Suppose you have read somewhere that the dynamic range of your camera at a certain ISO setting is 11 stops. And here comes the immediate question – how can one use such a treasure to its full potential? Optimal exposure for RAW is the answer. But now we need to explain what we mean when we say, “optimal exposure for RAW”. Let’s start with one of the problems, which arises as a result of non-optimal exposure for RAW. Here is a typical wide dynamic range low-light scene. According to Sekonic spot-meter, it is wider than 11 stops:
We made two shots of the same scene – one (#4152) was exposed according to the in-camera exposure meter set to matrix metering mode. The other shot (#4153, Figure 1.) was exposed 1 2/3 EV hotter (spot-metered from white dome, for the procedure to determine the amount of correction please read on, or you can simply bracket). Matrix metering seems to have gotten fooled by the specular highlights on that white dome, hence it underexposed a little more than usual.
We opened them both in FastRawViewer, applied Shadow Boost to both of them to open the shadows, and applied +1.67 EV exposure correction to shot #4152 (the one that was exposed according to in-camera exposure meter recommendation, so that both of the shots had the same overall brightness).
As we can see on to the RAW histogram for shot #4153 (Figure 2) the capture definitely has more than 11 stops. Now we look at shot #4152.
As we can see, shot #4152 exhibits significant noise everywhere below midtone. The thing with a camera’s dynamic range is that those exciting numbers become valid only when the signal reaches the maximum; that is, the exposure is the hottest possible for the given camera at a given ISO. Imagine a bucket having a volume of 2 gallons. If you do not fill it up to the top, some volume is wasted. It’s the same with a camera. If your whitest white is not exposed to the maximum, the top portion of the dynamic range is not being used. This immediately means that in order to use the dynamic range of the camera to its fullest and simultaneously to have as little shadow noise as possible, the whitest whites where you want to keep some texture details need to be exposed so that they are just below clipping. In a sense, one can say that dynamic range characterizes the depth, and starts from the top, while the bottom is sort of fuzzy because of individual tolerance to noise, viewing conditions, size of the output, etc.
Back to our bucket analogy – if one tries to pour, say, 2.5 gallons in it, an overflow happens, and half a gallon will end up on the floor. In digital photography we know this as highlight clipping. That is, if we try to expose a pixel hotter than its capacity allows, everything that is above the capacity will pour out, causing blown-out highlights. By the way, if one tries to set exposure low to keep from clipping all the light sources and specular highlights, he is, most probably, also wasting a couple of stops of dynamic range.
Unfortunately the metering systems of our cameras are not designed to provide the proper placement of the highlights right out of the box (this is partially because to do so, the camera’s AI needs to make a guess as to what a photographer considers to be “important highlights”, and if it guesses wrongly, it may arrive at an unacceptable exposure). One can try different metering modes with the camera, with mixed results when it comes to optimal exposure for RAW. You can use RawDigger or FastRawViewer to see how close you are; be prepared for more misses than hits, in part because the complex metering modes are optimized for JPEGs rather than for RAW. Reliability and repeatability, however, come with manual metering in the spot meter mode. While placing important highlights at the top, let’s take advantage of one very well-known and stable feature of the camera: cameras are calibrated in such a way as to provide the exposure that will result in the rendering of whatever was under the spot meter as “middle grey”.
For a JPEG middle grey is defined as 118 RGB in sRGB color space, which corresponds to the proverbial 18% grey (and L* = 50 in Lab color model, too). For RAW, middle grey is not fixed by common standards, and camera makers have some freedom while placing it. Why is this so? Modern cameras have relatively low noise, and customers often complain that the highlights in their exposures are blown out. To provide extra room for the highlights, camera makers tend to calibrate the exposure so that it shifts middle grey down to 13% (instead of 18%), and often even lower than that. This is what often causes the rumors that camera makers cheat with ISO. In fact it is not cheating, because in most cases the ISO speed is defined for the processed output (JPEG) and not for RAW. Lets look at the shots of the scene we have already used in the article “Why Bother Shooting RAW if Culling JPEGs?”.
Shot #1861 was exposed according to the in-camera exposure meter and it has the hottest possible exposure for JPEG. Shot #1866 had exposure increased by 1 2/3 EV, and it has the optimal exposure for RAW (Exposing to the Right / ETTR). If we look at the shadows of those two shots after running them through ACR (shot #1866 was converted without any adjustments; while shot #1861 was converted with +1.67 exposure correction applied in ACR, to “match” the exposure difference between the shots), we see that there’s quite a difference in the shadows between these two shots. By the way, the smaller the sensor is, the more difference optimal exposure will make.
A tone curve applied when rendering a JPEG takes care of matching of whatever the middle grey is in RAW to the standard 18% grey in JPEG. But if we are shooting RAW and trying to exploit the dynamic range of our cameras to its full potential we need to set an exposure which is close to optimal, and place the highlights in the RAW where they belong – at the top of the dynamic range of your camera.
In other words, the top of the dynamic range of the scene (and by scene we mean everything you want to capture, and not the irrelevant / specular highlights) needs to be placed at the top of the dynamic range of the camera.
Back to the question of achieving this goal in a reliable and repeatable manner:
- First, we need to know the value for the middle grey in RAW numbers. This varies by camera model and sometimes by ISO setting.
- Next we need to know the maximum value that the RAW file may contain. This also varies by camera model and may depend on ISO setting; it may be the same as the maximum value for ADC (12 bits, 4095; 14 bits, 16383), or it may be less.
- Finally we need to calculate the result – the number of stops between the middle grey in RAW and the maximum in RAW for this specific camera and given ISO setting, to use this number as the exposure compensation when metering off the highlights that we want to keep.
Again, to use the result we need to meter from the whitest white in the scene where we want to preserve some texture and details, and apply exposure compensation in the camera that equals the number we have just determined. Consequently, the important whitest whites of the scene will receive the maximum allowed exposure, thus being placed at the top of the dynamic rage of the camera. This way you have achieved optimal exposure, also known as ETTR. The whole thing hinges on the theory that whatever is presented to an exposure meter will be recorded as middle grey as soon as the exposure meter recommendations are followed. Let’s check this theory.
We made six shots of the ColorChecker (Figure 1):
For each shot, we spot-metered one of those neutral grey patches at the bottom row, starting from the white patch, setting the exposure according to the recommendations of the spot meter. Next, we opened those shots one by one in RawDigger, sampled the patch we spot-metered, and checked the RAW values for this sample.
As you can see, the RAW values for the patch under the meter are very close to each other:
Because the camera was set to record both RAW and JPEG, we can check the RGB values in Photoshop too. As expected, they also turned to be very close to each other and to the target value for the mid-tone for JPEG, RGB=118.
Now that we are confident that the spot meter indeed guides the exposure in such a way that whatever shade of grey we put under the spot meter, the result will be the same number in RAW, with the number representing middle grey in terms of camera calibration for RAW. Because the theory proves to be true, you don’t need to meter and shoot all 6 patches: one is enough, or you can use a grey card.
- To get the most accurate reading for the middle grey in RAW, it is more practical to use a light grey patch / card (ColorChecker Passport has a separate grey page for white balance, it has L*=81, or, in other words, it is 60% grey, bright enough to use with confidence) or matte white. The reason for using lighter tones is to keep flare as low as possible for both metering and capture. The higher reflection values of the lighter samples are less prone to flare compared to lower reflection values for dark grey and black samples.
So to get GMiddleGrey, make a shot with exposure setting according to the spot meter recommendation for a white or a light grey patch, open the shot in RawDigger, place a sample on the patch from which you metered, and read average value for the Green channel, Avg G. In our case, GMiddleGrey = 1435.
- Now let’s determine the maximum value for the G channel that RAW can contain (maximum values may be different for different channels).
There are basically 2 ways to determine the maximum:- One way, which does not require additional shots, is to include from the beginning some source of relatively small specular highlights in the scene (a shiny metal ball, for example, is what is often used); however, this can create flare in the shot, and that flare will throw the metering off; maybe just a little, but sometimes significantly.
- The other way is to simply shoot with long exposure, ensuring that part of the shot reaches clipping.
The truly clipped region will usually have a very small sigma (σ, standard deviation, Std.Dev.) value, in the order of single digits, in many cases very close to 0. The mean (average, Avg) value of the green channel for the clipped region is the maximum raw value we are after. On Figure 5, the average for green (Avg G) is 14,594.4; the standard deviation for green (σG) is 59.4. This means that there is no clipping.
Figure 10 – RawDigger. Shot #2254. Large sigma (σ) values for all channels On Figure 6, the average for green (Avg G) is 15,478.7; the standard deviation for green (σG) is 2.03. That is the G channel clipping all right, and increasing exposure further will not help decrease σG in any significant manner – see the next Figure, #7.
Figure 11 – RawDigger. Shot #2255. Small sigma (σ) value in Green (and Red) channels On Figure 7, the average for green (Avg G) is 15,477.3; the standard deviation for green (σG) is 1.81; that is, the green channel G is clipped, like it is on the previous one:
Figure 12 – RawDigger. Shot #2256. Small sigma (σ) values in all channels Nevertheless, the Avg G from the previous shot, #2255 (Figure 6) is a better choice: as you can see the Avg G is larger here, and this is because we triggered intense anti-blooming on shot #2256 (Figure 7), and it acts like solarisation, also known as a Sabattier effect. On film, increasing exposure results in decreased density of the negative; however, it is much milder on digital.
So, we will consider Avg G for a sample area from shot #2255 (Figure 6) as maximum raw value: Gmax = 15,478.7
- Finally, let’s calculate the value of exposure compensation (the number of stops, EV, between the middle grey in RAW and maximum / clipping point in RAW) for the given camera at a given ISO. We are going to use the formula:
EV = log2(GMax / GMiddleGrey) (1)
If you are interested in what the corresponding percent of grey is, you can calculate it as
percent = 100% * GMiddleGrey / GMax (2)
In our case GMiddleGrey = 1435 and Gmax = 15,478.7, so using the formula (1) we get:
log2(15,478.7 / 1435) ≈ 3 1/3 EV
That is this value of exposure compensation turns to be +3 1/3 EV (meaning the calibration for our camera puts the mid-tone in raw to 9.27%, nearly a stop lower than the “expected” 18%).
Let’s shoot with calculated exposure compensation and see… For the next shot #2257, we metered off the white patch, applying to the meter readings the exposure compensation we just determined, +3 1/3 EV. Now we open it in RawDigger and check raw values of the white patch:
As you can see, for a white patch average RAW value for Green channel (Gavg) is 14592.5. That is we are very close to the clipping point but still below it.
If we look at the JPEG of the shot #2257 (Figure 9) we will see that we “overexposed”: R=G=B=255. That is, of course, not so. OOC JPEGs do not use the full dynamic range available in RAW.
If you hate calculations, you can use a brute force approach. Shoot some neutral patch or grey card, setting the exposure as per your camera’s spot-meter recommendations and start adding exposure compensation, EC (you can set EC initially to +2 1/3 EV, which corresponds to calibration of the mid-tone in raw to 20% grey; we suggest this because calibration shouldn’t be higher than 20%) until the raw values for the patch or card comes close to clipping, within 1/3 EV from it.
Incidentally, you can often hear advice to meter off of the 18% grey card to set the correct exposure. This exposure is not likely to be optimal for raw, as in most cases it does not place the highlights high enough on the DR scale, but let’s check it.
Shot #2253 was metered from the N5 patch of ColorChecker (which by design has L*=50, or, in other words, it is 18% grey) with zero exposure compensation. As you can see in RawDigger, the white patch is recorded 1 stop below clipping, and we have not achieved optimal exposure (Figure 10). Not even close.
If we check OOC JPEG for the same shot (Figure 11) we will see that while middle grey was correctly placed at 118 (sample 2), the RGB values for the white patch (sample 1) are (220, 220, 225), that is, seriously below maximum value.
Sekonic 758 in incident metering mode recommended 1/4s for shutter speed, matrix and center-weighted metering in the camera suggested 1/6s and 1/5s, respectively, while the shutter speed for ETTR, as we already know from shot #2257, is 1/2.5s. All out-of-the-box metering turned to be inefficient in terms of ETTR, but very consistent with the ≈1 EV difference we observed between calibration for raw for this camera (≈9%) and calibration for JPEG (≈18%).
We are very happy to hear that a new camera has ½ of a stop more DR than a previous model. We consider those ½ stop to be a huge improvement. However, if we are not using ETTR, we will be getting from this new camera the same or even lower image quality an ETTR practitioner is getting from the previous model.
Hi Iliah,
I’m sorry, but having read this, I’m not clear on the ‘so what?’ in practicals terms . In other words, what does this result mean you’d do differently in the field?
