rramesh Posted October 26, 2015 Share #1 Posted October 26, 2015 Advertisement (gone after registration) Could someone throw some light on the SL's full frame sensor? Is it designed like that of the M with micro-lenses to improve edges. While I don't think this will be needed for SL and S lenses, I believe this will be a requirement for the wider M lenses. Could this mean that M lenses will not perform as well on the SL as on the M? Link to post Share on other sites More sharing options...
Advertisement Posted October 26, 2015 Posted October 26, 2015 Hi rramesh, Take a look here Leica SL sensor. I'm sure you'll find what you were looking for!
rramesh Posted October 27, 2015 Author Share #2 Posted October 27, 2015 This is what I see on the Leica website. However, I am unclear as to how the SL sensor can be optimised for SL, R and M lenses. 24-Megapixel CMOS Sensor Highest imaging quality in all situationsThe sensor of the Leica SL has 4000 × 6000 pixels with a pixel pitch of 6 × 6 µm on an active surface area with dimensions of 24 × 36 mm2. This means that you can depend on the imaging quality of a 35 mm, full-frame format. To ensure maximum image sharpness, Leica has intentionally left out the low-pass filter. At the same time, a specially designed pixel architecture ensures that more light falls on each individual photodiode. The result is impressive dynamic range, excellent contrast rendition, exceptional sharpness, the highest resolution, and noise-free images in almost all lighting conditions. Particularly in available-light situations, the interplay of the sensor with the Leica SL lenses reveals its full potential and captivates with atmospheric images of superior quality at ISO values of up to 50000. The sensor also plays a key role in terms of the compatibility of the Leica SL-System. For instance, it is optimized for use with Leica M-Lenses and, in addition to supporting all functions, delivers a picture quality that can be achieved only by Leica cameras. Link to post Share on other sites More sharing options...
40mm f/2 Posted October 27, 2015 Share #3 Posted October 27, 2015 I am curious what Leica calls "impressive dynamic range" compares to most recent Sony etc.. Not much is know about this sensor! Link to post Share on other sites More sharing options...
IkarusJohn Posted October 27, 2015 Share #4 Posted October 27, 2015 If you read around the reviews, you will see some useful comments about the dynamic range and useable ISO range for the sensor. My reading is that for a 24MP sensor, both dynamic range and usable ISO are class leading. Don't take my word for it though. Cheers John Link to post Share on other sites More sharing options...
viramati Posted October 27, 2015 Share #5 Posted October 27, 2015 I only hope that it doesn't produce the shadow banding that you get with the Q and the M. I shot at old friends funeral wake yesterday using the Q and Sony A7rII and again hit the banding problem when working on the files from the Q. Link to post Share on other sites More sharing options...
Willy ͡° ͜ʖ ͡° Posted October 28, 2015 Share #6 Posted October 28, 2015 What happens (crop ratio) when you attach a S lens to the SL body? Link to post Share on other sites More sharing options...
psss Posted October 28, 2015 Share #7 Posted October 28, 2015 Advertisement (gone after registration) What happens (crop ratio) when you attach a S lens to the SL body? the S lenses cover more then 35mmFF so no issues there Link to post Share on other sites More sharing options...
psss Posted October 28, 2015 Share #8 Posted October 28, 2015 If you read around the reviews, you will see some useful comments about the dynamic range and useable ISO range for the sensor. My reading is that for a 24MP sensor, both dynamic range and usable ISO are class leading. Don't take my word for it though. Cheers John i thought i have read every review available and i haven't seen anything that would support "class leading"....dpreview has files up and available for download...for all cameras, so you can compare and see for yourself..... leica does not list specs for DR, which is a tricky thing to do anyway.... from my own comparisons and the files i got out of the SL i played with, i would say the chip is definitely competitive with existing products...high iso is a little behind sony but the grain/texture is more pleasing to me (less chroma noise)....of course the problem is that the SL is going head to head with cameras costing a fraction..... Link to post Share on other sites More sharing options...
viramati Posted October 28, 2015 Share #9 Posted October 28, 2015 Preview has done some testing http://www.dpreview.com/articles/9955093579/leica-sl-typ601-in-depth-camera-review/3 Unfortunately the sensor seems to show the same shadow are banding as the Q and while I am happy to deal with it on the Q when it does occur there is no way that I would invest in a 'Pro' camera with this going on. Link to post Share on other sites More sharing options...
ramarren Posted October 28, 2015 Share #10 Posted October 28, 2015 I don't quite understand the basis of these comparisons. Who pushes shadow values six stops rather than making proper exposures to begin with? Then again, I've never trusted these on-line reviews in the first place ... Neither when they say good things nor when they say bad things. So I guess it's just another "Eh, so what? I'll see how it behaves when I test mine." Link to post Share on other sites More sharing options...
mjh Posted October 28, 2015 Share #11 Posted October 28, 2015 What happens (crop ratio) when you attach a S lens to the SL body? The crop factor is 1.25. Link to post Share on other sites More sharing options...
