Nikon SE 8x32 vs. Zeiss Victory FL 7x42 T* ??? (1 Viewer)

henry link said:

Too much here for me to read,

It doesn't matter at all whether you focus at the edge or the center first. I do it both ways. You're simply following the same curve from different directions. Either way the focus difference between best focus at the edge (midpoint between tangential and sagittal foci) and best focus at the center will have exactly the same value expressed in diopters.

Henry

Click to expand...

Henry,
when you do get the time to read what I wrote, you will find that I explained to you that this measurement does not measure curvature. It measures all other residual aberrations. As I said before, stop and think for a minute about what you are doing.

The best focus at edge represents all other aberrations minus curvature. the best focus in center represents no aberrations. Subtracting one from the other gives all other aberrations.

Frankly Henry, If you think about it just a bit longer, you can't even get curvature by starting focus out in the field. You can only get it by starting focus in the center, then moving out to the field to refocus. A little bit of thought will prove that out. That's a big difference than what you are doing.

Subtracting what you get from total would give curvature. But you never know what total is unless you start by focusing in the center.

I explained my use of units up above. As far as I know, all aberrations are measured by the degree of error they cause in the image. Therefore, I use arcseconds error. Others may do as they please.

edz

I did read that part, Ed. Once again, it makes no difference whether you establish best center focus or best edge focus first. The change in focus (the curvature of the field) between the two points is quite unaffected by the order. But, if you prefer to establish the plane of central focus first, there's no harm in doing it that way. It won't change anything. I do it that way sometimes too.

Perhaps I can make this clearer. Imagine you are measuring the FC of a binocular using a focusing eyepiece with an accurate diopter scale. You focus on a star at the center and note 0D on the diopter scale. Now you move the star to the edge and refocus for the lowest aberrations possible at the edge (the best focus). Now the scale reads -2D indicating 2 diopters of field curvature. Now move the star back to the center. It's now out of focus and you will need to rotate the eyepiece 2 diopters back to the zero position to regain focus. 2 diopters either way, it doesn't matter where you start.

Ron,

I decided to go ahead and post the Fujinon information. The 2D difference between the sagittal and tangential foci at the edge of the field indicates a bit more than negligible astigmatism, but field curvature (a line midway between the T and S curves not included in the graphic) would be quite low according to this, about 1D. How well do you think this reflects what you see through the binocular?

Henry

Henry,
Thanks for the plot and the interpretation.

It is not what I see, however, at least as well as I can understand its meaning. I just went out and checked the 7x50 again, for probably only the tenth time, naaathing, I know. But I saw what I saw the first time and every time.

If I focus carefully on a centered object at infinity, and move it right to the edge, I mean like touching, I can get the image back into sharp focus with -2.0D, or shall I say 2.0 white marks, of adjustment. -1.5 won't do, and -2.5 won't, but are equally bad, so I suspect the 2.0 is determined to within 0.1 or 0.2. And the refocused image at the edge is really sharp, maybe not quite, but very nearly as sharp as best focus at the middle. Of course color fringing is horrible out there if the light is bright and contrasty, but in non-CA situations, the refocused image is good.

Next I focused at the center with and without a pair of 2.25D reading glasses, and the adjustment required was precisely 2.2 or 2.3D, honest. Therefore, the Fujinon marks are diopters.

Then there is the plot's result that at the edge, tangential astigmatism is only 0.5D but sagittal is about 1.5D. I believe tangential means around the circumference of the field, and sagittal is radially directed, please correct me if this is wrong. Whichever, if I focus on a centered point, and move it to the edge without refocusing, although it spreads in both directions, it spreads more in the circumferential direction, as in most binoculars. I suppose this blur cannot be all defocus, which would make a round spot. Again, with refocusing, it becomes sharp as a nail.

So now I am a bit confused. Refocusing the edge implies the aberration is a nearly pure 2D of field curvature. But observing the edge without refocusing shows an asymmetric blur that seems to contain some astigmatism.

There's a lot of stuff I don't understand, but an outstanding mystery to me at the moment is: what is the difference between the appearance of positive, and negative, astigmatism? Maybe that would unravel my difficulty if I understood that.

It is not a warm fuzzy feeling to be disagreeing with Fujinon. Rarely does one get such technical dope from a manufacturer. One cherishes it, and it probably should take more than the likes of me to cast doubt on it. But that's what I see. I hope that if I am shown wrong I will at least learn something from it!

Sorry about all the verbage when you are sort of snowed under.
Ron

Henry,
I will try this once more.

