The Voigtlander 40mm f/1.2 Nokton is not a hefty lens. At 415g, it's rather light considering its f/1.2 aperture. In comparison, the Voigtlander 35mm f/1.7 Ultron with adapter and PCX is exactly 100g lighter at 315g.
It's designed with four aspherical surfaces as seen in this diagram:
Here are they side-by-side:
(CV 35/1.7 Without PCX 5m)
(CV 35/1.7 With PCX 5m)
The lens has impressive build similar to other Nokton lenses. It's designed for the E-mount and therefore there is EXIF transmission and auto focus magnification. The 10-blade straight aperture is very responsive and allows precise increments in 1/3 stops. It can also be de-clicked. The lens is rather large in diameter but short in length balancing quite well on my Sony A7RII.
Fred Miranda wrote:
This is an ongoing review of the CV 40/1.2 Nokton lens. I received it yesterday and so far, I was able to test it at infinity distance. I also compared it to the Voigtlander 35mm f/1.7 Ultron with a PCX (Plano-convex) front-lens side-by-side.
This thread will be updated with many other tests like OOF rendering, CA, Coma and sunstar rendition.
First Impressions:
The Voigtlander 40mm f/1.2 Nokton is not a hefty lens. At 415g, it's rather light considering its f/1.2 aperture. In comparison, the Voigtlander 35mm f/1.7 Ultron with adapter and PCX is exactly 100g lighter at 315g.
The lens has impressive build similar to other Nokton lenses. It's designed for the E-mount and therefore there is EXIF transmission and auto focus magnification. The 10-blade straight aperture is very responsive and allows precise increments in 1/3 stops. It can also be de-clicked. The lens is rather large in diameter but short in length balancing quite well on my Sony A7RII.
Focus shift
After reading about inconsistent performance results with this lens here and here, I wanted to first test it for RSA (residual spherical aberration). I found that the CV 40/1.2 is not well corrected for RSA and therefore focus shift is an issue.
Below is an example showing focus shift. The image on the LEFT was focused at f/1.2 and captured at f/1.2. The image on the RIGHT was focused at f/4 and captured at f/1.2. Big drop in resolution as you can see.
Lateral chromatic aberration is not well corrected which I consider normal for a small fast lens. Curiously, Lightroom does not apply a profile correction for this lens automatically as it does with other native lenses. When applying auto CA correction in post, this color aberration is not longer a big issue.
Here are is an example of CA at f/4 at the edge of the frame.
This is a difficult lens to test because it has very noticeable focus shift as well as field curvature. It will make a big difference in resolution where you focus and which aperture was used to achieve focus. For optimum results at infinity distance, I would advise focusing towards the mid-field area AND shoot at working aperture with this lens.
On my next post I will post 1:1 crops showing how this lens perform at infinity distance and how it compares to the CV 35/1.7 Ultron with PCX....Show more →
Thanks! Excited to see your further testing. Pretty happy to see that the auto CA correction seems to work so well.
Fred would you mind taking one more image of the lenses side-by-side with the 35/2.8 ZA thrown in there?
Here is the full image thumbnail showing the areas demonstrated at 1:1 magnification. (pixel level)
With all my infinity tests, the subject is focused manually and a "best of three" is used for each lens.
I use the default Lightroom settings except for sharpening which is the same for all lenses being compared.
CA, vignetting and distortion were NOT corrected for both lenses.
Let me start at f/1.7 at infinity distance:
BTW: The 40mm f/1.2 was focused AT shooting aperture and towards the mid-field of the frame. I got much better results using this method. The CV 35/1.7 was focused at center of the frame. (Not much focus shift with this lens)
Very similar performance at f/1.7 and center. However, keep in mind that the CV 35/1.7 Ultron is wide-open.
The CV 40/1.2 is weaker at mid-field and f/1.7. Remember this is the best it will look since the CV 40/1.2 crop was focused at mid-field and f/1.7.
Towards the edge, the CV 35/1.7 is slightly ahead and it's wide open. (keep in mind that without the PCX front lens, the results would be worse). I see some astigmatism on the 40/1.2 crop.
Here are the crops at f/2.8. The CV 40/1.2 improves a lot and actually is slightly superior at mid-field and extreme edges. I still give a slight advantage for the CV 35/1.7 at center though.
I wanted to be clear that the crops presented here show the best results these lenses can achieve, ignoring focus shift and field curvature. This is actually only an issue with the CV 40/1.2 Nokton as the CV 35/1.7 has very little focus shift and is pretty much flat field with the PCX lens.
