(New) Contender at 3.33x? 128 f/2.4 HR Linos Inspec.x
Moderators: rjlittlefield, ChrisR, Chris S., Pau
-
- Posts: 3413
- Joined: Sat Nov 20, 2010 10:40 am
- Location: Santa Clara, CA, USA
- Contact:
-
- Posts: 2627
- Joined: Thu Jan 17, 2013 9:34 pm
- Location: United States
- Contact:
$7031.
Thats less than I guessed
Specs are very nice although thats a lot of extension needed to hit 3.3x!
Focal length 128 mm
Extremely high resolution of up to 300 line pairs per mm in the object
• Magnification range: 3.2 - 3.5x
• Recommended for pixel sizes down to 3.5 μm
• Large image circle of 82 mm
• Working distance 58 ... 61.4 mm
• Spectral range: 400 - 750 nm
• Distortion < 0.1% (design value)
Thats less than I guessed
Specs are very nice although thats a lot of extension needed to hit 3.3x!
Focal length 128 mm
Extremely high resolution of up to 300 line pairs per mm in the object
• Magnification range: 3.2 - 3.5x
• Recommended for pixel sizes down to 3.5 μm
• Large image circle of 82 mm
• Working distance 58 ... 61.4 mm
• Spectral range: 400 - 750 nm
• Distortion < 0.1% (design value)
-
- Posts: 2627
- Joined: Thu Jan 17, 2013 9:34 pm
- Location: United States
- Contact:
-
- Posts: 3413
- Joined: Sat Nov 20, 2010 10:40 am
- Location: Santa Clara, CA, USA
- Contact:
Ahh, even more "reasonable" than I expected. That's for 1?RobertOToole wrote:$7031.
Thats less than I guessed
Specs are very nice although thats a lot of extension needed to hit 3.3x!
Focal length 128 mm
Extremely high resolution of up to 300 line pairs per mm in the object
• Magnification range: 3.2 - 3.5x
• Recommended for pixel sizes down to 3.5 μm
• Large image circle of 82 mm
• Working distance 58 ... 61.4 mm
• Spectral range: 400 - 750 nm
• Distortion < 0.1% (design value)
The long extension required for the target mag is why I have not yet built my SnS system for the RF3p5x and 105IXL3p5x. I have most of the components and have made most of the design decisions, but it will require a completely new system which in my case will be dedicated to the task. It's comforting that if I ever run across a 128IXHR it will fit in the system.
I'm not sure if that change in MTF is just due to straight calculation or if it's due to non-diffraction-limited performance of the lens. When I tested the RF3p5x, I saw only a small improvement at f2.8 vs f2.4. Maybe that's all I should have seen..RobertOToole wrote:MTF drops slightly closing down to f/2.4 to f/2.8.
At least in theory, that might be hard to see in a image.
-
- Posts: 1617
- Joined: Sat Apr 14, 2018 10:26 am
I dont have any experience with macro- and macroscopes, but the line scan lenses have an extremely large and uniform field - I doubt the macroscopes can reach 348lp/mm on the borders of a 80mm image circle.Scarodactyl wrote:Is 300 lp/mm extremely high resolution? The Wild macroscopes have a quoted max resolution of 348 lp/mm at 4x, and that's with a zoom lens from the 80s. Am I missing something, or are the specs perhaps not directly comparable?
there's also the questions if the numbers are directly comparable (same contrast target and image).
plus there are other aspects of image quality like CA, distortion, light falloff etc which I guess will all be much better with a modern line scan lens.
chris
I love that Linos includes little crosses/circles which indicate the limits of diffraction at the edges of their MTF curves, but publishing the calculated MTFs of the lens instead of the real ones is very strange to me.
I mean, they basically say: "if we made no calculation errors and we built the lens perfectly to specs this is what the lens would perform like" - but what if your calculation model is not totally accurate (likely) and what if manufacturing process introduces some inaccuracies (also very likely)?
obviously the lens will perform worse then calculated, but by how much?
