Leica SL2 Firmware 2.0: 187 MP Multishot Mode

[Mr.Q]

Advertisement (gone after registration)

So just the convenience of processing in-camera? I find stacking + stitching to be effortless in post. Quite effective for 200MP or even 500MP+ images too.

Or am I missing something?

[Joe Colson]

No, you're not missing anything. Multi-shot simply allows one shutter press to generate a 187MP RAW file plus a standard resolution file, with no post-processing. Some shooters might prefer that process (immediate gratification). What post-processing technique do you use that makes stacking + stitching "effortless"? That's a question asked solely for my education and improvement, not facetiously.

Joe

[Mr.Q]

Ok thanks. There's really not much technique involved but I use Photoshop to merge layers to stack + blend, and for stitching I use Lightroom for a simple 4-6 image panorama or PTGUI if a bit more work is involved. 

[Joe Colson]

Thanks for the explanation. I have done that too and will probably keep that in my tool kit. The multi-shot mode is just another tool. It doesn't replace the ones we have. When I added a battery-powered circular saw to my workshop, it didn't replace my hand saw(s). Although my wife asks, "Why do you need another saw?" To me, it's like manual focus vs. AF. One doesn't replace the other.

Joe

[Jon Warwick]

2 hours ago, Mr.Q said:

So just the convenience of processing in-camera? I find stacking + stitching to be effortless in post. Quite effective for 200MP or even 500MP+ images too.

Or am I missing something?

I think these are all different “tools” that we are discussing. But could myself be missing something!

1) multi shot (also known as high resolution mode) being minuscule variations in sensor position such that it quasi negates the downsides of a Bayer filter in terms of blur / false colours / moire, which can lead to higher perceived visibility of fine detail. A 50mm lens with depth of field at f 5.6 will have that same visual impact in both normal or multi shot mode.

2) stacking I take to mean “focus” stacking, where you create 1 image from multiple layers of different images that are taken with variations in focusing points. A 50mm lens would still frame like a 50mm here, but the perceived depth of focus goes from f5.6 to say f16.

3) stitching is adding multiple frames, sometimes in a panorama or via shifting in say a tech camera like an Alpa / Cambo. Here your 50mm lens’ angle of view won’t look like that in the final stitched image, likely being wider at (say) 28-35mm or however many stitches one does. By adding more pixels to a “wider” image, you are increasing resolution in effect. But it won’t be beneficial for avoiding moire or false colours as discussed in (1) above.

So in summary - I see the above as all being quite different things for different applications!!

Edited July 4, 2020 by Jon Warwick

[SrMi]

56 minutes ago, Mr.Q said:

Ok thanks. There's really not much technique involved but I use Photoshop to merge layers to stack + blend, and for stitching I use Lightroom for a simple 4-6 image panorama or PTGUI if a bit more work is involved. 

Multishot is different from merging layers to stack and blend, and stitching. 

Multishot increases dynamic range, improves colors, and eliminates aliasing. Both merging layers (not focus stacking) and Multishot reduce noise.

https://en.wikipedia.org/wiki/Pixel_shift

Edited July 4, 2020 by SrMi

[Joe Colson]

Advertisement (gone after registration)

1 hour ago, SrMi said:

Multishot is different from merging layers to stack and blend, and stitching. 

Multishot increases dynamic range, improves colors, and eliminates aliasing. Both merging layers (not focus stacking) and Multishot reduce noise.

https://en.wikipedia.org/wiki/Pixel_shift

 

1 hour ago, Jon Warwick said:

I think these are all different “tools” that we are discussing. But could myself be missing something!

1) multi shot (also known as high resolution mode) being minuscule variations in sensor position such that it quasi negates the downsides of a Bayer filter in terms of blur / false colours / moire, which can lead to higher perceived visibility of fine detail. A 50mm lens with depth of field at f 5.6 will have that same visual impact in both normal or multi shot mode.

2) stacking I take to mean “focus” stacking, where you create 1 image from multiple layers of different images that are taken with variations in focusing points. A 50mm lens would still frame like a 50mm here, but the perceived depth of focus goes from f5.6 to say f16.

3) stitching is adding multiple frames, sometimes in a panorama or via shifting in say a tech camera like an Alpa / Cambo. Here your 50mm lens’ angle of view won’t look like that in the final stitched image, likely being wider at (say) 28-35mm or however many stitches one does. By adding more pixels to a “wider” image, you are increasing resolution in effect. But it won’t be beneficial for avoiding moire or false colours as discussed in (1) above.

So in summary - I see the above as all being quite different things for different applications!!

By "stack + blend" Mr.Q was not referring to focus stacking. That's an entirely different tool. Stacking and blending refers to the process of taking multiple exposures of the same scene, often using eShutter mode, and stacking and blending them in Photoshop (or Capture One) to achieve a final image with higher DR, less noise, and [perhaps] better color. Phase One's IQ4-150 has achieved that in-camera with "frame averaging". Of course, multi-shot is not the same thing, nor is it accomplished in the same way. It's just another tool.

Joe

[Jon Warwick]

For the high resolution mode, it’s often discussed that a very heavy tripod is required. But, in reality, is there a need for anything too weighty? (assuming no wind to buffet the equipment) ..... given, as far as I can tell, the camera is completely vibration free due to the e-shutter? Or am I missing something here if I want to extract the most image quality possible?

