Raw vs jpeg part 2

[Iliah Borg]

Set your camera to ISO 100 and setup a low dynamic range studio scene.

Shoot a raw file with ETTR, and shoot a second raw file exposed for JPEG.

And upload the results.

 

[Ysarex]

Blown highlights are not fine.

When did blown highlights become a big no no for photography ?

Blown higlights are just white, reflections and light sources are too bright for the eye in real life too. It's natural with blown highligts. You did remove the highlights to save details, in real life you can't see details in highlights. Maybe with sunglasses but then everything becomes darker so it's not the same.

I see no objective difference in these reflections and the reflections on the flower for example. This is theory , not photography.

This really is Photo 101. http://digitalprotalk.blogspot.com/2008/03/technique-tuesday-specular-vs-diffuse.html

Let me know if you need more help.

--
" Use the shutter button on the headset cord " - Leonardo Da Vinci

 

[Michael Fryd]

in-camera sRGB JPEG is good.

In what way it is good?

Sometimes the level of quality offered by in-camera sRGB JPEGs is more than good enough for the job at hand

One can't predict how the image will be used 10 years from the moment it was shot. Or even 2 hours. There is no such thing in photography as "job at hand".

--
http://www.libraw.org/

True. There are rarely absolutes, but sometimes one can make a good guess.

I've done a number of product shots where the model release was very clear on exactly how long the image can be used. It is unlikely that the images will ever be used again.

If I am shooting someone's passport photo, 600px by 600px is all that I need. I could shoot 50 megapixels on the off chance that I might need the image some day, but it is not clear that the slight chance the image might be useful, justifies the added effort of taking such huge files, cataloging, storing and preserving the images.

Obviously, many situations are somewhere in between. It's possible that I may want to use an old image ten years from now, but I suspect it is far more likely that my skill level has continued to improve, and I will be taking far better images in 10 years, and won't be going back to old images.

One might be better off buying lottery tickets. Sure, you might win hundreds of millions of dollars, but it is more likely that the return will be less than the cost.

 

[digidog]

If you're going to talk about homework, you should set a good example. perhaps you could provide proof that the difference between ETTR and exposing the raw for JPEG will always make a noticeable and visible difference in the final image.

The absurd is the last refuge of a pundit without an argument.

I and other's have shown where quality differences exist that are visual. Now you expect us to suggest it always produces the same results? Really? As I wrote, you're at the end of your argument rope with that one!

Again, perhaps you should lead by example.

I provided two references (one by me). You? Nada yet. As usual.

Set your camera to ISO 100 and setup a low dynamic range studio scene.

Shoot a raw file with ETTR, and shoot a second raw file exposed for JPEG.

Post process the images such that they look as close to each other as possible.

Make prints and run tests to see if typical viewers can see a difference between the two.

Sure. Can you do that or not?

You're asking for proof that some people choose to shoot JPEG rather than raw?

I'm absolutely NOT asking for that. I've asked you to do your homework. Were you this difficult when you went to school? I feel for your mom. ;-)

Just read some of these forums. Lots of people choose to shoot JPEG.

Got nothing to do with the raw + JPEG exposure homework you need to get cracking on.

--
Andrew Rodney
Author: Color Management for Photographers
The Digital Dog
http://www.digitaldog.net

 

[digidog]

Set your camera to ISO 100 and setup a low dynamic range studio scene.

Shoot a raw file with ETTR, and shoot a second raw file exposed for JPEG.

And upload the results.

Please. Pretty please. With sugar on top.

 

[Michael Fryd]

If you're going to talk about homework, you should set a good example. perhaps you could provide proof that the difference between ETTR and exposing the raw for JPEG will always make a noticeable and visible difference in the final image.

The absurd is the last refuge of a pundit without an argument.

I and other's have shown where quality differences exist that are visual. Now you expect us to suggest it always produces the same results? Really? As I wrote, you're at the end of your argument rope with that one!

Then why did you disagree with my statement that it sometimes does not make a visible difference?

My position was that sometimes the difference was not visible. You have vehemently denied this. All I am asking is that you back up your statement that there is always a visible difference.

I suspect the issue is that you agree with me that there sometimes is no visible difference, but disagree that there are times when it is OK to expose for JPEG.

These are different issues that you seem to have a difficult time separating.

 

[digidog]

Michael Fryd wrote: I suspect the issue is that you agree with me that there sometimes is no visible difference, but disagree that there are times when it is OK to expose for JPEG.

