Evaluating Binoculars at High Magnification (2 Viewers)

[henry link]

Rather than continuing to highjack the Nikon thread with an off-topic discussion I ‘m starting a new one for those who are interested.

Here’s a link to the original thread: http://birdforum.net/showthread.php?t=120290

I have a busy week ahead so this post will likely be my only lengthy contribution. Edz and I approach testing binoculars at boosted magnification in significantly different ways. Rather than continuing to dispute particulars I thought I would post a description of my methods as used to test 4 binoculars today. I picked these four because they demonstrate a wide range of results. They are Nikon 7x50 Prostar, Nikon 8x32 SE, Zeiss 8x42FL, and Pentax 10x50 PIF.

I used two test targets today: a USAF 1951 Resolution test pattern to measure resolution and an artificial star (silver Christmas tree ball in sunlight ) set up near by to evaluate aberrations and defects. The targets were placed 78’ from the binoculars. I used a Takahashi FC-50 telescope of known resolution to evaluate the test conditions, mainly air stability, and as a resolution reference to be sure that the test set up was producing accurate results.

I used one barrel of a Nikon 8-16x40 XL zoom binocular (stopped down to 20mm to prevent light leakage) behind the eyepieces of the test binoculars to boost magnification. The zoom allowed me to easily arrive at a magnification for each binocular that clearly showed it’s resolving power. In the end I equalized the magnifications of all the binoculars to about 80X, comfortably high enough to see the resolution in all of them.
Here are the results for raw resolution of the best barrel within about 6%. The elements of the USAF chart are in 12% steps. I didn’t bother to tweak the distance for each binocular to get a more accurate figure.

Nikon 7x50 Prostar - 3.06”
Zeiss 8x42 FL - 3.06”
Pentax 10x50 - 3.87”
Nikon 8x32 SE – 3.87”

These raw numbers may look similar but they actually indicate very different levels of optical quality; something that is quite obvious in the appearance of the image at 80X in the different binoculars. The Zeiss and Nikon SE images look quite impressively clean and contrasty. The Nikon Prostar and especially the Pentax look gauzy and soft with low contrast and insecure focus. It’s possible to assign a number to the optical quality with a simple formula: Resolution X Aperture in mm. A perfect telescope will show a number around 115-120 in this test. The higher the number the worse the optical quality. Using that formula the binoculars sort out like this:

Nikon SE – 124
Zeiss FL – 129
Nikon Prostar – 153
Pentax PIF – 194

By binocular standards two of these are exceptional, one is unremarkable and one is rather poor, but far from the worst I’ve seen. Now the question is: what is wrong with the Prostar and especially the Pentax? This is where the artificial star comes in. A quick 80X star test (with and without a green filter to evaluate spherical aberration) reveals slight miscollimation , a little pinching and overcorrection of spherical aberration in the Prostar. The Pentax is a mess with overcorrection of SA, severe pinching and miscollimation (the other barrel is worse with astigmatism added). The Nikon SE star test is truly excellent for a binocular with very little spherical aberration and only a mild defect from a prism edge impinging on the aperture. The Zeiss is nearly as well corrected for SA, shows no assembly defects and has lower chromatic aberration than any of the others.

After these results you might think the Pentax would be headed for the garbage can, but remember this is what you see on axis at 80x, not 10x. In normal use at low magnification the Pentax is quite acceptable, really quite good in the barrel without astigmatism in spite of its sloppy optical construction, and its field flattener gives it quite nice off-axis performance. The Prostar shows no problems at all in normal use at an even more forgiving 7x and its off axis performance is as good as it gets in binoculars. I’ve tested many binoculars this way over the past 20 years, so I long ago got used to how bad the optics of binoculars can look at high magnification. Now I use these tests as part of a total evaluation. When I see something bad at high magnification it simply alerts me to look very carefully to see if it really causes damage to the image in normal use. Sometimes it does, sometimes it doesn’t.

Henry Link

 

[SUPPRESSOR]

Henry,
Once again thanks for an interesting and informing post,you seem to have the knack of conveying information that us mere mortals can grasp without our eyes glazing over. Now all we need is to have you come along when we purchase a new scope or bins!!
fiddler.

 

[Surveyor]

Thanks Henry, I enjoyed seeing your process. Mine is remarkably similar and, at the same time, somewhat different.

