Personally, I think that in some situations, vernier acuity is also used.

So using a steady tripod mounted 8x binocular with objective spacing of say 65mm I would expect those with very good sight to perceive genuine 3D up to 5,000 yards against a suitable high contrast distant background.
With hand held binoculars the distance would be considerably reduced, perhaps half or less.

With a 20% difference in spacing this would come down to 1,000 yards or 3,000ft assuming very clear conditions and very steady seeing. This would probably need 20/12 eyesight.

Furthermore, it is possible to see 3D with one eye at close distance, which uses the pupil size.
Also with large telescopes the aperture itself can be used to see 3D.

Binastro:

Although what you have said is true, you may be going into that glorious realm I walked into when trying to explain 3-axis collimation Vs. the ever-popular—but equally wrong—conditional alignment.

With stereopsis having been forever attributed to objective separation, saying it has little to do with that separation cries out for more clinical input. You have the background and if you have the time all could benefit. :cat:

Cheers,

Bill

PS I have worked on those Naval rangefinders ... unbelievable!

elkcub · May 11, 2018

Binastro said:
...

Furthermore, it is possible to see 3D with one eye at close distance, which uses the pupil size.
Also with large telescopes the aperture itself can be used to see 3D.

The attached article by James J. Gibson appeared in the 1958 book entitled "Readings in Perception" by Beardslee and Wertheimer. It provides important insights concerning 3-D spatial perception and is presented in the context of aviation primarily because the work was supported by a US Air Force grant. Otherwise, it provides a good introduction to his theory of visual gradients, which has never been disproven.

After opening (on your desktop) in Adobe Reader, rotate the view clockwise 90 deg.

I've long been convinced that Gibson came closer to understanding the 3-D perceptual process than anyone else.

Ed

Theo98 · May 11, 2018

elkcub said:
The attached article by James J. Gibson appeared in the 1958 book entitled "Readings in Perception" by Beardslee and Wertheimer. It provides important insights concerning 3-D spatial perception and is presented in the context of aviation primarily because the work was supported by a US Air Force grant. Otherwise, it provides a good introduction to his theory of visual gradients, which has never been disproven.

After opening (on your desktop) in Adobe Reader, rotate the view clockwise 90 deg.

I've long been convinced that Gibson came closer to understanding the 3-D perceptual process than anyone else.

Ed

Thanks for the informational reference Ed...Very Interesting! :t:

Ted

WJC · May 11, 2018

elkcub said:
The attached article by James J. Gibson appeared in the 1958 book entitled "Readings in Perception" by Beardslee and Wertheimer. It provides important insights concerning 3-D spatial perception and is presented in the context of aviation primarily because the work was supported by a US Air Force grant. Otherwise, it provides a good introduction to his theory of visual gradients, which has never been disproven.

After opening (on your desktop) in Adobe Reader, rotate the view clockwise 90 deg.

I've long been convinced that Gibson came closer to understanding the 3-D perceptual process than anyone else.

Ed

Super, Ed, and THANKS! :cat:

Bill

PS Stereopsis is congenital, and measurable performance varies from observer to observer. BINOCULAR VISION DOES NOT ALWAYS IMPLY STEREOSCOPIC ABILITY. — Applied Photographic Optics, Sidney F. Rey, Focal Press, 1988.

* The caps are mine.

Binastro · May 12, 2018

My two old computers could not access the pdfs referenced by Ed, but I was able to read an analysis of Gibson's ideas by E. Bruce Goldstein 1981.
I can read some pdf files but not the referenced one.

However, I consider Gibson's and Goldstein's work too complex regarding how far an image can be perceived in 3D with unaided eyes and with optical aid.
Vision is indeed very complex, but in essence as far as the distance where 3D is seen, rather simple.

My approach relies on simple mathematics and observational experience and may I suggest clarity of thinking about the problem.
I think that Holger Merlitz has the same conclusions, but I don't have the German language book. I wish it was available in English.
I have been interested in the matter of 3D and distance for many years.

