How does the Canon 16-35/4 work on Sony A7rII and what about other wide angles?

[ErikKaffehr]

Hi,

I have an A7rII and I find I need a good wide angle zoom. For some reasons (*) I am quite interested in the Canon 16-35/4.

Unfortunately I have seen some tests indicating that the 16-35/4 does not work very well on the A7rII.

The first one was by Ken Rockwell: How well do Canon lenses work on the Sony A7R II with an adapter: Sony A7R II vs Canon 5DSR, now I don't trust Ken Rockwell very much, especially as he extrapolates from Leica M wideangles not working well on the Sony.

But, Tony Northrup made a direct comparison between the Canon 16-35/4 and the Sony 16-35/4. My take on that was that the Canon performed better than the Sony, except on the short end near edges corners. Also they found that the Canon lens was much sharper than the Sony at the centers.

On the other hand, I am pretty sure Stefan Steib finds that the 16-35/4 Canon is usable on the A7rII with his HCam Master TSII.

Now, it is easy to say "cower glass" but Sony has around 2 mm of cover glass, while Canon's seem to vary between 1.0 - 2.0 mm ( LensRentals.com - Sensor Stack Thickness: When Does It Matter? ). Now, cover glass makes for a difference, Lensrentals found in their tests of Sony FE-mount lenses that 2 mm yielded best results (they tested 0, 1, 2 and 3 mm), but the FE-mount lenses probably have the outlet pupil closer to the sensor than the Canon lenses.

I also have a Canon 24/3.5 TSELII lens, I have a sort of mixed experience with it on the Sony A7rII. Mostly it works well and I don't really see issues on resolution tests, but I find it doesn't work well on thing like tree tops.

My best friend has a Canon 5DIII and we have run some shots on the Canon 24/3.5 TSE LII when it arrived, and we found that it was not that great in the corners.

http://echophoto.dnsalias.net/ekr/Articles/Shoots/TSE24_examples/20150725-_D4A0775.jpg

But this image was quite OK (it is stitched from two images with vertical shift):
http://echophoto.dnsalias.net/ekr/Articles/Shoots/TSE24_examples/20150725-_D4A0893-Pano.jpg

Shooting the 24/3.5 TSE LII on the Sony I would say the results are a bit similar. Here are some samples, all on the Sony A7rII if not indicated otherwise:
http://echophoto.dnsalias.net/ekr/Articles/Shoots/TSE24_examples/

Just to point out, DPReview did also test the Canon 24/3.5 TSE LII on Canon:
http://www.dpreview.com/lensreviews/canon_24_3p5_tse_c10/4

http://www.dpreview.com/lensreviews/canon_24_3p5_tse_c10/Samples/issues/0T7H8227-002.jpg

Hopefully we get some interesting feedback on this. Would be interesting for potential buyers of wide angles for the A7rII.

(*) The reasons I am interested in the Canon 16-35/4 over the Sony 16-35/4 is that it can be used on the HCam Master TSII I have. It is also less expensive than the Sony lens.

Best regards
Erik Kaffehr

 

[chrisd]

Back in 2014, before the FE 16-35mm was available, I used the Canon 16-35mm f/4 with a Metabones III (then IV) adapter on my A7R. After I purchased the FE 16-35, I kept the Canon for a period of time and compared the two with the idea that the Sony may go back if it didn't match the Canon. I did a number of test shots at varying apertures and subject distances, and compared them in LR, and I found that they were very close in terms of IQ (primarily sharpness). Zooming in to 100% on my MBPR, I could see the Canon had the slightest of advantages, but I could not tell the difference zoomed out to the full image. I decided to keep the Sony and sell the Canon because the AF and overall usability of the Sony was much more important than the slight loss in sharpness. I realize you asked about the A7RII, not the A7R, but I would be surprised if the results were any different.

