Televid 77: Actual magnification? (1 Viewer)

[Etiennef]

First of all, some background:
Some time ago i realized that the actual specified magnification on all of my binoculars actually are somewhat incorrect. In many cases the "real" magnification is actually outside what you would consider acceptable by rounding.

For example, all of my German binoculars (all with 8x specification) have had about 8,5 to almost 9 times actual magnification. In contrast, most Chinese optics I've had have been on the low side of the specification.

Nevertheless, a few fractions of an 'X' is perhaps not something that one cares too much about.


Back to the topic of this post;
Earlier today I was looking at camera-alternatives to my current digiscoping setup, which happens to be a APO Televid 77 paired with a Sony QX100.

Now you would think that simply multiplying the (35mm equivalent) focal length of your camera with the magnification of the spotting scope would provide the "digiscoping"-focal length.

I then compared these values to some pictures taken with a superzoom-bridge camera, and the values did simply just not add up...

After some frustration I remembered what I had concluded about binoculars, and thought that the deviation is specifications perhaps were proportional to the magnification rather than ~constant?

I then checked the actual magnification on my scope + combinations of all eyepieces and came to the following "real" magnifications:

32x eyepiece -> 35,7x actual

40x eyepiece -> 44,1x actual

20x-60x eyepiece, centered exactly over the "0" at the markings:
20x -> 21,8x actual
30x -> 32,6x actual
40x -> 43,7x actual
50x -> 54,9x actual
60x -> 67,2x actual
>60x -> 68,0x actual magnification (the zoom ring can be turned marginally past the center "60x"-marking on my eyepiece, about 2mm).

(Note: The above was done by photographing an object through the scope, then photographing the same object (from the same distance) without the scope, and then comparing the size of the object in the two pictures. There can be some minor errors due to slight changes in focal length during focusing and human error, but the effect of this should be absolutely minimal.)

Notice that all values are ~10% above the specified magnification. Taking this into account, the focal lengths in my previous calculations made perfect sense.

Has anyone else noticed the same thing on your spotting scope? I'm a bit interested to know if it's just this model of spotting scope that has such large deviations, or if you always can expect deviations like this from the specifications?

It actually makes quite a difference if you are considering upgrading your equipment...

 

[Hermann]

First of all, some background:
Some time ago i realized that the actual specified magnification on all of my binoculars actually are somewhat incorrect. In many cases the "real" magnification is actually outside what you would consider acceptable by rounding.

For example, all of my German binoculars (all with 8x specification) have had about 8,5 to almost 9 times actual magnification. In contrast, most Chinese optics I've had have been on the low side of the specification.

No idea about Chinese binoculars, but this doesn't agree at all with virtually all the published tests of German and Austrian binoculars I've seen. These tests were usually conducted in certified test laboratories or the laboratories of the big manufacturers.

Are you sure your way of determining the magnification of your binoculars isn't flawed?

Hermann

 

[Etiennef]

No idea about Chinese binoculars, but this doesn't agree at all with virtually all the published tests of German and Austrian binoculars I've seen. These tests were usually conducted in certified test laboratories or the laboratories of the big manufacturers.

Are you sure your way of determining the magnification of your binoculars isn't flawed?

Hermann

Well, that could of course be the case. However, if it makes any difference Iv'e only tried it on older pocket binoculars and monoculars (that's been "my thing"), and none of them have been from the side of the millennium. But when I think about it I have a pair of Docter Asph. 8x32's (from the 1990's still, but they are bigger at least) that I actually have not tested, but they might perhaps give other results, so I'll give those a check soon.


But sure, the method could be flawed. But on the other hand I don't really see why it should not work? I mean, the purpose of such an optical device would be to magnify an object X times, so I don't really see why the method would not work.
Also, It explains perfectly why the focal length numbers don't add up either.

But I'm sure there are plenty of experts on the subject here that can correct me if i'm wrong in the methodology.

