Jump to content

S2 and my impressions


Guest guy_mancuso

Recommended Posts

Advertisement (gone after registration)

A note on post processing images to remove lens distortion etc. It should be noted that in order to remove the barrel distortion from an image you need to translate pixels. When you perform a subpixel transformation (moving a pixel more or less than a full integer value), you essentially cut the image resolution in half. Everything is filtered during the transformation and you run the risk of losing extremely fine details and acutance. So, while you get rid of the distortion, there is no free lunch. Sharpening gets back some of the acutance, but it will not regenerate content that was lost during the transformation operation.

 

The S2 lenses appear to get things right at the moment of capture, which means that Leica doesn't have to perform these sort of post processing gymnastics and should be able to produce very, very clean and crisp images.

Link to post
Share on other sites

  • Replies 110
  • Created
  • Last Reply
Guest guy_mancuso

This is where I disagree . To me there is no way to make a 24mm without some barrel distortion in a retro focus design. Unless the OEM goes through huge hoops to do it and a cost and size that no one could afford. Bottom line on this correction will be come from the DNG when shot. Just like the M8 there are corrections made at capture that are in the DNG. Than you can use any raw processor to process. Same for other lens aberrations this will know what lens is on the body and the firmware will know exactly what to do to make further corrections. The M8 does this today and my bet it will carry through with the S2. Something you may never hear about but this is my prediction. Today I can fix almost all my lenses in C1 and my Phase back. Hassy does the same thing with Phocus. Leica IMHO is doing the same thing in camera

Link to post
Share on other sites

In the UK some Leica dealerships are "Premier Dealers", presumably the same as "Preferred Dealers", and they would most likely be the one's who would carry the full S2 system.

 

Don't get me wrong, the main Leica 'premier' dealers in London (Red Dot, RG Lewis, etc.) are great and I enjoy a visit to these shops, but they are a world away from the kind of London dealers (The Pro Centre, Teamwork, etc.) who have day to day dealings with the professionals that the S2 was meant to be targeted at. It seems strange to me that Leica would think that the traditional cosy dealer is the appropriate place to sell the S2. It's a bit like being expected to go to your local garden centre to buy a combine harvester.

 

I wonder if it's a tacit admission by Leica that, contrary to the pre-launch marketing spin, they actually expect the S2 system to sell mainly to affluent amateurs?

 

i don't believe in professional service by leica dealers. they are used to deal with dentist and lawyers and not with pros.

as long as calumet and others are not going to accept the S2 it will not work.

anyhow it's a nice camera for rich amateurs

 

Indeed.

Link to post
Share on other sites

Leica stated on the S2 site that they will work with Phase One on distribution and service of the S2 and believe this to still be the case. Some of the major Leica dealers (either exclusively Leica or just big Leica dealers who also stock other brands) may also be suitable for S2 sales and service, though Leica didn't state that would be the case.

 

But for widest possible distribution of this type of [Pro] camera it should get onto the communication lines of professionals; meaning the camera has to be out in the places where Pro and MF users come weekly for supplies and service. Some Pro customers might also visit the Leica stores such as Berlin for their current Leica needs and I think some of those should be allowed to sell S2. Also a lot of Leica users want to see it, and probably some affluent doctors may want one too.

Link to post
Share on other sites

Guest guy_mancuso

Seems they are changing there tune on Phase and even thinking of no tethered support with C1 and 100 percent sales from leica preferred dealers. Frankly it sounds like there completely closing this system in within themselves.

 

I am stepping out of this thread now since I see no more reason to participate, have a full week of shooting to prepare for. Have fun , you guy's know were to find me for more info if interested. Thanks:)

Link to post
Share on other sites

Advertisement (gone after registration)

I disagree. Print something on a piece of rubber. Now warp that by stretching some of it - have you lost resolution on that rubber print ? Concerning barrel distortion, I would say that everything depends how the correction algorithm is implemented. Of course I certainly have less knowledge of this topic than most others here :)

 

Edmund

 

 

 

A note on post processing images to remove lens distortion etc. It should be noted that in order to remove the barrel distortion from an image you need to translate pixels. When you perform a subpixel transformation (moving a pixel more or less than a full integer value), you essentially cut the image resolution in half. Everything is filtered during the transformation and you run the risk of losing extremely fine details and acutance. So, while you get rid of the distortion, there is no free lunch. Sharpening gets back some of the acutance, but it will not regenerate content that was lost during the transformation operation.

