Review of Canon 10x32 IS (1 Viewer)

Henry,

A big thank you for your detailed assessment of the Canon. :t: It sounds like a good one. For some reason, the IS stability of It's predecessor did not work well for me. Do you have any thoughts on how the stability performance might compare?

On the practical side, some time ago I did attempt to copy your photographic approach to illustrating colour and CA. Obviously the camera in auto mode was a disaster due to exposure and white balance correction. Even my attempts in manual and raw mode were confounded by fluctuation in luminance and spectrum of the natural light. Any tips on getting round the problem? In raw, I also realised I was also confusing of the CA of the camera lens with that of the binocular. I suspect the answer to that is a better camera! Any help appreciated.

David

Thank you Henry for your detailed and careful review.

I have always thought that Canon were using their very good lens designers to design the Canon IS binoculars from the start over 20 years ago.

In my experience, with perhaps ten, various Canon IS binoculars I have found performance to be very good except for false colour, which problem is compounded by the IS constant variation. With the Variprisms near their maximum offset all sorts of star images presented. The only lens system I have is the later 8x25 IS.
But still, the ability to show tiny faint stars over most of the field, and aircraft windows at night from far off aircraft was very impressive for a hand held binocular.
I still think that the Zeiss 20x60S is king, but a beast unless one is strong and fit.

The objectives in the 10x32 IS may have more in common with camera lenses than typical binocular objectives.

The reason why more Canon IS binoculars haven't been sold is that birdwatchers are rather conservative and the Canon IS binoculars have drawbacks. They are complicated, they have electronics, they need batteries for the IS too work. They are heavy and bulky compared to standard binoculars. The warranty is very short, similar to cameras and lenses and computer devices.

Also the IS is variable from one binocular to the next, and may have defects.
When I need fine resolution and for difficult observations I use a Canon IS binocular. For other observations various standard binoculars.
And where the Canon IS binoculars are not powerful enough I use a spotting scope. If that isn't good enough then I use an astro scope.
If that isn't good enough I accept defeat and go and watch the T.V. or make a cup of tea.

typo said:

Henry,

A big thank you for your detailed assessment of the Canon. :t: It sounds like a good one. For some reason, the IS stability of It's predecessor did not work well for me. Do you have any thoughts on how the stability performance might compare?

On the practical side, some time ago I did attempt to copy your photographic approach to illustrating colour and CA. Obviously the camera in auto mode was a disaster due to exposure and white balance correction. Even my attempts in manual and raw mode were confounded by fluctuation in luminance and spectrum of the natural light. Any tips on getting round the problem? In raw, I also realised I was also confusing of the CA of the camera lens with that of the binocular. I suspect the answer to that is a better camera! Any help appreciated.

David

Click to expand...

Hi David,

I'm afraid I don't have access to any of the old Canon models. Maybe somebody with both old a new ones, like Wolf Beam or Canip, can give us their impressions. I do have an opinion about the two IS modes on the 10x32. I find the Staliizer much easier to use. I'll address that in more detail in the last installment, which looks like it will be coming up sometime next year at the rate I'm going.

As for the other question, are you talking about the Powerpoint image from Post #20 or the CA target images from post #10?

Henry

Thanks Henry,

It's more the #20 colour/luminance issue that's been taxing me. Whenever I've checked out the ambient light levels, I've been surprised at how much the luminance and illuminance can shift in the time frame needed to do this kind of test. While I was writing the previous post I stopped to do another check. Even though the light looked totally steady to my eyes, the meter told me the luminance had actually oscillated between about 500 and 700cd/m2 over a two minute period. It can be worse than that. I've no way of measuring it, but the spectrum will shift in a comparable way. I think the answer must be to use artificial light, which has it's own set of problems, but haven't got round to trying to rig something up. Any thoughts?