On a side note: I have tried the same with a few riflescopes, and those are usually fairly accurate to the specs.

 

[Etiennef]

OK, the Docters 8x32's turned out to be around 8,6x according to the same method. And I tried it with both my phone camera and my RX100.
I'm thinking that perhaps the length of the device in relation to the distance to the reference object could have some effect. I'll see if I have some time tomorrow to do some tests over very long distances to minimize the effect of this (if it has any effect at all...).

I'll also try to see see if could be an eye-relief issue or something like that... I know that for example low magnification pistol scopes rated at or around 2x (even the good brands) always result in more or less no magnification at the specified eye relief - which is about half a meter for pistol scopes. It's basically like looking though a red dot sight; you can have both eyes open... So it may be something like that that's playing tricks on me..?

But how do you other test magnification? Just trust that the manufacturers actually always say the correct thing? FOV and other parameters are frequently quite off or mismatch, so why not the magnification?

-

Additionally, even if the method for calculating is flawed, something is making the pictures taken with a camera "bigger" than they should be. So even if the method does not work for calculation "human eye"-magnification it would be very interesting to know what actually causes the extra magnification in a camera. So regardless, it's very interesting from a digiscoping perspective.

 

[JerryLogan]

...

(Note: The above was done by photographing an object through the scope, then photographing the same object (from the same distance) without the scope, and then comparing the size of the object in the two pictures. There can be some minor errors due to slight changes in focal length during focusing and human error, but the effect of this should be absolutely minimal.)
...

Using this methodology you cannot be sure if it is a scope or a camera error. So, take a photo of an object of known size (like a meter stick for simplicity of calculation) and see if your camera performs to specifications.

Best,
Jerry

 

[Etiennef]

Using this methodology you cannot be sure if it is a scope or a camera error. So, take a photo of an object of known size (like a meter stick for simplicity of calculation) and see if your camera performs to specifications.

Best,
Jerry

Sorry, but I really don't see why knowing the objects dimensions would make any difference?

I mean, if I position the camera at a set distance from an object, take a picture as my reference, and after that I don't touch the settings of he camera at all.
I if then put 40x scope in front of the camera, should it then not magnify the object 40x?
That's at least what the manufacturers claim for human eyes...
The only explanation related to errors int he camera in that case would be if the camera for some reason zooms ~10% whenever i put a quality optic behind it, which sounds quite weird....

Nevertheless, if it makes any difference, here are the numbers when testing my Televid:

I used a label from a CD case as a reference, which is 117mm high. I did it indoors with uniform lighting.
The reference was ~9,5 meters from the camera. The camera was set at 28mm, (35mm eq.) but that really should of course not make a difference.

When photographing the reference without a scope in front of it it's more or less exactly 56 pixels high. I actually included the 2 outermost pixels that start to blur into the background. If i would take these into account also, the unexpected zoom would be slightly higher.

When I then stick the Televid with the 40x eyepiece in from of the camera (of course without changing any settings or zoom on the camera) the reference measures 2468 pixels high.

I of course repeated the same procedure for the other eyepieces. If you want those specific values i can provide them also, but I guess one example will suffice.

-

If someone can tell me were I'm going wrong I would be very glad...

Or in any case, I would appreciate if someone else could try the same thing with their setup and see if they get the same deviations? Perhaps it's just my stuff that's off?

 

[Etiennef]

...this doesn't agree at all with virtually all the published tests of German and Austrian binoculars I've seen. These tests were usually conducted in certified test laboratories or the laboratories of the big manufacturers.

By the way, can you link to a couple of these tests? It would be interesting to see how they calculated the magnification.

The only test-method in review I have come across is the one that allbinos.com uses. What they are doing is comparing the exit pupil size to that specified by the manufacturer. But that of course not necessarily accurate since this assumes that the full objective lens is actually put to use, and that's certainly not always the case. So as far as I see, that method rather measures what you could call magnification "effectiveness" rather than true magnification.

