Schneider Xenon-Sapphire 3.9/95-0001 Unscientific Testing

Have questions about the equipment used for macro- or micro- photography? Post those questions in this forum.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

Macro_Cosmos
Posts: 1511
Joined: Mon Jan 15, 2018 9:23 pm
Contact:

Schneider Xenon-Sapphire 3.9/95-0001 Unscientific Testing

Post by Macro_Cosmos »

This lens is optimised for 0.23x and does well at Mag∈[0.2x,0.28x].

Using it at 1x is using it out of spec, and the results are quite poor. Forward mounting gives this:
Image

I tried mounting it in retro, still pretty bad. This lens should do very well at around 4x (1/.23=4.3478) which is why I bought it in the first place. I hope it does well at least, it wasn't cheap.

The centre is actually very sharp, it's as good as the PN105A. Mid-frame is mediocre and corners are unusable. I've seen enlarging lenses designed for 1in sensors do better on fx in corners.

CA is completely non-existent. To calibrate for magnification, I used a ruler with 0.5mm resolution. Lenses with CA will reveal their traits here. This is a plus, CA can get annoying at smaller Mag.

This lens does have a glaring (haha I know silly puns) problem, it does horribly against light. Extremely bad, the worse I've ever seen on any lens. I thought the Laowa 12mm f/2.8 UWA was horrible, it looks solid now that I've used it. This renders a lens hood a must for use in the field or whatever scenario one wants to shoot against light. Luckily, the front threads are M52x0.75, metal lens hoods are cheap from China. If used in reverse, I just use a K3 female F-mount adapter and a M52 coupler with female internal threads to mount a metal hood on.

Here's the full resolution:
(Click on the link)
https://live.staticflickr.com/65535/487 ... 7001_o.jpg

I do hope I run into the 1.143x one soon. That might just be the reference lens for the ~1x Mag.
Last edited by Macro_Cosmos on Wed Sep 18, 2019 12:28 pm, edited 1 time in total.

ray_parkhurst
Posts: 3416
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

I considered this lens for use at 3.5x for stack and stitch, but the aperture would give only a slight improvement over the 105IXL3p5, assuming it is diffraction-limited wide open across an 82mm sensor, which is a big risk to take for the cost of this lens. Have you tested it at 3.5x? That seems to be a common mag for inspection, with test houses demanding "1um" resolution on 3.5um sensors.

I'm curious about your claim that the lens is as good at 1x as a 105PN in the center. Seems like the aperture of f3.9 is not going to beat a 105PN with its f2.8 wide open and f3.3 optimum aperture. I'd like to see some pics with both to prove that claim.

The flare problem is not surprising since these lenses were designed for very controlled conditions, and flare was probably not a design consideration.

Macro_Cosmos
Posts: 1511
Joined: Mon Jan 15, 2018 9:23 pm
Contact:

Post by Macro_Cosmos »

ray_parkhurst wrote:I considered this lens for use at 3.5x for stack and stitch, but the aperture would give only a slight improvement over the 105IXL3p5, assuming it is diffraction-limited wide open across an 82mm sensor, which is a big risk to take for the cost of this lens. Have you tested it at 3.5x? That seems to be a common mag for inspection, with test houses demanding "1um" resolution on 3.5um sensors.

I'm curious about your claim that the lens is as good at 1x as a 105PN in the center. Seems like the aperture of f3.9 is not going to beat a 105PN with its f2.8 wide open and f3.3 optimum aperture. I'd like to see some pics with both to prove that claim.

The flare problem is not surprising since these lenses were designed for very controlled conditions, and flare was probably not a design consideration.
Schneider claims in their datasheet that F4.8 shows maximum MTF with practically diffraction limited performance over the whole field, 0% vignetting realised.

I'll fire up some PN105A stacks of the same subject later in the night if I get the opportunity. I used to use my PN reversed since the shroud was removed. I have since fixed that issue so quality might improve. I don't think the PN105A was designed for retro (unlike the Rayfact 95).

chris_ma
Posts: 570
Joined: Fri Mar 22, 2019 2:23 pm
Location: Germany

Post by chris_ma »

Macro_Cosmos wrote:I used to use my PN reversed since the shroud was removed. I have since fixed that issue so quality might improve. I don't think the PN105A was designed for retro (unlike the Rayfact 95).
Interesting, I always thought with lenses optimised for 1:1 the lens orientation won't make any difference, but maybe my logic is flawed there? (like if the lens is a non symmetrical design or designed to be used with digital sensors with glas filter).

I would be very interested in a comparison of the PN105 normal and reversed.
chris

ray_parkhurst
Posts: 3416
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

chris_ma wrote:
Macro_Cosmos wrote:I used to use my PN reversed since the shroud was removed. I have since fixed that issue so quality might improve. I don't think the PN105A was designed for retro (unlike the Rayfact 95).
Interesting, I always thought with lenses optimised for 1:1 the lens orientation won't make any difference, but maybe my logic is flawed there? (like if the lens is a non symmetrical design or designed to be used with digital sensors with glas filter).

I would be very interested in a comparison of the PN105 normal and reversed.
chris
The lens is symmetrical so there is no difference, though there may be slight differences due to mfg tolerances or lens to lens variation.

In last few years I've been mounting my 105PN in retro since it's much easier than dealing with large adapters. The small M45-M42 adapter that Raf makes is perfect for this. It's also easy to swap from 105PN to 95PN since I use the 95PN in retro for 2x.

Edited to add: The "A" versions of the 105PN and 95PN have flat front and rear elements. I presume this may cause problems for re-reflections onto digital sensors, but the lenses are so far extended that I have never seen an issue.

Further edit: I have not tested a Rayfact version of the 105PN or 95PN, so am not sure if they also have flat front elements. Perhaps someone with one or both of these lenses could comment?

Macro_Cosmos
Posts: 1511
Joined: Mon Jan 15, 2018 9:23 pm
Contact:

Post by Macro_Cosmos »

chris_ma wrote:
Macro_Cosmos wrote:I used to use my PN reversed since the shroud was removed. I have since fixed that issue so quality might improve. I don't think the PN105A was designed for retro (unlike the Rayfact 95).
Interesting, I always thought with lenses optimised for 1:1 the lens orientation won't make any difference, but maybe my logic is flawed there? (like if the lens is a non symmetrical design or designed to be used with digital sensors with glas filter).

I would be very interested in a comparison of the PN105 normal and reversed.
chris
When I mounted mine in reverse, the results weren't that good.
This could also be a consequence of poor adaption which might bring in problems. I actually do intend to compare a bunch of lenses at 1x, it's going to be a big project, my current list is:

- Printing-Nikkor 105mm f/2.8A forward (reference)
- Printing-Nikkor 105mm f/2.8A Reversed
- Nikon Rayfact OFM20119MN (PN95) forward
- Nikon Rayfact OFM20119MN (PN95) reversed
- PC-E Nikkor 24mm f/3.5 with extension
- Minolta 5400 Scanner lens
- Scanner-nikkor 7 element
- $5 Kowa Scanner lens
- Schneider Xenon-Sapphire 3.9/96-0001 forward
- Schneider Xenon-Sapphire 3.9/96-0001 reversed

ray_parkhurst
Posts: 3416
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

Macro_Cosmos wrote:
chris_ma wrote:
Macro_Cosmos wrote:I used to use my PN reversed since the shroud was removed. I have since fixed that issue so quality might improve. I don't think the PN105A was designed for retro (unlike the Rayfact 95).
Interesting, I always thought with lenses optimised for 1:1 the lens orientation won't make any difference, but maybe my logic is flawed there? (like if the lens is a non symmetrical design or designed to be used with digital sensors with glas filter).

I would be very interested in a comparison of the PN105 normal and reversed.
chris
When I mounted mine in reverse, the results weren't that good.
This could also be a consequence of poor adaption which might bring in problems. I actually do intend to compare a bunch of lenses at 1x, it's going to be a big project, my current list is:

- Printing-Nikkor 105mm f/2.8A forward (reference)
- Printing-Nikkor 105mm f/2.8A Reversed
- Nikon Rayfact OFM20119MN (PN95) forward
- Nikon Rayfact OFM20119MN (PN95) reversed
- PC-E Nikkor 24mm f/3.5 with extension
- Minolta 5400 Scanner lens
- Scanner-nikkor 7 element
- $5 Kowa Scanner lens
- Schneider Xenon-Sapphire 3.9/96-0001 forward
- Schneider Xenon-Sapphire 3.9/96-0001 reversed
The biggest issue I've had with the 105PN is central hotspotting / lower contrast due to reflections off of extension / adapter internal surfaces. If your normal and retro mountings had different configurations of extensions and adapters, that might point to the problem.

Now, if the problem was one of coverage or central sharpness, then there could be a problem with the lens itself. I've owned and tested >20 105PNs and they were all essentially identical and perfect, forward or reversed. Nikon and later Rayfact/Nikon build these lenses like optical benches, with each element assembled by hand and tuned for spacings and such. Only other lenses I've seen that are this consistent are the 8000/9000ED scanner lenses, which are also built similarly.

I expect you will see excellent performance from the 95PN in reverse, but mediocre in forward, based on my tests.

I wish you luck in getting a setup configured for 1x on the scanner lenses. You'll likely need to mount lens directly to the camera, with minimal extension. I found it to be impossible in my setup on Canon. Maybe if you're using a mirrorless you will have better luck.

Macro_Cosmos
Posts: 1511
Joined: Mon Jan 15, 2018 9:23 pm
Contact:

Post by Macro_Cosmos »

ray_parkhurst wrote: Further edit: I have not tested a Rayfact version of the 105PN or 95PN, so am not sure if they also have flat front elements. Perhaps someone with one or both of these lenses could comment?
I can confirm that both Rayfact versions have flat front elements.
I own the Rayfact version of the PN95 and I know someone who has the Rayfact version of the 105/2.8 PN, which I've handled.

Here's the Chia Seeds at 1x, taken with the 105/2.8A PN:
Image
Full Res: https://live.staticflickr.com/65535/487 ... 4343_o.jpg

Centre:
Image
It's honestly hard to tell. The PN105 does have an edge. Take a guess which is which? (PN105 on the right)

The corners aren't worth a comparison. The difference is night and day.
Image

Because of the small acceptable circle of sharpness on the Xenon-Sapphire, that image actually "pops out more". If the subject is in the dead centre, such as a specific part of a flower or some small insect, the Xenon-Sapphire is alright. If one wants edge-to-edge sharpness, at 1x, my "reference lens" is still the winner.

chris_ma
Posts: 570
Joined: Fri Mar 22, 2019 2:23 pm
Location: Germany

Post by chris_ma »

Macro_Cosmos wrote:When I mounted mine in reverse, the results weren't that good.
This could also be a consequence of poor adaption which might bring in problems. I actually do intend to compare a bunch of lenses at 1x, it's going to be a big project, my current list is:
interesting that you got worse results in reverse.
maybe if it's a version already optimised for digital sensors then they calculated it to compensate for a CCD cover glass on one side?
(the inspect.x series for example mentions it's designed for use for a cover glass 0.76mm D263).
As far as I know most photographic cameras have a filter stack around 2mm (and M43 4mm), but then again in macro shots the extension should be long enough so that the glass doesn't change much. two links for those interested:
https://wordpress.lensrentals.com/blog/ ... ed-lenses/
https://www.lensrentals.com/blog/2014/0 ... it-matter/

In any case, very much looking forward to your 1x comparison, specially the Minolta 5400 with the PN105A (and also the coverage of the M5400)!

chris

Macro_Cosmos
Posts: 1511
Joined: Mon Jan 15, 2018 9:23 pm
Contact:

Post by Macro_Cosmos »

Image
55555 posts!

ray_parkhurst wrote: The biggest issue I've had with the 105PN is central hotspotting / lower contrast due to reflections off of extension / adapter internal surfaces. If your normal and retro mountings had different configurations of extensions and adapters, that might point to the problem.

Now, if the problem was one of coverage or central sharpness, then there could be a problem with the lens itself. I've owned and tested >20 105PNs and they were all essentially identical and perfect, forward or reversed. Nikon and later Rayfact/Nikon build these lenses like optical benches, with each element assembled by hand and tuned for spacings and such. Only other lenses I've seen that are this consistent are the 8000/9000ED scanner lenses, which are also built similarly.

I expect you will see excellent performance from the 95PN in reverse, but mediocre in forward, based on my tests.

I wish you luck in getting a setup configured for 1x on the scanner lenses. You'll likely need to mount lens directly to the camera, with minimal extension. I found it to be impossible in my setup on Canon. Maybe if you're using a mirrorless you will have better luck.
Back when I first used the PN105, I was forcing the M52x0.75 adapter into SM2 tubes (thread pitch does not match), which is very likely why I saw inconsistent results. Same with the RF95. I later on bought some M52x0.75 to SM2 adapters from Thorlabs and many issues went away.

I'm pretty sure the Scanner-Nikkor 8000/9000ED lenses are identical to the PN105.