David

David,

I've never had a problem with recording consistently repeatable color differences between the binocular outputs and the white background when all are photographed together in the same exposure. In that case there is no shift in light over time, but there has been a problem with getting even lighting over the surface of the board which can throw off the difference in brightness between the crops of the binocular outputs and a crop of the straight camera view in the same photo. I think the diffusing filter I tried in post 20 is not the best solution. It did appear to even out the reflection from the white board, but at the cost of introducing a time lag from multiple exposures. A lab quality light source with stable voltage would be ideal, but it's too much money

Your question got me thinking about eliminating the multiple exposures by using several diffusing glass discs (available from Edmund Optics) in a single exposure. The idea is to cover the binocular eyepieces with diffusing glass combined with a third disc (mounted in a tube to shade it from direct light), all photographed together pointing at the same spot on the white board. If I can get that together I'll post a redo the image in post 20.

Henry

Henry,

Sounds an interesting idea.

Unfortunately I no longer have the image analysis program I knew and trusted, but for what It's worth, the colour analysis app on my tablet gives the R:G:B values for your Habicht as 200:198:203 and the Canon as 198:194:191.

Late this afternoon the sky here was virtually cloudless. A rare event for the UK. I first set up a white photo target in direct sunlight. The results were better than the other day but I was still getting about a 15% fluctuation in luminance over the course of two or three minutes. When I put the target on the ground and shaded from direct sunlight there was no detectable variation on my metre over a similar time frame. Of course the light was pretty blue from the Rayleigh scattering, but significantly more stable as a light source.

I recall reading some scientific papers a while back that used what I recognised as as a very common 12v auto sidelight bulb for their spectral studies. The power supply was calibrated I think, but I guess a low cost, plug-in regulated power pack might be adequate for our relatively humble needs?

David

Henry:
A nice technical review, and how about your thoughts about handling and how you like them
in use.

Jerry

Hi Jerry,

I'll have summary of my personal likes and dislikes at the end.

Henry

Part Four: Resolution, Star-tests and Axial aberrations

I had planned to cover the aberrations revealed in star-testing and resolution measurements in the same post but I’m a bit stumped by some of the oddities in the star-tests and photographing them is proving to be so difficult that I’ve decided to report the resolution measurements now and add star-test information when that’s finished.

As usual I used a USAF 1951 glass slide placed a little over 10 meters from the binoculars. At full aperture the resolution of both sides was about 4.45”. In my experience that’s a solid enough result for a 32mm binocular, but a bit behind the best performing 32mm I have here (Nikon 8x32 SE, 3.9”).

I also measured resolution with the objectives stopped down to 25mm to simulate the effective aperture imposed by the eye in bright daylight. That dropped the raw resolution to about about 5”, which actually indicates an improvement in the optical quality from 142.4/D to 125/D (where D is the aperture in millimeters). Hand holding the binocular, using either IS mode, I find that I can resolve line pairs on the USAF chart to about 8.8”, the same as when the binoculars are tripod mounted.

Henry Link

Henry,

Thank you for the resolution results. Food for thought!

I think we established a few years back that your results are significantly different to mine for some reason, (possibly the choice of target?) so I'm not sure if those are good or bad results. I remember you reporting a result of 110/D. Wouldn't that make the 125/D stopped down result quite poor by comparison?

In that same thread I reported an apparent acuity of 85" with a 10x42, hand held, but again we are probably not comparing like with like.

Hi David,

I'm not sure why my resolution measurements tend to be higher than yours. Possibly because the glass slide I use can be set up to have very high contrast between the darkly shadowed black bars of the glass slide back lit by as bright a white surface as I can manage from either sunlight or by placing a halogen task lamp close to a white board. The extreme contrast of the target is definitely helpful to me for resolving the tiniest line pairs at the very high magnifications and small exit pupils I use.

To check myself I went back and remeasured everything about setup and introduced a reference telescope. I found the measured binocular to target distance was too long by about 1% and what I had thought was a 25mm mask was closer to 26mm. Off course there is always a 12% gap between the size of adjacent elements on the chart, so unless the the binocular to target distance is constantly adjusted the true resolution is likely to fall somewhere between elements. I've come to consider about 116/D as a perfect score on the USAF 1951 in my set-up, so I think if I ever reported 110/D there must have been a 6% error of some sort.

For a reference scope I used a Takahashi FC-50 Fluorite APO fitted with the same 26mm mask as the binoculars. Equal magnification of 80x was used for all three instruments.

