Ok, I'm going nuts to get to the bottom of this, so here we are.

 

I have this weird shadow* that appears on both short sides of the frame (even if in this picture is more evident on the left side), with the exact same weird shape, and with several kinds of adapters but ONLY with long lenses (the Contax 100-300 @ 200 and up and the Leica 180).

 

*I don't want to call it "vignette" because it doesn't look like one, it looks more like a piece of transparent foil was in the way; please keep in mind that to make it this visible I cranked contrast and clarity, otherwise it is a problem only with the sky in that portion of the frame

 

Every other lens is fine, with the same adapters. I've tried flocking the inside of the adapters, using different kinds, I even cleaned the sensor in the off chance it was some sort of halo on it that was causing this, but I'm still getting the "shadow".

 

Any idea of what the hell might it be? My (wild) guesses are:

 

a - that the camera lens mount itself might be a tad too small, and it interacts in some way with the parallel light rays projected by the long legacy lenses (I know, it's far fetched but I'm at a loss for more reasonable explanations).

 

b - that the sensor is reflecting back a ton of light, and this interacts in some way with the back of the adapter and the mount of the legacy lens

 

A sounds good, the last lens element (before the sensor) might be quite recessed in the lens barrel. Does it change with aperture?

A sounds good, the last lens element (before the sensor) might be quite recessed in the lens barrel. Does it change with aperture?

 

 

Not much. It gets a bit more noticeable stopping down, but it is still quite visible at full aperture.

Its an interesting pattern, the corner bits look like the image circle, and the left/right bits look like the inversion of the image circle. Perhaps there is some reflection somehow. Its odd.

 

I have a Novoflex adapter (leica m) and they (novoflex) have gone to some effort to reduce internal reflections, matt paint and what looks like a screw thread on the inner surface of the mount. Perhaps that can be a factor somehow?

Its an interesting pattern, the corner bits look like the image circle, and the left/right bits look like the inversion of the image circle. Perhaps there is some reflection somehow. Its odd.

 

I have a Novoflex adapter (leica m) and they (novoflex) have gone to some effort to reduce internal reflections, matt paint and what looks like a screw thread on the inner surface of the mount. Perhaps that can be a factor somehow?

It's exactly the strangeness of the shape that it's making me wonder about its origins.

 

Thanks for the suggestion, but I'm not sure the adapters are responsible. I tested several ones, and the adapters were already "threaded" inside; now I've added flocking felt inside and even when I put the lens+adapter combo, pointed at a strong light source, directly onto my eye I can't see a single reflection.

 

Again, a sensor might pick up light I can't see, so not definitive proof...but I would have expected reflections to be more of a problem with wide lenses (like the 18mm) or fast ones (loads of light bouncing around), instead wides and fast lenses are perfectly fine.

Which camera body ?

 

I would try:

Camera set at very ISO

Lens cap on

In a somewhat dark room

Use  a flash light to point all around the lens mount/ camera mount / adapter mount to see if any light is leaking by looking at the rear LCD of the camera ( and if it seems to offer the same pattern as you show)

 

Since these are somewhat heavy lenses, you may need to hand hold them the way you would in normal use while moving the flash light to ensure test is valid.

Which camera body ?

 

I would try:

Camera set at very ISO

Lens cap on

In a somewhat dark room

Use  a flash light to point all around the lens mount/ camera mount / adapter mount to see if any light is leaking by looking at the rear LCD of the camera ( and if it seems to offer the same pattern as you show)

 

Since these are somewhat heavy lenses, you may need to hand hold them the way you would in normal use while moving the flash light to ensure test is valid.

 

 

On the A7r.

 

Thanks for the suggestion, but these are not light leaks. I made the test anyway, with a pretty strong 800 lumens LED flashlight after maxing out the ISOs, and I couldn't see any problem in this regard.

try this, turn off electronic first shutter, or turn it on,

 

https://www.talkemount.com/threads/strange-light-pattern-with-telephoto-lenses-—-a7.5925/

 

http://www.dpreview.com/forums/thread/3617567

 

this makes some sense ...

 

 

Sony's native lenses have a short register distance, and the light reaches the corners at an angle,

while your adapted SLR lenses give a straighter path to the light, and the corners of the mount will shadow the sensor.

 

 

but ultimately the problem was solved with a different adapter

 

 

 

The Metabones has a flat black interior with multiple ridges to keep reflections to an absolute minimum.

 

 

 

 

Anyway, you are not the first to have this problem ... good luck.

A7r does not have EFCS (Electronic First curtain Shutter) so you can't disable what is not there. I thought of that and this is why i asked about the body in use.

 

Looking at the OP's website gear list, these 2 lenses are actually the longest in his arsenal ( longest after these 2 mentioned are MD Rokkor 135 and 100mm lenses.

 

What is weird in this situation is that the shadow is actually the same from left to right in spite of the difference in brightness from side to side

 

The only other thing that comes to my mind since the A7r has the pixels at an angle in the corners of the sensor ( to improve mostly wide angle lens performance since the light comes in at a pronounced angle ) is that it may do something clumsy like we see here with longer telephotos since these are projecting the light at an angle that is requiring more parallel pixels.

 

Just tried my A7r with Minolta MD 200mm F 2.8 with cheap chinese MD/E-Mount adapter at F 2.8/5.6/11 and can not see any such "shadow"

I'm kinda partial to your idea "B". 

