Use of Raw and JPEG (Page 14): Some time ago {it was when we were at our house in Marion IN, and we moved out 20 years ago}, I sent some Kodachrome25 slides to a pro to scan; I...

rehess Loc: South Bend, Indiana, USA

rehess wrote:

I’ve often had lenses that could receive more detail than the film/sensor could record, but I don’t look for more detail. I could always use Pentax’s “pixel shift” to improve resolution recorded, but I don’t need to see the fleas on the back of the squirrels or groundhogs in our backyard.

Some time ago {it was when we were at our house in Marion IN, and we moved out 20 years ago}, I sent some Kodachrome25 slides to a pro to scan; I selected some with small writing and other small detail. The pro returned 3000X2000 scans. Then I set myself up in what we called the “Library” in that house with a computer and the slide projector I’d always used {and trusted}. Every bit of detail I could locate on a slide I also found on the corresponding scan …… so I said I would switch to digital photography when they were producing 6MP cameras that were within my price range, and I said “6MP was within the range of lenses I was willing to buy”. Sometime later, I purchased a 16MP Pentax K-30, and used it with the “Pentax-A 50mm F/1.7” lens I had used to take most of the photos I had taken for that test. I was surprised at how sharp that lens turned out to be - the SOOC JPEGs were much sharper than that test had shown, and I dropped the “I was willing to buy” part of my statement.

davyboy Loc: Anoka Mn.

Canisdirus wrote:

There are levels to photography...as with anything... (show quote)

All I’m saying is most people will never notice the difference Don’t kid yourself jpegs can be beautiful

Longshadow Loc: Audubon, PA, United States

rehess wrote:

I’ve often had lenses that could receive more detail than the film/sensor could record, but I don’t look for more detail. I could always use Pentax’s “pixel shift” to improve resolution recorded, but I don’t need to see the fleas on the back of the squirrels or groundhogs in our backyard.

Amen!

Longshadow Loc: Audubon, PA, United States

davyboy wrote:

All I’m saying is most people will never notice the difference Don’t kid yourself jpegs can be beautiful

(or care)

selmslie Loc: Fernandina Beach, FL, USA

Ysarex wrote:

I didn't say my comparisons were comparing sharpness and detail. Why would you say that they were when I didn't?

You said, "The difference in rendered detail is basically in the demosaicing of the CFA." But the rendering of the detail involves mixing the raw pixels to come up with an equivalent set of RGB pixels.

What is inherent in the demosaicing process is a loss of sharpness due to the CFA and that is easy to demonstrate by removing the CFA and skipping the demosaicing process.

Are you claiming that Adobe (and everyone else) can do a better job of demosaicing the information in the raw file than Fuji can? Fujifilm would be interested to learn that since they invested a lot of effort to developing the X-Trans demosaicing step for the camera.

Ysarex wrote:

In what way specifically does a variation in camera input profile alter the demosaicing of the raw file?

Exactly my point. It doesn't!

It alters all of the other aspects of the image such as white balance, clarity, saturation, local contrast, etc.

Fuji's profiles imply that they can make the result look the way their film would have rendered it. That might apply to color positive film (E6) but not to negative film (C41) since all of those qualities are affected by the quality control during the film development process and the and paper on which it is printed and how it was processed. Anyone familiar with Fuji film products would be skeptical of those profiles.

Ysarex Loc: St. Louis

selmslie wrote:

You said, "The difference in rendered detail is basically in the demosaicing of the CFA."

Right, so your original response to me was off base. Got that.

selmslie wrote:

But the rendering of the detail involves mixing the raw pixels to come up with an equivalent set of RGB pixels.

What is inherent in the demosaicing process is a loss of sharpness due to the CFA and that is easy to demonstrate by removing the CFA and skipping the demosaicing process.

Inherent in the demosaicing process is a loss of fine detail.

selmslie wrote:

Are you claiming that Adobe (and everyone else) can do a better job of demosaicing the information in the raw file than Fuji can?

