Sigma 105 DG DN as tube lens?

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

Moderators: rjlittlefield, ChrisR, Chris S., Pau

mkbn
Posts: 276
Joined: Thu Jan 25, 2024 10:58 pm

Re: Sigma 105 DG DN as tube lens?

Post by mkbn »

Thanks for the discussion, y'all.

I'd still be curious if anyone with the sigma 105 has a chance to try out a larger image circle objective to see what the result is like, but I suppose no one around has yet.

I'm also very curious in general about understanding all the tidbits that go into whether a lens would theoretically vignette as I've read sooome of the considerations here again and again, but I lack the ability to connect the dots of what was said. Maybe I'll make a post about that at some point.

rjlittlefield
Site Admin
Posts: 24009
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Re: Sigma 105 DG DN as tube lens?

Post by rjlittlefield »

mkbn wrote:
Wed Oct 30, 2024 9:43 am
I'm also very curious in general about understanding all the tidbits that go into whether a lens would theoretically vignette
The main issues are the location and size of the entrance pupil, which is simply where the aperture appears to be, when you look into the front of the lens. If the entrance pupil is too small or located too far back, then some or all light rays that the objective is trying to send to the corners of the sensor get blocked by the edges of the entrance pupil. Zoom telephotos are more likely to have a small entrance pupil, located far back especially when set at longest length. The problem is less severe with wider apertures, say 200 mm f/2.8 versus f/4 or f/5.6. It is also less likely with fixed length lenses, which tend to have the entrance pupil farther forward.

The location and size of the entrance pupil is determined by details of the lens design that I have never seen published in any usable form. So unfortunately, the only robust way to know is by testing.

--Rik

mkbn
Posts: 276
Joined: Thu Jan 25, 2024 10:58 pm

Re: Sigma 105 DG DN as tube lens?

Post by mkbn »

rjlittlefield wrote:
Wed Oct 30, 2024 10:18 am
mkbn wrote:
Wed Oct 30, 2024 9:43 am
I'm also very curious in general about understanding all the tidbits that go into whether a lens would theoretically vignette
The main issues are the location and size of the entrance pupil, which is simply where the aperture appears to be, when you look into the front of the lens. If the entrance pupil is too small or located too far back, then some or all light rays that the objective is trying to send to the corners of the sensor get blocked by the edges of the entrance pupil. Zoom telephotos are more likely to have a small entrance pupil, located far back especially when set at longest length. The problem is less severe with wider apertures, say 200 mm f/2.8 versus f/4 or f/5.6. It is also less likely with fixed length lenses, which tend to have the entrance pupil farther forward.

The location and size of the entrance pupil is determined by details of the lens design that I have never seen published in any usable form. So unfortunately, the only robust way to know is by testing.

--Rik
Thanks, I had heard the "where the aperture appears to be" when viewed from the object side a few times, but I still need more optics primer understanding to formulate the questions I have since some of the dots aren't connecting. As in, If I did have a known or measured location for my entrance pupil for a given lens, I don't know enough to draw an outline of ray bundles from the objective to the entrance pupil to convince myself "ah, this is why it's vignetting and this is why it's this percent vignetted." That's on me soon to hopefully ask the right questions or find the right information out to formulate a question.

This has been in my mind the past few months since heartprairie mentioned he would expect a different model of 80-200mm Nikon to potentially vignette less, here: viewtopic.php?f=27&t=47519&p=299202&hil ... ns#p299202

Having looked at the optical bench for my 80-200mm for a bit and not understanding enough to make sense of the optical bench ( https://www.photonstophotos.net/General ... sO,OffAxis ) I had thought potentially I was barking up the right tree that there was a setting for "pupils" to be displayed on the optical bench, but I couldn't convince myself I was understanding anything enough to guesstimate why X or Y objective vignettes. I didn't realize that the details pertinent to determining the locations of the pupils might not be forthcoming enough to be displayed simply on the bench?

I have taken away from the forums that a physical aperture closer to the front may be promising (but not so simple enough to dictate that the entrance pupil is also forward enough) and that wide apertures would ameliorate some of the concerns.

To this end, and hoping for *something* that stacks natively on sony E in body, I started the thread hoping someone had the 105 macro and could try to pop on an objective for me. I Do know that (I believe macro_cosmos?) someone here mentioned testing the sony 70-200mm GM and said it had bad performance vignetting wise, but not sure if that was FF. I have no real huge utility for an expensive tele zoom -- I love regular photography but I don't do something enough to justify spending money on expensive AF glass that doesn't at least aid in the macro/objective use. I had considered the old, gen 1 sony 70-200mm F4 non-macro IF tele zoom as a native sony E tele. But outside of maybe posting on some local forums and asking if anyone wants to lend me some lenses for a day or two, I'm a bit stuck since it doesn't look like there are a ton of sony shoothers. I even considered just getting the ultra cheap and pretty unappealing sony SEL55210 ASPC tele and dealing with degraded image quality and potential vignetting, since this in body stacking adventure was just a potential little "nice to have in my pocket for the field" idea seeing as how I can't get great results at high NA by shifting the focus and stacking in body (IQ degrades off infinity design point at higher NA, as you had a demo thread of with I think a 0.55 NA objective a long while back that I found). As a side note and I will post a quick thread so no one else gets confused: the sony E adapters for non-native lenses don't allow stacking, even if AF works perfectly fine. I don't know if that's a sony limitation (I'd guess so), but I hate it, ha.

