This content has been marked as final. Show 8 replies
Basically you do both a capture sharpening in ACR to counteract blurring inherent to the sensor/lens combination and an output sharpening. Sometimes you do an intermediate step of creative sharpening. About your last question, make sure you use ACR at the native RAW resolution, optimize the image for sharpness at 100% in ACR, and subsequent upsampling should be as good as is possible.
I noted in Bruce's book "Imaging Sharpenning with...CS2" he has a four phase approach. 1) Source 2) Content 3) Creative 4) Output
RWCR for CS3 show's that ACR 4 now provides the "capture sharpening" which I take to be step #1. Where does content sharpening come into play? Is that the mask setting in ACR 4?
If you read the book carefully, you'll see that Bruce writes that the Source & Content are combined into a Capture Sharpening approach. The thing you are not grasping is that for Capture Sharpening there are two factors...the source of the images, capture size or film...and then the nature of the image itselfthe content.
As for the OP. I would suggest doing the capture sharpening in Camera Raw, yes. Then, if you do need to upsample (and you really should have a good reason, not just to get a bigger file), use Bicubic Smoother to upsample and you'll then need to do additional sharpening after the upsample (but not really just a Photoshop USM) and a final output sharpening ate the final output size.
If I use ACR 4.3 to sharpen an image at native camera res and then decide to down-sample in ACR - is the sharpening adjusted or do I need to revisit the sharpening settings?
Pretty sure sharpening is done before the down sample in Camera Raw in the pipeline and the down sample is a Lanczos algorithm (see http://en.wikipedia.org/wiki/Lanczos_algorithm for geek explanation) which has a "degree" of sharpening in it.
The only thing I would check is the impact a smaller output size would have on the radius setting. You can do this by setting your desired output size in the workflow settings and then viewing the image again at 100% in Detail. If it still looks "ok" at the down sampled size then you're good to go.
Thanks for the quick answer! I did look in your (most excellent) book before asking!
That is a great tip to check the Radius setting one more time before outputting, which at the lower resolution would be for quick client review, email, or web posting...so its not something Im going to loose a lot of sleep over so to speak.
Masters in Digital Photography Program Information
>Pretty sure sharpening is done before the down sample in Camera Raw in the pipeline and the down sample is a Lanczos algorithm (see http://en.wikipedia.org/wiki/Lanczos_algorithm for geek explanation) which has a "degree" of sharpening in it.
Your Wikipedia link is indeed very geeky--It talks about recovering Eigenvalues, but doesn't even mention resampling. Perhaps this link is more appropriate Lanczos Resampling.
I have no idea whether ACR sharpens before or after resampling, but to sharpen before resampling seems counterintuitive, since it is often advised to sharpen after resampling to offset the loss of sharpness occasioned by the resampling. With significant downsampling the sharpening halos would be resampled out of existence. Perhaps the Lanczos algorithm obviates these concerns.
Pretty sure both crop and resample is very early in the pipeline...but the issue of radius would still apply. And yes, that's a better link.