Workflow/Post processing

[bricynt]

Hello all. This is my second post on the site. I just purchased my 1st DSLR (Nikon D3000) three weeks ago. Previously, I used a Sony P&S. For editing software, I use Photoshop Elements (v6).

Here's where I need help. With the DSLR, I now have the choice to shoot RAW images in addition to JPEG (or both simultaneously). From what I've read, RAW gives you more control over edits than JPEG. I have experimented with Camera Raw (ACR 5.5) in Photoshop, but I'm not sure if this is the most efficient way to process RAW photos, or if I even need to shoot in RAW.

I would like to know how other shoot, then process. Any suggestions or tips would be greatly appreciated. Keep in mind I am an amateur with two small kids - i.e. limited budget . Thanks.

[Colin Southern]

JPEGs are structured around small image sizes - and one of the ways they achieve that is by throwing away parts of the image that the eye can't see (eg details buried in the shadows) and then compress the rest (a bit over-simplified, but you get the idea).

What this means in practice is if you need to make any large changes to jpeg files - expecially to levels - then you basically come unstuck pretty quickly. To put that another way, JPEGs are great as a final form for an image after all of the editing has been done.

The other part of the equation is that an in-camera generated JPEG also equates to a file that has been processed according to the camera instructions (with a little help from you in the form of selected picture styles). Cameras CAN take a good shot this way, but they can also muck it up badly as well.

So - with a RAW file, YOU get to choose how you want to process the file (in terms of things like white balance - exposure adjustments - black clipping point - fill light - contrast - brightness - saturation - vibrance - clarity etc) - PLUS - there is a LOT more information retained in the file to recover data from a badly taken shot (or one that pushes the limits of the camera).

RAW gives you flexability, control, and options. JPEG gives you convenience. Personally I shoot RAW exclusively - I'd be quite happy to have the JPEG option removed from my camera all together. In terms of image quality, I don't think I've ever seen a JPEG shot that I couldn't improve upon (especially if I start with a RAW capture).

Does this help?

[Bill44]

Totally agree with what Colin has said.

[bricynt]

Thanks Colin for the detailed explanation of JPEGs and RAWs. Can you give me an example of how you process a typical set of shots? Do you edit the RAW files in your main program, or do you go through a conversion type program first?

Thanks.

[Colin Southern]

Originally Posted by bricynt

Thanks Colin for the detailed explanation of JPEGs and RAWs. Can you give me an example of how you process a typical set of shots? Do you edit the RAW files in your main program, or do you go through a conversion type program first?

Thanks.

No worries

Pretty straight-forward ...

1. Card out of camera and into card-reader

2. Adobe DNG converter takes the Canon CR2 files from the card - renames them - converts them to DNG format - and stores them on the PC

3. I point Adobe Bridge at the converted files where I make my selects

4. I then open "the chosen ones" in Adobe Camera RAW, and adjust as much as I can in there (where there isn't as much image degradation), before (finally) ...

5. Opening in Adobe Photoshop CS4 for the final workover.

Does that help?

[Dave Humphries]

On the other hand, (he says, subverting 'the masters' cunning plan, he, he, he ), not everyone (read many) people work like that.

I shoot with a Nikon D5000 in RAW almost exclusively, only RAW+jpg if someone wants duplicates of everything.

Here's what I do:
1) Memory card into PC
2) Import to HDD using Nikon Transfer > Nikon ViewNX (it does all the renaming and duplicating stuff too )
3) Rapidly work through using the 'click 100% feature' and delete the soft ones and other 'no hopers', give the rest a star rating on their chances of becoming 'chosen ones'
3a) Filter on 2 - 4 stars, or 3 - 4 stars (in ViewNX)
4) Pick one, right click and Open with "Adobe Elements 6 (Editor)"
4a) This seamlessly opens ACR first (because they are all Nikon RAWs = .nef)
4b) Hit the "Auto" button, (I have a preset of Detail and Noise Reduction set that does for most of my stuff), tweak if the colour masks show clipping of black or white, or if the presets aren't working on this shot
5) Click "Open Image" and do the final workover in Elements

Hope that helps (and doesn't get me fired )

[Colin Southern]

Originally Posted by Dave Humphries

On the other hand

To be honest Dave, it looks like pretty much the same workflow to me, except that you're doing with Nikon View NX what I do with Bridge.