When you start at the outer edge and bring a target to best focus, you have already removed any error due to curvature. There is no curvature left affecting the image, only all other aberrations that cannot be focused out. That is what best focus out at the field edge represents.

You then move to the center and refocus to best at center. That removes all other aberrations. When you subtract one from the other, you get the focus in diopters to remove "all other aberrations".

For this it would not matter if you start at edge or center. But the fact of the matter is, you are not measuring curvature. You already removed the curvature when you decided to start by focusing at the outer edge.

Now, once again try it the other way.
1-Focus in the center.
2-move to the outer edge.
3-Now the first part of refocus goes from all aberrations at edge to best focus at edge. That gives diopter adjustment to remove curvature. That leaves you with all other aberration remaining.
4-Finally, you are now at best focus at edge. If you care to, you can move back to center and focus to best and the refocus difference will give you the diopter adjustment to remove all other aberrations.

your method of starting at edge completely misses the increment in step 3. Of course step 4 can be found by going from 1 to 4 or by going from 4 to 1. Step 3 can only be found by going from 1 to 3.

I'll reiterate, you are not measuring curvature.

edz

I'll also make one more effort. Then, barring one of us experiencing an epiphany I'm willing to allow the 3 or 4 people who are still reading this thread to make up their own minds based on what's been said.

I agree that your Step 3 is the problem. Establishing best focus of a star at the edge tells you nothing about field curvature "error" because field curvature doesn't exist for a single point in the field. FC is the difference in focus between two points in the field (the points of interest here being the center and the edge). If FC is present then focusing the edge defocuses the center. Focusing the center defocuses the edge. Bringing the edge to best focus is essentially nothing more than minimizing the appearance of the astigmatism that's always present to some extint at the edge. That is, making the image of the star into a tiny cross with bars of equal length (the shorter the bars the lower the astigmatism). That point is the midpoint between the sagittal and tangential foci (see the Fujinon graphic). If it falls on the plane of the center focus (the vertical line in the Fujinon plot) there is no field curvature and if you move the star to the center it will still be focused. If there is FC (the point of best focus at the edge doesn't fall on the vertical line) the star will be defocussed when it's moved to the center and refocusing using an eyepiece with an accurate diopter scale shows the difference in focus between the center and the edge in diopters (the scale at the bottom of the Fujinon plot). Doing everything in reverse will produce the same measurement.

I believe there are more than three or four people still reading this. I don't know if either of you will answer, but I am following Henry and not following Ed. So for the unscientific among us:

Ed, you are saying that if you focus at the edge you have "already removed the curvature" by focusing there. You haven't established that there is any curvature. Consequently I follow Henry by understanding that the difference between the center and the edge is the same as the difference between the edge and the center.

What I'm not following here is the degree to which aberrations other than FC can be removed by focusing, and you seem to be talking past each other on that point. Maybe this is stupidly elementary on my part, but I have not understood that astigmatism, coma, or other errors could be removed by focusing.

Person #2 here!

Ed, I don't follow you argument about how aberrations other than FC enter the picture. I know I'm sort of interrupting an exchange between you and Henry, and I apologize.

But, may I make a simplification? Assume that FC is the only aberration. Then, it seems to me that the difference in the diopter settings needed to focus an object sharply at center, and then at edge, is a measurement of field curvature. And if this is wrong, maybe it will be easier for you to explain why, in this simpler case.
Thanks,
Ron

I have no real scientific background for this, but I have been folowing this exchange with interest.

Like Ron I can't really understand Ed's methodology, or why it makes any difference whether you start in the center or at the edge. That said, simply using the 8x32 SE pretty much proves it doesn't have 3 diopters of curvature doesn't it? That's a LOT of curvature and the edges would simply be blurry if it were there, wouldn't they? The fact that they aren't suggests much less. I also wonder if depth of field may be an issue here as well, both in the binocular and the eye.

Just my 2 cents. I'm keenly interested in the outcome. Neat stuff.
Mark

the only aberration that can be focused out is curvature. other aberrations, or residual, cannot be focused out.

if you start by coming to best focus at some point out in the field, you have effectively already focused out all error due to curvature. as i've said repeatedly, best focus at some (any) point out in the field, leaves behind ONLY the residual aberrations.

subtracting the difference in focus from best out in field to best at center represents the degree of focus for ALL OTHER aberrations. refer tho steps 1-4 outlined above.

on the contrary, if you focus in the center and then move to any other point in the field, the amount of defocus (ERROR DUE TO TOTAL ABERRATION) seen in the image represents the combination of curvature AND all other aberrations. the key is step 3, which separates the curvature from all other.

the methods do not give the same results.

