Just tested the 50 Lux and Cron 50mm my M9...

[DFV]

Advertisement (gone after registration)

The Summilux 50 asph should be eye-wateringly sharp at 5.6. You results are not typical by any means.Somebody get out the MTF curves...

 

Yes, this is what I thought (and still think). I am certain that mine is not right since no mater what I do I just can't get proper results. With the Summicron I get incredible results.

 

This makes me change my mind about the M9. I was not impressed at first yet this is because I used it mainly with this "bad" Summilux. With the Summicron (and my other lenses) it is a frighteningly good camera. Still think there is a display missing and would have loved a safire back yet the sensor is right now clearly better than the M8. With my Summilux I was not able to see that.

 

I am sending it back to Leica on Monday...

[jaapv]

The photographs are a bit embarrasing, but just an example and two crops of the Summilux asph: 18% of smart sharpen to compensate for JPG conversion.

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!

[DFV]

The photograph is a bit embarrasing, but just an example and two crops of the Summilux asph: 18% of smart sharpen to compensate for JPG conversion.

 

Yes, this is what I am supposed to get with mine but am not getting. Really annoying since I am leaving for Paris tomorrow and I wanted to take it with me for some low-light shots... Although I am glad I have "re-discovered" the Summicron and right now it looks better than ever.

[Logic108]

I'm interested in this thread because I own a 50 'lux asph. The two images you posted have a slightly different white balance. I loaded both images into Lightroom and and I use a calibrated ACD. I adjusted the white balance (even though they are .jpegs) so that I could get a match.

Yes I can see that you are right about the 'lux being slightly softer than the 'cron.

Well I thought that I would run the 'lux image through my pre-sharpening workflow and I found the resultant image very pleasing having more detail than the 'cron.

Obviously you are not happy with the 'lux but if you get no joy from Solms then you might like to know that the 'lux images take sharpening very well.

Personally I find that I sharpen most images I take before outputting to the web or to print.

[Guest trond]

Have you tested these with your M9?

 

By the way, many thanks for your help.

 

Yes, I have identical shots taken at the same time, with the camera tripod mounted in the same position with all four lenses.

 

Images are taken at every full f-stop on all four lenses in the same position.

 

I notice also that the JPEG files of the Summicron are approximately 1 MB larger than the corresponding Summilux files for every apperture.

 

I really do not understand why, since this should mean that there is more information in the Summicron file, yet the Summilux images look more crisp and contrasty.

 

However, my direct comparison above, of both lenses wide open at 400%, shows that both lenses go beyond the resolution of the M9 sensor.

 

Best regards

 

Trond

[DFV]

Take a look at this... Do you see the three cables fastening the antenna? On the Summicron (bottom) it seems that the resolution is much higher that the Summilux (top).

 

 

This is a 1:1 crop with no sharpening or tampering with the pictures other than converting to jpg. At first I thought it could be a software problem yet I deactivated the Lens Recognition and yielded the same result.

[Jerry_R]

Advertisement (gone after registration)

As I remember, I have read, that SUMMILUX lenses are only better when wide opened. They are designed like this. This is their goal.

 

But they start EARLIER diffraction than SUMMICRONS and SUMMARITS. They are worse when closed down. Isn't this thread showing that? Based on what I read, at f/5.6 or f/8 - even SUMMARITS should excell.

 

Example from Reid Reviews

- @ f/5.6 SUMMARIT 35mm is slightly sharper than SUMMICRON 35mm at center

- @ f/5.6 SUMMARIT 35mm is slightly sharper than SUMMICRON 35mm at corners

- @ f/2 SUMMICRON 50mm is slightly sharper than SUMMILUX 50mm at corners

- @ f/8 SUMMICRON 50mm is not loosing so much due to difraction as SUMMILUX 50mm

 

It is repeated many times - that very often - slower lenses provide higher resolution when stopped down.

 

PS: I just checked what Tao of Leica says. It alse reminds diffraction. And states, that close-up performance is better with Summicron.

But general opinion is that Summilux is better than Summicron.

Hmm...

[gwelland]

I found the results to be disturbingly different. The Summicron is incredibly sharp edge to edge with an incredible amount of contrast. The superiority with the Summilux is such that I am beginning to wonder if there is something wrong with it.

 

If any can take a look at both shots please comment. I am linking to jpeg and both are full resolution about 10 meg... Careful. The motif is horrendous yet it serves me very well to se contrast and sharpness. Both shots where taken at f5,6 and 1/3000 sec. I have only converted the pictures from DNG to JPG in Lightroom. No sharpening and no retouching or editing.

