Observations on focusing an adapted lens

[pdh]

I've been using a Voigtlander Ultron 28mm f/1.9 (M39) on my E-P2 for a couple of months now, and have been more than happy (in fact, dumbstruck sometimes) with the image quality I can achieve ... but there is a downside: I find the image in the viewfinder (EVF) dim, and hard for me to bring to sharp focus accurately.

I also have a number of Konica Hexanon AR lenses (40mm, 50mm & 57mm). By contrast, the images in the viewfinder are sharp and bright and easy to bring to sharp focus.

There are two significant differences that I am thinking about:

1. The FL of the Ultron is 28mm, and I have read many suggestions that using adapted lenses with a FL of less than 35mm on :43: bodies can be problematic -- is this one of the problems?

2. The distance between the rearmost element of the Ultron and the sensor is much smaller than the distance between the rearmost element of the Hexanons and the sensor.

Then today I happened upon a thread with this comment

a used Leica 28mm/2.8 Elmarit ... The GF-1 w/ a Leica M Lens is definitely a compromise, focusing is not always clear and takes some getting used to -- using the EVF I need to employ the MF assist a lot

Interesting ...

 

[usayit]

Interesting indeed.

I will have to see if I notice the same behavior but on an E-PL1 (I'm at work).

I just so happen to have a 28mm Ultron and I can compare it to an adapted Takumar of the same focal length.

 

[pdh]

thanks ... be good to hear your report

 

[~tc~]

With some lenses, the back is not multicoated, so the reflection off the sensor can negatively impact the image quality. Obviously, this was not a problem with film.

 

[usayit]

tc,

Interesting, I learned something new today.


In a dim room, I set my E-PL1 to ISO 3200, shutter to 1/125, and the lens at f/5.6. The in-camera meter read -1 EV for both of my lenses; 28mm f/3.5 SMC Takumar (M42 screwmount) and a Voigtlander 28mm f/1.9 Ultron (LTM). Sample photos from both look pretty much the same. Kinda leads me to believe that the sensor "sees" the same amount of light for both these lenses (of different flange distances).

I did discover something I had not realized. In this dim room, it is a bit too dark to focus easily on the EVF. As such, I turned on the live view boost option on my camera. It helped a lot and brighted up the EVF for focusing. Now here's the part that I think is a bug. With the EVF in boost mode, you'd expect that magnifying would maintain the same EVF would maintain the same EVF brightness. It does not. So in this dark room with EVF view boost on, I only get the "boosted" live view when not magnifying... kinda seems wrong.

In conclusion, I am not seeing the same EVF brightness difference as pdh. I have read reports of problems with focal lengths less than 35mm on these non-retrofocus lenses. From my understanding, this only impacts the far corners of the frame; color shifts, less focus, vignette. This was attributed to the angle at which the light exits the rear element of the lens and reaches the sensor.

 

[robertro]

Perhaps two things:

Wide angle lenses can seem harder to manually focus accurately, just because they are often used to photograph broader subjects (eg landscapes with many small details) that are inherently harder to focus on, whereas telephoto lenses are more often used for larger/more distinct subjects.

I have used wide angle lenses that were not highly contrasty and they were difficult to focus accurately. This could be made worse if the focus movement was too long and it was challenging to rock focus back and forth.

I have not used the Ultron, but my favorite wide angle has been the Tamron Adaptall 28mm F2.5, because it has high contrast and a focus movement that lends itself to manual focus.

I tried a few lenses to test focus accuracy; this was an unscientific test where I focused on a couple of objects (wooden toy, painting, photo of Santa, clock face) wide open in poor interior lighting using normal view, and then went to magnified view to check my focus:

Vivitar 19mm/3.8: 75%+
Nikon 20/3.5 75%+
Vivitar 24/2 <25%
Vivitar 28/2 <25%
Tamron 28/2.5 <25%

What did this suggest to me? That wide angle lenses with wide apertures are REALLY hard to focus accurately without using the magnifier feature, but that I can get reasonable accuracy at f3.5 or smaller. With the magnifier on my E-PL1, it's simple to focus any of the above.

