Must I edit RAW images?

If you're going to do this kind of thing processing for editing is a good idea.
Oh boy, I dunno what that means. Searching for it now, tho. Tks.
The exposure that yields the best looking camera-produced JPEG is usually not the exposure that yields the best data for editing.

If you want a good looking out of camera JPEG, you generally expose such that (in the context of the selected ISO) the midtones in the scene map to midtones in the JPEG.

If you want the best data for editing, you want to use the maximin exposure that doesn't blow out unwanted highlights. You can get a reasonable approximation of blown highlights by enabling the "highlight alert" feature. This will cause blown (and almost blown) highlights to blink when reviewing the image.

To get the maximum exposure, set your 70D to ISO 100 (this is the base ISO). and use the largest exposure that doesn't result in important highlights blinking.

An "important" highlight, is any highlight where you want to retain detail. For instance you probably want to retain detail in the whitewalls on a tire, but you may not need detail in a small glint of the sun reflecting off a piece of chrome.

You increase exposure by using a longer shutter speed and/or a wider aperture.

In terms of sharpness, you want to use an aperture that yields sufficient depth of field. Be wary of stopping down too much, as you will get noticeable softness from diffraction at small apertures.

Once you have an aperture dialed in, adjust the exposure by altering your shutter speed. If the resulting speed is too slow for you to reliably hand hold, use a tripod. if you don't have a tripod, use the slowest shutter that you can handhold, and use Auto-ISO to get an appropriate ISO.

If you are able to get the high exposure at ISO, then your camera produced JPEG may look to light. However, you will end up having more detail, and less noise in the shadows, allowing you to brighten the shadows and still retain quality.

The technique of shooting at base ISO and maximizing exposure is call Expose To The Right (ETTR) because it usually results in an image with a histogram shifted towards the right.

Once you have the raw data, you can do your adjustments when you process the raw. If you need more flexibility, you can process to 16 bit RGB, and do the adjustments with layers and other techniques.
 
Last edited:
If you're going to do this kind of thing processing for editing is a good idea.
Oh boy, I dunno what that means. Searching for it now, tho. Tks.
I think that I talked about it up the top of the thread. What it means is processing your raw file to produce an output file that's good for editing, rather than viewing. So, when you process you set the white balance, use a really low black point, a high white point and a linear tone curve. Save it to a wide data format such as 16-bit or 32-bit floating point TIFF.
Are you sure about a linear output? I forget, but I think that would be for a special purpose, which a normal user would ordinarily never use. By default isn't it encoded with the same curve as a jpeg? It should be just done correctly by default, and I never have to worry about that as long as I save as a .png or 16-bit tiff file. I think the important thing is, you don't have to worry about it.

And unless I have to rescue a difficult file, usually the only thing I do with the white and black levels is to make sure they are not clipped, or sometimes add highlight compression. As you mentioned, I think any finesse is best left to a photo editor like Gimp or PS. I think otherwise we risk making this way too difficult for beginners.
It's kind of the stills equivalent of a video 'log' file. It'll look to grey and uncontrasty to the eye, but it means that you haven't lost any (or maybe much) raw information. Then you're going to set the white, black points and tone curve in the editor. It's a bit of a faff for normal use, but if you're using different layers it gives you much more flexibility with each of the layers. Whilst a lot of raw processing tools offer some 'editing' functions they rarely offer as much as a tool like the Gimp or Photoshop. Typically they don't have layers and all the paraphernalia that goes with them
Also, to tell you the truth, I think jpegs work just fine most of the time -- much better than expected. Even if working from raw files, I can rarely see any improvement over using intermediate tiff files, as long as I don't have to make large changes.
 
Last edited:
I just think asking beginners to wrestle with the camera, plus raw developers plus photo editors is cruel and unusual punishment. The less done with a raw developer the better.
 
If you're going to do this kind of thing processing for editing is a good idea.
Oh boy, I dunno what that means. Searching for it now, tho. Tks.
The exposure that yields the best looking camera-produced JPEG is usually not the exposure that yields the best data for editing.

If you want a good looking out of camera JPEG, you generally expose such that (in the context of the selected ISO) the midtones in the scene map to midtones in the JPEG.

If you want the best data for editing, you want to use the maximin exposure that doesn't blow out unwanted highlights. You can get a reasonable approximation of blown highlights by enabling the "highlight alert" feature. This will cause blown (and almost blown) highlights to blink when reviewing the image.

