Sam Bennett said: Dismissing the 50/1.2 with it's 5 times higher price because it's not 5 times sharper ...
Is it actually sharper at all (in your tests)? I mean f>2.8 aperture range in particular. I know, the lens should be only used wide opened. What about CA though? Could you please tell me which findings of William Castleman's test you might disagree with?
Sam Bennett wrote:
You guys are funny. Dismissing the 50/1.2 with it's 5 times higher price because it's not 5 times sharper than the 50/1.4 wide-open is like dismissing the 1D MKII because it doesn't have image quality 5 times better than the XT.
You obviously missed the point. Price has nothing to do with it for me personally. If the lens price was $500 how that would sharpen it wide open?
I will not use the 50L if my 85L and 35L perform much better just as I'm not using my 50 1.4 much for that reason. I'm not desperate for this focal length and will not buy a soft lens just because I can afford it.
mfurman wrote:
Is it actually sharper at all (in your tests)? I mean f>2.8 aperture range in particular. I know, the lens should be only used wide opened. What about CA though? Could you please tell me which findings of William Castleman's test you might disagree with?
I don't necessarily disagree with William at all, I just disagree with how you guys are using his findings to justify writing off the lens. William verified that the lens is A) sharper than the 50/1.4 (even at f/1.2), B) has better bokeh than the 50/1.4. I can't recall if he commented on the color rendition, but in my findings it's significantly better than the 85/1.8 (which was similar to my 50/1.4, iirc) and is what I would come to expect from L glass like the 35L and the 135L, both of which have great neutral color rendition.
The only thing in William's review that is dismaying, and appears to be backed up by my findings is that focus accuracy may not have improved at all - that, genuinely, is a bummer. But I never expected this lens to be significantly sharper than the 50/1.4 - that lens once stopped down to even f/1.8 was pretty sharp to begin with. What I did expect was for it to have better bokeh (which it does), better colors (which it does), better build (which it does), weather sealing (which it does), and better focus consistency (which it may not). So... am I ready to write of the lens right now? No.
EDIT: In terms of CA, what are you talking about, specifically? Birefringence (purple fringing) or lateral/transverse CA?
FWIW my specific need is better focusing in lower-light situations, sans flash. I couldn't tell whether the new lens handles that situation better than the f/1.4 or not, but it seems logical to deduce it doesn't based on the other data.
Fovea35 wrote:
William, thanks for the review.
I can deal with the f/1.4 slight flare under Photoshop, and the f/1.2 extra sharpness is not critical for me.
Did you test autofocus accuracy and consitency in lower light?
If there is a difference, it could make me plunge.
Xavier.
This is a important point that I didn't examine. The autofocus test was done with fairly bright modeling lighting. Low light autofocus of the 1.2L might be better.
responding to Lord-malones's question...would i buy the lens if there was no internet i would say this....i do understand that the internet and pixel peepers seem to find faults that never would have been seen or likely noticed years ago,i grant that...on the other hand the internet if you can sift through the crap and find the good stuff is a great source of information ....many times you will see negative reviews in one place and ignore it because there are tons of positive ones elsewhere...in this case the reviews all seem to be similiar...good but not great,not as good as expected for the price etc....that is my concern in general that there seems to be a consensus forming
Sam Bennett said: In terms of CA, what are you talking about, specifically? Birefringence (purple fringing) or lateral/transverse CA?
I only meant the results of William Castleman's tests - see that "Sharpness and CA table".
I did want this lens very much and that is why I feel so strongly about its "problems". I should also add that I understand the challenges of building an f/1.2 lens but my question would be what was the problem with just introducing 50 f/1.4 L (mkII) if you want)? 50 f/1.4 is excellent in f/2.8-5.6 range (if it focuses properly) and I use is almost as reference for lens sharpness at f/4.0.
wcastleman wrote:
This is a important point that I didn't examine. The autofocus test was done with fairly bright modeling lighting. Low light autofocus of the 1.2L might be better.
Needless to say, the focus accuracy question is a really tricky one. As I've said repeatedly, my lens couldn't focus accurately on a test chart to save its life. But out in the field, the lens was easily more accurate and consistent than my 85/1.8 which I've always regarded as having good accuracy and consistency. So, it's a real head scratcher at this point.
Just for the record, I hold your tests in very high regard and as I said early I don't really disagree with your tests at all. I just think some people may be drawing some conclusions from them a bit hastily.
Sam Bennett wrote:
I just think some people may be drawing some conclusions from them a bit hastily.
The 1.2L is a great lens, but very expensive. If I shot a lot of weddings and group portraits with a prime 50, I would have kept it. I use the lens mostly to shoot gymnastics, and slight autofocus variations make a bigger difference in image quality for my use than the small differences in optimal lens sharpness. So it wasn't worth keeping....for me.
Canon should be praised for going through the trouble and expense to build a better 50mm prime. They have. Perceptions of cost/benefit ratios will vary between photographers and influence the decision to purchase the lens.
wcastleman wrote:
The 1.2L is a great lens, but very expensive. If I shot a lot of weddings and group portraits with a prime 50, I would have kept it.
Yeah, this is what I figured. Ultimately if a lens isn't going to get used for what its best suited to, then it doesn't make sense to keep it.
wcastleman wrote:
I use the lens mostly to shoot gymnastics, and slight autofocus variations make a bigger difference in image quality for my use than the small differences in optimal lens sharpness. So it wasn't worth keeping....for me.