Let’s say I can’t be bothered with the maths (which I can’t) but I’m happy to shoot my grey card using the ‘brute’ approach you describe. If my ‘calibration’ (using centre spot metering mode’ is +2 1/3 eV and I find I can push it to +3 2/3 eV without clipping (difference = 1 1/3 eV) does this mean that, practically speaking, if I spot meter and lock exposure off the brightest part of my image in the field, I should be able to set set +1 1/3 eV in EC without clipping?
Sorry if my ideas lack refinement. I’m just trying to see if I can juice my sensor for more range in a practical way. By the way, I always shoot in RAW so I’m not concerned with the JPEG comparison.
> I can push it to +3 2/3 eV without clipping
> if I spot meter and lock exposure off the brightest part of my image in the field, I should be able to set set +1 1/3 eV in EC without clipping?
It is full +3 2/3 in this case, the full amount you’ve determined.
Hi Iliah, very informative ! Although the calcualtions may be bit tough to keep in memory, yet a very nice write up. I would like to know exactly what makes the difference that many reviewers say that a particular DSLR ,say Nikon D810 has a better dynamic range in comparison to say D 750 ? How much useful would more or better dynamic range prove to be ?
It is not very useful to speak of dynamic range without taking print size (and other things pertaining to viewing conditions, like viewing distance) into consideration. I do not know how the reviewers you mentioned arrived at their opinions, but D750 and D810 are very close when it comes to dynamic range, up to the point where one needs to upscale D750 image. In practice, tiny differences in dynamic range are less important than good exposure technique.
“IN PRACTICE, TINY DIFFERENCES IN DYNAMIC RANGE ARE LESS IMPORTANT THAN GOOD EXPOSURE TECHNIQUE.”
Excuse the block capitals, but this statement needs to be tattooed on some photographers’ foreheads.
So many obsess about technical minutiae while having no grasp of the fundamentals.
Hi Iliah, I read that in the DP Reviews ( Digital Photography Review,considered to be trustworthy) review & comparison between Nikon D 810 & D 750. They said that ‘if you want more dynamic range the opt for D810…..something like that ! Anyway ,I would like to know, how do I take an instant metering decision while using Spot Metering , when I don’t have time to check with blinkies in histogram?….situation may be for e.g, cloudy sky with landscape & when shooting portraits under strobe lights like Elinchrome?
If you have clouds, you can meter off the white cloud. In a studio I check the light setup and set the exposure before the actual shoot.
You can also use Continuous Low or High and do a 3-shot bracket (-1,0,+1) with something like +2/3 EC. Net result: -1/3, +2/3, +1 2/3. You can experiment with various bracket/EC values in order to determine the best combination to use under specific shooting and lighting conditions.
1 stop difference may be a little too crude; but of course one can use bracketing and select shots based on raw histogram.
Thank you for this most informative article. This explains quite clearly an issue that I’ve been struggling with over the past couple of months. The issue can be simply described in a question…”Why is the image on my LCD about 2/3 to 1 stop underexposed compared to the RAW file?”. Many of the articles on the net simple instruct you to disregard the LCD as an exposure instrument and instruct you to use the ETTR method. Many of these articles also instruct you to use the histogram to determine exposure and to avoid clipping the whites. These articles helped somewhat, but in now way answered the question for me. Being quite a technical person, your article not only explained the problem and therefore answered the question, but also provided me a solution. Thanks again and I appreciate the articles and website greatly!
Just a correction to the question I posted above…. I meant to say: ”Why is the RAW image about 2/3 to 1 stop undererexposed compared to the image on my LCD screen?
Try reducing the LCD brightness if you want it to “match” the actual exposure. The histogram is a much better tool than the LCD for judging exposure.
With respect, the LCD brightness adjustment is to make viewing a preview image more comfortable under different ambient lighting conditions. Trying to assess exposure by looking at an LCD display is just plain silly.
Trying to assess exposure by looking at an LCD display is akin to trying to assess the sound pressure level in a nightclub by using our ears.
Hi Iliah
Sorry to be pedantic, but I think you’ve got caught up with some word association as a result of your travels to Yosemite. The sensor on the spot meter is covered by a dome, not a half-dome!
Thanks for all your great articles.
Mike
Thank you, Mike. Somehow I contracted “hemispherical dome” to that “half-dome”.
For those who don’t/won’t use a spot meter-rant warning you should stupid- could we turn on the histogram manually expose to the right and over/underexpose as you said at the top of the article.
It reminds me of my 4×5 film days where we calibrated for the zone system it was about the relationship of the hand held meter tied to your dvelopment chemestry and development times. Less tonal range = just enough additional development time to increase contrast. More tonal range = less development time.
The related question is where to place the highlights or shadows calibrated to your system for the tonal range you want in your final image – this is what pre visualization in the zone system was all about
Perhaps you could add a class on this subject to the courses you offer.
Great article.
> could we turn on the histogram manually expose to the right and over/underexpose as you said at the top of the article.
The spot-metering procedure relies on the fact that highlights are usually neutral or close to neutral; with the histogram, it is much less predictable. The difference may be as large as 1 2/3 stops, as we had it on the shot on the beach.
@WallyinAustin
“The related question is where to place the highlights or shadows calibrated to your system for the tonal range you want in your final image – this is what pre visualization in the zone system was all about”
There is no zone system previsualisation in RAW digital capture, so no class is necessary.
First you capture as much data as is possible short of saturating the sensor.
Once you have the data you rearrange it (and thus the tonal range) in post process to your heart’s content to correspond to whatever visualisation you want.
You appear not to have understood the article – previsualisation of the kind you allude to belongs to film exposure and has no relevance in digital photography.
Thank you Iliah for an excellent article – long overdue if I may say!
It finally puts the nail in the coffin of those who have consistently refused to understand that the exposure meter is an accurate, but very dumb instrument, essentially calibrated (from the days of film) to mid grey which takes no account of the fact that digital exposure is linear and, by extrapolation, that ETTR is the only way to optimise digital exposure.
In fact, as you say, the sensor is capable of even more efficient capture if the additional steps to side step the JPEG interpretation, which you describe, are taken.
It also gives the lie to those who persist in clinging to the old nonsense that the in camera appearance of the image is, in some way, a personal, artistic or creative choice. It isn’t. It is, as you clearly point out, just throwing away valuable image data with all the penalties that accompany that ‘choice’. The best RAW capture is ironically likely the one which looks utterly horrible in the camera LCD – both too bright and too flat.
It also demonstrates that those who rely on any form of auto exposure without intelligent compensation (or even manual exposure in ‘mindless mode’ ) are fated to suffer from consistently inaccurate exposure and that all those techy arguments about dynamic range, sensor noise and image quality are rendered ridiculous without an understanding of how digital exposure actually works.
Spot metering highlights is perhaps a counsel of perfection for many photographers, but the underlying message holds true for all.
Exposure as far to the right as possible (just short of clipping important highlights) is the one and only way to get the best out of a digital camera.
Thank you for putting the spotlight on this misunderstood subject.
I think part of the counter-argument would be the time needed to make good use of that extra range may not be worth it for most images but ETTR kinda forces that. For some people and some images it’s certainly worth extracting the max whether it’s this technique or something like UniWB. For most users, and I’d say quite the majority, good enough is good enough. And when you factor the low DR methods that images are viewed (paper, prints, screen) its value is undercut.
Note, I’m not saying there’s no value and I’m certainly not saying it’s wrong. I’m just saying that for most images for most people for most outputs it’s not worth the time to maximize the exposure for what is usually a minor to moderate gain.
“I think part of the counter-argument would be the time needed to make good use of that extra range may not be worth it”
It’s not ‘extra range’ you are losing, it’s base exposure.
The first stop of less than optimal loses you a full 50% of the data right there and then.
I would argue that basic ETTR is not too much trouble for anyone – in fact it’s easy – just shoot a test frame and check the blinkies. If the blinkies are on highlights that you want detail in, back off the exposure until they (only just) disappear and if there are no blinkies check the histogram and increase exposure until they (almost) appear.
Or meter off a blue sky and add +2 stops or +3 off a white sky – not perfect, but you are in the ballpark. Check for blinkies.
That approach will only lose about a 1 stop possible exposure – that’s probably good enough if you can’t be bothered with inputting a calibration value in your meter.
“I’m just saying that for most images for most people for most outputs it’s not worth the time to maximize the exposure for what is usually a minor to moderate gain.”
I would have to disagree – 75% (or more) is hardly a ‘minor to moderate gain’.
1 stop underexposure throws a full 50% of your data in the trashcan. The next stop loses you half of what’s left – that would be another 25%. And so on. By the time you’re 3 stops underexposed (quite common), you are down to 12.5% of your original data. When you normalise the histogram in post to get an acceptable brightness, you may have only about 6 levels to play with in the shadows. Hello banding and posterisation.
That’s the problem with, and the beauty of, digital exposure. When it’s good it’s really good, but when it’s bad there’s no hiding place. Rather like the good old days with Kodachrome 25. Well, sort of.
That is also assuming that there is something useful in that area that you actually want in your pictures. 75% sounds a lot but it’s highly dependent upon the scene how much of that is actually data you *want* that is actually in there. Again, I’m not saying there’s no advantage but I think for most people the relative gain does not make up for the time required in processing and attention to detail. For those who want that extra it’s great. And thinking of all the people who love to crush the black tones anyway…
@Neil
“That is also assuming that there is something useful in that area that you actually want in your pictures.”
What ‘area’ are you referring to?
Your statement makes no sense to me.
When you underexpose by say,1 stop, you lose 50% of all the data – the whole histogram is pushed to the left – highlights, shadows and mid tones are all lost equally or, more accurately, half the data fails to record across the board. It is not a selective process. Digital exposure is linear. There is no choice about what part or ‘area’ of the data you want or don’t want. It’s simply not there (or rather it’s depleted) because half of it it didn’t get captured.
Oh yeah, I forgot (since I’m not on this forum as much as in the past), it’s not worth getting into discussions with some of the people here since they’re always right. In the mean time it makes 0 difference to my enjoyment of photography and 0 difference to my results. No doubt you have many zingers aimed at any opinions or counter-points I would offer but I’m in no mood for it so I’ll leave it for your imagination and, no doubt, witty rejoinders.
Neil
“…it’s not worth getting into discussions with some of the people here since they’re always right.”
If Pete A, Iliah and Nasim are “always right”, you might perhaps try a change of tack and take the opportunity to ask intelligent, probing questions, discuss their answers and perhaps even learn something of benefit to your photography – instead of trying to peddle your own personal cocktail of pseudo scientific quackery, outdated dogma and photographic folklore as some kind of divine revealed truth?
“In the mean time it makes 0 difference to my enjoyment of photography and 0 difference to my results.”
It’s made zero difference to your results because you have learned nothing – and having learned nothing, your results can be nothing other than the same. Zero in – Zero out.
…And it’s made zero difference to your enjoyment of photography because ignorance (if accompanied by a sufficient measure of complacency and self satisfaction) can be bliss.
Betty, you have some issues… Where is your aggressive tone coming from?
David
>>Betty, you have some issues…<>Where is your aggressive tone coming from?<<
Talking to smart asses who, when confronted with the inconvenience of evidence-based fact prefer to cling to long embedded, irrational beliefs by inventing ever more absurd and deluded ways of dissipating their cognitive dissonance.
Also, low self-esteem, a hyper-inflated sense of my own importance, late nights on the internet and getting out of the wrong side of the bed in the morning. But mostly I blame my parents.
Betty, have you tried counseling? It helped me in the past. The challenge is to find a good psychiatrist or psychologist, preferably one who is experienced with cognitive behavioral therapy.
Harvey
>>Betty, have you tried counseling? <>It helped me in the past.<<
From an objective standpoint, I can see no evidence of that…Harvey.
Harvey
>>Betty, have you tried counseling?<>It helped me in the past.<<
From an objective standpoint Harvey, I can see no evidence of that.
?
Thanks for your knowlageable input Betty… Great tips for sure…
Hi Betty
Always happy to read your comments. I have a question about theses statements of yours.
“Or meter off a blue sky and add +2 stops or +3 off a white sky – not perfect, but you are in the ballpark. Check for blinkies.”
Its the last statement that I don’t understand “Check for blinkies”. Why check for blinkies on the jpeg histogram ? The highlights on the histogram of the camera aren’t they under-exposed for achieving ETTR ?.
regards
Luc
I say check for blinkies so that the highlights *don’t* get clipped.
In other words, if you see blinkies, you reduce the exposure until they just disappear from the highlights *in which you want to retain detail*.
If say, it’s a bird with a little white wing patch, you would not want to lose detail in that. If however you only saw blinkies in a patch of sky in a corner of the frame which you were going to crop off later anyway and the bird’s feathers were OK, you might choose to let them go.
The main source of noise is underexposure. Not the camera, not the sensor, but underexposure. If one is not exposing “to the right” while recording raw files, he is underexposing, plain and simple. Correct exposure is not that difficult, and with digital we often have an option to bracket, too.
