• 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.

Review of Canon 10x32 IS (1 Viewer)

henry link

Well-known member
I finally just joined the ranks of Canon IS binocular owners. I’ve been sitting on the sidelines for many years, attracted to the idea and by other people’s good opinions, especially Kimmo Absetz’s enthusiasm for the 10x42 L, but also discouraged by my own unhappy, but admittedly brief and superficial experiences with trying a few of them.

Two things taken together finally moved me to take the plunge. One was some positive comments from fellow Birdforum member Canip about the new generation 10x32 and the other was a bargain basement price of $699.95 offered by B&H Photo.

I’ve been testing and using the 10x32 for several weeks and I think I’m now ready to start posting my results. I’ve decided to do this in four or five installments over the next few days, partly because there will be too many attachments for a single post and partly because I’m too lazy to write it all down at once.


Part One: Off-axis Aberrations and Distortion

With the one exception of lateral color, which I’ll come to in Part Two, the off-axis behavior of the 10x32 sets a new standard for excellence in my experience. Both field curvature and astigmatism are extremely well corrected over the measured 57º apparent field, easily to within a small fraction of a diopter. For my 20/15 eyes, using the USAF 1951 Resolution Chart, there is only about a 12% reduction in resolution at the very edge compared to the center. Any better correction than that would be pointless.

Distortion is also handled more adroitly than I’ve seen in any other binocular. The line of circles in the left image below shows approximately the outer 20º of the right side of the 10x32 field. There is only very slight evidence of angular magnification distortion in the shape of the last circle near the right edge. Compare its shape to the same circle severely distorted into an oval in the top row of three binoculars I previously imaged the same way (top: Swarovski 8x32 EL SV, middle: Nikon 8x32 SE, Bottom: Zeiss 8x56 FL). In these photos you can also see the Canon’s exemplary correction of field curvature, which makes all the circles nearly equally sharp, and the absence of astigmatism, which can be seen well on the Zeiss FL middle circle as a single focused line at the “poles” that defocuses toward the “equator”. On the down side you can also see chromatic aberration in the Canon image.

The distortion profile of the Canon demonstrates clearly that the “rolling ball” inducing mustache distortion used in the Swaro EL SVs (and the Zeiss SF) is totally unnecessary for correcting field curvature and astigmatism. Nearly perfect field flatness can be achieved with a distortion profile that should cause no panning disturbances for anybody.

The grid pattern image shows the Canon’s mild pincushion distortion (reversed to barrel when photographed through the objective end). Looking at the grid through the eyepiece suggests there may be a very subtle reverse mustache distortion near the edge, which slightly increases rather than decreases pincushion in the last 5 or so degrees, with the good effect of further correcting angular magnification distortion near the field edge, but with very little observable effect on straight lines because they’re so short near the edge.

The internal diagram of the 10x32 below shows the relative simplicity of the eyepiece design that accomplishes all the above. It appears to be a 3 element Konig combined with a doublet field flattener. The field flattener looks identical to the one in the 10x42 L diagram below, with the unfortunate exception of no UD glass indicated in the 10x32. Kimmo has praised the off-axis performance of the more complex 10x42 L eyepiece. I haven’t tried it, but I can easily believe that it would extend the good qualities of the 10x32 eyepiece to a wider apparent field and probably with less lateral color. Most discussion of the Canons centers on their IS feature, but I think if they were conventional binoculars we would be holding them up as exemplars for correcting off-axis aberrations and managing distortion.

More on the 10x32’s lateral color in the next installment.

Henry Link
 

Attachments

  • DSC_0470.jpg
    DSC_0470.jpg
    34.7 KB · Views: 322
  • Slide1.jpg
    Slide1.jpg
    65 KB · Views: 428
  • DSC_0493.jpg
    DSC_0493.jpg
    81.4 KB · Views: 400
  • lens-shift_2_.jpg
    lens-shift_2_.jpg
    13.3 KB · Views: 260
  • DSC_0238.JPG
    DSC_0238.JPG
    57.1 KB · Views: 208
Last edited:
Thanks Henry.
Looking forward to further installments.

Perhaps a bargain basement 10x42L IS will come your way later :)
 
My thanks too Henry for all your efforts. It's great to see an objective report (again), on BF and the insights you bring to binocular design are much appreciated. I can see now while it's taken awhile to appear as there is a lot of info there.

On a different note, I seem to remember you said you were going to report on the classic Zeiss 7x42 BGAT. I presume circumstances didn't align for that to come to fruition?
 
Thanks Henry for your initial comments. I have been considering one of the 10x Canon IS models to pair with my 7x glass.
Given the quality of these binoculars even disregarding the IS(particularly the L model), I would be interested in seeing Canon come out with a Non-IS binocular to see how it would compare to the current alphas.

Justin
 
Additionally, has anyone compared in particular the old ISII 10x30 and the new ISIII 10x32? The 10x32 is significantly heavier and I am wondering if that is simply due to the (small) increase in objective size, or something else entirely...

Justin
 
jremmons;3815827 The 10x32 is significantly heavier and I am wondering if that is simply due to the (small) increase in objective size said:
Hi Justin,

Something else entirely. The eyepieces are probably unchanged from the old 10x30, but the objective group and the IS mechanism in the 10x32 are completely new. The new objective is much more complex than the achromatic doublet in the 10x30. You can get an idea of the difference by comparing the diagrams in post #1. Just subtract the cover glass and the first lens from the 10x42 L diagram and you have something pretty close to the 10x30 doublet (without UD glass) and its IS prism.

Dipped,

That goes back a long way! I still know the guy with the 7x42 Dialyt, so maybe someday.

Next installment probably won't happen until tomorrow.

Henry
 
Henry,

Excellent review, as usual, thanks! A few questions and comments:
-your corrected vision is 20/15?
-do you use binos w/o glasses? If you do, what's your opinion about Canon eyecups? On the other hand, if you wear glasses then the eyecups should be fine.
-can you please measure the dioptric compensation range?
Is it exactly -/+3d, or a bit larger?
-when you said the "outer 20 deg" did you mean 20%? (that is, quite close to the field stop).
-there is a lot of CA in the Canon image; SE image, which has less CA but more field curvature, seems to be the closest to Canon's.

Peter
 
Last edited:
Hi Peter,

My acuity is about 20/15 without eyeglasses.

The extended eyecups are useless for me. I use them folded down without glasses. That would be a problem for side light reaching my eyes, but happily the 10x32 works well for me with the Bino Bandit neoprene face shield.

Exactly -/+3d is what I measure for the dioptric compensation range using a pair of +3d reading glasses, not very generous

I meant 20º of apparent field. Each circle subtends a little less than 4º of apparent field. The camera lens couldn't quite include the field center on the left without vignetting. Yes, the distortion profile of the Nikon SE is the closest of that group to the Canon, but the Canon is a little better at corrected AMD near the edge.

Henry
 
There have been rather many small changes in the Canon IS binoculars during the last twenty years.
Some were to extend battery life, or maybe to affect the IS tune or working systems.
Maybe some changes were made to make manufacture or tuning easier. The 10x42L was originally supposedly 38mm clear aperture, and extra cover was made to the electronics.
Possibly coating changes also.

As to the eyepiece, there may have been changes, I don't know.
But some had only one field flattener, and some two. There could have been changes to the curves or glass used also.
In general the smaller IS binoculars had one or maybe no FF, and the larger ones two FF elements.

It may be that the 10x30 IS had one or no FF element and later two FF elements. I cannot remember, and Canon doesn't disclose what changes it makes, and even their published information or the published information by sellers may be wrong.

The lens design has always been interesting.
The 18x50 IS had particularly small star images with fainter stars, and when the IS was engaged tiny stars were seen that were quite invisible without IS. The field was also very flat with faint stars seen over most of the field.
I have seen some reviews of different models that claim the IS revealed stars 0.5 magnitude fainter with IS on, But I think that it is at least one magnitude or even 1.5 magnitudes.
In my opinion, the original 10x30 IS in practice revealed stars as faint or fainter than a hand held normal 10x50.

So although it may be that the Canon 10x30 Mk2 has the same eyepiece and FF as the 10x32, this may not have always been the case.
 