There are many pitfalls, for instance looking at a blue post not very far away against an exactly the same shade of blue background, it is likely one just sees a blue vista with no post.
This is where camouflage comes in. Animals and birds use it as well as magicians and the military.

I once saw a Woodcock hiding in our garden. I have never seen another one. It was completely still and almost completely hidden in plain sight.

elkcub · May 13, 2018

David,

I sent you an email with the article attached. :smoke:

Ed

typo · May 13, 2018

elkcub said:
David,

I sent you an email with the article attached. :smoke:

Ed

Ed,

I think there has been a bit of a mix-up. It's a topic we have discussed before, but I haven't contributed to this particular thread. I think you sent it to the wrong person.

For those who haven't seen it there is a straight forward Wikipedia artical on stereoscopic acuity. The 0.5 arcsecond value seems to hold up for me in my back yard testing. The chapter from Gibson's work that Ed posted illustrates there are other factors at play as well.

https://en.m.wikipedia.org/wiki/Stereoscopic_acuity

David

Binastro · May 13, 2018

Hi Ed,
Thanks. I haven't yet read the article but will comment on your post 24.

I think you misunderstood me by highlighting the comments on using pupil size and telescope aperture.

I am not talking of the perception of space, distance and size, which I think Gibson discusses.
I am talking about real rangefinding 3D.
This uses the two optics of a rangefinder, binocular or our eyes to measure the two images produced from slightly different angles and comparing the margins of a high contrast well seen target.

If one starts off at medium distance, say 30 yards, one sees with unaided eyes the edge of a building against unmoving cloud or the sky and some, but not all people, will see two differing images, which are produced because of the small angle difference in the two views.
This is the principle of many optical rangefinders.

One then lengthens the distance until one is unsure whether there are two different images or not.
One then comes back nearer until one is sure there are two differing images because of the angle difference.
With experience one gets a good estimation that, say the rangefinding abilities of the two unaided, if necessary eyeglass corrected, eyes produces real rangefinding 3D between say 150 yards and 200 yards.
The measurement of this distance cannot be estimated by the eyes. This distance has to be independently measured by tape, laser rangefinder or accurate high scale map.

With magnification, the two differences in the image are magnified by say 8 times in a 8x tripod mounted binocular.
so one can repeat this process but at longer distances in very clear, very steady air.

The use of one eye's pupil at say 2 feet distance is using the same principles but using the rays coming from the left and right side of the pupil. One can lengthen the distance and again come up with a range of distance where one is sure or not sure the angle effect is seen.

Perception is quite different and a real can of worms.
There are so many false clues and interpretations of what one sees that many red herrings are produced.

This is instanced by the UFO believing population.
I had a fruitless long exchange with an intelligent guy who saw a 20mile wide, I think, space ship hovering in the sky to the north of Manchester. He wouldn't budge. He knew it was 20 miles wide and contained aliens.
I then went to a site that he suggested as proof and was staggered by the vast numbers of firm believers in alien UFOs that are seen in large quantities daily.
These reports are listed and run into maybe thousands of alien UFO sightings.

People's beliefs are beliefs, and one doesn't query them as it is pointless.
I am also struck by the basic non understanding of commercial pilots of angles and distance in the air around them. One just cannot judge the size of an object in space with no possibility of knowing whether it is one mile or ten miles away.

As far as I can judge it is only observational astronomers who understand these angles and observations. There may be others, but I haven't really met them.

So in summary, I am not talking about visual perception, but rangefinding 3D, based on the small angle differences in two images being compared and recognised.

elkcub · May 14, 2018

typo said:
Ed,

I think there has been a bit of a mix-up. It's a topic we have discussed before, but I haven't contributed to this particular thread. I think you sent it to the wrong person.

