mirekti Posted May 27, 2014 Share #81 Posted May 27, 2014 Advertisement (gone after registration) If everything Leica makes is made with high precision, the sensors should fit the same logic. Hence, the Embedded profile should be precise, and work for all Ms. Even the Adobe's standard profile should work as that's the software camera gets delivered with. Even though the procedure sounds simple, I'd had different profiles created based on a same source photo but using diff tool Adobe vs Color-checker. Ah, nevermind... Link to post Share on other sites More sharing options...
Advertisement Posted May 27, 2014 Posted May 27, 2014 Hi mirekti, Take a look here New Firmware on May 24th. I'm sure you'll find what you were looking for!
mjh Posted May 27, 2014 Share #82 Posted May 27, 2014 If everything Leica makes is made with high precision, the sensors should fit the same logic. Hence, the Embedded profile should be precise, and work for all Ms. Even the Adobe's standard profile should work as that's the software camera gets delivered with. The embedded profile is just two matrices (for two illuminants) while Adobe’s standard profile also comprises a LUT for much finer grained corrections. I see no reason one would ever use the embedded profile. Even though the procedure sounds simple, I'd had different profiles created based on a same source photo but using diff tool Adobe vs Color-checker. Adobe’s DNG Profile Editor and X-Rite’s ColorChecker Passport use different algorithms for deriving the LUTs from a target image. For one thing, Adobe’s LUTs have many more entries that X-Rite’s (for the same source). Adobe distinguishes 90 hues and 25 levels of saturation versus X-Rite’s 6 hues, 6 levels of saturation, and 3 levels of brightness. Link to post Share on other sites More sharing options...
edwardkaraa Posted May 27, 2014 Share #83 Posted May 27, 2014 The embedded profile is just two matrices (for two illuminants) while Adobe’s standard profile also comprises a LUT for much finer grained corrections. I see no reason one would ever use the embedded profile. That's very interesting. I have been using the embedded profile for both the M9 and M240 because I found it produced the more interesting color and closer to the camera jpg. Would it be possible to give us more info about this subject? That would be greatly appreciated! Link to post Share on other sites More sharing options...
pieterpronk Posted May 27, 2014 Share #84 Posted May 27, 2014 Whatever we can do to deal with the colorprofiles, I'm just hoping Leica will fix the color-output. If it really were that simple to make better colorprofiles I see no reason why Leica wouldn't just use it themselves. Surely a company that sells 6000€ cameras can invest in making the best color profile and ask Adobe to use it? Meanwhile I spend way too much time trying to fix color and white balance especially when photographing indoors or with various lightsources. And it's kinda sad that Leica is losing on this point when compared to most other mirrorless cameras. Hopefully the new firmware closes this gap. Link to post Share on other sites More sharing options...
jaapv Posted May 27, 2014 Share #85 Posted May 27, 2014 I must say most users seem to be happy with the present output, so it appears there is not too much to be fixed, if anything. Personally I prefer to use IR filters for optimum result and I find the differences between a user-generated profile and the embedded one marginal. The only situation where I like my own profile a lot better is with Tungsten light. Having said that, the appreciation of colour is a highly personal affair. Link to post Share on other sites More sharing options...
pieterpronk Posted May 27, 2014 Share #86 Posted May 27, 2014 Well, the colorproblem seems to pop up quite often compared how often colors of other cameras get mentioned. Especially in places beyond this forum. Obviously it seems that the discussion is over, but I would say that's more because it doesn't seem likely there will be any changes or fixes anyway, than that people are especially happy with the color output or white balance. The fact that people are still using third party tools to calibrate colors and suggesting the use of uv/ir filters points is quite telling imo. Link to post Share on other sites More sharing options...
jaapv Posted May 27, 2014 Share #87 Posted May 27, 2014 Advertisement (gone after registration) I have always done that, on various brands.. I used IR filters long before the M8 came out, on Epson... Profiles since the M9. Link to post Share on other sites More sharing options...
mjh Posted May 27, 2014 Share #88 Posted May 27, 2014 That's very interesting. I have been using the embedded profile for both the M9 and M240 because I found it produced the more interesting color and closer to the camera jpg. Would it be possible to give us more info about this subject? That would be greatly appreciated! For example, the Adobe Standard profile for the M (Typ 240) has two look-up tables, one (HueSatDeltas1) for standard illuminant A, i.e. tungsten lighting, and another (HueSatDeltas2) for D65, i.e. the daylight on a sunny day at noon. These tables define how colours should be mapped within the colour space to obtain an optimum colour rendering, depending on the illuminant. If the actual colour temperature is between those of A and D65 the raw converter interpolates the look-up table to be applied. And then there is a third look-up table (LookTable) that is applied in a second step. This third look-up table also differentiates between different levels of brightness, whereas HueSatDeltas1 and HueSatDeltas2 do not (they could, in principle, but in this profile they don’t). What these look-up tables do can be visualised within a colour circle: HueSatDeltas1: HueSatDeltas2: LookTable (only the version for maximum brightness is shown here; there are 15 more): These diagrams visualise the mapping by indicating the original colour by a circle and the resulting colour by a square. Btw, look-up tables such as these could be stored in a DNG file as Adobe’s DNG camera profiles are really just the subset of DNG that applies to camera profiling, but as far as I know, camera vendors only embed matrices, but no look-up tables. Link to post Share on other sites More sharing options...
