Canon 8x25 IS brief test (1 Viewer)

typo said:

... I recently discovered a complication. I had been using anything between 7m and 30m and had a feeling that I was getting better results at longer range but the differences were small. Recently a forum member was kind enough to donate a better chart which allowed me to get down to 2.5m, (which would be great for indoor testing) and it became totally obvious that I was getting poorer resolution results at near focus than longer range with a new acquisition. I was waiting for better weather to retest all my roof binos to see if the phenomenon is more widespread. I'll stick with 10m+ for this comparison.

Click to expand...

I think sticking to 10m+ is a very good idea. It seems quite possible that the image quality of binoculars (including the resolution) doesn't stay the same across the entire focus range. They normally reach their maximum resolution at infinity, and the closer you focus the more the image deteriorates. Tobias Mennle comments on this in his review of the SLC 8x42. He found a marked deterioration of the image quality at short distances: http://www.greatestbinoculars.com/allpages/reviews/swarovski/swarovskislc8x42/swarovskislc8x42.html

Interestingly he didn't find the same deterioration in the Habicht 8x30 (external eyepiece focusing). Is it possible that this is caused by the internal focusing of modern roofs?

Hermann

Hermann said:

I think sticking to 10m+ is a very good idea. It seems quite possible that the image quality of binoculars (including the resolution) doesn't stay the same across the entire focus range. They normally reach their maximum resolution at infinity, and the closer you focus the more the image deteriorates. Tobias Mennle comments on this in his review of the SLC 8x42. He found a marked deterioration of the image quality at short distances: http://www.greatestbinoculars.com/allpages/reviews/swarovski/swarovskislc8x42/swarovskislc8x42.html

Interestingly he didn't find the same deterioration in the Habicht 8x30 (external eyepiece focusing). Is it possible that this is caused by the internal focusing of modern roofs?

Hermann

Click to expand...

I'd previously noted changes in CA, field curvature and pincushion with distance in both a reverse porro and roofs I've owned but I don't think I've seen anything yet that was very obviously softer at close focus, but then I hadn't paid that much attention to near focus performance in casual testing. It was just under an arcsecond worse at 2.5m stopped down than at 30m which is only borderline detectable by eye. I suspect you are right that the focussing element is the likely candidate. Unfortunately I no longer have the reverse porro for comparison.

David

David,

Thanks for posting the first results. What was your impression of using a time limit? This is something I was wondering about, but thought that it would more closely resemble real-life binocular use if one did not allow for unlimited time to see that little glimpse of detail amidst constant swaying of the image.

As for my part of testing, there are supposed to be a couple of cold days next week, and that may mean sunshine also. However, I have one work-related thing I need to get done this week, so time might be a problem.

I dug out some of my notes for tripod-mounted line pair resolution for sharp binoculars taken at my usual indoor testing setup with the standard lighting I use. These amount to about 66" apparent VA at 10 meter distance, the same both for the best 10x and the best 8x binoculars yet tested, so I guess this is not that far from what you report getting. What I haven't tried to do is adjust the distance to get the absolute maximum threshold.

Kimmo

Kimmo,

Today I needed most of that minute to get to 72" but it was apparent in the first few seconds I was close. By the time I tested the Vanguard the light had dropped a fraction and it was obvious I wasn't going to match it. With the chart increment's around 10% per pattern it's tempting to tinker with the distance. It's worth mentioning that early on in the fog my apparent VA was around 110". My eyes are getting old!

Normally with better light conditions then around 65" would be my usual benchmark for a good binocular but my eyesight does vary a bit day to day. On a very good day I have managed 54" but that's rare. Both the binocular stopped down resolution and the the light conditions need to be perfect for that. Stopped down boosted testing is rather more reliable than my eyes alone.

The Dawes limit is 116/D, but with a printed line chart I'm really doubtful that is possible. I've had a few stopped down 20mm results around 6.4" level but my best is 6.2" which would make 124/D as the diffraction limit by the chart method. How does that compare with your results?

