Problems getting sharp images by 60MP

[lct]

Advertisement (gone after registration)

Same more or less at 1/f (1/60s) and 18mp upsampled to 24mp. Not sure camera shake is the culprit there since the M240 has more resolution at 24mp than the M11 at 18mp. Anyway, for old M11 users with steady hands like yours truly, seems like the only way to avoid motion blur at 1/f is to shoot at 18mp or to use a tripod pending the arrival of IBIS on the M12 hopefully. At 36mp, 1/2f should be good enough for me in most cases and i don't know if i will ever try 60mp as i feel ridiculous to have to shoot at 1/3f let alone 1/4f on still subject matters but it's just me.
M11: https://photos.smugmug.com/Diverse/n-QFBj4/Leica-M11-Leica-5028-v2/i-rHjxRG5/0/1671577f/X4/L1000541_mag-X4.jpg
M240: https://photos.smugmug.com/Diverse/n-QFBj4/Leica-M11-Leica-5028-v2/i-hB5THbk/0/d6c91bfd/X4/M2401711-X4.jpg

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!

 

Edited May 4, 2022 by lct

[Guest]

1 hour ago, lct said:

Same more or less at 1/f (1/60s) and 18mp upsampled to 24mp. Not sure camera shake is the culprit there since the M240 has more resolution at 24mp than the M11 at 18mp. Anyway, for old M11 users with steady hands like yours truly, seems like the only way to avoid motion blur at 1/f is to shoot at 18mp or to use a tripod pending the arrival of IBIS on the M12 hopefully. At 36mp, 1/2f should be good enough for me in most cases and i don't know if i will ever try 60mp as i feel ridiculous to have to shoot at 1/3f let alone 1/4f on still subject matters but it's just me.
M11: https://photos.smugmug.com/Diverse/n-QFBj4/Leica-M11-Leica-5028-v2/i-rHjxRG5/0/1671577f/X4/L1000541_mag-X4.jpg
M240: https://photos.smugmug.com/Diverse/n-QFBj4/Leica-M11-Leica-5028-v2/i-hB5THbk/0/d6c91bfd/X4/M2401711-X4.jpg

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!

 

Have you tried Topaz Sharpen AI or Photoshop Smart Sharpen? When used moderately I have found these useful in some cases to bring otherwise nice pictures back into “acceptable for non critical work” shape …

I know that people may start throwing Vintage Leica lenses or other objects at me for suggesting this, but with e.g. Noctilux wide open in particular I have found using a Monopod with dedicated Monopod head (single axis of movement) useful to not just reduce shake by reliably 2 stops but also stabilizing accurate framing and critical focus (esp. at f/0.95 …).

Edited May 4, 2022 by Guest

[lct]

5 hours ago, mzbe said:

Have you tried Topaz Sharpen AI or Photoshop Smart Sharpen? When used moderately I have found these useful in some cases to bring otherwise nice pictures back into “acceptable for non critical work” shape …

I know that people may start throwing Vintage Leica lenses or other objects at me for suggesting this, but with e.g. Noctilux wide open in particular I have found using a Monopod with dedicated Monopod head (single axis of movement) useful to not just reduce shake by reliably 2 stops but also stabilizing accurate framing and critical focus (esp. at f/0.95 …).

I dislike both NR softwares and tripods i'm afraid so i never tried Topaz sorry. Their first page showing a bird resembling a plastic toy deterred me from going any further but i know Topaz has a good reputation on the LUF .

[Guest]

4 hours ago, lct said:

I dislike both NR softwares and tripods i'm afraid so i never tried Topaz sorry. Their first page showing a bird resembling a plastic toy deterred me from going any further but i know Topaz has a good reputation on the LUF .

I directionally agree with your sentiment, my proposed workarounds are slightly annoying ... At least the Topaz thing doesn't distract you while shooting, and a monopod or tripod (for me it's a completely different experience between those two) have some upside beyond just fixing your shutter speed related sharpness concerns (and the downside of being clunky, and slowing you down while taking pictures). YMMV.

[lct]

2 hours ago, mzbe said:

I directionally agree with your sentiment, my proposed workarounds are slightly annoying ... At least the Topaz thing doesn't distract you while shooting, and a monopod or tripod (for me it's a completely different experience between those two) have some upside beyond just fixing your shutter speed related sharpness concerns (and the downside of being clunky, and slowing you down while taking pictures). YMMV.

I would have said the same in my film days i guess but i'm also a Sony user and IBIS has become as normal as a tripod or a monopod for me. YMMV too of course .

[hdmesa]

Cross-posted to FM Forums > Leica & Alternative Gear > Official: Leica M11

This has been done before by others more thoroughly and correctly, but I wanted to see how the 60/36/18 resolution DNG files compared for how I work with my images.

M11 -|- CV 50 APO -|- Manual mode set to ISO 64, f/5.6, and 1/250/sec -|- infinity focus -|- crops are from the bottom third of a horizontal image

Lightroom M11 profile, identical adjustments to all three for exposure and sharpness. Sharpness set to what looked good to me at 100%, which was 114 (40 is default), radius set to 0.8 (1.0 is default). The default sharpening on the M11 in LR is very conservative. Files exported to PSD 16-bit and the two smaller files given just a squeak more sharpening via Camera Raw filter.

Why would I add a bit of sharpening to the 36/18 files? Because when reducing the 60mp files Photoshop to smaller sizes, I used Bicubic Sharper mode (as likely do other testers since it's the default for reduction), which in effect sharpens for detail during reduction. Reducing the 60mp files with straight Bicubic (so as to preserve gentle gradients) might be more fair to the binned files and produce a less sharp reduction, but I didn't test that.

Conclusions:

If pre-sharpened and sharpened slightly after export, the 36mp and 18mp binned DNGs are nearly identical to the reduced 60mp DNG when viewed at 100%. At 300% magnification, the 60mp files appear to resolve slightly more detail and show slightly less noise. There seem to be no IQ benefits to binning, only space savings as other testers have shown. However, there is also very little penalty for binning if the files are thoughtfully edited in post. I hope that Leica further refines the binned DNGs in a firmware update to deliver a benefit other than space savings.

On this forum you must click on the images to see the sharpness of the comparisons – the inline images below will look soft for everything. If you have trouble with that, try viewing them here.

///

///

100% magnification –> 60 / 36 / 18:

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!

///

///

100% magnification –> left: 60 reduced to 36 / right: 36 native:

///

///

300% magnification –> left: 60 reduced to 36 / right: 36 native

Edited May 5, 2022 by hdmesa

[LocalHero1953]

Advertisement (gone after registration)

4 hours ago, hdmesa said:

Cross-posted to FM Forums > Leica & Alternative Gear > Official: Leica M11

This has been done before by others more thoroughly and correctly, but I wanted to see how the 60/36/18 resolution DNG files compared for how I work with my images.

M11 -|- CV 50 APO -|- Manual mode set to ISO 64, f/5.6, and 1/250/sec -|- infinity focus -|- crops are from the bottom third of a horizontal image

Lightroom M11 profile, identical adjustments to all three for exposure and sharpness. Sharpness set to what looked good to me at 100%, which was 114 (40 is default), radius set to 0.8 (1.0 is default). The default sharpening on the M11 in LR is very conservative. Files exported to PSD 16-bit and the two smaller files given just a squeak more sharpening via Camera Raw filter.

Why would I add a bit of sharpening to the 36/18 files? Because when reducing the 60mp files Photoshop to smaller sizes, I used Bicubic Sharper mode (as likely do other testers since it's the default for reduction), which in effect sharpens for detail during reduction. Reducing the 60mp files with straight Bicubic (so as to preserve gentle gradients) might be more fair to the binned files and produce a less sharp reduction, but I didn't test that.

Conclusions:

If pre-sharpened and sharpened slightly after export, the 36mp and 18mp binned DNGs are nearly identical to the reduced 60mp DNG when viewed at 100%. At 300% magnification, the 60mp files appear to resolve slightly more detail and show slightly less noise. There seem to be no IQ benefits to binning, only space savings as other testers have shown. However, there is also very little penalty for binning if the files are thoughtfully edited in post. I hope that Leica further refines the binned DNGs in a firmware update to deliver a benefit other than space savings.

