• Welcome to BirdForum, the internet's largest birding community with thousands of members from all over the world. The forums are dedicated to wild birds, birding, binoculars and equipment and all that goes with it.

    Please register for an account to take part in the discussions in the forum, post your pictures in the gallery and more.
ZEISS DTI thermal imaging cameras. For more discoveries at night, and during the day.

Binocular Resolution Test Results (1 Viewer)

Atomic Chicken

Registered with the D.O.E.
Greetings!

Today, I began a project that will eventually result in a new binocular oriented website I am getting ready to create. My main goal with this project is to do completely objective scientific comparisons of various binocular models, presenting information to help potential buyers select binoculars based on what is most important to them. Of course, the view and how the buyer feels about the view is really the ultimate determining factor, but in some cases binoculars are so close that it helps to compare specifications and objective scientific measurements the binocular companies seem hesitant to publish.

The first aspect of this project is to define a standardized measurement system for resolution measurements, and to do a sample run using several models of binoculars I own. I will be doing MUCH more of this type of measurement in the future, both at optics shops and with my own optics as my binocular collection grows. I think I have arrived at a method that should eliminate almost all extraneous factors, and provide solid and reliable data.

In order to present meaningful, scientifically derived, and (most importantly) OBJECTIVE information, I have approached the resolution testing using the following method:

#1 - All binoculars are mounted to a tripod mounted universal binocular clamp, devised and machined by me in order to secure the binoculars under test securely and eliminate all vibration and shaking. This also assures that the binocular can be correctly pointed, with the test chart properly centered in the view.

#2 - The test chart used, also devised by myself, contains a series of 54 zones, each approx. 1" square. Each zone contains either horizontal or vertical parallel lines that are of a fixed width and a fixed spacing, both the same dimension within that zone. The chart contains zones for 0.005"-0.020" in 0.001" increments, 0.022"-0.030" in 0.002" increments, 0.035"-0.050" in 0.005" increments, and finally a 0.060" and 0.070" zone. The chart is divided into two halves, with the upper half consisting of vertical parallel line zones for determining the binoculars horizontal discrimination, and the lower half of the chart is composed of the same measurement zones repeated with horizontal parallel line zones for determining the binoculars vertical resolution discrimination. The chart is printed on a 1200DPI laser printer on 98 brightness 24lb thick paper, and checked carefully for printing defects before use. This test chart was devised because it will give solid information on the actual resolution capabilities of the binocular in both axis, and the resulting measurements can easily be converted mathematically into an arc measurement if needed (example: 6 arc seconds or 5.4 arc seconds).
Edit: I've posted the chart as a .pdf file in the following thread message for download.

#3 - The test chart is hung on a wall at the end of a hallway, and the binocular objective lens is set to the exact same height as the center of the test chart. The binocular objective lens is set exactly 30 feet from the target, measured to +/- 0.5" tolerance.

#4 - All lighting in the area is extinguished. The test chart is illuminated by a single 75 watt bare light bulb which is located exactly 6 feet from the chart, and 2 feet to one side. The bulb is shielded completely in the direction of the binoculars by a large book, which has a matte black cover. This keeps light from reaching the binoculars from the bulb directly, which could cause internal reflections or other visual artifacts. The matte black book cover keeps the book blocking the light source from reflecting extra light back onto the test chart, avoiding extra light that might skew the results. The other end of the hall, where the binoculars are located, is kept completely dark during the testing. The testing is performed after midnight, to assure no sunlight enters the house.

#5 - After focusing, assuring that the diopter adjustment is properly set, and getting the clearest image of the test chart possible, the chart zones are examined one by one from left to right, and top to bottom, until a zone is reached where the "grayness" of the fused lines is replaced by actual visible lines. Even if the lines are fuzzy or hazy, the point is that the zone where the binocular begins to separate the lines into visible separate entities is the zone that determines the resolution discrimination for the test. The next zone to the left should remain as a uniform gray patch, with no visible line separation. The process is repeated for both vertical and horizontal resolution, and the results are logged.

