Good & bad samples, cherrypicking, etc (1 Viewer)

[SeldomPerched]

A general set of questions and so no special reason to put them in the Zeiss forum except it was a Zeiss binocular I was just reading about: the 54 HTs, which received a lukewarm reception compared with its 56mm FL predecessors some years ago. On the other hand that has reminded me that I have a Leica binocular that might fit the question also.

Often mention is made of sample variation in binoculars and reviewers and users will talk of cherry samples and lemon samples - no idea why lemon is used to mean bad but there you go; that's another question.

We know that tolerances exist because absolutely identical assembly is presumably impossible or not cost-effective. My questions are:

-with so-called lemon samples, what are the aspects that bring about the undesirable reduction in quality? Is it faulty assembly, is it defects in the dimensions of the glass surfaces, could it be poorly ground glass {I would assume that glass production even of aspherical surfaces is computer-controlled and/or blank press technology (is this right?)}, is it simply sometimes insufficently clean glass surfaces, or are packaging and transport responsible for a loss of accurate adjustment in transport?

-can a poor sample always be brought up to standard by return to the factory for adjustment and reassembly?

-is a weaker sample usually capable of being 'cherrified' by recollimation?

Of course I am not talking here of generally inferior designs but of designs that are represented by excellent samples as well as some not so good.

Thank you,

Tom

 

[b-lilja]

I look forward to responses on this. I wonder if variability is mostly related to lesser bins and scopes.

 

[etudiant]

Not an expert, but binoculars have multiple lenses lined up to perform, so if any of them are slightly tilted or ground off center or the glass composition is a bit off, there will be visible consequences. Bill Cook has often pointed out that the objective lenses are in fact aligned to an orientation, so just repositioning them will affect the view. The above suggests that the result is usually a compromise, juggling the parameters to maximize the performance. That does not bode well for making cherries out of lemons. Rather it may be analogous to the packaging of orange juice, lesser quality product is brought up to some minimum standard by adding better.
The hope is that those minimum standards are higher for the alpha products.

PS The lens teardown discussions posted by Roger Cicala at Lens Rentals (https://www.lensrentals.com/blog/ ) are super informative and pretty matter of fact about all the issues involved in producing a reliable quality optic.

 

[Troubador]

'Lemons' is a term used on Birdforum to describe an instrument that performs worse than another example of the same model. Likewise a cherry is one that performs better than another example. But it seems to me that for all we know, the lemon may meet the manufacturer's specifications and that the cherry should perhaps be regarded as an unexpected bonus. Usually these assessments do not involve using nature observations so it is not known what effect the lemon's shortcomings have on identifying birds at different distances or on observing behaviour for extended periods. In this context the term 'lemon' may be quite inappropriate and misleading if the unit meets the manufacturer's specifications and it performs satisfactorily in the field. Of course if when using the lemon birds cannot be identified at normal distances or eye fatigue prevents observing behaviour for extended periods one would be justified in questioning whether the manufacturer's specifications were adequate or competitive.

As regards HT54 I note that this model appears to have found a secure place in the market and is still offered by Zeiss, whereas HT42 has been discontinued.

Lee

 

[jring]

Often mention is made of sample variation in binoculars and reviewers and users will talk of cherry samples and lemon samples - no idea why lemon is used to mean bad but there you go; that's another question.

We know that tolerances exist because absolutely identical assembly is presumably impossible or not cost-effective. My questions are:

-with so-called lemon samples, what are the aspects that bring about the undesirable reduction in quality? Is it faulty assembly, is it defects in the dimensions of the glass surfaces, could it be poorly ground glass {I would assume that glass production even of aspherical surfaces is computer-controlled and/or blank press technology (is this right?)}, is it simply sometimes insufficently clean glass surfaces, or are packaging and transport responsible for a loss of accurate adjustment in transport?

-can a poor sample always be brought up to standard by return to the factory for adjustment and reassembly?