But regardless of this, there is examples on allbinos were Leica binoculars overshoot the specifications quite a bit. For example this one, which overshoot by close to 5% magnification:
http://www.allbinos.com/206-binoculars_review-Leica_Geovid_8x56_HD-M.html

So even by that test it's definitely not totally weird to get values that differ quite significantly from what the manufacturers say.

 

[JerryLogan]

Check this url, you will see that at a given distance an object of given size will project an image of known size. A 10x magnification means the image formed by the optical device should be 10x the size of the unmagnifyed image.

https://www.khanacademy.org/science...object-image-height-and-distance-relationship

In your procedure you have not calculated an expected image size. The scope magnification specifications refer to the relationship between expected image size and the image size produced at the eyepiece when the scope is at infinity focus.

You are basing your scope magnification numbers on the image size your camera produces but you don't know if your camera's magnification specifications are accurate so you cannot know if the problem lies with the camera or the scope.

Concrete example from long ago and far away in the time of film and military service. Photograph a 1 meter stick located 1000 meters away. Develop film. Unmagnifyed image should be 1/1000 or 1mm. If the stick measures 30mm on the negative then it's a 30x system!

In your procedure you're not measuring the stick on the negative, you're assuming it's 30mm because the camera manufacturers say it's 30x. Maybe the camera guys are wrong and the scope guys are right? You'll never know for sure until you measure.

Best,
Jerry

 

[JerryLogan]

...

Nevertheless, if it makes any difference, here are the numbers when testing my Televid:

...
The reference was ~9,5 meters from the camera.....

So, aside from general methodological considerations, this is way under infinity focus. Which means that the effective focal length of the lens system has increased. Which means that the magnification of objects will be greater than at infinity which is where the focal length/magnification specifications are set. So a totally expected property of optical systems in general, not a scope manufacturing problem.

See this thread,
http://www.cambridgeincolour.com/forums/thread2411.htm

Best,
Jerry

 

[Etiennef]

Ok, thanks for the input.

I'll simply continue experimenting when I have the possibility. I'll be away for a few days, so it will probably not be for a couple of days.

But I have to say that some of the things really make no sense to me...
I don't see why the size of the reference object would make any difference, since the magnification should be relative.

And that magnification would change over distance, and that the specification is only accurate at infinity, is also very strange if you ask me...
I guess I'll have to study the subject even more and see if I come to any clarifying insights...

Regardless of what is causing this deviation I'm NOT claiming that it's a manufacturing problem or "defect". It's rather a question about how to calculate the correct camera-comparable focal length.

 

[JerryLogan]

Ok, thanks for the input.

I'll simply continue experimenting when I have the possibility. I'll be away for a few days, so it will probably not be for a couple of days.

But I have to say that some of the things really make no sense to me...
I don't see why the size of the reference object would make any difference, since the magnification should be relative.

And that magnification would change over distance, and that the specification is only accurate at infinity, is also very strange if you ask me...
I guess I'll have to study the subject even more and see if I come to any clarifying insights...

Regardless of what is causing this deviation I'm NOT claiming that it's a manufacturing problem or "defect". It's rather a question about how to calculate the correct camera-comparable focal length.

Well OK, no worries! I've been writing late at night so I hope my tone has been all right. Geezers before coffee aren't polite or pretty but that's no excuse for being offensive. I sincerely hope I haven't offended.

So you're right, it's important to understand this stuff if we're going to be accurate about magnifications when we're digiscoping. Try to forget about the specifics of your experiments for a while. Taking it a step at a time and starting at the most basic, do you agree (and understand the reason why) that the magnification of an optical system is determined by the focal length of the lenses in the system? We can work this through together step by step if you like and I assure you that it does make sense. But you're definitely right, it's not necessarily common sense :>).

Best,
Jerry

 

[Etiennef]

Thanks,

Yes, I would indeed appreciate the support.