These are different issues that you seem to have a difficult time separating.

Ask yourself these questions while doing your homework. Ask us when providing the uploaded images with how each was captured.

--
Andrew Rodney
Author: Color Management for Photographers
The Digital Dog
http://www.digitaldog.net

 

[cookedraw]

Blown highlights are not fine.

When did blown highlights become a big no no for photography ?

Blown higlights are just white, reflections and light sources are too bright for the eye in real life too. It's natural with blown highligts. You did remove the highlights to save details, in real life you can't see details in highlights. Maybe with sunglasses but then everything becomes darker so it's not the same.

I see no objective difference in these reflections and the reflections on the flower for example. This is theory , not photography.

This really is Photo 101. http://digitalprotalk.blogspot.com/2008/03/technique-tuesday-specular-vs-diffuse.html

Let me know if you need more help.

Doesn't matter what you call the highlights. They are highlights cause they are bright and the reason the Jpeg make them bright is cause that is what they are exposed for. I don't care if you call the reflections specular or diffuse highlights , sorry.

Look at the whole image of the flower, these highlights is not an issue at all.

You can't say that this kind of highlight is good and this is bad, it's just you that got a hangup on blown highlights cause you pixel peep.

--
" Use the shutter button on the headset cord " - Leonardo Da Vinci

--
" Use the shutter button on the headset cord " - Leonardo Da Vinci

 

[Ysarex]

Blown highlights are not fine.

When did blown highlights become a big no no for photography ?

Blown higlights are just white, reflections and light sources are too bright for the eye in real life too. It's natural with blown highligts. You did remove the highlights to save details, in real life you can't see details in highlights. Maybe with sunglasses but then everything becomes darker so it's not the same.

I see no objective difference in these reflections and the reflections on the flower for example. This is theory , not photography.

This really is Photo 101. http://digitalprotalk.blogspot.com/2008/03/technique-tuesday-specular-vs-diffuse.html

Let me know if you need more help.

Doesn't matter what you call the highlights.

Do you think I made this up? You give me too much credit.

You need to head over to The Strobist and get them straightened out:

Loading…

strobist.blogspot.com

Sorry you missed class that day, but this really is Photo 101

Do your homework:

Loading…

books.google.com

Loading…

en.wikipedia.org

Loading…

www.photokonnexion.com

Loading…

digitalprotalk.blogspot.com

Loading…

www.huevaluechroma.com

Loading…

www.mdavid.com.au

Loading…

books.google.com

Loading…

books.google.com

Loading…

www.slrlounge.com

Loading…

www.digital-photo-secrets.com

Loading…

www.breathingcolor.com

They are highlights cause they are bright and the reason the Jpeg make them bright is cause that is what they are exposed for.

Well duuuuuh; if that is what they are exposed for then why aren't they exposed that way in the raw file? Explain why the color and texture in those petals is recorded in the raw file.

I don't care if you call the reflections specular or diffuse highlights , sorry.

Like I said, head on over to The Strobist and get them straightened out before they misinform more unsuspecting photographers.

Look at the whole image of the flower, these highlights is not an issue at all.

You can't say that this kind of highlight is good and this is bad, it's just you that got a hangup on blown highlights cause you pixel peep.

There it is -- finally the logical argument raises it's head. I should have known! I'm a pixel peeper!!

--
" Use the shutter button on the headset cord " - Leonardo Da Vinci

--
" Use the shutter button on the headset cord " - Leonardo Da Vinci

 

[Old Greenlander]

...

The in camera JPEG is either over exposed because you wish to ideally expose the raw or the raw is under exposed because you ideally exposed the JPEG.

Although you don't address it, this brings up the question as to why cameras are designed in this fashion?

If ETTR is the strategy that provides the best quality, why don't cameras do this by default?

It does, for raw! You just have to understand the basics of CFA's, image capture, raw data and it's encoding and more importantly, how to properly exposure for it.

When you buy film, there's no assumption you're not a rube have have zero idea how to ideally expose for that media. You test. At least some of us in the old days did with film, just as we do with digital.

I'm sorry, I don't see how your answer addresses the question.

If I understand your position, it is that one should use a different exposure strategy when one wants to maximize the quality of the information in the RAW file, as opposed to when one wants to maximize the quality of an in-camera JPEG.