With the 26mm mask in place I measured both the Tak and the Nikon at 4.48" or 116.5/D. The Tak showed a little sharper image, with essentially zero chromatic aberration. The Canon was one element off at 5.02" or about 130/D using the new 26mm aperture value and had more longitudinal chromatic aberration than the Nikon. I would say the Tak performed as expected for a Fluorite doublet stopped down to about f/16, the Nikon performed superbly well for a binocular with fast achromatic optics and the Canon was about average for a binocular, not good enough for any bragging rights nor bad enough to be considered a lemon.

I've given up on good photos of star-tests, so Ill put together what I have and try to wrap this up in the next few days.

Henry

Henry,

I did try to send you a PM explaining my questions and concerns, but unfortunately it bounced.

I'm hesitant to post the link, as some of the ensuing discussion became somewhat nonsensical , and is probably best forgotten, but your 110/D result was for an AP Stowaway. In the same thread Kimmo claimed 106/D for his Leica 8x20. I've always suspected that these bizarre results were most likely due to your use of a backlit, negative USAF 1951 target.

Many years ago I was told that that the original 1951 military manual contained quite rigid regulations on the nature of the light source, it's luminance and spectrum, but I don't known if positive and negative targets were mentioned. It would make little difference to a camera, (at the correct exposure), but scientific reports tell us that the eye adapts differently to positive and negative targets. It significantly alters parameters like visual contrast sensitivity. How much will depend on the luminance, and presumably the exit pupil. Unfortunately the publications don't predict what would happen in this case. I don't have a negative USAF target to compare, but have seen spurious patterning at very high luminance when I've tried to mimic one. It might be a plausible explaination for those very odd results. Perhaps something to be aware of. Personally, I will stick with a positive target.

I've got as close as mentally debating whether a result might be closer to 115/D or 116/D ( by juggling the distance), but never better than that, so our current results now appear to be more closely aligned. The 130/D for the Canon should be a quite acceptable resolution for the large majority of users, even with the IS engaged. Can't help hoping for better though.:-O

I known you have set great store by your star tests in the past, but my experience has been rather the opposite. The Airy disks have been normally highly distorted in the binoculars I've checked, even on one binocular that was borderline 115/D. I can't find a relationship with effective optical performance myself, but will try to keep and open mind on the subject.

Thanks again for the interesting data.

David

David,

FWIW, I think what I have is also a positive target, at least that's the way it's labeled by Edmund Optics (opaque bars and numbers deposited on transparent glass).

I guess I'll have to remain an advocate for using a very bright light source for the background in these high magnification tests, either sunlight or as close as possible with artificial light. Whatever the Air Force thinking was about light levels for their purposes it seems appropriate to me to use a source with a reasonably repeatable light level and color spectrum corresponding to the field conditions the optic will be used under. As a practical matter it's also useful to start with a very bright source since it will wind up looking pretty dim at very small exit pupils anyway.

It's certainly true that choosing a lower light level will affect the results. Just now I repeated measurements for the Tak FC-50 and the Canon 10x32 with the halogen lamp pulled back and slightly turned away from the white board behind my target. For my eye both suffer a loss of resolution at the lower light level, one element worse for the Canon and two worse for the FC-50. The Tak image looks "better", but it still can't resolve (or more properly my eyesight can't resolve) the next smaller element. If these results are to be believed both are now rather poor 147/D telescopes. Surely this has to be in inaccurate representation of the actual resolving power and the difference between these two telescopes, caused by the eye receiving insufficient illumination at high magnification. One is, after all, diffraction limited and essentially aberration free while the other has obvious chromatic and spherical aberrations revealed by a star-test. I guess we'll get around to our differences on that subject next.

Henry

Henry,

Seems I misunderstood you about the negative target. The mystery over the 110/D result deepens!

Naturally, there is the risk of errors with both excess and too little light.

