85mm 1.4: zeiss, nikon or sony alpha?

Started May 3, 2013 | Discussions thread
jpr2 Forum Pro • Posts: 15,529
re: yes, infamous and notoriously bad QC of Sigmas :(

verybiglebowski wrote:

jpr2 wrote:

verybiglebowski wrote:

jpr2 wrote:

verybiglebowski wrote:

I had Sigma 85mm f1.4 for Canon, but wasn't impressed at all, same as with Zeiss 85 f1.4 ZE.  Those two are only two 85mm lenses from my collection that I sold.

Planar's 85/1.4 bokeh is rather harsh and nervous, was it the same with the Sigma?

But nothing special either. When I compared 85' on Canon, Sigma was a good performer, but neither it was sharpest, or had nicest bokeh, great price, build quality or fastest AF. Some results are still on my Flickr .

I believe (if I remember well), that rather ugly bokeh for a portrait lens was my reason to sell it. With Zeiss it was similar, but Zeiss was also rather soft wide open.

I had really bad luck with Sgma lenses until NEX. 19 and 30 are optically really good and that's what matter most to me. (Build quality was  crap however (version 1), but for that money...)

I do wonder what gives it such a very high rating, the only higher standing are from when mounted on d800, which is hardly surprising given... the respective sensor's diffs. in resolution.

Who knows... I tested only one sample, but it was really rather avarage performer. There is always distance issue with any test, but I tried it at all - close, medium and infinity, and it never went over Canon E 85 f1.2 L II at comparable apertures...

BTW, there are "new data" for this lens here, and if I understand well, those were taken with Canon 5D MII. (same as part of my tests, the rest was on 1D M IV).

Looking at the sharpness - it's 17 instead of 20 in your reference. How's that possible?

...and such bad quality control is really hurting both S. (mostly Sigmas, but Sony's "decenterd lens issue" seems also pretty nasty judging by the number of such reports ).

I'd very much prefer for DXOMark and e.g. photozone.de to report their findings not only as absolute lines per image height/width, but also as a percentage of what is theoretically possible for each given sensor = this way comparisons across different bodies etc. would be much easier (e.g. as they stand now, the DXO scores only make sense for comparing ALL lenses of user's interest each time on exactly the same camera body; be it d800, or 5d3, or whatever...)

jpr2

Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow