ISO and 60D

WOW!!!!!
THANK YOU FOR THAT RESPONSE!!!!
I’m getting it. As my kid would say “I’m picking up what you’re saying”!

And, thank you for the positive comment on my Asheville people shoot. It was FUN. I know in NYC I’ll have to keep an eye out and a roll of dollars with me, :-)

Please correct me here, a quick look at my histogram and as long as I’m weighted to the right without clipping it’s a large capture with little “noise” problems.

The 60D does have a “C” mode that I believe can handle all the settings you suggested, I will make sure tonight.

AGAIN, thank you for your time and thoughts…..

Mark
Try to get the biggest exposure you can,
OK, not sure I’ve ever heard that expression. Please explain?
Ok - exposure measures the amount of light per unit area that the sensor (or film) collects. It is measured, properly, in lux seconds. The more lux seconds the bigger the exposure. In practice, it simply means get as much light into the image that you can. For a fixed scene brightness, that means as long a shutter speed and as large an aperture as you can. Shutter speed is likely to be limited by reducing or eliminating motion blur. The aperture will be limited either by your DOF requirement, the maximum aperture of the lens, or a desire to use the lens at its 'best' aperture. So far as noise goes, you always want the most light in the image, how important to you in relation to camera shake or lens resolution or DOF only you can judge.
Where you might want to set a limit is the other end (not sure Canon auto ISO will do this) where you'd want it to reduce shutter speed if the exposure got too much at 100 ISO, to avoid blown highlights.
Bob
Again, please have patience here with a newbie, can you expand on that?
A digital sensor just count photons, so whatever exposure it has it will count them, and you can process the output to get the range of output tones that you want. The exception is when it reaches its maximum count at base ISO. If the light levels in the image are higher than that, you will lose the highlights in the image irrecoverably. So, if you hit that problem, it is the one case that you need to reduce exposure, either by closing down the aperture or reducing the shutter speed (usually the latter, since it leaves the picture looking the same, assuming it was short enough to freeze motion blur in the first place)
Using a 70-200 f4 w/IS.

Here are some shots from last Saturday on my practice walk around Asheville where I live. Used AV for the first time, I’ve been shooting Manual since I started and will return to it for NY.
http://www.flickr.com/photos/retsurfer/sets/72157626729792564/
I wish I had your eye for the grab shot - some nice pictures there. Now, the problem you have with using auto ISO is that Canon just doesn't get it. As I said, what you'd like to be able to do is set the slowest shutter speed and largest aperture you can, and just have the camera set the ISO for you, except when you hit base ISO, when you want the camera to reduce exposure via the shutter speed and/or aperture. Unfortunately, you can't set Canon's auto ISO up like that (you can, I understand, with Nikon, although it's not exactly intuitive, it means the right combination of auto ISO limits and auto exposure mode).

So, back to your specific case. I can't see with this kind of work and a lens as good as yours, any reason to use anything other than f/4. The exposure time assuming IS gives 2 stops, could be 1/60 to cover all focal lengths on the zoom (if you were really cute, you'd want to change shutter speed as you zoom). Now as a rough estimate using the 'sunny 16' rule, in most normal daylight illumination, you will risk blowing the highlights at 100 ISO, so in that situation, you'd want to reduce the shutter speed.

The best compromise might be to stick to Av, f/4 at 100ISO, as soon as you see the shutter speed drop below 1/60 swap to M, f/4, 1/60, auto ISO. If the ISO goes back down to 100, swap to Av again. On my canons, as I remember, M mode retains settings between uses, so you can set the f/4, 1/60 in advance, and just change mode. You have to go into a menu to select auto ISO, though.

Canon manages to catch you allways. Another way of working would be Av, with the rear wheel assigned to ISO - then when the shutter speed goes down to your limit, you just boost the ISO to keep it there. So far as I know, youi can't do that, either.

Maybe someone with more familiarity with the 60D's auto ISO can suggest something better, or even more helpfully, point out that I'm wrong about the 60D's limitations..
--
Bob
 
WOW!!!!!
THANK YOU FOR THAT RESPONSE!!!!
You're welcome
I’m getting it. As my kid would say “I’m picking up what you’re saying”!

