21mm Summilux on M9

[sclamb]

Advertisement (gone after registration)

Here another one.

 

24/1.4 full frame at f/1.4 with no PP:

 

 

100% crop from left edge of frame with no PP:

 

 

100% crop from left edge of frame with Purple Fringing checkbox enabled in C1:

 

 

Note that it is hard to get rid of the green fringing around the windows and black post of the house in the background, although by f/4 is is all but gone.

 

Simon

[Googaliser]

Hi,

Leica have opined on my images.

The footer of the Email prevents me from reproducing it (confidential).

Lets just say that I do not see any need to test another 21mm 1.4 since I believe I will see the same results.

This would also be consistent with a quote I saw earlier in this thread..........

 

"In reply to the question of the customer you should know that the colour fringe the customer complains is a system (lens) inherent effect and there is no need to adjust the lens. To compensate colour fringe you can use a special tool in Adobe Lightroom 2.5 (resp. Camera Raw 5.5) DNG converting named defringing. By using this tool the effect of colour fringing is noticeable reduced."

[markowich]

Hi,

Leica have opined on my images.

The footer of the Email prevents me from reproducing it (confidential).

Lets just say that I do not see any need to test another 21mm 1.4 since I believe I will see the same results.

This would also be consistent with a quote I saw earlier in this thread..........

 

so leica gives confidential answers to customers with service requests and product information?

peter

[wlaidlaw]

Hi,

Leica have opined on my images.

The footer of the Email prevents me from reproducing it (confidential).

Lets just say that I do not see any need to test another 21mm 1.4 since I believe I will see the same results.

This would also be consistent with a quote I saw earlier in this thread..........

 

Sounds like when you take your car to a garage and the mechanic says "oh they all do that". Well they all may but it does not make it acceptable. It may be a case of back to the drawing board. If the lens has this inherent problem, then it should carry a health warning to that effect, given that it is one of the world's most expensive lenses. I am not convinced that any processing will eradicate it totally.

 

Wilson

[Googaliser]

Sounds like when you take your car to a garage and the mechanic says "oh they all do that". Well they all may but it does not make it acceptable. It may be a case of back to the drawing board. If the lens has this inherent problem, then it should carry a health warning to that effect, given that it is one of the world's most expensive lenses. I am not convinced that any processing will eradicate it totally.

 

Wilson

 

The footer is a fairly generic one preventing the reproduction of the Email and protecting its contents as confidential and for the intended recipient only yada yada yada - so I wouldn't read anything sinister into that. I do however take these things seriously.

 

I am voting with my feet though and will not be buying another fast WA lens for my Leica system (not my primary use for the M9 anyway) unless its has been specifically built for digital (like the Hassy HCD range). I sense we are reaching the point with some manufacturers now where there is a palpable shift away from backwards compatability - towards the present and future. The risk is that it starts to become a 'closed' system.

 

I heard a rumour that there may be a new 35mm Lux next year. A digital version perhaps ?

[georg]

Have you made some tests, propably with some newspapers (something 2D) on each edge and some with film? As said, your examples have some normal artifacts and some that look not ok. It's difficult to say from our perspective.

 

Have you talked to Leica in Solms?

 

Leica won't offer "digital"-lenses within the M-system because they would be unusable with film. But I'm sure sensor-technology will further improve. As far as I know, the HCD-lenses from Fuji are not "digital" in that matter, they just have a smaller image circle and optical aberrations which can be hidden to a certain degree in a digital workflow - as well as the aberrations from the SX21/M9.

 

@sclamb

As mentioned, what happens within the unsharp zones is not an issue of optical performance but "character/bokeh".

 

Here's a quick test I made with the M8 and the 90AA @f5.6 - this lens is free of any visible CA, the artifacts shown are due to the sensor or in the unsharp background:

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!

[sclamb]

Advertisement (gone after registration)

I still wonder from looking at your pictures whether you had a seriously bad sample of the 21/1.4. Leica have sold enough that I am sure we would have heard if there was a generic design fault.

 

Your CA looks so dramatically worse than mine that either you really did have a duff sample of the lens or perhaps Leica pushed the boundaries too far in making a 21/1.4 and maybe 24mm was the limit to get satisfactory performance.

 

Either way, if you like the 21mm focal length I would try the other samples that you said the dealer had access to.

 

As for the response from Leica, I cannot understand how they could say that what you had at f/16 was in any way acceptable.

 

Simon

[Googaliser]

Have you made some tests, propably with some newspapers (something 2D) on each edge and some with film? As said, your examples have some normal artifacts and some that look not ok. It's difficult to say from our perspective.

