"Camera RAW-to-Web" What gives?

[mistybreeze]

Does anyone know the optimum workflow "From Camera RAW to Web JPEG?"

I pose this question for all photographers seeking to optimize their RAW images in 16 bit-CS3 for specific use on web-based personal portfolio sites. It seems there's plenty of camera-to-print workflow information available but almost no camera-to-web info (for professional photographers). Plus, every time I turn around it seems there's one more photographer doing something different (without explaining why).

In this thread, MarkDS was kind enough to provide his 16 bit, Pro Photo workflow from PS Image Size to web. Here's what he wrote:

Image Size with resamplingto 96 ppi and 600~800 pixels on the large dimension; interpolation bi-cubic sharper;
PK Output Sharpen for the web; (set opacity to taste);
Convert to sRGB with RelCol Intent and BPC on;
Convert Mode to 8 bit;
Save for Web in JPEG format, 60% quality, Optimized, with embedded sRGB profile.

I guess if one can read shorthand, all this sounds reasonable. But there's a lot of missing information. Such as:

1. 96 ppi? Since when? Has the web standard of 72 ppi changed? Or, with larger monitor screen sizes, does 96 ppi provide better quality for larger sized (800+ pixels wide) images?

2. What if PK Output Sharpen is not an option? At what point in the workflow should one use Photoshop's final sharpening options?

3. What if you forget to Edit > Convert to Profile before converting to 8 bit? Is it OK to then convert the profile in the new File > Save for Web & Devices?

4. Isn't the Quality percentage an image-to-image consideration? Is 60% a good starting point? What do you look for?

5. If you converted your image to sRGB before the Save for Web & Devices, isn't the sRBG profile already embedded? Or must someone click on another confirmation?

6. What about Flatten Image? Where does this fall into the workflow and what if you need to tweak with Curves after the image has already been flattened? Can you continually Flatten with no harm to the file? Do you Flatten before or after converting to 8 bit?

7. Martin Evening used to recommend down-sampling in small-ish increments to have more control over the data being discarded. Does this recommendation apply to CS3? Given the huge size of a RAW file, will the human eye see any difference if you chop out a big chunk of data in a single decrease?

I don't mean to criticize MarkDS's offering from the other thread. It merely gave me food for thought on the difficulty of finding trusted information on this very important workflow. Many professional photographers are building their own websites or are required to supply designers with web-ready images. Maybe it's time for Reichmann/Schewe and Company to address this need in a video tutorial and cover the various RAW-to-Web options.

At least Lynda.com offers Photoshop CS3 for Photographers but Chris Orwig begins his Resizing for the Web tutorial with an 8 bit image. Where does one go for comprehensive information you can trust?

[kikashi]

I don't pretend to the expertise of others on this forum, but FWIW,

1. 96 ppi? Since when? Has the web standard of 72 ppi changed? Or, with larger monitor screen sizes, does 96 ppi provide better quality for larger sized (800+ pixels wide) images?

I don't think there is any web browser which will respect a ppi measurement. It will put the image up on the screen pixel-for-pixel. If your image is 600 pixels wide on your site, it will appear 600 pixels wide on the viewer's screen, whatever ppi you set.

2. What if PK Output Sharpen is not an option? At what point in the workflow should one use Photoshop's final sharpening options?

At the end, according to conventional wisdom. Capture sharpening; process, including creative sharpening; output sharpening tuned to the destination.

6. What about Flatten Image? Where does this fall into the workflow and what if you need to tweak with Curves after the image has already been flattened? Can you continually Flatten with no harm to the file? Do you Flatten before or after converting to 8 bit?

Done as you save: jpegs don't have layers.

Jeremy
[a href=\"index.php?act=findpost&pid=154856\"][{POST_SNAPBACK}][/a]

[mistybreeze]

I did find this:
http://livedocs.adobe.com/en_US/Photoshop/...2B3C566F84.html

It's helpful if you enjoy reading from an Adobe User Guide and, of course, it doesn't address common workflow questions.

