em_dee_aitch
Senior Member
I vote for the Nikon 50/1.4G over the Canon 50/1.2L. The Canon 1.2L has notorious focus shift issues. I have seen some of my Canon shooting friends go back to the 1.4 USM after using the 1.2L for that reason. But to compare it to the 1.4G from Nikon... If you go to a certain Canon-biased website whose initials are TDP, and you look at their lens comparator, you'd see the Nikon 1.4G beating the Canon 1.2L at f/1.4. At stopped down apertures, such as 2.0, 4.0, and 5.6, the Nikon easily beats the 1.2L in the corners, can match it in the center, and they trade blows for balance in the mid frame, with the Nikon being generally better balanced across the frame due to the Canon's weak corners... I've edited many images from the 1.2L when said friends shoot with me, and I have never been impressed with it overall. It turns out brilliant images from time to time, but it misses and turns out mushy way too often.
I have had focus issues with the 50/1.4G to be fair, but they have not been shift issues. Rather, it has been that I find the newest Nikon fast primes (the 50/1.4G and 24/1.4G) to be less capable of focusing on certain textures as compared to older primes and current zooms such as 14-24G and 24-70G. Nikon has no answer for this phenomenon, but I'm hoping they'll eventually address it in camera firmware. I do believe at this point that it is a software issue due to a couple of pieces of evidence from my own testing: First, if you manually focus the lens with "electronic rangefinder" focus confirmation in the viewfinder (uses the same phase detect sensor as regular autofocus), the worst and widest focus misses disappear on the 50/1.4G, which is to say that taking the AF algorithm out of the equation eliminates the worst errors (ironically it may also eliminate the very best instances of focus, depending on how nimble your fingertips are). Secondly, both of these lenses are extremely accurate in low light on any texture when using Live View contrast detect focus, which indicates that their motors are accurate and that they can "see" well; for instance, I have found that Live View contrast detect can focus on a fine wood grain pattern of a closet door from a distance of 10 feet in a room where the exposure was 1/125s f/1.4 @ ISO 6400. That is something that phase detect focus could never do.
All that being said, I think the 50/1.4G is optically a bargain. If you want a better 50, you should consider converting the last version of the Leica R Summilux 50mm. For way less money, you could convert a Summicron, but the gain would be less dramatic and not necessarily visible until stopped down around 5.6.
--
David Hill
http://www.bayareaweddingphotographer.com
San Francisco & San Jose, CA | Austin, TX
Wedding Photographer and Apparent Gearhead
I have had focus issues with the 50/1.4G to be fair, but they have not been shift issues. Rather, it has been that I find the newest Nikon fast primes (the 50/1.4G and 24/1.4G) to be less capable of focusing on certain textures as compared to older primes and current zooms such as 14-24G and 24-70G. Nikon has no answer for this phenomenon, but I'm hoping they'll eventually address it in camera firmware. I do believe at this point that it is a software issue due to a couple of pieces of evidence from my own testing: First, if you manually focus the lens with "electronic rangefinder" focus confirmation in the viewfinder (uses the same phase detect sensor as regular autofocus), the worst and widest focus misses disappear on the 50/1.4G, which is to say that taking the AF algorithm out of the equation eliminates the worst errors (ironically it may also eliminate the very best instances of focus, depending on how nimble your fingertips are). Secondly, both of these lenses are extremely accurate in low light on any texture when using Live View contrast detect focus, which indicates that their motors are accurate and that they can "see" well; for instance, I have found that Live View contrast detect can focus on a fine wood grain pattern of a closet door from a distance of 10 feet in a room where the exposure was 1/125s f/1.4 @ ISO 6400. That is something that phase detect focus could never do.
All that being said, I think the 50/1.4G is optically a bargain. If you want a better 50, you should consider converting the last version of the Leica R Summilux 50mm. For way less money, you could convert a Summicron, but the gain would be less dramatic and not necessarily visible until stopped down around 5.6.
--
David Hill
http://www.bayareaweddingphotographer.com
San Francisco & San Jose, CA | Austin, TX
Wedding Photographer and Apparent Gearhead