mirekti Posted May 27, 2014 Share #89 Posted May 27, 2014 Btw, look-up tables such as these could be stored in a DNG file as Adobe’s DNG camera profiles are really just the subset of DNG that applies to camera profiling, but as far as I know, camera vendors only embed matrices, but no look-up tables. Thanks for the clarification. Would it be irrational to ask Leica to provide tables in the DNG files or any other way so it could work smoothly, and without need for us to make custom profiles? Link to post Share on other sites More sharing options...
luigi bertolotti Posted May 27, 2014 Share #90 Posted May 27, 2014 Thanks for the clarification.Would it be irrational to ask Leica to provide tables in the DNG files or any other way so it could work smoothly, and without need for us to make custom profiles? I doubt that, at the end, it would result in less work to fine tuning color profiling... and, probably this could make more complex the structure of the DNG files, with some possible collateral issue. (btw, I think that Leica embedded matrices/table can vary during production lifecycle... maybe in consequence of variations of the sensor from batch to batch, or anyway during its manufacturing history) Link to post Share on other sites More sharing options...
mirekti Posted May 27, 2014 Share #91 Posted May 27, 2014 Not sure if this makes any sense, but if the rangefinder can be calibrated per camera so could the sensor. Link to post Share on other sites More sharing options...
mjh Posted May 28, 2014 Share #92 Posted May 28, 2014 Not sure if this makes any sense, but if the rangefinder can be calibrated per camera so could the sensor. Yes, but it is simpler to create a new profile than to have Leica re-calibrate the sensor (and store an updated profile in the camera’s firmware so it can be embedded in each DNG file). The result would be the same. Link to post Share on other sites More sharing options...
jto555 Posted May 29, 2014 Share #93 Posted May 29, 2014 A quick info for all of you waiting desperately: In a few days Leica will finish the new Leica M (Type 240) firmware. Our Leica Forum beta tester will have a short and intense field test - and if no severe bugs show up the firmware will be released. Thus there is nor a fixed release date neither a detailed list of improvements. What I’m allowed to say: The new firmware will bring Auto-ISO! Any update as to the firmware release date? Link to post Share on other sites More sharing options...
farnz Posted May 29, 2014 Share #94 Posted May 29, 2014 Any update as to the firmware release date? Patience, Grasshopper, patience. It's only been three days since Andreas's (Admin's) announcement. Kung Fu Pete. Link to post Share on other sites More sharing options...
jcraf Posted May 29, 2014 Share #95 Posted May 29, 2014 I've just sold my M, so the firmware release will now be imminent...... Link to post Share on other sites More sharing options...
AlanJW Posted May 29, 2014 Share #96 Posted May 29, 2014 I've just sold my M, so the firmware release will now be imminent...... Is it the firmware that caused you to bolt? Link to post Share on other sites More sharing options...
jcraf Posted May 30, 2014 Share #97 Posted May 30, 2014 Is it the firmware that caused you to bolt? A variety of reasons, the main of which is called the MM. Just wasn't using the M. Regards Link to post Share on other sites More sharing options...
Exodies Posted May 30, 2014 Share #98 Posted May 30, 2014 I wonder what Admin means by "auto ISO" Link to post Share on other sites More sharing options...
dante Posted May 30, 2014 Share #99 Posted May 30, 2014 I wonder what Admin means by "auto ISO" I'm more fascinated that new firmware gets some testing by some unknown forum member and then gets unleashed. This must explain why the current firmware seems so crashy. Dante Link to post Share on other sites More sharing options...
genefama Posted May 31, 2014 Share #100 Posted May 31, 2014 The colors aren't rich and natural looking. Sometimes the images even look colorized. There's a kind of a pink and yellow bias to everything: photos err on the side of being too warm. I tend to convert to monochrome more than I did with other cameras. The highlights are too hot and the dynamic range in the highlight area is narrow. Highlights tend to blow out quickly and are comparatively hard to recover for a modern expensive camera. The much-vaunted dynamic range might be there, but not at the high end. I tend to arrange my shots to avoid sunny areas and highlights in scenes that I normally wouldn't worry about. I can tell M240 shots on the net instantly by these traits. Not a first rate sensor or a first rate processor, but a nice camera. Wish I could rely on it more for professional jobs. Link to post Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.