And, thank you for the positive comment on my Asheville people shoot. It was FUN. I know in NYC I’ll have to keep an eye out and a roll of dollars with me, :-)
That kind of photography has just about gone in the UK, street photography carries a big risk of arrest, and if you have a roll of used fivers, you'll be in even more trouble.
Please correct me here, a quick look at my histogram and as long as I’m weighted to the right without clipping it’s a large capture with little “noise” problems.
Exactly, up to you how far to the right you want to go. Interesting to think what's the best way of doing street photography with a 60D, LV with the flippy screen might be rather useful, then you'd have the histogram straight away.
The 60D does have a “C” mode that I believe can handle all the settings you suggested, I will make sure tonight.
I looked it up, I believe it would, set it up to M, shatter speed and aperture you want, and Auto ISO, then just flip from Av to C when the shutter speed hits your limit.
AGAIN, thank you for your time and thoughts…..

Mark
Try to get the biggest exposure you can,
OK, not sure I’ve ever heard that expression. Please explain?
Ok - exposure measures the amount of light per unit area that the sensor (or film) collects. It is measured, properly, in lux seconds. The more lux seconds the bigger the exposure. In practice, it simply means get as much light into the image that you can. For a fixed scene brightness, that means as long a shutter speed and as large an aperture as you can. Shutter speed is likely to be limited by reducing or eliminating motion blur. The aperture will be limited either by your DOF requirement, the maximum aperture of the lens, or a desire to use the lens at its 'best' aperture. So far as noise goes, you always want the most light in the image, how important to you in relation to camera shake or lens resolution or DOF only you can judge.
Where you might want to set a limit is the other end (not sure Canon auto ISO will do this) where you'd want it to reduce shutter speed if the exposure got too much at 100 ISO, to avoid blown highlights.
Bob
Again, please have patience here with a newbie, can you expand on that?
A digital sensor just count photons, so whatever exposure it has it will count them, and you can process the output to get the range of output tones that you want. The exception is when it reaches its maximum count at base ISO. If the light levels in the image are higher than that, you will lose the highlights in the image irrecoverably. So, if you hit that problem, it is the one case that you need to reduce exposure, either by closing down the aperture or reducing the shutter speed (usually the latter, since it leaves the picture looking the same, assuming it was short enough to freeze motion blur in the first place)
Using a 70-200 f4 w/IS.

Here are some shots from last Saturday on my practice walk around Asheville where I live. Used AV for the first time, I’ve been shooting Manual since I started and will return to it for NY.
http://www.flickr.com/photos/retsurfer/sets/72157626729792564/
I wish I had your eye for the grab shot - some nice pictures there. Now, the problem you have with using auto ISO is that Canon just doesn't get it. As I said, what you'd like to be able to do is set the slowest shutter speed and largest aperture you can, and just have the camera set the ISO for you, except when you hit base ISO, when you want the camera to reduce exposure via the shutter speed and/or aperture. Unfortunately, you can't set Canon's auto ISO up like that (you can, I understand, with Nikon, although it's not exactly intuitive, it means the right combination of auto ISO limits and auto exposure mode).

So, back to your specific case. I can't see with this kind of work and a lens as good as yours, any reason to use anything other than f/4. The exposure time assuming IS gives 2 stops, could be 1/60 to cover all focal lengths on the zoom (if you were really cute, you'd want to change shutter speed as you zoom). Now as a rough estimate using the 'sunny 16' rule, in most normal daylight illumination, you will risk blowing the highlights at 100 ISO, so in that situation, you'd want to reduce the shutter speed.

The best compromise might be to stick to Av, f/4 at 100ISO, as soon as you see the shutter speed drop below 1/60 swap to M, f/4, 1/60, auto ISO. If the ISO goes back down to 100, swap to Av again. On my canons, as I remember, M mode retains settings between uses, so you can set the f/4, 1/60 in advance, and just change mode. You have to go into a menu to select auto ISO, though.