To get the maximum exposure, set your 70D to ISO 100 (this is the base ISO). and use the largest exposure that doesn't result in important highlights blinking.

An "important" highlight, is any highlight where you want to retain detail. For instance you probably want to retain detail in the whitewalls on a tire, but you may not need detail in a small glint of the sun reflecting off a piece of chrome.

You increase exposure by using a longer shutter speed and/or a wider aperture.

In terms of sharpness, you want to use an aperture that yields sufficient depth of field. Be wary of stopping down too much, as you will get noticeable softness from diffraction at small apertures.

Once you have an aperture dialed in, adjust the exposure by altering your shutter speed. If the resulting speed is too slow for you to reliably hand hold, use a tripod. if you don't have a tripod, use the slowest shutter that you can handhold, and use Auto-ISO to get an appropriate ISO.

If you are able to get the high exposure at ISO, then your camera produced JPEG may look to light. However, you will end up having more detail, and less noise in the shadows, allowing you to brighten the shadows and still retain quality.

The technique of shooting at base ISO and maximizing exposure is call Expose To The Right (ETTR) because it usually results in an image with a histogram shifted towards the right.

Once you have the raw data, you can do your adjustments when you process the raw. If you need more flexibility, you can process to 16 bit RGB, and do the adjustments with layers and other techniques.
I do have the highlight alert 'blinkies' turned on on my 70D, and I watch the histogram to set the exposure. And PS's raw app has a "Highlight clipping warning" button on its histogram (and a "Shadow clipping warning" button.

Thank you, and again I sure appreciate all the help. I'm preparing a list of all the images I need, and I'm currently at 300 ... maybe more 🙂
 
Last edited:
Also, to tell you the truth, I think jpegs work just fine most of the time -- much better than expected. Even if working from raw files, I can rarely see any improvement over using intermediate tiff files, as long as I don't have to make large changes.
In many cases carefully converting the raw to a JPEG is more than good enough. However, there are situations where different parts of the image need significantly processing. In that sort of situation, some people will find it easier to convert the raw to a 16 bit per channel RGB file, and then make the adjustments.

Consider a scene where we have the interior of a room. One side is lit by tungsten, the other side is much brighter due to sunlight streaming through the window. There’s an open interior door open to a dimly lit interior hallway, and a window showing a brightly lit exterior scene. Now, someone who is familiar with Adobe’s camera raw might be able to address all of these issues when processing the raw. The varying color temps of different parts of the scene, and the dramatic difference in illumination between the interior and the exterior. On the other hand, some may be more comfortable processing the raw to a 16 bit per channel RGB, and then taking their time to address these issues using layers.

It’s all a matter of the needs of the image, and the photographer’s preference. There are usually multiple reasonable workflows to choose from.
 
Also, to tell you the truth, I think jpegs work just fine most of the time -- much better than expected. Even if working from raw files, I can rarely see any improvement over using intermediate tiff files, as long as I don't have to make large changes.
In many cases carefully converting the raw to a JPEG is more than good enough. However, there are situations where different parts of the image need significantly processing. In that sort of situation, some people will find it easier to convert the raw to a 16 bit per channel RGB file, and then make the adjustments.

Consider a scene where we have the interior of a room. One side is lit by tungsten, the other side is much brighter due to sunlight streaming through the window. There’s an open interior door open to a dimly lit interior hallway, and a window showing a brightly lit exterior scene. Now, someone who is familiar with Adobe’s camera raw might be able to address all of these issues when processing the raw. The varying color temps of different parts of the scene, and the dramatic difference in illumination between the interior and the exterior. On the other hand, some may be more comfortable processing the raw to a 16 bit per channel RGB, and then taking their time to address these issues using layers.

It’s all a matter of the needs of the image, and the photographer’s preference. There are usually multiple reasonable workflows to choose from.
Ah, I'm with you now. For what I need now, I don't think I better delve that deep into things. I don't have an eye, yet, and what I find is it's already very easy for me to over-edit ... I fiddle and tweak, think I got it lookin' good, look away for a few moments then back and think, What the heck?! 😀 🤷

I have a feeling my next shoot with raw+jpeg is going to reveal to me that jpeg is sufficient for what I need now, and I'll continue editing and see if I can get a feel for "some but not too much". 🤞 👍
 
If you're going to do this kind of thing processing for editing is a good idea.
Oh boy, I dunno what that means. Searching for it now, tho. Tks.
The exposure that yields the best looking camera-produced JPEG is usually not the exposure that yields the best data for editing.