Certainly not Adobe, and "better" is subjective, but yes, some 3rd party raw processors use different demosaicing algorithms that do a better job with the X-Trans CFA in rendering fine detail than the demosaicing embedded in the camera -- well known.

ysarex wrote:

In what way specifically does a variation in camera input profile alter the demosaicing of the raw file?

selmslie wrote:

Exactly my point. It doesn't!

My point as well -- glad you've come to agree with me.

selmslie Loc: Fernandina Beach, FL, USA

Ysarex wrote:

Inherent in the demosaicing process is a loss of fine detail.

AKA a loss of sharpness.

Ysarex wrote:

some 3rd party raw processors use different demosaicing algorithms that do a better job with the X-Trans CFA in rendering fine detail than the demosaicing embedded in the camera

Are you implying that Fuji can't match their rendering or that they don't want to because it would slow the camera down?

Ysarex wrote:

My point as well -- glad you've come to agree with me.

There is a problem with the evidence you posted since you left out information such as the JPEG profile and other settings in the camera.

Bill_de Loc: US

selmslie wrote:

Sorry Charlie, StarKist doesn’t want tuna with good taste, but tuna that tastes good!.

Now that I understand!

Thanks very much.

---

Ysarex Loc: St. Louis

ysarex wrote:

Inherent in the demosaicing process is a loss of fine detail.

selmslie wrote:

AKA a loss of sharpness.

I don't agree those are exactly the same thing.

ysarex wrote:

some 3rd party raw processors use different demosaicing algorithms that do a better job with the X-Trans CFA in rendering fine detail than the demosaicing embedded in the camera

selmslie wrote:

Are you implying that Fuji can't match their rendering or that they don't want to because it would slow the camera down?

Don't know Fuji's motivations and I doubt they'd tell us. I know the difference is real.

ysarex wrote:

My point as well -- glad you've come to agree with me.

selmslie wrote:

There is a problem with the evidence you posted since you left out information such as the JPEG profile and other settings in the camera.

All of which does not alter demosaicing in any way.

selmslie Loc: Fernandina Beach, FL, USA

Ysarex wrote:

I don't agree those are exactly the same thing.

Sharpness is a legitimate and measurable photographic term. Loss of detail is a bit vague. In your own words, can you define the difference?

Ysarex wrote:

Don't know Fuji's motivations and I doubt they'd tell us. I know the difference is real.

That's also a bit vague.

Ysarex wrote:

All of which does not alter demosaicing in any way.

In your own words, can you tell us about the difference between Fuji's in-camera demosaicing and the results from Adobe or Capture One?

All three of these seem to be your opinion - not much for us to go on.

Lucian Loc: From Wales, living in Ohio

I wonder how many members here are printing huge photos to hang on their walls, how many are printing on archival paper using the best archival inks? Probably 99% are not. How many are zooming in extremely close on every photo they display on a computer, as they scroll through their images, or give a slide show to friends and family and keep zooming in to show the image quality? Probably the same percentage of 99% don't.

Therefore, if you are not doing that stuff, you don't really need to shoot in RAW. Today's cameras are doing a rather good job of giving us a pleasing image in Jpeg format, when we take a properly exposed and focused photo. Is a RAW file better? Of course it is, but how often do you really need that image to be that much better? Probably almost never, but when you do, you have the RAW file for that. That is why we take RAW and Jpegs at the same time.

Shooting in RAW and Jpeg is the best way because, for the most part, your JPEG will suffice for what most of us are doing, with regards to display, be it a print or in a slide show on a computer, or TV screen. When the Jpeg shot is complete from the camera, you see a very good rendition of what the RAW file is going to look like, after working it.

That means you have the file in an instant. You do not need to open an editing programme and download the RAW file and then start playing about with sliders etc. to tweak that RAW file until you have it as you would like to see it. And I bet most of the time that Jpeg copy was very close and if you were to take a family or friends poll, they would all say that the Jpeg was about as good as your time consuming, RAW file tweaking image was.

