Home · Register · Search · View Winners · Software · Hosting · Software · Join Upload & Sell

Moderated by: Fred Miranda
Username   Password

  New fredmiranda.com Mobile Site
  

FM Forums | Alternative Gear & Lenses | Join Upload & Sell

1      
2
       3       end
  

Lens Adapters: There Is No Free Lunch
  
 
philber
Offline
• • • • •
Upload & Sell: On
p.2 #1 · Lens Adapters: There Is No Free Lunch


The fact that the most visible influence of such weaknesses in adapters would be in the corners for WAs suggests that this might not be so bad with the most commonly used adapters, i.e. legacy (FF) lenses used on APS-C or MFT cameras, because the corner part of the images are not part of the image. Or am I barking up the wrong tree?
Also, Roger quotes 10 microns as being enough to induce problems. Are native camera mounts and lenses really that precise in order to avoid this in the first place?



Oct 01, 2013 at 11:17 AM
molson
Offline
• • • • • •
Upload & Sell: On
p.2 #2 · Lens Adapters: There Is No Free Lunch


Roger's findings don't surprise me one bit. Unfortunately, I sold off a number of really nice manual-focus Nikkor lenses before I figured out that they would perform significantly better on a Nikon body than they would on my Canon cameras using adapters...


Oct 01, 2013 at 01:06 PM
RCicala
Offline
• • • •
Upload & Sell: On
p.2 #3 · Lens Adapters: There Is No Free Lunch


Ah, assumptions. I love assumptions.

Actually lensrentals does a fairly huge business in adapted lenses. They probably have 100 adapters of various kinds and in general an adapter goes out with 2 or 3 very expensive lenses (Leicas, Zeiss, wide-aperture primes) making a big rental. The majority of it is for video shooters but a fair amount goes to m4/3 and NEX photographers.

Most people here know I sold Lensrentals (I kept a minority interest, for full disclosure) 2 years ago. I still run their QA and repair department and write my blog. Part of the agreement is that they get no editorial control over the blog content and I get full access to Lensrentals gear. To be honest they have pretty mixed emotions about the blog, so it will probably leave the Lensrentals site in the next 6 months or so.

For me personally, I generally shoot alt for pleasure, and it's mostly legacy alt. When you spend all day working on modern stuff it gets boring. My current project is trying to adapt a nice 1862 Schneider globe lens to a Horseman to 5DIII bellows system.

Edited on Oct 01, 2013 at 01:20 PM · View previous versions



Oct 01, 2013 at 01:12 PM
RustyBug
Offline
• • • • • •
Upload & Sell: On
p.2 #4 · Lens Adapters: There Is No Free Lunch


philber wrote:
The fact that the most visible influence of such weaknesses in adapters would be in the corners for WAs suggests that this might not be so bad with the most commonly used adapters, i.e. legacy (FF) lenses used on APS-C or MFT cameras, because the corner part of the images are not part of the image. Or am I barking up the wrong tree?
Also, Roger quotes 10 microns as being enough to induce problems. Are native camera mounts and lenses really that precise in order to avoid this in the first place?


IIRC, elsewhere we have others discussing the impact that 5 microns makes in third party/OEM lenses (i.e. not adapted).

APS-C/MFT may cut off the corners where it seems most apparent, but that still doesn't change the fact that it is off ... just chopping the most noticeable parts. This is a different thing from achieving the best image possible.

For lenses that natively have weaker corners and sharper centers by design choice, the crop sensor does play nice there, but that really isn't the point ... i.e. tolerances matter and the wider you go, the more it shows (imo).



Oct 01, 2013 at 01:19 PM
JohnJ
Online
• • •
Upload & Sell: Off
p.2 #5 · Lens Adapters: There Is No Free Lunch


Roger, there's something about this report/article that just doesn't gel.

I think the data that you present as 'proof' is too vague, specifically that there isn't an explanation for the 'triggers' where readings change direction abruptly and head south. You explained it as per below, but that just says that the data is being manipulated (in a ‘black box’ kind of way) for a certain end but it doesn't really explain what is happening.

"The settings on the optical bench we used for this series make it look much worse than it really is. While the graph makes it look like the MTF drops to zero, that’s simply because the settings we’re using report zero if focusing distance changes greatly or vignetting become severe."

