DPChallenge: A Digital Photography Contest You are not logged in. (log in or register
 

DPChallenge Forums >> Tips, Tricks, and Q&A >> Raw workflow
Pages:  
Showing posts 1 - 14 of 14, (reverse)
AuthorThread
04/17/2007 10:14:32 AM · #1
Ok folks - its time to speak up (and be specific).
I shot most things in raw and tend to "play" with settings, rather then have any methodical approach to it. Most times the exposure "as shot" seems to be best - so sometimes I'm doing very little with adjustments. I've also noticed the colour space is "Adobe RGB 1998". Should I be changing this here ?
Also, what specifically is your workflow here - what do you adjust and what do you leave as defaults ? Do you have a methodical approach ?
Time to speak up.
04/17/2007 11:26:40 AM · #2
I used RSP before (still do for quick projects) and Lightroom now. In both cases, I have some "default settings" that I apply to every image on import in the system. That's because a raw file looks rather flat when it is first read in and interpreted by the raw converter. So my "defaults" get me about 90% of the way there in about 90% of the cases.

From there, I'll tweak WB and exposure as necessary. This is usually done by identifying one image in a sequence and applying that same setting to all images of the sequence. Sometimes I have to make minor tweaks to the exposure from image to image in that sequence.

From there, it really depends... am I posting proofs? If so, I may not do anything else. Just export the jpegs and upload to smugmug.

But if I'm looking for a more finished image, then I'll probably spend a little more time examining the highlights and shadows and making sure everything (that I want visible) is falling into a range that can be edited in PhotoShop.

And if I'm doing a large print (11x14 and larger) I'll export 16-bit tiff instead of jpeg. And for the really large prints (16x20 and larger) I'll actually switch converters and go over to DxO to convert the image. That's because I feel the quality of the output of DxO is much higher. But DxO is also tons slower, so I only use it when the highest quality is necessary.


04/17/2007 11:51:14 AM · #3
I shoot raw+jpg. When I plug the compact-flash card into the reader, I copy the entire card onto my hard drive, and back that up on an external hard drive (two copies, which never get touched). Then, I import the photos (from the card) into Aperture v1.5.2, which only imports the raw files. From there, Aperture pretty much takes care of the rest. Where needed, I adjust the white balance. For minor edits, I usually adjust in aperture, export as a full resolution .jpg and make final adjustments in photoshop. For major edits, I go straight to photoshop in one of two ways. If it's a change that I will want ready-access to in my Aperture library, I use Aperture's "Edit in External Editor" option, and the final product will be accessible from within Aperture. The other option is to export a full resolution 16 bit tiff file from Aperture, and then edit within photoshop. My camera is set to sRGB color space, as is Aperture and Photoshop. For special cases I will use Photomatix. I am investigating DxO for the future, and my fingers are crossed for them to create an Aperture plugin.
04/17/2007 11:55:49 AM · #4
I use canon's DPP and up until about 2 weeks ago shot in the Portrait Picture style with the sharpness at 4, and the tone kicked one to the right (less red).
Shot a wedding and the second shooter used a 5D on standard and I like the skin tones better, so I switched - still have sharpness on 4 and tone one stop less red.

I can print after JPG conversion from there just fine 90-95% of the time with no further PS work. Proper WB is the key. For web I try to remember to up the sharpness in PP at some point for more 'pop'.

DPP 3.0 is out and it's more user friendly, and they've added NR. Some stuff I shot at ISO 400 last week I don't like the look of (on screen at 66% or more) after conversion...white 'noise' like oversharpening and it's worse at 100% but not there at 50%...not sure the issue at this point, but something in DPP 3 has changed from 2.2.

I used to shoot AWB and use a target for studio and more or less guess on other stuff (presets or click). I got an exposdisc last month, and as long as I remember to use it, I get much better WB and less work afterwards.

I don't have time to work every image - for weddings I shoot 1000 and show 500-600 (ad a second shooter and I have another 300-400 images to go thru). These go out as rough proofs, and then ones slected for the album or prints get a more thorough going over (100-150 images on average per wedding). For HS seniors, babies, etc (typical studio sessions) I may shoot 150-250 images and for projection sales I show 70-85 finished (in PS) images.
04/17/2007 12:07:53 PM · #5
I mostly am using Rawshooter essentials. The first step for me it setting the white balance, for photos that were shot under the same lighting will get the same settings to they look consistent. Then I will adjust the exposure and contrast if needed then the sharpness and if needed I then adjust the Noise suppression.

