It's OK... Philip was likely lifting specifics to support his pro-Zeiss/Sony argument. It could go the other way, like why the 24-70/4 is a rather lackluster performer, or why so many people are complaining about decentered Sony lenses. Bottom line is no manufacturer is perfect, but I don't believe Canon deserves the same criticism they used to get ~10 years ago. Pretty much all of their latest lenses are solid performers and from LR's consistency tests, seem to be built better than in the past, too. Interesting to note that there's an outfit in Germany having the 11-24 custom modified by Canon Germany to work as a shift lens. Apparently it and the new 16-35/4 work quite well at some focal lengths for this purpose and rival, if not better, the performance of the TS-E 24. Pretty good discussion about it here at GetDPI. I may go this route with the 16-35/4 IS for future tilt-shift work, particularly paired with a 24MP+ APS-C body... Imagine the amount of shift possible!
Regarding the SL's colors... something at the back of my mind that I didn't mention when I posted my take on DPR's images, was that there wasn't a specific LR profile available for the camera, so I just used the 'embedded' option, because it was the only option. I do think I see similarities to the M240's colors, but time will tell once there is proper profile support in LR, C1, etc.
Regarding banding in the DPR samples, the worst examples are the school bus scene and the horizontal Javits Center interior. I believe both were shot at ISO 160. Not base ISO. The same camera at other ISOs shows little or no banding. I pushed a number of the files and didn't see anything worrying. So... is it a preproduction quirk? I don't know.
philip_pj wrote:
The latest Canon zoom - the 11-24 zoom - really struggles with the 50Mp camera
Well at least each copy is close to the same as the next.
The FE 35/1.4, 35/2.8, 1635 and 90/2.8 have such copy variation who knows what you will get. Canons manufacturing standard seems well above Sony/Zeiss at the moment.
This is not even to mention there are no high speed zooms at all for the E mount, and which is the excellent Sony mid-range zoom, at any aperture?
To me the quality issues of the native lenses are an outright scandal. I don't like when my friends are forced to go native for best performance, but if the natives are built below the industry standards for top dollar, relative to Canikon, that is one cost cut too far for my taste.
uhoh7 wrote:
Well at least each copy is close to the same as the next.
The FE 35/1.4, 35/2.8, 1635 and 90/2.8 have such copy variation who knows what you will get. Canons manufacturing standard seems well above Sony/Zeiss at the moment.
This is not even to mention there are no high speed zooms at all for the E mount, and which is the excellent Sony mid-range zoom, at any aperture?
To me the quality issues of the native lenses are an outright scandal. I don't like when my friends are forced to go native for best performance, but if the natives are built below the industry standards for top dollar, relative to Canikon, that is one cost cut too far for my taste....Show more →
Perhaps I missed it, but I did not see that the 35/2.8 had been tested yet. The 35/1.4 and 90/2.8 have not fared well though the 55/1.8 did well. It will be interesting to see how the Batis lenses do if they test them along with the Loxias.
Also, some caution is in order as to the significance of these tests as Lensrentals themselves urge caution as they are essentially "beta testing" their methodology and it would seem that software corrections are not considered. Still, I agree that manufacturing consistency appears poor on two of the native mount lenses tested.
Luvwine wrote:
Perhaps I missed it, but I did not see that the 35/2.8 had been tested yet. The 35/1.4 and 90/2.8 have not fared well though the 55/1.8 did well. It will be interesting to see how the Batis lenses do if they test them along with the Loxias.
Also, some caution is in order as to the significance of these tests as Lensrentals themselves urge caution as they are essentially "beta testing" their methodology and it would seem that software corrections are not considered. Still, I agree that manufacturing consistency appears poor on two of the native mount lenses tested.
The 35/2.8 decentering has been remarked upon from it's introduction. But it's hard to really raise the alarm from anecdotes, as there are also bad copies of the 28 cron and 50APO. Roger is doing a service here on the level and perhaps beyonds Erwin Puts' surveys of the M system with the clear descriptions of performance. But Puts could not look at many lenses.
I would like some M lenses users to band together and let Roger test some of our lenses to improve his sample base for the M system, which to be honest, cannot yet be held to account like the Sony lenses.