The reason is that the in-camera JPEG processing will produce an image that is "too bright" if the RAW file is properly exposed.

My question was why are modern cameras designed in this fashion? If the best quality is always obtained by ETTR, why doesn't the in-camera meter do this by default? The in-camera JPEG processing would know about this, and take it into account.

If this is the best strategy in the general case, why aren't cameras designed this way?

...

Short answer: Because the metering and processing systems in the cameras are automated.

Automation requires averages. I live in the US Midwest where the corn is now thigh high. When it's ready for harvest the work will all be done using automated machinery. When the corn is ready for harvest all the stalks will be about the same height and all the ears of corn will be about the same size. Consider the automation process if this wasn't the case. What if corn stalks varied in height randomly from .5 meter to 4 meters? What if ears of corn varied in size randomly between 8 centimeters and 80 centimeters. Automaton only works when average conditions exist. The more uniform the conditions the more successful the automation and the less uniform the conditions the less successful the automation.

When conditions aren't very uniform the designers of the automatic machinery must accept a degree of fudge in the results and a degree of loss. In this case they design the machinery to hit the average. Your camera's metering and processing systems are automated machines and since there is a fair degree of variability in lighting conditions (and user expectations) the designers have been forced to target the average.

Another word for average by the way is mean and mean shares it's root with the word mediocre.

Long answer: The general consensus in the photo industry identifies the worst exposure error as detail loss (clipping) in the diffuse highlights. This is the worst error in part because it's not only ugly, but there isn't any remediation for the error except to after-the-fact paint, draw, copy, clone, or just smear some other data into the burn hole. The designers of your camera's automated metering and processing systems are very aware of this of course and know the sting of a camera review that faults a camera as too easily blowing highlights. They're going to adjust a bias into those metering and processing systems to target the average away from rather than toward that error.

The average has to be there, you can't escape it. To deal with the variability of lighting conditions and user expectations the best the camera designers do is give you some limited controls that let you bias those automated systems, eg. EC, contrast and saturation settings, etc.

Now let's look at two photos:

View attachment 5749085

I'm retired now and I spend a lot of time walking around the neighborhood. We have a pet food manufacturer on the edge of the hood and the cat in the bucket was outside the door one day. Both of the photos above are the camera JPEGs. I thought the truck was pretty cool so I took a picture. All the camera picture controls were neutral with WB on auto and in both cases I intervened and set a +.67 EC. Both JPEGs suck and both have clipped highlights. Why is the cat too light and the truck too dark? They both got +.67 EC.

So there's a huge difference in the lighting. There isn't 7 stops of tonal range in the cat photo. My camera sensor records nearly 10 stops. Did I really need the +.67 EC? If I hadn't used it and instead tried to get a best possible JPEG would I still have just as good a raw file? Yeah, pretty much.

My camera lets me reload raw files onto the SD card and re-process them with changes to the picture controls. I popped the cat back in the camera and pulled the exposure -.33 and raised the color temp and contrast to get a better JPEG.

View attachment 5749086

On the left is my cat processed from the raw file and on the right is the new JPEG with -.33 EC and the raised color temp and contrast. If I had set a custom WB when I took the photo I'd be pretty close with the JPEG and the raw file would have -.33 less exposure and no big deal -- not the very best raw file but very serviceable.

Why did I use the +.67 then which produced the sucky JPEG? Habit for one. My goal is always a full sensor exposure. In this case it's exactly the habit I want to have.

Again, my sensor records 3 stops more DR than this photo has in total -- talk about room to spare. The truck is another story entirely. The DR in the truck photo is in the 11 plus stop range -- more than my sensor. When I took that photo I saw that and I thought about that and I made the effort to get that full sensor exposure. And unlike the cat I got a JPEG that looks overall too dark, but still with clipped highlights. What now? How do you deal with that? Tell your crew to start setting up your lights?

Let's look at the options: Back in the camera with the raw file I processed two alternative JPEGs. Here they are:

View attachment 5749087

On your left I pulled -.33 EC and set the contrast to it's lowest values. JPEG is still too dark but I've removed the highlight clipping. CRITICAL POINT: The -.33 EC is less exposure. I would have had to reduce exposure in camera, and fiddle with the picture controls and I still get a sucky JPEG. AND I get an inferior raw file -- SCREW THAT! The scene is lit from the side and has just begun to go backlit with high contrast sun. On the right I increased the exposure +.33 and set the contrast to lowest setting. Now I have what appears to be a decent overall exposure but with highlights nuked into oblivion -- another sucky JPEG. Where's the option in camera for a non-sucky JPEG? Call the lighting crew or settle for sucky.

