raid
Dad Photographer
I like using a mild wide angle lens, Doug. The Summilux 35/1.4 is a near perfect lens on Leica cameras.
The 24mm Summilux: does it make any sense in the age of good high ISO sensors?
- Thread starter sleepyhead
- Start date
- Latest activity Latest activity:
- Replies 65
- Views 7K
raid
Dad Photographer
Leica lens designers insist that users of Leica lenses should "almost always" use the lenses wide open.
I use apertures 4.0-8.0 quite often.
This has been an interesting thread... and it may lead one to do a little more reading. There doesn't seem a lot of talk about it. Sean Reid had a positive review of the 24 Summilux on an M8 before the M9 came along. KR looked at it with an M9.
Upon its introduction the lens listed for $6000, then less than a year later it was $6500. It's an expensive lens, now $7200. But if one were shopping, one might have seen that RFF supporter PopFlash had a 8/10 used one for $4788, until yesterday. There are two near-new ones at B&H for $5200, so there's a possibility of savings. Enough to easily afford sending the used lens off to DAG, say, for a thorough checkup. Hmmmm! Any temptation? 🙂
Upon its introduction the lens listed for $6000, then less than a year later it was $6500. It's an expensive lens, now $7200. But if one were shopping, one might have seen that RFF supporter PopFlash had a 8/10 used one for $4788, until yesterday. There are two near-new ones at B&H for $5200, so there's a possibility of savings. Enough to easily afford sending the used lens off to DAG, say, for a thorough checkup. Hmmmm! Any temptation? 🙂
sleepyhead
Well-known
THANKS - I already have one, and I am going to keep it for the time being, because I'm currently leaning against the 24mm Summilux, at least until i can try one before buying...
sleepyhead
Well-known
Yes, David, that's right. I have since sold the Voigtlander 21mm f/1.8, BUT it was a GREAT lens in every way, except for size. However, it was a little too wide for me, I was constantly cropping images made with it and throwing away pixels...
(It also didn't say Leica on it, and I'm becoming a snob I fear... ;-)
The 24mm Summilux is my Goldilocks lens, hopefully "just right", and ½ a stop faster than the voigtlander 21/1.8.
Roger Hicks
Veteran
2/3 stop, even.
The 1/3 stops are f/2, 1/1.8, 1/1.6, f/1.4. The 1/2 stops are f/2, f/1.7, f/1.4. In practice, differences of 1/6 stop are rarely perceptible and marked apertures are often slightly divorced from true apertures and even further divorced from T/stops.
Cheers,
R.
tightsqueez
Well-known
The 24 Lux produced some good images. If you dedicate a little effort and get to learn a lens, regardless of lens, images will follow. But, the success rate (for me) was and is a lot higher with the 24 Elmarit and the 28 Summicron. It's easy to discount how a lens handles, or how it "drives." The three lenses I mentioned are very awkward in this regard. It'd be like putting a Bugatti engine in a Porsche Cayman... you'd loose the finesse. Good luck going around a corner! I see the Leica as no different. I have a few images posted on Flickr but will admit that I've been really hacking at what I show there lately. I recently deleted almost half of my images.
https://www.flickr.com/photos/70355737@N00/tags/leicasummilux24mmf14asph/
The color blooming/purple fringing I mentioned is quite common with high contrast lenses in high contrast scenes (or subject as a street light) on an M9/M240. Daylight scenes are easy to fix but night scenes with street lights are a little more difficult, especially with the three lenses I mentioned. Here are a few links I found on the net (I can email you some of my blooming/purple fringing images if you'd like. Just let me know):
http://www.michael-letchford.com/blog/zeiss-sonnar-85mm-f2-zm-lens-review-with-leica-m9
https://www.dpreview.com/forums/thread/3764062
Also, I'd like to retract my 28 Lux statement; I may try one out one day.
willie_901
Veteran
tightsqueez
Interesting links.
I noticed these examples had two things in common: they were from a CCD sensor and the regions with purple artifacts were all over exposed to some degree.
I do not think a lens-based, optical artifact is the primary culprit. I feel overexposure of one or more RGB channels when the shutter was open is the main cause of these purple artifacts
These artifacts are consistent with CCD sensor bloom. Excess electrical charge generated from overexposed sensor sites electrically leaks into adjacent sensor-sites. The adjacent sites no longer have a charge that represents the correct amount of R, G or B light for that location in the Bayer array. The solution is not to clip any RGB channels – especially when using high contrast lenses.