David

David,

A printed chart I think is a limiting factor only when you are getting down to the limits of the said chart. In my sample of a printed ES USAF chart that goes down group 3, at least down to half-way group 2 is high enough print quality that I'd be confident it doesn't do anything to the results. So you can study the chart at close up to determine where the lines begin to be excessively thick or thin, and use a distance where you would not need the unreliable range yet. With the vapor-deposited glass slide, one can go as small as one wishes with no reduction in precision.

I have gotten to 116/D with a couple of the very best scopes I have tested, full aperture, but haven't done enough stopped-down testing of binoculars to know where I'd get with those.

Kimmo

Kimmo,

There is no problem with the professional chart I was given recently which is usable down to at least 4,2, but even my previous home print I checked with a measuring graticule and it was accurate enough for the 7m+ I was using.

Obviously a printed paper chart cannot match the contrast of double stars and I supposed that would be impact on the diffraction limit value and that seems to be reflected in my results so far. I imagine a back-lit slide would get pretty close though. Interestingly it was one of a number of points Gerold Dobler made on resolution testing that paper values could not match the DIN ISO standard values but I don't know what that statement was based on.

After a dull and wet morning the sun finally came out at 3:15 and I had about 10 minutes before the sun dipped behind the houses. No time to set a new distance so it was 28.5m again, but the light levels were about double Sunday's level and only just below my optimum.

Tripod mounted, pattern 0,1 @ 28.5m which gives 7.2" (apparent VA 72"+)
Hand Held, -1,4@ 28.5m which gives 10.2" (apparent VA 102"+)

This was actually the same as I recorded Sunday but this time it was en easy read and I was very close to the next pattern for both readings, hence the "+".

Vanguard Endeavour EDII 8x42

Tripod Mounted, pattern -1,6 @ 28m which gives 8.12" (apparent VA 65")
Hand Held, pattern -1,3 @ 28m which gives 11.5" (apparent VA 92")

I said at the out-set that I chose these two as I thought they represented two different levels of steadiness in my hands. Interestingly in both sets of readings the hand held were 42% worse than the tripod mounted for both binoculars, and for the Prime at least, worse than I'd previously measured. I needed to rush about to get the test in which may have played a part. I would still say the amplitude of the shake was higher in the Vanguard yet I was still able to get a relatively better apparent VA, which surprised me.

The Prime probably has the higher and more neutral transmission, but the Vanguard better perceived contrast might have just edged the results in it's favour today. I somewhat doubt the Vanguard's small resolution advantage would have contributed to today's result.

David

Kimmo,

We got some good sunshine at about 1 o'clock today that if anything was a little high for my optimum acuity producing 500 cd/m2 at the target. It's probably the best I'm going to get this time of year and although I didn't quite match my best ever results for these binoculars the results are much more in keeping with previous results in good light levels.

I included my Eden 10x56 in today's test. I know that technically not one of my best and in optimum light conditions it is marginally limiting on my eyesight under test conditions. However that is perfectly adequate in most light conditions, and I think very good in low light. ( A similar argument was put forward by Zeiss in defence of the HT 8x54 )

Zen-Ray Prime 10x42
Tripod mounted, pattern 0,4 @ 21.7m which gives 6.74" (apparent VA 67")
Hand Held, 0,2@ 21.7m which gives 8.5" (apparent VA 85") or 27% worse.

Eden XP 10x56
Tripod Mounted, pattern 0,[email protected] which gives 7.5" (apparent VA 75")
Hand held,pattern 0,[email protected] which gives 9.5" (apparent VA 95") or 27% worse.

Vanguard Endeavour EDII 8x42
Tripod mounted, pattern 0,[email protected] which gives 7.5" (apparent VA 60")
Hand Held, pattern -1,[email protected] which gives 10.7" (apparent VA 85.6") or 43% worse.

On this occasion despite being the Vanguard being the 'sharpest' binocular to my eyes,the hand shake caused a proportionally higher decrease in apparent VA than the two 10x binoculars. I consider the Prime to be slightly steadier and sharper than the Eden and this is reflected in the results. However with the Eden I have learned to counter the increased shake if held normally with a forward and back hand hold, otherwise it would have faired quite badly.

Kimmo, unless you have any further proposals for other comparisons I'll call it a day on this exercise. One thing I've learned is that light levels have a greater role on the affect of hand shake on apparent VA than I've realised. It will be interesting to hear how the Canon compares.

