Sigma 50mm f1.4 review

[mfbernstein]

It's great to see that the lens reviews are continuing apace, and I'm looking forward to seeing the rest of the lineup. Thanks to Andy for the excellent work.

That said, the title (on the main page and the news item) is currently a bit misleading since to my knowledge there is no Sigma 50mm f/1.8.

Cheers,

P.S. I presume that Sony's 50mm f/1.4 will be added to the mix once Sony has an FF body to test on? FWIW, I think it'd be worthwhile to include the Panasonic 25mm f1.4 as well, as it offers the same normal view for 4/3. On the other hand, seeing as it's darned difficult to get ahold of, that may present its own set of difficulties.

--
MFBernstein

'Wilderness is not a luxury but a necessity of the human spirit.' - Ed Abbey

 

[Andrew Westlake]

That said, the title (on the main page and the news item) is
currently a bit misleading since to my knowledge there is no Sigma
50mm f/1.8.

Oops, sorry about that. It's fixed now, but might take a few minutes to propagate though the system.

The Sony will certainly get a review, although quite possibly in the guise of the Minolta version we use for our camera tests.

--
Andy Westlake
dpreview.com/lensreviews

 

[mfbernstein]

Geez you guys are fast! Cheers,

--
MFBernstein

'Wilderness is not a luxury but a necessity of the human spirit.' - Ed Abbey

 

[Prognathous]

The Sony will certainly get a review, although quite possibly in the
guise of the Minolta version we use for our camera tests.

Are you sure that the Minolta is 100% identical to the Sony?

Also, in the comparison table in the first page of the review you wrote that the Pentax 50/1.4 doesn't have image stabilization. This is technically true, but the lens does enjoy the stabilization built into all current Pentax DSLR bodies. It would have been more fair to at least put an asterisk near that "None" and explain this situation in a footnote, or alternatively to just replace the "None" with "Body-based".

Other than this bit, thanks for an excellent review! ;-)

Prog.

 

[Andrew Westlake]

Are you sure that the Minolta is 100% identical to the Sony?

Clearly it isn't (they are cosmetically different for a start) so we'd call it a 'Minolta 50mm F1.4 review' to avoid any confusion. BTW it would be surprising for Sony to have recomputed the optical design so quickly after buying Minolta's camera business, and then not release it as a new lens, especially given Sony's undeniable marketing expertise.

Also, in the comparison table in the first page of the review you
wrote that the Pentax 50/1.4 doesn't have image stabilization. This
is technically true

'Technically true' is good enough for me.

--
Andy Westlake
dpreview.com/lensreviews

 

[Peter Gregg]

Nice review.

There are a couple of comments.

First, the 77mm size is being listed as a bad thing and labeled RED while the other 50's are obviously smaller with the smallest filter ring being labeled GREEN.

This can actually be backwards depending on your view. If everything has to be the smallest possible size then it would be bad. But for professionals to have to carry different filter sizes then the 77mm is pretty much THE standard and I am grateful they didn't make an odd size here so a step ring is not required. IMHO this lens is more likely for people with pro level lenses and looking for a high end 1.4 lens, especially the Nikon user that doesn't have a USM type lens available until now.

Second, a very important feature these days for some cameras - especially the pro level camera - is the passing of distance information. i didn't see that mentioned here so I am guessing that distance information is not passed along to the camera body in a usable method for ETTL and ITTL calculations. That would be a good thing to know even if it is obvious.

Regards,
peter

--
'Life is good - eternal life is better'

 

[Makinations]

'Technically true' is good enough for me.

"You are technically correct. The best kind of correct."

• Bureaucrat 1.0, Futurama.

 

[Andrew Westlake]

Thanks for your input, I've modified the table to make if more consistent.

As for distance information, Sigma make no mention as to whether they have incuded it, so it's probably safe to assume they haven't been able to reverse-engineer that protocol yet. But we have no hard information one way or the other, and I wouldn't like to add speculation to the factual section of the review.

--
Andy Westlake
dpreview.com/lensreviews

 

[Prognathous]

'Technically true' is good enough for me.

But is it useful enough for the reader? After all, a 50/1.4 costs several times more than a 50/1.8 and only offers a measly 0.6 stop advantage in low-light capability. The Pentax body-stabilized 50/1.4, on the other hand, offers a whopping 2.5 stop advantage (at 1/6" sec), so it should at least be worthwhile of a footnote.

Prog.

 

[Ernst Dinkla]

That said, the title (on the main page and the news item) is
currently a bit misleading since to my knowledge there is no Sigma
50mm f/1.8.

Oops, sorry about that. It's fixed now, but might take a few minutes
to propagate though the system.

The Sony will certainly get a review, although quite possibly in the
guise of the Minolta version we use for our camera tests.

--
Andy Westlake
dpreview.com/lensreviews

Add another third party lens: the Voigtländer Nokton 58mm f1.4 SL II that looks like it is in the same category but with its own characteristics. Based on the Sigma and Nokton as tested by Photozone.

Ernst Dinkla

 

[Andrew Westlake]

No, in-body stabilization is a camera feature and therefore covered in camera reviews, and whilst we're fully aware of your personal crusade to enlighten us, it won't change our approach on this in the slightest.

--
Andy Westlake
dpreview.com/lensreviews

 

[Prognathous]

No, in-body stabilization is a camera feature and therefore covered
in camera reviews, and whilst we're fully aware of your personal
crusade to enlighten us, it won't change our approach on this in the
slightest.

I have no doubt that your approach will change the second Nikon or Canon does the inevitable and adds in-body IS. It will take a while though until they do.