viramati Posted October 28, 2015 Share #12 Posted October 28, 2015 I don't quite understand the basis of these comparisons. Who pushes shadow values six stops rather than making proper exposures to begin with? Then again, I've never trusted these on-line reviews in the first place ... Neither when they say good things nor when they say bad things. So I guess it's just another "Eh, so what? I'll see how it behaves when I test mine." Unfortunately with the Q you certainly don't have to push by 6 stops to see the effect. Please don't get me wrong and think I am Leica bashing but as much as I love my Q (and the problem was the same with the M when I had it) even on a so called 'correctly' exposed shot when trying to draw out detail in shadow areas in contrasty situation it is not hard to start seeing banding. My Sony's (A7s and A7rII) show no sign of banding when performing similar PP to an image. So this comes from my own experience with the Q and the SL has basically the same sensor and processor and I would have hoped not to have seen this issue in such an otherwise state of the art camera. I feel it is summed up well in the dpreview piece at the end of the page "What's This Mean? The Raw dynamic range of the Leica SL shows some significant limitations. It falls well behind class-leaders like the Nikon D750, D810, or Sony a7R II (scroll back up to widget). However, from purely a random noise level, the SL shows more dynamic range than a comparable Canon full-frame camera. However, the reality is that banding creeps in pretty early in shadows at low ISO settings, which will limit the real-world 'push ability' of shadows. What this ultimately means is that you'll be limited in your ability to decrease exposure to expose so as to not blow highlights in high contrast scenes, because you'll be limited in your ability to correct (brighten) dark tones in post-processing. Do note that at higher ISOs (800 and upward), you probably won't notice these limitations in the slightest, though you will be limited, in such low light situations, in your ability to decrease the ISO setting to preserve highlight detail, as the camera is fairly ISO-variant." Link to post Share on other sites More sharing options...
fnuernberger Posted October 28, 2015 Share #13 Posted October 28, 2015 Leica have shot themselves in the foot with this sensor. A Nikon D750 - for about 1750 € - has a vastly superior sensor. If you need to deliver results in a hurry (and get a bit sloppy with your metering) or if you need to extract just that extra bit of shaddow information from that file, you better have a camera that allows for these operations.Their "let's go for the pros" campaign is seriously flawed. Just one variable-aperture zoom lens available. And even more rumors about yet another Leica "system". Why can't they just concentrate on producing the best rangefinder and the best medium-format camera? Do they really need to try to compete in every category of camera imaginable? This half-assed trials (also the T) cost just a lot of money. Link to post Share on other sites More sharing options...
psss Posted October 28, 2015 Share #14 Posted October 28, 2015 I don't quite understand the basis of these comparisons. Who pushes shadow values six stops rather than making proper exposures to begin with? Then again, I've never trusted these on-line reviews in the first place ... Neither when they say good things nor when they say bad things. So I guess it's just another "Eh, so what? I'll see how it behaves when I test mine." you don't have to trust anyone, just download the files and play around..... they are helping you by pushing and pulling files in extreme situations..... i have shot the SL (and have my own files) and it is pretty obvious that you can't recover as much from the shadows (or the highlights) as from sony sensors (in sony and nikon) and like the review says: it looks like it is better then canon but canon is really behind sony in this regard..... the SL seems to be somewhere between 12 (canon) and 14 (sony) stops DR....but banding IS a real problem, very hard to remove in post.....the review mentions leica working on it, FW should be able to take care of it.... Link to post Share on other sites More sharing options...
viramati Posted October 28, 2015 Share #15 Posted October 28, 2015 you don't have to trust anyone, just download the files and play around..... they are helping you by pushing and pulling files in extreme situations..... i have shot the SL (and have my own files) and it is pretty obvious that you can't recover as much from the shadows (or the highlights) as from sony sensors (in sony and nikon) and like the review says: it looks like it is better then canon but canon is really behind sony in this regard..... the SL seems to be somewhere between 12 (canon) and 14 (sony) stops DR....but banding IS a real problem, very hard to remove in post.....the review mentions leica working on it, FW should be able to take care of it.... Well if that is the case I hope Leica can sort the banding issue in my Q as well Link to post Share on other sites More sharing options...
CheshireCat Posted October 28, 2015 Share #16 Posted October 28, 2015 but banding IS a real problem, very hard to remove in post.....the review mentions leica working on it, FW should be able to take care of it.... Uhm... how can they remove the banding in firmware (software) ? The only way is pre-cooking the raw with a possibly unwanted denoise filter. Link to post Share on other sites More sharing options...