Cheers,

David

I realise many find arcseconds confusing. We can look at it anther way. If we use the tripod value for the Vanguard 8x42 as the reference value for 8x magnification then hand held would be 5.6x. The Z-R Prime 10x42 on a tripod would be 8.9x and 7.1x hand held. The Eden 10x56 would be 8x on a tripod and 6.3x hand held.

Hand held, the Ren-Ray Prime had 25% higher magnification than the Vanguard and produced a 27% more detail in very good light conditions. The Eden on the other hand only produced a 10.5% advantage and slightly less steadiness and possibly apparent resolution could accounted for the difference. The Vanguard was the sharpest but relatively unsteady in my judgement.

In only slightly poorer light conditions the story is more confusing with my reduced visual acuity and possibly transmission characteristics of the binoculars coming into play.

David

David,

You have certainly done your part for now, and it is up to me to answer. I should have more time available in the coming days, and will start the testing.

Another way to offer relative resolving performance information would be to calculate distances to target, which is also easier to understand than arcseconds values. Thus you could say that for seeing the detail you get with the Vanguard tripod-mounted at 100 meters, you would have to view it hand-held at 70 meters.

Kimmo

kabsetz said:

Another way to offer relative resolving performance information would be to calculate distances to target, which is also easier to understand than arcseconds values. Thus you could say that for seeing the detail you get with the Vanguard tripod-mounted at 100 meters, you would have to view it hand-held at 70 meters.

Kimmo

Click to expand...

I'll give it a go.

I was trying to think of a way to convert the line pair value I got with the Vanguard and tripod into something more tangible. If I've done the conversion right it should be about the same as reading 26 point Rockwell Bold at 100m (109.4 yards). The capitals should be about 9.1mm tall.

Vanguard Endeavour EDII 8x42
Tripod 100m
Hand Held 70m

Zen Ray Prime 10x42
Tripod 111.3m
Hand Held 88.2m

Eden XP 10x56
Tripod 100m
Hand Held 79m

David

Kimmo,

To correspond with optimum acuity at 2.5mm pupil diameter I should do my stopped down testing at 20mm for an 8x, 25mm for a 10x etc. I confess I've got lazy and usually only test 20mm irrespective of magnification. It's easier to relate the numbers to my perception of sharpness when comparing the same objective diameter. In this case it does reveal a flaw in this approach.

At 20mm stopped down the Vanguard was 6.2" both barrels, the Eden 8.1" both and the Zen-Ray 6.3" left and 7.2" right barrel. In the Primes case normally the right would be paired with my weaker eye with no impact on overall performance. On a tripod I need to turn it upside down so the weaker lens is now matched with my stronger eye and that shows in the results above. You might note that for both the 10x I was able to get slightly better readings than the 20mm result would suggest. I presume the 25mm resolutions would correspond more directly.

Perhaps I should point out that all three of these binoculars are pretty good and are in the range of effective resolution performance I've seen in certain alpha samples. These small differences I've referred to would be quite undetectable by the majority of users, particularly hand held. I've got a Wow! reaction for all three of them from experienced birders.

David

Kimmo, David, Binoastro;

A little information about ISO light levels for resolution testing.

“The resolution test target shall be illuminated uniformly (± 5 %) by means f a light source with a correlated colour temperature of 5 000 K to 6 000 K, condenser and diffuser for non-dazzling observation conditions. The illuminance of the resolution test target shall be optimum for observation of its image.”

I use a standard light source that can be brought to the 5000 K point by increasing the voltage and setting the output illuminance by means of controllers. The specs for the bulb are shown below. This bulb puts out about 75,300 lux (close to EV=15 or noontime sunlight) at 25 mm, which is fed to a diffuser, the output of which is determined by the port selected, I get a very repeatable lux of 13,500 @ the 50mm aperture and 17,300 lux @ the 36mm port. Neither of these are enough to cause dazzling of the target in my collimator even with the naked eye. I can only see about group 2, 1 or 2, my VA averages about 100”, 90 on a good day. Note this light level is very close to what I see if I put the collimator on a tripod and point it at the north sky on a clear cold day around noon.

