• 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Forums > Back > next PZ lens test report: Zeiss Touit 32mm f/1.8 (Sony E)
#1
Honestly ... the Sony E 35mm f/1.8 OSS in better:

 

http://www.opticallimits.com/sony_nex/884-zeiss32f18e

 

Nice bokeh though.

  Reply
#2
Besides the center it's very disappointing. Looks like the 24MP Sony sensor requires specifically designed lenses to get good corner resolution at wide apertures. The Fuji test of the same lens is much better.

Sony have nice bodies, but a pretty bad lens selection IMO. Except for a few lenses, most of them kinda suck.

--Florent

Flickr gallery
  Reply
#3
Yep, a big question mark with Sony's Nex and it's registration distance, I'm not sure if my Pentax K3 doesn't use the same sensor, (similar if not)  but with it's 43mm approx. reg distance it has no problems with sharp corners and borders, in fact quite the contrary.

 

  Pentax abandoned the APSc mirror-less race after falling at an early sales hurdle with the K01, and in spite of, no VF, slow CDAF, bulky (reg. dist.)  eccentric/ bizzarre looks (Mark Newson's "proactive" design), when coupled with the DA 35mm F2.4 cheapie and the K01's greatly reduced to clear price (together they sold for often less than $450) you got a camera that produces tack sharp images even at wide apertures, coupled with a hoodman style viewer and the AF firmware update things came together. 

 

 When I look at the results of the Sony with the Touit and the amount of money still in my pocket, somehow it doesn't seem to make sense, so all in all  I'm not so disappointed!

  Reply
#4
Quote:Yep, a big question mark with Sony's Nex and it's registration distance, I'm not sure if my Pentax K3 doesn't use the same sensor, (similar if not)  but with it's 43mm approx. reg distance it has no problems with sharp corners and borders, in fact quite the contrary.
 

It can't be registration distance alone - Fuji X registration distance is even shorter than NEX and the Touit performed significantly better on the Fuji. There is probably some issue with the design of the 24 megapixel Sony sensor in combination with the short register of the E-mount. 

 

I wonder if the A6000, A5100 (both 24mp) and the A5000 (20,1mp) have the same problems with corner performance... 
  Reply
#5
Quote:It can't be registration distance alone - Fuji X registration distance is even shorter than NEX and the Touit performed significantly better on the Fuji. There is probably some issue with the design of the 24 megapixel Sony sensor in combination with the short register of the E-mount. 

 

I wonder if the A6000, A5100 (both 24mp) and the A5000 (20,1mp) have the same problems with corner performance... 
Agreed, I looked at the same lens tested on the Fuji and no real problems there, although the border resolution only just creeps into very good territory. It's the Sony sensor that's the problem which is why I said the K3 does not suffer in the same way, giving the impression that the sensor's pixels are not designed for light striking at an angle, one would expect vignetting rather than softening of the image though, all in all a flaw, perhaps that is why Sony is experimenting with a curved sensor. 

  Fuji look as if they have designed their sensor with short distances in mind and have avoided this problem. I'm very impressed with Fuji's general design philosophy.

  Reply
#6
Quote:Besides the center it's very disappointing. Looks like the 24MP Sony sensor requires specifically designed lenses to get good corner resolution at wide apertures. The Fuji test of the same lens is much better.

Sony have nice bodies, but a pretty bad lens selection IMO. Except for a few lenses, most of them kinda suck.
What's so bad about their lens selection? It was fair to say that in the early stages of the system but I think it's fairly comprehensive now. More so if you add third party stuff (although admittedly, too little of that has AF).
  Reply
#7
Quote:What's so bad about their lens selection? It was fair to say that in the early stages of the system but I think it's fairly comprehensive now. More so if you add third party stuff (although admittedly, too little of that has AF).
 

If I look at the NEX tests here at PZ with native Sony AF lenses, here are the results I get:
  • Sony E 18-200mm f/3.5-6.3 OSS: descent from 18 to 50 then pretty crappy after that.
  • Sony E 16mm f/2.8: probably the worse lens tested on this site.
  • Sony E 30mm f/3.5 macro: the worst result I've seen from a macro lens; center is good, but borders are abysmal.
  • Sony E 35mm f/1.8 OSS: good lens.
  • Sony E 50mm f/1.8 OSS: good from f/2.8 on; before that, borders are abysmal.
  • Sony E 10-18mm f/4 OSS: ok lens.
  • Sony E 16-50mm f/3.5-5.6 OSS PZ: one of the worst modern kit lens.
  • Sony E 18-55mm f/3.5-5.6 OSS: I take the previous line back: this one is a champion.
  • Sony E 55-210mm f/4.5-6.3 OSS: ok until 135mm if you close it down by 1 stop; after that it's abysmal.
As one can see, besides the 35mm f/1.8 and perhaps the 10-18, results are pretty poor.

There are also many holes in the lineup compared to what MFT and Fuji offer.

Furthermore, Sony seems to have one of the worse QC of any manufacturers.

Over the years, Sony have been releasing tons of new bodies, yet they didn't address the major issue: quality lenses (whether fast or slow) as well as a line-up covering most people's needs.

Even Samsung has a much better selection of quality glass.

--Florent

Flickr gallery
  Reply
#8
Quote:If I look at the NEX tests here at PZ with native Sony AF lenses, here are the results I get:

 

[etc.]
 

Mostly irrelevant.  These are the combination of lens + sensor system.   You can't make strong conclusions on the lenses alone.  

The confounding NEX-7 sensor not only seems to give relatively poor corner results, but conversely seems to give exceptional center results which makes the graphs look even more skewed.  

 

As long as PZ insists on keeping on testing on the outstanding-and-flawed NEX-7, it's a waste of PZ's time and the reader's time.  Sorry to say.  The results just keep on repeating what we already know, which is that the sensor dominates the lens-sensor system.
/Dave

http://dave9t5.zenfolio.com
  Reply
#9
My experience with the A6000 does not suggest that the situation has improved here. Just to mention.

  Reply
#10
Quote:The results just keep on repeating what we already know, which is that the sensor dominates the lens-sensor system.
 

Quote:My experience with the A6000 does not suggest that the situation has improved here. Just to mention.
 

Well that's quite a problem then.

 

On the one hand, I tend to agree that tests using the NEX 7 may not reflect the full potential of some lenses, due to the sensor's corner-issue. On the other hand, it seems that this is actually what you have to expect using a current-generation NEX (sorry, Alpha  Big Grin ) camera. 

 

I don't know if anyone has done it yet, but it would be interesting to test the resolving power of the same lens on an A6000, NEX 7 and a NEX 6. I fear the corners of the NEX 7/A6000 image could be worse even when downscaled to NEX 6 resolution, due to smearing.

Another question - and this could be the really interesting one with regard to Sony's behavior in this respect - is whether all of this is noticeable without pixel-peeping or printing relatively large. My guess here is that the vast majority of Sony's customers don't notice the problems on the image borders and thus Sony does not care enough to change something. Doesn't make things better though...  :ph34r:

  Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)