I am curious about how the 24mm Lux behaves with CMOS M bodies. While some early generation CMOS sensors would exhibit bloom in extreme conditions, the M's technology is developed way beyond this stage.
CMOS photo-diode arrays are inherently bloom resistant so bloom artifact levels are significantly lower when the brightest highlight regions are over exposed. The M10 has a higher signal-to-noise ratio which means its analog dynamic range is higher. So, compared to the M9, minimizing over exposed highlights should have less significantly impact on shadow region IQ .
As an aside, purple hue artifacts from lens flare observed during extreme intentional overexposure or extremely bright point-source light sources just outside the frame are not sensor blooming. Purple hue artifacts from flare are not restricted to high contrast edges.
Interesting links.
I noticed these examples had two things in common: they were from a CCD sensor and the regions with purple artifacts were all over exposed to some degree.
I do not think a lens-based, optical artifact is the primary culprit. I feel overexposure of one or more RGB channels when the shutter was open is the main cause of these purple artifacts
These artifacts are consistent with CCD sensor bloom. Excess electrical charge generated from overexposed sensor sites electrically leaks into adjacent sensor-sites. The adjacent sites no longer have a charge that represents the correct amount of R, G or B light for that location in the Bayer array. The solution is not to clip any RGB channels – especially when using high contrast lenses.
I am curious about how the 24mm Lux behaves with CMOS M bodies. While some early generation CMOS sensors would exhibit bloom in extreme conditions, the M's technology is developed way beyond this stage.
CMOS photo-diode arrays are inherently bloom resistant so bloom artifact levels are significantly lower when the brightest highlight regions are over exposed. The M10 has a higher signal-to-noise ratio which means its analog dynamic range is higher. So, compared to the M9, minimizing over exposed highlights should have less significantly impact on shadow region IQ .
As an aside, purple hue artifacts from lens flare observed during extreme intentional overexposure or extremely bright point-source light sources just outside the frame are not sensor blooming. Purple hue artifacts from flare are not restricted to high contrast edges.
tightsqueez
Well-known
Yes, both observations are correct. However, I've seen the same behavior with the 35 Lux FLE and 50 Lux ASPH; both "bloom" as they did on the M9 as on the M240. The 50 APO has none on either as a side note. The M9 had a tendency to exaggerate these issues but the 21/24 Luxes and the 50 Noct 0.95 were about the total package. But as low-light lenses... I thought their slower counterparts were smarter choices. I have no personal experience with the 24 Lux and the newer CMOS bodies. The 35 Lux is my only 1.4 lens now, after recently selling the 50 Lux. The 28/2 and 50/2 are adequate and I really never feel needing more light with the M240. Film though... That's a whole different story.
willie_901
Veteran
Well, the M240 should not bloom (high contrast areas with edge artifacts) unless there is extreme over exposure. Lenses can produce edge artifacts due to lateral chromatic aberration. Generally lateral CA is not highly dependent of over exposure.
I guess a proper test would be to use these lenses with film and see if edge artifacts occur when highlight regions are over exposed.
I guess a proper test would be to use these lenses with film and see if edge artifacts occur when highlight regions are over exposed.
sleepyhead
Well-known
Very interesting discussion guys, although a bit out of my depth of technical knowledge.
Tightsqueez, thanks fo rthe links to your photos.
I think that there is no way for me to know if I will like the 24mm Summilux without trying one myself...
Tightsqueez, thanks fo rthe links to your photos.
I think that there is no way for me to know if I will like the 24mm Summilux without trying one myself...
I agree, very likely. Or you could just rely on comments from owners... 😉 Mine just arrived yesterday from PopFlash! After a walk around the neighborhood I do like it, but I also wish it had a focusing tab...
MCTuomey
Veteran
congrats on the new lens, doug!
tightsqueez
Well-known
Your conclusion is what you sorta knew the whole time. Try one! One thing is for sure, there are PLENTY of them out on the market, so it's in your favor.
Contarama
Well-known
Glass always first!
Get a tripod if the M9 needs it.
Yes fast lens speed is still important because of faster shutter. DOF. Plus digital sensors just perform better at lower ISO ratings (my opinion)...just because one can be cranked up to get the shot doesnt mean that is the final answer.
Get a tripod if the M9 needs it.