I will have to leave the conversion of lux to cd/m^2 to David since I am not familiar with the conversion of different photometric (luminance and illuminance) units. I think that 1 lx = 1 lm/m2 = 1 cd•sr/m2 and a typical f:3 optic is .088 SR and a target at target at 10-20 meters would be considerably smaller than the FOV, I do not know what the cd/m^2 would be but assume about 1500 using Davids ratio of 3-4000 lux=300 cd/m^2 or 17,300 lux times the f:3 SR of 0.088. This level is not enough to cause dazzling of the target though I think it is close enough and repeatable for bino comparisons. I can get dazzling effects with 100 watt bulbs but cannot get the right color temp.

David, let me know what the value of the conversion would be for 75,300 and 17,300 lux, I may not even be close.

About the ISO method and a paper target. The ISO method uses a closed collimator with a diffuse light at the back of the target and direct collimated rays from target to observer. A paper target has a loss due to reflectance and some diffusion of the rays towards the observer.

Thank you Surveyor.
Do you give yourself a time limit for testing?

In real life, for me, instantaneous resolution, say 1 to at most 3 seconds determines whether I spot something new and unknown, before moving on.
How one would measure this I don't know.

With filtered unaided eyes sunspot observations, I see a penumbral size 60 arcsecond sunspot immmediately, less than a second, the Sun being 30 arcminutes, about, diameter.
50" sunspot maybe 3 seconds.
40" 10 to 20 seconds.
34", my limit, 1 or 2 minutes or never.
This is with rested eyes and the filter tilted to optimum illumination level.

There is a great difference between looking and seeing or observing.
Known objects are very much easier to see than unknown objects.

Regards, Binastro

Hello Binastro;

No, no time limit. I do not do any photometrics with optics; I am only really concerned with the difference in the hardware (radiometric) itself. Everyone’s eye/brain interpretation is different to some extend so I quit testing at the exit pupil.

All of my resolution testing is with a collimator and a much boosted image so as to take my VA out of the equation and I only deal with full aperture.

Ron,

As you pointed out to me a number of years ago cd/m2 is a measure of luminance or how 'bright' a surface is and lux is a measure of illuminance or the amount of light falling on a surface. They are not convertible. However if you misuse a light meter and use the lux range to measure surface brightness there is about a 10 fold difference in the units. So 1cd/m2 reads as 10 lux and vice versa.

When I started this stuff I was in the habit of taking illuminance readings of the light falling on the target, whereas now I find illuminance at the surface as a better measurement as direct sunlight or scattered light can result in different luminance values.

Nearly all the information I've found on the subject references Koenig's study in the 1880s where the peak acuity for that single subject was around 200cd/m2 in SI units. According to my light meter (which is very cheap and may not be accurate) I peak at 300-400cd/m2 and with a pronounce decline below this range that has most certainly got worse over the years. There is a more modest reduction above this level with only a pronounced decline over about 3000cd/m2. One thing that I know I need to pay better attention to is the whole view illumination as this better controls the pupil diameter and therefore acuity. In the kind of experiment we've been discussing here placing a 300-400cd/m2 luminance target against a dark background can still result in glare and poor results. Of course if it occupied a high proportion of the view there is no problem. It might be with more uniform illumination I would get a different peak value.

Cheers,

David

PS. The illumination levels recommended for normal acuity testing varies from country to country and may differ with diagnostic practice but the USA normally uses 85cd/m2, Germany 300cd/m2 and Britain 'at least' 120 cd/m2 in standard tests as far as I can work out.

typo said:

The rain we've had for a few days has stopped and after a very foggy start it cleared after lunch and I tried a few readings. The ambient light (needed to contract the pupil) and target illumination (contrast) where way below what I would normally use for testing, but with little prospect of an improvement in the next few days I thought I'd post the results anyway.

I initially used 10.5m, but the light (and results) improved a little to about 2000 lux when I had moved the chart to 28.5m and it's those I'm reporting. I used two binoculars, my Zen Ray Prime 10x42 which I consider one of my steadier pairs in my hands and also my Vanguard Endeavour EDII 8x42 which I don't find so steady. In previous boosted test results the Vanguard gave slightly better full aperture and stopped down resolution results than the Zen Ray, but I would not expect to see any difference at these light levels.

