Fred Miranda wrote:
Here are some crops (warning: Big images!)
They show both lenses at infinity distance focused at the same location (mid-field). Best of 3 for both at 12.5x magnification.
LR default settings except for the same sharpening used for both lenses.
For these crops they were both at f/5.6 (which is the best aperture for the ZM 35/1.4)
Notice the slight better contrast for the FE 35/2.8 thumbnail image.
For all crops: ZM 35/1.4 PCX (LEFT) | FE 35/2.8 ZA (RIGHT)
Those 35 FE crops look extremely strong, especially given the tiny size and weight of the lens. The ZM wins, but given the price and size/weight difference if I could have a copy of the FE35 that looked like that I'd take it every time (if you want to sell it let me know )
I'd always written the lens off as being a bit overrated - I had never seen crops that presented it as a compelling lens optically. Seems perhaps that is down to sample variation. Is it just me or is sample variation on E mount truly atrocious across all lens manufacturers? Maybe I just have bad luck. Had to try three Loxia 21's for a keeper and five VC15s. My old FE 16-35 was unimpressive and likely a bad copy but I never had another one to compare it to.
Fred Miranda wrote:
I was testing the ZM 35/1.4 (with 5m PCX) against the FE 35/2.8 today just for fun and was impressed how good the little FE lens is. (Well, I finally got a good copy of it...)
The ZM 35/1.4 PCX peaks at f/5.6. That's the aperture that finally brings the mid-field to the max resolution and yields sharpness across the entire field. All zones are very impressive at this aperture (center, mid and edges). I have not seen a 35mm performing better than this on any system.
The FE 35/2.8 ZA on the other hand does very well wide-open. At f/2.8, its mid-field is actually noticeably better than the ZM while center is very similar. The ZM has a slight advantage for the edges. The image has a lot of contrast and it's actually on pair with the ZM. I was surprised to see this. This tiny lens is definitely impressive at f/2.8. Colors also look very, very similar...Vignetting is stronger for the FE lens.
The FE 35/2.8 peaks at f/7.1 but is just as good as f/5.6, except for the corners and mid with a very slightly improvement at f/7.1. (f/6.3 looks identical to f/5.6)
So, how do they compare? I would say that center and mid-field are very similar starting at f/5.6. The mid-field looks better on the FE at f/2.8 and f/4 though.
Once we start approaching the edges, the ZM jumps ahead and it's absolutely superb. The FE 35/2.8 maintains sharpness while showing some astigmatism. From my tests, the FE 35/2.8 performs very similarly to the RX1RII at infinity distance but I was only able to see that with my latest copy.
Sunstar rendering looks strong after f/16 and is has 14 points. Best at f/22. The ZM 35/1.4's sunstar has 10 points and looks impressive starting at f/4.
Would this difference show in a print? Well, at 1:2 mag, both lenses at their optimum aperture look very similar at center and mid-field. Towards the edges, I can see better sharpness (resolution) on the ZM image but the difference is not that big and perhaps would go unnoticeable depending on the image. Of course, this different is much bigger at pixels level (1:1)
Contrast, believe it or not is slightly better for the FE!
I can tell you from experience, there is copy variation for this FE. I tested many copies and this one is finally well centered and performs great optically. (I bought it as an open box from Best Buy for $520)
I can post some crops later but after studying images from both lenses, that's my consensus....Show more →
I have an idea that the little FE 35 is a bit fragile too.
I had a copy which was absolutely great at f2.8; just didn't really improve much so that by the time you were stopped down other lenses, like the lox, were a bit better.
Used it for hiking satisfactorily, except for contrast loss against the light (and indifferent sunstars at normal apertures)
But: at some point it acquired significant tilt and the left side became much worse overall. I never dropped it as far as I know. Presumably there must have been some mild impact, but it must have been mild, because I am very alert to things like that.
It's the only lens I've ever had that has done that; so I gave it away.
I wouldn't mind having it again as a hiking lens, but the search for a good copy, and the fear that it won't stay good, put me off...
Aztatlan wrote:
Those 35 FE crops look extremely strong, especially given the tiny size and weight of the lens. The ZM wins, but given the price and size/weight difference if I could have a copy of the FE35 that looked like that I'd take it every time (if you want to sell it let me know )
I'd always written the lens off as being a bit overrated - I had never seen crops that presented it as a compelling lens optically. Seems perhaps that is down to sample variation. Is it just me or is sample variation on E mount truly atrocious across all lens manufacturers? Maybe I just have bad luck. Had to try three Loxia 21's for a keeper and five VC15s. My old FE 16-35 was unimpressive and likely a bad copy but I never had another one to compare it to....Show more →
Testing four FE 35/2.8 lenses, I can tell you there is variation and it's noticeable. I also have a great copy of the 16-35/4 and I will test them against each other (I will post the results in another thread). I know for sure it will beat the zoom up to f/5.6 @35mm but the zoom does improve further at f/8 and f/9. I'm curious to compare them.
As you wrote, the ZM wins especially towards the corners and that's not surprising to me as I think it's currently the best 35mm lens for the A7 series....but the FE 35 is better from f/2.8 to about f/4 at mid-field. (infinity). The ZM 35/1.4 has a noticeable mid-field dip at wider apertures (with or without the PCX lens) which is only completely masked at f/5.6. Mid-field is very important for landscapes and therefore I advise shooting at f/5.6 or smaller with the ZM. With the FE 35, f/5.6 looks very impressive and there is only a very slight improvement at f/6.3 and f/7.1 away from the center.
DavidBM wrote:
I have an idea that the little FE 35 is a bit fragile too.
I had a copy which was absolutely great at f2.8; just didn't really improve much so that by the time you were stopped down other lenses, like the lox, were a bit better.
Used it for hiking satisfactorily, except for contrast loss against the light (and indifferent sunstars at normal apertures)
But: at some point it acquired significant tilt and the left side became much worse overall. I never dropped it as far as I know. Presumably there must have been some mild impact, but it must have been mild, because I am very alert to things like that.
It's the only lens I've ever had that has done that; so I gave it away.
I wouldn't mind having it again as a hiking lens, but the search for a good copy, and the fear that it won't stay good, put me off...
David,
It could be build quality but I have not seen any other reports on this so we can't draw any conclusions. I believe this is a lens worth searching for a good copy because of its size/weight and very high IQ. The fact that it even compares to the mighty 35/1.4 ZM should tell us something.
I believe this FE has 3 aspherical elements which contribute to the unattractive bokeh balls in certain situations. This is my main negative about the lens.
Fred Miranda wrote:
David,
It could be build quality but I have not seen any other reports on this so we can't draw any conclusions. I believe this is a lens worth searching for a good copy because of its size/weight and very high IQ. The fact that it even compares to the mighty 35/1.4 ZM should tell us something.
I believe this FE has 3 aspherical elements which contribute to the unattractive bokeh balls in certain situations. This is my main negative about the lens.
Sure; it was just a bit upsetting. Maybe my copy took a big knock when hiking and I didn't notice, or maybe I just got unlucky. The only thing I didn't like about it was the look shooting in to the sun (and the extreme corners stopped down were fine but not great). I'd get another copy in a heartbeat if I was sure it was a decent one (it's a struggle returning lenses here in Aus - the shops want to return them to Sony, who then tell you it's in spec. The alternative is to get them from Adorama and pay postage on international returns.)
But yes; it was deeply impressive wide open! Here's an example that blew me away at the time of a shot wide open (if you notice little reflections, it was that this was taken through the glass of a hotel room - and still managed to be this sharp)
DavidBM wrote:
I'd get another copy in a heartbeat if I was sure it was a decent one (it's a struggle returning lenses here in Aus - the shops want to return them to Sony, who then tell you it's in spec. The alternative is to get them from Adorama and pay postage on international returns.)
Teds do have a 14 day exchange guarantee with two fairly big caveats - you can only exchange for store credit / another copy (no refunds) and it doesn't apply to "special orders" which I take to be anything they don't stock normally. Which is to say - a lot of things, especially in E mount. Most/all of the Batis and Loxia lenses are "special order", as is the 24-70 GM, etc. So I find the policy fairly unhelpful for Sony gear - it's a bit better for Canon/Nikon as they have a broader range off the shelf.
I have never needed to use it to return a copy more than once though, so I am not sure how they would react to potentially trying several samples of a lens.
I had a real corker of a return experience with a different retailer (not Teds) with one of my lenses. It was a horrifically bad sample, basically unusable before f8. The retailer was reluctant to swap it and rang the distributor. They put me on the phone to someone at the distributor who tried to blame "the megapixels" of my camera When I pointed out that "the megapixels" don't explain why performance improves at f8, they ummed and ahhed and eventually agreed to authorise the store to refund me. I miss B&H - my time in NYC was night and day in comparison. The returns department there just kept bringing down copy after copy of the VC15 for me to walk outside with, quickly test and summarily reject for smeared corners.
Fred Miranda wrote:
Testing four FE 35/2.8 lenses, I can tell you there is variation and it's noticeable. I also have a great copy of the 16-35/4 and I will test them against each other (I will post the results in another thread). I know for sure it will beat the zoom up to f/5.6 @35mm@ but the zoom does improve further at f/8 and f/9. I'm curious to compare them.
As you wrote, the ZM wins especially towards the corners and that's not surprising to me as I think it's currently the best 35mm lens for the A7 series....but the FE 35 is better from f/2.8 to about f/4 at mid-field. (infinity). The ZM 35/1.4 has a noticeable mid-field dip at wider apertures (with or without the PCX lens) which is only completely masked at f/5.6. Mid-field is very important for landscapes and therefore I advise shooting at f/5.6 or smaller with the ZM. With the FE 35, f/5.6 looks very impressive and there is only a very slight improvement at f/6.3 and f/7.1 away from the center....Show more →
I look forward to seeing your comparison of the 16-35FE. If I could get a fantastic copy of that lens I'd probably be happy with it but just don't have the inclination to play the lottery. My VC15 and L21 are a nice substitute for now. I did try out the 16-35 2.8 III from Canon in a store the other day and it looked excellent on my A7RII and MC-11, although I need to rent it to conduct some proper testing.
Aztatlan wrote:
Teds do have a 14 day exchange guarantee with two fairly big caveats - you can only exchange for store credit / another copy (no refunds) and it doesn't apply to "special orders" which I take to be anything they don't stock normally. Which is to say - a lot of things, especially in E mount. Most/all of the Batis and Loxia lenses are "special order", as is the 24-70 GM, etc. So I find the policy fairly unhelpful for Sony gear - it's a bit better for Canon/Nikon as they have a broader range off the shelf.
I have never needed to use it to return a copy more than once though, so I am not sure how they would react to potentially trying several samples of a lens.
I had a real corker of a return experience with a different retailer (not Teds) with one of my lenses. It was a horrifically bad sample, basically unusable before f8. The retailer was reluctant to swap it and rang the distributor. They put me on the phone to someone at the distributor who tried to blame "the megapixels" of my camera When I pointed out that "the megapixels" don't explain why performance improves at f8, they ummed and ahhed and eventually agreed to authorise the store to refund me. I miss B&H - my time in NYC was night and day in comparison. The returns department there just kept bringing down copy after copy of the VC15 for me to walk outside with, quickly test and summarily reject for smeared corners....Show more →
Huh; didn't know Ted's did that. They probably keep the 2.8/35 in stock, so if I want another copy it might be worth a try.
Who was the Other Retailer? Was it Digidirect? I found, when I had a horribly skewed copy of the 2.8/90 that they tried to make me talk to Sony. I had to email them copies of the relevant section of the consumer protection laws which say it's up to me whether the retailer or manufacturer addresses the issue. After muttering about ombudsmen etc. they replaced my copy, and the new copy I'm happy to say was spectacular. But it's not pleasant - I don't have the psychology that likes fighting the good fight when it can be reasonably avoided, so I am reluctant to buy something in the expectation of having a fight about returns!
just got the new kolari .2mm v4 thin sensor stack conversion and ran a comparison of the zm 35/1.4 vs the cv 35/1.7 today. the results were pretty surprising to me. both lenses perform dramatically better on the thinner sensor stack – you could get decent landscape shots wide open with both. on the stock sensor the cv has a better midzone while the zm has better corners (focusing at infinity in the center of the frame), but on the thinner sensor stack the cv appears to edge out the zm across the frame, if only by tiny margin.
sebboh wrote:
just got the new kolari .2mm v4 thin sensor stack conversion and ran a comparison of the zm 35/1.4 vs the cv 35/1.7 today. the results were pretty surprising to me. both lenses perform dramatically better on the thinner sensor stack – you could get decent landscape shots wide open with both. on the stock sensor the cv has a better midzone while the zm has better corners (focusing at infinity in the center of the frame), but on the thinner sensor stack the cv appears to edge out the zm across the frame, if only by tiny margin.
sebboh wrote:
just got the new kolari .2mm v4 thin sensor stack conversion and ran a comparison of the zm 35/1.4 vs the cv 35/1.7 today. the results were pretty surprising to me. both lenses perform dramatically better on the thinner sensor stack – you could get decent landscape shots wide open with both. on the stock sensor the cv has a better midzone while the zm has better corners (focusing at infinity in the center of the frame), but on the thinner sensor stack the cv appears to edge out the zm across the frame, if only by tiny margin.
here's a link to the full res comparison, and here's the downsized one: http://c1.staticflickr.com/1/624/33355308885_1edf1bf722_o.jpg
big thanks to nehemiah for lending the lenses to test out. ...Show more →
Would it be possible for you to perform the same comparison (A7v4 vs A7 stock) with a Loxia 21mm 2.8?
I would call it one of the best performing lenses on the stock A7 cameras which should have
the most issues with a thinner filter (being a wide angle lens with rear lens close to the sensor).
PS: I think by f/2.8 the ZM +5m Optosigma offers the same across frame sharpness (at least, as I am comparing ZM +5m on 42mp vs ZM +thin filter on 24mp).
BastianK wrote:
Would it be possible for you to perform the same comparison (A7v4 vs A7 stock) with a Loxia 21mm 2.8?
I would call it one of the best performing lenses on the stock A7 cameras which should have
the most issues with a thinner filter (being a wide angle lens with rear lens close to the sensor).
i'd love too, but i don't have loxia 21 or know anybody who does. i'm hoping that the contax g 21/2.8 will do well enough on the thin sensor for me to take a pass on the larger more expensive loxia. i don't actually have any native mount lenses to compare right now.
BastianK wrote:
PS: I think by f/2.8 the ZM +5m Optosigma offers the same across frame sharpness (at least, as I am comparing ZM +5m on 42mp vs ZM +thin filter on 24mp).
it's possible, it's hard to tell exactly with the different pixel counts. on the v4 a7 the zm is showing a lot of moire across the frame, it's hard to tell what it would look like at 42mp.
sebboh wrote:
i'd love too, but i don't have loxia 21 or know anybody who does. i'm hoping that the contax g 21/2.8 will do well enough on the thin sensor for me to take a pass on the larger more expensive loxia. i don't actually have any native mount lenses to compare right now.
Maybe someone here volunteers :-)
I tend to say: if there is no downside when using the Loxia 21 there will hardly be any with other lenses.
But honestly, I expect corners to look much worse when using the Loxia 21 on the thin filter v4 mod.
sebboh wrote:
it's possible, it's hard to tell exactly with the different pixel counts.
Yes, exactly.
Furthermore, and back to topic: it is very interesting for me to see the midzone dip of the ZM is also visible on the thin filter.
BastianK wrote:
Maybe someone here volunteers :-)
I tend to say: if there is no downside when using the Loxia 21 there will hardly be any with other lenses.
Well we know, that there are issues without the thicker filter stack for some lenses:
BastianK wrote:
Maybe someone here volunteers :-)
I tend to say: if there is no downside when using the Loxia 21 there will hardly be any with other lenses.
But honestly, I expect corners to look much worse when using the Loxia 21 on the thin filter v4 mod.
Yes, exactly.
Furthermore, and back to topic: it is very interesting for me to see the midzone dip of the ZM is also visible on the thin filter.
We know the Loxia 21 wouldn't perform well on the thinner filter mod but we are not sure on the level of degradation.
Perhaps the Loxia 35/2 and 50/2 wouldn't get affected that much or perhaps they may get even better.
If Kolari would send me a modded A7RII, I would test many FE and all Loxia lenses comparing stock and modded bodies at infinity.
I'd be more interested to see this comparison with stock A7RII vs A7RII V4 thin mod. Would it be expected for the results to be similar as using an older 24MP A7 body?
I believe the ZM35 1.4 is optimized for a current (modern) Leica stack, so going that thin might well be under the thickness for which the design is optimized. The CV35 1.7 might (and I emphasize might, because I don't know) be an older film design and, therefore, benefits from "as close to zero" as possible.
Thinner is not always better, as someone mentioned and Roger's measurements with the Otus showed. Getting it right - meaning as-designed for - will result in the lens delivering "as designed" results.
It is always possible that you have different priorities than the lens designer did. For example, in a related thread on the use of close-up lenses to improve corner sharpness, I suspect a trade-off is being made between geometric distortion and corner sharpness. That might be a good trade for some people and less so for others.
Don't interpret this as a slam on exploring other options. After all, I'm the guy who started this thin filter stack adventure with Ilija. So I'm all for him being able to offer solutions to meet different people's priorities. What does concern me somewhat is people searching for a single "Holy Grail" on this. There isn't one and the science is well understood.
Know what you are optimizing for.
Michael
sebboh wrote:
just got the new kolari .2mm v4 thin sensor stack conversion and ran a comparison of the zm 35/1.4 vs the cv 35/1.7 today. the results were pretty surprising to me. both lenses perform dramatically better on the thinner sensor stack – you could get decent landscape shots wide open with both. on the stock sensor the cv has a better midzone while the zm has better corners (focusing at infinity in the center of the frame), but on the thinner sensor stack the cv appears to edge out the zm across the frame, if only by tiny margin.
Gary Clennan wrote:
I'd be more interested to see this comparison with stock A7RII vs A7RII V4 thin mod. Would it be expected for the results to be similar as using an older 24MP A7 body?
i'd love to show you some but, i only have an a7. it seems like the general behavior of rangefinder lenses is the same on all the a7 versions with regard to astigmatism and field curvature. color shift goes away on the a7rii though due to the backlit sensor. some people have reported very slightly improved performance of some lenses on the a7rii versus other models but i haven't seen any actual comparisons that bear this out. both these lenses are easily outresolving my a7 from pretty much wide open, so i would expect the difference to be even bigger at large apertures between modded and unmodded sensors.
mdemeyer wrote:
I believe the ZM35 1.4 is optimized for a current (modern) Leica stack, so going that thin might well be under the thickness for which the design is optimized. The CV35 1.7 might (and I emphasize might, because I don't know) be an older film design and, therefore, benefits from "as close to zero" as possible.
Thinner is not always better, as someone mentioned and Roger's measurements with the Otus showed. Getting it right - meaning as-designed for - will result in the lens delivering "as designed" results.
It is always possible that you have different priorities than the lens designer did. For example, in a related thread on the use of close-up lenses to improve corner sharpness, I suspect a trade-off is being made between geometric distortion and corner sharpness. That might be a good trade for some people and less so for others.
Don't interpret this as a slam on exploring other options. After all, I'm the guy who started this thin filter stack adventure with Ilija. So I'm all for him being able to offer solutions to meet different people's priorities. What does concern me somewhat is people searching for a single "Holy Grail" on this. There isn't one and the science is well understood.
this is a very good point, both these 35mm lenses are designed for digital leicas (according to their manufacturer). this suggests that the closer you get to the m9 sensor stack the better, and going to thin would probably degrade their performance a small amount. the kolari v4 seems very close to the effective thickness of the m9's and these two lenses seem pretty much perfectly corrected for field curvature with it – no real difference between focus on the corner and focus on the center. my contax g 28mm and 21mm , which where designed for film however, still show a bit of outward field curvature such that best focus for across the frame sharpness is partway between center and corner focus (i'll post more about this in the thin sensor and corrective filter threads).
BastianK wrote:
Maybe someone here volunteers :-)
I tend to say: if there is no downside when using the Loxia 21 there will hardly be any with other lenses.
But honestly, I expect corners to look much worse when using the Loxia 21 on the thin filter v4 mod.
Yes, exactly.
Furthermore, and back to topic: it is very interesting for me to see the midzone dip of the ZM is also visible on the thin filter.
I have already volunteered a bunch but I also have a FE 16-35 Derek can use if that is helpful?
DavidBM wrote:
Huh; didn't know Ted's did that. They probably keep the 2.8/35 in stock, so if I want another copy it might be worth a try.
Who was the Other Retailer? Was it Digidirect? I found, when I had a horribly skewed copy of the 2.8/90 that they tried to make me talk to Sony. I had to email them copies of the relevant section of the consumer protection laws which say it's up to me whether the retailer or manufacturer addresses the issue. After muttering about ombudsmen etc. they replaced my copy, and the new copy I'm happy to say was spectacular. But it's not pleasant - I don't have the psychology that likes fighting the good fight when it can be reasonably avoided, so I am reluctant to buy something in the expectation of having a fight about returns!...Show more →
Yep, DD indeed. They were actually fairly willing to replace it with a different sample off the shelf but testing revealed that too was less-than-stellar. As it was the only other copy they had, I pushed for a refund.
)
When I pointed out that "the megapixels" don't explain why performance improves at f8, they ummed and ahhed and eventually agreed to authorise the store to refund me. I miss B&H - my time in NYC was night and day in comparison. The returns department there just kept bringing down copy after copy of the VC15 for me to walk outside with, quickly test and summarily reject for smeared corners.