Review: Canon 50L vs f1.4

[kenghor]

I've nearly completed my test comparing the new 50L against the f1.4.

The review may not go down well with some people. Initially I can't even bring

myself to believe my own test results. But after several re-tests, I just have

to admit the fact. Unless the copy that I'm having is a bad copy.

<p>

<a

href="http://www.lens-scape.com/article/50mm-12vs14/50mm12vs14.htm">http://www.lens-scape.com/article/50mm-12vs14/50mm12vs14.htm</a>

<p>

PS:<br>

I've contact the photo.net staffs to make this a permanent article here but have

not received any reply.

[gdanmitchell]

The sharpness results on the f/1.2 are disturbing, aren't they? I suspect there are a number

of people out there hoping that you somehow came up with a very good f/1.4 lens and an

unusually bad sample of the f/1.2.

 

If nothing else, your test does suggest again that the quality of the f/1.4 lens can be quite

good, especially regarding sharpness, color, and contrast.

 

Thanks for your work on this.

[waltz]

Nice test.

 

As far as sharpness goes... Were these manual focus or auto-focus? If AF, Canon only says the lens will be within one depth-of-field. So this may just be lens-to-lens variation. If you owned the L lens, you might be able to improve its performance by having Canon calibrate that lens to your body.

 

The sharpness test indicates filters were being used. Were they also used during the out of focus highlights test? I wonder if this could account for the cut-away edges. Both lenses show some in the bottom-right corner, but the L lens definitely has more of them.

[goran1]

I love my 50 1.4:) That's why I fuss so much about finding a "new 50" for the 1.6:).

 

Maybe you could do a similar test between the 35L and the 28 1.8?

 

Thanks

[carbon_dragon]

These ultra-fast lenses seem to always sacrifice general performance for that last stop or half stop of light. I think anyone who buys this lens or the Leica Noctilux or the old Canon "Dream" 50/0.95 should be buying them because they're going to be using them wide open for subjects that they couldn't get otherwise (maybe dark smoky nightclubs where ultimate sharpness isn't a big deal).

 

For the rest of us, the 1.4's, 1.8's and 2.0's are more than adequate (especially since the slower zooms spend a lot of time on our camera bodies anyway).

[trothwell]

Could you possibly show some more "real world" examples? Some photos of actual people, or using the lens in unusually dark situations where someone who bought one might want to use it?

 

As to the photo.net contributions, I too have submitted articles without receiving any response. I'm guessing (hoping?) that they are just too busy with the site redesign at the moment, rather than simply ignoring us. :-)

[beauh44]

Nicely done! The 50L has the 1.4 on vignetting, contrast, bokeh and a bit on the CA but Canon sure seemed asleep at the wheel on the sharpness. That's a huge difference to my eye.

 

Perhaps there's variation from lens to lens in that regard but for $1600... I think I'll crank up the contrast in PS a little, use ACR along with the vignette and CA sliders when necessary and hang on to my f/1.4.

 

Just for grins it would have been interesting to have seen the 50mm f/2.5 CM thrown in with regards to sharpness and distortion. I find it hard to beat in those areas. That was very illuminating - no pun intended.

[marknagel]

Very nice test, as I said in a post earlier, adding ring usm to the 1.4 is all they need to do!

 

M

[ed_hurst]

Re. Beau's comments about the f2.5 CM being hard to beat in some areas, it would be good to know for sure. All the test data I've seen suggests that the f1.4 is high resolution (in terms of MTF data), but most comments that people make suggests that they believe that the f2.5 is sharper subjectively. Not sure what to make of that one! Any thoughts anyone...?

[bobatkins]

"I've contact the photo.net staffs to make this a permanent article here but have not received any reply."

 

Well you may have tried but you may not have succeeded!

 

Philip has said that there probably won't be any new articles added to the site until the software and hardware upgrades have been done. I can't say I'm in 100% agreement with this policy, but that's the current situation to the best of my knowledge.

[lovcom_photo]

Wee, send your 50L to Canon for calibration, then report back. I wonder if the rush through manufacturing is causing quality issues.

 

I'd love to hear what you think of the post-calibrated 50L....perhaps we all would.

 

If the post-cal tests show razor sharpness wide open then this will be a very good thing...if they can do it with the 35L and the 85L then why not provide wide open sharpness with the 50L?

 

If Canon cannot improve wide-open performance then they really blew it big time :-( and in that case my order for the 50L will be cancelled, and back to the F1.4 for me...but then it too is too soft wide open...could it be the Canon has a disturbing gap in quality at 50mm? This would be horrible.

[bob_mcbob]

Were the filters used for the flare tests?

[larry h.]

I have no real interest in the new lens. I find it interesting to note that William Castleman's review (refernced a few threads back) found the L lens to be better for flare while this test showed the f/1.4 lens to be better for flare. Is this variability or different methodologies? In short, though, condidering the difference in flare is indeterminate, there does not seem to be any benefit to the new lens.

[PuppyDigs]

Nice test Wee. Thanks.

 

I wonder how much quality control varies? My EF 50 1.4 USM was feeble, even at F5.6 was

noticebly soft and low contrast at infinity compared to my L zooms in the same range. It

was totally unusable wide open. Oddly, it was pretty sharp close up but suffered from

softness at infinity. So I sold my EF 50 1.4 USM and was planning buying the EF 50 1.2L

USM when the price drifts down a bit.