-is a weaker sample usually capable of being 'cherrified' by recollimation?

Hi,

first of all - in german we value the lemon and thus call a bad example a cucumber (Gurke). There is no fruit or veggie term for a very good example though...

As for why optics does not perform as designed, there is unfortunately many different aberrations and some are often due to incorrect assembly or collimation and thus can be remedied by readjusting while others are usually due to problems during lens grinding and thus can only be fixed by changing the faulty elements.

Just for some of the main aberrations - there is many more.

- Spherical aberration - can be both, bad grind or incorrect lens spacing.

- Astigmatism - some amount is usually tolerated in the individual elements and then removed by pairing and rotating the elements of an objective so it cancels out. This means that the individual elements must not be rotated against each other - there should be a marking of the correct position on the lens edges.
Also overly tight fits of lenses in lens cells often lead to astigmatism which can be removed by loosening that retaining ring a bit. For astronomers a slight rattle in an objective lens is not a fault!

- Coma - in refractors usually an assembly problem - lenses either not mounted with a common optical axis or tilted. Can be fixed by correct reassembly, but in triplet or multiplet objectives getting things correct without ways to measure the optical performance is next to impossible.

- Zones - always a problem in grinding - cannot be adjusted.

Joachim

 

[henry link]

This is way too big a subject to handle briefly, but here we go.

The first thing that needs to be understood is that the two little telescopes that form a binocular are not even designed to be high quality low aberration devices. This is just as true whether the binocular cost $300 or $3000. The designers take advantage of low magnification and the reduced aperture of the eye in daylight to essentially get away with optical murder.

When subjected to a high magnification star test nearly all the binoculars I've tested in the last 30 years, cheap or expensive, look pretty terrible, with gobs of chromatic and spherical aberrations and often with a few other defects like astigmatism, coma, pinching or poorly made roof prisms thrown in for good measure. So why to people wax so poetical about the beauty of the images?

Firstly, it's because the magnification of small birding binoculars is so low that only bad defects would be visible when hand holding and, secondly, for daylight birding purposes, the binoculars are effectively stopped down to some smaller aperture than the full objective lens aperture. The smaller that effective aperture is compared to the full aperture the better corrected most of the aberrations become. A poor 8x42 may become a decent or even good 8x20 in bright daylight. Typically binoculars are designed to be no better than "just good enough" once all the advantages of low magnification and reduced aperture are factored in.

Sometimes, even with all the design sloppiness that binoculars allow the "just good enough" standard isn't quite met and the image actually looks inferior, at least to a close examination. That was the case with the Zeiss 8x54 I reviewed a few years ago.

All of the above applies to axial aberrations and defects, not off-axis aberrations or collimation between the two sides. Collimation can be fine even when the individual telescopes are optically poor or vice versa. Some axial defects are likely to be design limitations, like chromatic and spherical aberrations and will look essentially identical in both sides when star tested. Other defects may be sample dependent and may not appear in both sides.

Henry

 

[etudiant]

This is way too big a subject to handle briefly, but here we go.

The first thing that needs to be understood is that the two little telescopes that form a binocular are not even designed to be high quality low aberration devices. This is just as true whether the binocular cost $300 or $3000. The designers take advantage of low magnification and the reduced aperture of the eye in daylight to essentially get away with optical murder.

When subjected to a high magnification star test nearly all the binoculars I've tested in the last 30 years, cheap or expensive, look pretty terrible, with gobs of chromatic and spherical aberrations and often with a few other defects like astigmatism, coma, pinching or poorly made roof prisms thrown in for good measure. So why to people wax so poetical about the beauty of the images?

Firstly, it's because the magnification of small birding binoculars is so low that only bad defects would be visible when hand holding and, secondly, for daylight birding purposes, the binoculars are effectively stopped down to some smaller aperture than the full objective lens aperture. The smaller that effective aperture is compared to the full aperture the better corrected most of the aberrations become. A poor 8x42 may become a decent or even good 8x20 in bright daylight. Typically binoculars are designed to be no better than "just good enough" once all the advantages of low magnification and reduced aperture are factored in.