Not to change the subject but some search features are awful, and Luminous Landscape is among the worst. I wish Michael could find a better resource. Every time I put some words in the search window here, I get "sorry" 90% of the time. You have to commit to playing with your word choices and then, if you get results, they're often slim and you can't be sure your words did the best search. I quit after five attempts, which is three too many.

Adobe is only slightly better. The url above didn't appear until I typed "Save for Web." I have no idea why I didn't think to use those words yesterday when I made five attempts.

[digitaldog]

1. 96 ppi? Since when? Has the web standard of 72 ppi changed? Or, with larger monitor screen sizes, does 96 ppi provide better quality for larger sized (800+ pixels wide) images?

Its not a web standard. Displays have differing output resolutions.

2. What if PK Output Sharpen is not an option? At what point in the workflow should one use Photoshop's final sharpening options?

Yes. Does it look good on your display at 100% zoom? It may (it likely will not) look identical on anyone else's display. Its the web! Its not color managed and you have people looking at images with 12 year old CRTs and new, $3000K LCDs. But ever walk into a TV store? And those pups are all new!

3. What if you forget to Edit > Convert to Profile before converting to 8 bit? Is it OK to then convert the profile in the new File > Save for Web & Devices?

Yes.

4. Isn't the Quality percentage an image-to-image consideration? Is 60% a good starting point? What do you look for?

You look at the image on your display and understand the limitations discussed in #2. How does it look to you? How big do you want the uploads (and do you worry about download and bandwidth from others who, god forbid may be on dial up?).

5. If you converted your image to sRGB before the Save for Web & Devices, isn't the sRBG profile already embedded? Or must someone click on another confirmation?

Depends on what you select in SFW and what version of Photoshop. CS3 gives up both options.

6. What about Flatten Image? Where does this fall into the workflow and what if you need to tweak with Curves after the image has already been flattened? Can you continually Flatten with no harm to the file? Do you Flatten before or after converting to 8 bit?

You flatten. When doesn't really matter. You can flatten and convert at the same time in Convert to Profile.

7. Martin Evening used to recommend down-sampling in small-ish increments to have more control over the data being discarded. Does this recommendation apply to CS3? Given the huge size of a RAW file, will the human eye see any difference if you chop out a big chunk of data in a single decrease?

For the web? Only if I have a lot of time on my hands (and at the very least, an action).

At least Lynda.com offers Photoshop CS3 for Photographers but Chris Orwig begins his Resizing for the Web tutorial with an 8 bit image. Where does one go for comprehensive information you can trust?

Come on, its just the web kids. Its not a huge fine art ink jet print you control and can view as everyone else can. Lets not make this more complicated than it needs to be. When you consider the vast majority of web browsers and users using them are not color managed, that what anyone sees is likely to be vastly different from what you saw AND you can't control this, what's the use of banging your head? Its like getting up real close to a billboard and complaining the dots are real big, not small like your ink jet.

KISS. Get Lightroom, let it build web galleries and handle all the sizing, color space conversions and uploading. Spend time polishing pixels you have control over.

[mistybreeze]

Its not a web standard. Displays have differing output resolutions.

Fine. So then why use 96? Is 96 ppi the new universal output resolution or does someone (MarkDS) make this decision based on what 96 ppi looks like on his screen?

Does it look good on your display at 100% zoom? It may (it likely will not) look identical on anyone else's display.

"Identical" was never considered a realistic goal. Some of us know at least that much.

Its the web! Its not color managed and you have people looking at images with 12 year old CRTs

I just hate to sound snobbish  but maybe New York photographers, artists, models and their agents view web portfolios differently than those in Santa Fe or other parts of the country. Maybe earning three million+ a year forces an art professional to view all aspects of presentation with an obsessive desire to be as optimal as possible. I don't know anyone in advertising, modeling, or photography who is working on a 12 year old CRT. I get your exaggerated point, though.

How big do you want the uploads (and do you worry about download and bandwidth from others who, god forbid may be on dial up?).

Obviously, no one wants to wait for anything, including me. Slow-going Flash sites receive the most industry criticism. So knowing how far is too far seems crucial to perfectionist control freaks. I don't know any client or potential client who uses dial up.