On this forum you must click on the images to see the sharpness of the comparisons – the inline images below will look soft for everything. If you have trouble with that, try viewing them here.

///

///

100% magnification –> 60 / 36 / 18:

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!

///

///

100% magnification –> left: 60 reduced to 36 / right: 36 native:

///

///

300% magnification –> left: 60 reduced to 36 / right: 36 native

You might want to post a link to this post on the thread by @adan where he has done a similar test looking at low light noise at the different M11 resolutions - both very useful records.

[lct]

6 hours ago, hdmesa said:

At 300% magnification, the 60mp files appear to resolve slightly more detail and show slightly less noise.

Would you mind to explain me how it works? I always thought that magnifying at more than 100% can only add artefacts created by the computer or the screen but i am no techie at all. Just curious.

[hdmesa]

4 hours ago, lct said:

Would you mind to explain me how it works? I always thought that magnifying at more than 100% can only add artefacts created by the computer or the screen but i am no techie at all. Just curious.

The method of comparing at high magnification is done almost universally by testers (Jim Kasson and Fred Miranda are just two examples of testers who do this) mainly because on the computer screen, our eyes may not be able to discern subtle differences at 100%, especially on very dense, high-resolution screens (which sometimes need a 200% zoom just to see small details in a shot anyway).

I'm not sure on exactly what happens past 100% from a technical standpoint, but it affects all the test photos equally if they are the same pixel size to begin with. Zooming past 100% can reveal minor differences between shots of identical subjects even if the images look blurry at that magnification. High magnification is useful to me in everyday shooting because it can help me decide between a series of identically-framed shots to find the sharpest. At high zoom levels with two images side-by-side, it's much easier to spot things like motion blur, misfocus, etc. that you wouldn't catch otherwise. So while I will never print that big, it still makes sense to choose the image that has the most detail.

So the point of showing the 300% comparison in this case (60mp downsized to 36mp versus 36mp binned) is to show how little difference there is between the files. Because you have heard and will continue to hear that shooting at 60mp and downsizing is better, but I think it's important to know just how little difference there is in actual shooting if you edit your files well with regard to sharpening.

One conclusion for me after doing this –> it's not really fair for testers to reduce 60mp images in Photoshop using Bicubic-sharper (which prioritizes sharpening during reduction) and compare them with native lower resolution binned files that have not been sharpened. When they test that way, the 36mp and 18mp files appear to suffer much more than they should, IMO.

 

Edited May 5, 2022 by hdmesa

[hdmesa]

8 hours ago, LocalHero1953 said:

You might want to post a link to this post on the thread by @adan where he has done a similar test looking at low light noise at the different M11 resolutions - both very useful records.

Thanks. I feel like I would need to have compared the results with the M10/M10-R to be aligned with that thread, but I don't have my M10-P and R any longer. At some point I may compare downsized M11 60mp files against my SL2-S in very low light, and that may better align with the other thread.

[Dennis]

Don't publish elsewhere the M240 vs M11 motion blur tests. Inexperienced photographer would then buy the M240 thinking is better  

[hdmesa]

4 hours ago, Dennis said:

Don't publish elsewhere the M240 vs M11 motion blur tests. Inexperienced photographer would then buy the M240 thinking is better  

Lol. The M are all awkward for the average photographer used to button-infested mirrorless cameras 

I still prefer my SL2-S files to the binned M11 if comparing unedited DNG to DNG. The native lower res Leica sensors really sparkle, which is great so long as there are no aliasing artifacts introduced with highly detailed subjects of certain frequencies (tin roofs, fabric, distant fences, etc.).

 

[fsprow]

Just finished making some large (30” x 50”) enlargements from the M11.  50mm lens at 1/125th.  Beautiful and the 60 MP resolution really shows it’s stuff in fine detail.   My wife pointed out that these photos have the same “3 dimensional look” as those from my Hasselblad H6D-100c.  Extremely happy with this camera — the first “full frame” (or 35mm) one I’ve had that competes with larger sensor cameras. I have found no difference in needed shutter speeds from those I’ve used in my old M3 on up.