 

Summilux...

 

Summicron...

 

Please let me know what you think.

 

Assuming everything is even, you can definitely see in your examples that the Summicron is rendering a sharper image. It is most evident in the way that the 'cron image actually invokes moire in the details of the circular air-con fan yet the 'lux smoothes out these high frequency transitions. Ditto with the views of the trellis fences. I can certainly see sharper detail in the cron shot than the 'lux shot - be that due to increased micro-contrast or whatever, it definitely looks 'sharper' to me. You can read the manufacturer's label on the air-con unit in the cron shot but it's softer in the 'lux example.

 

I did a similar exercise with my 50 'cron I just got back from coding and compared it with my 50 'lux ASPH the other day. I wouldn't class my exercise as scientific, although like you it was based on shooting from a tripod with identical (as far as I could manage) shots of a rocky stream at f/2 & f/5.6. I was also basically shocked at the apparent difference between the 'cron images and my 'lux - putting the 'lux images down to critical differences in focus point etc. The 'cron was bitingly sharp, like yours - noticeably in fact. My buddy who'd bought it from me got a super-sharp lens!

[bybrett]

The sweet spot of a lens is related to its maximum aperture. It wouldn't be that amazing if "a Summicron" out performed "a Noctilux" at f5.6. I would expect a 24mm Elmar to be "sharper" than the Summilux at f3.8? Maybe pitch a 50 Summicron and a Summilux against each other at f2... have a look at f1.4 and then sell a lens!

[Guest trond]

Take a look at this... Do you see the three cables fastening the antenna? On the Summicron (bottom) it seems that the resolution is much higher that the Summilux (top).

 

Dear DFV,

 

This seems a bit odd; the jagged wires on the antenna is due to spatial aliasing between the projected image by the lens, and the pixel grid of the sensor. This should also be visible on the Summilux if the crop where identically aligned.

 

I suggest you re-shoot with the camera tripod mounted so the crop will be identical.

 

I would say that the resolution of the Summilux is actually higher that the Summicron:

 

The wires on the rightmost antenna is clearly resolved on the top image. While, on the Summicron image, these wires are barely visible.

 

In additinon, the single thin antenna rod on the far right of the image is visible on the Summilux image, but not so on the Summicron image.

 

This can result from tiny differences in focus distance, but also because the Summicron image is tilted slightly upwards. Again, a tripod will avoid this.

 

I agree that the contrast is higher on the Summicron, but the resolution of fine details is just a tad higher on the Summilux image. Sharpening will resolve this.

 

A super critical test like this it is important to remove as many variable factors as possible. Use a tripod, and make several shots vit a small amount of focus braceting to make up for slight focus calibration of the two lenses.

 

Best regards

 

Trond

[okram]

I think these two lenses how different drawing, an that for some kind of motive, the Cron's drawing suits better, and for some Lux's drawing.

But the difference between the two is so small at 5.6, that it will be visible only in very large print (if even so ).

 

In my opinion, the best way to compare is too shoot two photpgraphs of a person, full length horizontally, and look at the face. For corner res. put the head on both in the cornes while framing, and for center to center but sustain the full length distance. I think thereis nothing harder for a non antialiased sensor- it can resolve better straight lines.

 

I woud also use C1 pro or Raw Developer, since LR shows slightly less resolution (LR3)- not to mention LR 2.

 

 

M

[okram]

Trond, these could be two wires, rolled over each other...but also aliasing, only the op knows that.

 

I dont see any difference in resolution between yours and op's Lux, compared to Cron.

I just think the motif is different, an it appears yours suits better to Lux.

The differences are very small.

[okram]

DFV, take a look at the right, there is a red antenna, and on its left side (towards center) there is a very thin black antenna- it is bettter resolved with Lux.

There is also one very small one down the step on the right, that is invisible with Cron and there with Lux.

 

I just think that these two lenses draw differently, and that for some motifs and M9 sensor, 50/2 suits better, and we are talking f/4-5-5.6, more open than this and Cron is much weaker one. Btw, lux , in my experience, will show some focus shift at 4-5.6, and this an be part of its performance...but I use it solely at f/2 or 2.8.

 

Try shooting people or forest (jagged lines- not straight ones)...

[Guest trond]

Trond, these could be two wires, rolled over each other...but also aliasing, only the op knows that.

 

I don't see any difference in resolution between yours and op's Lux, compared to Cron.