Step one would be to try this at longer distances. If what you say is correct, it would minimise the deviation in magnification.

Then there is of course a difference in establishing what is causing the issue, versus how to calculate it. It's really the latter that I'm most interested in.

Obviously, the frequently presented method for calculating the corresponding camera-comparable focal length is off, at least at practical digiscoping distances. And I would really like to know how to calculate the correct value.

However, like I said before, I can't resume practical testing for a couple of days.

 

[seaspirit]

To get into the lower digiscoping magnification range with a good quality camera lens you will have to invest serious cash. Mid digiscoping magnification range is realistically unobtainable with camera lenses.

A brand name 500 mm prime lens plus 2x multiplier will cost new ~10K.
The 3rd party x-600 zoom lenses are good on their own and offer value and performance for the money, but a multiplier will limit functions and IQ.

UH

 

[JerryLogan]

Thanks,

Yes, I would indeed appreciate the support.

Step one would be to try this at longer distances. If what you say is correct, it would minimise the deviation in magnification.

Then there is of course a difference in establishing what is causing the issue, versus how to calculate it. It's really the latter that I'm most interested in.

Obviously, the frequently presented method for calculating the corresponding camera-comparable focal length is off, at least at practical digiscoping distances. And I would really like to know how to calculate the correct value.

However, like I said before, I can't resume practical testing for a couple of days.

Great! I'll be around so whenever you have time :>)

As a matter of curiosity, why the emphasis on precise magnification numbers? We cared when I was armed forces because knowing magnifications made computing distances simpler. But 40x v 44x isn't really meaningful as long as the bird (or whatever) fits the frame right.

It's interesting that you're using a QX100. How are you linking to the scope eyepiece? I've an Olympus Air on pre order because I thought interchangeable lenses would allow maximum flexibility, but I love the whole QX/Air approach to digiscoping - the Air is just going to pop on the end of my Kowa eyepiece with a T2. No fuss and no camera!

Best,
Jerry

 

[Etiennef]

Great! I'll be around so whenever you have time :>)

And now I have had time to check. And you are definitely right about the magnification being closer to the specs at longer distances. This time I did the same test against a tall industrial chimney about 2,8 km away.

The numbers are still a bit on the high side, but we are talking about a few percent. If those few percent would go away at infinity I can't say for sure. But regardless of this, a few percent is within what's typical variation according to Allbinos.com.

-

However, the focal length of the camera seems to make no (or minimal) difference. I could not detect any significant difference between 25 and 35 mm. I tried with 2 cameras at 25, 28 and 35 mm, and the magnification was always around 2-3% over the specification in the above test.

I have not confirmed that the focal lengths of the cameras are perfectly according to the spec. But there is clearly an observable difference by eye, so I'm without doubt testing at different focal lengths at least. I would in any case assume that the magnification is (should be?) relative to the cameras focal length, or at least not cause more than negligible variation.

-

But what's interesting is that none of my cameras act this way. As far as I can see I don't overshoot the (relative) specifications at shorter distances compared to longer, and certainly not by anything close to 10%... But I guess this is because it's a different optical system to some extent? Or modern cameras perhaps have some sort of correction for this effect..??

Going back to the original issue; the conclusion so far is that the frequently presented method for calculating camera-equivalent focal length when digiscoping is not as simple as just multiplying the cameras 35mm equivalent with the magnification (unless digiscoping at very long distances).

As a matter of curiosity, why the emphasis on precise magnification numbers? We cared when I was armed forces because knowing magnifications made computing distances simpler. But 40x v 44x isn't really meaningful as long as the bird (or whatever) fits the frame right.

Of course, from a practical point of view, and only considering the equipment at hand, it's absolutely meaningless to have exact figures.

However, the reason I find it interesting is because I was considering moving on to a super-zoom camera instead of going with a digiscoping setup.
About 1-2 years ago, there was not much that could compete with the a digiscoping setup in say ~1000 Euro budget range.