For those who haven't seen it there is a straight forward Wikipedia artical on stereoscopic acuity. The 0.5 arcsecond value seems to hold up for me in my back yard testing. The chapter from Gibson's work that Ed posted illustrates there are other factors at play as well.

https://en.m.wikipedia.org/wiki/Stereoscopic_acuity

David

So sorry guys, my bad. :-C

I thought my email was sent to David Binastro, but instead, it went to David Typo. (And as it turns out I don't really have David's email anyway.)

The title of this thread is: "Which roofs are better for their 3D view?," which I thought pertained to tridimensional perception, not rangefinding. If the OP meant: "Which roofs are better for rangefinding?" then I'm guilty of introducing unnecessary confusion. Framing the question makes a big difference, as Bill said earlier.

Assuming the OP was asking about perception, however, i.e., the experience of distance in extended space, then J. J. Gibson's theory of stimulus gradients is the place to start. Although I'm not going to proselytize his work, one really should read the article before drawing conclusions. On page 417 he discusses "The Diminishing of Binocular Cues with Distance," and correctly concludes that "... the actual range of stereoscopic vision, ... is not known."

Stereoscopic acuity, like all the other visual 'acuities,' is a man-made performance measure, which in and of itself does not explain tridimensional perception any more than do grating acuity or vernier acuity. Arguing that tridimensional perception requires binocular vision is tantamount to saying that a one-eyed person can't navigate effectively in a three-dimensional world, which Gibson shows is demonstrably incorrect.

I've attached a memoir written by Julian Hochberg about Gibson's work which is being valued increasingly in this age of AI and computer vision.

Ed

elkcub · May 14, 2018

Binastro said:
Hi Ed,
Thanks. I haven't yet read the article but will comment on your post 24.

I think you misunderstood me by highlighting the comments on using pupil size and telescope aperture.

I am not talking of the perception of space, distance and size, which I think Gibson discusses.
I am talking about real rangefinding 3D.
This uses the two optics of a rangefinder, binocular or our eyes to measure the two images produced from slightly different angles and comparing the margins of a high contrast well seen target.

If one starts off at medium distance, say 30 yards, one sees with unaided eyes the edge of a building against unmoving cloud or the sky and some, but not all people, will see two differing images, which are produced because of the small angle difference in the two views.
This is the principle of many optical rangefinders.

One then lengthens the distance until one is unsure whether there are two different images or not.
One then comes back nearer until one is sure there are two differing images because of the angle difference.
With experience one gets a good estimation that, say the rangefinding abilities of the two unaided, if necessary eyeglass corrected, eyes produces real rangefinding 3D between say 150 yards and 200 yards.
The measurement of this distance cannot be estimated by the eyes. This distance has to be independently measured by tape, laser rangefinder or accurate high scale map.

With magnification, the two differences in the image are magnified by say 8 times in a 8x tripod mounted binocular.
so one can repeat this process but at longer distances in very clear, very steady air.

The use of one eye's pupil at say 2 feet distance is using the same principles but using the rays coming from the left and right side of the pupil. One can lengthen the distance and again come up with a range of distance where one is sure or not sure the angle effect is seen.

Perception is quite different and a real can of worms.
There are so many false clues and interpretations of what one sees that many red herrings are produced.

...

David,

Many thanks for your discussion of 3D rangefinding, which as I recall you've mentioned several times before. When I get to the second highlighted paragraph, unfortunately, I can no longer follow. How does one eye distinguish between "...rays coming from the left and right side of the pupil" either with or without a telescope? In binocular vision, there is a left and a right image to be distinguished, but with one eye there is only a single image. Rays from the left and right sides of the pupil, and those in between, form the image. I'm missing something here.

Ed

PS. Viewing perception as a "can of worms" is a definitely a defeatist attitude.
PPS. Open the attachment to post #31 above and look at Gibson's portrait on the screen. Study it for a while with both eyes ... and then close one eye. Does the impression of depth increase, decrease, or stay the same?