The US eye test should be conducted with a chart luminance or 300cd/m2, which would typically correspond to a 2.5mm pupil. From my own tests this seems to be a reasonable guide to work from. As you reduce a binocular (or telescope) exit pupil below 2.5mm through boosting and/or stopping down, the target luminance appears to need a proportional increase in luminance to maintain retinal illuminance (trolands). So if I boost an 8x42 seven times, the EP will be 0.75mm and the change in effective EP area will need a luminance increased of 11 fold or ~3300cd/m2 (compared to the 2.5mm eye pupil) to compensate. If that binocular is stopped down to 20mm as well it should be almost 50 fold brighter at nearer 15000cd/m2. Fortunately things probably don't need to be too precise, but I like to get within +/- 30% if I can. Some days It's just not possible.

Unfortunately it's pretty difficult to guess luminance without a metre. Just to give you an idea. A piece of white paper 4" from a regular 40w incandescent light bulb gave me 300cd/m2. 4" from a 40w halogen spot gave me 2000 cd/m2. Frosted glass 2" in front of a 200w halogen tube gave me 20000cd/m2. A sheet of white paper in direct sunshine has given me readings well over 50000cd/m2. It is possible to calculate target luminance using camera settings. I can try to track down the formula again if you are interested?

Look forward to round 4, or is it round 5? :-O

David

P.S. The equation relating camera setting to luminance is about half way down this Wikipedia page.
https://en.m.wikipedia.org/wiki/Light_meter

The K value for different makes of camera is given in the following section. Don't forget to use spot metering.

Hi Henry,
I enjoyed reading this. I especially enjoyed reading about your CA test. I may have missed it(probably did) but at what distance from the binocular did you set up the white/black bar? I may try this myself.

I just picked up the 10x30 IS II, which I know is not quite the same quality as this 10x32, but man are they interesting. Ease of view isn't the best due to the small EP, and Chromatic Aberrations are occasionally very poor (is this due to the way the lens adjusts?), but in terms of resolving fine detail these are very nice; additionally, once the IS settles in, eye-strain is greatly reduced.

I'd not buy them as my only binocular, but for my intended uses (waterfowl, eagle nest monitoring, general winter bird counts when foliage is gone) they seem to have a lot of value.

Justin

Hi Chuck,

My target is 72" long with the centers of the 1/2" white plastic tape strips spaced 2" apart. I want the 2" spacings to correspond to about 2º of apparent field for every binocular tested regardless of magnification. The distance to the target doesn't have to be known as long as the AFOV is accurately known. For instance, if I know a binocular has a 64º AFOV I just move the tripod back and forth until I find the distance where 32 white bars bisect the binocular field and the cross bar is centered. Of course this means a 10x binocular has to be 25% farther away than an 8x binocular to maintain the same 2º increments.

I think a target scaled down to half the size of mine should work just as well and require less space.

Henry

jremmons said:

I just picked up the 10x30 IS II, which I know is not quite the same quality as this 10x32, but man are they interesting. Ease of view isn't the best due to the small EP, and Chromatic Aberrations are occasionally very poor (is this due to the way the lens adjusts?), but in terms of resolving fine detail these are very nice; additionally, once the IS settles in, eye-strain is greatly reduced.

I'd not buy them as my only binocular, but for my intended uses (waterfowl, eagle nest monitoring, general winter bird counts when foliage is gone) they seem to have a lot of value.

Justin

Click to expand...

That pretty much sums up my experience with the 10x32. An effective tool thanks to the IS, but thanks to the tiny exit pupil and the excessive lateral color often an unpleasant thing to look through.

Henry

Thank you for this excellently comprehensive review.
Is there any chance that you might be able to give Canon's flagship 10x42ISL a similar workout?

henry link said:

Hi Chuck,

My target is 72" long with the centers of the 1/2" white plastic tape strips spaced 2" apart. I want the 2" spacings to correspond to about 2º of apparent field for every binocular tested regardless of magnification. The distance to the target doesn't have to be known as long as the AFOV is accurately known. For instance, if I know a binocular has a 64º AFOV I just move the tripod back and forth until I find the distance where 32 white bars bisect the binocular field and the cross bar is centered. Of course this means a 10x binocular has to be 25% farther away than an 8x binocular to maintain the same 2º increments.

I think a target scaled down to half the size of mine should work just as well and require less space.

Henry

Click to expand...

Thanks Henry!