I don't know if this has any bearing on your VC lens since these are likely all retrofocus lenses, but I had fun!

 

[dsteady]

PDH -- The 28mm Elmarit I'm using is pretty new (to me) so it could really be me getting the hang of it. I do get more confident with focus the more I use it, but in shaded high-to-low contrast moments I find the Panny EVF really challenging for manual focus. It's just not designed to handle complex light I think. I would love to see a more sophisticated EVF for the GF-1.

I've taken to using the MF assist (magnifier) most of the time that I use the Elmarit. the results appear to be getting better, but I've been wondering if it has something to do with the 28mm focal length being too compromised by the EVF. With time and patience the lens and camera perform great together. They just don't do fast work so well.

daniel

 

[dsteady]

PDH -- The 28mm Elmarit I'm using is pretty new (to me) so it could really be me getting the hang of it. I do get more confident with focus the more I use it, but in shaded high-to-low contrast moments I find the Panny EVF really challenging for manual focus. It's just not designed to handle complex light I think. I would love to see a more sophisticated EVF for the GF-1.

I've taken to using the MF assist (magnifier) most of the time that I use the Elmarit. the results appear to be getting better, but I've been wondering if it has something to do with the 28mm focal length being too compromised by the EVF. The LCD does a better job resolving focus without the MF-assist, but I don't like shooting that way. With time and patience the lens and camera perform great together. They just don't do fast work so well.

daniel

 

[Luckypenguin]

Wide angle lenses can seem harder to manually focus accurately, just because they are often used to photograph broader subjects (eg landscapes with many small details) that are inherently harder to focus on, whereas telephoto lenses are more often used for larger/more distinct subjects.

I would agree that this is the main reason it seems harder to manually focus wider lenses. I regularly use MF 24mm and 28mm lenses and it would be rare that I don't check the focus using the magnified view.

 

[dsteady]

Perhaps two things:

Wide angle lenses can seem harder to manually focus accurately, just because they are often used to photograph broader subjects (eg landscapes with many small details) that are inherently harder to focus on, whereas telephoto lenses are more often used for larger/more distinct subjects.

....

I see the truth in this, so I'm not disputing it, but in my case I've been using the Elmarit 28mm for street shots and it's still pretty challenging. But then, part of the challenge is that street shots often require fast focusing and that's just not easy when toggling between MF-Assist and normal view. With a streetscape things get much easier.

Thanks for your posting your chart above. I assume the percentage numbers correspond to how frequently you were in focus with a given lens?

 

[pdh]

Wide angle lenses can seem harder to manually focus accurately

perhaps true, but then the even wider 17mm kit pancake I often use (and with MF too) I find as easy as the Hexanons ... so my idea is that it may be something about the design of particular lenses that produces a significant effect when allied with a :43: body?

 

[dsteady]

perhaps true, but then the even wider 17mm kit pancake I often use (and with MF too) I find as easy as the Hexanons ... so my idea is that it may be something about the design of particular lenses that produces a significant effect when allied with a :43: body?

I'm not so sure yet. I did some test shots with mine today and am seeing more and more to like about it. And, when focused through LCD screen it is accurate and easier to read without the MF-assist magnifier. Honestly, I think if I had the Oly Pen EVF I'd be in heaven with this set up.

 

[robertro]

Focus accuracy with manual focus wide angle lenses....

I see the truth in this, so I'm not disputing it, but in my case I've been using the Elmarit 28mm for street shots and it's still pretty challenging. But then, part of the challenge is that street shots often require fast focusing and that's just not easy when toggling between MF-Assist and normal view. With a streetscape things get much easier.

Thanks for your posting your chart above. I assume the percentage numbers correspond to how frequently you were in focus with a given lens?

Yes - I had 4 subjects - focused on each two times - and tabulated the results. I'd say it's a VERY rough test, but it gave me some food for thought. It convinced me that when I use a manual focus wide angle lens faster than f2.8, I should verify focus with magnification or bracket my focus.