Yes fast lens speed is still important because of faster shutter. DOF. Plus digital sensors just perform better at lower ISO ratings (my opinion)...just because one can be cranked up to get the shot doesnt mean that is the final answer.
sleepyhead
Well-known
Congradulations - let us know your impressions once you have formed some... (...and maybe a few photos?)
willie_901
Veteran
At their native or base ISO (not necessarily the lowest possible ISO) practically all digital still cameras deliver the highest possible signal-to-noise ratio and dynamic range. The only restraints are the need to freeze camera and, or subject motion and maximizing DOF. The tripod you mentioned usually addresses everything but subject motion.
Alway use the lowest possible ISO down to the camera's native ISO.
As mentioned, I miss the focusing tab... so I put on a wide/short rubber band, a refugee from the produce section, perfectly fits the focusing ring with adequate tension... importantly it feels quite distinct from other parts of the lens barrel. White, looks very "professional." 😀
This now is the second Leica lens adorned with an aftermarket rubber band! The 50 Summicron's collapsing lens-hood is kept extended by a small green rubber band just behind it...
Also, I'd been using the Voigtlander 25mm finder that came with the SnapShot Skopar, no framelines... that and its barrel distortion made it hard to get lined up and framed during a walk around the college last night. So I changed to the vastly superior Zeiss 25/28 finder for today's walk.
This now is the second Leica lens adorned with an aftermarket rubber band! The 50 Summicron's collapsing lens-hood is kept extended by a small green rubber band just behind it...
Also, I'd been using the Voigtlander 25mm finder that came with the SnapShot Skopar, no framelines... that and its barrel distortion made it hard to get lined up and framed during a walk around the college last night. So I changed to the vastly superior Zeiss 25/28 finder for today's walk.
Sleepyhead, it's now 4 months later; how goes it? 🙂
I have just now started to process my first 24 Summilux photos on an M240, I was that far behind after the Cuba trip using a newer 28 Summicron.
Even wide open, the lens is admirably sharp, less so in the corners. But it seems the plane of focus is bowl-shaped (or perhaps more complex) so the wide-open sharpness in outer zones depends on the nature of the subject. I'm still getting to know the lens...
One thing that can be obvious in some shots and not in others is the odd optical distortion... barrel distortion in the center area transitioning to pincushion in the other areas. It results in straight lines along the edges showing as a wave, thus "mustache" distortion. Hard to correct manually but easy in Lightroom using the lens profile correction.
Of course perspective distortion can be an issue too as with any lens, stronger in wide angles. I often need to rotate and correct some keystoning, but that's not related to the optical character of the lens.
Here are a couple of shots with the 24mm Summilux on an M240... the obligatory brick wall test! First the uncorrected mustache distortion, next a similar shot with that distortion corrected in LR. Bit of (uncorrected) perspective distortion in both as well. At f//2.8 for the bricks, wide open for the bike...
I have just now started to process my first 24 Summilux photos on an M240, I was that far behind after the Cuba trip using a newer 28 Summicron.
Even wide open, the lens is admirably sharp, less so in the corners. But it seems the plane of focus is bowl-shaped (or perhaps more complex) so the wide-open sharpness in outer zones depends on the nature of the subject. I'm still getting to know the lens...
One thing that can be obvious in some shots and not in others is the odd optical distortion... barrel distortion in the center area transitioning to pincushion in the other areas. It results in straight lines along the edges showing as a wave, thus "mustache" distortion. Hard to correct manually but easy in Lightroom using the lens profile correction.
Of course perspective distortion can be an issue too as with any lens, stronger in wide angles. I often need to rotate and correct some keystoning, but that's not related to the optical character of the lens.
Here are a couple of shots with the 24mm Summilux on an M240... the obligatory brick wall test! First the uncorrected mustache distortion, next a similar shot with that distortion corrected in LR. Bit of (uncorrected) perspective distortion in both as well. At f//2.8 for the bricks, wide open for the bike...
sleepyhead
Well-known
Well Doug thanks for the update. Whoa, that distortion is strange! (But, I will use light room if I ever get the lens, so I guess it'll be alright.
I am still saving up cash, I have about ⅔ of what I would need for a secondhand 24mm Summilux.
In the meantime, I am also considering the 28mm Summilux instead. It is a little smaller and seems better behaved. But even more expensive and harder to find. Also, the 24mm will pair better with my 35mm Summilux.
Will continue saving and considering...
Thanks for posting.
I am still saving up cash, I have about ⅔ of what I would need for a secondhand 24mm Summilux.
In the meantime, I am also considering the 28mm Summilux instead. It is a little smaller and seems better behaved. But even more expensive and harder to find. Also, the 24mm will pair better with my 35mm Summilux.
Will continue saving and considering...
Thanks for posting.