Zen Ray Prime 10x42

Tripod mounted, pattern 0,1 @ 28.5m which gives 7.2" (apparent VA 72")
Hand Held, -1,4@ 28.5m which gives 10.2" (apparent VA 102")

I should say that I felt I was struggling with the light levels and taking longer to decide a reading than I normally would, though still kept within the minute Kimmo suggested. The Tripod mounted apparent VA is about what I expected but the hand held is worse. This works out as a 42% decrease in apparent VA which is a rather larger difference than I've previously seen for this binocular.

Vanguard Endeavour EDII 8x42

Tripod mounted, -1,4 @ 28.5m which gives 10.2". (apparent VA 81.6")
Hand Held, -1,1 @ 28.4m which gives 14.5" (apparent VA 14.5 116")

This again works out as a 42% decrease in apparent VA which is only a little more than I've seen previously. The light had dropped a little more by the time I did this test.

I intend to repeat this, if or when we get better weather.

David

Click to expand...

I did a series of tests some time ago and found a 35-45% reduction in
the font size I could read at 100ft with tripod. I don't think there is any
question about the gains, whatever the illumination or chart. It's the
big improvement that matters, not a few percent. Who counts
hairs at the abattoir? It's real, of course....stabilizers on traffic
copters are glaring proof.

If you strap on a laser to your barrel and watch the dot move with
your shaking, you will see that your cortex removes almost all the low-
frequency shaking. This applies to stabilized binoculars because, while
they may still swing around a bit, the high-frequency wiggle has been
wiped out....so they should be almost the same as with-tripod.

I noticed Sancho's comment regarding compact binoculars and that he didn't like the Canon 8×25 IS.
I agree that these binoculars are plasticky and maybe not so well built.

I don't have the new build 8×25 IS possibly made in April 2014.
However, I do still have the old 8×25 which is possibly about six years old and where the optical window fell off and I have still not repaired it.
With the optical window in place but not retained by the plastic front retainer I just tried it out.
I'm just able to make out the pyramid tiny pimple on the top of the chimney block 400 feet away. But the detail was more difficult than the new 8×25. It is, however, a dull day and when I tested the new 8×25 it was a bright sunny day, so my eyes might be working at 3.1 mm today and only 2.5 mm or even 2.0 mm when I tested the new one.

The stabiliser on the old binocular is not nearly as good as on the new one, and it is jerky. I also have moments of slightly out of focus images with the old one and then it gets back into focus. The new one could be seen to be working with very small amplitude movements.

Looking into the front of the old binocular it seems that maybe the whole front cell moves when you press the stabiliser button. At least the front objectives move and I think actually the whole front cell moves. I don't know if this also has a third element behind the objective. In the new 8×25 I suppose that the third element could be a flat optical element although I suppose it could be curved.

I think I read recently that the new 8×25 employs a new tilt mechanism.

I don't think that either the new or the old have variable prisms but I'm not privy to this information.

I remember that when I tried the older binocular perhaps six years ago I didn't think there was much improvement with the stabiliser on.

However, the new Canon 8×25 IS really surprised and impressed me.
I was testing it against a 10×42 Zeiss Conquest HD and although the larger binocular had a much brighter image even in bright sunshine when the whole aperture would not be used, the Canon 8×25 resolved very much finer detail when the stabiliser button was pressed. This detail was on a much smaller scale than with the 10×42 Zeiss conquest HD but it was fully resolved and could be seen the whole time I pressed the stabiliser button. With the 10×42 Zeiss Conquest HD I could only see this fine detail perhaps 40% of the time. Of course it depends how steadily I can hold a binocular, but I think I am at least average. And I regularly use 12 times standard binoculars handheld.

In other words, the new 8×25 easily outresolved the 10×42 Zeiss Conquest HD.
To see this tiny pimple 400 feet away is for me very difficult with a 10 times binocular.
I had thought that it would be impossible for me to see this with an eight times binocular, but it is now clear to me that a tripod mounted or stabilised eight times binocular easily shows it if it is of good optical quality. In fact I think that a tripod mounted or stabilised six times binocular might just show this tiny detail to my eyes if they are well rested.

