New 35 Lux bokeh, foliage - and focus

[tashley]

Advertisement (gone after registration)

I hope the mods see fit to leave this thread here since it relates reasonably specifically to the performance of this interesting new lens on the M9.

 

Here's a link Tim Ashley | 35 Lux Forest Bokeh

 

to a series of shots I took this evening using the new 35 lux on an M9. Recent threads including one of my own have been concerned as to the nature of the bokeh of this lens, especially WRT foliage. I also had, with my first version, a copy which didn't focus correctly. Now I have a copy which does - but with a twist, which I believe to be endemic to the design, which is that at at all wide-ish apertures (say 1.4 thru 5.6 but especially at F4) the lens, when focussed through a well-calibrated RF, delivers images which are slightly soft at the centre of the frame but sharp elsewhere, and very good to the corners. If you pull focus slightly closer to get the centre perfectly sharp, you throw the periphery slightly off.

 

In practice, with focus and recompose, I find that this gives good results since my main point of interest is rarely bang in the centre of the frame. And in my stroll this evening the things on which I had intended to focus were in the great majority in focus so though I'd naturally prefer to have a lens that had a flat field of focus, I'll live with this one because it is fast and sharp and small and has very little focus shift.

 

So these shots (no artistic merit, very little if any processing and dodgy WB) are really there to show that it does focus pretty well, and to give examples at a variety of subject distances and apertures, of what the bokeh and DOF are like.

 

I hope it's useful to someone! The shot below is small-ish but if anyone wants to they can follow the link above to a series of others, all of which can be downloaded at full size though due to bandwidth constraints at 85% quality JPEG.

 

All shot RAW and processed in LR to defaults are thereabouts. Adobe RGB rather than sRGB because, well, if your monitor doesn't do the Adobe space yet, you might need to get one that does!

 

 

Tim

 

[01af]

Adobe RGB rather than sRGB because, well, if your monitor doesn't do the Adobe space yet, you might need to get one that does!

This is foolish because proper colour rendition doesn't depend on the viewer's monitor's gamut but on your image file being tagged correctly. And yours is tagged incorrectly. So everybody will see the wrong colours, no matter what their monitor's gamut may be.

 

Furthermore, it would have been helpful if you cared to disclose the aperture your image has been taken with. Yes, I know—EXIF says f/2.8 but we all know how dependable the M9's aperture value in the EXIF data is.

[tashley]

This is foolish because proper colour rendition doesn't depend on the viewer's monitor's gamut but on your image file being tagged correctly. And yours is tagged incorrectly. So everybody will see the wrong colours, no matter what their monitor's gamut may be.

 

Furthermore, it would have been helpful if you cared to disclose the aperture your image has been taken with. Yes, I know—EXIF says f/2.8 but we all know how dependable the M9's aperture value in the EXIF data is.

 

I'd there's nothing useful you can draw from this exercise then please feel free to pass. No obligation here.

[microview]

The amount of purple fringing showing on L1000862 is in my view somewhat extreme for a £3k+ lens, and I noticed something similar when trying the new lens courtesy of Leica Mayfair* some weeks ago. Any observations?

 

* I took my v4 Summicron along for comparisons and similar shots in Bruton Place showed a relative immunity from fringing with the non-ASPH 35!

[tashley]

The amount of purple fringing showing on L1000862 is in my view somewhat extreme for a £3k+ lens, and I noticed something similar when trying the new lens courtesy of Leica Mayfair* some weeks ago. Any observations?

 

* I took my v4 Summicron along for comparisons and similar shots in Bruton Place showed a relative immunity from fringing with the non-ASPH 35!

 

It's pretty strong yes... and I'm not the world's biggest defender of this lens! But I would say that given this shot was at F1.4 and against the light with strong contrast, exposed for the mid ground, it's not unexpected wide open. In general I would rarely expect to use such an aperture in bright daylight... which is why I did it on this occasion, just to see what would happen. It would clean up OK in most good RAW editors but hey, you're right, it's there and it's strong-ish.

 

Try reading Sean Reid's review... it makes the point that the performance of this lens in these conditions is pretty good all other factors considered and I sort of agree, though of course I wish it were better! And of course the 'cron doesn't go to F1.4!

[lct]

...I hope it's useful to someone...

Sure it is. Thank you much for sharing. I'm a bit concerned about your focussing issue though. Please correct me if i'm wrong but should i understand that you must focus behing or in front of the subject matter? On this pic for instance, did you focus on the yellow flower in the center?

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[archi4]

Advertisement (gone after registration)

Tim,

Thanks for posting these.

After following your post on the first lens, I rechecked my (old) 35 lux asph again for focus shift using 5 printed charts set up staggered as I had done before. With my M8 focus was spot on wide open at various distances from 3 meters to infinity. On the stagger test at about 10 feet there was no focus shift at f2 and f2.8. At f4 the card approx 1cm behind appeared very slightly sharper when pixel peeping. Using my Panasonic 4/3 with adapter gave the same results. This is my 3rd 35 s'lux Asph and the first one which has practically no focus shift without front focus. My first one was the Aspherical one and that one had a very strong flare problem.

I did like the pictures, and when downloaded and opened in my CS5 Prophoto working space I got the message that your images were Adobe RGB so they were correctly tagged.

regards, maurice

[microview]

I would rarely expect to use such an aperture in bright daylight... which is why I did it on this occasion, just to see what would happen.

 

Forgive my ignorance but are you saying that by stopping down fringing can be minimized under such lighting conditions?

(I was very impressed by the detail in the city-scape you posted elsewhere [p179501073.jpg] and am accordingly tempted by the lens. I recently found in mint condition a black paint 50mm pre-ASPH Summilux and it's an outstanding performer – I was more than pleased to ditch my 50mm ASPH Summilux, which always seemed soft under 1m distance. Now I am finding the v4 35 Summicron relatively disappointing!)

[edlaurpic]

Interesting discussion, where did you obtain a 35/1.4 Lux Asph (11663) in the first place?! No one has them in the U.S., no one of whom I am aware, anyway.

[tashley]

Sure it is. Thank you much for sharing. I'm a bit concerned about your focussing issue though. Please correct me if i'm wrong but should i understand that you must focus behing or in front of the subject matter? On this pic for instance, did you focus on the yellow flower in the center?

 

I did focus on the yellow flower, yes.

 

The issue of focus with this lens, or at least my copy of it, which I think is a well calibrated one, is complicated. At F1.4, F2 and F8 or smaller, and particularly at relatively short distances, focus is quite accurate. At F2.8 and worst at F4 as far as I can see, focus is accurate BUT the field of accurate focus is not as you expect: focussing accurately on a subject in the centre of the RF will mean that subjects at that same distance in the rest of the frame will be in good focus but the central section will be slightly OOF. If you pull focus slightly closer so as to get the centre in exact focus, you will throw the rest of the frame slightly out. You have to try it to understand it. And the effect is worse at medium to longer distances, on my lens at least. I found the exact same thing with the previous version of the 35 lux and to a lesser extent with the cron but I don't own either now so can't accurately check my memory! However the old lux had focus shift too, which complicated matters to the point where I couldn't be bothered to deal with it.

 

So to answer your question: at F2.8 and F4 and a little at 5.6, if you want the exact centre of the frame to be in perfect focus, practice pulling focus a tiny bit closer than the RF indicates but be willing to lose perfect sharpness elsewhere.

 

If you can be bothered!

[tashley]

Forgive my ignorance but are you saying that by stopping down fringing can be minimized under such lighting conditions?

(I was very impressed by the detail in the city-scape you posted elsewhere [p179501073.jpg] and am accordingly tempted by the lens. I recently found in mint condition a black paint 50mm pre-ASPH Summilux and it's an outstanding performer – I was more than pleased to ditch my 50mm ASPH Summilux, which always seemed soft under 1m distance. Now I am finding the v4 35 Summicron relatively disappointing!)

 

Most lenses have less bad fringing when stopped down a bit - very fast lenses are prone to fringing wide open. If you can, subscribe to reid reviews - it's really worth it. His review of fast 35mm lenses on the M9 will show you all you need to know but basically with the exception of the previous version of the 35 Lux, the new 35 lux has the best wide open CA performance of its peer group and this CA disappears when you stop down even to F2 in most shots.

 

You had a bad 50 lux. A good copy (actually I've not seen a bad one, mine is perfect) will blow you away at all distances and apertures. It is the best lens I have ever used, accurate, poetic, gorgeous!

 

If you're tempted by the 35 lux then also check Sean's review of the CV 35mm F2.5 lens. I have one and on centre it is reliably sharper than my new lux and is a great sunny day lens because it handles contrast slightly better BUT it has to be stopped down to F8 to get the corners sharp and much beyond that you start to get diffraction effects. However, it is really cheap, tiny, and at F8 pretty damned good!

[tashley]

Tim,

Thanks for posting these.

After following your post on the first lens, I rechecked my (old) 35 lux asph again for focus shift using 5 printed charts set up staggered as I had done before. With my M8 focus was spot on wide open at various distances from 3 meters to infinity. On the stagger test at about 10 feet there was no focus shift at f2 and f2.8. At f4 the card approx 1cm behind appeared very slightly sharper when pixel peeping. Using my Panasonic 4/3 with adapter gave the same results. This is my 3rd 35 s'lux Asph and the first one which has practically no focus shift without front focus. My first one was the Aspherical one and that one had a very strong flare problem.

I did like the pictures, and when downloaded and opened in my CS5 Prophoto working space I got the message that your images were Adobe RGB so they were correctly tagged.

regards, maurice

 

Thanks Maurice... as you say, the files are correctly tagged for color space so I'm not sure what the other guy was talking about but hey..!

 

Glad you got a 'good' old version 35 lux - still the holy grail as far as I am concerned, I've just never come across one, though Jamie Roberts has one too and I've been meaning to steal it for years!

[tashley]

Interesting discussion, where did you obtain a 35/1.4 Lux Asph (11663) in the first place?! No one has them in the U.S., no one of whom I am aware, anyway.

 

I got it from Caplan in London - I've been on the list for months so got first bite at the cherry!

[mjh]

Forgive my ignorance but are you saying that by stopping down fringing can be minimized under such lighting conditions?

This being purple fringing rather than (lateral) chromatic aberration, it is indeed to be expected that stopping down should help quite a bit.

[lct]

...So to answer your question: at F2.8 and F4 and a little at 5.6, if you want the exact centre of the frame to be in perfect focus, practice pulling focus a tiny bit closer than the RF indicates but be willing to lose perfect sharpness elsewhere.

If you can be bothered!...

Not sure i can but thank you much anyway. Sounds like floating lenses don't simplify the matter.

[DFV]

I hope the mods see fit to leave this thread here since it relates reasonably specifically to the performance of this interesting new lens on the M9.

 

 

Thanks for posting again. I found the comparison of you other two shots quite intriguing. It is the first time I have seen this by only changing the focus plane. What strikes me as very odd in your demo shots is that on the picture that has the edges sharp (near and far) only the center is slightly blurry. This was done by changing the focus and not the aperture, right?

 

Now, asking the engineers here; Could this be due to to the design of the lens... ergo, the new floating element? I would really like to go a bit deeper into these results or at least get more info.

[tashley]

Thanks for posting again. I found the comparison of you other two shots quite intriguing. It is the first time I have seen this by only changing the focus plane. What strikes me as very odd in your demo shots is that on the picture that has the edges sharp (near and far) only the center is slightly blurry. This was done by changing the focus and not the aperture, right?

 

To clarify, if you're referring to the roofline shot of London, you're exactly correct I think; if I had pulled focus slightly closer than the RF indicated on the round window in the middle distance, it would have been in great focus but other things at its distance would have been slightly less sharp. This is mainly a focus effect but it seems to me to be worst at 4 and thereabouts.

 

 

Now, asking the engineers here; Could this be due to to the design of the lens... ergo, the new floating element? I would really like to go a bit deeper into these results or at least get more info.

 

I am no engineer (of lenses or anything else!) but I would love to know what is going on here! I have an idea for a little experiment which I will now conduct and report back!

 

Best

 

Tim

[DFV]

To clarify, if you're referring to the roofline shot of London, you're exactly correct I think; if I had pulled focus slightly closer than the RF indicated on the round window in the middle distance, it would have been in great focus but other things at its distance would have been slightly less sharp. This is mainly a focus effect but it seems to me to be worst at 4 and thereabouts.

 

Yes, the roofline of London. That is exactly what puzzles me, on the second shot the sides are in focus at practically all distances. Now, that is quite peculiar if the aperture is not modified. I know that some lenses have a moment where the center is slightly less sharp in the center in benefit of the edges at optimum aperture (about f 8.0) but not when you change focus. At least this is the first time I see this effect...

[mjh]

I am no engineer (of lenses or anything else!) but I would love to know what is going on here!

Curvature of field would be the technical term; the image surface isn’t planar but curved, resulting in a similarly curved plane of maximum sharpness.

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[01af]

Curvature of field would be the technical term ...

No, it wouldn't.

 

If field curvature was the culprit, then the distribution of sharpness in the 'London roofs' image would not be the way it is.