What's your workflow?

[Rebecca West]

I thought it would be interesting if we each describe our workflow, from camera to printer. You can be as high-level or as detailed as you like. I'm talking typical workflow here.

Here's mine:

1. Take picture, usually in aperture priority mode and usually on auto white balance. Pay special attention to lighting and composition.
2. Download pictures from camera using a photodownloader like this one. That's not the one I have, but I can't find it on the net at the moment. That downloader puts files into a directory that it creates (named after the day: month-day-year) under a folder called 2006.
3. Use an image browser (Acdsee) to look at the photos, deleting as I go.
4. Call up good ones in photoshop.
5. Edit using levels, curves, and some actions I've accumulated.
6. Save as psp file in a "good" directory under the appropriate date.
7. Use Qimage to print from "good" directory. Printer is profiled, and profile is loaded in Qimage.

Voila!

[Max@Home]

1 take picture
2 connect camera to printer
3 press direct print button

All Hail the 30d and 5D

They save a LOT of typing!



(expect a more serious answer tomorrow morning, nighty-night !)

My kindest regards!

Max@Home

[cyclohexane]

I'm going to use "D2H" below, but you can assume this for just about any digital camera I'm shooting:

1. Whip out D2H, and turn it on, checking the bottom LCD for ISO and WB settings.* Correct ISO if necessary; set custom WB if needed.* The bottom LCD also contains QUAL, but I don't change this often (JPEG for high volume sports/instant, RAW+JPEG for most other uses; occasional RAW only). *Check autofocus settings.
2. Take a test shot: often, I'll meter off my hand/arm, choosing aperture first and shutter speed second.* Once dialed-in, I halve the shutter speed to add a stop of exposure, and this is usually close, if not correct.
3. Shoot away.* Chimp ocassionally to check histogram and composition.
4. Yank the card(s) out of the D2H. *Copy the images from the card using a card reader into a folder.* Folders are named mm_dd_yyyy as subfolders under each month (mm_yyyy).* Often, I attach a quick reminder after the date ("Robin Hood Rehearsal", "Basketball") to facilitate searching.
5. Review the images using Window Explorer and Photoshop File Browser; delete bad images. *Convert RAW files in Photoshop.
6. Edits: Curves > LAB Color > Gaussian Blur (raidus 5 to 8, usually) in a, b, channels if needed > USM lightness channel (58%, radius 2.8, threshold 4 is a good start) > Save as TIFF in a subfolder for easy access.
7. Printing is either sent off or done at home.

Of course, this tends to get modified quite frequently. *I've been experimenting with Nikon Capture NX, which gives much better color and has a neat U-point interface (I suck at masks). *For black-and-whites, I'll reconvert to RGB (working mode is usually Adobe 1998 to get close to the camera's Nikon Adobe RGB output) and run channel mixer.

[jfrancho]

The actual photo taking process varies so much depending on situation, so I'll skip that and start with a card full of images.
1. Put the card in the card reader and copy all files to folder on desktop.
2. Copy this directory to a remote drive.
3. Set the card aside with a small post-it it with the shoot details.
4. Move all the image files to a single folder.
5. Open Bridge and browse to that folder.
6. Batch rename all files in Bridge.
7. Start coding junk files for deletion.
8. Hide flagged files.
9. Second Pass for junk.
10. Put all the junk files in a folder called 'almost junk'.
11. Go through the remaining files and give good, better best ratings.
12. Anything unlabeled gets thrown in the 'almost junk' folder.
13. Combine the good and better and send any that don't make the cut to the almost junk folder.
14. Now I have my initial keepers that are worth developing to find out if they really are keepers.
15. Open in ACR and make general adjustments to WB, Exposure, etc.
16. I often apply settings to many images at once, this is just a quick and dirty raw setup.
17. Open all in ACR and convert to 8 bpc sRGB .jpg in the same dir as the raws.
18. Repeat steps 7-14, except I just flag the good raw file, and delete all the .jpgs when I'm done.
19. Adjust each image in ACR.
20. Let the images simmer for a day, since I'll need "fresh eyes" to finalize ACR settings.
21. Get all the ACR settings finalized, and save as 16 bpc .psd in some large gamut colorspace of my choice, and another folder called dev.
The rest of the steps are actions that point to scripts that are usually run as a batch, unless there is an issue with the edit. I have a a bunch of actions prebuilt that I snap together into one big action for batching groups of similar images. Sometimes, I have to backtrack and change the intensity of an effect, but generally the process is the same and looks like this:
22. Create surface masks, make selection, create layer mask, run noise reduction.
23. Inspect.
24. This is where I clone out goobers, do some controlled burn and dodge using a 50% grey overlay layer, or any other weird editing like this.
25. Inspect.
26. 1st pass of sharpening - I like to use modified TLR Capture scripts for this.
28. Inspect. Sometimes this where the really hard to spot goobers appear, so they get zapped here.
29. If there are any other issues with the picture, this where that gets handled. It's also when I'll convert to L*a*b color mode, if I want to add a certain "look" to the picture, or just to add pop with a Curves adjustment layer. Any other "creative editing" happens at this point.
30. 2nd pass of sharpening - TLR Creative sharpening. Interesting tidbit: in the quest for sharpness, a little bit of gaussian blur in the right places can go a long way. Think "unsharp mask" the hard way.
31. Final inspection.
32. Start the archive process.
33. Basically, a flattened .psd file gets put in about 3 redundant locations - some off-site.
34. Now I can go ahead and reformat the card. I'll go through the 'almost junk' ocassionally, and purge it.
Now I have nice finished, 16 bpc images ready for web or print. Well, not really. Here is the "weberize" process:
35. Using Image>Resize, to 800x533, or some other arbitrary dimension. Use Bicubic Sharper.
36. Convert to sRGB colorspace.
37. Convert to 8 bpc mode.
38. Save as .jpg quality 12.
39. Upload to hosting site.
40. Start a new topic here and keep hitting F5 when on the "Show new replies to your posts" page.

I think I got it all. If I missed something big, let me know and I'll fix it. I do realize it is a somewhat radical workflow, but it is cantered around quality at every step. Too bad I can't get more quality at the "press the shutter" step .

[Rokcet Scientist]

Whatever happened to good old chemical film photography? You took pictures, had them developed and framed, and voilá: your slides!
Life was so simple then.

[jfrancho]

I like it better this way. I am a control freak.

[aam1234]

Convert to jpeg in DPP... that's it!