#6 - Meaning of resulting measurements. Each binocular is given an h (horizontal resolution, from parallel vertical lines) rating, and a v (vertical resolution, from parallel horizontal lines) rating. The smaller the number, the better. An example would be:

Brand X 8x42
0.013h 0.014v

This would mean that the binocular was able to resolve vertical parallel lines that are 0.013" thick spaced 0.013" apart, meaning that the binocular is capable of horizontal discrimination of 0.013". It can also resolve parallel horizontal lines 0.014" thick spaced 0.014" apart, meaning that the binocular is capable of vertical resolution of 0.014". This result would mean that the binocular is capable of slightly better horizontal resolution than vertical. Since lenses are circular, you would expect the two numbers to be identical - but I've actually been surprised so far at how much difference between the two measurements some binoculars exhibit.


Resolution test results for my binoculars (Lower numbers are better):
===================================================

Rugged Exposure 10x25 Waterproof
0.020h 0.016v

Barska 16x32 Compact
0.013h 0.012v

Rugged Exposure 10x42 Waterproof Roof Prism
0.012h 0.012v

Bausch&Lomb 7x42 Discoverer
0.013h 0.013v

Zeiss Victory 8x20 Compact
0.012h 0.012v

Nikon HG 8x32
0.012h 0.011v

Pentax SP 10x50
0.010h 0.010v

Note that the cheapest binocular (the Rugged Exposure 10x25 Waterproof at $20.00) had the worst deviation between horizontal and vertical resolution results. Also, interestingly enough, the Nikon HG 8x32 seems slightly better at resolving vertical detail than horizontal (or worse at resolving horizontal detail than vertical, depending on how you look at it! ;) )- a trait not shared by any other high end binocular I tested.

Keep in mind that resolution says NOTHING about the overall image quality of the binocular, only it's ability to resolve detail. For example, the Barska 16x32 is my dimmest pair of binoculars, these are like looking down dark tunnels at a flashlight at the other end. After testing those binoculars, the next pair seemed like someone had suddenly turned on the lights! Also, the other cheap pair, the Rugged Exposure 10x25 had such incredibly poor optical quality they were like looking down two tubes of fog, oil, and dimness all mixed together. These are EXTREMELY unpleasant to use, they are definite headache specials. I purchased them recently to disassemble for a tutorial I will be placing on my website, so this is probably the last time I will be looking through them - thank god! The Rugged Exposure 10x42, on the otherhand, is a sheer joy to look through and continue to completely blow me away with their magnificent view. How someone can make a pair of binoculars like that and only charge $60 for them completely mystifies me. The Nikon HG 8x32 easily had the best view of all binoculars tested, with the Bausch&Lomb 7x42 Discoverer coming in VERY close 2nd. Neither of these binoculars had the best resolution, however, which was achieved by the Pentax SP 10x50. Once again, resolution and overall image quality are not necessarily related.

In any case, I hope you have all enjoyed reading this extra-long post, I will keep the results coming as I test more binoculars in the future. I am hoping to have my website up in 2 or 3 weeks, there is a lot of work to do before I will be ready for the test run.

Edit: One more thing to note is that magnification has no relationship to resolution. Bigger is not always better... sometimes a good 7x or 8x will resolve more detail than a mediocre 10x. The Barska 16x32 and Rugged Exposure 10x25 results above, when compared to the other binoculars, clearly demonstrate this interesting effect.

Best wishes,
Bawko
 
Last edited:
Greetings again!

Here is a .pdf file I created for the test chart. Print it at normal size, do NOT turn on page shrink or expand in your print driver. If you print it out to use, please use at LEAST a 1200DPI laser printer, preferrably a 2400DPI or better. If you print it out on anything less, the dimensions on the 0.005"-0.010" zones will not be correct, and moirre-like band patterns will appear where the line widths and spacings in the smaller width zones interfere with the laser printer resolution limitations.

Also, this is Copyright (C) 2004 by myself, you are free to use it for your own testing but I am not transferring commercial rights. In short - don't profit financially from my work... many thanks.

Best wishes,
Bawko
 

Attachments

  • CHART1.PDF
    13.1 KB · Views: 3,149
Blimey, Bawko - a tour de force there!

But I can imagine your site being in everyone's Favorites in no time - thanks for the time and effort you're putting into this.
 
Hey Bawko,

Lot of thought and work gone into this project already, great stuff! One thing that jangles a little for me (may just show my complete lack of understanding of things optical ;) ) is the way that the distance is set from the objective ; I'm thinking it would be best measured from the other end, maybe from the glass surface plus 15mm (or whatever the quoted/measured eye relief may be).

Andy.
 
Andy,

Thanks for the feedback! The reason I chose the objective lens as the determining distance from the chart is that this is where the alteration of the incoming light begins, and binoculars of varying length between the objective and eyepiece can do essentially the same thing - providing the same magnification, etc. with different arrangements of lenses and prisms. If I were to use the eyepiece as the determining distance from the chart, then binoculars with longer tubes would have an unfair advantage over shorter tubed models, everything else being equal. Also, porro prism and compact binoculars would typically be at a disadvantage as well, being as they are smaller lengthwise. Think of the binoculars as "black boxes" of varying length that manipulate the light, with a light entrance and light exit. If the chart is placed the same distance from the entrance of all "black boxes" being tested, then meaningful test results can be derived regardless of the length of the boxes themselves or whatever is between the entrance and exit.

Hope this makes sense... it's not the easiest thing to explain clearly! ;)

Best wishes,
Bawko
 
Bawko,

This is an interesting project. I have two suggestions, which come from having done similar things.

1) I think it is a good idea to express resolution in the universaly accepted form of arc seconds, so your results can be directly compared to others. Edmund Scientifics sells the 1951 USAF chart in various forms which can be used to arrive at measurements in arc seconds.

2) I think if you want to measure actual resolution rather than perceived sharpness you will need to boost the binoculars' magnification with a "booster" like the Zeiss, Swaro or eagle optics, or a small low power scope placed behind the eyepiece. Every good binocular I have measured has had higher actual resolution than can be seen by just looking through it, even on a tripod. This also assures that you are measuriing the resolution of the entire objective and not just the part that the size of your pupil dilation allows.

Regards,

Henry
 
Henry,


henry link said:
Bawko,

This is an interesting project. I have two suggestions, which come from having done similar things.

1) I think it is a good idea to express resolution in the universaly accepted form of arc seconds, so your results can be directly compared to others. Edmund Scientifics sells the 1951 USAF chart in various forms which can be used to arrive at measurements in arc seconds.

This is exactly why I chose to create a test chart that is accurately calibrated in thousandths of an inch, so that conversion to arc seconds is easy and fast. I will, in all future testing, convert the results into arc seconds before posting them. I was previously aware of the 1951 USAF chart, and chose not to use it for my testing because I wanted to create a more detailed chart of my own which had more lines for each test section, and equivalent area test sections so that the area of each section is identical. I don't know for certain if having differing surface areas for each test section (like the 1951 USAF chart has) can affect the test results or not, but it just seems like one more variable that can easily be removed from the testing equation by designing my own chart the way I did. Also, I wanted to create a chart to upload that anyone can print out at a cost of $0.25 or so at their local copy shop using a decent laser printer and paper, without requiring the cost or shipping time to order through Edmund Scientific or another supplier.


henry link said:
2) I think if you want to measure actual resolution rather than perceived sharpness you will need to boost the binoculars' magnification with a "booster" like the Zeiss, Swaro or eagle optics, or a small low power scope placed behind the eyepiece. Every good binocular I have measured has had higher actual resolution than can be seen by just looking through it, even on a tripod. This also assures that you are measuriing the resolution of the entire objective and not just the part that the size of your pupil dilation allows.

This is something that I have struggled with long and hard. While it is nice to know the exact optical limitations of a set of binoculars, if these limitations are imperceptible under real world conditions then to me it equates to meaningless data with no real-world application. If we were all carrying optical boosters around with us and using them with our binoculars regularly, then I would be strongly in favor of testing with such a booster. As it is, I decided that the best approach to obtaining "actual use" resolution data would be to test the binoculars in a situation of maximum pupil dilation (best-case scenario), which is the reason for having the binocular end of the hallway completely darkened during the test and also part of the reason for shielding the lightbulb in the direction of the binoculars (and user). Perhaps in future testing I will consider taking both measurements, idealized and real-world, and posting both. I'm open to feedback on this issue.