I start by taking the objective size of the test optic (example: 8x32) and dividing that into 116 to get an expected max resolution (116/32=3.6”) and then divide the max expected into 200 (200/3.6=55.6x) to get to the minimum power I can use and then divide this by the bino power (55.6/8=6.9x) to get a minimum booster power and go to next larger size booster I have available. The 200 figure allows me about a 30% safety factor over my normal eyesight. When low power, big objective instruments require larger boosters than I have, I will cheat back towards the 150 apparent resolution mark.

My available boosters are:
A 3x at Inf. to 30x at 4 inches, inverted image commercial alignment scope (not used much).

A 4.7x dioptometer for some occasional curvature and astigmatism measurements.

A 5x fixed focus scope with a 1 and 3 degree reticle.

A 6x double collimator with a 5’ grid, inverted image, commercial.

A 8x Zeiss monocular.

A 10x LXL bino.

A few 30x commercial alignment scopes, but way to much trouble to use.

Instead of an open-air system, I use a closed system to control light and atmospheric conditions. I use a collimator with a 400 mm focal length with a USAF 1951 negative resolution reticle, supplied with about 14 or 15 EV of 5000K-6000K (usually 5200K) light through a diffuser. At this time I do not have an internal condenser lens between the light source and diffuser, but use an external one when needed. I just have not bought the lens or had the case machined yet. I have not noticed this having any effect on low power instrument measurements though.

The collimator always presents the test optics and boosters with an infinity image. The process is to make sure the boosters are focused on the collimator image, then the test optics are inserted between and then focused to bring the booster image to sharp focus. All components are then, hopefully, focused to infinity. This system allows for very close spacing of booster, bino and collimator to control light and air currents and controls power differences to those at infinity focus. Normally tests are made in a darkened room, enough light to write and work, but low enough to keep stray light problems down.

After making the readings of the USAF chart and reducing to a resolution figure I compare the results to the approximate expected figure I started with. I usually expect this to be about 110% to 150% larger than the estimated starting value. If the value exceeds 160% or otherwise does not appear acceptable, then like Henry, I start looking elsewhere.

Since this discussion pertains only to resolution I will quit here.

Best to all.

Ron

 

[edz]

Actually Henry and I approach the testing in a nearly identical fashion. But I find it useful to record additional data, and I use that to report my results differently.

The post Henry linked to above contains several days of my written explanations on what I look for and how I use it.

Primarily, IMO, the one piece of the equation often missing is the magnification used to obtain the test result. There are numerous scenarios outlined in the linked posts that describe the importance of knowing the magnification and how it can be used.

For instance, while Henry might attempt to use the highest magnification that permits seeing the maximum resolution, I find a useful value would be the lowest magnification that will show the maximum resolution. Having this value for two closely compared and equal-sized optics may show that one reaches maximum resolution at an appreciable lower power than the other, a clear sign of finer optics, similar to the difference one might record if testing equal sized scopes, but one an achromat and one an apochromat.

One other thing is certain, poor optics show up quickly and do not improve. You can see the test for a poor optic well below the maximum power and generally you can’t improve a poor resolution reading by increasing power.

Henry rightfully questioned that I might not be approaching the maximum resolution if I do not test at a high enough power. That may be true. But a bit more on that in a minute to explain why that might not be an issue at all when using apparent resolution.

How I record my data:
First I test the binocular optic to determine its true power, for instance for my 12 different nominal 8x binoculars, that ranges between 7.5x and 8.4x. Then I test the power of my auxiliary scope. I apply auxiliary magnification and record total power for each optic tested to record resolution and power factor. When I obtain all my results of raw resolution, I multiply raw res x power and get a value for apparent resolution.

How much power do you need.
Optics will approach very very close to their ultimate resolution even when tested at magnification somewhat below power needed to see the limit of resolution. Testing a very fine optic, I got these results.
80mm scope, tested in daylight on line pairs
Magnification range 20x to 180x,
resolution range from 4.8” to 1.35”
at 100x, 67% of the magnification, it had reached 90% of the best resolution.
at 115x, 80% of the magnification, it had reached 95% of the best resolution.
At 140x, (100% magnification, beyond 140x is empty) it reached maximum resolution
At 180x it still gave the same reading for resolution (a fine scope indeed)

Resolution will plateau for a period, but then potentially, in some optics if you go too high in power, the resolution results deteriorate. I tested a premium 10x50 and 10x70 both at 60x and at 80x and found resolution did not improve, it stayed the same. It had already reached the resolution plateau at 60x. I also tested a fine 12x50 at 72x and again at 84x. Resolution at 84x deteriorated. Beyond 72x, I had exceeded the limit of magnification this optic could handle. I tested a fine 16x70 at 96x but also at 40x. This particular binocular had reached with .25” of ultimate resolution, less than one step below max, by 40x.