By the way, with Bridge, it pre-converts them all at whatever you've set ACR defaults to and then feeds you the updated thumbnails

[bricynt]

OK, so it looks like you guys are using 1) Bridge or 2)View NX to rename/tag etc., and throw out the losers, right? This is where I'm using the Organizer in PSE6. Then you go into Camera Raw for the first edits (white balance, etc), then you finish it off in Elements or CS4. I think I follow pretty good now. Thanks to both of you


As for final ouput - I assume JPEGs mainly? How about the original RAWs? Do you guys keep them for future usage? How about storage in general for your photos - on the main HD, or a separate HD for photos only?

[Colin Southern]

Originally Posted by bricynt

Then you go into Camera Raw for the first edits (white balance, etc), then you finish it off in Elements or CS4.

Kinda. ACR actually kicks into action as soom as you open Bridge (replacing embedded thumbnails with RAW-converted previews), so you get updated info in Bridge to make better decisions with (which can make quite a difference).

[willgoss]

Just for fun....here is my Gimp work order....also use the same basic order for other Graphic Processors...

GIMP (2.6x) (photo fix steps)...starting with jpeg or tiff image.....
The following work order is typical to work on a Photo

*** Open the Image
*** Crop to Size - Use Rectangular Select (set fixed Aspect ratio), then under Image-Crop to Selection

*** Do Fix-ups (Smudge, Spray, Clones, Red Eye, Blurs, Clouds, etc.)

Alt M Median (Despeckle 2, 1, 256) for high noise images….
-Plus are optional
Alt 0 for High Iso (100, 7) then Alt 1
Alt 1 Remove Noise - GREYCStoration
-plus Ctrl 1 Unsharp Mask (1, 1, 1)
Alt 2 Zero - Clarity Plug at 128-3-50
--or-- Ctrl 1 Unsharp Mask (20-0.5-0)
-plus Ctrl H for details (10-20)
-plus Ctrl 2 Fake HDR

Impact fixing approach

Alt 3 Levels-Auto -or- Manual WB
-plus Ctrl 3 for Inc. Dynamic Ratio
-plus Ctrl B for Range Adjustment
-plus Ctrl R for Unbalanced Lighting
-plus Ctrl 1 (50-0.5-0) Reduce Haze
Alt 4 Auto Stretch Contrast – Option
-plus Ctrl 4 Tone Mapping
Alt 5 Auto Contrast (+Auto Levels)
-plus Ctrl 5 Mask (inc. shadows +40)
-plus Ctrl L Roy’s Local Contrast
Alt 6 Brightness Contrast (final)
Alt 7 Saturation Enhance (last fix)
-plus Ctrl 7 Auto Color Enhance
Alt 8 Sharp Redux Sharpen (default) Blur is included – Good Finish Step… Set Sharpened Layer Opacity (flatten)
-plus Ctrl 8 D’CIM - Unsharp Mask
(Default Amt 1-4)
Alt 9 Option Selective Gaussian Blur option for soft blur finish (portraits)

Shift I to flatten image before Ctrl P
Ctrl P Filters Unlimited Plugin for Borders, B&W Conversion, effects etc.


My Gimp Shortcuts (set in Preference)
* Undo ---------------------------- Ctrl Z
* Redo ---------------------------- Ctrl-Shift Z
* Fit to Window ---------------- Ctrl F
* G’MIC Plugin (Effects) ----- Ctrl G
* Quick Mask ------------------- Ctrl Q
* Color Balance ---------------- Ctrl B
* Levels --------------------------- Ctrl L
* Retinex ------------------------- Ctrl R
* Channel Mixer --------------- Ctrl M
* Unsharp Mask ---------------- Ctrl U
* High pass Sharpen………… Ctrl H
* Select All ---------------------- Ctrl A
* Select None (Deselect) ---- Ctrl D
* Copy Selection -------------- Ctrl C
* D’CIM --------------------------- Ctrl G
* Paste into Selection ------- Ctrl V
* MV’s - Wire Worm----------- Ctrl W
* iWarp --------------------------- Ctrl I
* Zoom Larger ----------------- = (Equal +)
* Zoom Smaller --------------- - (Minus)
* Mask Layer On/Off -------- Ctrl Click
* Mask View On/Off---------- Alt Click
* Median filter------------------ Alt M
* Iso noise Fix------------------ Alt 0
* Photo Fix Process ---------- Alt 1 to 8
* Unsharp Mask ---------------- Ctrl 1
* Fake HDR----------------------- Ctrl 2
* Increased Dynamic Range Ctrl 3
* Tone Mapping ---------------- Ctrl 4
* Contrast Mask (HMS) ------ Ctrl 5
* Roy’s Max Local Contrast--Ctrl L
* Auto Color Enhance--------- Ctrl 7
* Unsharp Mask D’CIM------- Ctrl 8
* Flatten Image ---------------- Shift I
* Filters Unlimited ------------ Ctrl P