Henry is correct that one of us needs to experience an epiphany.

edz

Let's say you go from best focus at center to best focus at edge, then take a measurement. If the two methods do not give the same results then going back from from best at edge to best at center would, presumably, give you different results yes? Of course it wouldn't. They'd be the same. If they differed, then whatever "best focus" is must be an unstable commodity.

That's the best my mind can muster, epiphany or no.

Mark

edz said:

the only aberration that can be focused out is curvature. other aberrations, or residual, cannot be focused out.

if you start by coming to best focus at some point out in the field, you have effectively already focused out all error due to curvature. as i've said repeatedly, best focus at some (any) point out in the field, leaves behind ONLY the residual aberrations.

subtracting the difference in focus from best out in field to best at center represents the degree of focus for ALL OTHER aberrations. refer tho steps 1-4 outlined above.

on the contrary, if you focus in the center and then move to any other point in the field, the amount of defocus (ERROR DUE TO TOTAL ABERRATION) seen in the image represents the combination of curvature AND all other aberrations. the key is step 3, which separates the curvature from all other.

the methods do not give the same results.

Henry is correct that one of us needs to experience an epiphany.

edz

Click to expand...

Here is what I am (mis?)understanding: The only thing that can be focused out is FC. No matter where we achieve focus, all other errors remain (edge), no or few other errors are present (center), or we are at some point in between. The only measurable difference--using focus--between two points in the field (here center and edge) is a change in focus dictated by FC. I conclude that all other errors are irrelevant because they are unaffected by change in focus. I cannot understand from Ed's posts how other errors can contribute to the measurable change in focus, if FC is the only error that can be focused out.

What I can see on a chart viewed through the binocular after adjusting for FC has nothing to do with the change in focus that I had to make to be able to read the chart.

Let's say that the 'other aberrations' are absent in the centre, but at the edge all measured aberrations are, let's say, 100 units, 50 of which is curvature and the other 50 is 'other aberrations'.
Then, re-focusing at the edge removes 50 of the curvature. The 'other aberrations' are CONSTANT value at this point of FOV.
Then, if you move back to FOV centre, you have back the same amount of curvature (50 units) but no 'other aberrations'. Now, to improve that, takes the same amount of re-focusing.
Therefore, IMO it doesn't matter which way we go. That's how I understand it.
Interesting thread, BTW!!!

Regards,

Maciek

I wish to echo FrankD's sentiments regarding the Zeiss 7x42FL. I have had the good fortune to own a number of high-end optics over the last 10-15 years, including Swarovski SLC, Leica Trinovid and Ultravid, Nikon Venturer and Bushnell Elite. I have also had brief experience with friends' Swarovski ELs and Nikon EDGs. However, the one glass that has left an indelible mark on me is the Zeiss 7x42FL. Perfect? No, but close enough for me. If I were in the market for one today, however, the astronomical price would likely be enough for me to look elsewhere. I believe Zeiss FLs are approaching 2K currently (in fairness to Zeiss, however, the other alphas are also priced in the stratosphere, some even higher than Zeiss FLs.) My unit is a first generation version ('non-LotuTec') that I bought on the used market, so I saved mucho dinero (this was also when the U.S. Dollar was stronger than the Euro.) You may wish to look at the used market if you're working within a budget (and who isn't these days?) Zeiss FLs have been on the market for some time, so there are conceivably more used units available. I don't have experience with the Nikon SE, but the fact that it isn't water- and fog-proof is a deal breaker. Furthermore, the porro design of the SE is less comfortable for me than roof designs, but that's partly because I have small hands. However, I read nothing but superlatives about the image. So with all that said, I would hold off and save your pennies for an FL.

Howard220 said:

...But if you start in the center and take a reading there, assuming there are no aberrations present, and THEN move to the edge, there at the edge you can work at getting best focus based on field curvature alone. Wait - now I may be getting lost again. :scribe:

Click to expand...

Yes, at the edge you correct field curvature, so when moving back to the centre, you have to correct the same amount of field curvature again to obtain perfect focus.

Which of these two divergent threads was the original?

N P

Well Henry, I had my epihany.

And I apologize for belaboring the point about where you start focus to use your method. It doesn't matter. I must have had my dumb hat on, as I said it myself, the only error that can be focused out is curvature.

If all you want to do is measure how far the diopter turns to correct curvature, start in the center or start at the edge, you can get the same result. You'll find out how far the diopter needs to turn to correct curvature.

However, if you want to actually measure the total aberration in the image, and then the amount within that total that is from curvature and how much is all other aberrations, and actually find the image error, which is what I do, then you need to start in the center and refocus out in the field, otherwise you never know the total and can't find the difference. Some of you figured that out above.