Sometimes, even with all the design sloppiness that binoculars allow the "just good enough" standard isn't quite met and the image actually looks inferior, at least to a close examination. That was the case with the Zeiss 8x54 I reviewed a few years ago.

All of the above applies to axial aberrations and defects, not off-axis aberrations or collimation between the two sides. Collimation can be fine even when the individual telescopes are optically poor or vice versa. Some axial defects are likely to be design limitations, like chromatic and spherical aberrations and will look essentially identical in both sides when star tested. Other defects may be sample dependent and may not appear in both sides.

Henry

Just asking, if anyone actually designed/built a binocular to high quality optical standards, would the consumer notice?
I assume the Nikon WX is the pinnacle of current binocular performance.
How would it measure up to the standards of say a Questar, my 'unobtainium' optical instrument from when I got started?

 

[[email protected]]

Henry. Isn't a $3000 binocular designed to higher optical standards than a $300 binocular even though it may not be as precision as a Questar?

 

[henry link]

Henry. Isn't a $3000 binocular designed to higher optical standards than a $300 binocular even though it may not be as precision as a Questar?

Doesn't work that way. The best corrected binocular I've ever tested for axial aberrations at full aperture is a Fujunon 8x30 FMT-SX. I paid $310 for it about 20 years ago. It easily beats the Sawro 8x42 NL and my long time favorite Zeiss 8x56 at their full apertures. What saves the day for the NL and FL as daytime birding binoculars is that their aberrations clean up well once their apertures are reduced to 30mm or less by the eye's pupil in bright light.

 

[henry link]

Just asking, if anyone actually designed/built a binocular to high quality optical standards, would the consumer notice?
I assume the Nikon WX is the pinnacle of current binocular performance.
How would it measure up to the standards of say a Questar, my 'unobtainium' optical instrument from when I got started?

I can tell the difference when I compare a sensibly perfect 8.25x50 telescope (Takahashi FC-50 scope with Brandon 48mm eyepiece) to any 8x binocular, but I think I've been at this too long to qualify as a good stand-in for "the consumer".

I haven't seen any high magnification star tests of the Nikon WXs. The axial aberrations may not be that good.

I think any really good modern APO is better than a Quester, since the central obstruction cuts the Strehl ratio by 20% before the optical aberrations are even considered.

 

[SeldomPerched]

Thank you very much, Lee, Henry, Joachim and everybody else; I wasn't expecting so many answers so soon to these questions. Much appreciated.

Depressing but informative. But I'm also reassured because this all explains some anomalies I've encountered from viewing in different conditions and also it's reminded me not to sell my 8x56 T*FL.

Lee's point struck a chord. When out and observing nature all seems fine; it's 'assessment' or fiddling about that leads me to disappointment.

Another topic but it strikes me that high end photographic lenses don't suffer nearly as much from some of these problems?... or do they?

Tom

 

[Troubador]

Thank you very much, Lee, Henry, Joachim and everybody else; I wasn't expecting so many answers so soon to these questions. Much appreciated.

Depressing but informative. But I'm also reassured because this all explains some anomalies I've encountered from viewing in different conditions and also it's reminded me not to sell my 8x56 T*FL.

Lee's point struck a chord. When out and observing nature all seems fine; it's 'assessment' or fiddling about that leads me to disappointment.

Another topic but it strikes me that high end photographic lenses don't suffer nearly as much from some of these problems?... or do they?

Tom

Photo lenses are a quite different category of instrument Tom, because they have to deliver an image to a recording medium, film or sensor, not the human eye. Plus, with professional lenses, there is an expectation that a tripod will be available, and so saving weight is not such a priority. This allows additional lenses to be added to the optics as well as lens groups that move, and of course photo lenses work 'alone', they don't have another optical tube alongside with which it must focus in unison.

