NPeterman From United States of America, joined Apr 2009, 163 posts, RR: 0 Posted (2 months 3 weeks 4 days 3 hours ago) and read 2210 times:
Since I am not asking for redress on a particular photo, and because I know this area of the form gets more traffic than the feedback section, I wanted to pose this question here. Over the past few weeks, I have seen an increasing number of oversharp rejections. After 1 or 2 I started uploading images without any sharpening applied in post processing whatsoever, however I am still receiving some oversharpened rejections. So I am curious, what would you do? Aside from simply not applying any sharpness at all and uploading out of camera minus crop and exposure corrections (which I am already doing), I am kind of at a loss. Surely I need not apply some sort of blur filter here...
mjgbtv From United States of America, joined Jan 2008, 587 posts, RR: 0 Reply 1, posted (2 months 3 weeks 4 days 2 hours ago) and read 2200 times:
How do you re-size your images? I found that re-sizing in a single step often left jagged areas even before I applied any sharpening. Per a suggestion in the forum I switched to re-sizing in a few stages and found that I get better results.
NPeterman From United States of America, joined Apr 2009, 163 posts, RR: 0 Reply 2, posted (2 months 3 weeks 4 days 2 hours ago) and read 2199 times:
That is something I can certainly try doing, although Having always basically done it in one step it seems funny that I should start having this problem now. I guess my lens is getting sharper with age.
photopilot From Canada, joined Jul 2002, 2540 posts, RR: 19 Reply 3, posted (2 months 3 weeks 4 days 2 hours ago) and read 2179 times:
You should also check to see what (if any) level of sharpening your camera is applying to images as it writes them to the chip. Most cameras will allow you to either adjust this level, or de-select it altogether so that the camera doesn't modify the image.
NPeterman From United States of America, joined Apr 2009, 163 posts, RR: 0 Reply 4, posted (2 months 3 weeks 4 days 2 hours ago) and read 2167 times:
My in camera sharpening is set to off, as it has been for the year + that I have owned this camera. One theory that a friend of mine has put forward is that it could have something to do with the method that photoshop is using to resample the image when I resize, which is something else I can try as well.
N243NW From United States of America, joined Jul 2003, 1570 posts, RR: 21 Reply 5, posted (2 months 3 weeks 4 days 2 hours ago) and read 2161 times:
Not only can the number of resizings affect a photo's sharpness, the resampling method (in Photoshop, at least) will make a huge difference.
Here's a photo (before sharpening), resized using the "Bilinear" option in the Resize dialog (my preferred method).
And here's the same photo, resized using the "Bicubic Automatic" dialog. Notice the areas of high contrast appear to have some strange halos around them. I had no trouble using this Automatic option until I upgraded to CS6 - then something funky started happening.
Until I figured out that this was the cause of the weird editing halos in my photos, I got several oversharp rejections and really wasn't happy with their overall sharpness. To me, the "Bicubic Automatic" option makes photos appear softer than they actually are, often leading to oversharpening. Others may find a different editing workflow and resizing method that works better for them, of course.
Just curious - who here uses the Bicubic Automatic option and who uses the Bilinear option? Anyone else notice some pretty big differences between the two?
ckw From UK - England, joined Aug 2010, 660 posts, RR: 17 Reply 6, posted (2 months 3 weeks 3 days 19 hours ago) and read 2081 times:
Ryan's comments (which I may have misundertsood) suggest he is sharpening AFTER resizing. Resizing should be the very last step in any process.
Of course depending on your method of resizing you may need to adjust your sharpening technique - for example, if you sharpen the full size image to within an inch of it's life, and then resize using bicubic sharper, you are quite likely to get an oversharp image.
Doing it the other way round might therefore seem more sensible, but the problem is resizing throws away a lot of data and tends to 'bake' the image (especially round the edges) so applying further processing to a resized image will result in less than optimum results.
RCoulter From United States of America, joined Apr 2007, 507 posts, RR: 0 Reply 7, posted (2 months 3 weeks 3 days 19 hours ago) and read 2076 times:
Quoting ckw (Reply 6): Resizing should be the very last step in any process.
Not always. My last step is checking for dust spots via equalizing in PS. I do most adjustments in Lr4 (exposure, color/wb,curves,sharpening,nr) then go into CS6 for final cropping/leveling > resize > sharpen > dust spots > save as JPG.
Kaphias From United States of America, joined Nov 2010, 237 posts, RR: 0 Reply 8, posted (2 months 3 weeks 3 days 18 hours ago) and read 2057 times:
Quoting ckw (Reply 6): Resizing should be the very last step in any process.
Curious about this: I've always resized then sharpened as my last two steps, as the editing guide recommends I believe. I've only ever had one over-sharp rejection and just a handful of grainy rejections– the latter most likely due to using a point-and-shoot in overcast/rainy conditions. So while I understand why resizing should be the last step, I've never had an issue with my sharpening methods.
ckw From UK - England, joined Aug 2010, 660 posts, RR: 17 Reply 10, posted (2 months 3 weeks 3 days 10 hours ago) and read 2001 times:
Quoting Kaphias (Reply 8): Curious about this: I've always resized then sharpened as my last two steps, as the editing guide recommends I believe.
I'm not suggesting you can't get good results working any other way - if it works for you go with it. But my philosophy is that every edit in Photoshop (or lighroom for that matter) is fundamentally destructive - yes, I know you can use layers and change your mind.
Each step in the editing process throws away a certain amount of data, or at least alters it irrecoverably. Theoretically
the order you do things should take into account how each step might influence the next. Personally I like to do some intial editing (colour balance, contrast, noise reduction, leveling) in raw conversion as provides the full data of the RAW file. I;d adjust curves and sharpen here as well except it doesn't have the flexibility of these tasks in PS.
Once in PS, I go curves -> sharpen -> resize. If for example you sharpen before curves, you are altering the edge contrast significantly which greatly limits what you can do with curves without getting unwanted effects.
It is of course possible that after resizing you may find the sharpening needs a little adjustment - hence its probably safer to undersharpen in the intial stage until you get a feel for what settings will work.
Any dust spotting or similar, I do after sharpening (as this helps reveal any issues) but before resizing.
But let me stress - this is one of many ways of achieving a similar end result. If you don't have issues with what you do, stick with what works. If you are having issues, might be worth a try.
vikkyvik From United States of America, joined Jul 2003, 9017 posts, RR: 28 Reply 11, posted (2 months 3 weeks 3 days 3 hours ago) and read 1952 times:
Quoting N243NW (Reply 5): Just curious - who here uses the Bicubic Automatic option and who uses the Bilinear option? Anyone else notice some pretty big differences between the two?
I just use the standard Bicubic (I think it's the "best for smooth gradients" one). I don't recall my PS CS3 having a "Bicubic Automatic" option. Perhaps I've never really looked, though.
Bicubic Sharper results in a resized image with variable sharpness, in my experience. It makes further sharpening a pain, especially since some parts might already be oversharp. I feel I have much more control using unsharp mask after resizing.
"Two and a Half Men" was filmed in front of a live ostrich.