I then convert to a jpeg image using sRGB color space, if more edits are needed I will do them using Photoshop Elements. A lot of people don’t like to edit jpeg images worried that they will loose too much quality, but for the most part this really is not an issue. There are some images that I choose to edit instead as 16 bit/color tiffs, but I find the need for this to be very rare.

Scott
04/17/2007 12:50:44 PM · #6
My fledgling RAW workflow goes something like this:
-Shoot in just RAW (I was going RAW+, but I never use those jpegs anyway so why bother? waste of space...)
-Import to Lightroom (which I just started using)
-Add metadata, keywords
-Sort. Stupid photo? Deleted! (why keep 16mb files of a blown out sky?)
-Make adjustments. Its a slow process now, but as I add more and more presets and learn the program, I expect this to speed up significantly.
-Export hi-res DNG, prophoto colour space
-Export low-res (1000px, 80 qual) jpeg, sRGB
-If further adjustments are needed, I take the DNG to photoshop.
-Save photoshop edit as a prophoto PSD if incomplete, TIFF in colourspace of printer (+ sRGB jpeg for web).

Sooo...thats my new workflow. Ive only used it a couple times so far, but it seems to work pretty well. I believe its based loosely on the D-65 workflow, and was recommended to me by one of the top photogs in the snowboard world. I'm missing some of the file naming and triple backup-ing that his workflow contained, and maybe I'll adopt those once I get my own solid backup system in place... but for now, this is my standard procedure.
04/17/2007 02:50:33 PM · #7
I, too, am interested in a good RAW workflow process. Interesting the wide variety of applications used for RAW processing and the wide variety of purposes that affect the workflows.

This workflow philosophy assumes that it is better to do as much general contrast/color processing as possible in the RAW converter and that once you get into Photoshop all you need to do is selective adjustments to localized areas of the image. For a single image RAW process using ACR (Adobe Camera RAW) this is what I do:

1-Open RAW file in ACR.

2-Basically, in this step the intention is to make all the "standard" image level color and contrast adjustments that you'd normally make in Photoshop adjustment layers. Adjustments on the tabs are ordered well, so I do them in the order they occur in the ACR tabs:
A - On "Adjust" tab first set White balance either from popup or color and tint sliders.
B - On "Adjust" tab fix Exposure, Shadows, Brightness, Contrast and Saturation sliders. This fixes colors and light curves kinda like you would with Photoshop "Levels" and other PS adjustment layers.
C - Generally not needed but on "Detail" tab adjust sharpness, luminence smoothing and/or color noise reduction if necessary.
D - Generally not needed, but on "Lens" tab adjust Chromatic aberration and vignetting. I might use vignetting but only if cropping the image in RAW which is not that often.
E - If needed, On "Curves" tab adjust for standard "S" curve contrast.
F - This is an important adjustment. On "Calibrate" tab adjust red, green, blue saturation and hue.

3-Go back through all the ACR tabs and make fine adjustments as needed.

4-General conversion settings used are - Space:Prophoto RGB, Depth:16 bits/channel, Resolution: 240 dpi.
These setting are used because it preserves the most data from the RAW file that ACR allows (with the possible exception of DPI).

5-Open into Photoshop.
Now all I should have to do is make selective changes to localized areas.

6-Make selective fine tuning adjustments to localized areas of the image.

7-Perform the rest of the "regular" workflow processes to generate a master file and outputs from that master file.

A VERY positive side effect of doing as much work in ACR as possible is that it creates a TINY file that saves all your ACR adjustments much like adjustment layers do inside Photoshop but which in PS requires many 10s of megabytes of disk space of adjustment layers AND the results are better!

ACR actually saves disk space for standard global image adjustments!
04/17/2007 05:41:47 PM · #8
Thanks Steve - very informative.Just what I was after.
Do you adjust any sharpness, luminence & colour noise reduction ? I'm never sure whether to adjust these to zero (do all through PS later) or have a small setting in ACR and then finalise in PS .
04/17/2007 06:11:02 PM · #9
Originally posted by Tajhad:

Thanks Steve - very informative.Just what I was after.
Do you adjust any sharpness, luminence & colour noise reduction ? I'm never sure whether to adjust these to zero (do all through PS later) or have a small setting in ACR and then finalise in PS .

I rarely touch these adjustments.