Binos are designed to work with the human eye/brain and I find that appropriate rather than a point of weakness. And as long as they are good enough to provide useful and enjoyable nature observation experiences then IMHO they are achieving their goals.

Lee

 

[Binastro]

My low price Skywatcher 90 Maksutov is the equal of my Questars optically.
It is only mechanically that the Questars are better.

Horace Dall did not rate Questars very highly.

My 1950s 100mm Soviet Maksutov is way better than the Questars.

Very few camera lenses are of astronomical quality.
The nearest I found in thousands may be the c.1960 Tewe 600mm f/5 that was of astronomical standard. It is a 4 element Petzval. Very good views of Saturn at 150x. 120mm aperture.
It was not of fine or very fine astro quality

The Vivitar 600mm f/8 solid Cat is another. Beautiful at 180x.

A Den Oude Delft military Maksutov lens about 140mm aperture was of fine astro standard. Beautiful high powers views of Saturn.

The Ross 50 inch f/8 telephoto lens is of near astro quality

The 115mm Swarovski module seems to be getting into camera lens territory. Petzval, Aviar, Tessar, Heliar, double Gauss. I suppose it is actually a high end triplet.

The 22x60mm Takahashi binocular is probably of astro quality.
The Zeiss 20x60S binocular is of near astro quality with a very curved field.

I would not discount optics with central obstructions.
Horace Dall's 8 inch Maksutov would outperform most refractors in actual use,
My friend's 14 inch Celestron SCT is way better than his 7 inch Astrophysics refractor in actual use.
Others high quality well collimated Celestron 14s are also the choice, not large refractors, Apo or otherwise.

Although refractor optics can be good for military use, such as the Baker 40 inch f/5 and the 144 inch f/8, it is obstructed optics that are mostly used. Such as the 70 inch aperture Big Bird etc.

The Hubble telescope is not a refractror.

Regards,
B.

 

[doug el10x32]

Hi,

If there are cherries and lemons, do manufacturers compare barrel lens sets and assemble lemons or cherry sets in pairs deliberately or could you get a pair with a lemon say on the left side and a cherry on right. How noticeable that would it be if we are already able to determine whether a pair of binoculars was a lemon or a cherry ?

Doug.....

 

[Binastro]

In my experience one side is usually worse than the other.

Also the actual collimation process probably automatically reduces the quality of one side.

I don't think that manufacturers have the time or inclination to select.

However, it may be that the IQ range of Zeiss Jena did choose two good sides.

There were about a dozen Soviet 20x60 binoculars for the U.K. where the maker deliberately selected the best optics they could make. I have one of these binoculars that were sold at very ordinary prices to selected keen astronomers.

One can be sure that Putin gets the best or else.

Further to my post above, even in small optics, say 6 inch aperture, the military seems to prefer obstructed mirror optics to Apo refractors. Not sure why.

B.

 

[etudiant]

I think the picture is beginning to become clear, that optics has so many variables that no manufacturer can afford to optimize them all.
So buyers choose their priorities, obviously robustness for the military, resolution for the astronomers, all others somewhere in between.

Horace Dall sounds like a person I would have been privileged to meet.
Too bad he never chose to found a business that would carry on his genius.

Henry Link sagely suggests that we should be careful about translating what we see into 'optical quality'. I remember being blown away by some fellow birder's Questar at the Forsythe NWR, looking at terns way out.
Strehl ratio was not a term I was familiar with at the time and I wonder whether it fully captures the requirements of the spotting scope reality, multiple targets, dubious light and questionable contrast.

 

[Patudo]

I can't help but feel that criticising a binocular for not performing well when boosted to twice or thrice its magnification seems a bit like criticising a Prius for not driving well at 200mph - that's just not what the vehicle in question was ever designed to do?

That Takahashi FC-50 certainly sounds like it's brilliant optically, and portable to boot. Has anyone ever tried one for birding?