My Fuji has a special function called DR. It's designed for just this kind of situation and I could have brought that function to bear. This function (or some variant) is common in most cameras now. Nikon has ADL and Canon has HTP. They all do much the same thing. They apply a special tone curve and processing to the JPEG to really haul down the contrast and they manage to keep the highlights from clipping (see bad example above right -- ouch) by forcing an underexposure. Let me repeat that: They force an underexposure. Nooooooooooo! I want as much data recorded as possible. Not less! That's exactly what I don't want.

I'm sorry I can't re-process the raw file and show you this option but it's not possible. To record maximum data I had my ISO at base. To use the DR function in my camera it's required that you raise the ISO (there's the underexposure). For what it's worth that is an option to get a non-sucky JPEG. I could use that and get a pretty good JPEG but not the same JPEG I processed from the raw file. The forced underexposure comes with a hefty price: less recorded data. In this photo I want as much data as I can get.

View attachment 5749088

I'll handle the processing, I won't clip the diffuse highlights because they're not clipped in the raw file and note how I opened up the shadows -- I needed all that data to do that.

Given the lighting condition, subject and my user expectations this photo is way outside the range of the average that our camera's automation is designed to handle. And so all the way back to Andrew's quoted comment that started this thread. I exposed for full sensor saturation and got the best possible raw file with as much data recorded as possible and that allowed me to make my photo my way. From that raw file the camera can only produce sucky. To get a JPEG that's decent the camera's automated systems would force less exposure and less data recorded. You'll still get a photo and even a photo that doesn't suck but you won't get the photo I got -- you'll have to settle for less.

Ysarex

We had this discussion about JPEG/RAW some time ago and you seemed to be very compelling posting this truck with the two ways of processing the picture> I didn't comment at the time

I spent maybe 1 min (nothing fancy) and processed your JPEG picture just out of curiosity. Can anyone tell me which is what?

It is very little practical difference between them, and I used your tiny 0.07 Mb picture. On a real picture differences can be insignificant













--
Old Greenlander
"I show the world the way I see it"
35 years of photography and still learning

 

[digidog]

It is very little practical difference between them, and I used your tiny 0.07 Mb picture. On a real picture differences can be insignificant

There's a substantial difference in the highlight data in the lower image. It's visible in the image, it's visible in the histogram. It is hardly insignificant IMHO.

--
Andrew Rodney
Author: Color Management for Photographers
The Digital Dog

Home

 

[Old Greenlander]

It is very little practical difference between them, and I used your tiny 0.07 Mb picture. On a real picture differences can be insignificant

There's a substantial difference in the highlight data in the lower image. It's visible in the image, it's visible in the histogram. It is hardly insignificant IMHO.

--
Andrew Rodney
Author: Color Management for Photographers
The Digital Dog
http://www.digitaldog.net

So which is RAW and which is Jpeg?

--
Old Greenlander
"I show the world the way I see it"
35 years of photography and still learning

 

[Biological_Viewfinder]

And show us by processing the JPEG to match the processed raw.

He doesn't know how to process a JPEG as well as he does a raw file. That's one of the biggest problems with talking to these people. He can't process a JPEG very well, therefore RAW must be vastly superior. It isn't.

About the only way to do this right is to get a whole lot of people to take different pictures in RAW+JPEG, and the pick out some nice pictures based solely on the content, not anything else except that whatever it's about is pleasant to look at. Then have all of those pictures processed by everyone and find some particularly nice renditions from both the JPEG and the RAW, and finally to have people vote on which is which.

My guess is that almost no one will pick the RAWs out more than just over half the time. And that it is more important about how pleasant the image is in both its capture and its post-processed rendering, rather than its technical merit.

Regardless, this is a silly argument that never ends.
And I'll use whatever format I want on my camera. It's my camera!!

--
There are 10 types of people in this world, those who understand binary and those who don't.

 

[Ysarex]

It is very little practical difference between them, and I used your tiny 0.07 Mb picture. On a real picture differences can be insignificant

There's a substantial difference in the highlight data in the lower image. It's visible in the image, it's visible in the histogram. It is hardly insignificant IMHO.