Of course you are trying to keep things simple but you can’t do that if you want to present evidence to prove a point. In a sense you are telling us to ‘trust you’! I’ve been watching the X-Files for too long, I ‘Trust No One’.

None of the 6 examples (with adapters) really help to illustrate your point other than to show that some things are happening to 'trigger' the measurements but we don't know what those things are. Is an adapter too thick or thin, is the adapter warped, are its springs too weak, is the lens positioning being warped by gravity, are there other testing process/procedure related factors or errors being introduced? None of this is analysed and correlated with test results but a sweeping conclusion is offered. I don’t know what further testing or analysis you’ve done on the adapters but some further information about that would help your cause. Would it help if you could confirm or match the results on the Imatest?

Everyone knows there are bad adapters out there. Everyone. But your conclusion seems to be that there isn’t a single good one either (correct me if my understanding is wrong). I simply think the evidence you’ve provided is insufficient.



Oct 01, 2013 at 01:28 PM
RCicala
Offline
• • • •
Upload & Sell: On
p.2 #6 · Lens Adapters: There Is No Free Lunch


JohnJ wrote:
Roger, there's something about this report/article that just doesn't gel.

I think the data that you present as 'proof' is too vague, specifically that there isn't an explanation for the 'triggers' where readings change direction abruptly and head south. You explained it as per below, but that just says that the data is being manipulated (in a ‘black box’ kind of way) for a certain end but it doesn't really explain what is happening.

"The settings on the optical bench we used for this series make it look much worse than it really is. While the graph makes it look like the
...Show more


John, I don't disagree. This wasn't meant to be an exhaustive report, it was basically a report of me going 'I thought out of 20 adapters I'd find one that didn't make significant changes to measurements on an optical bench and I didn't". It's interesting enough that I'll repeat it on Imatest since that gives us a full 2-D image of the lens, rather an series of liniear cuts across the lens like an optical bench does. That will also give a different distance result since the bench is working at infinity focus and Imatest will be in the 8 to 15 foot range.

I didn't take it further than +/- 40 microns of focusing because anything more than that meant it wouldn't work for me (what appears to be off the chart on the print is simply inability to focus within 40 microns of the center OR inability to get an MTF reading in either tangential or sagittal plane OR too much light fall off to read. Not sufficient for what's happening type investigation, simply sufficient (in my mind) to say "alteration from baseline sufficient to make it unsuitable for testing'.

Similarly, to determine accurately what planes are worsened I'll need to repeat each lens every 30 degrees for 180 degrees (standard is simply 0 and 90 degrees).

I was quite surprised, actually. My expectation was some would be off a bit, but some would be fine and I'd use those for testing. That none were fine surprised me a lot.



Oct 01, 2013 at 02:01 PM
Mirek Elsner
Offline
• • •
Upload & Sell: Off
p.2 #7 · Lens Adapters: There Is No Free Lunch


Krosavcheg wrote:
I am not questioning a good spirit of an investigation, but cynic in me will pay direct attention to the correlation between subject and relevance of business interest.


Roger mentioned that they rent adapters as well, so I don't see any collisions here.



Oct 01, 2013 at 02:31 PM
Mirek Elsner
Offline
• • •
Upload & Sell: Off
p.2 #8 · Lens Adapters: There Is No Free Lunch


Roger, adapter is a piece of metal. Bad adapter can be thinner than it should be, thicker, or it could be thinner on one side and thicker on the other. Your charts show drop of MTF on both ends - how do you explain this?


Oct 01, 2013 at 02:39 PM
Bifurcator
Offline
• • • • •
Upload & Sell: On
p.2 #9 · Lens Adapters: There Is No Free Lunch


Jeff Kott wrote:
Roger Cicala has written a note explaining that not one lens adapter tested by him would be acceptable to use for lens testing.


I thought maybe I was reading the writings of a total tard till I came across this one sentence:


    What Does It Mean in the Real World?
    Like a lot of laboratory testing, probably not a lot.


So at least he's not a total tard... ok.. whew. But that's the only sentence I found useful in the entire article. How many people here are professional lens testers? I'm betting none. Or maybe one out of the hundred thousand FM subscribers...

