mjh Posted October 30, 2014 Share #21 Posted October 30, 2014 Advertisement (gone after registration) Isn't tonal gradation totally dependent on PP (adjusting curve)? In digital photography, tonal gradation gets finer towards the highlights, so ETTR gives a finer resolution when you push the histogram as far to the right as possible without clipping. The term ‘correct exposure’ is a red herring. When the contrast in the scene equals the dynamic range the sensor can capture then there is indeed just one exposure that would be deemed correct – namely an exposure exactly fitting the range of tonal values within the sensor’s dynamic range. When the contrast is lower than the dynamic range then there are many possible exposures that are equally correct in that they all preserve the tonal values in the scene. ETTR says which of the equally correct exposures is optimal in that it maximises the resolution of tonal values and minimises noise (namely one which amounts to overexposure). When the contrast is higher than the dynamic range then all exposures will be lacking in some respect – they will either clip highlights, produce shadow noise, or both. ETTR suggests that you should absolutely avoid clipping highlights (i.e. it recommends underexposure) and rather deal with the inevitable noise – noise can be suppressed but clipped highlights are gone for good. And btw, there is nothing special about the M (Typ 240) in this regard. Link to post Share on other sites More sharing options...
Advertisement Posted October 30, 2014 Posted October 30, 2014 Hi mjh, Take a look here ETTR with the M 240?. I'm sure you'll find what you were looking for!
edwardkaraa Posted October 30, 2014 Author Share #22 Posted October 30, 2014 In digital photography, tonal gradation gets finer towards the highlights, so ETTR gives a finer resolution when you push the histogram as far to the right as possible without clipping. The term ‘correct exposure’ is a red herring. When the contrast in the scene equals the dynamic range the sensor can capture then there is indeed just one exposure that would be deemed correct – namely an exposure exactly fitting the range of tonal values within the sensor’s dynamic range. When the contrast is lower than the dynamic range then there are many possible exposures that are equally correct in that they all preserve the tonal values in the scene. ETTR says which of the equally correct exposures is optimal in that it maximises the resolution of tonal values and minimises noise (namely one which amounts to overexposure). When the contrast is higher than the dynamic range then all exposures will be lacking in some respect – they will either clip highlights, produce shadow noise, or both. ETTR suggests that you should absolutely avoid clipping highlights (i.e. it recommends underexposure) and rather deal with the inevitable noise – noise can be suppressed but clipped highlights are gone for good. And btw, there is nothing special about the M (Typ 240) in this regard. In my humble experience with the M240, I find it to be different than current CMOS based FF cameras in that there is almost no latitude in the highlights. This camera does not like overexposure. In this respect, if ETTR means absolutely avoiding clipping highlights, I totally agree. However, in my experience, there are many instances where actually contrast is lower than DR, and ETTR mostly leads to overexposure. In these cases, the M240 does not like recovering highlights, even if they are not clipped. I have seen RGB values in the 230-245 produce not so nice colors when they are pulled back. The closer you get to the clipping point, the worse it gets. Link to post Share on other sites More sharing options...
AlanJW Posted October 30, 2014 Share #23 Posted October 30, 2014 When the contrast is higher than the dynamic range then all exposures will be lacking in some respect – they will either clip highlights, produce shadow noise, or both. ETTR suggests that you should absolutely avoid clipping highlights (i.e. it recommends underexposure) and rather deal with the inevitable noise – noise can be suppressed but clipped highlights are gone for good. Michael: Now I am confused, as I thought ETTR was "open up" until just before the highlights clip. Therefore I did not understand that ETTR suggests underexposure. I suppose I may be doing ETTR after all, as my big objective is not to blow out highlights even if some shadow areas are underexposed. So if the ETTR recommendation is to tend to underexpose those areas of a scene that are in shadows in order not to blow out those areas that are highlights, I guess that is what I am doing without labeling it as "ETTR". Am I missing something? Link to post Share on other sites More sharing options...
edwardkaraa Posted October 30, 2014 Author Share #24 Posted October 30, 2014 Alan, in some low contrast scenes that contain no significant highlights, you can overexpose from the optimal exposure by 3-4 stops and still do not clip any color channel. In this case, ETTR means gross overexposure even if there are no clipped highlights. Link to post Share on other sites More sharing options...