[lct]

On 5/5/2022 at 7:32 PM, Dennis said:

Don't publish elsewhere the M240 vs M11 motion blur tests. Inexperienced photographer would then buy the M240 thinking is better  

Hehe yes but the M240 cannot do that (M11, 60mp, 90/4 macro, 1/250s, FF & 100% crop). I was testing auto iso on the M11 wondering if shutter speeds are higher in 60mp mode. They are not apparently but the auto speed settles at 1/250s for the 90/4 macro i.e. a bit lower than 1/3f. Fast enough for 36mp i guess but i expected more softness at 60mp so there is some hope for old timers like me.  

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!

 

 

Edited May 6, 2022 by lct

[Tailwagger]

3 hours ago, lct said:

Fast enough for 36mp i guess but i expected more softness at 60mp so there is some hope for old timers like me.  

Actual experience does have a tendency to be more accurate than speculation, though perhaps not so much on these pages.  😃 

[lct]

Fair enough if one can live sans 1/f. I did the same test as above in auto iso mode with an Elmar-M 50/2.8 instead of 90/4 macro. Auto shutter speed was 1/100s i.e. 1/2f. Results are mostly sharp but similar shots are softer with the same settings, sometimes within the same second. See 100% crops below. Only relevant for pixel peeping i guess but i want to know if 1/3f let alone 1/4f are a prerequisite on this camera. 

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!

 

 

[lct]

End of my boring tests on the M11. Apo-Telyt 135/3.4, f/11, auto 1/320s (in between 1/2f and 1/3f). Results are mostly sharp but some shots are softer with the same settings as with 90mm and 50mm lenses tested previously. See 100% crops below. Reason why the M11 appears like a 1/2f or 1/3f camera to me. Allows me to set 1/2f as maximum setting in auto iso mode but i choose 1/3f manually if necessary. Could be worse but the M11 remains disappointing at 1/f. It is the price to pay for the lack of IBIS i guess but i don't understand why results don't improve at 36mp or 18mp.

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!

 

 

Edited May 10, 2022 by lct

[hdmesa]

32 minutes ago, lct said:

End of my boring tests on the M11. Apo-Telyt 135/3.4, f/11, auto 1/320s (in between 1/2f and 1/3f). Results are mostly sharp but some shots are softer with the same settings as with 90mm and 50mm lenses tested previously. See 100% crops below. Reason why the M11 appears like a 1/2f or 1/3f camera to me. Allows me to set 1/2f as maximum setting in auto iso mode but i choose 1/3f manually if necessary. Could be worst but the M11 remains disappointing at 1/f. It is the price to pay for the lack of IBIS i guess but i don't understand why results don't improve at 36mp or 18mp.

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!

 

 

Might sound crazy, but I was having to use 1/4f on the M11 until I added a soft release. I now can easily shoot at 1/1f out to 75mm (with 1/60th sec. as the bottom limit just to be safe).

For 135mm on any non-IBIS system at or over 50mp I've had (5DsR, GFX 50S/R) –> I call that a 1/4f lens not a 1/4f camera.

I probably missed it in the thread somewhere, but did you try the electronic shutter on the M11 to rule out shutter shock?

Edited May 10, 2022 by hdmesa

[lct]

2 minutes ago, hdmesa said:

[...] did you try the electronic shutter on the M11 to rule out shutter shock?

Yes i did at 1/f but i did not notice a significant improvement. Curiously enough the mechanical shutter sounds so quiet and well damped that i prefer using it contrary to my other cameras. 

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[KeyofG]

On 2/16/2022 at 7:37 PM, TrickyMrT said:

could you solve the issue by setup to 18MP ? 

because i have same issues to 18MP as well...... 

 

just wondering if anyone knows a usable solutions for handheld shots by lower shutterspeed?

Depends on the focal length as you already know. 

proper holding is one thing. I shoot 50mm. Never below 1/125th when people move. If I’m moving also 1/500th. Default is 1/250th for me, when I’m standing and people are moving. If people aren’t moving much or not at all then 1/125th. Freezing birds 1/1000  

I guess a building I can shoot at 1/80th or a little less maybe but, I wouldn’t do it. It’ll take a few tries. With the M10-R there’s not a lot of room to hide. M11 should be the same. 

Edited May 12, 2022 by KeyofG