Canon manages to catch you allways. Another way of working would be Av, with the rear wheel assigned to ISO - then when the shutter speed goes down to your limit, you just boost the ISO to keep it there. So far as I know, youi can't do that, either.

Maybe someone with more familiarity with the 60D's auto ISO can suggest something better, or even more helpfully, point out that I'm wrong about the 60D's limitations..
--
Bob
--
Bob
 
I don't use auto ISO because there is no way to compensate the exposure in that mode. If you're happy with default metering there's no reason not to use auto ISO.
You can use HTP to get around that.
I must admit I never thought of that, but 0 or 1 step under is too limited a range of choices for me.
If you're shooting RAW, and perfecting the tone curve in a converter, apparent (embedded JPEG) exposure is irrelevant at high ISOs; only absolute sensor exposure affects noise in any significant way. Even JPEG allows small adjustments to the tone curve.
I checked this out a few days ago with some real-world test shots at 1600 and 3200, and found that I prefer to suffer a little noise to preserve the shadows - the 3200 image looked better after my normal PP. I need to do a little more testing.
In fact, HTP should be default at high ISOs, because it makes the RAW files smaller and gives an extra stop of highlights as free lunch (no noise increase of any significance).
Interesting. I've not heard anyone else say that (HTP should be default at high ISO). Would you say more about your thinking on this?
The advantage of analog gain decreases as you climb the ISO scale with Canon DSLRs (it has almost no value at all on many other manufacturers' sensors).
Not sure what you mean by "the advantage of analog gain". Advantage over what, in what way?
At a certain point, it pays to just move middle gray down to a lower RAW value, as this reduces the number of useless, incompressible noisy bits, and leaves more room for highlights. You get more highlights if you need them, and smaller files if you don't use them extensively across the frame.
Fair enough, but as above, when I do that I don't like the crushed/blocked shadows, at least in the case of the scene I shot in that test.
 
If you're shooting RAW, and perfecting the tone curve in a converter, apparent (embedded JPEG) exposure is irrelevant at high ISOs; only absolute sensor exposure affects noise in any significant way. Even JPEG allows small adjustments to the tone curve.
I checked this out a few days ago with some real-world test shots at 1600 and 3200, and found that I prefer to suffer a little noise to preserve the shadows - the 3200 image looked better after my normal PP. I need to do a little more testing.
See below.
In fact, HTP should be default at high ISOs, because it makes the RAW files smaller and gives an extra stop of highlights as free lunch (no noise increase of any significance).
Interesting. I've not heard anyone else say that (HTP should be default at high ISO). Would you say more about your thinking on this?
The advantage of analog gain decreases as you climb the ISO scale with Canon DSLRs (it has almost no value at all on many other manufacturers' sensors).
Not sure what you mean by "the advantage of analog gain". Advantage over what, in what way?
Advantage over simply scaling the RAW data. Gain is only meaningful when something is gained upon.
.
At a certain point, it pays to just move middle gray down to a lower RAW value, as this reduces the number of useless, incompressible noisy bits, and leaves more room for highlights. You get more highlights if you need them, and smaller files if you don't use them extensively across the frame.
Fair enough, but as above, when I do that I don't like the crushed/blocked shadows, at least in the case of the scene I shot in that test.
That's not the RAW data; that's your converter or its settings. For example, ACR's "Exposure" slider is not a linear gain adjustment; "Blacks" must be moved, too.