For that price you can basically (at least in Europe) get a second hand Televid + eyepiece or similar quality "older"-generation APO/ED/HD scope, combined with a quality compact camera, for example a QX100.

At comparable image quality at those focal lengths, and at that prince range, there simply was not much.
However, now we have for example the Nikon P900. Which actually seems to be the first camera within that price range that can produce images that can put up a fight with my current setup. There are some other contenders also.

Now to the point; If I decide to upgrade I would like to be able to zoom in a bit more. Currently, I often find myself in situations were I'm just not close as I would like to be.
But without a way of accurately determining the camera-equivalent focal length of my setup, It's impossible to establish what my spec-requirements are.

It's interesting that you're using a QX100. How are you linking to the scope eyepiece? I've an Olympus Air on pre order because I thought interchangeable lenses would allow maximum flexibility, but I love the whole QX/Air approach to digiscoping - the Air is just going to pop on the end of my Kowa eyepiece with a T2. No fuss and no camera!

It's actually just connected via a rather simple "made in china"-adapter. I have a Leica original digiscoping adapter also, but first of all it does not accept the camera due to the large diameter lens barrel, and secondly it's not as stable, nor as easy to adjust at the Chinese one. o

The main point of this is that it allows me to have zero-contact with the scope+camera, which takes camera shade to the absolute minimum.
It also allows me to touch-focus on a tablet, which gives me quite good control.

Earlier I used a Panasonic LF1 (Wifi enabled), which also gave very good results. But optically, the QX100 is slightly better.

 

[JerryLogan]

Now to the point; If I decide to upgrade I would like to be able to zoom in a bit more. Currently, I often find myself in situations were I'm just not close as I would like to be.
But without a way of accurately determining the camera-equivalent focal length of my setup, It's impossible to establish what my spec-requirements are.

This makes perfect sense! I see what you're after - let me think a bit on procedures. Two things to keep in mind, a) magnification may well be hard to nail down from adverts and specs alone because the term is inconsistently used in scope v. camera industry b) concrete example: in the camera world a zoom that goes from 35mm to 350mm is a 10x lens because they are measuring the change in apparent image size at the focal plane of the camera; the same lens on a telescope has a maximum magnification of 7x because magnification is measured as a ratio of object size to image size.

The short version of the above is that a 50x super zoom is definitely not going to put up the same image as a 50x spotting scope so you have to get apples to apples before you can choose what you want. There's a rough rule of thumb if you can pry the 35mm maximum equivalent focal length out of the camera manufacturers but I've never really tested the rule - so more thought!

Best,
Jerry

 

[Etiennef]

Yeah, I'm fully aware that the "zoom" in a camera in nothing else than the ratio between max/min focal length.

A camera starting at 25mm and has a 12x zoom will take it to 300mm at max zoom.
A camera starting at 30 mm and has 10x zoom, will take it to the same value as above at max zoom.

So basically, just looking at the zoom-value is definitely a bad idea.

-

To solve the problem I guess that I could simply go back to the first post and look at that principle again.
Those numbers were based on ~10 meters, and I would say that that's quite a good "close range"-distance to work with (and closer ranges seem to provide the most "additional" magnification). I's a short, but yet reasonable distance from a practical perspective.

So based on that, I've got an "actual" magnification of ~44x on the scope, and usually I have the camera at ~30mm. This takes me to just above 1300mm ( 44 * 30 = 1300).
This means that if I want to match or exceed my current setup, I would have to look for cameras that can zoom to more than 1300mm, and preferably significantly more since I usually feel I would like to come a bit closer.

This is of course provided that there are no more "surprises" along the way, and that the calculation actually gives a fairly correct equivalent (which I'm pretty unsure about )...