jaapv Posted October 30, 2014 Share #25 Posted October 30, 2014 As ETTR was advocated to reduce shadow noise in the first place the idea is a bit obsolete nowadays with modern low-noise and high DR sensors, added to sophisticated noise-reduction software. (see Schewe and Fraser, Real World Image Sharpening for the original rationale. They used a Canon 10D to demonstrate the effect). Link to post Share on other sites More sharing options...
cirke Posted October 30, 2014 Share #26 Posted October 30, 2014 Same here. Never exposed to the right with the M8, M9, M240, S2. A bit of noise in the shadows always preferable to burnt highlights and the problems that you describe.burning highlights has nothing to do with ETTR As ETTR was advocated to reduce shadow noise in the first place the idea is a bit obsolete nowadays with modern low-noise and high DR sensors, added to sophisticated noise-reduction software. (see Schewe and Fraser, Real World Image Sharpening for the original rationale. They used a Canon 10D to demonstrate the effect).If you want the best IQ it will not be obsolete for a while , there is a huge difference with and without Link to post Share on other sites More sharing options...
cirke Posted October 30, 2014 Share #27 Posted October 30, 2014 Advertisement (gone after registration) I am using spot metering with any cameras I had before or now (the M240) and of course I use always ETTR which is from far the best method Link to post Share on other sites More sharing options...
flyalf Posted October 30, 2014 Share #28 Posted October 30, 2014 As far as I understand and experience ETTR and one-zone has clear IQ benfits also for M. My I propose that we are more presice in wording: "Correct exposure", "optimal exposure" , "under" or "over" must be written with refernce to something, like grey card measurment, in-camera, or other. The result of the exposure must relate to out-of-camera jpeg or “the final processed exposure” after post. After all its the final outcome that counts. Link to post Share on other sites More sharing options...
mjh Posted October 30, 2014 Share #29 Posted October 30, 2014 In these cases, the M240 does not like recovering highlights, even if they are not clipped. I have seen RGB values in the 230-245 produce not so nice colors when they are pulled back. The closer you get to the clipping point, the worse it gets. Recovering highlights can mean two different things: First there is some clipping occurring in internal image processing and this kind of clipping that only affects JPEGs is easily removed in raw development. The latitude solely depends on how severe the in-camera clipping is – the less clipping there is the less the amount of highlights to be recovered. So when the in-camera processing strives for a high contrast it will throw over the top much of the highlights and consequently there is also much to be recoved. This doesn’t imply that the camera was much better at preserving highlights – far from it actually. Another kind of highlight recovery can be applied when one colour channel starts to show clipping but the others do not – the raw converter can then try to reconstruct the missing info from the other channels. How much can be recovered depends on how different the exposure of the three channels is, which in turn depends on subject colour, colour temperature, and the differences in the sensor’s quantum efficiencies for red, green, and blue. Link to post Share on other sites More sharing options...
mjh Posted October 30, 2014 Share #30 Posted October 30, 2014 "Correct exposure", "optimal exposure" , "under" or "over" must be written with refernce to something, like grey card measurment, in-camera, or other. Grey card measurement (also incident light metering etc.) ignores the actual distribution of tonal values as shown in the histogram which is what ETTR is about. Link to post Share on other sites More sharing options...
mjh Posted October 30, 2014 Share #31 Posted October 30, 2014 Now I am confused, as I thought ETTR was "open up" until just before the highlights clip. Therefore I did not understand that ETTR suggests underexposure. ETTR means that you shift the tonal values as far to the right as possible without clipping the highlights. When the contrast is high, chances are that the highlights are over the top already in which case you have to shift left – i.e. underexpose. Link to post Share on other sites More sharing options...
pop Posted October 30, 2014 Share #32 Posted October 30, 2014 ETTR means that you shift the tonal values as far to the right as possible without clipping the highlights. When the contrast is high, chances are that the highlights are over the top already in which case you have to shift left – i.e. underexpose. "Underexpose" with respect to what? Link to post Share on other sites More sharing options...
cirke Posted October 30, 2014 Share #33 Posted October 30, 2014 ETTR means that you shift the tonal values as far to the right as possible without clipping the highlights.without clipping any RGB color Link to post Share on other sites More sharing options...
