nehemiahphoto wrote:
Yeah, it would have been a very attractive option if the 21/1.4 was very good coma wise (like the 24 GM/20 G) WO. Thanks for checking for me
The 24/1.4 GM has outstanding coma performance. I don't remember testing the 20/1.8 G for that.
The Sony 20 f/1.8 G seems to have quite low coma. Here is a review from dpreview in which they provide a 100% crop of the night sky. It seems to have really nice coma performance which is consistent with my experience of the lens:
Steve Spencer wrote:
The Sony 20 f/1.8 G seems to have quite low coma. Here is a review from dpreview in which they provide a 100% crop of the night sky. It seems to have really nice coma performance which is consistent with my experience of the lens:
I agree Steve. I have formally tested the 20G vs 24GM for coma--both are superb. A bit different. The GM has better across the frame coma. When pixel peeping, the 20g actually has less in the center, but as you extended to the midframe, and especially the corners, you do get a tiny batwings. The GM 24 has more LoCA, throughout the entire frame, but coma performance remains the same across the entire frame. GM 24 correction is phenomenal, the 20G is not far behind. I have shot both in the field multiple times, and both are among the best, offer stellar results. Even on 42MP sensors, both lenses hold up quite well when looking at crops:
When the CV 24/1.4 was announced, I was hoping that it would be usable, coma-wise, at f1.4. For my taste it didn't measure up still stopped down, but I am curious to see if the VM 21/1.4 is better.
Yes Fred, I have been eying the CV 21/3.5 for sometime and I tabbed it in my mind when you reported the IQ was not too sensitive to sensor stack as it surprised me. But I have my beloved G21 which is nearly identical size. Having said that, I have a CV 35/2 Ultron, 75/1.5 and might get a 21/3.5 in m-mount as I have an m film camera, NKIR modded sensor coming back and a7c. The 35 and 75 no focus shift, or so little I can shoot them WO on film with no issue, same on my m240 when I had it. The 21/3.5 would be a minimal fuss/max compatibility across those bodies, and it's small and priced well and a high usage FL for me. The only stinker is the 35/2 Ultron is wonky on a Sony sensor stack, but I survive--the lens in wonky anyway lol.
nehemiahphoto wrote:
I agree Steve. I have formally tested the 20G vs 24GM for coma--both are superb. A bit different. The GM has better across the frame coma. When pixel peeping, the 20g actually has less in the center, but as you extended to the midframe, and especially the corners, you do get a tiny batwings. The GM 24 has more LoCA, throughout the entire frame, but coma performance remains the same across the entire frame. GM 24 correction is phenomenal, the 20G is not far behind. I have shot both in the field multiple times, and both are among the best, offer stellar results. Even on 42MP sensors, both lenses hold up quite well when looking at crops:
When the CV 24/1.4 was announced, I was hoping that it would be usable, coma-wise, at f1.4. For my taste it didn't measure up still stopped down, but I am curious to see if the VM 21/1.4 is better.
Yes Fred, I have been eying the CV 21/3.5 for sometime, but I have my beloved G21 which is nearly identical size. Having said that, I have a CV 35/2 Ultron, 75/1.5 and might get a 21/3.4 in m-mount as I have an m film camera, NKIR modded sensor coming back and a7c. The 35 and 75 no focus shift, or so little I can shoot them WO on film with no issue, same on my m240 when I had it. The 21/3.5 would be a minimal fuss/max compatibility across those bodies, and it's small and priced well and a high usage FL for me. ...Show more →
For the modded Sony, the Leica 21/3.4 SEM offers great performance. Mid-zone resolution will be higher than what the CV 21/3.5 provides. As for the corners, the CV 21/1.4 Nokton has slightly bettter performance starting at f/3.5. The latter is quite amazing stopped down.
Fred Miranda wrote:
For the modded Sony, the Leica 21/3.4 SEM offers great performance. Mid-zone resolution will be higher than what the CV 21/3.5 provides. As for the corners, the CV 21/1.4 Nokton has slightly better performance starting at f/3.5. The latter is quite amazing stopped down.