I really DO appreciate your input very much, I know that you have probably forgotten more about optical measurement and theory than I have yet learned. I look forward to more suggestions if you have the time, and based on your current feedback I am shifting to arc-seconds only reporting for future resolution tests. I will also re-do the above binocular results in arc second units, and post the results. I am also planning to modify the resolution test chart to include both the 1/1000" measures and the arc-seconds equivalents.

Best wishes,
Bawko
 
Last edited:
John,


John Traynor said:
Bawko,
Are you limiting your resolution/sharpness analysis to the centermost part of the field?
John

So far, yes. When I get really serious about this testing and am ready to start posting results for various binoculars on the website I'm currently building, I'll find the center resolution, note it, then re-check on the left edge and upper edge of field to see if the edge performance of the binocular in question degrades resolution or not, and by how much. I'll publish all 3 figures for both horizontal and vertical resolution. For now, I'm working out the "glitches" in the method, in order to arrive at the most accurate method of real-world binocular resolution testing that can be easily duplicated by anyone who wants to check the results independently. Henry Link has already made a suggestion that will change how the results are presented, and I'm considering other minor alterations to the test methodology to make it more applicable to testing under a wider variety of settings. My current line of experimentation is to see how much, if any, the lighting plays into the results - if I can eliminate the currently defined strict lighting requirement and not skew test results, it will be far easier to gather data in optics retail stores, etc.

Best wishes,
Bawko
 
Just a brief comment along with best wishes and encouragement to your endeavours.

Henry suggested above that you consider using a booster for determining the resolution, and you had your reasons for not doing so for the present. However, if you do not use a booster, you are inroducing a non-duplicatable variable in the form of your eyes. Unless you happen to have visual acuity about 2.5-3 times better than normal 20/20 vision, or visus 1.0, any good binocular will have a significantly higher resolution than you are able to see. Also, the performance of your eyes varies over time and from minute to minute as well.

So, under identical conditions and with an identical pair of binoculars, person A will get a different result from person B if they both use their own individual eyes for the measurement. If you use a booster (of at least 3x magnification), the binocular becomes the weakest link in just about all cases, and your measurement becomes more objective.

You do have a good point about naked eye measurements correlating much better with actual use and perhaps taking better into account factors such as contrast, brightness and colour balance differences, so ideally you should provide both naked eye and boosted resluts.

In any case, to minimise variation, you might consider doing two things: 1) disclose your naked eye resolution at the time of doing each test (measured in arcseconds with the same test target but without any optical aids except for you glasses if you view with them on) and 2) always check with a designated and always same reference binocular first to see whether your eyes are giving your normal results for that binocular before you measure the bin you are about to test.

Steve Ingraham did much of this in his N.E.E.D. testing, although he gave up on providing booster-measured resolutions at some point, and was not as methodologically consistent in his reviews over time as one could have hoped for.

Kimmo
 
Kimmo,

Thank you for the EXCELLENT input. You have convinced me - I will most definitely obtain a booster and give both boosted and naked eye results for future tests - although I do have excellent vision (probably better than 90% of the population, based on my past experiences comparing notes).

I am highly motivated to pick up where Steve Ingraham left off... I would like to eventually achieve his level of reviewing skill and knowledge - hopefully without his well-known biases with regard to certain manufacturers.

I am very frustrated by the lack of objective binocular information available on the internet, I would say that over 75% of binoculars I attempt to research come up with zero usable information when I try internet searches. Sure, I get 5,000 companies trying to sell me the binoculars (4,000 of which are affiliated with Amazon.com in some way!), but reviews and actual test data are very rare and hard to come by. I am hoping to eventually change that, I know it is a lot of hard work but I definitely have the willpower and the time to put into this project.

I am making steady progress on the website development, I will be ready to unveil the prototype in a few short weeks.

Best wishes,
Bawko
 
Atomic Chicken said:
Kimmo,

Thank you for the EXCELLENT input. You have convinced me - I will most definitely obtain a booster and give both boosted and naked eye results for future tests - although I do have excellent vision (probably better than 90% of the population, based on my past experiences comparing notes).

I am highly motivated to pick up where Steve Ingraham left off... I would like to eventually achieve his level of reviewing skill and knowledge - hopefully without his well-known biases with regard to certain manufacturers.

I am very frustrated by the lack of objective binocular information available on the internet, I would say that over 75% of binoculars I attempt to research come up with zero usable information when I try internet searches. Sure, I get 5,000 companies trying to sell me the binoculars (4,000 of which are affiliated with Amazon.com in some way!), but reviews and actual test data are very rare and hard to come by. I am hoping to eventually change that, I know it is a lot of hard work but I definitely have the willpower and the time to put into this project.

I am making steady progress on the website development, I will be ready to unveil the prototype in a few short weeks.

Best wishes,
Bawko


A serious problem for using a booster is to fix it proper to the EP of the binocular if there is no adapter available because every shaking of the image will disturb the measurements and the comparableness of your findings.

Steve
 
hinnark said:
A serious problem for using a booster is to fix it proper to the EP of the binocular if there is no adapter available because every shaking of the image will disturb the measurements and the comparableness of your findings.

Steve

Steve,

I own a machine shop, devising and fabricating a universal adapter will not be too difficult. Do you have any recommendations as to what the best booster to purchase for this endeavor would be?

Best wishes,
Bawko
 
Atomic Chicken said:
Steve,

I own a machine shop, devising and fabricating a universal adapter will not be too difficult. Do you have any recommendations as to what the best booster to purchase for this endeavor would be?

Best wishes,
Bawko

Bawko,

I only know the 2x booster of Swarovski that fits only with their own binoculars and the Zeiss Mono 3x12 B Henry and Kimmo are using. The latter would also be my favorite because it works universal and it has a function of its own as a miniscope and a loupe.

Steve
 
Zeiss 3x12 (which I use) is fine, sharp enough and very handy size, but as of yet it is not T* coated, which makes it less bright than it could be. Jan Meijerink has used the Zeiss 4x12 Design Selection monocular, which does have T*. I have not compared these two. However, I consider 3x to be the ideal magnification for a booster, partly because I have also used with telescopes, and there a 3x booster gives a nice continuity with zoom eyepieces having a 3x range. The Swarovski booster is much heavier, and having only 2x magnification, does not provide enough headroom with truly sharp binoculars. As an actual booster in field use with its designated binoculars, it is very likely better for the same reasons, though.

Zeiss sells inexpensive plastic adapter rings for mating the 3x Classic with their Victory and Dialyt binoculars, but I have not used these since I do not own any of these bins. Instead, I have used self-made adapter rings made out of cell foam gasket strip material obtained from a local rubber and plastics supplier. I glue rings out of various thickness strip such that they fit snugly over the booster and inside various eyecups. For small adjustments in diameter, I use rings cut out of rubber bike inner tube or if very small increases are needed, sometimes masking tape. This works well enough so that the booster does not have to be touched and stays centered and aligned.

I'm sure you can machine a much more elegant system, however.

Kimmo
 
I usually use a 5X25 Takahashi finder scope mounted on a tripod behind the eyepiece, but not touching it. Any inexpensive small finderscope would work just as well. It's a bit of a pain in the rear to get everything aligned, but the crosshairs help. You might consider a zoom riflescope.

If you take things this far maybe you should also consider star testing with an artificial star while the magnification is boosted. That would allow you to quickly detect sample defects.

Henry
 
Although there's a good practical value in testing everything from the same distance (a bird won't let you get closer just because you've got lower power bins, for instance!), I can't help thinking that it's not objective enough to compare different magnification optics from the same point.

When using test-charts to test camera lenses testers have to fill the frame in the viewfinder with the chart (which means very close-up with wide angles and 'half-way down the garden' with super-telephotos!).