By the way, what is good enough for us may be not good enough for grandchildren in case they will want to process those raw files, and print or display them large ;)
At the light levels that suggest ISO 1600, 2/3 of a stop of underexposure results in very visible difference in noise even with a high-end FF camera like Nikon Df, making an underexposed image not good enough if one needs to open the shadows.
The dynamic range of the black and white paper used to be 7 stops. Why would Adams suggest 11 zones? It’s up to us how we map input to output, how we use those 11 stops.
The article is an attempt to help folks who feel they have not enough dynamic range in their cameras, and to suggest some metering strategy that may be used to address their issue.
Hi Iliah! I’m interested in hearing what you think about my solution to this problem. I just use a mirrorless camera with an electronic viewfinder (e.g. Sony a7R) that permits me to see the histogram BEFORE taking the picture. I use the exposure compensation dial to shift the histogram up against the right edge, take the shot, and I do not even have to look at the histogram (or the exposure) afterwards! Jacek
As the picture which is shown to you in the electronic viewfinder relies on your jpg-settings you will get a perfectly exposed jpg ooc – but the raw file won’t be exposed to the right – depending on the jpg-Settings your raw will be underexposed by 1-2,5 EV.
To minimize this problem you could either configure a very flat jpg-style (but your picture in the viewfinder will look horrible, too bright and too flat… or you can try to take a number of pictures with your normal jpg-Settings adding more and more exposure compensation. Opening the files in your raw-converter afterwards (no automatic picture style should be active) will show you how much more headroom the raw has compared to your jpg. Let’s assume 2 EV… from now on you know that if the zebra in the viewfinder start to show you are still able to add around 2 EV more exposure
It’s no different than looking at Live View or seeing the jpeg after the shot with a DSLR. If you want to use jpeg to approximate the raw histogram look into a UniWB approach. It’s much closer and you definitely can’t trust the image displayed in the VF.
Neil
Have you tried UniWB with wildlife. sports or BIF? Tricky.
I tried it based on Thom Hogan. It’s not fun but it’s possible. Then I realized that since I’m not a pro I don’t need to go through all those machinations since it’s only for personal satisfaction. That and a $1 will buy me a soda.
Actually I don’t know any pros who go through this rigamarole. Taking many extra steps in order to make sure you have as much shadow detail as possible, whether or not you need it, is more of an amateur technique.
Maybe Ansel Adams had already answered this.
Landscape and some kinds of studio work aside of course – these are by their nature slow processes.
@Andy from Boston
Have you considered that the only slow process going on here might be the one in your head?
This is what happens when people spend more time doing math on web sites than shooting. You get jerks who can do math and insult people, and have no idea how to work in the field.
Of all the pros I know, who are mainly shooting weddings, events, PJ and portraits, none of them use this technique. The only times I see this used with good effect are in limited landscape and studio work, and by fussy amateurs. Make whatever rude comments you want, that’s the truth.
Not only you are wrong in so many ways, your language is totally inappropriate.
@Andy from Boston
Well Andy, I have never been good at math, but it doesn’t take a great mathematician to apply a +1EV compensation before shooting does it? You just have to know that it’s a sensible thing to do and why you are doing it.
Also, you have no idea what other photographers are doing. The ones who know what to do are quietly doing it, while the ones who don’t are unaware of what they are losing.
And those who have just found out, either embrace the opportunity to learn and improve or sink into a state of denial and anger and steadfastly cling to their ingrained dogmas.
It’s a choice.
You can take it or leave it.
Andy from Boston wrote: “You get jerks who can do math and insult people, and have no idea how to work in the field.”
You are an exemplar of the jerks who are far too intellectually lazy to “do the math[sic]” and have, instead, chosen the modus operandi of insulting those who fully understand the mathematics of the relevant branch of science.
Mathematics is a formal descriptive language, it is not a science per se, that underpins the whole of science. Your camera and lenses would not, and could not, exist without the scientists and engineers who have dedicated themselves to learning their art and craft from first principles.
You have made it blindingly obvious that you have not learnt your art and craft from first principles. Shame on you.
@Andy from Boston
This kind of statement generally comes from those who know little and care even less. It’s just complacency.
Striving for excellence is not a selective process. You are either interested in improving or you are not.
Inputting a simple compensation value is hardly “many extra steps”.
Arguing at infinitum about the (usually small) differences in dynamic range of various cameras while ignoring the best way of achieving it is laughable.
The mismatched ratio between the forcefulness and informativeness of your statements notwithstanding, we have different methods of evaluating excellence. You believe excellence occurs when you max out a camera’s “stats”. Evidently with an emphasis on shadow detail, at the cost of highlight detail. I don’t think there’s much excellence to be found in that, except in some rare situations such as certain very constructed and deliberate studio work and some landscape scenes.
@Andy from Boston
“You believe excellence occurs when you max out a camera’s “stats”. Evidently with an emphasis on shadow detail, at the cost of highlight detail.”
You are putting words in my mouth, so to clarify..
If by excellence you mean technical excellence, then yes, this can only be achieved by extracting from the sensor all the data it is capable of collecting. That kind of excellence is evidence based and irrefutable.
However, if by excellence you mean photographic/artistic or creative excellence, then no, no amount of maxing out data will help that.
That kind of excellence is based on talent which in turn, is a matter of opinion.
If you believe that fullest possible digital exposure or ‘maxing out the stats’ as you put it, in some way ’emphasises shadow detail at the cost of highlight detail’, you are mistaken and clearly still do not understand the basics of digital exposure.
I suggest you read Nasim’s article again.
To be clear, a good digital exposure maximises both shadow and highlight detail.
And yes, I do think there is a great deal of excellence to be found in that.
Suboptimal exposure on the other hand loses information, and hence detail, across part/the whole tonal range.
Overexposure saturates the sensor (clipping) and thus irretrievably loses image highlight information/detail.
Underexposure fails to capture all the data that is available and thus also irretrivably loses image information/detail – from shadows, midtones and highlights.
And so no, I don’t think there’s much excellence to be found in that.
“I suggest you read Nasim’s article again.”
Oops, apologies to Iliah.
Absolutely no problem. The subject is more important than the name under the text in any case.
To be clear, I don’t believe in a correlation between shadow detail and excellence. And I think that articles like this are likely to have a negative effect on beginners who may be fooled into thinking that they should be doing this type of analysis instead of practicing photography, or intermediate shooters who don’t get it well enough to know that ETTR should only be used when you need to bring in more shadow detail or can do it while still shooting at base ISO. (Spencer Cox wrote a good article a while ago that made that point.)
Shoot a slant edge using ETTR and your normal routine. Load both shots into MTF Mapper and compare resolution. Next, shoot some grey step wedge using both procedures and measure the noise in midtones. That may help you see that proper exposure is not just about shadows, and in many cases it has nothing to do with shadows at all.
@Andy from Boston
On the contrary, beginners are the very people who benefit most from learning sound photographic principles – infinitely preferable to listening to entrenched, negative, old die hards who can’t let go of their ingrained, film based, belief systems.
Nobody is suggesting that every beginner should replicate the steps carried out by Iliah in this article.
But shooting a series of exposures (just once) from a grey card to establish a RAW clipping point and setting an appropriate compensation is hardly beyond the abilities of even the rawest (excuse the pun) beginner.
If you still think that ETTR should only be used when you need to bring in more shadow detail, then there is still much that you do not understand.
ETTR is almost universally applicable, indeed essential, if the fullest range of tones with maximum detail and minimum noise is be achieved from a digital exposure.
> with an emphasis on shadow detail, at the cost of highlight detail
And why would that be at the cost of highlight detail?
Correct exposure for raw is not about details in shadows, it is about noise control. The lower is the exposure, the more noise it contains. Compare 2 exposures of a typical 8-stop wide scene, one exposure is “to the right”, and one exposed 1 stop lower. What is the difference in noise between those two exposures? Approximately half a stop. This does not depend much on a sensor, or on a camera. That is the property of light, called photon shot noise.
Half a stop of noise is negligible in real world results at base ISO. You only see it when pixel peeping. Even then it’s usually not an unpleasant effect until you get into high ISO. At more than base ISO, ETTR should be avoided because increasing ISO to increase exposure will offset any noise reduction you might have had from using ETTR.
>Half a stop of noise is negligible in real world results at base ISO
Depends on how large you print, and what you are going to do with the image.
As to your objection to ETTR with higher ISO settings, it is completely wrong.
1. there are cameras where increasing ISO results in decreased noise. All current 1-digit cameras from Canon and Nikon are this way.
2. Increasing ISO does not increase exposure, exposure is light + shutter speed + T-stop.
3. With ISO-less cameras, increasing ISO does not increase noise (it clips highlights). Think of this: why a sane engineer would put an amplifier into the camera if that amplifier makes things worse?
@Andy from Boston
I do understand where you are coming from to a degree – ISO variance/invariance and the Law of Diminishing Returns etc? Forgive me if I am wrong.
There is some truth in what you say, but to get the benefit of full exposure, the base exposure has to be ETTR with meter compensation as well. Yes, changing ISO from that point is just signal amplification (image brightening) rather than a true change in exposure, but that does not take away from the principle. Simply underexposing and cranking up the brightness in post process is not an answer. It’s just sloppy technique and carries penalties.
Knowledgeable photographers are aware that there is also a penalty to pay when moving from base exposure at native ISO in order to satisfy the requirements of higher shutter speed or greater depth of field, but that’s a mindful choice, not an assumption that you can do what you like and it will make no difference.
Also ISO variance/invariance is still something of a minefield in that manufacturers are not very forthcoming about the characteristics of their sensors, so cameras may be ISO variant, invariant or partially variant. In addition this variance may occur in a step wise manner at different ISO settings (and even individual camera bodies vary), so trying to do what’s best becomes rather impractical ‘in the field’. Trying to work that out really would be “taking many extra steps” as you put it.
I will stay with ‘tried and tested’ until someone proves that another way is better. Then I will change.
@Andy from Boston
“Actually I don’t know any pros who go through this rigamarole. Taking many extra steps in order to make sure you have as much shadow detail as possible, whether or not you need it, is more of an amateur technique.”
1. Number of steps required = 1. Input the compensation to your meter and expose to the right.
2. It’s not just about shadow detail – it’s about all the detail. It affects the shadows most because these are recorded with the lowest number of levels, so the more data you capture the more levels you will have to play with in post.
3. If it’s an “amateur technique”, what constitutes professional technique in your view – point and shoot in P mode?
All this would be so much easier if only the camera manufacturers gave us the option to see a RAW based histogram, not one from the cooked JPEG that the camera produces. Sigh …
I’ve thought that for years but now I think it might not be that easy. All sensors have 1 sensitivity and are either using a gain or digital multiplier to approximate other ISO values. What would the raw histogram show? Would it only show the native sensor ISO no matter what? That could be misleading or not even desirable if you’re needing fast shooting. Would it show the scaled values? Maybe misleading as well. Would it be luminance or would be one value for each primary color? I think I can see why the jpeg way is what they display plus almost all photos are viewed in that manner so it’s much closer to the final result that most people take/process/view images.
A RAW histogram with a user defined clipping value would do nicely thank you.
This could even be relatively easily incorporated into auto metering.
Neil,
You would definitely want the three color channel separate (ever try shooting a poppy ? red will saturate quite handily). For ISO you want the ISO that generated the RAW file, there may be analogue gain (before the A to D converter) as well as digital gain (after the ADC), but both lead to “higher” values for the pixels, and that is what we are concerned about here. Of course if you are at high ISO you already struggle with light, so if you increase ISO to ETTR it would not buy you much as you are first multiplying the data (to ETTR) and then in post processing you divide (to get the desired exposure).
As to technical feasibility that should be quite easy, my guess is we don’t get it because the camera companies are afraid of user error or confusion; also most people, even with DSLRs do not shoot RAW but JPEG.
On the shots of flowers red is saturated in JPEG due to white balance, mostly. It is not saturated in raw most of the cases – check the raw histogram with FastRawViewer. The effect of clipping of red or blue channel in JPEG because of white balance being applied in JPEG is well-known. White balance is not applied to raw data in the camera, however; thus in a lot of cases you can rely on JPEG histogram – it indicates false clipping of colour channels.
Magic Lantern firmware for Canon cameras display raw histogram for years, and proved to be very useful.
Very few cameras have pure digital ISO these days, mostly Sigma’s. In any case, it is a simple option in user interface to select histogram display mode.
Hey guys,
I am very sorry for letting this one slip through, but the original article author is the great Iliah Borg, not me – I fixed the mistake earlier this morning! My apologies for this mess, I would never claim to be as knowledgeable in the field of camera sensors as Iliah is!
Sincerely,
Nasim
Okay, here comes a silly question:
In the Nikon system, I have the ability to modify what the camera will display on the LCD via the Picture Controls setting.