If you want a good looking out of camera JPEG, you generally expose such that (in the context of the selected ISO) the midtones in the scene map to midtones in the JPEG.

If you want the best data for editing, you want to use the maximin exposure that doesn't blow out unwanted highlights. You can get a reasonable approximation of blown highlights by enabling the "highlight alert" feature. This will cause blown (and almost blown) highlights to blink when reviewing the image.

To get the maximum exposure, set your 70D to ISO 100 (this is the base ISO). and use the largest exposure that doesn't result in important highlights blinking.

An "important" highlight, is any highlight where you want to retain detail. For instance you probably want to retain detail in the whitewalls on a tire, but you may not need detail in a small glint of the sun reflecting off a piece of chrome.

You increase exposure by using a longer shutter speed and/or a wider aperture.

In terms of sharpness, you want to use an aperture that yields sufficient depth of field. Be wary of stopping down too much, as you will get noticeable softness from diffraction at small apertures.

Once you have an aperture dialed in, adjust the exposure by altering your shutter speed. If the resulting speed is too slow for you to reliably hand hold, use a tripod. if you don't have a tripod, use the slowest shutter that you can handhold, and use Auto-ISO to get an appropriate ISO.

If you are able to get the high exposure at ISO, then your camera produced JPEG may look to light. However, you will end up having more detail, and less noise in the shadows, allowing you to brighten the shadows and still retain quality.

The technique of shooting at base ISO and maximizing exposure is call Expose To The Right (ETTR) because it usually results in an image with a histogram shifted towards the right.

Once you have the raw data, you can do your adjustments when you process the raw. If you need more flexibility, you can process to 16 bit RGB, and do the adjustments with layers and other techniques.
I do have the highlight alert 'blinkies' turned on on my 70D, and I watch the histogram to set the exposure. And PS's raw app has a "Highlight clipping warning" button on its histogram (and a "Shadow clipping warning" button.

Thank you, and again I sure appreciate all the help. I'm preparing a list of all the images I need, and I'm currently at 300 ... maybe more 🙂
My understanding is that the “blinkies” on your 70D are based on the camera-produced JPEG, and not the actual raw data. While they are good enough to get your exposure into the ballpark, they are not definitive as to whether those pixels are blown out in the raw data. For instance, changing the white balance can change whether or not a JPEG pixel is blown out, but the white balance doesn’t affect the raw data. If you wanted to make your histogram more accurate, you can play games with the white balance setting. That helps with the histogram, but screws with the colors in the camera-produced JPEG.

I am not suggesting that you shouldn’t rely on the blinkies. Just that you should be aware that while they are a useful tool, they are not definitive. If you absolutely wanted to maximize your exposure, you need to do more than look at the blinkies.
 
My understanding is that the “blinkies” on your 70D are based on the camera-produced JPEG, and not the actual raw data. While they are good enough to get your exposure into the ballpark, they are not definitive as to whether those pixels are blown out in the raw data. For instance, changing the white balance can change whether or not a JPEG pixel is blown out, but the white balance doesn’t affect the raw data. If you wanted to make your histogram more accurate, you can play games with the white balance setting. That helps with the histogram, but screws with the colors in the camera-produced JPEG.

I am not suggesting that you shouldn’t rely on the blinkies. Just that you should be aware that while they are a useful tool, they are not definitive. If you absolutely wanted to maximize your exposure, you need to do more than look at the blinkies.
Got it, and yes, several sources say the 70D's highlight alert uses the jpeg not the raw. But I had to start somewhere, and since my next batch will be raw+jpeg, shouldn't the blinkies I see on the 70D's display should match the jpeg I edit on my PC?
 