I checked two decent 8x42s at 48x and 56x. For one of them, I got a one step improvement in resolution by moving up to 56x. For the other I got no improvement.

(I’ll concede that the 6x booster is just to low to test the limit of 7x50s.) I tested my premium 7x50 at 56x and lowered the reading from 4.3 at 42x to 3.4 at 56x. I’ll need to boost all my 7x50s again. Although interesting to note is that it almost gives no change in apparent resolution.

Nearly all the retests with additional boosted power to check values showed less than 5% change in apparent resolution value. In other words, once close enough to maximum resolution, which should occur easily within 75% to 80% of needed magnification, the apparent resolution value will change by only a very small amount. As power is increased (to a very particular point explained below) resolution gets better to a point. The product of the two changes very little.


What point are we looking for?
If you were to take a group of different optics and test them at a variety of magnifications from normal up to the limit of resolution, then plot curves for magnification and res, when you plot resolution against power, you would see a graph that slopes down slowly then forms a nadir plateau (levels to a flat line at the best res). Eventually res will then turn upward, increasing again, getting worse with too much magnification. For magnification, if you plot it against improving resolution, you would see a slow steady climb, then at the lowest limit of (best) resolution it spikes straight upwards.

The most meaningful point of these two curves is the point at which magnification spikes upward. That is the same point that resolution first reaches the nadir and begins to plateau. If it takes one optic only 60x to reach 3 arcsec res (180 arcsec apparent) and it takes another same size optic 70x to reach the same 3 arcsec res (210 arcsec apparent), clearly there is a marked difference in the two optics. One of them is producing an image quality that allows it to reach maximum resolution at a lower power. While resolution gives you significant information about the optics, it does not give the whole story. The point at which maximum resolution occurs gives more information. That point is defined as the product of maximum resolution and lowest magnification at which it occurs, the apparent maximum resolution.

With each increase in objective size, potential resolution gets smaller. It will take more magnification to see it. You will have a lower limit that requires you to apply enough magnification just to see the resolution. Theoretically, if all the binoculars had best possible resolution for their aperture, you would need to apply magnification to each differently just to reach that level of seeing it. My 80 scope test results, the finest resolution tested here, show I can see 1.35 arcseconds resolution, well below the diffraction limit, when I raise it to 190 arcseconds apparent. Few if any of these binoculars are even reaching the diffraction limit. A few of my best binoculars also test to a lowest apparent reading near 185-200 arcseconds, but many are in the range of 200-275 arcseconds and quite a few of the poorer readings are over 300 arcseconds.

If you convert all these best reading values to apparent resolution by multiplying best res by magnification at which it occurs, this allows comparing different sizes and powers of optics directly, as the ones with the lowest apparent resolution generally show they have reached the optimum for that size at the lowest power. It is a common denominator that would allow you to compare various binoculars. Many do not reach their best resolutions at these lowest magnifications. The wide range at which you can see close to the limit and also at which the apparent resolution changes little allows some latitude in obtaining readings.


edz

 

[FrankD]

Gentleman,

I readily admit that much of what you discuss is a bit beyond my understanding most of the time (though I am guessing that if I re-read everything a few times I might get more of the gist of it).

What I would like to say though is that I do truly appreciate your efforts. After being a member here for four or five years now I feel I have a much better grasp of optical concepts and designs and I owe that all to you folks.

Thank you.

 

[edz]

The use of magnification as a variable in the reporting of resolution is not restricted to the high power test. For all straight on normal power binocular resolution testing, rather than simply record the resolution seen, multiplying the res recorded by the magnification of the binocular to arrive at apparent resolution provides the ability to compare ALL different sizes of binocular to one another. Keep in mind, actual power is often different than nominal stated power. And using closer focus raises power even higher, so resolution testing should be performed at the same distance at which power was tested, otherwise you've lost your control power values.