Binastro · May 14, 2018

Thanks Ed,
I will try to read your references.

What bothers me about perception is highlighted by the excellent artist and seemingly intelligent guy who definitely saw a 20 mile wide spacecraft at night hovering in the sky to the north of Manchester and his numerous pals who see similar spacecraft daily.
I suggested eventually that he asks the captain of said craft to tether to the top of Blackpool tower, so I could visit. He could also make a tidy sum giving conducted tours of the vast spacecraft.

What are they seeing and how are they interpreting their observations?
I feel that perception sometimes results in delusion.

I have spent a lifetime observing the sky and only three times have I been left bewildered. In each case an explanation was found rather quickly.
Amongst my observational astronomer friends, who have great experience, they sometimes see unexplained events, sometimes resolved, sometimes not.

I agree with you that perception of stereo is difference from rangefinding stereo.
In rangefinding stereo I make sure that I don't move my head, and the background, say clouds is not moving. Also trees are not moving in the wind.
Surprisingly trees with uneven leaf outlines appear well in 3D.

With stereo perception, from what I gleaned, ones head and eyes are moving and information is obtained by walking to or away or sideways from the target.

I also see perception stereo, but I am wary of this.

The wonderful stereo view of the full moon in the Soviet 20x60 was beautiful, but I know that it is perception, however real it looks.

Binastro · May 14, 2018

With one eyed stereo.

I am looking at a tumbler of water on the table at 2ft and it has depth.
I think this is the result of the image on the retina, in which the edge rays are at slightly different angles and the brain recognises this.
It could also be a result of different depth focus, but I am not sure.

I do not think that this is perception, but again I am not sure.
I may ask my optics lecturer friend.

elkcub · May 14, 2018

Binastro said:
With one eyed stereo.

I am looking at a tumbler of water on the table at 2ft and it has depth.
I think this is the result of the image on the retina, in which the edge rays are at slightly different angles and the brain recognises this.[b/]
It could also be a result of different depth focus, but I am not sure.

I do not think that this is perception, but again I am not sure.
I may ask my optics lecturer friend.

Gibson would explain your one-eyed perception of depth while statically viewing a real 3D object (e.g., water tumbler) as due to a collection of visual gradients within the retinal image. The gradients result from lighting, perspective, depth-of-focus, and others.

In the picture of Gibson (below) do you experience greater depth while viewing one-eyed or two-eyed? Notice how much better the face stands out from the background using only one eye. For me, the effect is so dramatic that the face appears to come to life when viewed monocularly. The same results when the picture is viewed at distance with binoculars, i.e., monocularly vs. binocularly.

But getting back to binocular viewing at a distance, it should be understood that only for a relatively small number of image points that lie exactly on the visual horopter do we not see double images. A large aspect of binocular vision, therefore, must also involve the suppression from consciousness of position gradients, or we would be in a constant diplopic state.

Regards,
Ed

Binastro · May 14, 2018

The photo stands out better with one eye, but not the other due to the need for new glasses.

It is flatter using both eyes.

b3rd · May 15, 2018

But.. the object is flat. It is a photo (or a screen emitting light). So doesn't that make sense? And with a single eye, it would make sense that other cues for depth might be used for perception?

WJC · May 15, 2018

elkcub said:
So sorry guys, my bad. :-C

I thought my email was sent to David Binastro, but instead, it went to David Typo. (And as it turns out I don't really have David's email anyway.)

The title of this thread is: "Which roofs are better for their 3D view?," which I thought pertained to tridimensional perception, not rangefinding. If the OP meant: "Which roofs are better for rangefinding?" then I'm guilty of introducing unnecessary confusion. Framing the question makes a big difference, as Bill said earlier.