--
John

 
The advantage of analog gain decreases as you climb the ISO scale with Canon DSLRs (it has almost no value at all on many other manufacturers' sensors).
Not sure what you mean by "the advantage of analog gain". Advantage over what, in what way?
Advantage over simply scaling the RAW data.
So you're saying, IOW, once you get into high ISO, you're better off "scaling the raw data", right? I've heard this before, hence the tests I did, but I've yet to see the advantage of it.
At a certain point, it pays to just move middle gray down to a lower RAW value, as this reduces the number of useless, incompressible noisy bits, and leaves more room for highlights. You get more highlights if you need them, and smaller files if you don't use them extensively across the frame.
I understand "You get more highlights if you need them," but I don't understand "if you don't use them extensively across the frame." What are "them" - highlights? How do you (or don't you) "use them extensively across the frame"?
Fair enough, but as above, when I do that I don't like the crushed/blocked shadows, at least in the case of the scene I shot in that test.
That's not the RAW data; that's your converter or its settings. For example, ACR's "Exposure" slider is not a linear gain adjustment; "Blacks" must be moved, too.
So that's not "scaling the raw data"? Okay, how do I scale the raw data? IOW (this is what I really want to know), how do I avoid blocked shadows with the kind of "middle grey down" raw data you're advocating?

(I get a feeling I'm close to understanding something new here. Thanks for sticking with me. :-))
 
The advantage of analog gain decreases as you climb the ISO scale with Canon DSLRs (it has almost no value at all on many other manufacturers' sensors).
Not sure what you mean by "the advantage of analog gain". Advantage over what, in what way?
Advantage over simply scaling the RAW data.
So you're saying, IOW, once you get into high ISO, you're better off "scaling the raw data", right? I've heard this before, hence the tests I did, but I've yet to see the advantage of it.
There is a theoretical advantage that digital amplification is noiseless, if properly done the and the digitisation is good enough to capture all the analog information. Typically, variable analog gain stops at that point (when all the sensor information is digitised) and the camera uses digital gain changes from there on. The problem is that sometimes the implementation is not 'good enough'. Typically, camera processors use 16 bit integer arithmentic, and rounding errors can cause something quite similar to noise. Do it on your computer and the same thing can be done with accurate floating point arritmetic, so long as your software tools allow.
At a certain point, it pays to just move middle gray down to a lower RAW value, as this reduces the number of useless, incompressible noisy bits, and leaves more room for highlights. You get more highlights if you need them, and smaller files if you don't use them extensively across the frame.
I understand "You get more highlights if you need them," but I don't understand "if you don't use them extensively across the frame." What are "them" - highlights? How do you (or don't you) "use them extensively across the frame"?
John is talking about file size after the lossless compression that occurs in raw file formats. This compression is foiled by real or false detail. Real detail would be highlights captured rather than clipped. false detail would be read noise captured unnecessarily due to too high an analog gain.
That's not the RAW data; that's your converter or its settings. For example, ACR's "Exposure" slider is not a linear gain adjustment; "Blacks" must be moved, too.
So that's not "scaling the raw data"? Okay, how do I scale the raw data? IOW (this is what I really want to know), how do I avoid blocked shadows with the kind of "middle grey down" raw data you're advocating?
I don't know how John does it. I use dcraw, which can be made to make a linear 16 bit output file. Then you can use an image editor that can handle 16 bit files (PS will do, but I use Cinepaint) to apply exactly the tone curve that you want.
--
Bob
 
The advantage of analog gain decreases as you climb the ISO scale with Canon DSLRs (it has almost no value at all on many other manufacturers' sensors).
Not sure what you mean by "the advantage of analog gain". Advantage over what, in what way?
Advantage over simply scaling the RAW data.
So you're saying, IOW, once you get into high ISO, you're better off "scaling the raw data", right? I've heard this before, hence the tests I did, but I've yet to see the advantage of it.
Did you look at the size of the RAW files? Compare highlight retention?
At a certain point, it pays to just move middle gray down to a lower RAW value, as this reduces the number of useless, incompressible noisy bits, and leaves more room for highlights. You get more highlights if you need them, and smaller files if you don't use them extensively across the frame.
I understand "You get more highlights if you need them," but I don't understand "if you don't use them extensively across the frame." What are "them" - highlights? How do you (or don't you) "use them extensively across the frame"?
I brought in no new plural nouns after "highlights", so that is what it refers to. You don't use them extensively across the frame when you just have a few small neon lights in a scene, for example.
Fair enough, but as above, when I do that I don't like the crushed/blocked shadows, at least in the case of the scene I shot in that test.
That's not the RAW data; that's your converter or its settings. For example, ACR's "Exposure" slider is not a linear gain adjustment; "Blacks" must be moved, too.
So that's not "scaling the raw data"? Okay, how do I scale the raw data? IOW (this is what I really want to know), how do I avoid blocked shadows with the kind of "middle grey down" raw data you're advocating?
You have to use the "Blacks" and "Recovery" sliders (or equivalent) in many converters to get a normal-looking tone curve (or the one you want), because the converters are not always designed well for pushing and pulling. Their "exposure" sliders often give special treatment to shadows and highlights, anchored to the original RAW values. Only the midtones can be truly assumed to be adjusted like the numbers on the slider suggest.

