-
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.
Encounters with Zeiss 8x42 HT and SF (1 Viewer)
- Thread starter henry link
- Start date
Tobias Mennle
Well-known member
Just had a closer look at the 8x42 SF distortion again. It does not have proper pincushion to 75% of the field, but lines will slightly bend towards the center, not the edge when reaching the outer field. It´s a "wavy" distortion, whatever the exact term might be. Outer field looks pretty rectilinear to me.
The 8.5x42 SV - as far as I remember - is real pincushion over most of the field and then going to rectilinear.
That explains the big difference in RG between the two. But then, the SV 8x32 may successfully compete with the Zeiss for the title "worst pseudo RG effect".
The 8.5x42 SV - as far as I remember - is real pincushion over most of the field and then going to rectilinear.
That explains the big difference in RG between the two. But then, the SV 8x32 may successfully compete with the Zeiss for the title "worst pseudo RG effect".
henry link
Well-known member
Thanks Tobias, I also enjoyed reading your review of the same binoculars.
Yes, I used a 64x star-test, but that's not hard to do if you have two 8x binoculars and a tripod. Only the tested binocular absolutely requires tripod mounting. You can hold the other one up to the eyepiece steadily enough to see the basic out of focus diffraction patterns.
What I saw that made me say "overcorrected" was a set of well defined rings outside of focus and no rings inside of focus, the opposite of what most binocular show. Looking more closely at the FL yesterday I see that the situation is more ambiguous. The overly bright outer ring and overly bright central spot are on the sides of focus they should be for undercorrection. So, what seems to be unusual about the FL/HT is the strong pattern of well defined rings outside of focus rather than inside.
The 64x star-test is also helpful in distinguishing "aberrations resulting from sloppy manufacture and aberrations resulting resulting from sloppy design". Since you're essentially testing two specimens of a single telescope design any problem that's confined to only one side has to be sloppy manufacturing. Something that looks the same in both the left and right telescopes is a candidate for sloppy design. Confidence that the design is at fault increases with more specimens showing the same thing. My star-test results for spherical aberration in the 8x54 HT were corroborated by another BF member who star-tested a different pair, so that's four specimens of the same telescope with the same aberration. I think that's pretty good evidence for design rather than manufacture, but of course more samples would make the confidence higher.
Henry
Last edited:
John Dracon
John Dracon
Henry - Would the glass composition, i.e., lead v. non-lead (eco glass) yield anything your 64x test would differentiate? Or is is essentially design driven?
John
John
Henry,
I had hoped for your report on the 8x42 HT for a long time now, ever since you led the discrediting of the 8x54 HT, while James and others here championed the 8x42 mightily. The inconsistency, although across different specifications, was so striking that I didn't know what to think. It appears now that it's mostly confirmed: The 8x54 HT suffers from easily noticeable (that is, bad) spherical aberration while the 8x42 sibling is pretty much up to Z-snuff (although no match for the larger aperture FL). This is surprising, since I see no fundamental difference between my 42 and 56 mm FLs, so correct me if I don't have that right.
Nice place, the Outer Banks. I spent a few days there in the early 90s and had a fine time, but we got out just before a hurricane hit. And too bad about the Passat, I guess you mean the neat old boxy model.
Ron
I had hoped for your report on the 8x42 HT for a long time now, ever since you led the discrediting of the 8x54 HT, while James and others here championed the 8x42 mightily. The inconsistency, although across different specifications, was so striking that I didn't know what to think. It appears now that it's mostly confirmed: The 8x54 HT suffers from easily noticeable (that is, bad) spherical aberration while the 8x42 sibling is pretty much up to Z-snuff (although no match for the larger aperture FL). This is surprising, since I see no fundamental difference between my 42 and 56 mm FLs, so correct me if I don't have that right.
Nice place, the Outer Banks. I spent a few days there in the early 90s and had a fine time, but we got out just before a hurricane hit. And too bad about the Passat, I guess you mean the neat old boxy model.
Ron
henry link
Well-known member
I would say no to a star-test revealing anything about the contribution of lead or non-lead glass to aberrations, but it well show assembly defects as well as design driven aberrations.