I can always go forward with my plans to make some remote motorized focus ring stacking for a non in-body solution and one that doesn't move the camera itself, but I really did like the idea of having multiple options with differing amounts of gear that I'd have to bring for remote work. I was hoping to travel soon to look at fungi and refining my in-field gear to suit the job and portability has always been useful in my short macro journeys so far.

Also, to this end, I had considered whether a fast 135mm like F/2 would have some potential. But again, the sony ecosystem with its myriad of lenses seems to lack some stock in third party, longer AF telephotos. I had also considered the Tamron 35-150mm f2-2.8, or the cheaper 70-180, but they're not IF and would be less useful for my favorite aspect of using the nikon 80-200mm ED: It's easy to find my subject and framing by zooming down to 100MM before popping back up to the 140+ range to remove vignetting. and I'm in the same place with the unknowns even eschewing internal focusing when used with an objective :)

enricosavazzi
Posts: 1515
Joined: Sat Nov 21, 2009 2:41 pm
Location: Västerås, Sweden
Contact:

Re: Sigma 105 DG DN as tube lens?

Post by enricosavazzi »

mkbn wrote:
Wed Oct 30, 2024 11:44 am
Thanks, I had heard the "where the aperture appears to be" when viewed from the object side a few times, but I still need more optics primer understanding to formulate the questions I have since some of the dots aren't connecting. As in, If I did have a known or measured location for my entrance pupil for a given lens, I don't know enough to draw an outline of ray bundles from the objective to the entrance pupil to convince myself "ah, this is why it's vignetting and this is why it's this percent vignetted." That's on me soon to hopefully ask the right questions or find the right information out to formulate a question.[...]
As for the first part (locating the entrance pupil), the simplest and most practical way to do this is by focusing onto and photographing the entrance pupil (if necessary stopped down) though the front element of the lens with a camera with macro lens, and taking a note of (1) the physical distance between the front of the lens being tested and the sensor plane of the camera used for the test, and (2) the focus distance as reported by the macro lens. The difference between the two distances is the optical "depth" of the entrance pupil from the front end of the tested lens. It is actually easier to do than to explain.

The second part (calculating whether there will be vignetting) is not so easy, and depends on lens design and a few other unknown factors. As Rik said, the best way is just to test. However, an entrance pupil located deep within the lens is generally not a good sign.
--ES

Lou Jost
Posts: 6212
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Re: Sigma 105 DG DN as tube lens?

Post by Lou Jost »

I would add a little tidbit about the entrance pupil of the 105mm Sigma. Unlike most lenses, its entrance pupil does not move as the lens focuses. This is an important feature for macro lenses that do focus bracketing, but even some macro lenses designed for focus bracketing, such as the 70mm Sigma Art lens, do not have this feature. I've tested both of them directly. Only the 105mm exhibited no perspective shift during focus bracketing,

rjlittlefield
Site Admin
Posts: 24009
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Re: Sigma 105 DG DN as tube lens?

Post by rjlittlefield »

mkbn wrote:
Wed Oct 30, 2024 11:44 am
Thanks, I had heard the "where the aperture appears to be" when viewed from the object side a few times, but I still need more optics primer understanding to formulate the questions I have since some of the dots aren't connecting. As in, If I did have a known or measured location for my entrance pupil for a given lens, I don't know enough to draw an outline of ray bundles from the objective to the entrance pupil to convince myself "ah, this is why it's vignetting and this is why it's this percent vignetted." That's on me soon to hopefully ask the right questions or find the right information out to formulate a question.

This has been in my mind the past few months since heartprairie mentioned he would expect a different model of 80-200mm Nikon to potentially vignette less, here: viewtopic.php?f=27&t=47519&p=299202&hil ... ns#p299202

Having looked at the optical bench for my 80-200mm for a bit and not understanding enough to make sense of the optical bench ( https://www.photonstophotos.net/General ... sO,OffAxis ) I had thought potentially I was barking up the right tree that there was a setting for "pupils" to be displayed on the optical bench, but I couldn't convince myself I was understanding anything enough to guesstimate why X or Y objective vignettes. I didn't realize that the details pertinent to determining the locations of the pupils might not be forthcoming enough to be displayed simply on the bench?
Yes, that's a good tree. In principle all that's needed is the size and location of the entrance pupil, so if the photonstophotos optical bench is accurate for the lens under consideration, than the pupil it shows should work fine. All you need to do -- again in principle -- is to get the photonstophotos diagram properly scaled and then overlay it with another diagram that shows the off-axis bundle of light as it emerges from the rear of the objective. If the entire bundle lies inside the pupil than it won't vignette; if it lies entirely outside the pupil then the vignette will be pure black; and if the bundle is partially in and partially out, then it will darken in proportion to the fraction of bundle that is outside the pupil. You don't need to trace rays through the lens because that's already taken into account by using the pupil rather than the physical aperture.

As for choosing the bundle you care about, the key thing is to get the right angle. That angle won't depend on details of the tube lens, so for simplicity you can imagine putting a thin lens right behind the objective, placed one focal length away from the sensor. In that case the chief ray to the corner of the sensor will be angled at atan2(focalLength,sensorDiagonal/2) with respect to the optical axis, and the remainder of the bundle will be at the same angle because the objective's image is virtually at infinity. In principle...

I keep writing "in principle" mostly because I have not confirmed this method by physical experiment so I may have messed up something in the description. That would be an interesting experiment, but there's a lot on my plate at this moment so I have no idea whether & when it might get done. If you get to it first, please let us know.

--Rik

Post Reply Previous topicNext topic