Some reversed lenses should be shot through thick glass

A forum to ask questions, post setups, and generally discuss anything having to do with photomacrography and photomicroscopy.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

Lou Jost
Posts: 5944
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

Metabones Speedboosters take the filter stack into account. The Canon-to-MFT Speedbooster for example takes the MFT thick filter stack into account. So perhaps Speedboosters are the best way to use non-native lenses on MFT cameras.

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

Lou Jost wrote:Rik, maybe telecentricity doesn't make a lens escape the problem of filter stacks
Correct. It only addresses the issue of things getting worse away from center due to increasing slants of the ray cones.

--Rik

Lou Jost
Posts: 5944
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

An update to this thread:

The newer 0.71x Ultra Metabones Speedbooster for MFT is designed to take into account the thick filter pack of the MFT sensor, so it effectively converts normal Nikon and Canon lenses (designed to shoot through a 0 to 1mm glass filter stack) to lenses designed for the 4mm sensor stack of MFT. That's in addition to its stated purpose of reducing magnification and increasing speed. So it can solve this mismatch issue for legacy lenses on MFT.

Edit-- I see I already made this point before--sorry! I saw the info for this specific model and got excited....
Last edited by Lou Jost on Thu Apr 20, 2017 5:08 am, edited 1 time in total.

ChrisR
Site Admin
Posts: 8668
Joined: Sat Mar 14, 2009 3:58 am
Location: Near London, UK

Post by ChrisR »

A UV /IR filter remover man told me Canon filter packs can be over 2mm. :smt102
Chris R

Lou Jost
Posts: 5944
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

I was relying on my memory for that. The point is that it is much less than the filter pack for MFT. Of course legacy film lenses, which many of us use for our tube lenses, are designed for 0mm.

harisA
Posts: 513
Joined: Sun Jul 03, 2011 9:31 am
Location: Greece

Post by harisA »

Although i haven't read all the posts here, i carefully read both lensrentals articles mentioned by Lou Jost in the first two posts of this thread.

In the first article
https://www.lensrentals.com/blog/2014/0 ... ed-lenses/

my understanding is,that they tested zeiss otus at 1.4 with and without focal reducer on a sensor that had no glass thickness in front of it.
If we assume that otus has been made with a 2mm thickness correction why it produced a magnificent mtf curved when tested on a sensor with no glass in front of it?

ChrisR
Site Admin
Posts: 8668
Joined: Sat Mar 14, 2009 3:58 am
Location: Near London, UK

Post by ChrisR »

Tangential question - Why do µ4/3rds cameras have such thick glass?
Could it be to render dust out of focus?
Chris R

Lou Jost
Posts: 5944
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

harisA, the graphs in the first article show some comparisons between the Otus with 2mm glass and the Otus with no glass in front of the sensor. The "no-glass" case was worse, though still decent. It looks like the response to glass thickness is non-linear with respect to thickness. Rik will have to tell us why that is.....

Chris, I read somewhere that out-of-focus dust- particles were indeed one reason why the sensor stack was thick.

enricosavazzi
Posts: 1474
Joined: Sat Nov 21, 2009 2:41 pm
Location: Västerås, Sweden
Contact:

Post by enricosavazzi »

ChrisR wrote:Tangential question - Why do µ4/3rds cameras have such thick glass?
Could it be to render dust out of focus?
Increasing the distance between the front surface of the filter package and the sensor chip surface is indeed a way to defocus any dust particles on the glass. However, it is entirely possible to achieve the same effect by mounting a thin glass spaced a few mm away from the chip package window.

The space between the filter package and the sensor window is sealed to avoid contamination. One possible drawback of increasing the spacing is that the air/gas in this sealed space expands and contracts with temperature changes, so a lower distance makes the pressure changes less strong. Hard disk drives have a micropore filter to allow gas exchanges with the environment while avoiding contamination. The cost of such a filter is very low, but might be regarded as significant in mass-produced items, so a thicker filter package might be regarded as the simplest and laziest solution, since lenses can be designed anyway to take the package thickness into account.

I seem to remember looking at a Nikon 1 body (or possibly another small mirrorless) years ago. It had a thin greenish filter (probably UV/IR-cut) quite distant from the sensor, probably for the dust reason. Given the very small sensor, dust shadows projected onto the sensor would be a bigger problem. Since the Nikon 1 was designed to be compatible with Nikon F lenses, the filter package thickness likely could not be increased arbitrarily beyond what was already in use in Nikon DSLRs.

The next factor is the placement of the focal-plane shutter. It would not be a good idea to place the mechanical shutter between filter stack and sensor, because of contamination caused by wear of the shutter. This limits the possible separation between filter package and sensor. With an electronic-only shutter, the problem of course does not exist.
--ES

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

Lou Jost wrote:It looks like the response to glass thickness is non-linear with respect to thickness.
Yes. Wavefront error is linear with respect to thickness, but image degradation is highly non-linear with respect to wavefront error.

For illustration, download any of the image sets at http://www.photomacrography.net/forum/v ... hp?t=23751 and compare wavefront errors 0.25 (slight sag in the MTF curve), 0.5 (huge sag), and 0.75 (catastrophic loss of resolution). Those were done for focus errors, but the results would be similar for spherical aberration.
Rik will have to tell us why that is.....
"Why" is always a hard question. I think of it as relating to the waviness of waves. When you're near the top, you can change phase quite a bit without going up or down much. But when you're near the middle of the slope, changing phase makes a huge difference. What increasing wavefront error does is to move more of the aperture farther onto those sides. Eventually you get to a point where the center and edge of the aperture are out of phase and cancelling each other, while some middle ring of the aperture is at maximum slope. That's the 0.5 case, where things are bad already and getting worse fast.

--Rik

Post Reply Previous topicNext topic