Assuming the OP was asking about perception, however, i.e., the experience of distance in extended space, then J. J. Gibson's theory of stimulus gradients is the place to start. Although I'm not going to proselytize his work, one really should read the article before drawing conclusions. On page 417 he discusses "The Diminishing of Binocular Cues with Distance," and correctly concludes that "... the actual range of stereoscopic vision, ... is not known."

Stereoscopic acuity, like all the other visual 'acuities,' is a man-made performance measure, which in and of itself does not explain tridimensional perception any more than do grating acuity or vernier acuity. Arguing that tridimensional perception requires binocular vision is tantamount to saying that a one-eyed person can't navigate effectively in a three-dimensional world, which Gibson shows is demonstrably incorrect.

I've attached a memoir written by Julian Hochberg about Gibson's work which is being valued increasingly in this age of AI and computer vision.

Ed

Let’s say, for the sake of argument, that depth perception can be obtained with ONE eye. I find it fascinating, especially if it has any basis in science. However, with someone having recently challenged the idea of objective spacing as the major cause of stereopsis, how is it possible—comparing an objective separation of 2.5 to 5 inches in binoculars, adding such little data, to tiny pencils coming from opposite sides of the same eye as even anything worth discussing? I know I’m just a screw turner. However, are we not stacking BBs on the topic as never before? I am not trying to cause trouble; I truly want to know.

Please correct me if I am wrong, but with empirical data from an authoritative source. If the point really matters, and has already been addressed, please point my little pointy head at it once again. :cat:

Bill

Binastro · May 15, 2018

If one looks at an object 2 feet (600mm) away with one eye in room lighting with a 4mm eye pupil, the angle difference between the left and right side of the pupil is 23 arcminutes (1 part in 150). This is a significant angle.

It is the same angle as using two eyes 2.5 inches (63mm) apart at 31 feet or about 10 yards.
No one doubts that 3D can be seen at 10 yards using two eyes, by persons who see 3D.

The question is, can the eye/brain use this information to see genuine rangefinding 3D close up with one eye.
I will try to ask a senior optics lecturer whether this happens or not.

With perception, this is using extra clues, and can lead to all sorts of problems.

The photo above of the face is an optical illusion of which there are many, deliberately man made or in nature.
As an observational astronomer when I make an observation, I have to ask myself, What am I actually seeing? This is the difference between potential discovery observations or false observations.

Lowell was convinced there were Canals on Mars made by the local inhabitants.
Ufologists believe they are seeing alien spacecraft.

Seeing is believing and believing is seeing are pitfalls.

As the comet section director of a national organisation I was presented by two amateurs with a new comet discovery. They were totally convinced.
I got world class astronomers to get up in the middle of the night to open their observatories and photograph the comet. It didn't exist and the two amateurs didn't have the grace to apologise or thank anyone, and my credibilty suffered.

On the other hand my friend discovered a Supernova in an external galaxy and was too cautious. Someone else got credit for a later observation.

Observations need common sense and logic as well as seeing.

P.S.
It might be better to consider a 2.5mm eye pupil in good light where the eye has its best acuity.
This is 1 part in 240 for an object at 2 feet.
This gives an angle of 14 arcminutes, which is the same angle as two eyes 2.5 inches apart at 17 yards.

elkcub · May 15, 2018

The basic problem with what you're saying, Binastro, is that there is no known mechanism, anatomical, optical, or neurological, for the brain to distinguish light rays from the left and right sides of a single eye's entry pupil.

Ed

Which roofs are better for their 3D view? (1 Viewer)

Well-known member

Well-known member

Well-known member

Silicon Valley, California

Attachments

Eurasian Goldfinch

Well-known member

Well-known member

Silicon Valley, California

Well-known member

Well-known member

Silicon Valley, California

Attachments

Silicon Valley, California

Well-known member

Well-known member

Silicon Valley, California

Attachments

Well-known member

Reg1stered User

Well-known member

Well-known member

Silicon Valley, California

Similar threads

Users who are viewing this thread