Mike, thanks for posting the photo from the 12. Cool shot! I think I recall seeing that before.
That shot does look pretty good, then again, it's hard to tell with the nicely saturated sky. I find that a lot of images can mask the effect, and in really neutral conditions it suddenly rears it's head. I'm talking only about lenses that I've mentioned already above, but it's encouraging to hear that the 12 is looking good on the 5N. Like Uhoh says, I will soon be posting some samples.
The Leica Elmarit-M 24/2.8 ASPH was the best wide I used on the 5. Very low color shift issues, and nearly as perfect wide open as it was stopped down to f/8. I sold it before I got back into the NEX with the 5N, but I can't imagine the performance would degrade any.
For SLR lenses, the OM 24/2 is hard to beat in the price/performance/size evaluation.
Still leaving the 5D2 at home with its all fast cheap and super-performing lenses - so I have against all odds started to think about really slow 18 or 21 lenses. The ZM 18/4 is praised very much, but what's the word on the ZM 21/4.5? That's super slow - on par with a long white excellent Canon tele. But what the heck, it's small
Or would it be betterto get the large ZE 21/3.5? - which I could use it on the 5D2 when eh... I want some slow optics on that.
EDIT: and then there is the Voigtländer Color Skopar 21/4, but I guess that's inferiour to the above choices...(?)Thanks for all the congratulations...yes I must admit, one of my better lens designs
But seriously, this issue of the great 21 is interesting.
The skopar is a hellva lens with centers like you cannot believe, tiny, weightless and cheap. But its downfall is the edges, as with so many afforable options. You don't even notice it much till you are working in large landscapes--and it may be possible to mitigate.
The ZM 21/4.5: great question. Who has one? My main source of gossip havs been down for a week--RFF.
My thinking right now is: daylight ZM 18 + leica 28/2; night time ZM 21/2.8 and CV 35/1.2, But the truth is I have not seen more than a handfull of ZM 21 samples on 5n-- from the fast or the slow.
here from nick at LL
"In my informal tests against the CV21, the lenses had matching center-resolution at all working apertures, though the ZM21 had perceptibly better resolution toward the edges and corners at f/4 and f/5.6. That said, the difference was small and only noticeable in a relatively close-up subject with very fine detail. By f/8 the two lenses were effectively identical in resolution. For daily use, I would take the CV21 over this lens on account of the size difference, notwithstanding the Biogon's marginal quality advantage."
While Rock Huffwell states:
"The Voigtländer 21mm f/4 or even the best 21mm SLR lens or zoom is nowhere near the performance of this Zeiss or the LEICA 21mm lenses.' (refering to the 21/2.8)
The ZM is 258 grams or so.
Apparently the rear node of the 4.5 is very deep....Show more →
So is the CV15 and like the ZM 21 f/4.5 is a problem lens on the M9, yet on the 5N the CV15 is great. I too wonder about the 21mm f/4.5, but unfortunately don't have access to one...
Jack Flesher on the CV 21:
“ Well, as it just so happens... I had an opportunity to try out one of these little gems. Simply stated, in terms of resolution it performed at least as well as the 21P if not a bit better, and as such is not really very far behind the 21A. Its big faults were more significant falloff in the corners (maybe 2 1/2 stops?), more visible distortion in the corners, and the fact it is more flare-prone than either Leica 21.
I think I need to do some more shooting with the 21 CV!...Show more →
My experience with the ZM21 f/2.8 on the NEX-5N indicates it's not a great performer into the corners. You need to stop it down to f/8 to get sharp corners. I believe this is an issue of the lens design rather than it being a wide RF lens on a NEX because if you look at the Zeiss MTF curves for it, you'll see that it falls off into the corners at wider apertures, and this carries over to the APS-C crop of the 5N. The CV15 is a much, much more problematic lens on the M9, but performs great on the 5N, so I don't suspect it's an issue with the 5N causing the poor ZM21 results.
I posted a blog write-up with full-rez downloads, including my results with the CV12 and CV15.
I think my CV12 is a bit out of alignment, and looks kind of bad on the NEX, though on the M9 it's not as noticeable...
dpap1978 wrote:
Ok, I have both the 5N and the Zeiss 21mm ZM 4.5. Today I did some rough tests. The lens exhibits some color vignetting, but less so that the M8. I examined the corners, and there is no evidence of corner smearing at f4.5. When I get some time, I will post some corner crops.
fantastic--I think we'd all love to see a number of samples and links to full size from that lens when you have a chance. It must be alot smaller than the 2.8.
@freaklikeme
I have an older 28 elmarit from the 80's, which also never really shifted on the n5---made pretty good edges too, but the little CV 28/3.5 blasted it in the centers. It was not quite fast enough for easy use indoors on the n5, but is prolly OK with n5n
dpap1978 wrote:
Ok, I have both the 5N and the Zeiss 21mm ZM 4.5. Today I did some rough tests. The lens exhibits some color vignetting, but less so that the M8. I examined the corners, and there is no evidence of corner smearing at f4.5. When I get some time, I will post some corner crops.
Then it would seem to be a better alternative to the ZM 21 f/2.8 if you can live with f/4.5. See my post just above for the link to samples of the ZM 21 f/2.8 on the 5N.
uhoh7 wrote:
I have an older 28 elmarit from the 80's, which also never really shifted on the n5---made pretty good edges too, but the little CV 28/3.5 blasted it in the centers. It was not quite fast enough for easy use indoors on the n5, but is prolly OK with n5n
the asph is so small and seems REALLY sharp-- some say it's identical to the cron from 2.8 on.
My understanding is that the 28 Elmarit-M ASPH is a considerable improvement over the previous f/2.8 versions. Also, the new 24 Elmar is supposed to be super-duper fantastic wide open, though it's an f/3.8 lens...
Looking forward to seeing what you can do with the 28 Cron. If you don't like it, let me know. I could use it on the M9
rscheffler wrote: dpap1978 wroteOk, I have both the 5N and the Zeiss 21mm ZM 4.5. Today I did some rough tests. The lens exhibits some color vignetting, but less so that the M8. I examined the corners, and there is no evidence of corner smearing at f4.5. When I get some time, I will post some corner crops.Then it would seem to be a better alternative to the ZM 21 f/2.8 if you can live with f/4.5. See my post just above for the link to samples of the ZM 21 f/2.8 on the 5N.
Ron
WOW fantastic, Ron. What a shame--I'd been dreaming the zm 21/2.8 would run right out to the edges, but it looks just like the CV 21--till way down there anyway.
Not a landscape animal, but maybe great for mood and lowlight. Frankly--I wonder if it's better than the contax version.
I'm coming around on that CV 15, which you make a strong case about. I turned one down at 300USD a few weeks ago LTM.
But philber has me 100% sold on the zm18, so better put my money there.
uhoh7 wrote:
the asph is so small and seems REALLY sharp-- some say it's identical to the cron from 2.8 on.
I had the 28 Elmarit ASPH for a while - in terms of resolving power the Elmarit ASPH holds its own against the 28 Cron, but they have their own style. The Elmarit is much contrastier. The Cron has a very distinctive style, I can't really verbalize it. You'll know it when you see it.
The one thing the Elmarit has on the Cron is handling - so small!
Ron - fantastic tests and write up - really appreciate your hard work on this.
I have the M mount 15mm on the 5N and love it - i've been impressed.
However, your sample looks even better! It looks pretty much perfect wide open. Mine shows a slight softening on the extreme left edge wide open but is gone by F5.6 - it seemed like field curvature becasue if I focussed on the extreme left edge it was fine - could be my adaptor thickness, but its not noticeable in real world use.
But... if you ever sell your one - i'll buy it anyway haha...
Thanks for the tests, Ron. I noticed that you mentioned that you're focusing wide open and then stopping down to shoot, and I'm wondering if that could be causing some of the issues you mentioned in the test?
Jacob D wrote: uhoh7 wroteor just buy a grey card on ebay for $6--but you may not have to bother, as noted above.
The real problem child was the 15, which is much better now, so maybe the 12 is as decent as the 21.
You will show us soon!
congrats on great lens!!Thanks!
I think I will give Douglas' suggestion a try. The problem with a gray card is, it's not large enough to cover the entire frame at working distance. I suppose it could be placed close to the lens, as focus won't matter anyway (as in the tissue example).
I actually found the 21/4 to be "bad" compared to the 35/2.5, 40/1.4 (no colorshift to speak of), and Oly 50/1.8 which seems to have a little. If the 12 is in the ball park of the 21 I'll definitely want to Cornerfix it.
I suppose this is a little like the white balance trick with a coffee filter over the lens.
Douglas, have you made cornerfix profiles this way? I tried a couple on a light colored neutral wall, defocused it so there was no texture, and made my profile images. However when I use them it seems to brighten the photo quite a bit. I assumed this was due to my use of a light colored wall. I don't have any prior experience with Cornerfix though, so I don't know if my observation was correct or not. I thought I read somewhere that 18% gray is best to use for the profile. I know this is a little OT but in the context of wides for the NEX it's sort of relevant I guess....Show more →
I haven't had to use Cornerfix profiles too much, but it seems to have to do with the aperture and focus distance that you use. If you make a profile wide open, your correction is likely to be more extreme than at f5.6. So, I think you should shoot a few at different aperture and focus settings and see which is the best compromise for you. I believe that I settled on focus at infinity and f5.6, if I remember correctly.
FWIW, the problem with shooting a grey card or wall is that the light usually ends up being at least a little directional. I would use the tissue method that I mentioned, because it diffuses the light pretty evenly (maybe use a lens hood so that the tissue doesn't touch the lens.)
uhoh7 wrote: rscheffler wrotedpap1978 wroteOk, I have both the 5N and the Zeiss 21mm ZM 4.5. Today I did some rough tests. The lens exhibits some color vignetting, but less so that the M8. I examined the corners, and there is no evidence of corner smearing at f4.5. When I get some time, I will post some corner crops.Then it would seem to be a better alternative to the ZM 21 f/2.8 if you can live with f/4.5. See my post just above for the link to samples of the ZM 21 f/2.8 on the 5N.
RonWOW fantastic, Ron. What a shame--I'd been dreaming the zm 21/2.8 would run right out to the edges, but it looks just like the CV 21--till way down there anyway.
Not a landscape animal, but maybe great for mood and lowlight. Frankly--I wonder if it's better than the contax version.
I'm coming around on that CV 15, which you make a strong case about. I turned one down at 300USD a few weeks ago LTM.
But philber has me 100% sold on the zm18, so better put my money there.
So...... who has a 21 summilux?
Your posts here expressing interest in the ZM 21 f/2.8 are what drove me to finally get around to posting these test images. I didn't want you to spend $1300 and be disappointed.
I think if you're shooting landscapes or anything where you want good edge resolution and you can't stop down to f/8, then look for something other than the ZM21 f/2.8. The f/4.5 version sounds promising, but I think the ZM18 is the best bet. Of the three, it has the best MTF values for an APS-C sensor. It does fall off into the corners, but that's on FF. For low light work where the subject will be fairly centralized, or where it's just one person and you've focused on them off-center, I think the ZM 2.8 version will be OK. But the lens I suspect you really want for corner to corner sharpness is the Leica Super-Elmar 21mm f/3.4. At 'only' $3K new, it's a relative bargain for a Leica. It's supposed to be as good if not better than the ZE/ZF 21mm and is likely the next RF lens I hope to acquire. (BTW, the ZM 21 f/2.8 is not in the same league, on paper, as the ZE/ZF 21, though in real life use on a FF camera is very good). Actually, I'd love to have the SE21 and the 21 Lux because 21mm on FF is one of my favourite focal lengths... but I doubt I'll have a Lux in the near future.
Sosua wrote:
Ron - fantastic tests and write up - really appreciate your hard work on this.
I have the M mount 15mm on the 5N and love it - i've been impressed.
However, your sample looks even better! It looks pretty much perfect wide open. Mine shows a slight softening on the extreme left edge wide open but is gone by F5.6 - it seemed like field curvature becasue if I focussed on the extreme left edge it was fine - could be my adaptor thickness, but its not noticeable in real world use.
But... if you ever sell your one - i'll buy it anyway haha......Show more →
Thanks Sam for your feedback here and on my blog post! I think I got pretty lucky with the CV15 and somewhat unlucky with the CV12. My hunch is there's a Voigtlander lottery and is also alluded to by Jeff Hapeman in his review of these lenses, IIRC. And I think there are a number of variables at play, such as copy to copy quality, adapters, the camera's lens mount... On the M9 my CV12 is pretty decent, just a bit softish along the left 'B' zone wide open but very decent by f/8. On the NEX my copy sucks. But I really should try another adapter to rule out that variable. BTW, my CV15 is the M version and the CV12 is the LTM (because I wanted a smaller, compact lens without that huge probably useless permanent hood of the M version). Another variable could be the LTM-M adapter... But it shouldn't really make a difference in terms of optical quality because both versions of both lenses are optically identical.
Doug: I'm pretty confident that the ZM results are OK and not thrown off by my focus method because it's consistent with what I see from the ZM21 on the M9. But on the M9 the central APS-C crop area is not as high rez as the 5N, and therefore doesn't amplify the effect as dramatically. For the building scene I focused in the center of the frame, wide open. For the forest scene, on the near tree trunk at the left 1/3 of the scene. My observations of the ZM images are that the sharpness falloff is consistent into all corners and is representative of the lens's MTF curves. It has a sharp central zone wide open and I suspect there is some field curvature because on the M9 I can focus on something 2-3m away and along the edges of the frame the background at maybe 10-15m will be in focus. It means that any scene with near foreground objects in the corners will look nasty, such as the forest scene. But I think if everything is at a similar farther distance, the ZM 21 f/2.8 will probably be reasonable at smaller than f/5.6. I did test the ZM 21 f4.5 on the M9 when I got the ZM 21 f/2.8 and the f/4.5 was better towards the edges, so this would correlate with dpap1978's findings, which I look forward to seeing... but stated earlier, I would agree that the ZM18 is probably the best of the three. Unfortunately it's not super compact for a RF lens, but still pretty small.
rscheffler wrote:
Your posts here expressing interest in the ZM 21 f/2.8 are what drove me to finally get around to posting these test images. I didn't want you to spend $1300 and be disappointed.
You don't know how close i came---missed one at DPI by about 10 mins---Thank you so much, Ron.
The 28 summicron seems like it should be OK http://www.flickr.com/search/?q=28mm%20summicron%20nex
most of those taken with n5 and seem to show the lenses normal vignetting. Really does seem to have a beautiful look with sharpness and milder contrast..
Might be fantastic on the n7
Sigma or tokina should take note and make us a nice fast flat 20
Found a ZM 18 tonite---so I'm done!!
Thanks for all the great input so far.
I'm still mystified by the 21/2.8 zeiss edges---I shouldn't be since I see similar issues with the CVs. How is that lens on the m9?
Not sure if people had seen this: interesting writeup of experiences with a number of wides on the 5N, including the Summicron 28, the ZM 25/2.8, and the Elmarit-R 19, which sounds like an intriguing seldom-discussed option.
alwang wrote:
Not sure if people had seen this: interesting writeup of experiences with a number of wides on the 5N, including the Summicron 28, the ZM 25/2.8, and the Elmarit-R 19, which sounds like an intriguing seldom-discussed option.
some people should be required to get a license for adjectives Literary waterboarding. With high fructose corn syrup, hehe. Perhaps it reads better in francais......
Not that I could do any better.
plenty of samples though.
good find there.
Does the ZM 25 live up to its potential on ANY digital camera?
the leica 28/2 looks promising knock on wood for a good copy
If you don't want to correct red edges I would say stay away from the ZM 21 f/4.5 on the NEX 5N. At infinity it is definitely there. If you move in a little closer the red edge goes away. It does smear slightly in the corners and heavy vignetting is there. Lack of distortion though is wonderful
uhoh7 wrote:
Does the ZM 25 live up to its potential on ANY digital camera?
At infinity / for general landscapes I'm so far underwhelmed by the ZM 25 on the NEX-5N, which is in part why I thought I'd have a good look at the GXR.
It is great on a Hawk adapter for close up work... you can get very close indeed, maybe about 6cm from the front of the lens. But that's not the primary reason I bought it...
The ZM 35/2 on the other hand delivers nicely on the 5N. Haven't received my 18 yet.
rsolti13 wrote:
If you don't want to correct red edges I would say stay away from the ZM 21 f/4.5 on the NEX 5N. At infinity it is definitely there. If you move in a little closer the red edge goes away. It does smear slightly in the corners and heavy vignetting is there. Lack of distortion though is wonderful
TY for these samples. Would you perhaps have a link to full size for #1? Or some sort of edge/corner rich sample?
What do we think of the edge corner performance from the ZM 21/4.5 on the 5n?
It's funny, RFF resurfaced for some time last nite and I was reading up on the 21/4.5---which is widely considered the best of all the landscape 21s over there---well besides the super elmar maybe.
I was all ready to bring it up here, naively anticipating perfection, as usual
I'm beggining to think the CV 21 may be as good as any on the 5n---pending edge results on the ZM 4.5 and leica 3.4
I was also chasing after reports of the leica 28/2 at RFF--which are surprisingly few and far between. The 35 lux is a weed by comparision.
My favorite review is Stephen Gandy's
"28/2 Summicron ASPH to be introduced at Photokina 2000. Presumably the best and sharpest of the Leica 28's. "
uhoh7 wrote:
(...)
I was also chasing after reports of the leica 28/2 at RFF--which are surprisingly few and far between. The 35 lux is a weed by comparision.
A very nice weed though...
I'm looking forward to some samples from your 28Cron. I'm still not sure about the exact favourite focal length for my walk around lens. The FA31/1.8 is about right, but a big lens and a little too much of not perfect.
uhoh7 wrote:
The lens is more mythological than legendary
28mm snow leopard...
In case you don't know these yet, there's a pretty extensive collection of pictures with that lens by denoir/Luka to be found here (on the M9 though): http://peltarion.eu/img/m9/cron28.html