psss Posted October 29, 2015 Share #17 Posted October 29, 2015 Uhm... how can they remove the banding in firmware (software) ? The only way is pre-cooking the raw with a possibly unwanted denoise filter. that is of course a very good question....do we know of anyone else using the same sensor? anyone who is getting the data off clean? sony just had a similar issue with the new 42mpix sensor...or the compressed raw from it.....the usually do a 11bit+4 (i have no clue what actually means) for their raw which keeps file sizes down considerably......but people were freaking out because there was visible banding in dark shadows when people went nuts with adjustments....not a normal situation but still easily replicated.....sony came out with a firmware upgrade and now provides 14bit raw files along with the other raw files....the general opinion is that it makes no sense to shoot the 14bit raws but you can if you want and go nuts with sliders and have no banding in the underexposed dark shadows.....now i am pretty sure there is no detail there either but banding is ugly and options are good to have..... anyway: i think leica knows that this is an issue with this sensor and i am pretty sure they can come up with something....especially since this (SL) is the new professional flagship:) Link to post Share on other sites More sharing options...
ramarren Posted October 29, 2015 Share #18 Posted October 29, 2015 Hmm. So let's get this straight. You have a contrasty scene, a scene in which the dynamic range exceeds the sensor's dynamic range. So you attempt to fix it by underexposing to minimize highlight saturation, and then pulling up shadow values in rendering. That's when you see these problems... Do I have that right? I guess that's just not how I work the problem. If I have a contrasty scene that exceeds the sensor's range, I either decide where I'm going to clip tones to stay within the dynamic range I can record or I add light to the shadows to reduce contrast and fit the sensor better. I actually end up pushing shadow values down a little bit in rendering most of the time. Old habits from working with far more intractable film materials ... "Capture low contrast, add contrast to suit in rendering." I expect the SL will show up with a lot of interesting insights when it arrives. I have a Nikon D750 already (and M-P) so I'll be doing a lot of testing and experimentation for the next year. :-) Unfortunately with the Q you certainly don't have to push by 6 stops to see the effect. Please don't get me wrong and think I am Leica bashing but as much as I love my Q (and the problem was the same with the M when I had it) even on a so called 'correctly' exposed shot when trying to draw out detail in shadow areas in contrasty situation it is not hard to start seeing banding. My Sony's (A7s and A7rII) show no sign of banding when performing similar PP to an image. So this comes from my own experience with the Q and the SL has basically the same sensor and processor and I would have hoped not to have seen this issue in such an otherwise state of the art camera. I feel it is summed up well in the dpreview piece at the end of the page "What's This Mean? The Raw dynamic range of the Leica SL shows some significant limitations. It falls well behind class-leaders like the Nikon D750, D810, or Sony a7R II (scroll back up to widget). However, from purely a random noise level, the SL shows more dynamic range than a comparable Canon full-frame camera. However, the reality is that banding creeps in pretty early in shadows at low ISO settings, which will limit the real-world 'push ability' of shadows. What this ultimately means is that you'll be limited in your ability to decrease exposure to expose so as to not blow highlights in high contrast scenes, because you'll be limited in your ability to correct (brighten) dark tones in post-processing. Do note that at higher ISOs (800 and upward), you probably won't notice these limitations in the slightest, though you will be limited, in such low light situations, in your ability to decrease the ISO setting to preserve highlight detail, as the camera is fairly ISO-variant." Link to post Share on other sites More sharing options...
CheshireCat Posted October 29, 2015 Share #19 Posted October 29, 2015 sony just had a similar issue with the new 42mpix sensor...or the compressed raw from it.....the usually do a 11bit+4 (i have no clue what actually means) for their raw which keeps file sizes down considerably The difference here is that Sony's issue was software-only: they were using a surprisingly stupid lossy-compression scheme to keep the file size down. The Sony sensor is perfectly fine, and indeed, once the stupid compression scheme is disabled, the problem is gone. The next step for Sony is to update the firmware to use a lossless-compression scheme like any other respectable manufacturer. In any case, I have checked the SL files for the "Studio Scene" comparison in DPReview, and honestly, I don't see any noticeable banding. What I see instead is that the shots are quite softer than any other 24MP camera, including the Leica Q files (which is supposed to use the same sensor) even at low ISO. It may be the lens, but really, I am puzzled... Link to post Share on other sites More sharing options...
psss Posted October 29, 2015 Share #20 Posted October 29, 2015 The difference here is that Sony's issue was software-only: they were using a surprisingly stupid lossy-compression scheme to keep the file size down. The Sony sensor is perfectly fine, and indeed, once the stupid compression scheme is disabled, the problem is gone. The next step for Sony is to update the firmware to use a lossless-compression scheme like any other respectable manufacturer. In any case, I have checked the SL files for the "Studio Scene" comparison in DPReview, and honestly, I don't see any noticeable banding. What I see instead is that the shots are quite softer than any other 24MP camera, including the Leica Q files (which is supposed to use the same sensor) even at low ISO. It may be the lens, but really, I am puzzled... i would say that the compressed lossless format sony is using is a pretty good compromise but options are good..... i was surprised about the softness as well....especially since i got to try the 24-90 and have files that everything but soft....DPreview also mentions that the lens was soft in their tests.....the copy i got to handle was definitely not soft in any way.....compared to pretty much all 24-70/24-105 i have owned and tested it was noticeably sharper (which it should be at 5x the price but that is a different story) Link to post Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.