 

[craigosh]

I'm interested in this as well. Unfortunately can't get hold of a rental lens to try. My 24mm ts-e seems to be uneven across the frame, its certainly sharper to the right and top, than the left and bottom. I don't think I noticed until I tried the Actus as I was shifting and stitching wider views diagonally with the lens, so think the stitched images where using all the sharp parts of the image circle. Stitching with the Actus though I was framing round the image circle so it showed up the softer areas in the results.

From what Stefan has mentioned about the 16-35mm it sounds good, especially as I could use something wider than the 24mm with shift, but not as wide as the 17mm. But keen to see some results, which Stefan hasn't posted yet.

 

[turtle]

Just a quick one on the 24mm TSE II - ensure you stop down to f8 or more. I find f10-11 optimal if doing large amounts of rise. Even f8 is inferior when used this way. I find my copy to be sometimes inferior to my 24-70 L II at 24mm and sometimes notably superior. I think it is due to field curvature in both lenses manifesting differently, but there is no doubt that under f8 the zoom has better edges and corners.

From what I see the Canon 16-35 f4 is slightly superior overall, but the usability issue will either matter to you or not, depending on application. For a keen landscape shooter prepared to take the additional weight, I'd go for the Canon.

 

[ErikKaffehr]

Hi,

I decided to order the Canon 16-35/4 and I assume it arrives this friday. I would assume I will have some initial tests pretty soon. I need to do significant testing while I have the option to return the lens.

Best regards
Erik

Just a quick one on the 24mm TSE II - ensure you stop down to f8 or more. I find f10-11 optimal if doing large amounts of rise. Even f8 is inferior when used this way. I find my copy to be sometimes inferior to my 24-70 L II at 24mm and sometimes notably superior. I think it is due to field curvature in both lenses manifesting differently, but there is no doubt that under f8 the zoom has better edges and corners.

From what I see the Canon 16-35 f4 is slightly superior overall, but the usability issue will either matter to you or not, depending on application. For a keen landscape shooter prepared to take the additional weight, I'd go for the Canon.

 

[ErikKaffehr]

Hi,

I got my 16-35/4 Canon lens today. Some early impressions:

• Autofocus works on the Metabones IV, but it is a bit slow.
• I have looked at corner sharpness and astigmatism. I would say corners look OK. Significant lateral colour, but that is easily corrected in Lightroom. Astigmatism is observable, but not bad.

Hopefully I can do some real world shooting tomorrow!

Best regards
Erik

 

[ErikKaffehr]

Hi,

No weather today…

I did a quick and dirty comparison of corners between three of my lenses at 24 mm. The lenses were 16-35/4, 24/3.5 TSE, 24-70/2.8.

For this test I used one of my favourite test targets a couple 1$ bills one of Norman Koren's sinusiodal charts and Bart van der Wolf's test target. This time I added a USAF test target oriented to detect astigmatism. Differences in cover glass thickness would lead astigmatism and the USAF target would detect it. Focus was on the USAF target.

The first image shows a comparison of the two Canon lenses 16-35/4 at f/8 on the left:

What I can see here that the USAF target is quite a bit sharper on the TS. Astigmatism is present on both lenses. Moving inward the Canon 16-35/4 gets sharper than the TS lens. That may depend on different curvature of field. Curvature of field stuff is tricky because it can vary with focusing distance and aperture.

Next up was the Canon 16-35/4 Sony and 24-70/2.8 ZA pair, Canon on the left:


Here the Sony lens does show very little astigmatism, reinforcing my impression of it being fairly good at 24 mm.
The figure below shows the USAF target, enlarged (200%). Astigmatism is best observed on group -2, undergroups 2,3 and 4 are clearly resolved on Sony, on the canon the right column is resolved but not the left one. This indicates that astigmatism is present.


A general observation is that all three lenses are resolving near sensor limits. This being very extreme corner on 24 mm zooms indicates that no ways a 42 MP sensor outresolves good lenses.

I did no comparison at 16 mm, but the Canon lens still works well. In the samples below (shown at 100%) the left one shows lateral chromatic aberration while the one on the right has corrected chromatic aberration in LR.


This is quite extreme testing for weakness, specially for weakness possibly induced by the cover glass. Real world images may yield different results. But, now I am pretty sure the lens is a keeper.

Best regards
Erik

Hi,

I got my 16-35/4 Canon lens today. Some early impressions:

• Autofocus works on the Metabones IV, but it is a bit slow.
• I have looked at corner sharpness and astigmatism. I would say corners look OK. Significant lateral colour, but that is easily corrected in Lightroom. Astigmatism is observable, but not bad.

Hopefully I can do some real world shooting tomorrow!

Best regards
Erik

 

[ShooterSteve]

I was very close to buying the Sony 16-35 a few days ago but controlled myself. But now I'm again undecided. Having reliable and enhanced AF, (especially for video), on the Sony lens is very handy. I just hate to give up image quality....

 

[ErikKaffehr]

Hi,

Regarding AF, I just tested under indoor conditions. It seems to work but quite slow compared to my 90 macro.

Best regards
Erik

I was very close to buying the Sony 16-35 a few days ago but controlled myself. But now I'm again undecided. Having reliable and enhanced AF, (especially for video), on the Sony lens is very handy. I just hate to give up image quality....

 

[uhoh7]

Keep in mind copy variation on the Sony 1635 is high. A really good copy seems pretty good.

For ultimate UWA performance there is new Batis 21/2.8 and WA the Batis 25/2. On the A7r2, again assuming good copies, these lenses are unbeatable on the A7r2 because they are designed for the coverglass and high MP.

Otherwise still the Leica WATE is best wider than 21 with the CV 15 v3 pretty close.

28 is another FL without a top shelf solution on the A7r2. The FE 28/2 is pretty good. The new Otus would be very good but so huge. The Zeiss Hollywood seems good also.

Hope you guys take the tests out doors and make your foregrounds distant: the real test of how the cover glass is liking the lens is a true infinity shots, e.g. a town from a hill or roof top. Indoor tests are pretty useless, I fear

PS love to hear Tony rant about that batteries and single card slot. Boy, he is very slick with the youtube reviews, I must say.

 

[ErikKaffehr]

Hi,

I think zooms have a benefit. If I need a 19 mm lens a 21 mm won't help me. This is just my way of shooting. In real world if I need a slightly wider lens than what I have I would stitch.

One of the main reasons I am interested in the 16-35/4 is that it can be used with tilts and shifts on my HCam Master TSII.

I have only made some testing in corners, with focus on the target, to check for astigmatism. I did see some, same with the 24/3.5 TSE LII I also have.

Weather has not been very cooperative, but I did shoot some outdoor shots yesterday. I would say it performed decently across the field.

I don't publish those test shots, as they were not so well executed, but still they hold some promise.

By the way, you are wrong about the value of test chart shots… But I have also made some outdoor shots.

As a small comment on Tony Northup's rants. MFD's used to have a single CF-card slots and were still regarded as highly professional cameras. But, if you are used to dual slots, I can see that you miss it once you are on a single shot.

Battery live is not much of a problem for me, but I am a slow shooter. On my last week of travel with the Sony I used about 1.5 batteries each day.

Best regards
Erik

Keep in mind copy variation on the Sony 1635 is high. A really good copy seems pretty good.

For ultimate UWA performance there is new Batis 21/2.8 and WA the Batis 25/2. On the A7r2, again assuming good copies, these lenses are unbeatable on the A7r2 because they are designed for the coverglass and high MP.

Otherwise still the Leica WATE is best wider than 21 with the CV 15 v3 pretty close.

28 is another FL without a top shelf solution on the A7r2. The FE 28/2 is pretty good. The new Otus would be very good but so huge. The Zeiss Hollywood seems good also.

Hope you guys take the tests out doors and make your foregrounds distant: the real test of how the cover glass is liking the lens is a true infinity shots, e.g. a town from a hill or roof top. Indoor tests are pretty useless, I fear

PS love to hear Tony rant about that batteries and single card slot. Boy, he is very slick with the youtube reviews, I must say.

 

[ErikKaffehr]

Hi,

I finally got out to get some outdoors shots I care to show. What I have found that the corners are quite OK.

This was shot at 16mm and f/8.

These are extreme corner before and after correcting chromatic aberration:


At the centre it produced pretty hefty moiré which indicates that the lens outperforms the sensor:


These images were processed in Lightroom CC with "Landscape" sharpening.

So, I don't see the issues reported by KR. Tony Northrup has found some edge/corner softness compared to the Canon 5DsR and also compared to the Sony 16-35/4 lens. My samples neither comfirm or contradict this, but the corners are good enough for me and I would say they are quite OK.

I have also tried some shifts. At 16 mm I could shift just a few mm ( less than five) and corner sharpness was nothing to write home about. At 24 mm it was possible to use something like 10 mm of shift and sharpness was acceptable. I will clearly not sell of my 24/3.5 TSE LII…

I have not tested tilts, yet, but I guess that tilts will be quite OK.

How to stop down on the HCam? Simples way is to shoot an exposure with lenscap on and remove lens during exposure.

So, what I have found is that the lens is definitively a keeper. It doesn't blow my socks off, but it does the job it is supposed to do.

Best regards
Erik

 

[cdavis32]

Hi Erik,

Did you remove the plastic cover from the rear of the lens? Just wondering because HCam reports closer to 15mm of movement in the 24mm range.

I just received my 16-35 and tested it with a metabones adapter(while I wait for my actus from CI and canon board from SK Grimes). I'll have everything later this coming week, and interested to compare it to the 24 TSII. Straight on, it looks to equal a decent copy of the TS, but maybe not a great copy in terms of sharpness. There is definitely some strong CA's in the corners in the 16mm range.

I'm not sure it can replace the 17 and 24 in terms of absolute quality, but when you think about the flexibility the zoom offers along with the $3k in savings, it starts sounding better and better!

 

[Stefan Steib]

I would definitely recommend to use f11 or even f11,5 when shifting.
(even if diffraction may cause a bit less contrast and a little bit less sharpness, this is easy to recover by software).
And yes - the plastic cover of the 4/16-35mm has to come out for full shift performance.

Regards
Stefan

 

[cdavis32]

Thanks Stefan... I appreciate all your help, and you being a trailblazer with the use of movements on the sony! I tested it at f11 and f16 and both were plenty sharp. F16 had some diffraction, but still looked great.

I see one of your modified 11-24's in my future - especially if it can replace the 17TS(and be more versatile). I have to try it out on some shoots, but appears that the 16-35 is a decent replacement for the 24TS. Being able to use it at 22-30mm regularly(and on either extreme in a pinch) is game changing. If the 11-24 has a similarly large sweet spot(say 14-20) I'll be a convert to zooms on the wide end...

I would definitely recommend to use f11 or even f11,5 when shifting.
(even if diffraction may cause a bit less contrast and a little bit less sharpness, this is easy to recover by software).
And yes - the plastic cover of the 4/16-35mm has to come out for full shift performance.

Regards
Stefan

 

[ErikKaffehr]

Hi,

I did shoot with baffle in place, the last shot I made at 16 mm and for that one I removed the baffle. I would say that it is sitting there for a reason. It is easy to remove it and not difficult to put it back.

This page has a couple of images shot with that lens: http://echophoto.dnsalias.net/ekr/Articles/Canon_16_35_4/ . I will add more in due time.

My take on shifting is right now that I have a 24/3.5 TSE LII already, so I don't really need 24. On the other hand, 35mm is quite interesting. At 16 mm there is very little shift.

I have a bunch of Hasselblad V lenses, 40, 60, 100, 120 and 180 mm and I can use all those with the HCam

Best regards
Erik

Hi Erik,

Did you remove the plastic cover from the rear of the lens? Just wondering because HCam reports closer to 15mm of movement in the 24mm range.

I just received my 16-35 and tested it with a metabones adapter(while I wait for my actus from CI and canon board from SK Grimes). I'll have everything later this coming week, and interested to compare it to the 24 TSII. Straight on, it looks to equal a decent copy of the TS, but maybe not a great copy in terms of sharpness. There is definitely some strong CA's in the corners in the 16mm range.

I'm not sure it can replace the 17 and 24 in terms of absolute quality, but when you think about the flexibility the zoom offers along with the $3k in savings, it starts sounding better and better!

 

[Stefan Steib]

Thanks Stefan... I appreciate all your help, and you being a trailblazer with the use of movements on the sony! I tested it at f11 and f16 and both were plenty sharp. F16 had some diffraction, but still looked great.

I see one of your modified 11-24's in my future - especially if it can replace the 17TS(and be more versatile). I have to try it out on some shoots, but appears that the 16-35 is a decent replacement for the 24TS. Being able to use it at 22-30mm regularly(and on either extreme in a pinch) is game changing. If the 11-24 has a similarly large sweet spot(say 14-20) I'll be a convert to zooms on the wide end...

Hi Erik

The sweetspot of the EF-L 11-24mm is 15-24mm whereas the 17-24mm can fully substitute both 17 and 24mm TS-E. (on the Sony E-Mount !)
The TS-E´s have 2 advantages: an even bigger image circle for usage maybe on an HCam B1 or Alpa FPS - and second the 17 TSE is "STRAIGHT!",
there is no straigther superwideangle on the planet, but - the 11-24mm is not much worse, the little bit of distortion can easily be corrected with a lens profile.

About the baffle: the HCam Master TS full takes the functionality of the baffle when mounted on it, so it is not needed any more !

Regards
Stefan

 

[ErikKaffehr]

Hi,

One interesting observation may be that the 11-24mm can replace both the 17/4TSE and the 24/3.5 TSE, in addition it can really replace the 16-35/4, too.

So buying the 11-24 is quite a bit cheaper than buying all those other lenses.

On the baffle thing, I use the 16-35/4 as a general purpose lens, so it would normally be used on the Metabones and not the HCam.

I have shot some images at 20mm recently, and having shift was indeed helpful and I felt there was a significant amount of possible shift, more than what I needed.

Best regards
ERik

Hi Erik

The sweetspot of the EF-L 11-24mm is 15-24mm whereas the 17-24mm can fully substitute both 17 and 24mm TS-E. (on the Sony E-Mount !)
The TS-E´s have 2 advantages: an even bigger image circle for usage maybe on an HCam B1 or Alpa FPS - and second the 17 TSE is "STRAIGHT!",
there is no straigther superwideangle on the planet, but - the 11-24mm is not much worse, the little bit of distortion can easily be corrected with a lens profile.

About the baffle: the HCam Master TS full takes the functionality of the baffle when mounted on it, so it is not needed any more !

Regards
Stefan

 

[cdavis32]

Except that the 24-30 range is what's so appealing! I often find myself in situations where 24 is too wide and 35 is too tight.

Anyone try PT Lens? - http://epaperpress.com/ptlens/shift.html - Seems like with a custom correction it could go a long way toward improving the results with the zooms. The 16-35 is sharp, but has a lot of distortion at the wide end.

Hi,

One interesting observation may be that the 11-24mm can replace both the 17/4TSE and the 24/3.5 TSE, in addition it can really replace the 16-35/4, too.

So buying the 11-24 is quite a bit cheaper than buying all those other lenses.

On the baffle thing, I use the 16-35/4 as a general purpose lens, so it would normally be used on the Metabones and not the HCam.

I have shot some images at 20mm recently, and having shift was indeed helpful and I felt there was a significant amount of possible shift, more than what I needed.

Best regards
ERik

 

[daf]

Except that the 24-30 range is what's so appealing! I often find myself in situations where 24 is too wide and 35 is too tight.

Anyone try PT Lens? - PTLens | Shift Lenses - Seems like with a custom correction it could go a long way toward improving the results with the zooms. The 16-35 is sharp, but has a lot of distortion at the wide end.

Capture One has a pretty good correction profile for the 16-35f4canon ... and the very very best things is that you can correct the distorsion/CA with shift movement X/Y !