Of course, the fixed distance test would be OK for comparing binoculars of the same power with one another.
 
henry link said:
I usually use a 5X25 Takahashi finder scope mounted on a tripod behind the eyepiece, but not touching it. Any inexpensive small finderscope would work just as well. It's a bit of a pain in the rear to get everything aligned, but the crosshairs help. You might consider a zoom riflescope.

If you take things this far maybe you should also consider star testing with an artificial star while the magnification is boosted. That would allow you to quickly detect sample defects.

Henry

Henry,

I´m sorry for the error. I thought you use the Zeiss Mono too. For astronomical use Barlow lenses are quite common. In fact the Swarovski doubler is a barlow lens. So I assume that other barlow lenses used for astronomical purposes should do the same job with scopes and bins.

Kimmo: I once asked Zeiss about the reason for price differences between Design Selection miniscopes and the Mono B since the 3x12 Mono ist quite more expensive. They answered that the 3x12 has better and more costly optics. Unfortunaltely I didn´t ask for more details at this time.

Steve
 
hinnark said:
Henry,

I´m sorry for the error. I thought you use the Zeiss Mono too. For astronomical use Barlow lenses are quite common. In fact the Swarovski doubler is a barlow lens. So I assume that other barlow lenses used for astronomical purposes should do the same job with scopes and bins.

Kimmo: I once asked Zeiss about the reason for price differences between Design Selection miniscopes and the Mono B since the 3x12 Mono ist quite more expensive. They answered that the 3x12 has better and more costly optics. Unfortunaltely I didn´t ask for more details at this time.

Steve

I also think the price is remarkably high for the zeiss 3x12 mono. Is it any good for practical birding in use with binos or scope?
 
Adey Baker said:
Although there's a good practical value in testing everything from the same distance (a bird won't let you get closer just because you've got lower power bins, for instance!), I can't help thinking that it's not objective enough to compare different magnification optics from the same point.

When using test-charts to test camera lenses testers have to fill the frame in the viewfinder with the chart (which means very close-up with wide angles and 'half-way down the garden' with super-telephotos!).

Of course, the fixed distance test would be OK for comparing binoculars of the same power with one another.

Adey,

For my intent (finding the arc-second limitations of binocular resolution), the distance doesn't really matter, as long as the whole chart can be fit in the view for evaluation. As long as the calculation is done correctly, any distance can be used within the limitations of the chart - at greater distances only the larger test patterns would be visible, and at shorter distances the smaller test patterns would become visible. However, after noting this and doing the calculation with the distance plugged in, the differing distances will yield the same arcsecond result for resolution. I chose 30 feet arbitrarily, because of a few factors - I have a hallway in my house that is just a bit longer than that, and it is a convenient distance for testing optics at optics retailers, etc. where you don't want to set up a target at the other end of the parking lot or across the street! ;). Also, this distance seemed to place the range of smaller patterns on my chart (0.005 to 0.020) right in the middle of 7x-10x optics, with average results seeming to come up 0.010 to 0.015, making the 30 foot distance work well with the chart devised.

I think the chart you are referring to that camera testers use (and fill the full lens field with the chart) is probably not a resolution chart, but rather a chart devised to detect lens aberrations such as barrel distortion, edge effects, etc. I am currently looking at a few test charts of these types (typically for video equipment evaluation and calibration) in order to get some ideas for making my own chart of that type. When using that type of chart, my testing will very definitely be at varying distances in order to fill the binocular view with the entire chart. If the chart you are referring to IS a resolution chart, then my guess is that the distance between the lens and chart must be precisely measured in order to have that equation variable for the resolution calculation... not much different from what I'm doing except that a variable distance variable is exchanged for not filling the view with the chart. Personally, I'd rather have one less thing to measure - as it is I just have a plumb-bob string hanging from the cealing at the correct distance to tell me where the objective lens goes, making it fast and easy to change binoculars under test.

Best wishes,
Bawko
 
Warning! This thread is more than 20 years ago old.
It's likely that no further discussion is required, in which case we recommend starting a new thread. If however you feel your response is required you can still do so.

Users who are viewing this thread

Back
Top