All real Pro wedding photographers shoot in Raw and Jpeg and for the most part, use the Jpeg files for printing and making the wedding books. When shooting weddings, we just don't have the time to go through every image in RAW and adjust things to see a finished image. A Jpeg shows us what it will probably look like and if you have done your job well, you should have mostly good shots to show. For the ones that are off a bit or really count, then you can play with tweaking the RAW files. However, no wedding photographer ever has the time to go through all the RAW files to get to a finished item. That is why we shoot Jpeg.

Although not as much of an issue today as it used to be in the past, there is the question of storage space. Your RAW files will take up soooo much more space than your Jpeg files. Therefore, if you can manage with Jpegs, then just do that. I think the people telling us we must always shoot only in RAW and work on those files only, are probably good friends with those people that tell us any tweaking of a file after it comes out of the camera, are not True Photographer, because we should not be touching that image if we want to be true to the real meaning of a photograph.

Most of those never give thought to how much post processing went on in the dark rooms in the old days of negatives and printing. Basically, a Jpeg will be just as good for those who happen to see our images, as the RAW files you might have taken time to work on, so why bother with all that extra work? It won't really ever get appreciated.

If you say you just do it for your own satisfaction, then you don't really need to be part of this discussion anyway, since none of it matters to you. You just do what you need to do and don't need to tell us about it.

rehess Loc: South Bend, Indiana, USA

Ysarex wrote:

All of which does not alter demosaicing in any way.

selmslie wrote:

In your own words, can you tell us about the difference between Fuji's in-camera demosaicing and the results from Adobe or Capture One?

All three of these seem to be your opinion - not much for us to go on.

What are you two arguing about???

Doesn’t what what Fuji does differ from what other companies do?

How does Fuji’s attempt to match film profiles change their task?

selmslie Loc: Fernandina Beach, FL, USA

Lucian wrote:

I wonder how many members here are printing huge photos to hang on their walls, how many are printing on archival paper using the best archival inks? Probably 99% are not. How many are zooming in extremely close on every photo they display on a computer, as they scroll through their images, or give a slide show to friends and family? Probably the same percentage.

Therefore, if you are not doing that stuff, you don't really need to shoot in RAW. ...

The reason to develop the raw image on your computer has less to do with sharpness and detail than with properly rendering tonal distribution (shadows and highlights), white balance, color, clarity, tonality, local contrast, the correction of lens issued like distortion (barrel and pincushion), chromatic aberration as well as leveling, cropping and perspective (removing keystone) - not to mention the removal of sensor dust.

These issues seldom occur in the same image but they are all easier to cake care of on the computer than within the camera.

selmslie Loc: Fernandina Beach, FL, USA

rehess wrote:

Doesn’t what what Fuji does differ from what other companies do?

How does Fuji’s attempt to match film profiles change their task?

Most cameras provide a way to render the SOOC JPEG differently.

Fuji just does it with built-in profiles with film names.

Ysarex Loc: St. Louis

selmslie wrote:

In your own words, can you tell us about the difference between Fuji's in-camera demosaicing and the results from Adobe or Capture One?

All three of these seem to be your opinion - not much for us to go on.

Not just my opinion. Massive amount of available info to go on for anyone capable of a Google search. Do your own homework.

https://blog.thomasfitzgeraldphotography.com/blog/2022/1/processing-fujifilm-raw-files-in-2022-the-best-solutions
https://fstoppers.com/gear/which-program-handles-fujifilm-raw-files-better-lightroom-or-capture-one-594075
https://lenscraft.co.uk/photography-blog/lightroom-fuji-raw-files/
https://fstoplounge.com/fuji-x-tran-raw-processing-done-right/
https://medium.com/@nevermindhim/x-trans-vs-bayer-fantastic-claims-and-how-to-test-them-475b4f1b7fae
https://www.dxo.com/tech-news/unique-sensor/
https://forum.luminous-landscape.com/index.php?topic=127994.0
https://www.sandrofabris-photography.com/blog/fujifilm-raw-conversion/
http://ijourneys.co.uk/fujifilm-x-trans-sensor/
https://www.lindsaydobsonphotography.com/personal/watercolour-effect-and-worms-fuji-xt20-x-trans-iii-raw-processing-with-adobe-lightroom