Last edited:
Part Two: Chromatic Aberrations


To test Chromatic Aberrations in a systematic way I use a target I designed for the purpose. It’s a long bar, painted black, with strips of white plastic tape placed at regular intervals. I set the distance from binocular to target so that the tape strips are spaced about 2º of apparent field apart. I try to sight the target through the binoculars so that a cross of tape is centered in the field both horizontally and vertically, so that the bar bisects the field. Observing longitudinal CA usually requires boosting the magnification, but lateral CA often starts to become visible at normal magnification around 2-4º off-axis and can be very vivid by 8-10º off axis. Every binocular shows some lateral color off-axis, but some are better than others, particularly near the center where false color is the most troublesome.

The two images below show the target photographed through the Canon 10x32 on the left and a Nikon 10x35 E II on the right. The Nikon, like most old porros with simple cemented doublet objectives, has the expected high longitudinal CA of a fast achromat, but performs quite well when it comes to lateral color. The Canon is similar to the Nikon for longitudinal CA, but is notably poor for lateral color, especially near the field center. If anything the photos fail to reveal the full extent of the lateral color in the Canon when it’s hand held with the IS engaged.

I don’t know for sure why the Canon is so poor, but I can think of a few possibilities. First is the absence of UD glass in either the objective group or the eyepiece. A comparison with the current 10x30 IS II would be instructive since it also doesn’t use UD glass and appears to use the same eyepiece design (another potential source of lateral color). Another possible source is the 10x32's very complex six element objective design with several wide air spaces. The 10x30 has a simple cemented doublet objective like the Nikon 10x35. Finally there is the IS system which is constantly decentering one of the objective elements to stabilize the image. Whatever the source or sources, lateral color appears to be the one really weak performance area of an otherwise very good to state-of-the-art set of optical characteristics.

Next installment: Light transmission, Color Accuracy and Glare Resistance.

Henry Link
 

Attachments

  • DSC_0455.jpg
    DSC_0455.jpg
    39.9 KB · Views: 380
  • DSC_0449.jpg
    DSC_0449.jpg
    39.3 KB · Views: 281
Last edited:
All the photos I made were tripod mounted, but I experimented a few times with engaging the IS. I don't believe the image above was one of those. I found the "Powered IS" setting too "swimmy" to allow accurate positioning of the target in the field and I think it's also a bit worse than the "Stabilizer" setting for hand held lateral color.

Henry
 
Hi Henry,

Thank you for another excellent report! A few quick comments on both part 1 and part 2 of your review:
-I was a bit surprised to see that the SV has some visible astigmatism (much less than FL's but more than Canon's and SE's); I expected the view thru the SV to be virtually free of astigmatism.
-The use of the Bandit Bino shield seems to be a reasonable solution to the eyecup problem of some Canon binos. A possible drawback is that it does not seem to offer protection in the rain, unlike a standard rainguard.
-Regarding the use of * vs % to express off-axis distances, I understand your use of * but I find it a bit ambiguous without info about the AFoV. For instance I believe that the image in your 2nd report covers about 30* (28* to be more exact) of the apparent field, but that does not tell me how far from the center it extends unless I know the AFoV---imo this would become clearer if you gave the distance from center in %, for instance saying that the image ends are at about 50% off axis.


Peter
 
Hi Henry,

Thanks for some splendid work yet again.

One thing strikes me as odd in the Canon lateral CA image: the aberrations closer to the center are purple fringing which is very prominent on the bottom edges of the white crosses, but there's virtually nothing on the top edges, neither purple nor yellow-green. Does this mean that the lateral CA is unsymmetrical with respect to the vertical axis, or that the target was above the middle of the view horizontally?

Were the results equally bad for both barrels of the binocular?

Kimmo
 
I own the old Canon 15x50s and the new14x32s. I find it very odd that I have to look hard to see any CA with the 15x50s whereas in the 14x32s it is all over the place. Center, edges, everywhere...
 
One thing strikes me as odd in the Canon lateral CA image: the aberrations closer to the center are purple fringing which is very prominent on the bottom edges of the white crosses, but there's virtually nothing on the top edges, neither purple nor yellow-green. Does this mean that the lateral CA is unsymmetrical with respect to the vertical axis, or that the target was above the middle of the view horizontally?

Were the results equally bad for both barrels of the binocular?

Kimmo

Thanks Kimmo,

I've been making these images by hand holding the camera behind a tripod mounted binocular. I can't see the lateral color very well through the dim viewfinder, so I just make many images and hope that one will be close to properly aligned. It's much easier to make visual evaluations looking directly though the binos.