I sometimes scale the actual RAW data, with no tone curve applied, and this is where you can see that ISO 6400, for example, has no advantage over 3200 with -1EC more, and very little over 1600 with -2EC, marginally better than 800 with -3EC, but ISO 100 with -6EC would be an absolute disaster with a Canon DSLR.
(I get a feeling I'm close to understanding something new here. Thanks for sticking with me. :-))
--
John

 
So you're saying, IOW, once you get into high ISO, you're better off "scaling the raw data", right? I've heard this before, hence the tests I did, but I've yet to see the advantage of it.
Did you look at the size of the RAW files?
No, I don't have a problem with file size. It's not why I did the test.
Compare highlight retention?
Highlights looked similarly fine in both shots, but the shadows looked awful in the pushed 1600 shot. So the solution to a problem I don't have causes a big new problem.
You don't use them [highlights] extensively across the frame when you just have a few small neon lights in a scene, for example.
I'm thrown by the idea of "using highlights" in an image file or PP. It sounds like a mistranslation, or some special meaning that requires a significant explanation. You just mean "when there are a lot of highlights in the image", no?

(Hey, this would be much easier if you said what you mean in simple every-day terms and explained how to unblock the shadows, rather than me playing Twenty Questions. :-))
How do I avoid blocked shadows with the kind of "middle grey down" raw data you're advocating?
You have to use the "Blacks" and "Recovery" sliders (or equivalent) in many converters to get a normal-looking tone curve (or the one you want), because the converters are not always designed well for pushing and pulling. Their "exposure" sliders often give special treatment to shadows and highlights, anchored to the original RAW values. Only the midtones can be truly assumed to be adjusted like the numbers on the slider suggest.
Okay, but there's no way the blacks and recovery sliders alone can make the 1600 shot resemble the 3200 shot (I'm using ACR). In fact they work in entirely the wrong direction, raising the black and white points rather than lowering them.
I sometimes scale the actual RAW data, with no tone curve applied, and this is where you can see that ISO 6400, for example, has no advantage over 3200 with -1EC more, and very little over 1600 with -2EC, marginally better than 800 with -3EC, but ISO 100 with -6EC would be an absolute disaster with a Canon DSLR.
What software do you use for that? Is it possible with ACR?
(I get a feeling I'm close to understanding something new here. Thanks for sticking with me. :-))
(Not feeling any closer to understanding, but thanks for sticking with me. :-))
 
I understand "You get more highlights if you need them," but I don't understand "if you don't use them extensively across the frame." What are "them" - highlights? How do you (or don't you) "use them extensively across the frame"?
John is talking about file size after the lossless compression that occurs in raw file formats. This compression is foiled by real or false detail. Real detail would be highlights captured rather than clipped. false detail would be read noise captured unnecessarily due to too high an analog gain.
It's not about file size. I don't understand what he means by " use highlights extensively across the frame". Use them for what, to do what, how, why?
How do I scale the raw data? IOW (this is what I really want to know), how do I avoid blocked shadows with the kind of "middle grey down" raw data you're advocating?
I don't know how John does it. I use dcraw, which can be made to make a linear 16 bit output file. Then you can use an image editor that can handle 16 bit files (PS will do, but I use Cinepaint) to apply exactly the tone curve that you want.
So you couldn't do that with DPP, for instance? The "shoot at 1600 and push" idea only works with specialist software and esoteric PP?
 