I just think the motif is different, an it appears yours suits better to Lux.

The differences are very small.

 

In my case the resolution of both lenses are better that the M9 sensor can resolve.

 

The landscape image with buildings and threes contain both straight and bent lines in all directions, with details well beyond the resolution capability of the M9 sensor.

 

Yet, with both lenses, even wide open, the M9 sensor is the limiting factor, not the Summicron, and certainly not the Summilux!

 

I see (at least my copy) the Summilux as an absolute fantastic lens, the best I have ever seen with a 50mm lens.

 

Only stopped down to f5.6 my Summicron is as good as the Summilux.

 

At f1.4, the Summilux is better than the Summicron at f4.0.

 

Again this is in my case, with a different set of lenses it can well be the opposite.

 

I have seen this kind of small differences between two different samples of the same lens.

 

According to the Leica published MTFs, the Summilux should be slightly better than the Summicron at f 5.6.

 

The Summicron has a more "wavey" MTF pattern. This can favorize sharpness in parts of the image (corners). However, the Summilux should have a clear lead in the center of the image.

 

Best regards

 

Trond

[250swb]

DFV, take a look at the right, there is a red antenna, and on its left side (towards center) there is a very thin black antenna- it is bettter resolved with Lux.

There is also one very small one down the step on the right, that is invisible with Cron and there with Lux.

 

 

Well I can see that aerial on the Summicron photo. But the problem you introduce looking at aerials and wires for justification is that a gust of wind at the time of exposure sort of spoils the test, as would a slight variance in exposure where contrast of the aerial is changed against the sky.

 

The only way to get to the bottom of this is an indoor test with a stable lighting setup, but the Summicron is a very sharp lens, and its a strange thing that people actually seem disapointed by this fact.

 

Steve

[Guest trond]

Dear All,

 

In the link below, the all famous Leica guru, Erwin Puts, describes in great detail the virtues and capabilities of the Summilux 50mm ASPH.

 

The writeup is very well written and sheds a great deal of light onto the above discussion.

 

Highly recommended reading!

 

Erwin Puts on Summilix 50mm ASPH.

 

Best regards

 

Trond

[Overgaard]

Just an interesting comparison here as well which I think illustrates the traditional qualities of the Summilux. Forget the sharpness, look how it handles light in all from detail, tonality, dynamic range, etc:

 

The deifinitoon of shadow detail is to me what really make the Summilux ASPH stand out.

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!

[D&A]

Part of what I am seeing in images posted is contrast differences between Lux Asph and current Summicron, with the latter having slightly higher contrast which also gives the appearance of greater apparent sharpness as well as blocking up shadow detail in certain instances. This increase in contrast gives an increase in apparent edge definition, which in the case of the pair of images posted, seems like the Summicron is out resolving the Lux. A quick leveling of these contrast differences may in some instances "level" out some (not all) of the differences that are being observed. Try lowering contrast in Summmicron image and compare to the Native Lux....and conversly raise contrast in Lux image and compare to native Cron.

 

Dave (D&A)

[cbretteville]

Interesting, the rep of the modern ASPH lenses, like the 50 Summilux, is that they're (almost too) sharp and very contrasty. The older double gauss designs, like the 50 Summicron, have a rep as being lower in contrast and by some considered as better lenses for digital capture for that very reason.

Carl

[Advertisement]

Hallo,

Look here: Leica M

[gwelland]

Interesting, the rep of the modern ASPH lenses, like the 50 Summilux, is that they're (almost too) sharp and very contrasty. The older double gauss designs, like the 50 Summicron, have a rep as being lower in contrast and by some considered as better lenses for digital capture for that very reason.

Carl

 

Generally I would absolutely agree with you on this. (and the Puts examples too).

 

It does seem though that there's a lot of 'Lux ASPH apologists for the particular comparison here because, assuming everything is equal & the same, the two images posted by the OP clearly show the Summicron appearing to be sharper with higher apparent contrast to the 'lux.

 

I think we all agree that the 'lux should perform better than is shown here, and as evidenced by some of the other posted examples. It could just be a sample variation or a problem with the OP's 'lux. I know that I had a similar issue with a 28 'cron that I had to send off to Leica for calibration where I only noticed the softness of my lens when compared to another example side by side.

 

More testing and sampling to ensure that what's being seen is consistent is definitely a good starting point. Having just two images does introduce the possibility of the shooting conditions & focusing to not be the same between the two lens shots.