5%?
10%?
30%?
there's no way of telling :/
I have two of the older inspec.x L 105 lenses and I'm very happy with them, and I'm sure this new lens is spectacular - but it's still a strange marketing decision to me.
I mean, they basically say: "if we made no calculation errors and we built the lens perfectly to specs this is what the lens would perform like" - but what if your calculation model is not totally accurate (likely) and what if manufacturing process introduces some inaccuracies (also very likely)?
obviously the lens will perform worse then calculated, but by how much?
5%?
10%?
30%?
there's no way of telling :/
I have two of the older inspec.x L 105 lenses and I'm very happy with them, and I'm sure this new lens is spectacular - but it's still a strange marketing decision to me.
-
- Posts: 2627
- Joined: Thu Jan 17, 2013 9:34 pm
- Location: United States
- Contact:
Hi Chris,chris_ma wrote:I love that Linos includes little crosses/circles which indicate the limits of diffraction at the edges of their MTF curves, but publishing the calculated MTFs of the lens instead of the real ones is very strange to me.
I mean, they basically say: "if we made no calculation errors and we built the lens perfectly to specs this is what the lens would perform like" - but what if your calculation model is not totally accurate (likely) and what if manufacturing process introduces some inaccuracies (also very likely)?
obviously the lens will perform worse then calculated, but by how much?
5%?
10%?
30%?
there's no way of telling :/
I have two of the older inspec.x L 105 lenses and I'm very happy with them, and I'm sure this new lens is spectacular - but it's still a strange marketing decision to me.
The old US distributor for Rodenstock posted this years ago (on an online forum).
Response from Rodenstock HQ in Germany about MTF test results:
"Rodenstock MTF curves are calculated. The reality is when you measure MTF curves on the MTF machine with a lens it could have a tolerance of -10 % at the most. In other words there is a difference when you measure the MTF together with the lens."
Also one more thing to think about is digital sensors and cover glass/low-pass filter assy. since these MTFs are not system MTFs with a camera.
LensRentals.com compared 70-200s years ago and Sonys new 70-200 came out but the MTF results placed it near the bottom of the results compared to Canon and Nikon (I believe Nikon had the best sharpness). Sony sent LR a low-pass filter assy to use on their bench and sure enough the MTF did improve. I thought that was a really unfair but the results were interesting to read anyway.
Hope this helps.
Robert
hi Robert,
thanks for the quote from Rodenstock. I can definitely see how getting the right setup for measuring MTF on different lenses could be a tough challenge.
I'm not really sure what they mean with "measure the MTF together with a lens". I mean how would you measure MTF without a lens?
But if I understand them right, they mean that the measured MTF performance would be 90+% of the calculated performance?
so let's say a graph says for a given aperture and lp/mm has a calculated MTF of 50% - the real world performance could be as low as 45% then?
Not that it creates much of a problem in the real world I guess, calculated MTF are very helpful to me for baseline estimates (I wish rayfact had some for comparison), but it still feels a bit sloppy without giving a tolerance range at the same time (I'm sure some lenses are constructed to much tighter standards then others, having less sample variation and being closer to the theoretical values).
just some thoughts and sorry for rambling
chris
thanks for the quote from Rodenstock. I can definitely see how getting the right setup for measuring MTF on different lenses could be a tough challenge.
I'm not really sure what they mean with "measure the MTF together with a lens". I mean how would you measure MTF without a lens?
But if I understand them right, they mean that the measured MTF performance would be 90+% of the calculated performance?
so let's say a graph says for a given aperture and lp/mm has a calculated MTF of 50% - the real world performance could be as low as 45% then?