In normal power testing the primary influence on the resolution seen is the magnification, not the aperture. Raw resolution results for binoculars (assuming all relatively same quality) will be progressively smaller as you increase magnification. So a 7x50, 10x50 and 12x50 will all get different readings. It’s a relatively easy matter to compare 6 different 8x40s. Since they are all the same size, generally, the best raw resolution wins. But how do you compare a 7x50, 10x50 and 12x50? Obviously the 12x50 will have a lower res reading than the 10x50 and both will be lower than the 7x50. The raw readings cannot be compared to see which is better. But when you multiply the resolution readings by the magnifications, it provides the common denominator. Now you can compare the apparent resolution of the three different powers, something you could not do without calculating apparent resolution. A 7x binocular that can see 11 arcseconds (=77) is a much finer instrument than a 10x binocular that can see 9 arcseconds (=90). For the 10x binocular to equal the performance of the 7x, the 10x would need to see 7.7 arcsec. Spotting relative quality by this method becomes a simple matter of looking for the best apparent resolution value.

One argument has been this shows nothing more than your limit of acuity. Not so. I've tested nearly 6 dozen binoculars. The lowest value I’ve recorded may be establishing my acuity (76 arcseconds apparent), but few binoculars approach the lowest reading I've recorded. A dozen or so are in the low 80s, perhaps another dozen are under 90, and the range extends up to 100 arcseconds. So for perhaps 58 out of 60 binoculars my values are not limited by my acuity.

edz

 

[Surveyor]

A 7x binocular that can see 11 arcseconds (=77) is a much finer instrument than a 10x binocular that can see 9 arcseconds (=90). For the 10x binocular to equal the performance of the 7x, the 10x would need to see 7.7 arcsec. Spotting relative quality by this method becomes a simple matter of looking for the best apparent resolution value.
edz

Hi Edz;

I guess we will just have to agree that we have a difference of opinion here.

The way I see your example of the 7x 11” and 10x 9” comparison, assuming that you are seeing the limiting resolution with your good eyesight, is that the 9” instrument is better. It has 9” resolution in object space and that is not going to change regardless of which eyepiece magnification you choose to get into image space or apparent resolution. Same with the 7x at 11”, the 11” is in object space and not going to change however the image is manipulated into image space. If you change both eyepieces to 7x, the 9” would be better, if you changed both eyepieces to 10x, the 9” is still better.

Have a good day.
Ron

Hmmm, just noticed that I somehow missed post 4 and 5 this morning. Am catching up.

 

[henry link]

Ron,

I also use an indoor 10m test set-up and thanks to your help I have what I need to assemble a collimator. I've just been too lazy to do it. It's on my list.

Ed,

I try to generate all the "data" I consider to be meaningful and reliable. If I don't trust a test, I don't do it, particularly if it's a pain in the rear. At high magnification I find it's really pretty obvious from image quality alone whether a binocular has excellent axial performance or has some sort of problem. Measuring the resolution may roughly quantify how bad the problem is, but will never indicate the cause. Surely there's not much point in laboriously plotting a resolution curve for a defective binocular. A high magnification star test prevents that sort of wasted effort. It's much easier to do and much more informative than any combination of resolution tests. Usually it's the only test of axial performance I do before a buy a binocular. You should try it.

Henry

 

[edz]

Ed,

Surely there's not much point in laboriously plotting a resolution curve for a defective binocular. A high magnification star test prevents that sort of wasted effort.
Henry

Henry,

didn't mean to mislead you into thinking i do this for every binocular. I don't. In fact it's not necessary. I did it once, for my premium scope, several years ago. That taught me what can be seen from the visual of the graph. I tried to give you the same visual. Couldn't think of a better way to do that, than to describe what the graphs look like.

edz

 

[edz]

Hi Edz;

I guess we will just have to agree that we have a difference of opinion here.

The way I see your example of the 7x 11” and 10x 9” comparison, assuming that you are seeing the limiting resolution with your good eyesight, is that the 9” instrument is better. It has 9” resolution in object space and that is not going to change regardless of which eyepiece magnification you choose to get into image space or apparent resolution. Same with the 7x at 11”, the 11” is in object space and not going to change however the image is manipulated into image space. If you change both eyepieces to 7x, the 9” would be better, if you changed both eyepieces to 10x, the 9” is still better.

Have a good day.
Ron

Hmmm, just noticed that I somehow missed post 4 and 5 this morning. Am catching up.