I sometimes scale the actual RAW data, with no tone curve applied, and this is where you can see that ISO 6400, for example, has no advantage over 3200 with -1EC more, and very little over 1600 with -2EC, marginally better than 800 with -3EC, but ISO 100 with -6EC would be an absolute disaster with a Canon DSLR.
What software do you use for that? Is it possible with ACR?
Can anyone tell me how to "scale the actual RAW data, with no tone curve applied"?
 
I understand "You get more highlights if you need them," but I don't understand "if you don't use them extensively across the frame." What are "them" - highlights? How do you (or don't you) "use them extensively across the frame"?
John is talking about file size after the lossless compression that occurs in raw file formats. This compression is foiled by real or false detail. Real detail would be highlights captured rather than clipped. false detail would be read noise captured unnecessarily due to too high an analog gain.
It's not about file size. I don't understand what he means by " use highlights extensively across the frame". Use them for what, to do what, how, why?
What I would guess hem means is 'take a photograph of a scene with extended tonal range in which there are many points of maximum brightness spread across the frame'.
How do I scale the raw data? IOW (this is what I really want to know), how do I avoid blocked shadows with the kind of "middle grey down" raw data you're advocating?
I don't know how John does it. I use dcraw, which can be made to make a linear 16 bit output file. Then you can use an image editor that can handle 16 bit files (PS will do, but I use Cinepaint) to apply exactly the tone curve that you want.
So you couldn't do that with DPP, for instance? The "shoot at 1600 and push" idea only works with specialist software and esoteric PP?
Well, one of the problems with not conforming to the film metaphor which the camera manufacturers foist on us is that you have to find alternative means to an end. having said that, I wouldn't say that dcraw is 'specialist', nor that adjustment of tone curves in PS is 'esoteric PP'. Now I do it, I find it simpler than the alternative - batch convert with dcraw and then adjust to taste (in my case using cinepaint, not PS, but you pays your money, you takes your choice).
--
Bob
 
I sometimes scale the actual RAW data, with no tone curve applied, and this is where you can see that ISO 6400, for example, has no advantage over 3200 with -1EC more, and very little over 1600 with -2EC, marginally better than 800 with -3EC, but ISO 100 with -6EC would be an absolute disaster with a Canon DSLR.
What software do you use for that? Is it possible with ACR?
Can anyone tell me how to "scale the actual RAW data, with no tone curve applied"?
I think that I did.
--
Bob
 
I don't understand what he means by " use highlights extensively across the frame". Use them for what, to do what, how, why?
What I would guess hem means is 'take a photograph of a scene with extended tonal range in which there are many points of maximum brightness spread across the frame'.
That's the only sense I can make of it.
So you couldn't do that with DPP, for instance? The "shoot at 1600 and push" idea only works with specialist software and esoteric PP?
Well, one of the problems with not conforming to the film metaphor which the camera manufacturers foist on us is that you have to find alternative means to an end. having said that, I wouldn't say that dcraw is 'specialist', nor that adjustment of tone curves in PS is 'esoteric PP'.
If you can't do it with the software provided by the manufacturer, or with popular general purpose raw converters, I'd say it's a special application. I adjust tone curves all the time and have been for years, but I've never had to create one from scratch to "scale linear data", so I'd say that's an unusual task for the average raw-shooting photographer.
Now I do it, I find it simpler than the alternative - batch convert with dcraw and then adjust to taste (in my case using cinepaint, not PS, but you pays your money, you takes your choice).
So if I installed dcraw it would be obvious, or trivial to work out, what settings I need, file format, etc? I could use a little help here! (Jeez, anyone'd think you guys don't want me to be able to try this stuff.) :-)
 