Edited July 10, 2020 by Jon Warwick

[oka]

Having used SL2 around a month - the only complain is the firmware... still the latest 2.0 feels like it's pre-release / beta version.

Also I don't like that due to competition (Canon R5 with 8k video capture limitation due to overheating is maybe the best example) nowadays it's new normal to put features / design restrictions with just won't work properly. I haven't done any scientific comparisons, but SL2 feels like it's using batteries much faster than SL. Also I'm constantly getting 4k / burst low battery warnings when battery level is around 10-20% (and doing only stills).

I do hope that we don't have to wait a year or even more for more stable firmware like on SL.

[MediaFotografie]

vor 21 Stunden schrieb Jon Warwick:

For the high resolution mode, it’s often discussed that a very heavy tripod is required. But, in reality, is there a need for anything too weighty? (assuming no wind to buffet the equipment) ..... given, as far as I can tell, the camera is completely vibration free due to the e-shutter? Or am I missing something here if I want to extract the most image quality possible?

I'm using high res mode with Gitzo's (very light) Traveler Series 1 - even with 90-280 without any problems

[scott kirkpatrick]

I used it outdoors with a table top tripod braced on a stone wall.  It worked well. But I might not be getting the absolute most out of it even for a single shot under those conditions.

Edited July 11, 2020 by scott kirkpatrick

[trickness]

On 7/10/2020 at 4:20 PM, oka said:

I haven't done any scientific comparisons, but SL2 feels like it's using batteries much faster than SL. Also I'm constantly getting 4k / burst low battery warnings when battery level is around 10-20% (and doing only stills).

I just got my SL a couple of days ago and I'm shocked at how quickly the battery runs down vs the SL. I took like 10 photos and a fully charged battery was down to 75%. 

[caissa]

The precision of the battery level indicator is not high. So depending how you charged the battery it often takes very little to get from 4 stripes (100%) down to 3 (75%). In reality it was maybe just going from 88% (shown as 4 stripes) to 87% (shown as 3 stripes).That’s why I would like to see a more precise indicator, e.g. as a percentage. Most cameras and phones can do that, why not the SL2 ?

Battery usage is higher in the SL2 than the SL, but not that high. Turn off prefocus to save energy. (For me that made a big difference).

Edited July 21, 2020 by caissa

[scott kirkpatrick]

Also, if you want a better measure of battery charge, use one of the less expensive chargers (e.g. NiteCore) which shows the battery voltage.  This varies from just above 8 V when the battery is fully charged down to 7 volts when nearly empty.  For questions like when will Leica's internal alarm not let you shoot continuously, it's the voltage that they measure.

[Arrow]

As a side note: the multishot mode is great for digitalising slides and negatives. The added resolution is especially welcome when digitalising other than 1:1 formats like 6x6 or Xpan 24x65. 

[verwackelt]

But it seems to smooth the "scans" a little bit. I tried a 35mm negativ with my Lumix G9 in normal resolution and in Highresmode.
The Highresmode kills the filmgrain a bit instead of resolving it higher as i expected.
If you like grain, you should "scan" it in normal resolution. if you do not like grain try the highres mode:
The post is in german. Long text  short:
left is normal resolution mode and the image size is enhanced in PS to the same resolution as the highres shot at the right.

 

[SrMi]

2 hours ago, verwackelt said:

But it seems to smooth the "scans" a little bit. I tried a 35mm negativ with my Lumix G9 in normal resolution and in Highresmode.
The Highresmode kills the filmgrain a bit instead of resolving it higher as i expected.
If you like grain, you should "scan" it in normal resolution. if you do not like grain try the highres mode:
The post is in german. Long text  short:
left is normal resolution mode and the image size is enhanced in PS to the same resolution as the highres shot at the right.

 

The high res mode produces different results in different cameras. The results of Lumix G9 may not be applicable to the output of SL2.

[verwackelt]

Ok, i thought they are similar because of the same technology that is used. Would like to see a comparison with SL2 or S1R…

[scott kirkpatrick]

Leica and Panasonic take different approaches to aggregatng 8 Bayer images into two sets of RGB (Foveon-style) images, the second set displaced by half a pixel spacing in the x and y directions) and then transforming them back into one set of Bayer-style images with half the original pixel spacing and R, G, or B data synthesized at each new pixel.  Panasonic is faster, but has some artifacts at high contrast boundaries.  Leica runs slower, reduces acutance a bit and doesn't have the artifacts. The algorithms are proprietary. but I would guess the differences occur in that last step, which allows the results to be processed by the software we already have.  I wonder if there is a TIFF or PSD format that would accept the Foven-style intermediate result.

Here's an example, using the SL2 and SL 50 APO SC:

full image:

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!

U1040837 by scott kirkpatrick, on Flickr

and a 100% crop from near the center:

U1040837 copy by scott kirkpatrick, on Flickr

I shot this one with the camera on a monopod, held firmly against a pedestrian railing, but I still had to wait about a minute before the SL2 would take down its wiggle warning and let me finish squeezing the trigger.

 

[Advertisement]

Hallo,

Look here: Leica L-Mount

[SrMi]

6 hours ago, scott kirkpatrick said:

<snip>

I shot this one with the camera on a monopod, held firmly against a pedestrian railing, but I still had to wait about a minute before the SL2 would take down its wiggle warning and let me finish squeezing the trigger.

 

What setting for Motion Artefacts Correction did you use? Turning it on should reduce artifacts, turning it off should increase resolution slightly.