I typically shoot RAW so I usually choose the *Neutral Picture Control setting in order to get a better representation of what the RAW file will look like vs the other settings. ( Like Vivid, Standard, etc )
*In the D810 and later, Nikon introduced the FLAT setting which may be an even better choice than Neutral though I haven’t really played with it yet.
Would it make sense to simply use the histogram based on the Neutral or Flat Setting output, then expose to the right as far as possible without blowing out the highlights ?
@nehumanuscrede
Not silly at all,
That’s the best approach short of taking one more step and biasing your meter to compensate for the difference between the JPEG and the RAW rendering.
In addition, you could also set your cołour space to AdobeRGB rather than sRGB. That will make the LCD display look even flatter, but will be a little closer to a true RAW representation.
I have been being using FLAT for this very reason but I didn’t know that setting the color space to AdobeRGB also supports in getting blinkies which more closely show highlight clipping for RAW. Thank you, Betty.
Is there a rule of thumb for D810 14-bit raw?
Or on Sony mirrorless, use zebras at 109+, expose so whatever you want to clip is in the zebras, and set EC to +2.0. The clipping in RAW will correspond to the area in zebras every time.
Hmm, I might have worded that wrong. You want to set the exposure +2.0 over what you set after compensating for the zebras.
Yes, as the zebra shows the overexposure of the jpg you can add add some EC without having clipped highlights in the final raw. Only the mentioned amount of +2 EV is not fixed as this depends quite a bit on the selected jpg-settings (and might vary with different light sources/temperatures)
Yup, it needs Standard with 0 contrast and saturation, though DRO has no impact on the highlights, so it can be set to whatever is needed.
Thank you for this very detailed article. I tried the brute force approach using a Gary Fong grey dome. At +2 EC, there was no clipping but at + 2 1/3 EC minor clipping started to show in Capture One. I noticed that even if I push the EC above 4, Capture One can still recover the highlites. What are the drawbacks for overexposing beyond the point of clipping and relying on the software to recover the highlites?
To check for the clipping in raw, try FastRawViewer, I’m afraid C1 is not showing what is clipped in raw accurately enough, for the same reasons the clipping indication in the cameras is inaccurate for raw.
Highlight recovery in a converter is a two-stage process, first a converter applies a simple linear compensation to brightness, pulling back the highlight information that is present in the raw data; and after that it starts extrapolating the highlights. You want the first, but normally you do not want much of the second stage.
Thank you very much, Iliah.
In addition to the brute force method, I also tried spot metering off the brightest part of the clouds in the sky, then incrementally added exposure compensation. I was able to go as high as + 3 1/3 EC (I will have to try this in FastRawViewer opposed to C1) before any clipping. Is this method another alternative to the brute force approach that one can use? If not, what are its drawbacks?
Thanks again.
> as high as + 3 1/3 EC before any clipping
It is just the same “brute force”, with expected result; but to be safe, please check for clipping with FastRawViewer.
Iliah,
Thank you very much indeed for taking the time to write this highly-valuable technical article, and for taking the time to respond to the readers.
I also thank Betty for taking the time to further inform the readers.
There is something that deeply puzzles me about the comments from the readers who disagree with the science and/or the essence of the article. None of them have yet been able to identify any technical errors in either the article itself or in the replies from Iliah and Betty.
Nasim appreciates my technical input to his website, but in this instance I have chosen to refrain from providing it because I’m eagerly awaiting the day when one of the dissenters demonstrates having acquired a thorough understanding of a science-based subject *prior* to having gained their sense of entitlement to criticize it :-)
Kindest regards,
Pete
Hi, thanks for the article. One question: what can you say about the influence of lenses on metering. Using 1.8 primes there is a tendency to overexpose while slow zoomlenses tend to underexpose. The difference can be nearly a whole stop (usually 1 third to a half stop). This is using matrix metering so there can be a difference in interpretation but if the frame and the scene are exactly the same the difference still exists. Maybe spotmetering is a different story I have to try but for now I have to correct my lenses.
Vignetting could have an influence and would result in an overexposed mid of the frame – but as vignetting is usually more pronounced on fast lenses than on slow ones this would be quite the opposite from what you are describing. What kind of camera are you using?
I am using a Nikon D750. Vignetting can probably influence the (matrix) metering that is a good point, thanks. I think vignetting on fast lenses is less as long as you are not using them wide open. Comparing at F 4 or 5.6 I think the zooms will have more vignetting.
Apart from usual vignetting we have “digital vignetting”. Have a look at Figure 7 in http://www.kodak.com/ek/uploadedFiles/Content/Small_Business/Images_Sensor_Solutions/Datasheets(pdfs)/KAF-8300LongSpec.pdf The graph shows how the response changes with the angle of incidence. The angle of incidence depends on the telecentricity of the lens and is more pronounced with short lenses.
When the in-camera spot-meter is using a point close to the centre of the frame, the readings usually are not affected by vignetting.
Cees
Interesting observation – one which I haven’t noticed mostly because I have never looked for it.
An explanation (besides lens vignetting) which occurred to me…
Zoom lenses transmit light less efficiently than primes. Or, to put it another way, multi-element zooms suffer more internal light loss and so in effect are a bit slower than prime lenses. The F-stop tells us nothing about this light loss caused by light reflection on every lens element surface and absorption inside the lens barrel. The ‘real life aperture’ is the T-stop (F-stop+ light loss in the lens) and it is this value which determines how much light actually reaches the sensor.
I was wondering whether evaluative and matrix meters in inputting so many variables ( luminance, colour, contrast, distance, areas of similarity and difference) fail to account for the difference between T-stop vs F-stop transmission?
I may well be entirely wrong…
Perhaps Iliah or Pete A might cast some light on my darkness?
Betty, When the T-stop is different from the f-stop it affects only manual exposure mode when the exposure settings have been entered from an external light meter (or using the sunny 16 rule). In all auto exposure modes, the camera needs to determine only how many f-stops the lens will stop down from its wide-open meter reading when the shutter is released. The addition of lens filters alters the T-stop, but the camera doesn’t require this information for auto exposure.
NB: There is a caveat to the above with Nikon Matrix Metering, but it is far beyond the scope of this discussion.
I’m glad Cees mentioned his observations because I became aware of it ages ago while using various lenses with Nikon film cameras and high-contrast transparency film. Despite endless web searches, I still don’t know why it happens: I’ve read a plethora of guesses, but no definitive answer.
Pete A
Thanks for that.
I was of course aware that ultimately the exposure set by the camera is down to the exact amount of light landing on, and measured by, the sensor. No different to a hand held meter in that respect.
It occurred to me though, that dedicated pro video lenses are marked in T-stops rather than F-stops to ensure absolute uniformity of performance when changing from one lens to another.
Thus my speculation that there might be some connection between this difference and the way in which the CPU/algorithm (especially with matrix metering) ‘tells’ the camera what exposure to set.
Although wrong, I take comfort from you not knowing either!
In cinematography, it is extremely important to maintain not exposure accuracy per se, but exposure consistency between the various combinations of lenses and cameras — hence the lenses are marked in T-stops, not f-stops.
Forgot to add that lenses for cinematography have to honour their marked T-stop throughout the whole of their focus range and the whole of their zoom range; *and* they have to maintain a constant angle of view throughout the whole of their focus range at each zoom setting (which means that they must appropriately change their focal length as their focus distance is changed).
This is why superb zoom lenses cost hundreds of thousands of £/$!
Wow, great to read the artikel and al the comments.
I’am not a advanced photographer (try to studi and read a lot) when reading the comments follow question came up, maybe a silly one, but anyway:
Is it passible to choose a brand or type camera that have the best solution (for example, Canon + magic latern) and what about Fuji, everybody saying the jpeg are so good, just marketing? or gimmick? or best of both worlds. I now the photographer must do the brain work, but a good tool will help. The questions are a bit of topic, sorry for that, but for me verry helpful!
Thanks!
“3. With ISO-less cameras, increasing ISO does not increase noise (it clips highlights). Think of this: why a sane engineer would put an amplifier into the camera if that amplifier makes things worse?”
I am not sure that I understand the above comment by Iliah. Why would either option of amplifying the signal or clipping the highlights be ideal?
I do not think I said “ideal”.
Amplifying the signal makes sense if amplification reduces noise and artifacts, and though you can take guidance from Bill Claff’s pages on shadow noise improvements, those are only about read noise; and it is better also to check your particular camera sample – make/model do not account for sample variation, and the variation is sometimes significant.
Clipping highlights – I seldom account highlights that are 6 stops higher than the midpoint in raw. That means in low light I shoot setting ISO about 2 stops lower the metering suggests.
Would the noise resulting from an ISO-less camera that underexposes the image compare to the noise resulting from an amplified signal in an ISO camera?
That strongly depends on what cameras we are comparing here. For example, Nikon D2X is an “ISO-less” camera, but I do not expect it to beat Canon 5D original which is “ISO-full” at ISO 400.
This is correct from a technological point. And we certainly need the scientists and engineers to advance the technology of cameras and teach us how to get the best out of the sensors.
But with all due respect for the technologists, from the point of getting the most important photos, image content, emotional impact, composition, timing, expression, and storytelling always are more important. See all the horrid technical flaws of Warren Richardson’s World Press Photo of the year of Syrian refugees passing a baby through barbed wire at night. The flaws don’t matter.
We should keep a sense of perspective here.
@Hanna Becker
With respect, that’s a complete red herring.
“Technologists” created the cameras you are using. So if we try to help you to understand how those cameras work, maybe we know why you need it.
The title of the article is: How to Use the Full Dynamic Range of Your Camera.
It is not: Optimal Camera Settings for Photojournalism.
> Optimal Camera Settings for Photojournalism
Centre-weighted, Manual, ISO Auto, -2EV EC, exposure lock works for me ;)
You owe me a box of tissues: my current box is wiping away my tears of laughter :-)
Iliah
Is nothing sacred?
Why oh why have you given away the secret recipe for correct exposure – known (until you just blew it) only to a few photography and digital sensor research experts?
Now every Tom, Dick and Harry will be getting in on the act.
I’m afraid very few will even try it.
Iliah, I sincerely hope that you didn’t take my comment “You owe me a box of tissues…” to be laughing *at* you; I was laughing *with* you.
I know. But nothing wrong with “at” too ;)
LOL, that’s pretty funny Iliah!
I am confused: Iliah’s response below implies he was serious while the responses of Nasim and others implies he was joking.
Have you tried? Have you read how ISO Auto works?
Iliah, you answered my puzzled comments with a question (two questions, actually). I come to this site looking for straightforward information.
By the way, I have used Auto ISO before, although I am not knowledgeable about the science behind it. I know that Auto ISO automatically adjusts the ISO based on the metering mode set in-camera. Although, if someone were to ask me about the technicalities behind it, I would be unable to answer them.
I still remain puzzled why some are humoured by what appears (at least to me) as serious explanations.
Dear Sir:
I answered your comments several times, with explanations. But this site is hardly a place to quote camera manuals at length. By the way, “adjusts the ISO based on the metering mode set in-camera” is not correct. The adjustments take place based on meter readings, not mode.
Ultimately, one can try and see. It will not damage a camera.
Iliah, aren’t we getting caught up in semantics regarding meter readings vs. meter mode? After all, aren’t meter readings dependent on the meter mode?
Yes, you have answered my questions. I was not looking for camera-manual length responses. I am just puzzled why others here, including Nasim, are laughing as if you are joking when your responses to me appear serious in my eyes. Were you serious or joking?
Dear Sir:
Yes, we are getting into semantics. Semantics, in fact, is the study of meaning.
> aren’t meter readings dependent on the meter mode
ISO Auto knows nothing of the mode. Only the readings matter.
It’s semantics that leads to some of the ludicrous arguments on forums like this.
-Two people neither of whom understand the meaning of the words they are using, fighting to the death over concepts they don’t understand.
The upside is that it is very simple :) One is still in full control of exposure, DoF, shake/blur; but is not in danger of clipping highlights because of ISO amplification. Works even in strongly backlit situation, while noise increase is very tolerable with modern cameras.
Hi Iliah
your comment:
“One is still in full control of exposure, DoF, shake/blur; but is not in danger of clipping highlights because of ISO amplification”
Does it means that at any iso other than the base iso, there is a reduction of light reaching the sensor, and its because of this that there is less danger of clipping the highlights, but at the same time how do you prevent more shadows from being completely black. Is it the reason DR is dropping so much at the highest iso, for any digital camera.
regards
ISO setting: suppose the maximum allowed signal to ADC is 0.7 Volt. At base ISO that is the equivalent of the full usable charge of a pixel. However if ISO is raised 1 stop, that is the amplification is 2x, and now the maximum charge is 0.7 / 2 = 0.35 Volts. That means 1 stop in highlights is clipped. There are variations to this scheme, but the net result (excluding cameras that do not use ISO for raw recording, which are very few Sigma cameras currently) is the same: for each additional stop of raising ISO 1 additional stop in highlights is clipped.