Also, to tell you the truth, I think jpegs work just fine most of the time -- much better than expected. Even if working from raw files, I can rarely see any improvement over using intermediate tiff files, as long as I don't have to make large changes.
In many cases carefully converting the raw to a JPEG is more than good enough. However, there are situations where different parts of the image need significantly processing. In that sort of situation, some people will find it easier to convert the raw to a 16 bit per channel RGB file, and then make the adjustments.
Yes.
Consider a scene where we have the interior of a room. One side is lit by tungsten, the other side is much brighter due to sunlight streaming through the window. There’s an open interior door open to a dimly lit interior hallway, and a window showing a brightly lit exterior scene. Now, someone who is familiar with Adobe’s camera raw might be able to address all of these issues when processing the raw. The varying color temps of different parts of the scene, and the dramatic difference in illumination between the interior and the exterior. On the other hand, some may be more comfortable processing the raw to a 16 bit per channel RGB, and then taking their time to address these issues using layers.

It’s all a matter of the needs of the image, and the photographer’s preference. There are usually multiple reasonable workflows to choose from.
Yes, there are many such examples.
 
If you're going to do this kind of thing processing for editing is a good idea.
Oh boy, I dunno what that means. Searching for it now, tho. Tks.
The exposure that yields the best looking camera-produced JPEG is usually not the exposure that yields the best data for editing.

If you want a good looking out of camera JPEG, you generally expose such that (in the context of the selected ISO) the midtones in the scene map to midtones in the JPEG.

If you want the best data for editing, you want to use the maximin exposure that doesn't blow out unwanted highlights. You can get a reasonable approximation of blown highlights by enabling the "highlight alert" feature. This will cause blown (and almost blown) highlights to blink when reviewing the image.

To get the maximum exposure, set your 70D to ISO 100 (this is the base ISO). and use the largest exposure that doesn't result in important highlights blinking.

An "important" highlight, is any highlight where you want to retain detail. For instance you probably want to retain detail in the whitewalls on a tire, but you may not need detail in a small glint of the sun reflecting off a piece of chrome.

You increase exposure by using a longer shutter speed and/or a wider aperture.

In terms of sharpness, you want to use an aperture that yields sufficient depth of field. Be wary of stopping down too much, as you will get noticeable softness from diffraction at small apertures.

Once you have an aperture dialed in, adjust the exposure by altering your shutter speed. If the resulting speed is too slow for you to reliably hand hold, use a tripod. if you don't have a tripod, use the slowest shutter that you can handhold, and use Auto-ISO to get an appropriate ISO.

If you are able to get the high exposure at ISO, then your camera produced JPEG may look to light. However, you will end up having more detail, and less noise in the shadows, allowing you to brighten the shadows and still retain quality.

The technique of shooting at base ISO and maximizing exposure is call Expose To The Right (ETTR) because it usually results in an image with a histogram shifted towards the right.

Once you have the raw data, you can do your adjustments when you process the raw. If you need more flexibility, you can process to 16 bit RGB, and do the adjustments with layers and other techniques.
I do have the highlight alert 'blinkies' turned on on my 70D, and I watch the histogram to set the exposure. And PS's raw app has a "Highlight clipping warning" button on its histogram (and a "Shadow clipping warning" button.

Thank you, and again I sure appreciate all the help. I'm preparing a list of all the images I need, and I'm currently at 300 ... maybe more 🙂
My understanding is that the “blinkies” on your 70D are based on the camera-produced JPEG, and not the actual raw data. While they are good enough to get your exposure into the ballpark, they are not definitive as to whether those pixels are blown out in the raw data. For instance, changing the white balance can change whether or not a JPEG pixel is blown out, but the white balance doesn’t affect the raw data. If you wanted to make your histogram more accurate, you can play games with the white balance setting. That helps with the histogram, but screws with the colors in the camera-produced JPEG.
If it's not blown out in the histogram, you can be reasonably assured that the raw file is OK. Of course it's possible to miss some pixels. That's all we've got to go on without extraordinary distractions, and fortunately, it works.

Keep it simple. At least for now.
I am not suggesting that you shouldn’t rely on the blinkies. Just that you should be aware that while they are a useful tool, they are not definitive. If you absolutely wanted to maximize your exposure, you need to do more than look at the blinkies.
Now, if someone wants to explain how the blasted histogram x axis is mapped, that might be a subject for the Sci Tech forum.
 
So my question is: How common is it for a photographer to record images in RAW and not edit them, instead just saving as JPG / PNG?
"I can think of four different approaches to acquisition.
  1. I want the module (raw converter, ib) to do nothing more than open the file without damage. I understand that it will probably be flat and colorless if I do this. I intend to fix the problems in Photoshop.
  2. Although I will refine the image in Photoshop, I would like to be able to make quick, obvious moves in the acquisition module to make life easier later.
  3. I am not interested in the very best quality. I will do whatever is possible in the raw module but I will not work in Photoshop afterward. So I would like to be able to get attractive color from the module.
  4. I want to make the image as perfect as possible in the module. I will intervene later in Photoshop only as necessary."
 
I just think asking beginners to wrestle with the camera, plus raw developers plus photo editors is cruel and unusual punishment. The less done with a raw developer the better.
I agree. Too often these discussions are framed in terms of 'should I be doing x' rather than 'do I want to be doing x'. However, processing and editing of a shot can be a rewarding and enjoyable experience. I quite often 'remaster' my old shots just for the fun of it.
 
This is with 15 sec. in PSE. Obviously I didn't do any layers so I wasn't doing any selective edits. So you don't have to spend a lot of time with editing. Your dark photo of the front marker light is the reason that you need to know how to do a little bit of editing- so you can save photos like that. That one is starting to get marginal though- that one can probably be fixed by editing a JPG but if it was a little darker than that it would be much easier to save by working from a raw file.

87fc83555c634617a491e9f74c4fd528.jpg
 
AWB-White Priority / Standard
AWB-White Priority / Standard

Daylight/Landscape
Daylight/Landscape

This is not intense editing. It’s three or four clicks, then process as jpeg.
OK, got it. Again, I look forward to seeing the raw vs jpeg images I get out of the camera.

BTW I'm using a tripod for every shots, a wireless shutter release, and exposure bracketing. I also just got a circular polarizer since the reflections are out of control
Since you're using exposure bracketing and are getting crazy reflections. By crazy reflections do you mean specular highlights? Have you tried merging any of these into an HDR image? Might be a better cure/result than trying to PP a separate file.

Are you using Strobes or is it all natural light? Even a reflector could help a lot. I guess I'm not sure who i'm talking to here, if the OP or the poster of the images above. It's meant for the car images?!

John
 
OK, got it. Again, I look forward to seeing the raw vs jpeg images I get out of the camera.

BTW I'm using a tripod for every shots, a wireless shutter release, and exposure bracketing. I also just got a circular polarizer since the reflections are out of control
Since you're using exposure bracketing and are getting crazy reflections. By crazy reflections do you mean specular highlights? Have you tried merging any of these into an HDR image? Might be a better cure/result than trying to PP a separate file.

Are you using Strobes or is it all natural light? Even a reflector could help a lot. I guess I'm not sure who i'm talking to here, if the OP or the poster of the images above. It's meant for the car images?!

John
John - I'm the OP and the one who mentioned reflections (@JustUs7 posted those landscapes). And yes specular highlights; they're too harsh and make it so when I try to expose the entire image based on those highlights, the rest of the image is under-exposed ... probably the same thing all photographers encounter at times. I have not yet tried merging different exposures, but will work on that.

Just sunlight, no man-made lighting.
 
On Canon cameras, highlight tone protection (HTP) will help tame the reflections somewhat. Check your manual.
 
Since I already think the RAW shots are excellent, maybe the JPGs will look even better; I intended to test this next when it stops raining every day here in Florida. We'd love to hear your thoughts and feedback.
 
I save raw+jpeg. If the picture needs tweaking, I use the raw file, if the jpeg is fine, I use it as it is.

Although I agree with bobn2 that processing and editing are conceptually different, I often combine them in one session: lens correction, lifting the shadows, tweaking WB, cropping, removing a zit, straightening the horizon. All this can be done in DPP.
That's entirely true. What I'm getting at is that it pays to separate processing and editing in terms of understanding what is going on under the hood. Most tools these days combine operations of both processing and editing. I used to use a technique 'processing for editing', where I'd process the raw file not to be the final image, but to be a good subject for a powerful editor. It's useful when you need to do operations involving multiple layers. Essentially you process to a 16-bit or, better, a FP TIFF with a low black and high white point and a linear tone curve, but set up colour temperature and balance. It's kind of the equivalent of a video log file, designed to be suitable for onward grading.
BTW, raw is a word, not an acronym. These are literaly the raw data from the sensor.
I'm so glad someone else said that, so I didn't have to.
 

Keyboard shortcuts

Back
Top