I cannot really give a full assessment as the old 8×25 is well used, but I would think it's probably working as well as when it was new as far as the stabiliser is concerned.

It would seem that the new Canon 8×25 IS is a considerable improvement on the old version as far as the stabiliser is concerned. And I think the stabilising method has changed.
Unfortunately I could not compare the old and the new side-by-side as the new one has gone back to its owner.

Binastro,

I seem to remember we did a rough calculation on that pimple on the chimney pot a while back and came up with something like 15 arcseconds. 8x magnification would make that 120 arcseconds. If I understand what you said correctly then the IS makes a difference to how distinct the pimple is. I know an ink dot on white paper is not the same thing, but I made a range of spots of different sizes with a very fine marker pen and viewed them hand held and braced with my 8x.

The 0.9mm spot (16.9" x 8 = 135") was very clear hand held. The 0.5 spot (9.4" x 8 = 75") was clearly visible most of the time, disappearing briefly every heart beat hand held, but was quite distinct when braced. The 0.4mm spot (7.5" x 8 = 60") had less ink and more a mid grey rather than black and was only visible 10-20% of the time hand held and probably still under 50% of the time when braced. The bracing appeared to have eliminated the major oscillations not the micro ones. It's worth mentioning that the surface brightness was 1500 cd/m2 at midday and way too high for my best results but I thought you might be interested all the same.

It's 18 months since I last tried the Canon IS range so I'm struggling to recall what if any benefit I noted at the time. My best recollection was that some models, but I don't remember which, appeared to dampen the heart beat oscillations but seemingly not the micro ones so might have improved the 0.5mm results which should be obvious, but maybe not the others. All wild guessing of course.

On certain days we get the Luton flight path routed a mile or two from us, but I don't know the exact distance. Hand held the EasyJet lettering on the tail was very readable, but maybe a little fuzzy. It was sharper braced.

I don't know if this helps any?

David

I found a map of the flight path, and given the angle of view, I was observing the plane at a range of 2.5 to 3km or 1.55-1.86 miles.

Hi David,
I thought that the pimple was smaller, but I will photograph it and get the accurate size.
It is shaped like a small pyramid and is part of the top chimney brick. I think that British bricks may be about 9" x 4.5" but I will check. The building must be about 1930s or earlier and is a mansion block. Birds like sitting on the chimney, which is about 80 feet above the ground.

Unfortunately it is now cloudy and dull. I just went for a walk hoping to view the sunspots with a filter and although this morning was very bright and sunny I could not view the Sun today, except in the morning with the H alpha telescope.
There has been a protected unaided eyes sunspot for several days now, about 39 or 40 arc seconds penumbral size. It is actually composed of two close subvisual sunspots which are seen as one to my eyes. With the moons of Saturn and Jupiter I found that if two subvisual moons were separated by less than five times the Dawes limit I would see them as one small visual moon. I think I probably see about the same with two subvisual sunspots with protected unaided eyes, which can then be seen as one. I.e less than 5x Dawes limit apart.

When I was out I took photographs of the crows flying around.
What was very strange was a Boeing 747 flying at about 220 kn with half flaps down and at about 6000 feet altitude. It seemed to be flying sideways so I assume the crosswind was around 50 kn or more at that height.

You don't say what binocular you were using to read the Easy Jet name.

The one problem with the Canon 8×25 IS is that it uses a camera battery and although I have some unused ones they are quite old and I will have to buy some new ones. They cost around 4 pounds although you can get them cheaper from somewhere like Mathers of Lancashire, I think. I used to buy ten at a time for cameras but I don't need them nowadays. I think they were mainly used in film cameras.

I remember buying a daft tiny handheld television that was made by the British inventor and he used a Polaroid battery, which was typical of him. The successful Japanese ones used ordinary batteries.
I remember having a white small Cambridge calculator, which used for memory, reverse Polish logic.
Then he invented the daft electric tiny vehicle for driving around on the pavement.
Sinclair maybe?

I might be able to photograph the pimple but the light is very poor today.

The pimple was invisible with 8x25 IS unstabilised. Continually, easily visible with stabiliser on (2014 new version, more difficult old version).
With the 18x50 stabilised a lot of detail is seen in the pimple and no ordinary handheld binocular comes close.