Interesting--never shot the 21/3.4 SEM. Good to know. Aside from technical IQ, I notice the CV 21/1.4 is a lens where I see a lot of 3D or spacial seperation, much much more and fairly consistently from different users across different light situations. I have really enjoyed some of the photos from that lens in particular. Curious to see the new CV 50/1. Only so many pennies though!!
Did you find the midzone dip to be pretty bad/aggressive/a real-word issue on the CV 21/3.5?
nehemiahphoto wrote:
Interesting--never shot the 21/3.4 SEM. Good to know. Aside from technical IQ, I notice the CV 21/1.4 is a lens where I see a lot of 3D or spacial seperation, much much more and fairly consistently from different users across different light situations. I have really enjoyed some of the photos from that lens in particular. Curious to see the new CV 50/1. Only so many pennies though!!
Did you find the midzone dip to be pretty bad/aggressive/a real-word issue on the CV 21/3.5?
It shows up when I test and look for it. However, I think it's harder to spot in real world images, especially when stopping down and resizing. There is mid-zone weakness on the CV 21/1.4 as well but it recovers at smaller apertures. That's the uniqueness of the 21SEM where mid-zone is great from wide open. It's quite compact and the haptics is great.
I should get the Nokton 50/1 for review and will compare it to the f/1.2 version.
nehemiahphoto wrote:
Interesting--never shot the 21/3.4 SEM. Good to know. Aside from technical IQ, I notice the CV 21/1.4 is a lens where I see a lot of 3D or spacial seperation, much much more and fairly consistently from different users across different light situations. I have really enjoyed some of the photos from that lens in particular. Curious to see the new CV 50/1. Only so many pennies though!!
Did you find the midzone dip to be pretty bad/aggressive/a real-word issue on the CV 21/3.5?
Nehemiah, I have had the CV 21 f/1.4 and Loxia 21 f/2.8 on Sony and really really liked both lenses, but my favorite 21 by a pretty big margin is the 21 f/3.4 SEM. I love both the IQ and the haptics, and the complete absence of midzone dip is a plus (although the Loxia has basically no midzone dip as well). I also love the small filter size and the lovely little hood. My only real issue with the Loxia is that I think it is bigger than it needs to be for an f/2.8 lens. That said we are definitely spoiled for good choices at 21mm.
Steve Spencer wrote:
Nehemiah, I have had the CV 21 f/1.4 and Loxia 21 f/2.8 on Sony and really really liked both lenses, but my favorite 21 by a pretty big margin is the 21 f/3.4 SEM. I love both the IQ and the haptics, and the complete absence of midzone dip is a plus (although the Loxia has basically no midzone dip as well). I also love the small filter size and the lovely little hood. My only real issue with the Loxia is that I think it is bigger than it needs to be for an f/2.8 lens. That said we are definitely spoiled for good choices at 21mm....Show more →
My only complaint about the Loxia 21 is the rather prominent inwards field curvature and fortunately that's not the case with the SEM.
Steve and Fred (or Rich or anyone) have you used the ZM 21/2.8? And thanks for the insight on the 21 SEM—interesting and telling you both like it so much. The ZM seems to be reviewed well and praised. Wonder how it stack up against the CV 21/3.5 and 21 SEM.
nehemiahphoto wrote:
Steve and Fred (or Rich or anyone) have you used the ZM 21/2.8? And thanks for the insight on the 21 SEM—interesting and telling you both like it so much. The ZM seems to be reviewed well and praised. Wonder how it stack up against the CV 21/3.5 and 21 SEM.
That's a lens I've never tested. However, according to their MTF graphs, which may not give a very accurate comparison, the Leica performs better everywhere. Keep in mind the Leica's lower line is at 40 lp/mm while the Zeiss is at 30 lp/mm which is a big plus for the Leica. Imagine the ZM's performance being even worse at 40 lp/mm.
Their curves have a similar fall-off but the Leica provides higher contrast and it can discern higher frequency detail.
Fred Miranda wrote:
That's a lens I've never tested. However, according to their MTF graphs, which may not give a very accurate comparison, the Leica performs better everywhere. Keep in mind the Leica's lower line is at 40 lp/mm while the Zeiss is at 30 lp/mm which is a big plus for the Leica. Imagine the ZM's performance being even worse at 40 lp/mm.
Their curves have a similar fall-off but the Leica provides higher contrast and it can discern higher frequency detail.
Yeah, the SEM looks appreciably better from those MTF’s. A quick Google and user reports indicate most people feel the same and the SEM is a superlative lens. I don’t see a lot of testing though, but it makes sense given the age of the ZM as well.
Fred Miranda wrote:
Yes, at f/8 DOF masks the induced FC, so that's expected. Pretty much all M lenses perform ok on the stock Sony sensor at f/8.
However, at wide apertures, the CV 21/1.4 Nokton does not perform equally as wrote previously. That's misinformation.
I never meant that CV 21/1.4 performs equally good on stock and modified sensor stack Sony or M-cameras, apologies if it created confusion.
What I wanted to say is that in my experiments it outperformed CV 21/3.5 at all apertures (including at f3.5 in the corners) and that *to me* it is acceptably sharp in the corners starting with f2.4.
It evidently performs *much better* in the corners on a modified Sony sensor (thanks for the additional tests!) and on M-mount cameras, but that seems to be generally the case for all fast and wide VM lenses I’ve seen tested for that. This one is just *good enough* to me even when crippled by the thick sensor stack.
As to the coma, the VM lens on stock A7RII shows pretty much the same performance showed here for the E-mount version: at f2.4 it’s usable and at f2.8 it’s pretty good. Wider than f2.4 it’s rather bad. I.e. not even close to the G20 and GM24.
Dec 17, 2021 at 01:35 PM
Steve Spencer Offline • • • • • • • Upload & Sell: On
Fred Miranda wrote:
That's a lens I've never tested. However, according to their MTF graphs, which may not give a very accurate comparison, the Leica performs better everywhere. Keep in mind the Leica's lower line is at 40 lp/mm while the Zeiss is at 30 lp/mm which is a big plus for the Leica. Imagine the ZM's performance being even worse at 40 lp/mm.
Their curves have a similar fall-off but the Leica provides higher contrast and it can discern higher frequency detail.
The lower line for the Zeiss charts is 40 lp/mm. Here is the caption from the Zeiss page: Spatial frequencies R = 10, 20 and 40 cycles/mm.
That said the conclusion is the same the MTF charts are way better for the Leica 21 SEM than the Zeiss ZM 21 f/2.8 and I think the difference is even more striking at f/5.6 where the MTF graphs for the Zeiss ZM show a lot of astigmatism (or more exactly separation of the sagittal and tangential performance).
Steve Spencer wrote:
The lower line for the Zeiss charts is 40 lp/mm. Here is the caption from the Zeiss page: Spatial frequencies R = 10, 20 and 40 cycles/mm.
That said the conclusion is the same the MTF charts are way better for the Leica 21 SEM than the Zeiss ZM 21 f/2.8 and I think the difference is even more striking at f/5.6 where the MTF graphs for the Zeiss ZM show a lot of astigmatism (or more exactly separation of the sagittal and tangential performance).
Wow, I should've known better since the Loxia lenses are also tested at 40 cycles/mm.
simonedf wrote:
Has anyone used the Voigtlander Ultron 21mm f/1.8? How does it compare to the Nokton or the f/3.5?
I can't compare it, but until I got my Sigma 14-24 Art, it was the best wide that I had. It was one of the few lenses that was designed before Sony's FE cameras that had basically the same performance on both M and Sony cameras.
The Voigtlander 21/1.4 Nokton performs just as good on the M and SL. I have done controlled tests on this and actually corners perform a little better on the latter to my surprise. Not the case with the Leica 21/3.4 SEM.
(hint: not good!) Unfortunately (for me) the Lux shows its age. I agree the SEM is a jewel of a lens.
