Archive for the ‘Workflow’ Category
Saturday, June 5th, 2010
If you follow me on Twitter (and you should!), you know that I just got myself a copy of Adobe Creative Suite 5 - including Photoshop, of course, but also Illustrator, InDesign and Acrobat 9 - thanks to my newfound ability to take advantage of crazy academic discounts on software.
Above is one of my favorite crowd shots from Bamboozle Roadshow, which I’ll be posting about soon. Unfortunately, it was 2-3 stops underexposed because the stage lights went out just as I was taking the photo. The version you see above is what I managed to salvage using Adobe Camera Raw 6.1, Photoshop CS5 and the latest 64-bit version of Noise Ninja. If you hover your mouse over the image and wait for a second (sorry, the files are large), you’ll see the earlier version I originally posted, processed using Nikon Capture NX2, Photoshop 7.0 and an older version of Noise Ninja.
I’m not actually sure which version I like better. With Camera Raw, I was better able to salvage the shadows while still keeping the noise at acceptable levels. I screwed up the custom white balance a bit though; I like the cooler tones of the original image better than the orangey tones of the new one. That said, the original image feels like an underexposed image to me, which bothers me; the new one feels a lot less so. It’s all personal preference of course, but I’m pretty happy to have a new tool at my disposal. When processing images that have been properly exposed and need minimal post-processing work, it might not matter much. But when trying to salvage images that I screwed up in-camera, or when I want to use artistic effects for portraits, wedding shots, etc, my options just expanded enormously.
(Also: let’s take a moment to marvel at how an ISO 3200 image can be brought up nearly 3 stops in post - boosting effective ISO to 256000 - and still be usable at all.)
Monday, August 17th, 2009
For the curious, here is my standard post-processing workflow. It is highly inefficient but I’ve streamlined it enough that I can get things done relatively quickly. Nevertheless, I should probably buy a copy of Lightroom, which might (from what I’ve heard) combine a number of these steps into one.
Things to know beforehand: I shoot 12-bit RAW almost exclusively; about the only time I use JPEG is when I’m shooting outdoor sports in good light. I have a relatively powerful computer with dual displays, of which the primary display is calibrated regularly using a Spyder2Express tool. I use an ancient version of Photoshop that I hope to upgrade sometime in the near-ish future (I have several other things higher on my photography want-list, though). I generally process with Web usage in mind rather than print usage, so I live exclusively in the sRGB color space and am fairly clueless when it comes to advanced color management.
Anyway, here goes:
- Download photos from CompactFlash card to computer using Nikon Transfer. All my photos go in two places: one copy in a working directory on my main hard drive, and one backup copy onto my Drobo. I only use the Drobo for backup, never for working.
- Extract JPEGs from NEF files using PreviewExtractor. This is a great piece of freeware that pulls the preview JPEGs from each NEF - the JPEGs are full-size but limited quality and generally come to less than one megabyte each. I extract the previews only on my working hard drive, not on the Drobo.
- Browse the JPEG previews and select images to process. I use an old version of ACDSee (v4.01 I believe), which I like simply because it’s fast, completely bloat-free, and I’m comfortable with the interface. Within ACDSee I make copies of the images I want to process in a new directory on my working drive.
- Process NEFs using Nikon Capture NX2, and save as JPEGs. This can involve any combination, but rarely all, of the following: exposure tweaks, contrast tweaks (the contrast slider in NX2 is fantastic), white balance, picture control (I use STANDARD and D2XMODE1 probably 90% of the time), and noise reduction. I do this to each NEF individually; rarely am I comfortable with batch-processing since I try to get the best look for each individual image. I save the tweaked files as maximum-quality JPEGs.
- Process the resulting JPEGs in Adobe Photoshop 7.0. This can involve any combination, but rarely all, of the following: cropping, rotating, curves adjustment, various local adjustments (sharpening, dodging, burning, cloning etc), application of Noise Ninja, application of any preset actions or global tonal shifts (very rarely), resizing/sharpening/watermarking/saving for web. I save the full-size version as a maximum-quality JPEG, and the web-resized version is generally 800 pixels on the long side, also saved as a maximum-quality JPEG.
- Copy the final, processed JPEGs to the Drobo. I back up both the full-size processed versions and the web-resized versions on my Drobo, which has a directory structure organized by date (one folder per month) and then by shoot (one subfolder per concert, portrait shoot, etc).
- Upload to web. Generally to either my Flickr site or my Zenfolio site, depending on the intended usage.
Wednesday, June 3rd, 2009
Sorry, that’s me talking in marketing jargon. Above is a Drobo, which is sold as “the world’s first data storage robot” or some nonsense like that. It’s not a freaking robot. It’s a 4-bay hard drive enclosure that has a particularly clever RAID-like data striping/mirroring implementation built in. As far as I can tell it’s something like RAID 5, but it’s smart enough to be able to handle non-identical (in size, spec or brand) hard drives and changes its strategy based on the number and size of the drives it has to work with. Right now I have two 1.5 terabyte Western Digital Caviar Green drives installed in my Drobo, so basically all it’s doing (as far as I can tell) is a simple RAID 1 mirroring strategy. If I were to add a third or fourth drive, its strategy would change to something more sophisticated. The real amazing thing is that drives are hot-swappable while you work - as long as you have more than one drive installed, you can literally pull one out and still be able to access all your data with no interruption.