He should have been a little more straight when answering his own question though: Q. What Does It Mean in the Real World? A. Absolutely nothing at all! Heck the weight of most 50/1.4 SLR lenses can warp the plastic (which positions and supports the mount) more than the tolerances he's talking about

I suppose it's interesting to watch nerds being nerds tho, so I can see the fun in reading an article like this. Don't give it any credibility tho - it's actually meaningless!

I have used a micrometer on the thicknesses of my Chinese cheap-0 $20 adapters and 14 of 16 were right on (within the tolerances of the micrometer itself). One was just not seated right (fixed), and the other is a tilt-adapter which doesn't zero properly.




Edited on Oct 01, 2013 at 03:17 PM · View previous versions


Oct 01, 2013 at 02:57 PM
timballic
Offline
• • •
Upload & Sell: Off
p.2 #10 · Lens Adapters: There Is No Free Lunch


It's one of the reasons I'm surprised that the Contax 21/2.8, requiring an adapter, still commands higher prices than used Zeiss Z* 21/2.8. Surely if any lens would show up adapter weaknesses this would be it.


Oct 01, 2013 at 03:02 PM
 

Search in Used Dept. 



RCicala
Offline
• • • •
Upload & Sell: On
p.2 #11 · Lens Adapters: There Is No Free Lunch


Mirek Elsner wrote:
Roger, adapter is a piece of metal. Bad adapter can be thinner than it should be, thicker, or it could be thinner on one side and thicker on the other. Your charts show drop of MTF on both ends - how do you explain this?


Mirek, I'm speculating, but two things occur to me. 1) a tilt in focus plane (one side thicker than another) would drop both sides since we're looking at one 5 micron focusing slice; 2) decentering drops all four sides roughly equally (you didn't mention mount centration, but that's real too). Simply off-centering the lens would not decenter like a central element, but it could move the central point of the lens.

Edited on Oct 01, 2013 at 03:43 PM · View previous versions



Oct 01, 2013 at 03:42 PM
Bifurcator
Offline
• • • • •
Upload & Sell: On
p.2 #12 · Lens Adapters: There Is No Free Lunch


Krosavcheg wrote:
I am curious about assumptions in the statements.
Would it imply that marginal qualitative differences between "high-end" expensive OEM and adapted 3rd party lenses are invalidated by adaptor quality?
Or it is a generic statement implying literally any OEM glass would outperform adapted 3rd party alternative?

Additionally, wouldn't rigorous testing in ideal environment differ from everyday usage in most cases?

One would hope so. But I think you nailed it. This article is aimed at professional lens testers and not users, photographers, or artists.

He said on his site "I wouldn’t want or expect people to stop using adapters any more than they should stop [using] teleconverters or extension tubes. But lens reviewers and testers shouldn’t be testing lenses mounted to adapters and considering the results valid for what the lens would do in native mount." which is a give-away. And i suppose that's true to some degree. For old SLR lenses like most of us are concerned with here it doesn't seem to hold up in common sense. But it certainly is true that a tester or highly technical reviewer should probably not be basing his findings of say, a modern Nikkor lens from data he gets from it adapted onto a Canon body.




Oct 01, 2013 at 03:42 PM
zhangyue
Offline
• • • •
Upload & Sell: Off
p.2 #13 · Lens Adapters: There Is No Free Lunch




Mirek Elsner wrote:
Roger, adapter is a piece of metal. Bad adapter can be thinner than it should be, thicker, or it could be thinner on one side and thicker on the other. Your charts show drop of MTF on both ends - how do you explain this?

Unless the lens is optimized for flange distance it was designed for?



Oct 01, 2013 at 03:57 PM
artur5
Offline
• •
Upload & Sell: Off
p.2 #14 · Lens Adapters: There Is No Free Lunch


Just to put some numbers in.
Let's imagine a 21/2.8 lens which we use with an adapter in a different mount camera.
Let's imagine also that the adapter's thickness is a bit uneven and skews the lens inwards 0.01mm in one corner.
If we focus a subject located 3mt, away using the centre of the frame, the 'faulty' corner -being 0.01 mm. nearer to the focal plane than the centre- will focus at 3.2 mtr instead of 3.0 mtr.
With a 21mm lens open at f/2.8, even using a very small CoC of 0.015mm. the DOF will give sharp enough images from 2.3 to 4.2 mtr.
Granted, sharp enough isn't perfect but I bet that those possible small tolerances in adapters are almost always less noticeable than the inherent corner softness of most wideangles full open.
Of course if the adapter had a tolerance in the order of 0.03-0.05mm., that'd be a different question.

