hmm, the 21 lux actually looks better with regard to corner smearing than the 21 SEL at least till f/8 or f/11. the lux has that midzone dip though and much worse CA (also it's huge).
We still don't know. Ron's test are for the A7. He still doesn't have an A7r for running the tests. Additionally, I do not think that the Sony lens correction software was used or available.
naturephoto1 wrote:
We still don't know. Ron's test are for the A7. He still doesn't have an A7r for running the tests. Additionally, I do not think that the Sony lens correction software was used or available.
Lens correction software isn't going to correct the sort of smearing we are seeing on some of these examples on the A7. A different sensor might - so there's at least hope until the A7r is fully explored.
It is possible that some of the wider angle lenses showing smearing will perform markedly better on the A7r but I prefer wait for evidence rather than invest in hope.
k-h.a.w wrote:
So what is the reason for the smearing?
Could it have something to do with the width of the adapter?
Thanks.
no. it's refraction through the cover glass on the sensor. a poor adapter will just make things look misfocused. this refraction that varies with angle of light rays causes additional astigmatism that increases towards the corners.
Another 7r report in Chinese posts saying 7R is worse then 7 w/ 25 zm similar to 28g: met a man took zm25 to experiment with mark just next to the shaft redshifts, Coushang see the next 25 focal lengths basic redshift very slight, straight out perfectly acceptable results without amendment. The strange thing is, A7 lighter than A7R point, I wonder if the new technology did not work or because of the number of pixels a drag 36m.
Thanks. That seems to agree mostly with my observations.
I have NEX-5N and NEX-7.
NEX--7 also showed a tad of a color cast with the WATE 16-18-21 after the firmware update.
It didn't show it before to the same degree.
Tim Ashley has indicated that he observed some corner mush with Lux ASPH 50/1.4 on OM-D E-M1.
Interesting if true.
Ron, thanks so much for doing this. To my eyes the Summilux 50 looks decentered on the infinity scene, with the right much softer than the left. If I'm not imagining this I wonder if it's the lens or your adapter?
"I will be testing the A7 and A7r starting in about 9 days and will be using lenses like a 15 Voigtlander, 21 Zeiss, 35 1.2 Nokton or 35 Zeiss Biogon and 50 Summicron and 50 Nokton ASPH. I will also test the new Sony lenses as well or course and will have daily updates here starting on the 28th going through November 1st."
serhan_ wrote:
Another 7r report in Chinese posts saying 7R is worse then 7 w/ 25 zm similar to 28g:
I wonder if they're normalizing for resolution...most aberrations at 100% magnification will look worse on the A7r vs the A7, like the D800 vs D600 (or vs 5DM2/5DM3).
k-h.a.w. - just bear in mind that the NEX-7 and OM-D have smaller and smaller image circles. If there's a little smearing evident already on crop cameras, expect to see more smearing (generally speaking) on cameras with similar back focal length but with larger full frame sensors.
Even on a purpose built for M lens APS-C camera, the Ricoh GXR/M (12.3MP) softness was evident with some lenses in particular wide angle lenses of symmetrical design. All other things being equal (back focal length in particular) I assume they'll perform far worse on full frame sensors.
I just can't see Sony going through heroic measures to tune the A7r for such lenses when they have to support their own lens designs first and foremost. As we see from Ron's example image series the native FE wide angle (currently just the FE35/2.8) perform very well out to the edges, including the FE35/2.8 on at A7. It'll have to do even better on the A7r.
Whatever the lens construction of the FE line up, it apparently does not have the high incident light angle problem many wide angle RF lenses have, so Sony doesn't have to optimise the sensor for this.
If the A7r does treat these problematic lenses better then it's more by happy accident or coincidence (merely the removal of the AA filter?), because FE mount lens performance must always come first.
Whatever the case it is impossible to guess - the A7r sensor is completely different to the A7.
serhan_ wrote:
Another 7r report in Chinese posts saying 7R is worse then 7 w/ 25 zm similar to 28g:
I really wish these guys could distinguish between color-shift and smearing.. They're different issues with different consequences for the final result.
Ron this test is an eye opener. Thank you so much for taking the time and effort to do the tests and for being the first to achieve such a task in record time and to such a high standard!
Looking forward to the a7r tests when you get the camera - sorry we don't have B&H here in the UK unfortunately and I can not use your links but would most definitely do otherwise just glad I did not order the a7.
* Pure Conjecture * It might be that to allow for such high incident angle lenses, they would have had to sacrifice the performance of their A mount lenses using their adapter (or lenses with minimal incident angle.) If that's the case, you can guess which direction they'd go. In any case, I'm hoping my ZE lenses, CV 125mm, etc work well with the Metabones adapter.
Holy poop. Those are atrocious, at least with the lenses I've already got (21SEM and 28 Cron). Wow. Really, really bad smearing. Not just the corners, but the edges, even with the 28mm!
Unless the A7R is substantially better, I'm going to be cancelling my pre-order, sadly. I'll be bummed if I still need to use cornerfix for color cast, but smearing that bad just won't work for me at all. The CV15 on the Nex-7 appears to be a better 21-22mm solution than the 21SEM on the A7.