I don't understand what he means by " use highlights extensively across the frame". Use them for what, to do what, how, why?
What I would guess hem means is 'take a photograph of a scene with extended tonal range in which there are many points of maximum brightness spread across the frame'.
That's the only sense I can make of it.
So you couldn't do that with DPP, for instance? The "shoot at 1600 and push" idea only works with specialist software and esoteric PP?
Well, one of the problems with not conforming to the film metaphor which the camera manufacturers foist on us is that you have to find alternative means to an end. having said that, I wouldn't say that dcraw is 'specialist', nor that adjustment of tone curves in PS is 'esoteric PP'.
If you can't do it with the software provided by the manufacturer, or with popular general purpose raw converters, I'd say it's a special application.
dcraw is a popular general purpose raw converter. In fact, it's the basis for several popular, general purpose raw converters. So, it's not exactly fair to call it a 'special application'. It is also free, so it is hardly a financial drain.
I adjust tone curves all the time and have been for years, but I've never had to create one from scratch to "scale linear data", so I'd say that's an unusual task for the average raw-shooting photographer.
Well, it depends whether you want to do it in stages or not. If so, the first thing you can do is to make a straight curve to produce the full tonal range or else use the 'levels' adjustment to set 'white' at the top end of the histogram you get from the conversion.
Thereafter, do the curves adjustment as you would normally.
Now I do it, I find it simpler than the alternative - batch convert with dcraw and then adjust to taste (in my case using cinepaint, not PS, but you pays your money, you takes your choice).
So if I installed dcraw it would be obvious, or trivial to work out, what settings I need, file format, etc? I could use a little help here! (Jeez, anyone'd think you guys don't want me to be able to try this stuff.) :-)
I'm happy to help you along. If you get the raw dcraw, it's a command line tool. There are plenty of GUI's but for this the command line is easier. Put all the raw files in a working folder, cd to that folder and then enter
dcraw -4 -T *.CR2 (or whatever the file suffix for your brand or raw is)
'-4' means linear, 16 bit. -T means produce a TIFF file.

dcraw will convert every raw file in the folder to a 16 bit 'TIFF', named as the original raw. You can then read these one by one into PS, and do the desired tone mapping, colour balance and anything else you want, as above.
--
Bob
 
dcraw is a popular general purpose raw converter.
Talking about linear conversion, not dcraw specifically. Anyway, that's irrelevant. The problem is that whenever I see people advocating "shoot at 1600 and push", they don't add that that you have to process in a significantly different way to make it pay off.
If you get the raw dcraw, it's a command line tool. There are plenty of GUI's but for this the command line is easier. Put all the raw files in a working folder, cd to that folder and then enter
dcraw -4 -T *.CR2 (or whatever the file suffix for your brand or raw is)
'-4' means linear, 16 bit. -T means produce a TIFF file.
Excellent! Exactly what I asked for eleven posts ago. :-)

I need to shoot some fresh test shots. I'll push some that way and let you know how I go. Thanks a bunch.
 
dcraw is a popular general purpose raw converter.
Talking about linear conversion, not dcraw specifically. Anyway, that's irrelevant. The problem is that whenever I see people advocating "shoot at 1600 and push", they don't add that that you have to process in a significantly different way to make it pay off.
The other side of the coin is that unless you are suffiently into these things to be aware enough of that to think it's obvious, there probably isn't much benefit to 'shoot at 1600 and push' - the camera does an adequate digital push, it's just that you can do a fraction better in the computer. The main thing for me is not the output, but that losing the tyranny of ISO frees up your shooting style. Unfortunately, it makes some other things more complex, because the camera's UI is based around the film metaphor and ISO.
If you get the raw dcraw, it's a command line tool. There are plenty of GUI's but for this the command line is easier. Put all the raw files in a working folder, cd to that folder and then enter
dcraw -4 -T *.CR2 (or whatever the file suffix for your brand or raw is)
'-4' means linear, 16 bit. -T means produce a TIFF file.
Excellent! Exactly what I asked for eleven posts ago. :-)

I need to shoot some fresh test shots. I'll push some that way and let you know how I go. Thanks a bunch.
It can be fun as well, though the PP strategies you adopt change.
--
Bob
 
The other side of the coin is that unless you are suffiently into these things to be aware enough of that to think it's obvious, there probably isn't much benefit to 'shoot at 1600 and push' - the camera does an adequate digital push, it's just that you can do a fraction better in the computer.
To me the more important thing is the much increased headroom (DR) when using the 1600 and push in post approach. In fact, my approach when shooting in stage light situations is to find the brightest spot, set manual exposure to not blow this at ISO 1600 (or whatever your camera's best base for this approach is) and be done with it. No worrying about exposure anymore. I even did this back in the film days.