Edited on Oct 01, 2013 at 08:39 PM · View previous versions



Oct 01, 2013 at 05:10 PM
RustyBug
Offline
• • • • • •
Upload & Sell: On
p.2 #15 · Lens Adapters: There Is No Free Lunch


Well, I've not read Roger's article yet, but I can only speak from practical experience where I've seen the difference a good adapter can make from a lesser one ... as mentioned particularly in WA/UWA. As such, I no longer purchase cheap adapters (midrange @ Fotodiox Pro is my personal level of tolerance ... possible Leitax in the future for my Oly 18).

So, imo, the debate about whether or not it is limited to theory or translates into real world is very simple for me ... I've seen it in real world application. How many microns variance does it take to make a difference ... I can't answer that. But, imo, the tighter the tolerance, the better.

Now, I guess I should go read the article.



Oct 01, 2013 at 05:17 PM
alwang
Offline
• • •
Upload & Sell: Off
p.2 #16 · Lens Adapters: There Is No Free Lunch


I wonder if there is also a difference between heavier, longer SLR lenses (where the weight may cause a cheap adapter to warp out of alignment) and lighter rangefinder lenses. I definitely feel like SLR adapters are prone to a little more play.


Oct 01, 2013 at 05:26 PM
sebboh
Offline
• • • • •
Upload & Sell: Off
p.2 #17 · Lens Adapters: There Is No Free Lunch


Bifurcator wrote:
I have used a micrometer on the thicknesses of my Chinese cheap-0 $20 adapters and 14 of 16 were right on (within the tolerances of the micrometer itself). One was just not seated right (fixed), and the other is a tilt-adapter which doesn't zero properly.


it takes a very shoddy machining job to make an adapter that is uneven thickness. i feel sure 95% (or more) of the $20 adapters you buy will show no measurable deviation in thickness with a standard micrometer. the much more difficult part is machining the mount and insuring that it holds both the camera and the lens perfectly flush against the adapter surfaces. this is the source of most of the issues and even if the adapter is made by the OEM vender to the same tolerances as their camera and lens mounts (say a canon EF mount to eos-m mount adapter made by canon), it will still double the variance in focus plane to sensor distance (which are measurable using a native lens with no adapter). i believe roger previously demonstrated the differences in native mount alignment?




Oct 01, 2013 at 05:37 PM
U.C.
Offline
• • •
Upload & Sell: Off
p.2 #18 · Lens Adapters: There Is No Free Lunch


Maybe not the best evidence that an adapter makes always no difference in the real world, but I think it's pretty decent:

Zeiss 21/2.8 ZF on Nikon


Zeiss 21/2.8 ZF on Canon (with adapter)


Zeiss 25/2.8 ZF on Nikon


Zeiss 25/2.8 ZF on Canon (with adapter)


Zeiss 28/2 ZF on Nikon


Zeiss 28/2 ZF on Canon (with adapter)


Zeiss 35/2 ZF on Nikon


Zeiss 35/2 ZF on Canon (with adapter)


Keep in mind the resolution difference between the Nikon D3X (24MP) and Canon 5D2 (21MP).


And 2 uwa-lenses with native mounts for reference (to show what the difference in resolution does with the numbers).

Zeiss 18/3.5 ZF on Nikon


Zeiss 18/3.5 ZE on Canon



Oct 01, 2013 at 06:24 PM
RustyBug
Offline
• • • • • •
Upload & Sell: On
p.2 #19 · Lens Adapters: There Is No Free Lunch


Which adapter?


Oct 01, 2013 at 06:56 PM
Jeff Kott
Offline
• • •
Upload & Sell: Off
p.2 #20 · Lens Adapters: There Is No Free Lunch


RustyBug wrote:
Which adapter?


Yes, that's the one that Roger's been looking for.



Oct 01, 2013 at 07:03 PM
1      
2
       3       end




FM Forums | Alternative Gear & Lenses | Join Upload & Sell

1      
2
       3       end
    
 

You are not logged in. Login or Register

Username   Password    Retrive password