Henry
henry link
Well-known member
BOXY? NEVER! It was a 1999 Passat. I recall a reviewer describing its styling as consisting of "harmonious slow curves". A couple of years later the poor Passat was tarted up with fancy taillights, chrome and a menacing snout. It was all downhill from there.
Henry
Hi Henry
So sorry to hear about your Passat. If its in good shape otherwise and it suits your needs it might still be a good solution to replace the processors even if you won't get your money back on selling or trading later.
Lee
henry link
Well-known member
Thanks Lee. I considered doing just that, but my mechanic advised against it. He thought what had been diagnosed initially might only be "the tip of the iceberg" and the mildew smell would never go away. The event turned out to be covered by the comprehensive part of our auto insurance. The insurance company declared the car a total loss and made us a very generous offer we couldn't refuse, so the old Passat is now sitting in a junkyard.
Henry
Henry
Inspired by Hery's earlier work and report on the 8x42 SF, I took a few photos through a 10x42 SF. These are taken with the binocular mounted on a tripod, eyepieces towards and perpendicular to a board that has 1x1 cm square pattern on it. Distance from eyepiece to target is 15 cm. Photos are taken with a Pentax K5 with an F 1.8/50 mm prime lens. Quality of photography is not high, and illumination is pretty uneven to begin with. For comparison, the third photo is through a Canon 10x42 L IS, a binocular with a different distortion pattern but even less sharpness fall-off towards the edges than the SF has.
Since these pictures are taken backwards through the objective, the distortion pattern is reversed. What we here see as barrel distortion is pincushion when viewed through the binocular in the usual way, and vice versa.
Kimmo
Since these pictures are taken backwards through the objective, the distortion pattern is reversed. What we here see as barrel distortion is pincushion when viewed through the binocular in the usual way, and vice versa.
Kimmo
Attachments
henry link
Well-known member
Kimmo,
Thanks very much for posting the photos. I didn't have access to a grid pattern on the Outer Banks, so I'm happy to see images that confirm the mustache distortion in the SF.
I think the Canon image is just as interesting for what it tells us about distortion vs. edge sharpness. There seems to be a pervasive misconception that only binoculars with very low pincushion like the Swaro SV can have high edge sharpness. The Canon is a good example of how high edge sharpness can be combined with any amount of pincushion distortion. Other companies could follow that pattern if they choose to.
I found a couple of old threads in which the same grid technique was used to make images of the distortion in the Swaro 8.5x42/8x32 SV and a few other binocular models. Grid photos don't appear until post #22 in the very long second thread.
http://www.birdforum.net/showthread.php?t=175077
http://www.birdforum.net/showthread.php?t=237219
Henry
Thanks very much for posting the photos. I didn't have access to a grid pattern on the Outer Banks, so I'm happy to see images that confirm the mustache distortion in the SF.
I think the Canon image is just as interesting for what it tells us about distortion vs. edge sharpness. There seems to be a pervasive misconception that only binoculars with very low pincushion like the Swaro SV can have high edge sharpness. The Canon is a good example of how high edge sharpness can be combined with any amount of pincushion distortion. Other companies could follow that pattern if they choose to.
I found a couple of old threads in which the same grid technique was used to make images of the distortion in the Swaro 8.5x42/8x32 SV and a few other binocular models. Grid photos don't appear until post #22 in the very long second thread.
http://www.birdforum.net/showthread.php?t=175077
http://www.birdforum.net/showthread.php?t=237219
Henry
Last edited:
Henry,
Thanks for linking the earlier threads, that makes it much easier to compare images. When I look at your photos of the 8x32 vs 8.5x42 SV's, I'm all but certain that the left-hand image is the 8.5x and the right-hand image the 8x32. Two reasons for this assumption: Firstly, the right-hand image shows a larger field (more squares across) and I'm assuming that you had the target and tripod in more or less the same place for both photos. The second reason is that the image I'm assuming to be the 8.5x SV image looks all but identical to the one you posted in the first linked thread, while the right-hand image has some pretty pronounced additional twisting of the lines in the last 5% or so of the field edge.
Now as an addition plus commentary on your comment on edge sharpness vs pincushion, some data on off-axis resolution of the 10x42 SF and the 10x42 Canon. I measured these at 10 meter distance, where the true field of the Zeiss measured 121 cm and the Canon 111 cm. Incidentally, these fields correspond just about exactly to the specified figures, certainly well within measurement errors.
My procedure here was to have a binocular on a tripod, center the resolution target (Edmund 2" glass slide), view with both eyes, and focus for best resolution in the center. Both binoculars just barely resolved USAF group 1 element 5 with my eyes. Then I'd leave the focus where it was and check resolution at 10, 20, 30 etc centimeters off axis by tilting the binoculars up 10 cm at a time, so the bottom of the view field was used for the edge of field figures. This way, I can use both eyes for each reading, and don't have to move my head much relative to the binoculars as would be necessary if I'd go to either side.
Results: 10 cm off axis still 1/5 for both binoculars; 20 cm off axis 1/4- for both binoculars; 30 cm off axis 1/2 for both binoculars; 40 cm off axis 1/1 for Canon, 0/6 for Zeiss; 50 cm off axis 1/1 for Canon, 0/5 for Zeiss; 60 cm off axis 0/2 for Zeiss (this is just about the extreme edge). Canon does not reach 60 cm off axis due to its narrower FOV, but at its extreme edge of 55 cm it still resolves 1/1.
So, in this test, the two binoculars are equal and very sharp for half of their view field, after which the Canon begins to slightly pull ahead, being significantly sharper at the extreme edge. However, the edge performance of the Zeiss is very respectable and in my experience only bettered by the Canon and Swaro SV 10x42 & 50 models.
About the distortion patterns, to me the SF pattern looks most like the 8x32 SV pattern, while the Canon pattern looks very much like the 8x42 SLC HD pattern in your first linked thread. If you let your eyes wander over these images, you can actually get a bit of a sense of the feeling the images give when viewing.
I must say that I'm so used to the Canon view by now that my brain probably compensates for any and all distortions it has, but to me it has no rolling ball and only very mild pincushion bending of straight lines off axis. I don't see nearly any angular magnification distortion in it even when I specifically look for it by moving round or rectangular shapes from center to edge of field.
In the Zeiss, for my eyes, the AMD at the outermost 5-10% of the field is very pronounced, and there's quite a lot of RB when panning. This is more noticeable at short distances than long, and although I see it readily, it does not bother me very much and I'm sure that personally I could get used to this view just fine. However, the Canon example shows that equally good or even better edge performance is possible without AMD and RB as long as some pincushion is accepted. Another possible compromise concerning the Canon design is external dimensions of the eyepiece lens system. I don't know enough about optical design to know what the possibilities are, but at least in the Canon eyepiece, there is a very large diameter lens inside the e.p. behind the eye lens, and this necessitates very large eyepiece tubes which are ergonomically problematic. Already at the 6.5 degree view field, I think they are at or above the size limit that works well with binoculars.
Kimmo
Thanks for linking the earlier threads, that makes it much easier to compare images. When I look at your photos of the 8x32 vs 8.5x42 SV's, I'm all but certain that the left-hand image is the 8.5x and the right-hand image the 8x32. Two reasons for this assumption: Firstly, the right-hand image shows a larger field (more squares across) and I'm assuming that you had the target and tripod in more or less the same place for both photos. The second reason is that the image I'm assuming to be the 8.5x SV image looks all but identical to the one you posted in the first linked thread, while the right-hand image has some pretty pronounced additional twisting of the lines in the last 5% or so of the field edge.
Now as an addition plus commentary on your comment on edge sharpness vs pincushion, some data on off-axis resolution of the 10x42 SF and the 10x42 Canon. I measured these at 10 meter distance, where the true field of the Zeiss measured 121 cm and the Canon 111 cm. Incidentally, these fields correspond just about exactly to the specified figures, certainly well within measurement errors.