Light is not reduced with ISO going up and given the shutter speed and aperture are the same, it gets clipped.
Black shadows in raw are no problem, in 14-bit mode they still contain plenty of data. Noise is the problem, and noise goes down on some cameras when ISO is upped (important cases are Canons and Nikon current 1-digit cameras excluding D3X). However, the difference in terms of noise between “proper” ISO and ISO 1 or 2 steps lower will not break a reportage or contre-jour / backlit photo, while blown out highlights often do.
One can find a lot of suggestions of different degrees of validity on-line. This one is easy to check. Much of photography is about experimenting.
Thanks for taking the time to answer….
I thought it was: “F/8 and be there”.
Like it!
Hanna
I happen to agree with what you say, but it’s still a red herring.
“The flaws don’t matter.”
Well I suppose ultimately that’s true – ‘the image is everything’.
That doesn’t mean we should accept flaws as a starting point.
And surely the image would have been even better without the flaws?
So the flaws do matter – but can be forgiven when the image is compelling enough.
But it’s actually all irrelevant in a discussion about how to achieve optimal image quality, is it not?
PS
Commenting on another’s post in an open forum is neither badgering nor stalking.
Stalking… really?
You should be ashamed of yourself for applying derogatory and emotive labels to people with whom you disagree.
If you don’t want comment I suggest you put your contribution in an email and post it to yourself.
If flaws don’t matter, what’s the point of making better film, better sensors, lenses, printers, lights, software, colour management,…
At the risk of drawing criticism for what I am about to say, I feel that Betty’s tone is rather condescending to a number of posters. Based on the comments of some other posters, I believe that I am not alone in this sentiment. I feel that Betty has a lot to add in terms of content. If only she can be more polite to other posters who take the time to comment here.
Daniel C. Dennett has more than adequately addressed your issue:
QUOTE
I listen to all these complaints about rudeness and intemperateness, and the opinion that I come to is that there is no polite way of asking somebody:
“Have you considered the possibility that your entire life has been devoted to a delusion?”
But that’s a good question to ask. Of course we should ask that question and of course it’s going to offend people. Tough.
Harvey
Life is not a popularity contest.
It’s not a ‘she’. ‘Betty’ is a guy in the UK. Presumably a deeply embittered guy with a massive chip on his shoulder who seems to spend most of his time trawling through these pages belittling the contribution of others until they acquiesce to his own point of view. I can only guess he feels that hiding behind a woman’s name offers him some protection from reproach but I’m sure he’ll tell me I’m wrong. In any event someone so angry and bitter has my pity.
Kevin
Kevin, I had no idea you had paranormal, gender detecting powers. You should contact Uri Geller and monetise this unique gift.
As for belittling the contributions of others – not so.
What I try to do is challenge disseminators of misinformation, otherwise known as hogwash, to ‘acquiesce’ to evidence based fact and currently accepted scientific principles. No apologies for that.
This site, under Nasim’s banner, is dedicated to sharing knowledge, not reinforcing blind ignorance and outdated dogma. I am sure he will put me right if I have misunderstood.
Disavowing the occasional contributor of long held, deeply cherished, but entirely flawed, beliefs not infrequently causes feathers to become ruffled. Too bad. Have a little chat with yourself and get over it.
Unsurprisingly, applying the pin of scientific enlightenment to the balloon of an over inflated male ego sometimes results in a violent expulsion of hot air.
Unfortunately, “hiding behind woman’s name” has done little to shelter me from reproach – on the contrary, it has served merely to demonstrate how little some men have evolved from wrist trailing Neanderthals.
Kevin, the only people deserving of pity are those who live with their heads stuck in dark places where the sun don’t shine.
“Kevin, the only people deserving of pity are those who live with their heads stuck in dark places where the sun don’t shine.”
Cranioproctosis.
Warning: NSFW.
Don’t be so hard on Kevin.
Proctomancy is a difficult condition to cure and rubbing it in ?could well lead to an exacerbation of this unfortunate affliction.
“Don’t be so hard on Kevin.”
That’s as difficult to parse correctly as is: eats shoots and leaves.
“> Optimal Camera Settings for Photojournalism
Centre-weighted, Manual, ISO Auto, -2EV EC, exposure lock works for me ;)”
Wouldn’t the exposure compensation depend on the reflectivity of the subject matter?
For this type of work, centre-weighted and -2 EV EC does it for me.
It is much easier to accurately estimate the number of bullets in your head, from 0 to 1, than it is to accurately estimate “How to Use the Full Dynamic Range of Your Camera”; and how to accurately estimate whether or not it is wise to criticise those who have not just painstakingly learnt photography from first principles, they are more than willing to share their expertise with the readers of this website.
I strongly suggest that you reflect on the subject matter of this article, and you reflect on your comments appended to it.
Harvey
Iliah knows what he is talking about.
If I were you I really would stick with the recommended settings.
I know that Iliah knows what he is talking about, which is why I was asking him to clarify. I wasn’t challenging him, rather seeking more information on the topic.
Nothing wrong with challenging, too, IMHO. The settings I suggested are a starting point, they are rather robust and leave about 3 stops extra for the extreme highlights (if you ever shoot sweating bald-headed African American jazz musicians in a bar, you know what I mean), but the real underexposure is less than 1 stop, the other 2 stops are “false underexposure”, because of how ISO control works, it’s just extra room in highlights. On modern cameras those 2 stops may bring half a stop of noise max, Nikon D5 or Df, which I prefer for smaller places, result in about 1/3 of a stop of extra noise. Of course I use the fastest aperture I can get away with, and slowest possible shutter speed. Those are the real exposure controls. In less difficult environment I may use EC -1.5, rarely -1.0. If I’m shooting with a flash, it is a very different setup.
Harvey,
The settings Iliah suggested for using a digital camera for Photojournalism are based on the same fundamental principles that I would use to make a digital audio recording of the sound in those situations.
Whenever the dynamic range of the recording medium is greater than the dynamic range of the information that we wish to record and reproduce, only an incompetent fool would set the recording levels too high thereby causing some of the image highlights to clip, and some of the sound to clip resulting in horrible distortion; whereas a competent recording engineer would set the recording levels correctly.
Signal clipping cannot be undone in post-production. Whereas levels that were set too low can be amplified during post-production, and noise reduction applied if necessary.
E.g. a 24-bit digital audio mixing desk and recording system has a dynamic range of 144 dB. If I was using 24-bit digital audio to record the sounds that typically occur in Photojournalism scenarios, I would attempt to set the recording levels such that my guesstimate of the peak sound level that could occur during the recording session are in the region of 40–60 dB below the clipping level of the recording device.
There is no polite way of saying this: You do not understand anywhere near enough about digital photography to know what it is that you do not yet understand.
Whether or not *you* have been wasting *your* time “on silly banter on Iliah’s threads” is irrelevant. *You* have been wasting the time of Iliah, Betty, myself, and the readers. Shame on you!
We have 20 years worth of misconceptions in regards to noise and ISO showing their teeth. The common expression “high ISO noise” needs to be replaced with “low exposure noise”. The common understandings “too dark = underexposed” and “too bright = overexposed” need to be debunked for the case of raw.
Iliah
I have been saying this over a number of related PL articles for longer than I can remember and have been consistently rubbished, ridiculed and insulted for it.
It seems that the vast majority of photographers, even so called ‘pros’ and ‘experts’ have a mindset harking back to film and simply cannot get their heads around the fact that a digital sensor does not behave like film and that digital exposure is radically different to exposure for film.
That of course is not helped by camera manufacturers calibrating their meters to 12-18% grey – great for film but hopelessly confusing for digital.
Thank you for plugging away at shedding some light.
If we reach our final day of life on this planet, without ever having been insulted, all we have achieved is draining this planet of its valuable resources.
Pete A
Fortunately, and despite the regular and liberal application of clinically proven, rejuvenating horse liniment (with added snake oil), my skin is still tougher than some of the planks I have been beaten with.
I agree with you 100%.
We also have more than 20 years worth of misconceptions of the terms “digital” and “quantum” that are long overdue for being debunked, then laid to rest in the graveyards of pseudoscience and anti-science.
I always have some fun explaining to digital photographers how latent image in emulsion consists of 0s and 1s ;)
Yes indeed: molecular quantum mechanics, aka quantum chemistry. The shining example [pun intended] being the discovery of cathode rays by Michael Faraday circa 1838.
Harvey
“After all, aren’t meter readings dependent on the meter mode?”
No, they are not. Meter readings are just meter readings – regardless of mode.
The dumb meter just measures the amount of light falling on it, or the sensor.
All modes, given the same tone to read, will read it the same.
ISO determines how the sensor responds to that light and is set by the user – who then also sets an aperture and shutter speed combination which ideally fully saturates the sensor at that ISO.
“Regardless of Auto ISO being blind to the metering mode, nonetheless the metering mode is the starting point, which influences the ISO set by Auto ISO.”
No, the metering mode is not the starting point and does not influence the ISO set by Auto ISO.
Base ISO is set by the manufacturer of the sensor.
Auto ISO is influenced by meter readings (regardless of mode) and the user of the camera – who defines the maximum and minimum permissible ISO and the minimum permissible shutter speed.
“I think in the future I will stick to threads written by Nasim on Photography Life and avoid wasting my time on silly banter on Iliah’s threads.”
The principle cause of noise is underexposure.
Rudeness apart, you might perhaps worry less about controlling the Nth degree of ISO noise through ISO variance/invariance which, in most instances, is not that important and more about how to achieve a good exposure in the first place which, in all instances, is all important.
Betty, if the metering mode “does not influence the ISO set by Auto ISO”, then why does my Auto ISO on Sony give me different ISO values depending on the metering mode (i.e. Multi, Center, Spot for Sony), given the same scene in the same light?
Harvey
Because each of these metering modes ‘ looks’ at a different part of the scene.
Multi looks at the whole frame and makes a shed load of algorithm based calculations.
Centre looks mostly, but not entirely, at the central area.
Spot looks at a tiny 2-5 degree spot.
Each of these will almost certainly have a different tonality.
Try metering off an 18% grey card or a sheet of paper and see what you come up with.
Hi Iliah
Just a question about the settings in FastRawViewer, concerning DR setting . Can I use the values for DR as shown on DXOMARKS.com at the different iso DR was tested and set this value in FRV preferences for underexposure detection limit. For instance if DXO say for my camera at iso 100 I have a DR of 14, is it the value that should be set in FRV for all images shot at iso 100 ?
regards
DxO Labs publish so-called Engineering Dynamic Range numbers. Those are for signal-to-noise 1:1, that is about 3 stops higher compared to photographically usable dynamic range for a 8×12″ (20x30cm, S8R, 6PW) print at 300 dpi. When DxO say 14 stops, I set 11 stops (and test for banding and artifacts on a print, because it may turn out to be 10 stops).
To quote Kodak, “It is common in photography to designate SNR=40 as an ‘excellent image’ and SNR=10 as an ‘acceptable image’” — log2(10) is about 3 1/3 EV; turns out Kodak were right ;) The Kodak document is still available here: http://www.kodak.com
Thanks for your answer on my question, but is this also apply if you don’t print (its not my case ? Among all theses articles on internet this one is one of the best to apply to my landscape photography.
Thanks again
regards
Luc
Hello Iliah.
Thanks heaps for your insightful articles. The math does go over my head but I try to keep up, hoping to arrive at the magic value for my Nikon D40 (ancient, I know) :). I am a casual shooter but would like to ensure that I “juice” the sensor as much as I can for maximum usable DR. From your knowledge & exhaustive testing of your own software products, would you be able to provide these values to us? My own basic testing indicates I can ETTR the D40 by about 1 to 1.5 stops max. Does your testing agree with this value? Thanks again. Looking forward to the next one.
I’m afraid I never tested Nikon D40 for this, but I would gladly examine your raw files. You can e-mail links to [email protected]
“By the way, if one tries to set exposure low to keep from clipping all the light sources and specular highlights, he is, most probably, also wasting a couple of stops of dynamic range.”
Guilty as charged… I’m working on this issue currently, trying to improve exposure technique to minimize the need of post in LR. The article is very technical (at least for my limited understanding of these things) but I’ll go through it one more time to see if I can translate any of this into practical application in the field. Thanks for posting!
I think the most important thing to master is identifying which elements in each scene convey very little information to our intended audience. E.g., when the Sun is in the frame, there is no need to avoid highlight clipping unless the Sun is the primary subject of the image.
It’s very difficult to master this using only photography as the medium; it’s much easier to master starting with pencils and paper, then progressing to brushes and paints.
The basics of using spot-meter to expose for highlights is very simple:
– determine how many stops do you have in raw between the metered exposure and highlights
– meter from important highlights, having the amount from the previous as positive exposure compensation dialed in the camera
– do not hesitate to bracket the exposure until you get the grip
You are welcome to ask questions ;)
Betty, yes, I was aware how the varying metering modes work. I’ve also metered off a grey card, amongst other things, numerous times.