Well, this begs the question: did you try shooting gymnastics with it? Objective tests are fine and useful to a certain degree, but ultimately the proof is in the pudding. That's what makes the 50/1.2L so frustrating for me. When I try to "test" it under the sorts of controlled conditions like you have, the lens doesn't perform well - it in fact performs badly, but when I use it for what I bought it for... it performs superbly. Very confusing, very frustrating. The question really is, where is it going to start performing badly? That's not an open question I'm prepared to live with, so that's why I'm going to try to get Canon to "fix" the issue.
wcastleman wrote:
Canon should be praised for going through the trouble and expense to build a better 50mm prime. They have. Perceptions of cost/benefit ratios will vary between photographers and influence the decision to purchase the lens.
Question for those who have used and tested the 50L
What is your impression of manual focus in low light conditions?
Mine will be here this coming Friday. I ordered this for use on the 5D for low light candid work and walk around type photography.
Ideally I would have purchased a Leica M8 for this type of work, but cannot afford the stiff price tag for body and lens, plus all the reported problems with it.
As a reference, I find the 35L to be accurate both in af and mf situations.
I'm holding off for more reports on the issue of flare. Has the 1.2L lens really improved? My 1.4 is quite useless in outdoor daylight, even in tree shaded conditions where I need the speed.
Paul Gardner wrote:
I'm holding off for more reports on the issue of flare. Has the 1.2L lens really improved? My 1.4 is quite useless in outdoor daylight, even in tree shaded conditions where I need the speed.
William's tests seem to show that pretty conclusively. I can't make a direct comparison to the 50/1.4, but in my experience the flare control of the 50L is excellent. At the RollerGirls bout I shot I was faced with Tungsten light been shone towards me at each corner. This posed major problems for the 85/1.8 but the 50/1.2L took it in stride showing little flare and excellent contrast.
Summary: The superbly built EF 50mm f/1.2L lens provides modestly sharper images with better out of focus blurring (bokeh) at f/1.4 and f/2 and is much more resistant to flare and internal reflections than the EF 50mm f/1.4 lens. This improved performance come at the cost of increased size, weight and a greater than 5-fold higher cost. The f/1.2L lens does not focus faster or more accurately than the f/1.4 lens. Corner sharpness is lower and chromatic aberration is greater with the f/1.2L lens.
Thank you for the excellent test !
So, my take away message is that 50/1.4 is the sharper lens ( Actually much sharper ) for the apertures f/2.8 and smaller ?
Pondria wrote:
So, my take away message is that 50/1.4 is the sharper lens ( Actually much sharper ) for the apertures f/2.8 and smaller ?
I don't think this is the case. Or at least I don't think this is the case for most applications. The 50/1.4 has better corner performance at a few apertures settings, but most people looking a photographs don't look at the far corners. I doubt at f/4 and beyond that one could tell which photograph was taken with which lens in a print.
Edited by wcastleman on Dec 18, 2006 at 09:53 AM GMT
wcastleman wrote:
I don't think this is the case. Or at least I don't think this is the case for most applications. The 50/1.4 has better corner performance at a few apertures settings, but most people looking a photographs don't look at the far corners. I doubt at f/4 and beyond that one could tell which photograph was taken with which camera in a print.
I am focusing on the 50% center weighted MTF comparison. The 1.2L lens outperforms the 1.4 lens at f/1.4 and f/2.
At f/2.8 and f/4, the f/1.4 lens outperforms the L-lens.
Between f/2.8 and f/5.6, which happens to be my favorite range with this focal length, 1.4 version outperforms significantly.
Pondria wrote:
I am focusing on the 50% center weighted MTF comparison. The 1.2L lens outperforms the 1.4 lens at f/1.4 and f/2.
At f/2.8 and f/4, the f/1.4 lens outperforms the L-lens.
Between f/2.8 and f/5.6, which happens to be my favorite range with this focal length, 1.4 version outperforms significantly.
50% MTF is higher between f/2.8 and f/5.6. However, once a threshold is reached in 50% MTF, I'm not certain that you can see differences in prints. Any differences in the grossly enlarged images from f/4 to f/8 are barely distinguishable. Even though the absolute 50% MTF value differences are small at f/1.4 and f/2, I think the image sharpness differences at these apertures would show up in prints.
Few organizations or individual clients ever seem to notice the relatively small differences in image sharpness that we photographers notice as being important. I am confident that the athletic associations that use my gymnastics images could care less whether I shot their gymnast at f/2 with the 1.2L or the 1.4 lens even if the 1.2L were slightly sharper. Most of their applications never tap into that level of image sharpness.
If I spent significant time shooting backlit subjects (e.g., wedding couples against chapel windows), the flare control of the 1.2L would be a much bigger advantage.
I am visiting the threat regurally and so far failed to find a picture which truly impressed me. Reading through FM I have seen a lot of 85L pictures, which are really unique and cannot be made with otherwise excellent 85/1.8. 50/1.2 does not impress me in a similar manner. So far it is a letdown. Or may be it takes time until people will learn how to use 50L potential.
The question really is, where is it going to start performing badly? That's not an open question I'm prepared to live with, so that's why I'm going to try to get Canon to "fix" the issue.