Boise, US
|
I don't have the D700, but my D200, D2X, and D3 all deliver stunning prints at the 13x19 size (btw, I wish this were a more standard format; it's an impressive size, and a very minimal crop of the sensor's native 2x3 ratio). I routinely enlarge to 20x30 with either the D2X or D3, and lately every portrait customer has come into the session asking for a 24x36 up front. No problem, but the margin for error does begin to get very small once you go beyond doubling the file (roughly 19x28").
As mentioned elsewhere in the thread, there are two ways to get to an enlargement. One is to simply lower the print resolution, thereby enabling the pixels you recorded to cover a larger area in a print. Deselect the "resample" button in PS's Image/Size menu, and changing dpi from the industry-standard 300 to 200 bumps your print output size from +/- 9x14 to 14x21. Conversely, of course, if you were to select a 24x36 output size, the program would automatically reduce your resolution to 118 dpi. In fact, if you send your out-of-the-camera file to a lab and request a 20x30 print, that's likely how they'll do it. After all, that's how we enlarged film before the digital era enabled us to manufacture pixels after the fact through interpolation. Note, btw, that sized the same way on your monitor, all of these resolutions will look the same, which is to be expected. You're not actually altering any pixels in the computer environment; you're just inputting a formula for stretching those same pixels during printing.
If you don't want to sacrifice resolution to gain size, the other way to create an enlargement is to uprez your file by interpolating new pixels. If you leave the "resample" button checked and fix your dpi at 300, when you want to increase the natural 9x14 print size to 24x36, obviously you're going to have to find some more pixels to fill in the spaces you've just created. You can use PS's built-in enlarger or a specialty plug-in like Blow Up or Genuine Fractals; I've tried them all, and, though there are minor differences, I don't know that the after-markets provide enough increased quality to justify their hefty prices. Other factors are far more important.
First, if your want to make clean 24x36 prints (or larger) you absolutely have to start off with a perfect exposure. If you overexpose, blown-out faces that escape detection at 8x10 will haunt you like pasty ghosts at the larger sizes. While you can always bump up an underexposure in PS, the increase in noise will become annoyingly visible in a big print. Watch your histograms, manage your contrast, and make sure your main subject is where you set your primary exposure.
It should go without saying that any problems with focus will completely ruin a nice 8x10 when it grows to 20x30. If you know you're going to make a big print, put it on the tripod, give yourself another stop down for leeway, and use a cable release. In these circumstances, even though I'm tripod steady, I'm also still careful to stay within the traditional focal-length-inverted shutter-speed guidelines for the lens. Obviously, I also crop exactly for the finished print in the viewfinder; this isn't medium- or large-format, and you can't afford to throw away any pixels to fix composition problems you should have noticed at the time of exposure.
The bottom line is, you're really stretching the capacity of your pixels when you enlarge past double-size, and the margin for error seems to shrink exponentially. Shoot RAW in Adobe RGB, and use all of those available 14 bits. I also apply as little processing of the file as possible in the camera. You want to save this for PP, because you want to be able to apply these settings to your finished size, rather than having to stretch your fixes after they're done.
In the RAW converter (I always use ACR), I never apply any settings I can save for PS, particularly sharpening. I color-correct, adjust black/white points, and then send the file to Photoshop. Even in PS, I save sharpening to the very last, after the file has been up-rezed to the output size. It pays, btw, to have a computer loaded with RAM; a 300 dpi, 24x36 file with a few layers active will slow things to a crawl and, eventually, crash your machine.
As others have mentioned, the apparent sharpness of your finished print is also a function of the viewing distance. OTOH, when I show the print to a customer, the first thing he does is put his nose next to the print to see how it looks "up close." Personally, I prefer not to have to explain that his print will look better if he stands back. I'd like it to look sharp no matter where he views it from so I don't have to sound like I'm making excuses for shoddy work.
That said, imho, the new-generation digital cameras have long ago bypassed 35mm film in quality. I don't know very many photographers from the old days who could routinely expect to get a sharp 24x36 print out of 35mm film. Most (including me) would tell you that 11x14 is already pushing 35mm quality. Every high-end wedding photographer shot MF, as did any respectable portrait studio. I can't imagine dragging a MF camera to a wedding these days when my D3 produces comparable quality. I compete handily in the frameable portrait market with MF houses in town, and my customers are delighted with the price and the quality. Recently, while vacationing on the Oregon Coast, I stopped in to a local landscape photographer's gallery and was astonished at the breathtaking quality of his 30x40 prints. When I asked him what he was shooting, he told me, much to my surprise, that he was using Nikon digital cameras, and that his current flagship was a D2Xs. I admit, I'm in awe of his ability, and I certainly don't have the nerve to offer 30x40 portraits to my customers with my current gear, but clearly the gear is not the limitation.
Hope that helps.
Bruce Jones
Writer, rider, shooter
|