Of course, if don't have any of the bright spots in the image and DOF and shake or motion blur is not an issue, you gain IQ by increasing exposure, so ideally you adjust what you consider the bright spot when there is time for it.
 
The other side of the coin is that unless you are suffiently into these things to be aware enough of that to think it's obvious, there probably isn't much benefit to 'shoot at 1600 and push' - the camera does an adequate digital push, it's just that you can do a fraction better in the computer.
To me the more important thing is the much increased headroom (DR) when using the 1600 and push in post approach. In fact, my approach when shooting in stage light situations is to find the brightest spot, set manual exposure to not blow this at ISO 1600 (or whatever your camera's best base for this approach is) and be done with it. No worrying about exposure anymore. I even did this back in the film days.
Yes, you are absolutely right. that is the big extra freedom. This obsessing about 'correct' exposure is very unproductive - I think a lot of people have invested in it, a lot of books have been written about it, and people just aren't willing to let it go.
Of course, if don't have any of the bright spots in the image and DOF and shake or motion blur is not an issue, you gain IQ by increasing exposure, so ideally you adjust what you consider the bright spot when there is time for it.
Again, a key bit of understanding on you part - you gain IQ by increasing exposure , not by reducing the ISO. In part we are talking about 'worrying about exposure', but 'worrying' about it for the right reasons, to control what the picture looks like.
--
Bob
 
I need to shoot some fresh test shots. I'll push some that way and let you know how I go. Thanks a bunch.
It can be fun as well, though the PP strategies you adopt change.
They sure do. :-)

Here are some test images. It's not a great scene but it illustrates what I've generally found.

This one is ETTR at 1/80th at 6400, processed in ACR with default settings (no NR or sharpening), and resized in PS without sharpening:



The appearance is realistic enough so I've used this image as my baseline - I tried to get the other images to match the tonal range.

This one is 1/80th at 1600, processed in ACR as above with Exposure = +2, and resized in PS:



The colour is similar, it has a bit more contrast, and appears noisier (easily fixed).

1/80th at 1600 again, processed in DPP with default settings (no NR or sharpening), Brightness = +2, and resized in PS:



Very similar to the 6400 shot, with slightly less apparent noise in the full size image.

1/80th at 1600 again, following your advice to do an uncorrected linear conversion to 16bit TIFF (I used DPP with Brightness = 0), and processed in PS:



To get the RGB levels similar to those of the 6400 image in all three areas of the WhiBal card, I used Levels with highlights input level = 32 to get a similar white point, midtone input level = 2.2 to get a similar gamma, then a curve with one point at input = 100, output = 128, to brighten the middle grey.

Contrast is low. The colour is not too bad on the WhiBal card and the picture, but it's a complete mess in the shadows. Because of the difference in appearance I'm not convinced that the apparent noise is any lower than in the other images.

But what if we do some correction in the raw converter? 1/80th at 1600 again, DPP with Brightness = +2, and processed the TIFF in PS:



RGB levels equalisation required highlights input level = 128, midtone input level = 2.2, and the same 100/128 curve.

This one is much better than doing all the correction in PS, but not better looking than any of the other three shots due to the aberrant colour in the shadows.

So at least with the way I've approached it (trying to match the levels on the WhiBal card), shooting at 1600 and pushing doesn't give a decisive noise advantage, and there is a significant penalty in shadow colour. I suspect you're doing something quite different to get a normal-looking image from the unpushed linear TIFFs. Would you tell me more about what you do Bob? I'll make these raw files available if you'd like, or we could work on some of yours... Thanks.
 

Keyboard shortcuts

Back
Top