Prog.

 

[Andrew Westlake]

I have no doubt that your approach will change the second Nikon or
Canon does the inevitable and adds in-body IS.

A tediously predictable response. Why do you think we'd change our approach if Canon or Nikon adopted in-body IS? We'd still deal with it in camera reviews (and not lens reviews).

--
Andy Westlake
dpreview.com/lensreviews

 

[Two Truths]

Are you sure that the Minolta is 100% identical to the Sony?

Clearly it isn't (they are cosmetically different for a start) so
we'd call it a 'Minolta 50mm F1.4 review' to avoid any confusion. BTW
it would be surprising for Sony to have recomputed the optical design
so quickly after buying Minolta's camera business, and then not
release it as a new lens, especially given Sony's undeniable
marketing expertise.

They didn't.

To my knowedge; the "new" Sony 50mm F1.4 D was one of two lenses which had been updated by Minolta before Sony took over.

Minolta had already announced the Minolta 35mm F1.4 D G, which was a lens which never actually made it to market. It never even announced the 50mm F1.4 D but rumours are that it was about to.

However; as best anyone can tell, the only differences between these "new" lenses and their old versions is that they both have ADI distance encoding. As well as giving more advanced flash exposure, there are also rumours that anti-shake is more effective with ADI lenses than non-ADI lenses.

I am guessing then that you will be keeping to your discontinued Minolta 50mm for some time then!

--
Stuart / the Two Truths
http://www.flickr.com/photos/two_truths/
http://two-truths.deviantart.com/gallery/

 

[BJN]

Doesn't the in-camera stabilization work with third party lenses in the Pentax mount?

Whatever benefit Pentax has with in-camera stabilization, it's not worth a footnote in a lens review if all lenses for that mount benefit from the feature.

--
BJ Nicholls
SLC, UT

 

[Octane]

Maybe I'm the only one, but when I saw the different shaprness in the test depending on the camera used and the explanation for it in the article, I was pretty dissapointed. Basically it means that each test is a combination of lens and camera resolution plus sensor size, which in the end defeates the idea to compare lenses. The whole idea of being able to compare lenses doesn't work any more.

And it makes me see all these lens reviews in a different light now.

--
-------David-------
http://flickr.com/photos/childish/

 

[Art Vandelay II]

Doesn't the in-camera stabilization work with third party lenses in
the Pentax mount?

Whatever benefit Pentax has with in-camera stabilization, it's not
worth a footnote in a lens review if all lenses for that mount
benefit from the feature.

--

Yes, in-body stabilization works no matter what lens is mounted on the camera. Which is why I refuse to buy Canon or Nikon then pay yet another premium for a lens with built in stabilization, even if in-lens stabilization is slightly better.

As far as a footnote goes; I don't think it should be mentioned in this comparison since none of the lenses have it. However, if a comparison of 300mm primes comes up and you highlight green for Canon & Nikon since they are stabilized and make Sony, Pentax & Olympus versions red since they are not then I would have an issue with it. In that case I think clearly an asterisk is needed so not to give Canon & Nikon undue praise for a simple difference in philosophy.

I did notice one thing about this comparison that could make people think you give undo favor to Canikon though. In your 70-200mm comparisons you DON'T highlight the price in red of the Canon & Nikon 70-200mm's even though they are over twice the price of the Sigma & Tamron. However, in this comparison you DO highlight the price of the Sigma in red because it is twice the cost of the others.

That said; I love the lens reviews. Keep up the good work.

 

[Andrew Westlake]

However; as best anyone can tell, the only differences between these
"new" lenses and their old versions is that they both have ADI
distance encoding. As well as giving more advanced flash exposure,
there are also rumours that anti-shake is more effective with ADI
lenses than non-ADI lenses.

And I think it's fair to point out that these are no more than completely unsubstantiated rumours with (as far as I know) no hard data whatsoever to back them up; and the idea that ADI info could make any significant difference to IS does seem to run counter to the accepted wisdom of how stabilisation systems work (they measure and correct for tilt angle, which is fundamentally independant of subject distance).

I am guessing then that you will be keeping to your discontinued
Minolta 50mm for some time then!

Yes; as the Canon 50mm F1.4 data shows, these older designs can happily provide enough resolution for the 21Mp 1Ds III and 12Mp 450D, with room to spare at the apertures we use for camera reviews.

--
Andy Westlake
dpreview.com/lensreviews

 

[Ehrik]

I know two types of CA:

Lateral, where the colours are focussed on the same plane but at
different magnification, giving one colour at a dark-bright transition
and the complement colour at the opposite transition, when going
in a radial direction.

And longitudinal, aka axial, aka bokeh CA where the colours are
focussed on different planes, giving a halo of one colour in front
of the focus plane and the complement colour behind the focus plane.

But the Sigma 50/1.4 review first talks about a "complete lack of axial chromatic aberration" and then later about "quite stong bokeh chromatic aberration" so it seems DPR distinguishes between these? I'm a bit lost here.

Just my two oere
Erik from Sweden

 

[Two Truths]

I totally agree, their resolving power is more than enough, that is true. Distortions and vignetting should also be under control.

However; what about chromatic aberrations... none of the 50mms except for the Sigma are designed to reduce CAs and because of their ancient & simplistic design they are reasonably prone to them.

How about adopting the Sigma 50mm as a standard lens that is available on most mounts?

That would remove one element of randomness from your tests.

--
Stuart / the Two Truths
http://www.flickr.com/photos/two_truths/
http://two-truths.deviantart.com/gallery/