 

The S2 lenses appear to get things right at the moment of capture, which means that Leica doesn't have to perform these sort of post processing gymnastics and should be able to produce very, very clean and crisp images.

Link to post
Share on other sites

This is where I disagree . To me there is no way to make a 24mm without some barrel distortion in a retro focus design. Unless the OEM goes through huge hoops to do it and a cost and size that no one could afford. Bottom line on this correction will be come from the DNG when shot. Just like the M8 there are corrections made at capture that are in the DNG. Than you can use any raw processor to process. Same for other lens aberrations this will know what lens is on the body and the firmware will know exactly what to do to make further corrections. The M8 does this today and my bet it will carry through with the S2. Something you may never hear about but this is my prediction. Today I can fix almost all my lenses in C1 and my Phase back. Hassy does the same thing with Phocus. Leica IMHO is doing the same thing in camera

 

I'm not saying it's a bad thing and you're right, everyone does it. It's just that a lot of people don't realize that when you perform a subpixel transformation on a bit-mapped image, you essentially cut resolution in half. It doesn't matter if you do this on a RAW, tiff or jpeg file. A pixel is a pixel and translation operations have nothing to do with color depth. If you move a pixel a full integer value (1,2,3 etc) there is no penalty. But if you have to move it a by a fraction of a pixel (1.3432 etc) you need to interpolate and filter. Then you take a hit and your resolution is cut in half. You're essentially creating a new pixel from the surrounding ones.

 

The point is that if the S2 lenses are as good as we expect them to be, we will have to perform a lot less of these sort of operations on our files and will not have to pay the penalty others have to. So, if the 75mm spits out and image as flat as a pancake, we get to keep all the resolution those lenses and sensor deliver.

 

I'm very excited about the S2. I can't afford one (and really don't have a need for one), but the S2 represents a glimps behind the curtain and gives us an idea of the direction the new Leica is taking. The M8 is a legacy product from before Kaufmann took over. If anyone wants to guess what the R10 or M9 is going to be like, look at the S2.

Link to post
Share on other sites

I disagree. Print something on a piece of rubber. Now warp that by stretching some of it - have you lost resolution on that rubber print ? Concerning barrel distortion, I would say that everything depends how the correction algorithm is implemented. Of course I certainly have less knowledge of this topic than most others here :)

 

Edmund

 

Rubber is not a good analogy for a bit mapped image, which has a fixed resolution.

 

A vector drawing, like you would make in Adobe Illustrator is more like what you are describing, because it has 'unlimited resolution'.

 

Imagine the difference between bitmapped fonts and postscript (postscript would be like rubber. you can stretch it, without it pixelating and getting jaggies/stairsteps).

Link to post
Share on other sites

I'm not saying it's a bad thing and you're right, everyone does it. It's just that a lot of people don't realize that when you perform a subpixel transformation on a bit-mapped image, you essentially cut resolution in half. It doesn't matter if you do this on a RAW, tiff or jpeg file. A pixel is a pixel and translation operations have nothing to do with color depth. If you move a pixel a full integer value (1,2,3 etc) there is no penalty. But if you have to move it a by a fraction of a pixel (1.3432 etc) you need to interpolate and filter. Then you take a hit and your resolution is cut in half. You're essentially creating a new pixel from the surrounding ones.

 

The point is that if the S2 lenses are as good as we expect them to be, we will have to perform a lot less of these sort of operations on our files and will not have to pay the penalty others have to. So, if the 75mm spits out and image as flat as a pancake, we get to keep all the resolution those lenses and sensor deliver.

 

I'm very excited about the S2. I can't afford one (and really don't have a need for one), but the S2 represents a glimps behind the curtain and gives us an idea of the direction the new Leica is taking. The M8 is a legacy product from before Kaufmann took over. If anyone wants to guess what the R10 or M9 is going to be like, look at the S2.

 

Hy thrid,

 

I absolutely agree with all of your conclusions. I think the combination of the abilities of the new lenses and the developed software is finally what makes the best effort. Let´s see.

 

In that way it is also in my opinion a glimpse behind the curtain of what is really possible and wanted by professional photographers. I am not professional photographer but the S2 shows in fact the way Leica intends to go. So this is most fascinating, also concerning the R 10.