I do try to minimise extension tube combinations. My end goal is custom made tubes to the exact length for my Nikon DSLRs. If I ever buy into the mirrorless hype, Nikon's official adapter will do.
Image
5.5in to get 1.03x. I can make it exactly 1x with SM2 variable tubes but those are finicky, 1.03x is close enough anyway, I can't tell the difference unless a ruler is the subject. We don't take photos of rulers here (I hope not :P )

chris_ma wrote: interesting that you got worse results in reverse.
maybe if it's a version already optimised for digital sensors then they calculated it to compensate for a CCD cover glass on one side?
(the inspect.x series for example mentions it's designed for use for a cover glass 0.76mm D263).
As far as I know most photographic cameras have a filter stack around 2mm (and M43 4mm), but then again in macro shots the extension should be long enough so that the glass doesn't change much. two links for those interested:
https://wordpress.lensrentals.com/blog/ ... ed-lenses/
https://www.lensrentals.com/blog/2014/0 ... it-matter/

In any case, very much looking forward to your 1x comparison, specially the Minolta 5400 with the PN105A (and also the coverage of the M5400)!

chris
By coverage, do you mean how well the 5400 covers FX? It does, adequately.

I have an older comparison:
Image
https://live.staticflickr.com/7887/4712 ... 5b2a_o.jpg

I don't recommend using it to decide which is better, my old adaption setup is pretty bad.


The $5 Kowa is probably the ultimate winner already though, $5 compared to $150-$500 (MN5400) compared to $400-$1000 (PN105A), it's unbeatable.
Kowa on top. Corners aren't too good but far from bad.
Image
Image

Either way, a better more controlled comparison is required.

ray_parkhurst
Posts: 3416
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

Macro_Cosmos wrote: ...
I'm pretty sure the Scanner-Nikkor 8000/9000ED lenses are identical to the PN105.
...
For sure this is not true, they are significantly different.

Macro_Cosmos
Posts: 1511
Joined: Mon Jan 15, 2018 9:23 pm
Contact:

Post by Macro_Cosmos »

ray_parkhurst wrote:
Macro_Cosmos wrote: ...
I'm pretty sure the Scanner-Nikkor 8000/9000ED lenses are identical to the PN105.
...
For sure this is not true, they are significantly different.
What I mean is those two are just the PN105 being stripped of all its mechanical parts. The performance is identical. Unfortunately I don't have it anymore so I can't really verify this.

ray_parkhurst
Posts: 3416
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

Macro_Cosmos wrote:
ray_parkhurst wrote:
Macro_Cosmos wrote: ...
I'm pretty sure the Scanner-Nikkor 8000/9000ED lenses are identical to the PN105.
...
For sure this is not true, they are significantly different.
What I mean is those two are just the PN105 being stripped of all its mechanical parts. The performance is identical. Unfortunately I don't have it anymore so I can't really verify this.
No, they are very different internally. I'll pull mine out and take some pics to show.

Edited to add: the biggest physical differences are:

- 105PNA has flat front and rear lens surfaces, while the 8000ED lenses have significant curvature
- Although the 8000ED outer casing is larger, its retaining rings are much wider. The end result is the exposed glass element diameter is 38mm on the 105PN, and 35mm on the 8000ED.

The designs may indeed be similar, but for sure are not identical.

Further edit: the performance is also not identical.

Macro_Cosmos
Posts: 1511
Joined: Mon Jan 15, 2018 9:23 pm
Contact:

Post by Macro_Cosmos »

ray_parkhurst wrote:
Macro_Cosmos wrote:
ray_parkhurst wrote:
Macro_Cosmos wrote: ...
I'm pretty sure the Scanner-Nikkor 8000/9000ED lenses are identical to the PN105.
...
For sure this is not true, they are significantly different.
What I mean is those two are just the PN105 being stripped of all its mechanical parts. The performance is identical. Unfortunately I don't have it anymore so I can't really verify this.
No, they are very different internally. I'll pull mine out and take some pics to show.

Edited to add: the biggest physical differences are:

- 105PNA has flat front and rear lens surfaces, while the 8000ED lenses have significant curvature
- Although the 8000ED outer casing is larger, its retaining rings are much wider. The end result is the exposed glass element diameter is 38mm on the 105PN, and 35mm on the 8000ED.

The designs may indeed be similar, but for sure are not identical.

Further edit: the performance is also not identical.


:oops: :oops:
Don't make me feel regretful about letting the 8000ED go.

ray_parkhurst
Posts: 3416
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

Macro_Cosmos wrote: :oops: :oops:
Don't make me feel regretful about letting the 8000ED go.
No worries, they are not rare.

Post Reply Previous topicNext topic