Bill Cook has often pointed out that the objective lenses are in fact aligned to an orientation, so just repositioning them will affect the view.

He's also stated that it's impractical and foolish to compare two binoculars of similar design and performance, and indeed that sample variation is so great that comparing two binoculars of the same model - produced on the same day - is impractical and foolish as well... (see link).

Which makes about 99% of the discussions here impractical and foolish... But then, what else would we do with our spare time? 3

 

[Troubador]

I can't help but feel that criticising a binocular for not performing well when boosted to twice or thrice its magnification seems a bit like criticising a Prius for not driving well at 200mph - that's just not what the vehicle in question was ever designed to do?

:t::t:

Lee

 

[Chosun Juan]

A general set of questions and so no special reason to put them in the Zeiss forum except it was a Zeiss binocular I was just reading about: the 54 HTs, which received a lukewarm reception compared with its 56mm FL predecessors some years ago. On the other hand that has reminded me that I have a Leica binocular that might fit the question also.

Often mention is made of sample variation in binoculars and reviewers and users will talk of cherry samples and lemon samples - no idea why lemon is used to mean bad but there you go; that's another question.

We know that tolerances exist because absolutely identical assembly is presumably impossible or not cost-effective. My questions are:

-with so-called lemon samples, what are the aspects that bring about the undesirable reduction in quality? Is it faulty assembly, is it defects in the dimensions of the glass surfaces, could it be poorly ground glass {I would assume that glass production even of aspherical surfaces is computer-controlled and/or blank press technology (is this right?)}, is it simply sometimes insufficently clean glass surfaces, or are packaging and transport responsible for a loss of accurate adjustment in transport?

-can a poor sample always be brought up to standard by return to the factory for adjustment and reassembly?

-is a weaker sample usually capable of being 'cherrified' by recollimation?

Of course I am not talking here of generally inferior designs but of designs that are represented by excellent samples as well as some not so good.

Thank you,

Tom

I think the crux of the matter has been pretty well covered.

Just in terms of manufacturing - there are that many tolerances stacked on top of one another, that if the units test good enough to get out the door, then it is good enough to meet the range of the design spec. (I think the level of the standards could be improved for higher quality - but that's another matter).

Some units may have less marginal components or alignments - ie. smack bang in the middle of the design point. As such when they are shipped out the door, they meet the spec just the same as their lesser brethren, but will have more margin of error to accommodate different folks viewing characteristics and conditions.

If your face/eyes/brain is a bit off to the sides of the normal distribution curve of such parameters for the rest of the folks on the planet, then you may find things a bit more critical than others. A 'cherry' unit may be easier to view through than a 51%er ......

As for sending them back to be rejigged - I think improvements are unlikely - those particular components are likely to be more or less in the best positions and alignments possible. Theoretically it may be possible to change some more marginal components and then rejigger for a better result, but this would never happen as the sent back unit would be assessed to be 'in spec'. It would be very rare I would think for a manufactured unit to get out the door that was 'out of spec'. Possibly a unit may be damaged in the field though that sends it (or parts of it) out of spec.







Chosun :gh:

 

[Chosun Juan]

Hi,

If there are cherries and lemons, do manufacturers compare barrel lens sets and assemble lemons or cherry sets in pairs deliberately or could you get a pair with a lemon say on the left side and a cherry on right. How noticeable that would it be if we are already able to determine whether a pair of binoculars was a lemon or a cherry ?

Doug.....

Without knowing the inner machinations of various manufacturers I would say mostly no. Who would want to buy a perfectly matched lemon ? :-O

However, the question is interesting - do the cherry components get cherry assembled into cherry units ? Not that you are likely to see out in the wild .......

I will bet my hat though, that the odd "propeller head" special gets made (I know I have seen them in other industries) - ie using the top 90th percentile tested manufactured components and assembling them bang on in the centre of the design spec for a personal back room cherry unit :t:






Chosun :gh: