I've flipped through the various threads on these two lenses (in the current Zeiss line-up), and I'm still in a bit of a quandary as to which is likely the better landscaping lens (stopped down performance). The new f/1.4 seems to have better corner-to-corner performance and better CA control, but it also has significant field curvature. Any thoughts?
I have no experience with 35/2 and can't offer any comparison. But from my limited testing, the 35/1.4 is as suitable for landscapes as 21/2.8 (with the added benefits of low light performance and narrow dof). If the performance stopped down is comparable between 35/1.4 and 35/2, the size and weight may be the deciding factor for you.
Mirek Elsner wrote:
But from my limited testing, the 35/1.4 is as suitable for landscapes as 21/2.8 (with the added benefits of low light performance and narrow dof).
I would disagree with that. Although the 35/1.4 has better contrast in the fine details than the 35/2 it comes nowhere close to the 21 Distagon (exceptionally few lenses do). This is shown both in MTF charts and sample images (yours included). So while the 35/1.4 may be better than the 35/2 for landscapes, it's definitely not comparable to the 21.
So while the 35/1.4 may be better than the 35/2 for landscapes, it's definitely not comparable to the 21.
That's not what I had on my mind, maybe I need to provide some clarification: As much as 21 is probably one of the best choices for landscape around 20mm, this 35 may be one of the best choices for landscape at 35mm. With the disclaimer that I don't have any experience with 35/2.
edwardkaraa wrote:
Indeed, it's curious to see that the 1.4 is better than the 2 for landscapes but not as good at f/2
I remember that the 35/1.4 is better than the 35/2 stopped down
May 08, 2011 at 10:01 PM
Lars Johnsson Offline • • • • • • • Upload & Sell: Off
If you don't mind spending about twice as much for the f/1,4 lens, then buy it. I would not do it unless I needed the fast apertures also.
I also agree with those who think it's strange that the fast lens is better stopped down. But worse at f/2 than the other lens
The 35/2 is great for landscaping.
I would buy the 1.4 for the larger aperture and very nice bokeh. Not really the things you need when doing landscapes.
So in this case: 35/2, because you won't gain much when using the 1.4 stopped down.
The green vertical lines mark the short edge, long edge and corners of the image.
I've always found the 35/2 a bit lacking for landscapes and it looks like the 35/1.4 wouldn't suit me at least either. It's different with a better rendering of fine detail in the center, but that drops off to the edges. The field curvature seems to be significant something that isn't good in a landscape lens.
Going a bit offtopic, but on the general theme - how would the MTF curves of a good landscape lens look like? Well, I know from personal experience that the Zeiss 35/2 Biogon ZM (rangefinder lens) is really good. This is what we see when we look at the MTF charts:
black = 35/2 ZE (Distagon)
blue = 35/2 ZM (Biogon)
The Leica has higher overall contrast/resolution but it shows a lot of field curvature. I really love that lens but I've never been completely sold on it as a landscape lens. The wavy lines are probably the explanation. You don't want that field curvature - the ideal landscape lens would have high contrast in the fine detail (i.e the 40 lp/mm line) and would be flat across the field.
The 21 Distagon is an example of a lens that isn't far from the ideal. Two other candidates would be the Leica Summicron-M 28/2 ASPH:
--
So to round things up - 35/1.4 ZE vs 35/2 ZE for landscapes: Not easy - the 35/2 is more consistent across the field with less field curvature but its rendering of fine detail could be much better. The 35/1.4 produces better detail center/mid frame but suffers from field curvature meaning an uneven resolution across the frame. Take your pick
The best landscaping 35 that I've used is the 35/2 Biogon, a lens that I thoroughly dislike at larger apertures. It's an unfortunate reality that just because a lens is very good at one thing doesn't automatically mean it's good at everything.
Thanks for your analyses, Luka, very informative! It also reflects my skepticism about the 35 2.0 as a landscape lens. For me it's a lens from what I get the most impressing results at f2.0, but I am less overwhelmed with its stopped down performance for landscapes (relative to what I expect given its wide open performance). I even often grab the 21mm instead and do some cropping afterwards. Subjectively my Zeiss 35-70 3.4 impresses me more for stopped down landscapes at 35mm, but I have to compare it more thoroughly against the 35/2 (at least the ZE-color rendition is nicer) .
denoir wrote:
...
but the 35/2 overtakes it somewhere around 8 mm.
...
you mean that Sagital is more important than tangential for subjective sharpness ?
Sorry but if so, how much the sagital curve can be more important than tangential for sight perception ?
Until now, I thought two things were equally important.
BTW, I also use for now most of the time my ZE35/2 at full aperture.
I own the 2.0/35ZE and tested the 1.4/35ZE for a few hours. My experience is, that you need to stop down the 1.4/35 to f/8-f/11 to get the same corner sharpness as with the 2.0/35ZE. At the same time the center sharpness/resolution of very fine details (high frequency) is at all apertures slightly better with the 1.4/35ZE. But the 2.0/35 has a slightly higher contrast of lower frequency structures. I will not buy the 1.4/35ZE and keep using my 2.0/35ZE for landscape work.
I'm no scientist, but I've been very impressed with the ZE 35/2, especially in the center, but also better than I expected (but not perfect!) at the edges. I'm still in the process of getting familiar with it, but very encouraged so far.
trusty wrote:
you mean that Sagital is more important than tangential for subjective sharpness ?
That's a very good question. Both components are important, and it is dependent on the subject which will have more impact. I think the sagittal blur is more important for edge contrast as a blur in the radial direction will typically impact it the most. But that's not really the fine detail we're talking about here, but rather the 20 lp/mm or 10 lp/mm lines.
jffielde wrote:
Denoir, that you for taking the time to present your information to me. Much appreciated.
Your welcome, but I was very interested myself in the comparison.
Bobu wrote:
I own the 2.0/35ZE and tested the 1.4/35ZE for a few hours. My experience is, that you need to stop down the 1.4/35 to f/8-f/11 to get the same corner sharpness as with the 2.0/35ZE. At the same time the center sharpness/resolution of very fine details (high frequency) is at all apertures slightly better with the 1.4/35ZE. But the 2.0/35 has a slightly higher contrast of lower frequency structures.
That agrees generally with what the MTF charts say. As for the corners of the 35/1.4, it seems to be good in the sagittal direction (almost as good as center sharpness) while not as good as in the tangential direction. Of course the MTF chart above was just for f/4 - it's difficult to know what happens at f/8.
denoir wrote:
But that's not really the fine detail we're talking about here, but rather the 20 lp/mm or 10 lp/mm lines.
Do you have a source for that statement, or are you making that up? Canon has a page where they explain that the 10lp/mm curve is an indicator for global contrast and 20lp/mm for microcontrast (local contrast). A similar explanation is given in the article about understanding MTF by Dr. Nasse from Zeiss.
30-40lp/mm should be relevant for edge contrast and fine detail I'd think.
I guess it depends on your definition of edge contrast, but I think about acutance, or the contrast that sharpening enhances. To my understanding microcontrast describes the contrast of larger scale elements (local contrast).
"We must not forget that when we talk about ‘contrast’ we always mean micro contrast, i.e. structures, which we can just about see or just cannot see with the naked eye, for example on a slide."
The 10 lp/mm is typically the edge contrast, i.e relatively large scale detail. The 40 lp/mm is the really fine detail in the image and definitely not very relevant for the edge contrast.
Do you have a source for the Canon thing? It would really be quite stupid of them to call the 10 lp/mm for "global contrast". Global contrast is something completely different and relates to the image as a whole.