My underlying point, however, is we are getting caught up in semantics again. Particularly when some metering modes utilize an algorithm (or weighting), which ultimately influence Auto ISO. If the metering mode did not influence the Auto ISO (or the automatic, and semi-automatic, modes, such as P, A, S), then there would be no need for an algorithm (or weighting based approach).
Harvey, You previously stated: “Regardless of Auto ISO being blind to the metering mode, nonetheless the metering mode is the starting point, which influences the ISO set by Auto ISO.”
No! The starting point is mastering photography from first principles. Only when this has been achieved does the in-camera meter, with its various modes, serve any other function than pandering to the intellectually lazy for the sole purpose of increasing company turnover.
Unless you completely understand the metering algorithm that you have selected then it isn’t “the starting point”, it is yet another unknown and unpredictable variable in your image capturing process.
semantics (n.)
“science of meaning in language,” 1893, from French sémantique (1883); see semantic (also see -ics). Replaced semasiology (1847), from German Semasiologie (1829), from Greek semasia “signification, meaning.”
http://www.etymonline.com/index.php?term=semantics&allowed_in_frame=0
semantic (adj.)
1894, from French sémantique, applied by Michel Bréal (1883) to the psychology of language, from Greek semantikos “significant,” from semainein “to show by sign, signify, point out, indicate by a sign,” from sema “sign, mark, token; omen, portent; constellation; grave” (Doric sama), from PIE root *dheie- “to see, look” (source also of Sanskrit dhyati “he meditates;” see zen).
http://www.etymonline.com/index.php?term=semantic&allowed_in_frame=0
Harvey’s semantic filibustering is a modus operandi that is frequently deployed for the purpose of hampering, rather than helping, scientific discourse.
Harvey
I just did a quick and dirty test regarding your problem.
I took my D800E out of the cupboard and shot three frames of my Eizo (very uniform) 318-4K monitor of my default dark grey background only using matrix, centre weighted and spot readings.
The exposure was F5.0 1/50sec Auto ISO.
Unsurprisingly, the indicated ISO stayed at ISO 640 for all three exposures and all three histograms in the back of the camera were identical.
Either you are doing something weird or your camera is doing something weird.
Go figure…
Betty, repeat the process, trying it with different scenes. You should get varying results, depending on the reflectivity of the various subjects in the scene.
For the sake of exactitude (which some here apparently expect), my previous comment regarding “depending on the reflectivity of the various subjects in the scene” was assuming consistent lighting throughout the scene. Of course, varying light levels across the scene is another factor.
Harvey
“Betty, repeat the process, trying it with different scenes. You should get varying results, depending on the reflectivity of the various subjects in the scene.”
Dear Lord, give me strength…
That’s what I was trying to tell you in my previous answer.
You seem not to understand what a metering mode is, let alone how to use it. You either genuinely don’t know or you are taking the proverbial.
**Of course** you will get different results with different modes – if they all gave the same result there would be no point in having different modes.
Each of these modes meters *different areas of the scene*. The meter will only meter what you give it to meter and will then render whatever that is, a mid tone grey.
If you are matrix metering you are asking the meter to average all the tones in the frame and render them mid tone grey (simplified version).
If you ask the same question of a spot meter, it will make *whatever is under the spot in that scene* a mid tone grey regardless of whether the object under the spot is white or black or anything in between.
If you don’t understand metering (and you clearly don’t), stay away from spot metering and use matrix. It will keep you out of trouble pending the day you know what you are doing.
You are ripe for Nasim’s Level 1 Photography Course.
This is no knock against Nasim’s course, as I am sure it’s a good course. With that said, I don’t need a course in metering. Actually, nailing accurate exposures, manually, with spot metering is one of my strong suits. Please stop assuming what others here do and do not know. It is annoying. Anyway, I am done arguing on here.
Harvey
“I don’t need a course in metering. Actually, nailing accurate exposures, manually, with spot metering is one of my strong suits.”
Spot metering is one (you mean there are more?) of your strong suits! Silly me, I should have known.
Heavens above Pete A, Iliah and I clearly have much to learn from your unique grasp of this subject, after all, getting a different result every time one switches metering modes is the holy grail of photography, an inexhaustible source of surprise, delight and creative opportunity. Perhaps you might contribute an article on how you achieve your unique and ever varying results?
“Please stop assuming what others here do and do not know.”
You come on Photography Life and ask a series of questions that would embarrass a quantum mechanic and then have the nerve to come out with that?
Rest assured, there is no need to assume anything.
When you ask a question like, “..why does my Auto ISO on Sony give me different ISO values depending on the metering mode (i.e. Multi, Center, Spot for Sony), given the same scene in the same light?” AND you manage to get into the same mess metering from a grey card, it’s painfully obvious you don’t know your arse (English spelling in case you want to get into semantics) from your elbow.
Pete A put it best…
“There is no polite way of saying this: You do not understand anywhere near enough about digital photography to know what it is that you do not yet understand.”
Now go away and stop wasting everyone’s time.
Betty, is it physically possible for you not to be such a huge jerk? Your barrage of vitriol doesn’t warrant any more of my time responding. Surely, there will be others here for you to pick on. It surprises me that you have not been banned from this site yet. The day that happens, this site will be better for it. You are one angry, bitter, negative, condescending narcissist. I am out of here.
Harvey
There’s only one jerk here – the one who doesn’t know his mode from his meter.
Betty, if you spent less time here picking fights with people and more time actually shooting, you may have known by now how to properly operate your camera, including its metering modes.
Harvey
Golly Brian, I wonder if then you could help me?
Could you explain why my Auto ISO gives me different ISO values depending on the metering mode (i.e. Multi, Center, Spot ), given the same scene in the same light? Even with a grey card!
I point my meter at things and that ISO thingy keeps going all over the place. It’s really baffling and from reading your conversations with Pete and Iliah, you obviously you have a much deeper understanding of this difficult subject than any of us.
?
Betty, no, I certainly don’t know more than the experts who write for this site. Photographers like Nasim and Iliah are certainly experts, both of whom I have learned from. They are in a different league than I am. I know that. And I am thankful that they are here to educate.
With that all said, I do know enough to get the results I am looking for in the field. Yet, I am constantly looking to learn more and improve.
Anyway, good night.
Harvey
“Photographers like Nasim and Iliah are certainly experts, both of whom I have learned from. They are in a different league than I am. I know that. And I am thankful that they are here to educate.”
Well that’s a step in the right direction.
If you listened to the experts more instead of trying to prove them wrong, you might make faster progress.
I was not trying to prove Iliah “wrong”. I was just pointing out that he was nitpicking by saying I was “not correct” based on a technicality that lacked context. Anyway, talking about this ad nauseum is not enjoyable. I would rather be out shooting and processing my photos in Capture One (which I recently upgraded and absolutely love).
Harvey,
I am sure Iliah wasn’t nitpicking or indulging in semantics – he undoubtedly has better uses for his time, but sometimes what at first sight appears to be a ‘technical quibble’, actually goes to the heart of someone’s lack of understanding. In this case, about metering, metering modes and auto ISO – and how they interact – or don’t.
The fact that you dismiss it as a technicality or silly banter just serves to show how off base you seem to be on some of this. So my advice would be first, to be sure of your facts before telling an expert he is quibbling and second, do some research on metering because what you have said about your methods and results strongly suggests there is a gap in your understanding.
Neither was I, or am I, picking a fight with you or anyone else.
Science doesn’t hold polite opinions on matters of fact. You’re right or you’re wrong or you don’t know – all of which is OK as long as your motivation is to learn.
I just call it as I see it based on the science – for which I frequently get berated. Curiously, there is one characteristic common to people who shout the loudest – they frequently know the least. And when they fail to make their case with facts grounded in science, they often unfortunately resort to ad hominem abuse of one kind or another.
Some choose to continue their photography using methods based on their ingrained beliefs. That’s OK if that’s what they want to do, but when they declare their beliefs as fact, a correction is in order lest they mislead others and of course, it offers them an opportunity to ‘see the light’.
So, no hard feelings and happy shooting.
Betty, I am frequently berated on this website for the same and/or similar reasons. I have previously stated on this website:
QUOTE [2015-08-30]
I usually spend many hours writing [as in: editing, rewriting, and creating many typos during the process!] each of my long technical comments.
The reason that I do this is because I think Nasim has created a website that might qualify as being one of the best [as in both technically accurate and easy to understand] encyclopaedias of photography that is available on the World Wide Web — it’s a gold mine of interesting information and education.
When I read comments on articles claiming that the author is wrong, or is just promoting a product, I think it is important to challenge such claims, rather than ignoring them. Leaving such claims unchallenged will likely cause non-expert readers of this website to doubt its authenticity.
END QUOTE
QOUTE [2016-04-12]
I’m sometimes very harsh to those who write comments because I deeply value the Photography Life website for being by far the most accurate encyclopaedia of photography that I have ever encountered. Why is it so accurate? Because Nasim adheres to the core scientific principles that enable science to be self-correcting, therefore, this website is an extremely valuable resource, both now, and for future generations.
END QUOTE
I shall do my utmost to honour those statements, to the very best of my ability, for the remainder of my life.
Betty, are you speaking on behalf of Iliah? I think he is perfectly capable of speaking for himself. If you recall, the debate between Iliah and I ended off on the definition of “semantics”, not based on “science” as you describe.
I still believe that Iliah was nitpicking. Perhaps he does not see it that way. For example, in the world of computer science, in which RAW software is written, a certain exactitude is required when writing with computer language. Even something as seemingly minor as misplaced punctuation can create a bug in a computer program. So, I do understand why Iliah may have a tendency to nitpick.
Harvey
“Betty, are you speaking on behalf of Iliah?”
No Harvey, I wouldn’t presume to do that. I was speaking on my own behalf – as always. It is however perfectly possible to hold a discourse based on what was said by someone else.
To refresh your memory…and get back to the point.
“Iliah said, “adjusts the ISO based on the metering mode set in-camera” is not correct. The adjustments take place based on meter readings, not mode.”
To which you replied, , “Iliah, aren’t we getting caught up in semantics regarding meter readings vs. meter mode? After all, aren’t meter readings dependent on the meter mode?”
That is you diverting a genuine scientific point into a spurious semantic one to cover the fact that you have no grasp of the concept he was putting to you and revealing your ignorance in all its glory.
“I still believe that Iliah was nitpicking.”
Of course you do.
That’s because your mind is closed to the possibility that you might simply not understand. You evidently prefer to cling to your ‘beliefs’ rather than face the discomfort of admitting the possibility that you might be afflicted with chronic anatomical dyslexia (inability to tell arse from elbow).
In fact you are so determined to be ‘right’ that you reject correction from two acknowledged experts in the field and ignore the self evident nonsense that happens when you apply your muddled thinking to your own photography – preferring instead to scurry down a rabbit hole of ‘semantics’ where you think you can cover your basic photographic illiteracy under a cloak of dictionary definitions.
The semantic point you are missing however, is the fact that you do not get the point.
Betty, I am not debating with you over your interpretation of another writer’s intent.
Harvey
There was never any discussion of another writer’s intent. The discussion was about what was said and what was said directly to you could not be clearer:
>> “adjusts the ISO based on the metering mode set in-camera” is not correct.
The adjustments take place based on meter readings, not mode.<>Yes, we are getting into semantics. Semantics, in fact, is the study of meaning.
“aren’t meter readings dependent on the meter mode”
ISO Auto knows nothing of the mode. Only the readings matter.<<
But you know better.
And of course you’re not going to debate further – you are filibustering because you’ve run out of places to hide.
I could retort and debate with you until the world ends. But I won’t. I would rather spend my time productively.
Harvey
“I could retort and debate with you until the world ends.”
I am sure you could spout hot air till the cows come home, but a filibuster is not a debate.
Filibuster: An action which obstructs progress especially by speaking at inordinate length.
“I would rather spend my time productively.”
I would like to suggest you start by learning the difference between metering and a metering mode and then go on to study how one of these is intimately connected to auto ISO and how the other isn’t.
Someone who has nothing better to do with her time than troll this site, picking fights with people, putting others down in a condescending way, telling me how to better use my time productively. Isn’t that fresh?
Harvey
Ah yes, now we have it – the ignoramus’ last stand.
When all else fails, when the argument is shown to be hollow and the semantic diversions no longer work, what does an ignoramus do to assuage a bruised ego?
Yes of course, turn to the ad hominem attack.
Now the follower of a site becomes a ‘troll’, commenting becomes ‘picking a fight’, making a suggestion becomes ‘being condescending’ and making a factual correction becomes ‘forcing one’s opinion’ – all peppered with a gentle rain of insulting epithets…
It’s really very sad in that had you simply asked for clarification, one of the experts would gladly have given you a fulsome explanation and you could have learned something of lasting benefit.