  

Unsure xactlee why, but there's

something mysteriously saying

to me that "B" makes sense. 

  

Since the problem occurs with

vintage long focus lenses, and

since "B" includes mumbojumbo

about bounced back reflections

of the sensor yadyadayada ....

I offer this strange test:

  

Create a vintage, VERY vintage,

long focus lens, like a pinhole on

a 6-inch tube. Since sharpness

isn't critical here, and you need

enuf light to mimic reasonably

normal lens aperture brightness,

so the pinhole might be more of

a NAILhole. 

  

The main point here is that this

glassless optic can't cause any

bounceback reflections from the

sensor. You could make shorter

and longer tubes, and possibly

an alternative "front element"

with the nailhole way off-center

to see if the shadows shift with

the offset hole .... if the nailhole

optic produces any shadows. 

  

You can even ditch the adapter

for some tests to see if the lens

array on the adapter itself is the

problem. Who needs an 'official' 

lens mount if you got duct tape ?

And a nearly weightless "optic" ?

Did I mention this is fun ? !

try this, turn off electronic first shutter, or turn it on,

 

https://www.talkemount.com/threads/strange-light-pattern-with-telephoto-lenses-—-a7.5925/

 

http://www.dpreview.com/forums/thread/3617567

 

this makes some sense ...

 

 

 

but ultimately the problem was solved with a different adapter

 

 

 

 

 

Anyway, you are not the first to have this problem ... good luck.

 

THANKS!!! It is exactly the same problem. It's the first time I (well, actually you) managed to find someone else with the identical pattern showing. It looks like a Novoflex adapter might be in order...

 

I'm kinda partial to your idea "B". 

  

Unsure xactlee why, but there's

something mysteriously saying

to me that "B" makes sense. 

  

Since the problem occurs with

vintage long focus lenses, and

since "B" includes mumbojumbo

about bounced back reflections

of the sensor yadyadayada ....

I offer this strange test:

  

Create a vintage, VERY vintage,

long focus lens, like a pinhole on

a 6-inch tube. Since sharpness

isn't critical here, and you need

enuf light to mimic reasonably

normal lens aperture brightness,

so the pinhole might be more of

a NAILhole. 

  

The main point here is that this

glassless optic can't cause any

bounceback reflections from the

sensor. You could make shorter

and longer tubes, and possibly

an alternative "front element"

with the nailhole way off-center

to see if the shadows shift with

the offset hole .... if the nailhole

optic produces any shadows. 

  

You can even ditch the adapter

for some tests to see if the lens

array on the adapter itself is the

problem. Who needs an 'official' 

lens mount if you got duct tape ?

And a nearly weightless "optic" ?

 

I run the test you suggested (without the adapter) and I couldn't see any problem at all, not even cranking clarity contrast etc. all the way up.

 

I repeated the test with the adapter, and I still couldn't see a problem.

 

I tried then using only the lid I was employing as a base for the nail-hole (without the cardboard tube to simulate a tele lens), and then I started seeing vignetting from the lens mount itself (different pattern than the one showed above). The same thing happened shooting without a lens or a pinhole at all as soon as I inclined the camera even a tiny bit in relation to the light source (a big sunlit window).

 

So my guess is that is indeed a combination of A + B. A (FE mount too small) provides the vignette in the corners with some lens; B (light bouncing back from the sensor and reflecting either on the bayonet of the legacy lens or on the rear element or inside the adapter, or a combination of these factors) produces the weird looking lateral curved shadows.

 

Like I said, it looks like a Novoflex might be the solution (it should have a sort of small baffling inside to prevent this problem, according to the guy who had the same problem in the post at dpreview).

 

THANKS!!

The only other thing that comes to my mind since the A7r has the pixels at an angle in the corners of the sensor ( to improve mostly wide angle lens performance since the light comes in at a pronounced angle ) is that it may do something clumsy like we see here with longer telephotos since these are projecting the light at an angle that is requiring more parallel pixels.

 

 

 

I think you might be onto something, I didn't take into consideration the displaced micro lenses...

Did you get the adapter after buying the A7r or did you already have it from the NEX era before FF ?

 

Some of the older adapters that were designed before the FF era may have issues.

 

I know i have one i was using with NEX-7 and A7r that does not even fit on my A7II

Did you get the adapter after buying the A7r or did you already have it from the NEX era before FF ?

 

Some of the older adapters that were designed before the FF era may have issues.

 

I know i have one i was using with NEX-7 and A7r that does not even fit on my A7II

 

 

I think I've had the Leica R adapter since the Nex7 (not sure, though), but I bought the latest Contax maybe a few weeks ago (and it was advertised as full frame, besides it doesn't generate trouble with any other lens), so this shouldn't be a problem.

i don´t know if that helps but i had problems with reflections inside  adapter´s with 

 

Mamiya mediumformat lenses, and Contax G90

 

So i have added a velour tube inside the adapter to absorb the reflections. 

 

that was very effective 

 

so you could try something like this.....maybe this helps

 

i don´t know if that helps but i had problems with reflections inside  adapter´s with 

 

Mamiya mediumformat lenses, and Contax G90

 

So i have added a velour tube inside the adapter to absorb the reflections. 

 

that was very effective 

 

so you could try something like this.....maybe this helps

 

DSC04556.jpg

Thanks, as a matter of fact I already tried flocking the adapters using the same material (a sort of velour) camera makers use. Unfortunately in this case it didn't make a difference...