What would be the lure to send in your precious 28 cron for a lens test? You would get the results and find out where your copy stands in the mean. I know Roger shoots a MM, so maybe he would be up for it
If a prestige lens like the 35/1.4 has such problems of variable performance and poor mechanical design/construction its difficult to believe that any other lenses from the same company for the same product line would be better surely?
Plenty of reports right from the start of some problems with the original 35/2.8 and 55/1.8. And prices for these dont seem appropriate to the cost of 'competitors' from Can, Nikon and the independents.
The only E mount lenses I have are the kit zooms that came for next to nothing with the Nex-6 and the A7, the 16-50 seems to perform ok although I don't expect much from that, and it doesn't inspire any confidence for a long life. The 28-70 seems better made from the outside, but performance is pretty poor, when tested it gave softer edges than the 35 Summicron asph with all the problems that has. I don't know whether its faulty, (although its at least consistently poor!) or just a pretty poor lens. My 17-50 Tamron is very much better (on Nikon D7000) and the list price of the lens is less than Sony charges for the 28-70 on its own, even my friends humble Nikon 18-55 which goes for peanuts is better.
rscheffler wrote:
It's OK... Philip was likely lifting specifics to support his pro-Zeiss/Sony argument. It could go the other way, like why the 24-70/4 is a rather lackluster performer, or why so many people are complaining about decentered Sony lenses.
The E/FE lens-body system is built round a concept of achieving final accuracy in alignment and focus without needing precision in every component. The nominal 18mm mount to sensor register doesn’t have to be perfect (and seems to vary by at least ±0.1mm). All Sony E and FE mount lenses compensate for variations and use free-floating magnetic focus often combined with floating OSS – they don’t have fixed infinity stops. Just as the bodies don’t have to be all that precise, the lenses themselves don’t need to be. As long as both work with the sensor to AF perfectly, the overall system is self-correcting.'
'So, one overall issue is that despite the high cost, the Sony FE/A7 series range of bodies and lenses lacks the precision engineering of past systems and it’s designed that way. When you find one side of your pictures always seems soft with wide-angle, wide zoom or very fast lenses you have encountered the limitations of Sony precision and quality control.'
Thanks for pointing out those images for banding (I missed your comment about it in the original image post; sorry about that). I found some banding, which is surprising and I will track that as a real problem, but it was only visible when pushing exposure in Capture One. Admittedly, those pushes were modest, but I'm a painfully literal guy so I still disagree with the "SOOC" comment made by lightskyland. I also feel compelled to say that I was too generous to DPReview; their "minor exposure tweaks" appear to be at least a stop in places, which I find disingenuous and contrary to the minimal-editing philosophy I attributed to them earlier in this thread. Your assessment of the degree to which they massage the images was definitely more accurate.
'The nominal 18mm mount to sensor ... seems to vary by at least ±0.1mm. All Sony E and FE mount lenses compensate for variations and use free-floating magnetic focus often combined with floating OSS – they don’t have fixed infinity stops. ... As long as both work with the sensor to AF perfectly, the overall system is self-correcting.'
'...When you find one side of your pictures always seems soft with wide-angle, wide zoom or very fast lenses you have encountered the limitations of Sony precision and quality control.' ...Show more →
I don't think this argument is consistent. Loose tolerance of flange distance might lead to problems reaching infinity (or near) focus with adapted lenses, but it wouldn't lead to side-to-side variation in image quality. On the other hand, I suppose the camera/lens system could use optical stabilization's ability to adjust pitch and yaw to compensate for a tilted lens mount, resulting in a minor shift effect, but that would hurt non-OSS native lenses just as much as adapted lenses. Doing so would require some sort of way for the camera to measure and compensate for tilt, none of which are apparent.
Yes, finding decentering reveals limitations of design, manufacturing, and/or QA; but that's tautological. No, this is not due to the FE magnetic rail focus, which actually strikes me as a delightfully elegant solution (if sufficiently precise).
Still, it is a good reminder that Sony has a long history of designing systems that are good enough and cutting corners near extreme specifications in order to make the sweet spot larger. With these cameras -- lens mount, data processing, environmental sealing, and likely other aspects -- Sony has demonstrated that it continues this tradition.
Regardless of these handicaps, I still won't hesitate to shoot my Sony or recommend it to others because I'm almost entirely within that sweet spot that Sony has engineered for, but the SL system is that much more attractive because Leica isn't likely to have intentionally designed such compromises.
That is very interesting... Panasonic and Canon are doing this in their P&S as Leica did with Q eg lens does not cover the sensor, but this is the first time I saw it in a high end zoom esp coming from Leica....
rscheffler wrote:
Sandy at Chromasoft has a brief look at how much software correction is applied to 24-90 images...
serhan_ wrote:
That is very interesting... Panasonic and Canon are doing this in their P&S as Leica did with Q eg lens does not cover the sensor, but this is the first time I saw it in a high end zoom esp coming from Leica....
Look at the Olympus Pro (f/2.8) line of zooms for m4/3rds that many m4/3rds users love. They all do the same thing. Fuji does it with their zooms too, and again those are very well liked. If you have an OVF, this strategy won't work because you will see the errors in the viewfinder and the viewfinder and pictures will look quite different. It is in my opinion a bad side effect of cameras having an EVF. Expect many lenses on EVF cameras to make this tradeoff. Zeiss certainly did heavy distortion correction for the Batis f/1.8.
The E/FE lens-body system is built round a concept of achieving final accuracy in alignment and focus without needing precision in every component. The nominal 18mm mount to sensor register doesn’t have to be perfect (and seems to vary by at least ±0.1mm). All Sony E and FE mount lenses compensate for variations and use free-floating magnetic focus often combined with floating OSS – they don’t have fixed infinity stops. Just as the bodies don’t have to be all that precise, the lenses themselves don’t need to be. As long as both work with the sensor to AF perfectly, the overall system is self-correcting.'
'So, one overall issue is that despite the high cost, the Sony FE/A7 series range of bodies and lenses lacks the precision engineering of past systems and it’s designed that way. When you find one side of your pictures always seems soft with wide-angle, wide zoom or very fast lenses you have encountered the limitations of Sony precision and quality control.' ...Show more →
Thanks. Is there an independent source for the statements made in this reference?
I know the distortion correction which I don't use if I can. I was talking about the lens not covering the sensor... Then it looks like my $200 apsc zoom used on FF...
Steve Spencer wrote:
Look at the Olympus Pro (f/2.8) line of zooms for m4/3rds that many m4/3rds users love. They all do the same thing. Fuji does it with their zooms too, and again those are very well liked. If you have an OVF, this strategy won't work because you will see the errors in the viewfinder and the viewfinder and pictures will look quite different. It is in my opinion a bad side effect of cameras having an EVF. Expect many lenses on EVF cameras to make this tradeoff. Zeiss certainly did heavy distortion correction for the Batis f/1.8.
My small sample of E mount cameras, a nex-6 and an A7, seem to have the the same 'register', I have shimmed three adapters to give hard stop infinity focus with my Leica and a few of my Nikon lenses, and any combination of adapter on either camera gives results indistinguishable from each other. Do the Loxias not have hard stop infinity focussing? If so then they would show up any discrepancy between bodies.
I would be very dissapointed if I found a newer body was different.
I had a chance to play with the SL and 24-90 this morning. I was also able to shoot a bit with the Vario-Elmar R 35-70/4, Elmar-M 24/3.8, T 18-56 (?) and Summilux-M 75. Quite impressive. Size is actually quite nice, shutter quiet and responsive, EVF I wasn't allowed to use my card or keep any shots, but examined several on the LCD as well as a computer. IQ seems very good even with the wide M.
The 24-90 is big and heavy, but the IQ seems to justify it. That being said my 35-70 was certainly more comfortable to use and seemed to work well with the body. Manual focus even without magnification works nicely with the big, bright EVF...magnification makes it even better (don't care for the bottom left button placement for magnify though)
Nov 06, 2015 at 02:00 PM
Steve Spencer Offline • • • • • • • Upload & Sell: On
CVickery wrote:
I had a chance to play with the SL and 24-90 this morning. I was also able to shoot a bit with the Vario-Elmar R 35-70/4, Elmar-M 24/3.8, T 18-56 (?) and Summilux-M 75. Quite impressive. Size is actually quite nice, shutter quiet and responsive, EVF I wasn't allowed to use my card or keep any shots, but examined several on the LCD as well as a computer. IQ seems very good even with the wide M.
The 24-90 is big and heavy, but the IQ seems to justify it. That being said my 35-70 was certainly more comfortable to use and seemed to work well with the body. Manual focus even without magnification works nicely with the big, bright EVF...magnification makes it even better (don't care for the bottom left button placement for magnify though)...Show more →
Could you focus the 75 Summilux even without magnification? If so, that is quite impressive as that lens can be a bit tricky wide open? I do hope they let you move magnification to the upper right button.
Steve Spencer wrote:
Could you focus the 75 Summilux even without magnification? If so, that is quite impressive as that lens can be a bit tricky wide open? I do hope they let you move magnification to the upper right button.
About as well as I can on the M240 ...manual focus is much easier than I expected since the viewfinder is so large, bright, as well as the resolution. Swapped the SL a couple of times with a D810 and the EVF made the D810 VF feel like a crop body VF, the difference is that significant. If you try hard you can provoke behavior that shows that it's a EVF, but in use I'd say it's better than the OVF on my R8, which I'd rank as the best I've used. These are first impressions of course, based on a short (~1hr) session.
CVickery wrote:
About as well as I can on the M240 ...manual focus is much easier than I expected since the viewfinder is so large, bright, as well as the resolution. Swapped the SL a couple of times with a D810 and the EVF made the D810 VF feel like a crop body VF, the difference is that significant. If you try hard you can provoke behavior that shows that it's a EVF, but in use I'd say it's better than the OVF on my R8, which I'd rank as the best I've used. These are first impressions of course, based on a short (~1hr) session. ...Show more →
That's a very interesting advice. Some "summilux" forumers have tried it in Paris. Some of them were very upset about specifications of Leica SL and about his philosophy so different from Leica-M. But, after trying it, advice radically changed. The SL looks of top quality and EVF is so nice even using it in available light. But what about using it in daylight?
Besides, the sensor seems to be made for Leica-M lenses, especially for wide angles. It means no derive chromatics in angles and a good definition without vigneting as seen on Sony A7x. I hope it will be the same for Leica-R lenses on SL.
I've seen quite a few samples now from acquaintances that somehow have gotten them (3 people I know personally have their own Leica SL in hand w/zoom). And I'm really underwhelmed by the quality I'm seeing from them. And quite worried that it's not Leica-esq. Whatever that means
I'm not going to repost their photos, as I don't think it's polite. It's from private feeds etc.. But honestly, I'm a bit worried now that I spent a lot on a camera system I'm not going to be happy with.
Hopefully I can make the camera work better for me, but at this point I'm kinda stressed..
Hello adamdewilde,
Were the samples in jpeg or DNG? Leica DNG are known to be very neutral. It means you need to work on it...
Click on link to feel secure. Don't worry IMHO, this is a great system. https://www.fredmiranda.com/forum/topic/1393428/19#13262414
Whenever you look at a DNG, you're largely seeing the profile used by the RAW converter. I think it's too early to expect any good profiles. The S(007) has a similar problem - good results, but not always as beautiful as the older S cameras. It's easy to worry that it's CMOS vs. CCD, or Leica-has-lost-their-way, but I put it down to beta profiles.
It will get better. Then you can choose LR, C1, or some custom profile as a good starting point.
I wasn't allowed to use my card or keep any shots, but examined several on the LCD as well as a computer. IQ seems very good even with the wide M.
...manual focus is much easier than I expected since the viewfinder is so large, bright, as well as the resolution. Swapped the SL a couple of times with a D810 and the EVF made the D810 VF feel like a crop body VF, the difference is that significant. If you try hard you can provoke behavior that shows that it's a EVF, but in use I'd say it's better than the OVF on my R8, which I'd rank as the best I've used. These are first impressions of course, based on a short (~1hr) session.