Ron,

seems you are thinking in terms of what can be seen, which, regardless of quality is almost always dictated by magnification. The example cited describes the comparison of two different sized models and how they compare to each other for QUALITY, not quantity. The magnification will dictate the image resolution, but only to a point. Each will vary up or down from mean quality according to the specific quality of that instrument.

You can't play around with changing eyepieces here to support which will still be better, If you change the eyepieces, then the output changes also, because it is magnification dependant.

If you always select the lower resolution as better, then you would have a difficult time ever trying to say, I have a 7x binocular that is of higher quality than a 10x binocular. Selection based on raw better resolution would preculde that from ever being a possibility.

edz

 

[henry link]

Unlike Ed, I've never once been able to see a measurable difference in visible detail on the USAF chart between two binoculars of the same magnification, when they were measured at the same time under the same conditions. On the other hand I've measured different visible detail for the very same binocular on different occasions because my eyesight acuity is not completely stable.

I tried again today with several 8x binoculars of very different quality levels. I started with a tripod mounted Nikon 8x32 SE with measured resolution of a little better than 4 arcsec. I moved it until I could barely make out Group 1 Element 1 at 380". That works out to about 10.69 arcsec at about 8.2X at that distance. So my eyesight acuity today would be about 88 arcsec, about as good as it ever gets. Sometimes it's more like 95-100 arcsec. I replaced the SE with an old Nikon 8x30 E which has a very good left barrel (4.5"), but a rather poor right barrel (about 6.5" from miscollimation and astigmatism). Magnification of the E is also about 8.2X at this distance and of course the right and left barrels are matched. I found I could see the very same Element as the SE in both the good left barrel and the bad right barrel. Just for fun I tried an old uncoated Leitz Binuxit from 1930. Even with 50% light transmission and very low contrast I could still see the same Element. Of course there are profound differences in the image quality of these binoculars, but measurable detail differences on the USAF chart at 8x is not one of them. I have to conclude that for me to see a difference I would need much better eyesight acuity and/or some much worse binoculars to test.

 

[elkcub]

All,

Reading over these two interesting threads, it occurred to me that there might be a way to put the main ideas together into a broader framework.

Elementary texts describe the minimal objective resolution needed to design a binocular this way. If the objective's true resolution is multiplied by power, the result should equal or exceed the size of the retinal matrix, or visual performance will be compromised. This quantity, resolution times power, is essentially what EdZ refers to as "apparent-resolution," so I'll use that term. When apparent-resolution is greater than what is needed for visual acuity, the excess is essentially wasted — or said to be "more than the eye can use." If apparent-resolution is less than what's needed, however, acuity will suffer and no additional magnification will help. (That doesn't happen except for really bad products, but I have seen a few.) As Henry described, above, if one has several decent binoculars of a given power, they will all produce the same acuity test results because they each provide the essential data needed by the observer.

Some might quit at this point, feeling that good enough is really good enough. I tend to be one of these, since I don't like feeling envious of someone else's raw resolution. Still, there are those who just can't leave well enough alone. Actually, these guys are usually astronomers who need to push things to the limit. They stay up at night thinking about it.

To study the fascinating topic of raw resolution, a procedure has evolved to gradually boost power to the point where there is no excess apparent-resolution. At that point retinal requirements will just be met, but no longer exceeded, and additional magnification will no longer improve visual performance. That limit will be reached at a higher power for observers with poorer acuity than those with excellent acuity, and this will be reflected in their test score. Still, optics can be rank ordered based on the results from individual observers of different acuity, ... and that should nail down which is finest for astronomy tasks, but maybe not so relevant for birding tasks.

Within this rarified zone of magnification, one notes systematic variations in target detectability and image quality over a range of powers. Things don't always show a distinct transition from what can to what can't be seen. Why is this? Does the pattern of anomalies reflect quality differences between the optics — the holy grail for astronomers? First, note that we are talking about the limits of visual acuity at the extreme of instrument resolution. This is a far cry from what happens at lower powers when there is lots of excess apparent-resolution slopping around. In this situation, almost anything could influence the eye's demand, the instrument-eye coupling, or the instrument itself, — most, but not all of which, degrade visual performance. A little decentering of the eye could just change the chromatic pattern just enough to generate pseudo-resolution. A small change in pupil size or accommodation (they are coupled and dynamic), could change the aperture or focus of the whole system. The temperature or humidity could effect the internal stresses within the instrument. This is delicate work, at best, and fraught with measurement errors of all sorts.

Personally, I have no problem associating various properties of the retinal images seen under these conditions with the "quality" of the instrument, but for me it seems like murky waters and really tough to prove. I'll accept the experts' opinions.

I hope not to have offend anyone, as several comments are made tongue in cheek.

Henry, your first post was a real gem.

Ed

 

[henry link]

Thanks for a very sensible post, Ed. One small optogeek quibble with paragraph 4: I think those with better acuity see the limit at lower power. Since I've never been able to see any deficiency on the USAF chart at normal magnification with any binocular I've tried and my eyesight acuity is not going to improve I thought I might go out shopping for the worst possible binocular to test. The Optic 1050 (See Up to 35 MILES!) looked intriguing, but at $29.95 is out of my price range for this purpose. Besides, I don't have any other 1000X binoculars for comparison and I might want to see something more than 35 miles away. I saw a real binocular for $9.95 at a local drug store chain, but even that might be too good. Will the store give me a refund if it's not bad enough?

Henry

 

[Steve C]

Elkcub,

Thanks for that post. I've been making an attempt to follow this topic both in the Nikon forum where it showed up, and here. First I'd think that Henry maybe had it right. Then I'd think that no, Edz has it figured out, or, well heck Surveyor sure just made a good point! My skull full of mush was out of its depth. Your post seems to tie this together.

Now a question to all who have posted. Is this something we shouldn't try at home? Or how can an interested ameteur set up something like this to test before buying?

 

[Surveyor]

Ron,

seems you are thinking in terms of what can be seen, which, regardless of quality is almost always dictated by magnification. The example cited describes the comparison of two different sized models and how they compare to each other for QUALITY, not quantity. The magnification will dictate the image resolution, but only to a point. Each will vary up or down from mean quality according to the specific quality of that instrument.

You can't play around with changing eyepieces here to support which will still be better, If you change the eyepieces, then the output changes also, because it is magnification dependant.

If you always select the lower resolution as better, then you would have a difficult time ever trying to say, I have a 7x binocular that is of higher quality than a 10x binocular. Selection based on raw better resolution would preculde that from ever being a possibility.

edz

Edz;

Again I will offer a different view, maybe wrong, but good for my education on these things by the discussions.

First, we start by looking at a real object in the field. This image travels thru the objective, focusing lens if present, prisms to the focal/image plane. This focal plane is a real image whose quality is controlled by the resolution and aberrations of the components listed, but mostly the objective lens, analogous to the film plane of a camera. Like film, this real image controls the rest of the train past it and the details in this image can just be magnified, not really improved on, just made worse, same as enlarging a negative. If this particular objective system has a native resolution of 4”, any angle smaller than that will be fuzzy. All the eyepieces do is create a virtual image to magnify this image to allow you to see the detail that is available but the 4” fuzzy section is also magnified. In the case of a 10x eyepiece the fuzzy section will be 40” of your AFOV, or the apparent resolution. As long as your visual acuity number (75, 120, or whatever it is) is higher than the apparent resolution (most seem to think by a factor of 2) then the optic is good for general use. Once you pass your visual acuity (and with me, that is pretty variable), what you describe as the knee and plateau, then further magnification just results in a larger image size with no increase in detail.

To summarize, if my visual acuity at a particular time is 120 (varies from 110 to 140) arc seconds and the objective system is capable of 4 arc seconds resolution I will see increasing detail up to a magnification level of about 30x. As I go up in magnification above 30x I would not see much, or any, increase in detail, just a larger image. If my eyesight were as good as Henry and yours, say 85, this point would occur at around 21x. Neither indicates quality, just acuity, since the fixed values are the magnification and native resolution.

Just some thoughts. Have a good day.

Ron

 

[elkcub]

Thanks for the correction, Henry. And thank you too, Ron, for explaining it properly. Better acuity, lower power. Got it.

If you have a Big 5 sporting goods store around they are a treasure trove for bad optics. I looked through one once and could see less than with the naked eye. Can't remember the price. Actually, they charge more for those, because otherwise no one would buy one.

Ed

 

[edz]

Unlike Ed, I've never once been able to see a measurable difference in visible detail on the USAF chart between two binoculars of the same magnification, when they were measured at the same time under the same conditions. QUOTE]


When binoculars were tested at normal powers, I often see differences in resolution between models of same power. generally I bring out one or two used previously to benchmark.