Instead, you chose to question their bona fides, put your own ignorance on display and have learned nothing.
Wikipedia: “In Internet slang, a troll (/ˈtroʊl/, /ˈtrɒl/) is a person who sows discord on the Internet by starting arguments or upsetting people, by posting inflammatory,[1] extraneous, or off-topic messages in an online community (such as a newsgroup, forum, chat room, or blog) with the deliberate intent of provoking readers into an emotional response[2] or of otherwise disrupting normal on-topic discussion,[3] often for their own amusement.”
Harvey
I see you continue to be big on semantics.
Well, setting aside the fact that you began the whole thing by trying to disseminate your home brewed nonsense, let’s refer to your definition:
1. You were the one who started the arguments – with Iliah, Pete A and myself.
2. You were the one went off topic with your silly, semantic detours.
3. You were the one who was abusive.
I will leave it to you to work out who the troll is.
Oh, and dealing with terminal stupidity is anything but amusing.
1. If you check the history of comments, you will see that it was indeed you who first addressed me on the debate. As for the others you mentioned, I will leave it up to them to speak for themselves.
2. The question of who was the first to go off topic is a matter of semantics.
3. If I had nothing better to do with my time, I am sure that I can compile paragraphs, if not pages, of abusive comments made by you, Betty, here on Photography Life.
Harvey wrote on 2016-07-26: “If you check the history of comments…”
Here’s a short summary thus far:
2016-07-15
At the risk of drawing criticism for what I am about to say, I feel that Betty’s tone is rather condescending to a number of posters. Based on the comments of some other posters, I believe that I am not alone in this sentiment. I feel that Betty has a lot to add in terms of content. If only she can be more polite to other posters who take the time to comment here.
2016-07-20
… Anyway, I am done arguing on here.
2016-07-21
Betty, is it physically possible for you not to be such a huge jerk? Your barrage of vitriol doesn’t warrant any more of my time responding. Surely, there will be others here for you to pick on. It surprises me that you have not been banned from this site yet. The day that happens, this site will be better for it. You are one angry, bitter, negative, condescending narcissist. I am out of here.
2016-07-24
I still believe that Iliah was nitpicking.
2016-07-25
I could retort and debate with you until the world ends. But I won’t. I would rather spend my time productively.
2016-07-26
Someone who has nothing better to do with her time than troll this site, picking fights with people, putting others down in a condescending way, telling me how to better use my time productively. Isn’t that fresh?
2016-07-26
If I had nothing better to do with my time, I am sure that I can compile paragraphs, if not pages, of abusive comments made by you, Betty, here on Photography Life.
Pete A
Thank you.
Harvey
1. As Pete A points out, you ‘stuck your head in’ at the outset to make an unprovoked, personal criticism.
2. The matter of who first went off topic is anything but a matter of semantics.
As plus resident expert on semantics, you should know that.
3. Using humour, sarcasm, irony is part of the cut and thrust of debate. Guilty.
The inability to distinguish between that and abuse is infantile.
Yes, I am aware that it is a public forum. Sadly, I allowed myself to get dragged into a childish debate in a public forum. That’s not something that Betty or I should be proud of. It does not look good on Photography Life either.
Harvey
No one dragged you into anything.
You got there under your own steam by ‘provoking’ the expert writer of an excellent article; implying he did not understand the terminology used in his own field of expertise.
True, Betty, that you had not yet provoked me when I first mentioned that your tone is condescending. Although you had not yet provoked me, that does not mean that your condescending tone did not affect me and others here; I, and surely many others, come here to read information and learn, not to read about people putting others down in a very disrespectful way.
So, Betty, if your condescending comments are simply your attempts at humor, sarcasm, and irony, then why is nobody here laughing at what you say?
Harvey
First and with respect, if there was less concern with tone and more with content, there would be less hot air and more intake of useful information.
Second, joining a discussion is not a ‘provocation’. If you feel provoked, you would be better served attending an anger management class than an internet forum.
Third, condescension is in the eye of the beholder.
Unfortunately, some feel ’condescended’ when their unsubstantiated beliefs are questioned.
That is something for them to come to terms with; it is not for me, Pete A or others to pander to their pseudoscientific delusions under PL’s banner.
I can’t help feeling that some of the ruffled feathers would get a good deal less ruffled had the comment came from a Bill rather than a Betty. But perhaps I am being oversensitive? (It’s one of my failings).
If you care to actually read my opening comments in most discourses, they are almost invariably polite, straightforward and factual. This is frequently taken as ‘condescension’ and often ’provokes’ a heated defence – more often than not accompanied by abuse of one kind or another. At that point the gloves come off. Some no doubt gain some amusement or even learn from what ensues, others not.
C’est la vie.
Betty, you are condescending. Very condescending. I don’t think you realize just how condescending you are. And it has nothing to do with your gender. Nothing at all. It has everything to do with the way you talk to people.
Why was my response deleted?
Next time a police officer pulls me over for speeding, perhaps I ought to borrow a page from Iliah’s book of logic and argue that the speed of the car was not dependent on me because the engine was blind to me and my foot. The trouble with such an argument is that I will get laughed out of court.
Harvey
A car, like a camera, requires intelligent input.
Without that, all bets are off.
When I first learned how to drive, I did go online to find out how the engine, clutch etc roughly work…
Below is my sensored version of my deleted response.
Oxforddictionaries.com: “semantics: 1.1 The meaning of a word, phrase, or text: such quibbling over semantics may seem petty stuff”.
Merriam-Webster.com: “SEMANTICS: the meanings of words and phrases in a particular context”.
When debates lack “context” they can become “petty”.
Hopefully, this sensored version does not get deleted.
whinge [verb]: complain persistently and in a peevish or irritating way.
Pete, I just learned a new word. Thanks.
Thank you mr. Borg for a very thorough, and insightful article.
Reading this thread this morning, it struck me, that being right often differs vastly from being polite or even just nice. It also is a great example of how to shut down debate. The article discusses (as far as I can see) how to best to utilize the dynamic range in a sensor. The condescending tone in various replies obscures this, and just leaves you sad.
I honestly don’t understand all the confusing, or am I confused?
What I know is that you want to have the histogram to the right as much as possible without clipping to get the highest SNR.
Dear Pim,
There are 2 caveats to what you’ve said.
First, having JPEG histogram on the back of your camera to the right is not enough, as you can see from http://www.fastrawviewer.com/culling-raw-vs-jpegs the raw may still be very significantly underexposed.
Second, promoting the histogram to the right using ISO control does not have desired effect past certain ISO value, and is never an adequate compensation for low exposure.
Thanks for your reply, great article by the way.
Sure you need to use the histogram of the raw file actually. As for your second point, I think I dont understand ISO exactly, you mentioned earlier that there are camera models that decrease noise with raising ISO. Do you by any chance know where the technical details are explained behind this?
I would say that with raising the ISO, you double the amount of light and noise, and thus an increase of sqrt{2} in SNR…
Dear Pim,
ISO is an internal setting applied after the actual exposure took place, and as such it does not affect the amount of light. In the camera only shutter speed and aperture (exposure-time factors) control the light to the sensor. In auto exposure modes (except for Manual + Auto ISO) doubling ISO cuts the light in half.
Some cameras, mostly those that are based on Sony sensors, have very low read noise numbers at low ISO settings, thus raising ISO has very little effect on noise in shadows (read noise). However even with these cameras one still needs to be careful with the idea of shooting without adjusting ISO at all, as raising ISO 2-3 stops up from base helps decrease banding artifacts and shadow tint (magenta, usually), which is the result of black level drift.
Other cameras, like Canon dSLRs and most of current Nikon 1-digit bodies (including Df, but excluding D3X), have relatively high read noise levels at base ISO, and read noise goes down significantly with ISO going up to a certain point which is sometimes called “point of invariance”. The graphs on http://www.photonstophotos.net/Charts/PDR_Shadow.htm go very flat starting from that point of invariance.
Iliah,
Thanks a lot for your excellent article. I have been waiting for it ever since February and even feared you would never publish it due to the rather weary response to your midtone articles.
I have a couple of questions, though. In the article above you indicate that the exposure compensation value to be determined using your method depends on the camera model and possibly on the ISO value. Are there any ranges of ISO values with similar, comparable exposure compensation values for a given camera model? If yes, is there any shortcut for determining these ISO ranges (not the compensation values) for a given camera model, e.g. by using the sensor data published by Bill Claff (or do you even know these ranges for some camera models and are willing to publish them;-) ? If yes, how? I am just thinking about how to reduce the work to be done as well as the compensation values to remember to a manageable amount.
In addition to the excellent articles on PL, even the comments sections contain very useful and helpful hints, so I always try to read them. Although your article is about exposure, there are a lot of hints w.r.t. ISO in the comments above. Unfortunately, though, I do not understand them completely. And even after reading Bill Claff’s Sensor Analysis Primer (thanks for the link) I am still confused, especially w.r.t. PDR Shadow Improvement, which you also explained above.
Is my following interpretation correct? In case the dynamic range (DR) of a scene fits into the DR of a D810 at ISO 200, it is wise to use ISO 200 instead of ISO 64 (although the D810 as a larger DR at ISO 64) because below ISO 200 the PDR Shadow Improvement is less and above the PDR Shadow Improvement does not increase significantly (provided you expose to the right, as explained above). If this is correct, then it would be wise to set ISO to 3200 on a D5 (provided the DR of the scene fits in to the DR of the D5 at ISO 3200), which if find really puzzling. (Please excuse my awkward formulations. This is because I do not know what PDR Shadow Improvement really is, nor how it is defined nor in what unit it is measured.)
Iliah, how about writing an article on PL about ISO? Given my lack of understanding as well as the questions and quarrels above, this might be a good idea! (And no, Bill Claff’s Primer doesn’t suffice, for it does not explain the different photographic scenarios to consider.)
Dear Carsten,
> Thanks a lot for your excellent article.
Thank you for your kind words.
> I have been waiting for it ever since February and even feared you would never publish it due to the rather weary response to your midtone articles.
Sorry for the wait. As to the responses, in my experience they are not an accurate indicator of interest or attitude; on top of that, some things need to sink in, and that can take time: months and years.
> In the article above you indicate that the exposure compensation value to be determined using your method depends on the camera model and possibly on the ISO value.
First, let me be clear, it is not _my_ method ;) Metering for highlights is actually one of the things spot-meters were designed for; with film, one can observe the result directly to calculate the exposure shift (similar to the brute force approach I mentioned in the article), or one can use a densitometer (that is what RawDigger is for digital).
> Are there any ranges of ISO values with similar, comparable exposure compensation values for a given camera model?
Normally, camera metering (midpoint) is calibrated the same way for all ISO settings, leading to the same compensation value. Exceptions are: “Lo” settings (those are below the base ISO setting and are usually implemented through the shift of midpoint; in practice there is very little reason to use those, be it for raw or JPEG); some higher ISO settings, for example on Fujifilm cameras, also implemented as midpoint shift, but in the opposite direction, towards less exposure (no point in using those if one shoots raw); intermediate ISO settings on some cameras may need additional checking, see, for example, our article http://www.rawdigger.com/howtouse/the-riddle-of-intermediate-iso-setting
> In case the dynamic range (DR) of a scene fits into the DR of a D810 at ISO 200, it is wise to use ISO 200 instead of ISO 64 (although the D810 as a larger DR at ISO 64)
The first choice is to raise the exposure. That is, decrease shutter speed and/or open aperture up (ISO setting is not a part of exposure). The more light you allow onto the sensor during exposure, the lower the noise.
If the exposure can’t be increased further (because of considerations like shake, the speed of the subject’s movement, depth of field, risk of AF misses) it makes sense to start increasing the ISO. There are very few exceptions to this rule, but they are hard to analyse in a reply, and they probably merit a separate article, probably one not interesting to most.
Why would one increase ISO setting at all on ISO-less (ISO-invariant) cameras, or beyond the point of invariance (no further shadow improvement with the raising of the ISO setting)? – because read noise (and shadow improvement calculated from the decrease of read noise, measured in photographic stops on Bill’s site) is not all that matters when one starts to lift shadows, even for the best ISO-less cameras that are currently available. When the levels in shadows are scarce on EV scale, the relative error is large, resulting in black point drift, colour skews, banding, loss of contrast. Those problems are often not visible until we start to bring the shadows out. My rule of thumb is that for low light scenes, given that I’m limited by shutter speed and aperture, I set ISO about 2-3 stops above the point of invariance to ensure that the shadows are editable. If I’m not planning on opening the shadows, that’s another matter.
> it would be wise to set ISO to 3200 on a D5
D5 is not even close to ISO-less below the ISO setting of 2500. The colour in the shadows improves up to 2x of that, ISO 5000.
> how about writing an article on PL about ISO?