Depends on what you select in SFW and what version of Photoshop. CS3 gives up both options.

Yes, I'm now using CS3 Extended. I put CS3 in the title of this thread. Too many variables make learning a challenge.

You flatten. When doesn't really matter.

But can you flatten multiple times on the same image? If you flatten and sharpen a JPEG and find that the JPEG needs a tweak Curve, adding the Curve turns the JPEG into a PSD. Flattening once again after the tweak Curve, converts the PSD back to JPEG. Are there no file degradation issues here?

RE: Martin Evening down-sampling
For the web? Only if I have a lot of time on my hands (and at the very least, an action).

Cosmetic ads require perfection. Fine art pieces, too. At the prices these command, no one is willing to compromise on quality, even for web presentation. If you down-sample incrementally on the final TIFF, why wouldn't a perfectionist down-sample incrementally on a web-portfolio presentation? Is their a sound argument that says don't waste your time, even with an action?

Come on, its just the web kids.

First, my "kid" days are long over and, second, nobody at Thaddaeus Ropac thinks web presentation is a game. Perfectionists bang heads, it's their nature.

Its not a huge fine art ink jet print you control and can view as everyone else can.

I get that. But does that mean Fine Art does not have its own optimization considerations for web view?

Lets not make this more complicated than it needs to be.

I'm all for uncomplicated. But I need to compare results first and then smartly challenge the know-it-all tech nerd who says it ain't as good as it can get.

[mistybreeze]

And lastly,

Get Lightroom.

I've had Lightroom. That's my next upload. My fear is that I may not love it as much as CS3 Camera Raw. I don't end love relationships easily.  

[digitaldog]

I just hate to sound snobbish  but maybe New York photographers, artists, models and their agents view web portfolios differently than those in Santa Fe or other parts of the country.

Don't be so sure. Are you positive that 100% have calibrated displays and are using Safari? If not, you can't at all be sure that what YOU see and what THEY see match.

But can you flatten multiple times on the same image?

Sure. You can't flatten an image that doesn't have layers, its flat. You can build a layer for one task, keep it or flatten it, keep others non flattened. But one thing is for sure, before you upload, you're going to have a flat file. JPEG doesn't support layers!

If you flatten and sharpen a JPEG and find that the JPEG needs a tweak Curve, adding the Curve turns the JPEG into a PSD. Flattening once again after the tweak Curve, converts the PSD back to JPEG.

No, it flattens the file, it can be a PSD. You have to tell Photoshop what format you want. But if you want a JPEG, for the web, its flat.

Are there no file degradation issues here?

Every time you re-save a JEPG, there's more data loss. Do you see it?

Cosmetic ads require perfection. Fine art pieces, too. At the prices these command, no one is willing to compromise on quality, even for web presentation.

Sorry but you can't control this. You can control the quality of ink on paper, you can control the quality of a web image you view. You can't control the quality of what a zillion other people view from the web. Its just as simple as that.

If you down-sample incrementally on the final TIFF, why wouldn't a perfectionist down-sample incrementally on a web-portfolio presentation?

Because you'll never see the difference and even if you did, there's absolutely no guarantee anyone else would.

Is their a sound argument that says don't waste your time, even with an action?

Its your time, knock yourself out. And when it looks prefect on your display, just consider that it may look totally different on someone else. Does that make you feel better?

First, my "kid" days are long over and, second, nobody at Thaddaeus Ropac thinks web presentation is a game. Perfectionists bang heads, it's their nature.

You can't prefect something you have no control over. You can prefect something you can control.

I get that. But does that mean Fine Art does not have its own optimization considerations for web view?

Fine art printing is about producing a print YOU can view under a controlled environment (sometimes you can't control the environment). But the difference, unless you're viewing the print under some really odd illuminant isn't going to be anywhere as variable as the millions of users, working with all nature of displays, of differing ages and conditions, using non color managed browsers. Maybe some day this will change (one would hope). The facts remain, every user is working with vastly different displays and browsers and operating systems and bandwidth. Compare this to controlling how a print, or even a magazine reproduction appears is pretty simple.