pgk Posted October 30, 2014 Share #34 Posted October 30, 2014 When the contrast is lower than the dynamic range then there are many possible exposures that are equally correct in that they all preserve the tonal values in the scene. ETTR says which of the equally correct exposures is optimal in that it maximises the resolution of tonal values and minimises noise (namely one which amounts to overexposure). In my experience, the ability of ETTR to preserve tonal (and colour) values is not as good as it is all too often cited to be and using it can involve substantially more adjustments to produce a presentable image. I used ETTR for a time and then stopped because it was as much of a compromise in terms of time and adjustment and final image 'quality' as exposing 'normally' would produce. FWIW, having actually compared low contrast files differently exposed I'm far from convinced that ETTR is, or ever was, worth the effort. Link to post Share on other sites More sharing options...
fotofool Posted October 30, 2014 Share #35 Posted October 30, 2014 A member of this forum, sandymc, wrote a blog post on this subject which I found fascinating. He provides a technical explanation for the color issues some of seem to be seeing. ChromaSoft: Why "Expose to the Right" is just plain wrong Link to post Share on other sites More sharing options...
Exodies Posted October 30, 2014 Share #36 Posted October 30, 2014 I tried the test Edward suggested. The results are disturbing. Pictures were taken hand held, focused on the lowest red band for first shot only, 40 seconds apart, hopping about between shots. First is exposure metered using the centre weighted meter. Second is over exposed by +2 (quarter the shutter speed) exposure adjusted by -2 in Lightroom. Camera calibration profile is Adobe standard. Sky was uniformly overcast. WB and tint same in both. Cropped to about 1000*500 pixels, exported as jpeg with standard screen sharpening. 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! Here are the histograms of the uncropped photos, metered exposure first. Untreated over-exposed photo: What I see, Edward, is a good reason not to mess with your exposure in Lightroom; you'll only make things worse. Or is the lesson don't overexpose? Link to post Share on other sites Simply register for free here – We are always happy to welcome new members! Here are the histograms of the uncropped photos, metered exposure first. Untreated over-exposed photo: What I see, Edward, is a good reason not to mess with your exposure in Lightroom; you'll only make things worse. Or is the lesson don't overexpose? ' data-webShareUrl='https://www.l-camera-forum.com/topic/236259-ettr-with-the-m-240/?do=findComment&comment=2697777'>More sharing options...
edwardkaraa Posted October 31, 2014 Author Share #37 Posted October 31, 2014 What I see, Edward, is a good reason not to mess with your exposure in Lightroom; you'll only make things worse. Or is the lesson don't overexpose? Glad that you can see the problem in your own test The article by Sandy mentioned above documents very well this issue. Definitely, in my opinion, do not overexpose. The M240 likes accurate exposure or very slight underexposure, according to what I see in my photos. Only one exception, when shooting night scenes with nice colored light sources, neon signs... etc, it is better to underexpose by 1 or 2 stops to get the lights well exposed (at as low iso as possible) and then pushing the exposure in post. Link to post Share on other sites More sharing options...
k-hawinkler Posted October 31, 2014 Share #38 Posted October 31, 2014 So, how different is the behavior of the M240 from the M9 in light of the discussion here? http://www.l-camera-forum.com/leica-forum/leica-m9-forum/195838-expose-right-ettr-underexpose.html Post #163 to the end, in particular the measurements by Lindolfi? Link to post Share on other sites More sharing options...
edwardkaraa Posted October 31, 2014 Author Share #39 Posted October 31, 2014 So, how different is the behavior of the M240 from the M9 in light of the discussion here?http://www.l-camera-forum.com/leica-forum/leica-m9-forum/195838-expose-right-ettr-underexpose.html Post #163 to the end, in particular the measurements by Lindolfi? In my opinion, they behave pretty much the same, except for the more DR and greenish shadows of the M 240. Whatever applied to the M9 still applies to the M 240. Link to post Share on other sites More sharing options...
k-hawinkler Posted October 31, 2014 Share #40 Posted October 31, 2014 Thanks Edward. Here http://www.l-camera-forum.com/leica-forum/2220127-post244.html is an interesting quote by douglasf13, quote: "The D700, D3, D4 and others will behave like the 5Dii in this regard. The Nikon cameras with Sony sensors, the D7000, D3x, D800, etc., will perform more like the M9. It all depends on the design of the sensor and ADCs." So, aside from some DR differences M9, M240, D800 seem to behave similarly with regards to ETTR, no? Link to post Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.