[KentDub]

Here's my flow, using Adobe Bridge, ACR, and Photoshop.

1. Plug camera into computer, turn on
2. Import pictures using Adobe Bridge (Following steps are inside Adobe Bridge)
3. Group photos (Ctrl-G) - makes it easier to look at a large amount of shots, especially when you take 3+ photos of the same subject in the same position
4. Apply metadata/keywords to photos (Type of subject(s), type of picture, location, etc.)
5. Rate photos (1=garbage -> 5=I'll make $ on this). I don't delete any photos so a lot get 1 star.
6. Any photos 2 or up I do at minimum a quick once-over in Adobe Camera Raw (ACR), the higher the star the more time I usually spend
7. I then look for photos that I see opportunity to enhance using Photoshop and open them up
7.a. Apply capture sharpening
7.b. Levels/Curves and color correction
7.c. Retouching
7.d Subjective/creative sharpening
7.e Output sharpening
7.f Save as .psd
8. I take the psds and downsample to 8-bit, convert to sRGB, resize to web resolutions, and export a copy as a jpg
9. Upload the jpg to CiC and get feedback, then set the star rating to 1 (lol?)
10. Take more (hopefully better!) photos

[bricynt]

Thanks all for the feedback.

KentDub - based on your step 8, does this mean you do your RAW work at 16 bits?

Willgoss - I assume the 'gimp' is another processing program...also, can I assume it's best to crop early before making any white balance adjustments?

[willgoss]

bricynt.....

Yes Gimp is a freeware program somewhat like Photo Shop....

In my listing you will see, I first open the image and crop to size before any post processing.....

I then have 8 steps, in my case Alt 1 to Alt 8.....with a few extras and alternates thrown in.....I only use the steps I feel are required......I have identical steps in Paint Shop Pro with similar Short Cuts....

[KentDub]

Originally Posted by bricynt

KentDub - based on your step 8, does this mean you do your RAW work at 16 bits?

Yes bricynt, when I open the photos from ACR into photoshop I bring them in 16-bit (my camera captures them at 14-bit), and in AdobeRGB. This gives me the most flexability if I choose to go drastic with any effects/corrections. I even bought a Canon printer that accepts natively 16-bit AdobeRGB images (it makes fantastic prints).

There are arguments for and against 16-bit workflow. The way I generalize it, stealing a quote from a car commercial: "It's not more than you need, it's more than you're used to." The only, in my opinion, valid argument against 16-bit is that a significant amount of plugins/filters are not 16-bit compatible (Though, to be honest, I don't use them anyways!). I'm a firm believer in working at the maximum quality that is possible until you convert to your output format and size - but I always save the high-quality version seperatly.

Congrats on your new camera

[Colin Southern]

Originally Posted by KentDub

There are arguments for and against 16-bit workflow.

I like to think of it this way: "It's a lot easier to go from 16 bits of information to 8 bits than the other way around"

[Dave Humphries]

Originally Posted by bricynt

also, can I assume it's best to crop early before making any white balance adjustments?

Yes, but!

and the but is that if you need to Noise Reduce with a plug in like Neat Image. It will need to sample from an area with (ideally) no picture information in (just noise) which is exactly what you are likely to crop off!

So always have Noise Reduction before the crop in your workflow.

btw Neat Image works in 16 bit if you get, or upgrade to, the Pro version and can be used stand alone, or as a plug in from within PS Elements, Photoshop and half a dozen other popular image editing progs. Sounds like an ad, but my only connection with them is as a very satisfied (paying) customer.

Regards,

[Dave Humphries]

Originally Posted by Colin Southern

To be honest Dave, it looks like pretty much the same workflow to me, except that you're doing with Nikon View NX what I do with Bridge.