 

How's barrel distortion on the 50 1.2? The 50 1.4 has lots of barrel distortion under 2

meters. I was hoping for improvement in this area. When I shot door details they kinda

looked like wooden barrels.

[squareframe]

> there does not seem to be any benefit to the new lens.

 

unless you desire f1.2 -- and based on multiple and conflicting observations and analysis, perhaps it is best to say that decisions to purchase this lens should be based on more than dubious examples of flare and edge-resolution.

 

my personal opinion regarding fast optics, is that it is more of an emotional alignment with the wide-aperture gestalt than anything defined by optical-metrics. otherwise, every f1.8 lens (or stopped down optic) rises to the top .. generally, and those that can't grasp that should be banned from fast-primes and shackled to slow-zooms for the duration.

[bobatkins]

As I said in the other thread on the 50/1.2L, the 50/1.2L will mainly be a "prestige" lens, both for Canon and for most buyers.

 

If Canon actually wanted to make a lens that was sharper than the 50/1.4 at all apertures, with better flare conrol, higher contrast, better bokeh and faster focus, they'd have made a 50/1.4L USM, not a 50/1.2L USM. As an added bonus it would have been cheaper, smaller and lighter than the 50/1.2L as well. If they'd been able to put some sort of IS in it, it would have been more useful too.

[Stock-Photos]

Why would you do a sharpness test with a filter on the lens?

[dan_roche]

It looks like Canon has succeeded in creating the world's first $1600 L-series doorstop. Very soft, terrible CA and awful flare. A plastic kit lens looks better than this monster.

[lovcom_photo]

Given all the negativity one can read about the new 50L on the net, an asking price of even $800 would seem way too much. It seems Canon may have screwed up...still, I wait for my own copy, and if it is soft wide open then I'll send to Canon for calibration and after that if still too soft I'll return it for $ back and buy another F1.4.

 

What troubles me is that Canon got the 35L and 85L right...how could they miss with the 50L in terms of wide open performance...this is the part I just don't understand.

[John Crowe]

Why is one of your sink shots perfectly sharp, while the others are poor? I think it is the first sink shot that is sharp (in the enlargements). May have something to do with some of the other shots being so poor.

[PuppyDigs]

I don't think I'd want to draw any major conclusions about the 50 1.2L from this test. I'd

think you'd want multiple samples worked over by an expert with real test equipment.

There are too many varibles and holes in this test to be sure about many aspects of the 50

1.2.

 

I had a bad EF 70-200 4L USM. It was blurry only on the left side and any test would have

shown it to be subpar for an L optic. After Canon fixed it, it's sharp as tack. As I

understand, it's rare to see such problems with the 70-200, but I was blessed with them.

Perhaps the test 50 1.2 needs to be calibrated.

 

What I really wonder about is how the 50 1.2 does with infinity landscapes? My 50 1.4 was

rather poor beyond 10 feet but would have excelled with a newspaper on the wall test.

[steve_dunn2]

<cite>If AF, Canon only says the lens will be within one depth-of-field.</cite>

 

<p>The test was done with a 5D, according to the EXIF data. If the central AF point was used, then with both lenses, high-precision AF was active, and Canon's specs say that should get you to within 1/3 of DOF. If one of the other eight normal AF points were used, then yes, AF is supposed to work to within DOF only.</p>

[PuppyDigs]

<P>There's another similar test with somwhat different results in the sharpness and flare

areas:</P>

 

<P><A HREF="http://www.wlcastleman.com/equip/reviews/50mm_1.2L/index.htm"

TARGET="_blank">http://www.wlcastleman.com/equip/reviews/50mm_1.2L/index.htm</

A></P>

[saurabh1]

Nicely done. I originally had 50/1.8 and after reading your article about comparison between 1.8 and 1.4 I decided to go for 1.4. I mean 1.4 is a clear winner over 1.8 and I think cost is very well justified too. Now, however, I wonder where Canon is going by introducing such products. My understanding was that they would eventually want everyone to gradually shift from older generation products to their newer products. Along the way they would phase out old lenses. However, this new lens falls way short of providing incentives for me to upgrade from my current 50/1.4. For most people the cost itself is reason enough to shy away from it. I also think 50/1.4L would have been a better lens at this point, but who knows they may bring out f/1.4 II. The current 1.4 is definitely a very good lens and deserves a better USM at least.

 

Irrespective of whether lens is really better or just marginally better than non L counterparts, looking at high costs of 85/1.2L and 50/1.2L I do not see myself having these lenses. I wish Canon had some lens choices in the range between super expensive 1.2L's and could-have-been-better 1.4 and 1.8's... although I should not complain about 85/1.8. From from the reviews I understand it is a very fine lens.

[yakim_peled1]

>> If Canon actually wanted to make a lens that was sharper than the 50/1.4 at all apertures, with better flare conrol, higher contrast, better bokeh and faster focus, they'd have made a 50/1.4L USM, not a 50/1.2L USM. As an added bonus it would have been cheaper, smaller and lighter than the 50/1.2L as well.

 

IMHO there is no need to go to the L territory. Just make a 50/1.4 Mk II with IF and ring USM as the 85/1.8 USM and 28/1.8 USM.

 

>> If they'd been able to put some sort of IS in it, it would have been more useful too.

 

If they'd been able to put some sort of IS in it, it would have been immensely more useful.

 

 

Happy shooting,

Yakim.