--
Andrew Rodney
Author: Color Management for Photographers
The Digital Dog
http://www.digitaldog.net

So which is RAW and which is Jpeg?

They're both from the same raw file. They're too small to play this game, but the top one has the blown highlights that identify it as the camera JPEG.

--
Old Greenlander
"I show the world the way I see it"
35 years of photography and still learning

 

[Don Lacy]

It is very little practical difference between them, and I used your tiny 0.07 Mb picture. On a real picture differences can be insignificant

There's a substantial difference in the highlight data in the lower image. It's visible in the image, it's visible in the histogram. It is hardly insignificant IMHO.

--
Andrew Rodney
Author: Color Management for Photographers
The Digital Dog
http://www.digitaldog.net

So which is RAW and which is Jpeg?

--
Old Greenlander
"I show the world the way I see it"
35 years of photography and still learning

Where did you get the Raw file I do not see a link to it in the original post

--
Don Lacy

500px

500px.com

http://www.witnessnature.net/

 

[digidog]

It is very little practical difference between them, and I used your tiny 0.07 Mb picture. On a real picture differences can be insignificant

There's a substantial difference in the highlight data in the lower image. It's visible in the image, it's visible in the histogram. It is hardly insignificant IMHO.

--
Andrew Rodney
Author: Color Management for Photographers
The Digital Dog
http://www.digitaldog.net

So which is RAW and which is Jpeg?

IF the bottom one isnt the raw, it was pisspoorly rendered. And not blown out.

--
Andrew Rodney
Author: Color Management for Photographers
The Digital Dog
http://www.digitaldog.net

 

[Old Greenlander]

It is very little practical difference between them, and I used your tiny 0.07 Mb picture. On a real picture differences can be insignificant

There's a substantial difference in the highlight data in the lower image. It's visible in the image, it's visible in the histogram. It is hardly insignificant IMHO.

--
Andrew Rodney
Author: Color Management for Photographers
The Digital Dog
http://www.digitaldog.net

So which is RAW and which is Jpeg?

--
Old Greenlander
"I show the world the way I see it"
35 years of photography and still learning

Where did you get the Raw file I do not see a link to it in the original post

--
Don Lacy
https://500px.com/lacy
http://www.witnessnature.net/

I downloaded / saved his tiny pictures

--
Old Greenlander
"I show the world the way I see it"
35 years of photography and still learning

 

[Ysarex]

It is very little practical difference between them, and I used your tiny 0.07 Mb picture. On a real picture differences can be insignificant

There's a substantial difference in the highlight data in the lower image. It's visible in the image, it's visible in the histogram. It is hardly insignificant IMHO.

--
Andrew Rodney
Author: Color Management for Photographers
The Digital Dog
http://www.digitaldog.net

So which is RAW and which is Jpeg?

--
Old Greenlander
"I show the world the way I see it"
35 years of photography and still learning

Where did you get the Raw file I do not see a link to it in the original post

--
Don Lacy
https://500px.com/lacy
http://www.witnessnature.net/

I downloaded / saved his tiny pictures

--
Old Greenlander
"I show the world the way I see it"
35 years of photography and still learning

Those are both JPEGs I created from the raw file using the camera processor last night. Neither one is my file processed from the raw original. You misunderstood what's in the post.

 

[Old Greenlander]

It is very little practical difference between them, and I used your tiny 0.07 Mb picture. On a real picture differences can be insignificant

There's a substantial difference in the highlight data in the lower image. It's visible in the image, it's visible in the histogram. It is hardly insignificant IMHO.

--
Andrew Rodney
Author: Color Management for Photographers
The Digital Dog
http://www.digitaldog.net

So which is RAW and which is Jpeg?

IF the bottom one isnt the raw, it was pisspoorly rendered. And not blown out.

--
Andrew Rodney
Author: Color Management for Photographers
The Digital Dog
http://www.digitaldog.net

none of them is actually the raw

i processed the "poor one"from the left of his image and compared with the "good one" he posted on the right

his right side original is now on top and you can see the left of the picture is cropped

the bottom is my quick fix from his RAW processing

--
Old Greenlander
"I show the world the way I see it"
35 years of photography and still learning

 

[Iliah Borg]

I used your tiny 0.07 Mb picture

This one?

Loading…

1.static.img-dpreview.com

If so, it is the same raw converted in the camera in 2 different ways, with different conversion settings.

--

About LibRaw | LibRaw

www.libraw.org