Zeiss ZM 15 performance on the A7s

[hiepphotog]

So I have had this lens for almost 2 months and by now, I know the lens well enough. So I finally had a little bit of time to do an aperture series at infinity:

https://flic.kr/p/ptAyxN

(Dpreview doesn't allow Flickr BB code, so I'll just give you the link to the Full-res test).

--

http://www.flickr.com/photos/zeissaholic/

 

[ZhanMInG12]

Thanks for the test.

I'm considering this combination (have the ZM15 and the regular A7). Can you compare the wide-open performance to the A7r? Does the A7r smear with the lens wide open or just red corners?

Victor

 

[Tuloom]

Thanks for putting this up. That's pretty awful. Unusable for landscape as is the CV15.

I wonder if Zeiss will try a 15mm Loxia.

 

[steve vida]

Hi ,thanks for posting these ,there should be a lot of interest in this lens /camera combo.I really like its night sky performance ,what ISO do you use?I also liked the Breaking of dawn capture,as with most 15mm captures their success depends on the lighting conditions in combination with composition,a different way of seeing.Vignetting seemed more noticable in some & not others.Center sharpness is up there from F2.8.I guess you get what you pay for.

Regards,

Steve.

 

[stevo23]

So I have had this lens for almost 2 months and by now, I know the lens well enough. So I finally had a little bit of time to do an aperture series at infinity:

https://flic.kr/p/ptAyxN

I'd like to see it tested with focus not set to infinity - focusing on the foreground for instance. It's hard to tell what we're getting here. And I know the ZM18 Distagon is stellar on the A7.

 

[Tuloom]

So I have had this lens for almost 2 months and by now, I know the lens well enough. So I finally had a little bit of time to do an aperture series at infinity:

https://flic.kr/p/ptAyxN

I'd like to see it tested with focus not set to infinity - focusing on the foreground for instance. It's hard to tell what we're getting here. And I know the ZM18 Distagon is stellar on the A7.

He posted some with the A7R. The sides, not just the corners, are pretty bad. The ZM25 and the CV15 are also bad on the A7 at mid and infinity, but I would have no issue shooting B&W street with either. And Hiepphotog makes the ZM15 work super well for astro. They're just not going to work well for daylight landscape or anything like that. The new FE 16-35 is clearly much better even with so few samples around.

http://www.dpreview.com/forums/thread/3716392

Do you have a full size sample with the ZM18 on an A7x? It would be interesting to see as a comparisons with samples here.

 

[stevo23]

So I have had this lens for almost 2 months and by now, I know the lens well enough. So I finally had a little bit of time to do an aperture series at infinity:

https://flic.kr/p/ptAyxN

I'd like to see it tested with focus not set to infinity - focusing on the foreground for instance. It's hard to tell what we're getting here. And I know the ZM18 Distagon is stellar on the A7.

He posted some with the A7R. The sides, not just the corners, are pretty bad.

But I don't think his subject allowed for a great judge of sides or corners - at least not for me. A better test is to set it to focus in the near field. Or even a flat target.

The ZM25 and the CV15 are also bad on the A7 at mid and infinity, but I would have no issue shooting B&W street with either. And Hiepphotog makes the ZM15 work super well for astro. They're just not going to work well for daylight landscape or anything like that. The new FE 16-35 is clearly much better even with so few samples around.

http://www.dpreview.com/forums/thread/3716392

Do you have a full size sample with the ZM18 on an A7x? It would be interesting to see as a comparisons with samples here.

Not me personally - but I think there was a review on Leica Boss and I think even a couple of others. There is some color shift, but doesn't seem to be any smearing or blur. The color cast seems to occur with the ZM15 on some of the Leicas as well.

Here was Leica Boss:

http://www.leica-boss.com/2013/12/your-a7r-wants-a-wide-angle-lens-what-do-you-do/

 

[ZhanMInG12]

Taking a closer look I think his adapter has some planarity issues. With the f4 and f5.6 images the upper corner is clearly better than the lower, despite them being similar in distance.

I have more confidence for the combo - I suspect the lower corner smear is because he set the lens at infinity, and the adapter made the lower corner go to beyond infinity, thus causing the extra smear. Anyways I've ordered an A7S for my 15mm ZM. Looks like color fringe control is much better than the A7, and I get 1-2 extra stops of usable aperture. The ZM is very usable on the A7 from F8 onwards - and I typically don't open beyond f4.5 or so for landscapes.

 

[ZhanMInG12]

Look at the top right and bottom left corners. The top right is always sharper, even at f2.8.

I think the actual performance is probably somewhere between the two, but I don't know if the OP's lens is decentered or the adapter thickness by a hair on one side. If the lens is decentered then we can expect a good lens to perform like the top right corner, which IMO is good for landscapes at F4 and perfect at F5.6. If it's the adapter then probably 5.6 is still good, but the jury for f4 is out.

 

[stevo23]

Taking a closer look I think his adapter has some planarity issues. With the f4 and f5.6 images the upper corner is clearly better than the lower, despite them being similar in distance.

I have more confidence for the combo - I suspect the lower corner smear is because he set the lens at infinity, and the adapter made the lower corner go to beyond infinity, thus causing the extra smear.

Is it smear, or out-of-focus? I just don't think these samples are good tests.

Anyways I've ordered an A7S for my 15mm ZM. Looks like color fringe control is much better than the A7, and I get 1-2 extra stops of usable aperture. The ZM is very usable on the A7 from F8 onwards - and I typically don't open beyond f4.5 or so for landscapes.

 

[Tuloom]

So I have had this lens for almost 2 months and by now, I know the lens well enough. So I finally had a little bit of time to do an aperture series at infinity:

https://flic.kr/p/ptAyxN

I'd like to see it tested with focus not set to infinity - focusing on the foreground for instance. It's hard to tell what we're getting here. And I know the ZM18 Distagon is stellar on the A7.

He posted some with the A7R. The sides, not just the corners, are pretty bad.

But I don't think his subject allowed for a great judge of sides or corners - at least not for me. A better test is to set it to focus in the near field. Or even a flat target.

It's similar with just about every wide ZM you look at. My ZM25 is the same - smearing along the sides, not just the corners. You can't focus the smearing away.

The ZM25 and the CV15 are also bad on the A7 at mid and infinity, but I would have no issue shooting B&W street with either. And Hiepphotog makes the ZM15 work super well for astro. They're just not going to work well for daylight landscape or anything like that. The new FE 16-35 is clearly much better even with so few samples around.

http://www.dpreview.com/forums/thread/3716392

Do you have a full size sample with the ZM18 on an A7x? It would be interesting to see as a comparisons with samples here.

Not me personally - but I think there was a review on Leica Boss and I think even a couple of others. There is some color shift, but doesn't seem to be any smearing or blur. The color cast seems to occur with the ZM15 on some of the Leicas as well.

Here was Leica Boss:

http://www.leica-boss.com/2013/12/your-a7r-wants-a-wide-angle-lens-what-do-you-do/

Those samples are not large enough to tell anything about the lens on that site. I'm not sure why someone would post a center crop and not a corner or side crop? Centers are not in question.

Ron Scheffler has samples here. The ZM18 smears badly over about 1/5 of each entire side for a total of roughly 2/5 of the frame at landscape distance.

Here is what Ron says about it.

"Some color shift and wide open vignetting is stronger than when on the M9 and set as uncoded. Edge sharpness is poor compared to the M9 and might be usable at f/8 and higher, depending on the application. At f/4 the softness creeps fairly far towards the center of the frame."

I can't see the point buying these for the A7x with Voigtlander releasing a new 15mm, the FE1635 out, and the fact you can get the Zeiss ZE/ZF wide primes instead or even the Canon 17mm T/S. And then there's the $400 Samyang 14mm if you feel like playing Rokinon Roulette for a good copy.

Hiepphottog's own words from his last thread:

"I got the ZM around the same price of the ZE/ZF. I wouldn't pay the current retail since it doesn't play well with anything except on the native camera. However, if I were you, I would wait for the upcoming Photokina before I would buy the ZE 15. I went with the ZM instead because of both the bulk and weight. Otherwise, I would wholeheartedly recommend the ZE (Samyang is a lottery, I got 3 copies so far, and all of them had various degree of de-centering)."

 

[hiepphotog]

Thanks for the test.

I'm considering this combination (have the ZM15 and the regular A7). Can you compare the wide-open performance to the A7r? Does the A7r smear with the lens wide open or just red corners?

Victor

 

[hiepphotog]

But I don't think his subject allowed for a great judge of sides or corners - at least not for me. A better test is to set it to focus in the near field. Or even a flat target.

I would be open to any suggestion. But most landscape photographers I have talked to so far would say that the slanted infinity aperture test is their preferred method to evaluate a lens. To optimize for near-far distance in a landscape shot, hyper-focal is needed, not just focus on the near field (or just infinity for that matter). This would require more intimate testing for each lens.

Though for sure, flat target for close focus (infinity, horizon test is the flatest test target you can ask for) is even more useless for landscape purpose.... Tests from all the major review for WA/UWA lenses are hence useless as well for landscape purpose since they tested at close focus on a flat test chart. Optical bench from lensrental would be ideal.

For side performance, you definitely need to do a little math to figure out the performance on this slanted test.

 

[hiepphotog]

Look at the top right and bottom left corners. The top right is always sharper, even at f2.8.

I think the actual performance is probably somewhere between the two, but I don't know if the OP's lens is decentered or the adapter thickness by a hair on one side. If the lens is decentered then we can expect a good lens to perform like the top right corner, which IMO is good for landscapes at F4 and perfect at F5.6. If it's the adapter then probably 5.6 is still good, but the jury for f4 is out.

 

[hiepphotog]

So I have had this lens for almost 2 months and by now, I know the lens well enough. So I finally had a little bit of time to do an aperture series at infinity:

https://flic.kr/p/ptAyxN

I'd like to see it tested with focus not set to infinity - focusing on the foreground for instance. It's hard to tell what we're getting here. And I know the ZM18 Distagon is stellar on the A7.

He posted some with the A7R. The sides, not just the corners, are pretty bad.

But I don't think his subject allowed for a great judge of sides or corners - at least not for me. A better test is to set it to focus in the near field. Or even a flat target.

It's similar with just about every wide ZM you look at. My ZM25 is the same - smearing along the sides, not just the corners. You can't focus the smearing away.

The ZM25 and the CV15 are also bad on the A7 at mid and infinity, but I would have no issue shooting B&W street with either. And Hiepphotog makes the ZM15 work super well for astro. They're just not going to work well for daylight landscape or anything like that. The new FE 16-35 is clearly much better even with so few samples around.

http://www.dpreview.com/forums/thread/3716392

Do you have a full size sample with the ZM18 on an A7x? It would be interesting to see as a comparisons with samples here.

Not me personally - but I think there was a review on Leica Boss and I think even a couple of others. There is some color shift, but doesn't seem to be any smearing or blur. The color cast seems to occur with the ZM15 on some of the Leicas as well.

Here was Leica Boss:

http://www.leica-boss.com/2013/12/your-a7r-wants-a-wide-angle-lens-what-do-you-do/

Those samples are not large enough to tell anything about the lens on that site. I'm not sure why someone would post a center crop and not a corner or side crop? Centers are not in question.

Ron Scheffler has samples here. The ZM18 smears badly over about 1/5 of each entire side for a total of roughly 2/5 of the frame at landscape distance.

Here is what Ron says about it.

"Some color shift and wide open vignetting is stronger than when on the M9 and set as uncoded. Edge sharpness is poor compared to the M9 and might be usable at f/8 and higher, depending on the application. At f/4 the softness creeps fairly far towards the center of the frame."

I can't see the point buying these for the A7x with Voigtlander releasing a new 15mm, the FE1635 out, and the fact you can get the Zeiss ZE/ZF wide primes instead or even the Canon 17mm T/S. And then there's the $400 Samyang 14mm if you feel like playing Rokinon Roulette for a good copy.

Hiepphottog's own words from his last thread:

"I got the ZM around the same price of the ZE/ZF. I wouldn't pay the current retail since it doesn't play well with anything except on the native camera. However, if I were you, I would wait for the upcoming Photokina before I would buy the ZE 15. I went with the ZM instead because of both the bulk and weight. Otherwise, I would wholeheartedly recommend the ZE (Samyang is a lottery, I got 3 copies so far, and all of them had various degree of de-centering)."

I agree with you there regarding the point of this lens (or any WA/UWA RF lenses, with the exception of the WATE) on the A7 series. The verdict is still out for the new CV 15. The FE1635 is very tempting to me but it lacks the bite of the ZM 15 that I love so much. I don't consider the ZE/ZF since they're not as flexible, huge and too heavy for my use (I shot the A7s + ZM 15 17 hours yesterday without fatigue). Same goes with Canon (plus the inflexibility of filter usage). I would like to love that Samyang but I just couldn't for various reasons.

Like you, I'm looking forward to seeing that Loxia 15 (cross my fingers.)

 

[hiepphotog]

Hi ,thanks for posting these ,there should be a lot of interest in this lens /camera combo.I really like its night sky performance ,what ISO do you use?I also liked the Breaking of dawn capture,as with most 15mm captures their success depends on the lighting conditions in combination with composition,a different way of seeing.Vignetting seemed more noticable in some & not others.Center sharpness is up there from F2.8.I guess you get what you pay for.

Regards,

Steve.

Steve, thank you. Most night was done at 3200 to 6400 (I would avoid 1600 and would skip straight to 3200 because the the noise profile at 1600 is worse). Since I PP'ed all of my pics, I also processed the vignette profile to my liking. Night shots definitely incurred more apparent vignette.

Hope that helps,

Hiep

 

[ZhanMInG12]

I don't have an A7s yet, but my A7 and the 15mm zm IMO performs better than the bottom left corner at f5.6, but not as well as the top right corner. That's why I'm curious - as to whether the A7s is actually better in the corners because of the larger pixels. But for me the top right corner performance at 5.6 is good enough for landscapes - note that some of that will be corrected away once a distortion profile is applied.

My adapter is perfect, though. I hand-picked a Novoflex one by testing it against an MTF chart using a wide-open fast wide lens. I'll try mine on an A7s when it arrives.

 

[Tuloom]

I don't have an A7s yet, but my A7 and the 15mm zm IMO performs better than the bottom left corner at f5.6, but not as well as the top right corner. That's why I'm curious - as to whether the A7s is actually better in the corners because of the larger pixels.

Smearing is similar, color shading is better. It's the smearing that's the real problem. Color issues are easy to fix.

But for me the top right corner performance at 5.6 is good enough for landscapes - note that some of that will be corrected away once a distortion profile is applied.

How is smearing along an entire edge going to be corrected by a distortion profile?

Can you post a full size sample at 5.6 and infinity with the A7.

 

[ZhanMInG12]

The test is fine - another thanks to the OP here - I would do the exact same thing. Horizon tilted to have both edges on the corners of the frame. And I agree that part of the corners at f2.8 is most likely caused by OOF. But the corners are equally bad at f2.8, and not so onwards.

What surprises me is how the two corners differ so much even though they should be roughly at the same distance to the camera. And speaking from experience a 15mm lens at f5.6 has basically everything in front of the camera in focus.







This is what I think is happening - the adapter is skewed to one side, and part of the image is beyond the infinity plane of the lens. When this happens it is difficult to get good sharpness even at f16 - like trying to get sharp results beyond the close focus of a lens.



--
too lazy for DSLRs

 

[ZhanMInG12]

I was talking about the OP's sample. I only use the regular A7 combo at f8, where the only issue is in the extreme corners, which goes away once the 3% distortion is taken off.