But let me describe what I found yesterday. I picked about half a dozen binoculars off the shelf and went outside to view my target. I wanted to test this diopter method. So I tried both ways, and sure enough, get the same number of diopters both ways. But wait, there's more.

Now here's the clincher.
I have to assume all these binoculars have 5 to about 7 diopters plus or minus, since I looked thru the house and could only find maybe 2 out of 10 that have a diopter scale. I think 7 diopters is about the most I've ever seen. But somewhere between +/-5 to +/-7 diopters is a reasonable assumption, except that one (as noted) has a very very short range assumed +/- 4 diopters. I purposely chose binoculars with a wide range of actual measured curvature error.

I have previously measured and recorded the actual curvature error in arcseconds in the image for these binoculars.

Celestron Regal LX 8x42 rp = 25 arcseconds
ZRS 10x42 rp = 45
Nikon SE 10x42 = 60
Pentax PCF WP II 8x40 = 90
Bushnell Legend 8x42 rp =100
Oberwerk 12x50 sport rp= 120
Zen ED2 8x43 rp = 130
Nikon AE 8x40 = 150

Well, every one of these binoculars focused out the curvature with somewhere between 2 to 4 diopters turn of the dial. Perhaps some were as low as 1.5 diopters, but none (except one) were much more than half the diopter range from zero to end.
The Celestron Regal took 5 clicks out of 15 maximum.
The Nikon SE 10x42 took 2.5 diopters.
The Obie Sport 12x50 took only 2-2.5 out of 8 increments.
The ZRS 10x42 took 3 out of 8 marks.
The Pentax took 2 out of 4 diopters.
The Bushnell Legend took exactly half, appox 3 out of 6 increments of the diopter.
The Zen ED2 took 7 out of 11 clicks.
The Nikon AE 8x40 did take the full diopter range from 0 to end, but it has by far the narrowest diopter range, I suspect only 4+/-.


But how could that possibly be, I thought? If it takes 1.5 to 2.5 or 3 diopters to remove the curvature when curvature contributes only 20 or 25 arcseconds error in the image, or even 45 arcseconds, shouldn't it take 5 or 6 diopters to remove the curvature when it measures 90 or 100 arcseconds and shouldn't it take 10 to 15 diopters to remove the curvature when it measures 120 or 130 or 150 arcseconds?

It doesn't. None of these binoculars took more than about 4 diopters to remove the curvature.

What this tells me is that you can measure the diopter adjustment to remove curvature, but that gives you no clear indication of how much curvature error is in the image. All that tells you is the number of diopters it takes to remove it.

The number of diopters it takes to remove curvature does not correlate to the actual amount of curvature error in the image. In other words, you can't tell how much curvature contributes to the aberration error in the image by noting the adjustment of diopters that it takes to remove it.

Finding out that one binocular has a 2 diopter error and another has a 4 diopter error only tells you it took 2 or 4 diopters to remove it. It does not tell you that one has twice the amount of curvature image error than the other. It doesn't tell you that one of those binoculars may have 30 arcseconds curvature and the other has 150 arcseconds curvature, 5 x as much.

If you want to know how much curvature contributes to the total aberration (distortion, but not in the pure sense of the word) in the image, you need to actually take a reading of the curvature in the image. You can't rely on the number of turns of the diopter as an indication of how much curvature error (distortion) is in the image.

and I apologize to the OP for completely sidetracking his thread.

edz

No apology needed, at least on my part Ed. I found this thread very educational.

edz said:

Well Henry, I had my epihany. edz

Click to expand...

Ed, I've learned more from your and Henry's contributions here (and at CN) than from any other sources for years. I don't know why, but this exchange was more interesting to me than many. Thanks.

Jonathan

edz said:

But how could that possibly be, I thought? If it takes 1.5 to 2.5 or 3 diopters to remove the curvature when curvature contributes only 20 or 25 arcseconds error in the image, or even 45 arcseconds, shouldn't it take 5 or 6 diopters to remove the curvature when it measures 90 or 100 arcseconds and shouldn't it take 10 to 15 diopters to remove the curvature when it measures 120 or 130 or 150 arcseconds?

It doesn't. None of these binoculars took more than about 4 diopters to remove the curvature.

Click to expand...

Ed,

Thanks for being big enough to admit your mistake. At the risk of continuing a digression in this thread (not to mention starting some new unpleasantness) I can offer an alternative explanation for the above, but I don't have time to do it now. Perhaps I can get to it tonight or someone else can chime in.

Henry