6 - 7x50s ranged from 12.1” to 10.8”
14 - 8x40/42s ranged from 11.5” to 10.3”
4 -10x42s ranged from 9.7” to 8.1”
12 – 10x50s ranged from 9.7” to 8.1”
5 – 12x50s all ranged close, 7.5” to 7.2”
3 – 15x70s ranged from 6.5” to 5.8”
3 - 16x60/70/80 were all 5.8”
6 – 20x80s ranged from 5.4” to 4.6”
1 – 20x60 reached 4.4”

Results can be slightly deceiving. About half of all the binoculars actual magnification is not what is stated as nominal. For instance, of the 14 8x40/42s, 5 were greater than 8x and 4 were lower than 8x. Of the 12x50s, three are 12.0x, one is 12.2x and one is 12.5x.

edz

 

[edz]

In my last post I pointed to the range of results I have on record. You can see within any size, results range by 12% to 20%.

Earlier I mentioned an example comparison, and I stated a 7x binocular that can see 11 arcseconds (=77) is a much finer instrument than a 10x binocular that can see 9 arcseconds (=90).

In the 10x binocular group the three best 10x42/10x50s could see 8.1 arcseconds. Several other 10s were able to see apparent resolution around 84-87 arcseconds. There were a number of 10s that could see no better than 9.1" (9 in my ex.) and even two that were so poor they could see no better than 9.7". Not surprising, the several 10x binocs that could see resolution of 8.1 arcseconds also scored among the highest in total score of 15 different categories. The 10x binoculars that could see no better than 9 arcseconds were at best mediocre, and several of them were among the lowest total scores out of 60 binoculars, some binoculars that I would not recommend.

In the 7x binocular group, the best could see 10.8 (11 i my ex.) arcseconds. Two rather mediocre 7x binoculars could only see 12.1 arcseconds. The 7x binox that was able to see 11 arcseconds is among the highest scoring of all binoculars in overall quality out of 60 binoculars.

The fact that a higher powered binocular can see smaller resolution is no indicator of quality. You simply cannot compare raw resolution. You can see from the ranges of data above, that in almost all cases, no smaller binocular would ever be considered higher quality that any larger binocular, simply because the brute strength of magnification would overpower the smaller size. That does not define quality. Quality is defined by how fine can a specific power can see in comparison to how fine it should see. In some cases that range varies as much as 20%. Even the worst 10x binocular always beat out even the best 8x and 7x binoculars. I'm sure there are no shortage of users everywhere that would be hard pressed to accept that some medicre $99 10x50 binocular is better than their 8x32SE or a 7x50 Prostar.

edz

 

[henry link]

I'm going to offer a possible explanation for why Edz and I report such different experiences with visible detail on the USAF chart at normal magnification. I've been looking at too many good binoculars and he's been looking at too many bad ones. I have a collection of binoculars that have mostly minor defects. Not all are expensive (most recent purchase, $179) but all have passed some kind of minimum test. Ed has been testing all sorts of binoculars, some good, but also some among the low priced roof prism binoculars and inexpensive Chinese Porros that are probably pretty poor or defective. I'm not much interested in cheap roof prism bins for myself (if I buy cheap I buy Porro), so I've evaluated only a few when friends wanted an opinion. Those few have been bad in various ways that were apparent without any serious testing. On the other hand there's not much excuse for an inexpensive Porro to show poor axial detail, unless it's very badly made. So my best guess as to why some of the binoculars on Ed's list show such poor performance on the USAF chart at normal magnification (well below his typical eyesight acuity) is that they fall into one of two categories: defective Porros or inferior roof prism bins which are probably even more likely to be defective.

 

[henry link]

Now a question to all who have posted. Is this something we shouldn't try at home? Or how can an interested ameteur set up something like this to test before buying?

Steve,

I've been doing this so long that it's become easy and quick. I used to recommend it to everyone, but I've noticed that lining up tripods and using one binocular to look through the eyepiece of another one doesn't exactly come naturally to most people. Interpreting a star test is also tricky. Still, it doesn't take much to get started; two tripods, an extra pair of small binoculars or a small monocular, a small round shiny object in sunlight for artificial star testing, a cheap USAF 1951 test pattern plastic card from Edmond Optics and you're in business. Have fun.

Henry