I've set the "Sharpness" adjustment to "view" only in the options, prefering to do sharpening as the last step in photo editing. I have applied it before in the RAW converter, but only with soft images that need some pre-processing sharpening.
04/18/2007 08:12:15 PM · #10
Very nicely described, Steve. Thanks.

A couple of comments:

First, if there is a serious defect in your photo, it's usually best to deviate from the standard order and work on that first. For example, if the exposure is way off, adjust that before dealing with white balance. The settings all affect each other, and making large changes in one often require tweaks to the others. Of course, if everything is pretty clean, the order of the tabs works great.

For step 4, keep in mind both the final destination and the anticipated steps needed to get there. If you have a lot of Photoshop work to do and are going to print on a high gamut printer, ProPhoto RGB and 16 bits/channel make a lot of sense. If the photo needs minimal work and will be used on a website or sent to a printer that can only handle "standard" photos, the extra work of gamut and bit depth conversion won't buy much, and you may actually get better quality going directly to sRGB at 8 bits/channel.
04/18/2007 09:15:06 PM · #11
Originally posted by dr rick:

... A couple of comments:

First, if there is a serious defect in your photo, it's usually best to deviate from the standard order and work on that first. For example, if the exposure is way off, adjust that before dealing with white balance. The settings all affect each other, and making large changes in one often require tweaks to the others. Of course, if everything is pretty clean, the order of the tabs works great.

For step 4, keep in mind both the final destination and the anticipated steps needed to get there. If you have a lot of Photoshop work to do and are going to print on a high gamut printer, ProPhoto RGB and 16 bits/channel make a lot of sense. If the photo needs minimal work and will be used on a website or sent to a printer that can only handle "standard" photos, the extra work of gamut and bit depth conversion won't buy much, and you may actually get better quality going directly to sRGB at 8 bits/channel.

Agreed that an image with specific needs should normally have those needs addressed first and then move on to so-called 'normal' adjustments. Mine is just a workflow suggestion based on a 'normal' image and my own experience. Truth be known, I have no idea how things should really be done in RAW. I'm was kinda hoping you guys would tell me that. :)

Regarding step 4, I always post process an image with the thought of making large prints from it in the back of my mind. That is why I preserve the most data using the Prophoto RGB color space and 16 bits/channel. There is never a problem 'dumbing down" an image to 8-bit sRGB for the web, but there is ALWAYS a problem going the other way around. Worst case scenario in preserving as much data as possible is that you use more disk space and you have to learn and use additional software. Worth it to me but perhaps not worth it to others.

Message edited by author 2007-04-18 21:16:14.
04/18/2007 09:39:56 PM · #12
I shoot as:

RAW > uncheck any sharpening or colour boosting and convert to TIFF16 with canon's DPP > process in PS > save as TIFF16 if need be > convert to 8bit > save in JPEG
04/18/2007 10:16:38 PM · #13
I can tell by reading that some of you are wasting your megapixels. With digital raw always overexpose as much as posible. You have to figure out how much you can bring back in converstion and push it. That makes raw+jpg useless, except for raw as a back-up for jpg.
04/19/2007 12:20:05 AM · #14
Originally posted by Niten:

I can tell by reading that some of you are wasting your megapixels. With digital raw always overexpose as much as posible. You have to figure out how much you can bring back in converstion and push it. That makes raw+jpg useless, except for raw as a back-up for jpg.


That's easier said than done! At least I haven't had much success with it...

The theory behind deliberately overexposing (often called "exposing to the right") is that it increases the signal to noise ratio, so you get a cleaner photo. (When you reduce the exposure during RAW conversion you reduce the noise as well.) Unfortunately, you also lose some margin for error; if you overexpose too much, you can't correct it and your low-noise shot is ruined.

This can certainly be a useful technique. Especially for low key photos which are more prone to noise. (And it's certainly a test of your visualization skills to look at the somewhat bright preview and see the dark photo you envisioned!) But I personally stop short of always using it.
Pages:  
Current Server Time: 04/19/2024 03:23:57 PM

Please log in or register to post to the forums.


Home - Challenges - Community - League - Photos - Cameras - Lenses - Learn - Prints! - Help - Terms of Use - Privacy - Top ^
DPChallenge, and website content and design, Copyright © 2001-2024 Challenging Technologies, LLC.
All digital photo copyrights belong to the photographers and may not be used without permission.
Current Server Time: 04/19/2024 03:23:57 PM EDT.