With the 40/1.2, if one is not careful shooting at "working aperture" the resolution differences can be quite dramatic.
For example. Here is the same f/2.8 crop above. The one of the LEFT was focused at f/2.8 and the one one the right focused at f/1.4. Look at the difference!
teh_rebel wrote:
Fred, appreciate the lens comparison. Wasn't expecting the 40/1.2 to be shorter than the 35/1.7 with adapters/filters. Now I'm intrigued!
yeah, i figured it was a little longer than the 35/1.7, that's great.
Nice work Fred!
Yes I did my tests focussing at taking aperture, but noticed when going through some close focus crops where I had focussed at wide aperture, that there was significant focus shift..
It's a nuisance, since for critical focussing it's a lot easier to do wide open.
Landscape/Tripod shooters would hate it under certain circumstances.
DavidBM wrote:
Nice work Fred!
Yes I did my tests focussing at taking aperture, but noticed when going through some close focus crops where I had focussed at wide aperture, that there was significant focus shift..
It's a nuisance, since for critical focussing it's a lot easier to do wide open.
Thanks for the interesting test results, Fred! I wonder if you see the focus shift at center as well as the corners? In my test I found the corners to shift if focused at the same spot and changing apertures but I couldn't detect the shifting at center (didn't test extensively though and there might have been a much smaller shift at center). I almost always focus at taking aperture, though I didn't do so in my initial aperture series test.
Good to see that the optimum for infinity is f4 and already decent at f2.8, that's how it seemed to me as well so I'm mostly using f4 for long distance shots and f2.8 in low light when necessary.
This is a difficult lens to test because it has very noticeable focus shift as well as field curvature. Resolution changes depending where you focus and which aperture is used to achieve focus. For optimum results at infinity distance, I would advise focusing towards the mid-field area AND shooting at working aperture with this lens.
However, I found that placing the focusing ring at the hard stop combined with f/5.6 and small apertures is optimum for great resolution across the field.
On my next post I will post 1:1 crops showing how this lens perform at infinity distance and how it compares to the CV 35/1.7 Ultron with PCX.
Lateral chromatic aberration is not well corrected which I consider normal for a small fast lens. Curiously, Lightroom does not apply a profile correction for this lens automatically as it does with other native lenses. When applying auto CA correction in post, this color aberration is not longer a big issue.
Here are is an example of CA at f/4 at the edge of the frame.
After reading about inconsistent performance results with this lens here and here, I wanted to first test it for RSA (residual spherical aberration). I found that the CV 40/1.2 is not well corrected for RSA and therefore focus shift is an issue at all distances.
Below is an example showing focus shift. The image on the LEFT was focused at f/1.2 and captured at f/1.2. The image on the RIGHT was focused at f/4 and captured at f/1.2. Big drop in resolution as you can see.
davewolfs wrote:
I almost always focus at the f/stop I shoot.
When shooting landscapes, it's common practice to focus a couple stops wider to better discern your subject. (Also in low light). It shouldn't be much of an issue as long as we are aware of it.
DavidBM wrote:
Nice work Fred!
Yes I did my tests focussing at taking aperture, but noticed when going through some close focus crops where I had focussed at wide aperture, that there was significant focus shift..
It's a nuisance, since for critical focussing it's a lot easier to do wide open.
David,
If you refocused for every aperture, it's not a problem. Shooting distance is definitely a factor but I'm glad you already knew about the focus shift. Did you try hiding that information from us.
Juha Kannisto wrote:
Thanks for the interesting test results, Fred! I wonder if you see the focus shift at center as well as the corners? In my test I found the corners to shift if focused at the same spot and changing apertures but I couldn't detect the shifting at center (didn't test extensively though and there might have been a much smaller shift at center). I almost always focus at taking aperture, though I didn't do so in my initial aperture series test.
Good to see that the optimum for infinity is f4 and already decent at f2.8, that's how it seemed to me as well so I'm mostly using f4 for long distance shots and f2.8 in low light when necessary. ...Show more →
Definitely at center. (Actually all zones)
I don't think a lens can be 100% free of focus shift unless it's overcorrected for spherical aberration. However with some lenses it's pretty much unnoticeable even when checking at high magnification. The 40/1.2 seem to be under-corrected and as long as we are aware of this limitation and shoot at working aperture, it should not be much of a problem.
On the bright side, I've heard that under-corrected lenses actually have more pleasant bokeh. I will test that next.
And no problem reaching infinity, for sure.