 

Hi - no I decided not to test further. I work on the principle that if my first shots show material issues - I stop there and investigate. Time is money - and I feel I have spent a lot of time on the 21mm 'Lux already - let alone having to nurture its images into the world via C1 - a package I don't currently use. I can only talk from my own perspective though - and if the 21 Lux is a critical lens for your work - it sounds like the issues that may occur with the M9 can be largely mitigated through a C1 workflow. The question you have to ask yourself is *how* much is that extra hassle worth to you - and is that hassle you are prepared to (or should have to) go through having already forked out over $6,000 for a lens. For me - the answer is no on both counts sadly.

I will keep using the Nikon D3 for this aspect of my work. I can partially make up for the extra couple of stops (on the 14-24) with the higher noise floor of the D3 sensor

[innerimager]

I'm following this thread with great interest as I have been considering the 21 lux for my M8 to have an equivalent 28/1.4. This is a most used FL for me on my D3 where the nikon 28/1.4 gets a lot of action in clubs and other low light work. In fact I was considering selling my nikon to help fund the leica 21. While I would be concerned about any M9 in my future in terms of this problem, can anyone speak to whether this CA (or whatever) is significantly worse on the M9 than on the M8 as may well be the case? This would be very helpful to me, and probably other M8 owners. Thanks in advance....Peter

[wlaidlaw]

I still wonder from looking at your pictures whether you had a seriously bad sample of the 21/1.4. Leica have sold enough that I am sure we would have heard if there was a generic design fault.

 

Your CA looks so dramatically worse than mine that either you really did have a duff sample of the lens or perhaps Leica pushed the boundaries too far in making a 21/1.4 and maybe 24mm was the limit to get satisfactory performance.

 

Either way, if you like the 21mm focal length I would try the other samples that you said the dealer had access to.

 

As for the response from Leica, I cannot understand how they could say that what you had at f/16 was in any way acceptable.

 

Simon

 

It may well be with this extreme technology lens, rather like the 35 ASPH Lux, there is more sample variation than Leica would like but for economic reasons, they cannot reject all the less than 100% perfect ones, so they end up at dealers. This was one of the factors in the breakdown of the Zeiss/Kyocera relationship. Zeiss were said to be rejecting over 75% of the N and 645 lenses, which made the whole process uneconomic as far as Kyocera were concerned. I would not be surprised to see a new internal focussing 35/1.4, more for the economics of a higher acceptability rate on the production, rather than seeking a better lens.

 

Wilson

[rosuna]

Sounds like when you take your car to a garage and the mechanic says "oh they all do that". Well they all may but it does not make it acceptable. It may be a case of back to the drawing board. If the lens has this inherent problem, then it should carry a health warning to that effect, given that it is one of the world's most expensive lenses. I am not convinced that any processing will eradicate it totally.

 

Wilson

 

That fringe may be due to a combination of variables, related to the lens, the angle of incidence of the light, the microlenses, the sensor's filters, the sensor itself or even the RAW processing. I am not sure. I would like to see comparative tests with digital versus color film.

 

Those lenses may be used on film and digital cameras, so if the problem is not only in the lens, the only reasonable warning would be "consider the possible presence of fringe problems on the current generation of (digital) sensors".

 

I had a 24mm Summilux and the fringe problem was there (M8), that is true. And it was much more notorious than the occasional fringe problems I see in other Leica lenses (on a digital body).

 

On the other hand, the maximum magnification of a film based photography is determined by the size of the positive copy. We see digital files at 100% on screen, and this is a huge magnification (it is like seeing the negative magnified to more than 1.5x1 meter). Those fringe artifacts may be invisible on the final printed copy. So, there is also a problem of perception here (to some extent)?

[georg]

I would replace the Summilux by the dealer, a D3+14-24 is a rather poor replacement. I spoke to Nikon/Canon-guys when showing them loose focussing rings, CA and unsharp edges - I always got the answer not to look too close or to stop down - no replacement... Maybe appropriate for them, but not for this 5k€-lens...

[tashley]

I still wonder from looking at your pictures whether you had a seriously bad sample of the 21/1.4. Leica have sold enough that I am sure we would have heard if there was a generic design fault.

 

 

 

Simon

 

Simon, that's not a great assumption, however reasonable it sounds. People merrily purchase a couple of other lenses I can think of from the current line-up without ever noticing their endemic (or maybe pandemic) problems!

 

T

[sclamb]

Simon, that's not a great assumption, however reasonable it sounds. People merrily purchase a couple of other lenses I can think of from the current line-up without ever noticing their pandemic (if not quite endemic) problems!

 

T

 

Agreed. However, I know three people with the 21/1.4 and they have not mentioned such bad optical issues as the OP. I am sure they would have said something by now as I know they have used the lenses quite a bit, they are very critical of image quality and they don't like spending over £4,000 on something that is not fit for purpose.