-

Extra note: The manufacturers basically always provide 35mm equivalents for their cameras, so getting that figure is no problem.
Here is a list of the cameras that currently can produce (manufacturer specified) 1300mm equivalents: http://pricespy.co.uk/product.php?j=3028651,2802797,3004802,2487651,2501651
The only camera that significantly exceeds this value is the Nikon P900... The Sony DSC-H400 also comes close, but I seriously doubt that a product that cheap can deliver the image quality I'm looking for (perhaps in 10 years)...

 

[JerryLogan]

Yeah, I'm fully aware that the "zoom" in a camera in nothing else than the ratio between max/min focal length.

A camera starting at 25mm and has a 12x zoom will take it to 300mm at max zoom.
A camera starting at 30 mm and has 10x zoom, will take it to the same value as above at max zoom.

So basically, just looking at the zoom-value is definitely a bad idea.

-

To solve the problem I guess that I could simply go back to the first post and look at that principle again.
Those numbers were based on ~10 meters, and I would say that that's quite a good "close range"-distance to work with (and closer ranges seem to provide the most "additional" magnification). I's a short, but yet reasonable distance from a practical perspective.

So based on that, I've got an "actual" magnification of ~44x on the scope, and usually I have the camera at ~30mm. This takes me to just above 1300mm ( 44 * 30 = 1300).
This means that if I want to match or exceed my current setup, I would have to look for cameras that can zoom to more than 1300mm, and preferably significantly more since I usually feel I would like to come a bit closer.

This is of course provided that there are no more "surprises" along the way, and that the calculation actually gives a fairly correct equivalent (which I'm pretty unsure about )...

-

Extra note: The manufacturers basically always provide 35mm equivalents for their cameras, so getting that figure is no problem.
Here is a list of the cameras that currently can produce (manufacturer specified) 1300mm equivalents: http://pricespy.co.uk/product.php?j=3028651,2802797,3004802,2487651,2501651
The only camera that significantly exceeds this value is the Nikon P900... The Sony DSC-H400 also comes close, but I seriously doubt that a product that cheap can deliver the image quality I'm looking for (perhaps in 10 years)...

So I'm pretty sure the 1300mm focal length isn't right for what you want. More or less, 1300mm lens is 26x magnification - to get to where you are now with the Leica scope you'll need 2200mm just to get the scope part. Really, the P900 looks like the only camera with a shot of getting close to what you have now.

Best,
Jerry

 

[Etiennef]

More or less, 1300mm lens is 26x magnification - to get to where you are now with the Leica scope you'll need 2200mm...

OK, but how did you end up at 1300 = 26x more or less? This would imply that a typical camera would start at 50mm, which is quite a lot more than what I've come across on the cameras iv'e seen. Usually they are in the mid/high twenties...

(Not that I consider the zoom-ratio to be particularly important. The most important number is of course the maximum focal length).

 

[JerryLogan]

OK, but how did you end up at 1300 = 26x more or less? This would imply that a typical camera would start at 50mm, which is quite a lot more than what I've come across on the cameras iv'e seen. Usually they are in the mid/high twenties...

(Not that I consider the zoom-ratio to be particularly important. The most important number is of course the maximum focal length).

You can see formula and calculation here:
http://www.physicsclassroom.com/class/refrn/Lesson-5/The-Mathematics-of-Lenses

Sharp eye to notice the 50mm connection! But it's 50mm focal length, not 50x magnification or zoom range. The quick rule of thumb of which I wrote; A 50mm lens on 35mm camera, objects subtend the same arc as viewed by human so 50mm is 1x view. Notice this is different definition of magnification than ratio of object image to actual object. Also not precise re horizontal, vertical, stereo vision - but roughly correct.

Also, from personal experience, looking through a 600mm lens with 2x Tele extender (1200mm) is nowhere close to 40x spotting scope view so I'm sure the rough rule above is putting us close to working values.

So 1300 mm lens gives around 26x magnification, you're looking for at least 40-44x magnification.

Best,
Jerry