By the way, with Bridge, it pre-converts them all at whatever you've set ACR defaults to and then feeds you the updated thumbnails

I agree, and I'll probably discover the last bit when I give LR a trial soonish.
The hope of an issue free browser and the DNG benefit (no sidecar, no database) beckon.

Cheers,

[bricynt]

Originally Posted by KentDub

Here's my flow, using Adobe Bridge, ACR, and Photoshop.

1. Plug camera into computer, turn on
2. Import pictures using Adobe Bridge (Following steps are inside Adobe Bridge)
3. Group photos (Ctrl-G) - makes it easier to look at a large amount of shots, especially when you take 3+ photos of the same subject in the same position
4. Apply metadata/keywords to photos (Type of subject(s), type of picture, location, etc.)
5. Rate photos (1=garbage -> 5=I'll make $ on this). I don't delete any photos so a lot get 1 star.
6. Any photos 2 or up I do at minimum a quick once-over in Adobe Camera Raw (ACR), the higher the star the more time I usually spend
7. I then look for photos that I see opportunity to enhance using Photoshop and open them up
7.a. Apply capture sharpening
7.b. Levels/Curves and color correction
7.c. Retouching
7.d Subjective/creative sharpening
7.e Output sharpening
7.f Save as .psd
8. I take the psds and downsample to 8-bit, convert to sRGB, resize to web resolutions, and export a copy as a jpg
9. Upload the jpg to CiC and get feedback, then set the star rating to 1 (lol?)
10. Take more (hopefully better!) photos

KentDub - can you explain the different sharpening steps, specifically 7a and 7e? Also, when you finish, do you have three files types of each shot - RAW, .psd, and .jpg?

[KentDub]

Originally Posted by bricynt

KentDub - can you explain the different sharpening steps, specifically 7a and 7e?

Hi bricynt, sure no problem.

I have defaulted my ACR to apply no capture sharpening. There is a slider on one of the tabs that shows various sharpening sliders - turn the amount to 0 on all of your pictures (or default your settings like that). We do this because we like to have more control over the initial sharpening of the images than ACR provides us.

I'll give you an overview. We sharpen images in three stages. The first state, capture sharpening, is to offset filters (anti-aliasing) done by the camera. There is also a slight blur that is created by the nature of how we capture images. Capture sharpening corrects all of that and sets us up for our retouching. Colin can go into much greater detail than I on why we need to do capture sharpening if you're interested.

During the retouching phase, we apply subjective/creative sharpening - to taste. A portrait shot may have just the lips and eyes sharpened, for example. We sharpen using masks (some use the history brush) to sharpen only select areas of the photo to give them that pop -- to help the viewers eye. This is part of retouching and can be done anytime (or even multiple times) during the retouching phase, so 7.c and 7.d are really one in the same.

Output sharpening is interesting. You will want to do this as a non-destructive edit because you sharpen differently depending on the output (web/print/etc). If you are generating a file to put on the web or to send in an e-mail, you will want (save the origional first!) downsize the image so it fits on a screen without having to zoom out. Images will lose sharpness when this happens. You can correct this with output sharpening.

Sharpening for print is an art in its own. Inkjet printers have nozzels that naturally blur images as they spray ink onto the paper. To counteract this, we 'oversharpen' the images so when they print, they are correct. Some people apply this sharpening to taste (usually a bit more than they can stand on screen is where they stop). Some people, like me, use a plugin that you tell it what kind and size of paper, the DPI settings of your printer, and the DPI of the image and it calculates how much sharpening it will do (of course, you can always adjust from that). Using a plugin gives you very consistant results between prints, and to start with at least, takes a lot of guess work out of the equasion.

Originally Posted by bricynt

Also, when you finish, do you have three files types of each shot - RAW, .psd, and .jpg?

Yes. Using Adobe Bridge you can set a filter on the file type, so you don't see three of every shot. It eats up a fairly large amount of disk space, but its cheap and easy to extend that as needed. My RAW files are about 28mb, the typical photoshop file is about 200mb, and the jpegs (for web) are usually less than 1/2 or 1/4 mb (512 or 256kb). Each filetype is well suited for its purpose. The most important file is always the origional, so make sure it gets backed up.

Hope this helps, let me know if you have any other questions

[Colin Southern]

If it helps, we discussed sharpening some time ago here.