[Mark D Segal]

Does anyone know the optimum workflow "From Camera RAW to Web JPEG?"

I pose this question for all photographers seeking to optimize their RAW images in 16 bit-CS3 for specific use on web-based personal portfolio sites. It seems there's plenty of camera-to-print workflow information available but almost no camera-to-web info (for professional photographers). Plus, every time I turn around it seems there's one more photographer doing something different (without explaining why).

In this thread, MarkDS was kind enough to provide his 16 bit, Pro Photo workflow from PS Image Size to web. Here's what he wrote:

Image Size with resamplingto 96 ppi and 600~800 pixels on the large dimension; interpolation bi-cubic sharper;
PK Output Sharpen for the web; (set opacity to taste);
Convert to sRGB with RelCol Intent and BPC on;
Convert Mode to 8 bit;
Save for Web in JPEG format, 60% quality, Optimized, with embedded sRGB profile.

[{POST_SNAPBACK}][/a]

I only just now noticed this thread and I see Andrew has addressed most of your issues so I won't repeat. I also mentioned there that after implementing the above-referenced workflow, I import the folder of images to Lightroom, create a Collection out of it, then create a web gallery of the collection in Lightroom. The workflow and the settings I developed for creating website photographs is the result of experimentation with variants in both the sequence of steps and the settings, and this is what I found produced the best image quality relative to a manageable file size that I could develop. It is also completely amenable to automation, and in fact works from just after the image is opened through its saving as a jpeg ready for import to Lightroom. All happens in a real jiffy. The only issue I have with it is that periodically it oversharpens. If that happens I re-do with a lower sharpener opacity.

You are asking many questions about the merits of this workflow or other ideas about how to do it - that's fine, but why not TRY IT, and see whether it works for you. As well, you can judge for yourself what my results come out like on [a href=\"http://www.markdsegal.com]My Website[/url] - click on the link to Galleries and let me know your impressions of the image quality. Of course this is not a substitute for trying this workflow yourself on your images, because you know what your originals look like and how you want them to appear on the web, but you can't have the same degree of knowledge about mine.

[mistybreeze]

You are asking many questions about the merits of this workflow or other ideas about how to do it - that's fine, but why not TRY IT, and see whether it works for you.

Of course I try every smart suggestion. I didn't pose these questions simply for my own benefit. I have plenty of photography students who work for me and they ask these kinds of questions all the time. I figure this is a good venue to inquire further and I think the subject is worthy, assuming there are plenty of others who will appreciate the dialogue.

I totally understand Andrew's stand on monitor screen variation. If you can't control everyone's viewing environment, it makes little sense to fuss and stress. But I know many photographers who do fuss and stress. They will attempt to view their web images on as many screens as they can and try to tweak an acceptable version based on the variations actually seen. It's not a perfect science, for sure. But it helps them sleep better at night and take their sites live.

All I'm suggesting is that a smart camera-to-web workflow remains a mystery to many photographers who wish to prepare their images for website portfolios. Save for Web & Devices is a feature of Photoshop often ignored. Because many view printing as the only workflow worth a photographer's time and effort, photographers have few resources for getting easy to comprehend workflow advice to help them take 16 bit images to a great quality JPEG for the web.

[Mark D Segal]

OK, thanks for clarifying the context.

[Gordon Buck]

I don't have Lightroom.  Realizing that my post processing often uses the same procedures and settings, I have made several Actions for taking RAW to a web sized image.  However, most of such Action processed web sized images are emailed instead of posted on the web.

Once you find the post processing that suits you, it is very simple and convenient to set up an Action to automate those steps.  Another procedure that I use is to make an Action that produces a "High Quality JPG".  By this I mean a jpg that, to me anyway, is preferable to those generated in-camera.  My High Quality JPG does not have final sharpening.  With other Actions, that High Quality JPG (or 16 bit psd) is finished as a web image, a snapshot, full screen image, etc.