Not that it creates much of a problem in the real world I guess, calculated MTF are very helpful to me for baseline estimates (I wish rayfact had some for comparison), but it still feels a bit sloppy without giving a tolerance range at the same time (I'm sure some lenses are constructed to much tighter standards then others, having less sample variation and being closer to the theoretical values).
just some thoughts and sorry for rambling
chris
-
- Posts: 2627
- Joined: Thu Jan 17, 2013 9:34 pm
- Location: United States
- Contact:
I'm sure some of the trouble here is that its a German>English translation.chris_ma wrote:hi Robert,
thanks for the quote from Rodenstock. I can definitely see how getting the right setup for measuring MTF on different lenses could be a tough challenge.
I'm not really sure what they mean with "measure the MTF together with a lens". I mean how would you measure MTF without a lens?
I think they mean to say lens+MTF test bench.
Yes, thats how I take it also, around 90% would be expected.But if I understand them right, they mean that the measured MTF performance would be 90+% of the calculated performance?
so let's say a graph says for a given aperture and lp/mm has a calculated MTF of 50% - the real world performance could be as low as 45% then?
Yes, same here.Not that it creates much of a problem in the real world I guess, calculated MTF are very helpful to me for baseline estimates (I wish rayfact had some for comparison),
LensRentals.com blog has posted some interesting articles on lens sample variation, some brands are a lot worse than others, Canon has the least sample variation (due to factory automation), Sigma is second from what I remember.but it still feels a bit sloppy without giving a tolerance range at the same time (I'm sure some lenses are constructed to much tighter standards then others, having less sample variation and being closer to the theoretical values).
Also I believe I read on LensRentals blog that Imatest MTF results can run +-10% for each MTF run but I don't remember exactly, but that would make a difference in consistency also.
Best,
Robert
-
- Posts: 2627
- Joined: Thu Jan 17, 2013 9:34 pm
- Location: United States
- Contact:
Yes, I think that's true in the case of the Sony 70-200.Lou Jost wrote:I think it might not have been a low-pass filter that improved the resolution, it would more likely have been a sensor cover glass of the right thickness for Sony.
I've read that Zeiss lenses, like the premium upscale Otus line are very sensitive to cover glass thickness when it comes to performance on an MTF bench.
The high-end Otus line is produced in Nagano Japan at the Cosina factory BTW, I saw them when I was there in Feb. this year
A low-pass would have the opposite effect, so a production lens on a camera with a low-pass would probably not meet the theoretical MTF figures published by Rodenstock or most other manuf. Even with Zeiss figures that are MTFs made with an actual lens but this is for a lot of different reasons like:
Is the MTF made a standard test from center to edge at one rotation or is it an average with 4 rotations per lens like LensRentals MTF?
Full spectrum light or a more narrow band?
Is the MTF an average of 10 lenses minimum like LensRentals MTF or is it a single copy.
If the MTF was not an average was the sample lens an average performer or hand picked? LensRentals has tested 100 copies of the same lens and the top 10 or 20% were about 10% better in MTF results than the average lens.
Best,
Robert
On the same theme, measured MTF numbers are a function of the whole system, not just the lens. While I find it hard to imagine that lens manufacturers don't have access to a good enough measurement device whose effect on the measured lens MTF can be ignored, maybe that is harder than we think. If so, the calculation route would be needed. Of course, as mentioned by Chris, that assumes manufacturing tolerances are perfectly zero.
-
- Posts: 2627
- Joined: Thu Jan 17, 2013 9:34 pm
- Location: United States
- Contact:
Yes, the system, including the RAW conversion process.Lou Jost wrote:On the same theme, measured MTF numbers are a function of the whole system, not just the lens. While I find it hard to imagine that lens manufacturers don't have access to a good enough measurement device whose effect on the measured lens MTF can be ignored, maybe that is harder than we think. If so, the calculation route would be needed. Of course, as mentioned by Chris, that assumes manufacturing tolerances are perfectly zero.
An optical bench and staff must be very expensive. Just a couple of years ago I read that Sony did not even own MTF equipment in the US, so they used an outside business for that. Nikon has been budget cutting and downsizing including service staff and out sourcing lens testing/servicing more and more I am told.