I have it in mind, but “The Arithmetic of White Balance and Why Should I Care” comes first ;)
> Normally, camera metering (midpoint) is calibrated the same way for all ISO settings, leading to the same compensation value.
Just one compensation value to determine and to remember – perfect.
> The first choice is to raise the exposure.
This is clear, for this is what your article above is all about. But was it? Apparently not, because for some reason I thought I could expose to the right at base ISO and then increase ISO to the point of invariance (i.e. ISO 200 for a D810, if I read the PDR Shadow Improvement data correctly) in order to improve the shadows. This would cause the highlights to clip, though. So, my interpretation explained above is just nonsense. Thanks for sorting this out.
> Why would one increase ISO setting at all on ISO-less (ISO-invariant) cameras, or beyond the point of invariance (no further shadow improvement with the raising of the ISO setting)? – because read noise […] is not all that matters when one starts to lift shadows.
Ok, since there are other contributing factors not reflected in the PDR Shadow Improvement data and since I do not know of any additional data showing them (and wouldn’t be able to interpret them correctly anyway), I am back to my previous mode of operation: having and using the ISO dial on my D810 makes sense – if I cannot further improve exposure, I will raise ISO in order get a properly “exposed” image and can then do whatever I want in post processing. See my last reply below, though. :-)
> D5 is not even close to ISO-less below the ISO setting of 2500. The colour in the shadows improves up to 2x of that, ISO 5000.
I just took a second look at the PDR Shadow Improvement data, and yes, the point of invariance seems to be at ISO 5000 with an improvement of almost 3 1/3 stops. Nevertheless, my original interpretation is nonsense.
> > how about writing an article on PL about ISO?
> I have it in mind, but “The Arithmetic of White Balance and Why Should I Care” comes first ;)
Deal – I will wait for both of them now.
> First, let me be clear, it is not _my_ method ;)
Sure, but you are the one who clearly describes this method and provides the tools needed ;)
Pete, I said that Betty was the “one who first addressed me on the debate”. The operative words being “on the debate”. She stuck her head into my discussion between Iliah.
Harvey
PL is a global forum.
If you want to engage in private debate, stay off the internet.
Yes, I am aware that it is a public forum. Sadly, I allowed myself to get dragged into a childish debate in a public forum. That’s not something that Betty or I should be proud of. It does not look good on Photography Life either.
Harvey
“Yes, I am aware that it is a public forum.”
Good, then in future, you will, I hope, accept that comment from any quarter is OK.
The only childish part of this debate was not so much your being plain wrong on a scientific/photographic subject, or even your not taking the opportunity to learn, but that you instead chose to try to prove yourself right by picking a semantic fight with an acknowledged expert in his field of expertise! How silly is that.
– And of course your subsequent peevish whingeing, squirming and abusiveness to divert attention from the fact that you were in a hole.
“The only childish part…”. That is your opinion. An opinion which is blind to your narcisstic behaviour.
How do you know what I have, and have not, learned? Do you possess a special device that can read the knowledge base in my brain? But of course a narcissist does not need such a device.
I did not pick a fight with Iliah. We had a discussion, which he, smartly, long gave up.
Harvey
“The only childish part…”. That is your opinion”
Yes Harvey, it’s my opinion; that a half baked amateur trying to prove an expert wrong on his own subject is about as silly as it gets.
I don’t see what my narcissism has to do with it.
“How do you know what I have, and have not, learned?”
I don’t, but hope springs eternal and will perhaps this time triumph over reality.
Fingers crossed.
“Do you possess a special device that can read the knowledge base in my brain?”
No special device needed.
A plank is a plank – plain for all to see.
“I did not pick a fight with Iliah. We had a discussion, which he, smartly, long gave up.”
Call it what you will.
He told you, more than once, that you were flat wrong.
He smartly gave up because trying to educate a plank is futile.
I was not attempting to prove Iliah “wrong”. Rather, he was attempting to prove me wrong, based on a technicality that lacked context. The fact that our final contentious point was the definition of “semantics”, rather than my assertion that meter readings are dependent on the metering mode, is evidence to what I have said. Iliah never disputed the fact that meter readings are dependent on the metering mode.
“The first step toward change is awareness. The second step is acceptance”, Nathaniel Branden.
You have made the first steps in your recovery from narcissism. Good on you.
Harvey
Stop rewriting history.
Brush up your photography.
Don’t worry about me, I will be just fine.
“I was not attempting to prove Iliah “wrong”. Rather, he was attempting to prove me wrong, based on a technicality that lacked context.”
…..Iliah was attempting to prove you wrong….Yeah, right…….?
Iliah never disputed the fact that meter readings are dependent on the metering mode.
“The adjustments take place based on meter readings, not mode.”
You are living in a parallel universe, a sort of LaLa land where memory and reality merge into delusion.
Meter readings don’t take place in a vacuum, independent of other variables. For the sake of exactitude, there are a number of variables, in the process, which both Iliah and failed to mention. So, perhaps we were both “not correct”. At no point did Iliah explain the process from A to Z.
At this point, who, other than Betty, cares if one, or both of us, was not correct. After all, to err is human. I have been incorrect on numerous matters numerous times in my life and I will be incorrect numerous times in the future. It is part of being human. No shame in that.
Your quote of Iliah was his initial point of contention, not his response to my follow up point, which he failed to answer head on, presumably because doing so would have been an admission that he was nitpicking, which would have been okay if he had taken the time to educate by explaining the process from A to Z. Unfortunately, he appeared to be more motivated in proving me “not correct” on a technicality, rather than educating. This is why we had a problem.
So, no, I do not live in a parallel universe.
The subject of Iliah’s article is more than adequately conveyed in the title of his article: How to Use the Full Dynamic Range of Your Camera.
The detailed explanation of the subject is more than adequately conveyed in the article itself and in Iliah’s replies to the readers who wish to learn.
It is impossible to explain any advanced concept, from ‘A to Z’, to people who have not only failed to understand steps A, B, C, and D; they also delight in taunting those of us who do fully understand the advanced concept from first principles.
Taunt away to your heart’s content — we really don’t mind because *this* universe is never going to alter its laws of physics to comply with *your* wishful thinking :-)
Harvey
“Meter readings don’t take place in a vacuum, independent of other variables.”
Strictly speaking Harvey, yes they do.
The only variable, as far as the meter is concerned, is the light falling on it.
You are utterly confused about the ‘other variables’ because you don’t understand the differences between metering, metering modes, calibration, exposure, ISO, and Auto ISO, nor how they relate (or don’t relate) to each other. Until you do, there is little point in your trying to engage in a discussion on photography, still less insult an expert with your idiotic, semantic nitpicking.
With regard to your complaint that “At no point did Iliah explain the process from A to Z.”… Iliah said:
“Dear Sir:
I answered your comments several times, with explanations. But this site is hardly a place to quote camera manuals at length. By the way, “adjusts the ISO based on the metering mode set in-camera” is not correct. The adjustments take place based on meter readings, not mode.”
You are lucky he was that polite.
‘The whole process’ was not explained to you because the article assumes more than a rudimentary grasp of photographic principles. When it became clear that you do not understand even the most basic terms used, there was little point in him wasting his time further or quoting a camera manual at you. RTFM Harvey.
It’s not Iliah’s job to teach you elementary photography from first principles in the middle of a discourse on ‘How to Use the Full Dynamic Range of your Camera’. If you want the ‘whole process’ to be explained to you, go and do the PL Level 1 Course and tell them there about all the ‘variables’ involved in metering.
Be prepared for laughter.
I had hoped that at some point in this, a light might have come on for you, but no, the fuse box has evidently blown and you seem content to sit whistling the intellectual equivalent of Yankee Doodle in the dark.
I just watched the 15 minute trailer for PL Level 1. There was not a single thing mentioned that I have not already learned.
Harvey
Learning is one thing, understanding is quite another.
Understanding has to pass the test of peer review and reproducibility.
Expert peers with impeccable credentials have told you you are wrong.
Reproducing what you have learned (about metering, metering mode and auto ISO) leads to a self evident nonsense.
Draw your own conclusions.
With respect, you need more than a 15 minute trailer.
“I just watched the 15 minute trailer for PL Level 1. There was not a single thing mentioned that I have not already learned.”
That which can be asserted without evidence, can be dismissed without evidence:
https://en.wikipedia.org/wiki/Hitchens%27s_razor
Again with the condescending tone. Can’t I say anything without drawing a personal attack?
Harvey
“Again with the condescending tone. Can’t I say anything without drawing a personal attack?”
You take care of your tone and I’ll take care of mine.
No personal attacks from me, Harvey. Just so we are clear:
A personal or ad hominem attack is an attack on the person rather than on their argument; most commonly on their character or some personal attribute such as physical appearance, character, gender, race, family background etc, with the aim of undermining their case without actually having to engage with it.
This is not be confused with an attack, no matter how vitriolic, humorous or sarcastic, on another’s argument, statements, actions or beliefs. Barbed writing is a legitimate tool in written debate and the fact that the recipient may take it personally or is offended, does not make it ad hominem.
Ad hominems are used by immature or unintelligent people because they are unable to counter their opponent using logic and reason.
For instance, the comments, “Betty, is it physically possible for you not to be such a huge jerk?” or “You are one angry, bitter, negative, condescending narcissist.” are good examples of ad hominem attacks.
Pete, this is not a court of law. I simply made a genuine comment. If you don’t believe it, that is your prerogative. I honestly don’t care. I don’t have to prove anything to you or anyone here. It’s not like I am trying to sell you, or anyone here, something, for which my credentials and knowledge base may come into question. My personal knowledge base is completely irrelevant to this website. I am not a contributing author here. So, nobody here should really give a darn what I do and do not know. It is not relevant to this site. I am simply a reader of this (and many other) site(s).
Better, suggested a course to me. My genuine reply implied that the course appears to have little to no benefit to me, based on what I already know. I find it sad that a genuine comment, draws such cynicism and criticism.
“My personal knowledge base is completely irrelevant to this website.”
I’m very glad that — after posting 42 tiresome comments on Iliah’s article — you have finally been able to accept the truth.
Harvey
That’s just an evasive, circular, self serving, apologia.
If you had made a genuine comment, you would have accepted the genuine explanations that Iliah, Pete and I offered. I wrote at least three quite detailed comments for you.
Your response, (besides expending an inordinate amount of energy criticising my ‘tone’), was to dismiss what been offered in good faith as ‘nitpicking’, ‘semantics’ and ‘silly banter’ while at the same time referring to me as a troll, a jerk and a narcissist; all the while insisting that your misguided contentions were correct and that Iliah was ‘just trying to prove you wrong’.
So finally you are right – this is not a court of law, you don’t have to prove anything, your knowledge base is completely irrelevant and nobody here really gives a darn what you do or do not know.
I did not know anybody was keeping score of the number of posts.
Harvey
Now you do.
Pete A is the kind of guy who puts a high value on accuracy.
“I did not know anybody was keeping score of the number of posts.”
No person is “keeping score of the number of posts”: the website server computer is displaying most, but not all, of your posts; and it is extraordinarily easy to ask a web browser to count the number of posts written by you.
Betty, I did not read a word you just wrote. You know why? I don’t care. Life is good and too short to waste bickering over nothing.
Harvey
“Betty, I did not read a word you just wrote.”
You wrote a reply to something you did not read?
Peter, are you suggesting that I was having a conversation with myself?
I should add that my initial comment to Betty, which was about her condescending tone, was not a point of contention. The only debate that ensued from that was between Betty and Kevin regarding the identity of Betty’s gender. Further, the fact that Betty omitted Kevin’s name from the group of names she listed is evident that she was not referring to those comments.
Additionally, Pete, now that you are sticking your head into this, I don’t believe that initiated any conversation with you. I did not pick a fight with you.
“Peter, are you suggesting that I was having a conversation with myself?”
It does seem that way.
Much better for all concerned.
“I should add that my initial comment to Betty, which was about her condescending tone, was not a point of contention.’
It was to me.
I found it very hurtful…I had to take an extra Prozac to take the pain away…?
“Additionally, Pete, now that you are sticking your head into this…”
Pete has my permission to stick his head into this. His head is filled with wisdom.
On the other hand Harvey, your head is filled with brown stuff and it would be better if you stuck it back where it came from.
Hope the above is not too condescending…?
An excellent informative article and learnt more from comments section than from the main article , got cleared lot of misunderstandings of the concepts and doubts. The complete article including comments is great and informative including Betty’s clarifications on the subject except the unnecessary silly issues. Clearly Harvey did not know the difference between various metering modes and meter readings or might have misunderstood the subject, if that is the case ,there is noting wrong in accepting it , instead of diverting the to the discussion something else. Let me thank Iliah Borg, Nasim, Pete A and Betty for the most informative article and comments. Hope to see many more such articles.
KVS Setty: There is nothing wrong with my understanding of metering modes and meter readings. Don’t know why you need to poke your head into an old silly argument.