My procedure here was to have a binocular on a tripod, center the resolution target (Edmund 2" glass slide), view with both eyes, and focus for best resolution in the center. Both binoculars just barely resolved USAF group 1 element 5 with my eyes. Then I'd leave the focus where it was and check resolution at 10, 20, 30 etc centimeters off axis by tilting the binoculars up 10 cm at a time, so the bottom of the view field was used for the edge of field figures. This way, I can use both eyes for each reading, and don't have to move my head much relative to the binoculars as would be necessary if I'd go to either side.
Results: 10 cm off axis still 1/5 for both binoculars; 20 cm off axis 1/4- for both binoculars; 30 cm off axis 1/2 for both binoculars; 40 cm off axis 1/1 for Canon, 0/6 for Zeiss; 50 cm off axis 1/1 for Canon, 0/5 for Zeiss; 60 cm off axis 0/2 for Zeiss (this is just about the extreme edge). Canon does not reach 60 cm off axis due to its narrower FOV, but at its extreme edge of 55 cm it still resolves 1/1.
So, in this test, the two binoculars are equal and very sharp for half of their view field, after which the Canon begins to slightly pull ahead, being significantly sharper at the extreme edge. However, the edge performance of the Zeiss is very respectable and in my experience only bettered by the Canon and Swaro SV 10x42 & 50 models.
About the distortion patterns, to me the SF pattern looks most like the 8x32 SV pattern, while the Canon pattern looks very much like the 8x42 SLC HD pattern in your first linked thread. If you let your eyes wander over these images, you can actually get a bit of a sense of the feeling the images give when viewing.
I must say that I'm so used to the Canon view by now that my brain probably compensates for any and all distortions it has, but to me it has no rolling ball and only very mild pincushion bending of straight lines off axis. I don't see nearly any angular magnification distortion in it even when I specifically look for it by moving round or rectangular shapes from center to edge of field.
In the Zeiss, for my eyes, the AMD at the outermost 5-10% of the field is very pronounced, and there's quite a lot of RB when panning. This is more noticeable at short distances than long, and although I see it readily, it does not bother me very much and I'm sure that personally I could get used to this view just fine. However, the Canon example shows that equally good or even better edge performance is possible without AMD and RB as long as some pincushion is accepted. Another possible compromise concerning the Canon design is external dimensions of the eyepiece lens system. I don't know enough about optical design to know what the possibilities are, but at least in the Canon eyepiece, there is a very large diameter lens inside the e.p. behind the eye lens, and this necessitates very large eyepiece tubes which are ergonomically problematic. Already at the 6.5 degree view field, I think they are at or above the size limit that works well with binoculars.
Kimmo
Last edited:
henry link
Well-known member
Kimmo,
I went back and looked at the original uncropped photos from 2012. I agree with you that the left photo is the 8.5x42. I also found that I had experimented with a circular target in the store, but hadn't posted the images. The photos below show the distorted shape of the circle when it was placed close to the field edge in the 8.5x42 SV (left image) and the 8x32 SV (right image). By my measurements the vertical axis of the compressed circle in the 8.5x42 is about 8% longer than the horizontal axis while the vertical axis in the 8x32 is about 11% longer than the horizontal axis. So, the raw numbers indicate that the 8x32 has higher angular magnification distortion and should be a little more inclined toward RB. I think the true distortions of the two are a little bit closer than the raw numbers because the 8.5x42 image has a larger circle from the higher magnification and that would naturally show a little less AMD. If I had been more careful I would have used two circles of different sizes to compensate for the magnification difference.
Henry
I went back and looked at the original uncropped photos from 2012. I agree with you that the left photo is the 8.5x42. I also found that I had experimented with a circular target in the store, but hadn't posted the images. The photos below show the distorted shape of the circle when it was placed close to the field edge in the 8.5x42 SV (left image) and the 8x32 SV (right image). By my measurements the vertical axis of the compressed circle in the 8.5x42 is about 8% longer than the horizontal axis while the vertical axis in the 8x32 is about 11% longer than the horizontal axis. So, the raw numbers indicate that the 8x32 has higher angular magnification distortion and should be a little more inclined toward RB. I think the true distortions of the two are a little bit closer than the raw numbers because the 8.5x42 image has a larger circle from the higher magnification and that would naturally show a little less AMD. If I had been more careful I would have used two circles of different sizes to compensate for the magnification difference.
Henry
Attachments
Last edited:
Henry,
Thanks for digging these up. Also good to hear we are in agreement about which was which in the grid pattern photos.
I had a chance to briefly view with three New EL SV's yesterday, indoors as the weather sucked. They were an 8x32 and a 10x in both 42 and 50mm configurations. Of these, the 8x32 showed much more AMD in the extreme edge than either of the tens, and also much softer edges with more aberrations. This last observation is of course not surprising, since typically 10x binoculars have better relative off axis behavior than lower magnifications, and the 8x32 has an unusually wide field of view.
Of those two circles, the 8x32 circle looks more like what I see through the Zeiss SF's.
Kimmo
Thanks for digging these up. Also good to hear we are in agreement about which was which in the grid pattern photos.
I had a chance to briefly view with three New EL SV's yesterday, indoors as the weather sucked. They were an 8x32 and a 10x in both 42 and 50mm configurations. Of these, the 8x32 showed much more AMD in the extreme edge than either of the tens, and also much softer edges with more aberrations. This last observation is of course not surprising, since typically 10x binoculars have better relative off axis behavior than lower magnifications, and the 8x32 has an unusually wide field of view.
Of those two circles, the 8x32 circle looks more like what I see through the Zeiss SF's.
Kimmo
I thought it might be better to provide the off axis resolution measurements of the 10x42 Zeiss SF and Canon in a more easily comprehensible form for those who do not read USAF 1951.
So here they come in seconds of arc:
Center of field: Zeiss 3.28" Canon 3.28"
10 cm off axis: Zeiss 3.28" Canon 3.28"
20 cm off axis: Zeiss 3.9" Canon 3.9"
30 cm off axis: Zeiss 4.64" Canon 4.64"
40 cm off axis: Zeiss 5.8" Canon 5.2"
50 cm off axis: Zeiss 6.56" Canon 5.2"
55 cm off axis Canon 5.2"
60 cm off axis: Zeiss 9.28"
These figures refer to single line width in seconds of arc at 10 meters distance. For line-pair angle, multiply each figure by two.
Target line-pair dimensions here range between 3.17 lp/mm (the 3.28" figure) to 1.12 lp/mm (the 9,28" figure). The 5.2" measured for the Canon at the extreme edge of field is 2.0 lp/mm.
Kimmo
So here they come in seconds of arc:
Center of field: Zeiss 3.28" Canon 3.28"
10 cm off axis: Zeiss 3.28" Canon 3.28"
20 cm off axis: Zeiss 3.9" Canon 3.9"
30 cm off axis: Zeiss 4.64" Canon 4.64"
40 cm off axis: Zeiss 5.8" Canon 5.2"
50 cm off axis: Zeiss 6.56" Canon 5.2"
55 cm off axis Canon 5.2"
60 cm off axis: Zeiss 9.28"
These figures refer to single line width in seconds of arc at 10 meters distance. For line-pair angle, multiply each figure by two.
Target line-pair dimensions here range between 3.17 lp/mm (the 3.28" figure) to 1.12 lp/mm (the 9,28" figure). The 5.2" measured for the Canon at the extreme edge of field is 2.0 lp/mm.
Kimmo
Tobias Mennle
Well-known member
Thanks, Henry, for your detailed explanation, will try it with my Habicht.
Kimmo, great stuff.
So the Canon 10x42 with pincushion has better edge performance than the Zeiss SF... could Zeiss really get similar edge performance with pincushion in a field that wide... and if yes, why on earth didn´t they do it.
Kimmo, great stuff.
So the Canon 10x42 with pincushion has better edge performance than the Zeiss SF... could Zeiss really get similar edge performance with pincushion in a field that wide... and if yes, why on earth didn´t they do it.