Installing and uninstalling drives is a snap, too - no cables to play with, just take a 3.5″ SATA hard drive and slide it into a slot. Bingo. The ease of use of this thing is amazing, and it’s nice to know that my photos are backed up on a secure device that can withstand the failure of a hard drive (but not two at once). I was running out of space for all my photos, and this seemed like a much easier, and ultimately cheaper, solution than building a file server or an external RAID array. What’s more, it’s easily expandable, as I still have two open bays and I can always swap out older, smaller drives with newer, larger ones. But I should be good to go for a year or so.
This thing is apparently very, very, very, very, very, very popular among photographers.
Sunday, April 26th, 2009
My post-processing routine for concert photos involves cropping, curves, noise reduction, and sharpening. Sometimes, if the lights make skin tones go green or some subtle but sickly shade, I’ll correct for that. But I don’t generally go beyond that. Recently, though, I shot a show at Jaxx where the lighting consisted of nothing other than deeply saturated reds and blues. When the band members were lit with all red, I didn’t have many options either at the time or in post-processing, but when they were lit with blue, some serious color balancing was possible after the fact.
Roll your mouse over each image - above, Sweden’s The Haunted, and below, Canada’s The Agonist - to see what they looked like before color correction in Capture NX. (If you’re viewing this in an RSS feed reader, you’ll probably actually have to visit the blog to get this to work.)
I don’t normally like for my concert photos to be so unrepresentative of what I was actually seeing, but those deeply blue images are so unattractive that I’m willing to make exceptions in this case. By bringing down the blue channel and pumping up the red - which, of course, comes with a definite penalty in higher noise levels - I can make skin tones look reasonably decent while also making the background (if it happens to be red fog) much more dramatic. It’s a win-win as long as the original exposure is good enough that the image can handle all this manipulation without become a blizzard of noise.
Lots more photos from that show here, the majority of which were manipulated in this manner to get more pleasing colors.
Monday, January 26th, 2009
So I’m shooting a wedding in May, and as part of my educational process have been browsing through the portfolios of countless wedding photographers, from the very best (and most insanely expensive) to the more, er, down to earth. It occurs to me that in the field of wedding photography there’s a proliferation of a certain kind of hack: the photog who has a copy of Photoshop, wants to follow the trend of soft-focus, “flattering” or “romantic” post-processing, but doesn’t really have the chops to do it. I can’t even describe the number of otherwise perfectly fine wedding photos I’ve seen ruined with cheesily overdone tilt-shift effects or heavy-handed retouching that makes people look like plastic dolls. I don’t mind the idea of soft or selective focus, or retouching, or whatnot, in general, but man, if it’s not done right, it looks awful.
I started playing around with some wedding actions in Photoshop and I do sort of like the effect sometimes. But I’m very aware of the fact that this is not my style and I need to be really careful about overdoing it. My general rule of thumb regarding pretty much any Photoshop effect: if I think I might need to tone it down, I need to tone it down.
Wednesday, January 7th, 2009
Warning: excessive geek-talk follows.
Ever since I started using Nikon Capture NX to process the 12+ megabyte RAW files from my D300, I’ve known that my camera equipment has outpaced my computer equipment. (Odd, since I have been a computer nerd since long before I became a photo nerd.) High ISO noise reduction in particular slowed my system to a crawl, making post-processing concert photos a much longer process than necessary. I’ve been wanting to upgrade my old version of Photoshop (7) as well, but have been afraid that my computer wouldn’t run anything newer than CS2 very well.
I’m using a machine I built myself five and a half years ago, so it’s definitely due for an upgrade. The big photography-related question was: what monitor to get? Currently I use a Dell Ultrasharp 2005FPW, which is a 20″ widescreen LCD based on a high-quality S-IPS panel, and an old 15″ LCD that’s beginning to show signs of conking. I wanted like to get a new monitor to replace the 15-incher and still have a dual-monitor setup, and a bigger one with HDCP support would be great since my DVD drive replacement is a Blu-Ray compatible drive.
I tend to do lots of photo processing, lots of using Internet/e-mail/office apps, some movie watching, and a tiny bit of gaming, all of which require slightly different strengths in an LCD screen. Photo processing is the most demanding from a color reproduction standpoint, and S-IPS panels are the way to go for that. Trouble is, S-IPS panels are freakin’ expensive (we’re talking $1000+ for a good 24″ S-IPS panel). All the cheapest monitors are TN panels, which are 3-4 times cheaper than the S-IPS ones, and are great for gaming since they tend to have fast response rates, but poor at color reproduction and horrible in terms of viewing angles.
Realistically I had to settle for a PVA panel like that in the Dell Ultrasharp 2408WFP or 2707WFP, which are newer and larger versions of my current monitor, and which are available at quite reasonable prices refurbished. After some back and forth I have decided to go with the 2707. I’m pretty psyched about it; 27″ is a lot of screen real estate for all those Photoshop toolbars and the like. I’ll get the monitor, as well as the rest of my upgrade parts, later this week. Woo!
Check out this Anandtech thread for tons of info on selecting LCD monitors - very helpful indeed.
One concern: I am upgrading to Windows Vista 64-bit, and I was a bit worried that the color calibration software that came with my Spyder2 device isn’t Vista 64 compatible. Luckily, I Googled and found this. Nice.
After the jump, the full details on what I’m upgrading from and to.