In the crop of the center of the Canon image below it looks to me like the true center is really up and a little to the right of the center of the cross. I think the Nikon image is off a little to the left. I noticed that before I posted the images, but I didn't think they were so far off that the comparison was invalided. Visual results were so similar in both barrels that I didn't bother with making photos through the right barrel.

Henry
 

Attachments

  • DSC_0455.jpg
    DSC_0455.jpg
    41.6 KB · Views: 134
Hi Henry,

-Regarding the use of * vs % to express off-axis distances, I understand your use of * but I find it a bit ambiguous without info about the AFoV. For instance I believe that the image in your 2nd report covers about 30* (28* to be more exact) of the apparent field, but that does not tell me how far from the center it extends unless I know the AFoV---imo this would become clearer if you gave the distance from center in %, for instance saying that the image ends are at about 50% off axis.


Peter

Hi Peter,

I probably can't be pried from my preference for using degrees of apparent field rather than percentage, but in future I'll try to make sure I supply enough information so those who prefer percentage can make their own calculations.

Henry
 
Henry,

It is good to see you back on the warpath. It is a pleasure to read your objective and well explained test results.

The combined presence of large lateral chromatic error and extreme air spacing supports the cause and effect relationship between the two. I think testing the earlier 10x30 with the simpler objective but similar eyepiece, and comparing the two, is about as close to the scientific method that you could come in trying to prove or disprove this idea. The idea, as we have communicated privately, comes from the telescope community, and certainly has merit in principle, but has not been demonstrated to be actually significant in binoculars. But until that happens, I feel pretty confident that that is what's going on. That 32mm objective occupies almost half the optical path! That is getting in the ballpark of the case of a chromatic corrector shortly before a telescope eyepiece, which was used for the explanation I read of the idea, and which I suppose must therefore introduce a very noticeable error.

Ron
 
Thanks Ron,

I do wonder why Canon held off from using UD glass in the new models. Maybe a design with all those singlets and air spaces just couldn't be really well corrected and they thought what they did was good enough.

Henry
 
Last edited:
Part Three: Light Transmission, Color Accuracy and Glare Resistance


Subjectively, I found the light transmission and color accuracy to be outstandingly good, particularly given that there are no fewer than 22 glass to air surfaces in the new Canons, the most of any binoculars I know.

The left image below shows a comparison between the Canon 10x32 and a 2016 Swarovski 8x30 Habicht (12 glass to air surfaces). The image was made on a cloudy day by setting up a piece of white foam board facing the northern sky. I made 3 photos of the board: first a straight photo of the board center using a piece of white translucent plastic in front of the camera lens to diffuse the light returning from the board. Then, from the same tripod position as the camera, I made photos through the objective lenses of the two binoculars with the eyepieces pointed at the center of the board, also with the diffusing plastic placed in front of the eyepieces. The final image was assembled in Powerpoint from crops of the three photos. The background is the straight camera photo. The left small square is a crop of the view through the Swarovski objective lens and the right square is the Canon.

In my opinion the Swarovski represents the current state of the art for both light transmission and color accuracy, so I was quite surprised by how small the difference in transmission appears to be between it and the Canon, both subjectively when looking through the binoculars normally and in the Powerpoint image. Gijs van Ginkel has measured the Swarovski at around 95-96% @ 550 nm. The Canon looks close enough to me to be around 90%. The difference in color seen in the image can also be seen in normal use, with the Swarovski appearing completely neutral to my eyes and the Canon looking just a little bit warm/yellow.

The right photo shows the exit pupil and interior of the Canon as seen from the eyepiece. You can see that baffling is extremely well done. There are no significant unbaffled internal reflections and in fact I found it nearly impossible to induce any veiling glare at all no matter how difficult the lighting. This is a particularly valuable quality in a binocular with a dinky 3.2mm exit pupil and combined with the excellent AR coatings makes for an unusually bright high contrast image from such a small binocular.

Next up: Resolution, Star-tests and Axial Aberrations (probably delayed for a few days while I wait for a new artificial star to arrive from Hubble Optics).

Henry Link
 

Attachments

  • Slide1.jpg
    Slide1.jpg
    28.3 KB · Views: 213
  • DSC_0509.jpg
    DSC_0509.jpg
    76.9 KB · Views: 260
Last edited:
Warning! This thread is more than 5 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