 

Simon

[Googaliser]

I would replace the Summilux by the dealer, a D3+14-24 is a rather poor replacement. I spoke to Nikon/Canon-guys when showing them loose focussing rings, CA and unsharp edges - I always got the answer not to look too close or to stop down - no replacement... Maybe appropriate for them, but not for this 5k€-lens...

 

You're right of course - until recently my favourite available light combo was the D3 + 28/1.4. I did own a 55/1.2 Noct for a while - but I found focussing a challenge on the D3 screen (athough was seriously tempted to keep due to the out of focus characteristics and lack of coma).

But now with the M9 - the 50 summilux is my champ - I just love it (and more importantly so do my subjects) - and this is my prefered combo for indoor parties. I actually don't mind the M9 noise when converted to B&W and find the grain-like characteristics acceptable up to about 1250 as long as the subject isn't in shadow.

[tashley]

Agreed. However, I know three people with the 21/1.4 and they have not mentioned such bad optical issues as the OP. I am sure they would have said something by now as I know they have used the lenses quite a bit, they are very critical of image quality and they don't like spending over £4,000 on something that is not fit for purpose.

 

Simon

 

Fair do's. I must say, I'm tempted to get one myself given that I can't get a decent copy of the 18mm super-elmar. I just don't fancy the embarrassment of having to take two lenses back to my poor dealer in a row if I get a pup!

[sclamb]

Fair do's. I must say, I'm tempted to get one myself given that I can't get a decent copy of the 18mm super-elmar. I just don't fancy the embarrassment of having to take two lenses back to my poor dealer in a row if I get a pup!

 

That's a shame as I had an amazingly good 18/3.8 but found it too wide for me on the M9. As I liked the 24mm FOV that it gave on my old M8.2 I decided to trade it and my 24/3.8 for the 24/1.4.

 

Simon

[mwilliamsphotography]

I've been using the 24 Lux for some time now. When I first got it I did experience some CA and communicated with Leica about it. The response was that it was a mild compromise easily dealt with in LR, ACR in order to have a very fast 24mm M lens that didn't get too big ... "too big" being the operative words.

 

In practice Leica was right ... the lens is incredibly sharp out to the edges and only occasionally produces CA ... at which time I deal with it ... a process that takes all of 5 seconds when needed, which isn't often.

 

I'd say that if one bought a 21/1.4 lens at that price point to shoot in bright light, perhaps a 21/2.8 ASPH would be a better choice.

 

I tend to use it in dim ambient light with strong light sources against darker backgrounds, and experience very little CA ... and when it shows up usually about a 12 to 18 setting on the slider totally eliminates it.

 

IMO, this lens is optimized for low light for those that need that end of the performance scale ... of which I am one. However, even in daylight situations I've never experienced CA at the level being shown by the OP.

 

-Marc

 

(M9 ISO 800, 24/1.4 ASPH)

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!

[wlaidlaw]

Marc,

 

I hear what you say but one of the USP's of Leica fast lenses was that you could have your cake and eat it. You got the ultimate fast lens, albeit with a few minor compromises wide open but when you stopped down, you got all the benefits of a slower lens. Now from various posts, I am not sure that is the case with at least some 21 and 24 Luxes, for CA or purple fringing. My experience is limited to the Zeiss 21 ZM Biogon and the WATE. The 21 Biogon did show quite a lot of purple fringing at f2.8 and a trace at f4. The WATE is remarkably resistant at all three focal lengths, showing very little CA. I agree with you that if the purple fringing is mainly at the f1.4 on the wide Luxes, against bright light, that is liveable with but if it is still a problem at smaller apertures, I would go as far as to say it is a design fault.

 

Wilson

[Advertisement]

Hallo,

Look here: Leica M

[markowich]

Marc,

 

I hear what you say but one of the USP's of Leica fast lenses was that you could have your cake and eat it. You got the ultimate fast lens, albeit with a few minor compromises wide open but when you stopped down, you got all the benefits of a slower lens. Now from various posts, I am not sure that is the case with at least some 21 and 24 Luxes, for CA or purple fringing. My experience is limited to the Zeiss 21 ZM Biogon and the WATE. The 21 Biogon did show quite a lot of purple fringing at f2.8 and a trace at f4. The WATE is remarkably resistant at all three focal lengths, showing very little CA. I agree with you that if the purple fringing is mainly at the f1.4 on the wide Luxes, against bright light, that is liveable with but if it is still a problem at smaller apertures, I would go as far as to say it is a design fault.

 

Wilson

 

i do believe that leica was right and honest when they said that it was a design compromise to accept the CA in order not to make the 21mm and 24mm luxes too large. but a compromise for 5000.- euros????? this is really annoying and shows the bad attitude of leica towards its customers. if a nikon lens for 800.- euros shows that kind of performance i can live with it. but not for 5000.- euros.

btw, check out the CA of the noctilux 0.95. another design compromise in order not to make it too large? for 8000.- euros?

peter