 

Greetings from Berlin

 

Ralf

Link to post
Share on other sites

Could someone please send me the address of the adult Leica Forum where people are civil and grammar/spelling are respected and comments thoughtful. I would appreciate it very much. :(

 

Well said ! i find this entertaining - like reading page six of the NY post

Link to post
Share on other sites

I'm not saying it's a bad thing and you're right, everyone does it. It's just that a lot of people don't realize that when you perform a subpixel transformation on a bit-mapped image, you essentially cut resolution in half. It doesn't matter if you do this on a RAW, tiff or jpeg file.

 

I am not sure if you are correct. Maybe in theory, but I haven't seen it in practice. I use DXO to correct for lens distortion and in practice can't detect any loss of sharpness, let alone cutting the resolution in 1/2. (Not counting extreme cases of fisheye lenses.)

 

Here is an example from a fisheye lens. I use this lens when making virtual tours. I didn't adjust the color, just typical sharpening. I chose an off center area that had much much more distortion than would be common in any non-fisheye quality lens.

 

Also, just so you can see what extreme interpolation does, I've also attached images from about 3/4 of the way up on the left side. This part of the corrected image becomes larger and shows how the detail had to be stretched.

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!

Link to post
Share on other sites

I am not sure if you are correct. Maybe in theory, but I haven't seen it in practice. I use DXO to correct for lens distortion and can't detect any loss of sharpness, let alone cutting the resolution in 1/2.

 

Trust me on this one. I've worked in high-end digital imaging (not Photoshop) for over 15 years. Good code can go a long way to retain image quality, but once you perform the kind of transformation as in the example you gave, none of the pixels in that image are the same, as when they were captured.

Link to post
Share on other sites

I am not professional photographer but the S2 shows in fact the way Leica intends to go. So this is most fascinating, also concerning the R 10.

 

Greetings from Berlin

 

Ralf

 

The R10 is going to be very interesting, as will be the new R-AF glass. Can;t wait to see it

 

Alles gute von London.

 

 

Feli

Link to post
Share on other sites

The "corrected" version of the lamp image @ 100% has lost significant amounts of detail and sharpness, IMO.

 

I agree. It's not bad and could be helped with some smart sharpening, but what you are seeing is the result of the filtering. Every pixel in that new image has been re-generated via interpolation.

Link to post
Share on other sites

The "corrected" version of the lamp image @ 100% has lost significant amounts of detail and sharpness, IMO.

 

Did you also read what I wrote or only looked at the photos? I said - Not counting extreme cases of fisheye lenses. So I gave pretty much the most extreme case of barrel distortion that is possible.

 

In only an extreme case such as with a fisheye lens, where it has to stretch the image is it noticeable. With the amount of distortion that I see in typical lenses, I can't see any loss of detail when barrel distortion is corrected. The middle part of the image where the subject gets compressed may actually gain resolution per square inch. I don't think the detail is actually added or removed, it is just spread over a larger area or compressed into a smaller area depending on where it is in the photo. If you consider a lens that may just have 1 or 2% distortion, I think it will be very very hard, if not impossible to see a loss of detail.

 

Also, I wonder what percentage of MF files end up being used at 100%. At that point I bet the printing technology and ink spread is the limiting factor in resolution. Smaller than 100% and it is up to the software how to throw away detail. Larger than 100% it is up to the software to interpolate it. (Sections of the corrected fisheye image are interpolated to larger than 100% of the captured size.)

Link to post
Share on other sites

i don't believe in professional service by leica dealers. they are used to deal with dentist and lawyers and not with pros.

as long as calumet and others are not going to accept the S2 it will not work.

anyhow it's a nice camera for rich amateurs

 

Ah- but Calumet is more than highly interested in the S2.

Link to post
Share on other sites

With the amount of distortion that I see in typical lenses, I can't see any loss of detail when barrel distortion is corrected.

 

How many typical lenses challenge the resolution of the sensor? I suspect that you will see more resolution loss with a better-than-typical lens. Still might be better than the typical lens, but you will see a loss of image detail.

Link to post
Share on other sites

Consider how small and inexpensive a fisheye lens is compared with something such as the forthcoming 17mm TSE. For Canon it is about $650 vs $2500 and 12 onces vs. 29 ounces. It occurs to me that if a fisheye lens could be made that is very high resolution and the sensor is also high resolution, one could afford to lose some of the sensor resolution when correcting for the distortion of convergence in the fisheye photo.

Link to post
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...