cerkb wrote:gstark wrote:
Unsharp mask is something to be used carefully; it's all too easy to oversharpen an image, which will introduce other undesirable artefacts. And on a noisy image, USM will make the noise look even worse.
Interesting. I've thought I read some time ago in various message forums (possibly including this one) that most D70 shots with the kit lens with default settings can benefit from some additional sharpening. This seems like a pretty generalized statement, but it seems like I'm not getting a lot of sharpness in a lot of my sports shots.
OK ...
We're talking about a few different, but similar, and certainly related, things here.
First of all, you state "most D70 shots with the kit lens with default settings can benefit from some additional sharpening".
Let's add to that a little. Not just the D70. Not just the kit lens. Not just the default settings.
So now we have the statement that most digital camera shots can benefit from some addittional sharpening. That is true. Yes, it's a generalisation, but it's still true.
There are various ways to impose some sharpness into an image - camera settings is one such way, USM in PP is a second, and there are others, and they all will help.
But you need to be careful to not go overboard in this. I don't think I've seen an instance of in-camera sharpening going too far (over-sharpening) but I've seen lots of cases - including my own images - where it's been overdone in PP.
That's what I was talking about in the quoted section above, and it's an acquired art - knowing just what id the "right" amount of sharpening, and knowing when to stop.
There's no correct answer to this question either: it varies for every image you shoot, and it's only your experience and viewing of the image than can guide you as to when enough is enough.
The final point you raised in the section above relates to the sharpness you see in your sports shots. Without seeing some examples from you, it's really difficult to say of this is a result of insufficient in-camera sharpening, insufficient PP sharpening, or perhaps a problem whilst shooting.
But based upon your earlier comments in this thread, my guess is that it's the latter, so we need to concentrate on shooting technique and exposure variables first, and thus get those elements under control.
When we have exposure nailed, and we're consistantly getting some usable images to work with, we can then move on to making them look as good as they can with PP.
This is probably due to low light conditions and my limitations on glass as several of you have pointed out. But can I improve my shots in these situations by using some post processing for sharpness?
Possibly, but don't rely on it.
Your goal should be to get the best possible image you can in the camera. That gives you the best quality raw materials to work with. PP can often make a good image great, but it's a much harder task to make a lousy image even barely good.
If so, what is the best way to sharpen an image that appears to need it? Should I use Unsharp Mask or the Sharpen tool? What's the difference between these? What settings should I use on these and what should I look for to determine if I've gone too far?
The best way is whatever works for you. Try a few different methods, and compare results. See what you like.
To determine if you've gne too far, carefully examine the sharper edges that are rendered within the image. Perhaps a leg, or a forearm. Look along the edges for any signs of haloing. Difficult to describe, but you'll know it when you see it.
Learn to read the histogram. It's your friend, and it's here to help you.
If I understand correctly, having a "balanced, centered" histogram will indicate good exposure but does it also give indications of clarity? How can I tell from the histogram if the shot is "sharp" or "clear"?
No. The histogram is only a guide as to expoosure. It's just a graph of the pixels' brightness, in the range of 0-255. More darker pixels will tend towards 0 (left), while more brighter ones will tend towards 255 (right).
If your subject is wearing dark clothes, or is in a darker area, you should expect correct exposure to tend towards the left. That will be fine, given the circumstances of the subject.
But there is no indication of sharpeness within the histogram.
If you're shooting in A
mode, you won't need to - you won't be able to - set the shutter speed. Not directly. If your images seem too dark, if your histogram is falling off the left of the frame, then start to bump your ISO.
If they seem too bright, or the histogram is falling of the RH edge of the frame, pull the ISO, or bump the aperture.
It appears I can adjust the shutter speed while in AP
mode with the front thumb wheel. Should I not attempt to do this and rather adjust the ISO instead as you suggest?
With the default settings on the D70, the front command dial will adjust the aperture. The rear command dial adjusts the shutter speed. This behaviour can be altered through the menus,
If you're in A
mode and you adjust the aperture (as described above), this will have an effect on the other exposure variables you have set, and thus the shutter speed may also change as a result.
Similarly, if you're in S
mode, changing the shutter speed will have a run-on effect and alter the aperture that the camera's exposure system calculates as correct for the given conditions.
The only
modes I use on the camera are A, S, and M, and I'm not really qualified to speak to the other
modes, but I've yet to run into limitations of the hardare within those three
modes - the greatest limitation is the idiot driving the camera.
By the way, I've been shooting "Jpeg Fine Large" as my normal setting. Before doing any post processing, I normally create a Tiff version to do edit work on, and then may save as a jpeg depending on what my final goal is (e.g., printing, slideshows, archives, etc.). Several people have suggested shooting RAW. Would this introduce significant advantages to the quality of my shots or further post processing?
What you're doing is fine, but it does present you with some limitations.
JPG is an inherently lossy format. The first time that you save an image, you have lost data. While you're saving in tiff you are temporarily seeming to overcome this issue, the point is that, right from the very get-go, what you've started with has missing information. All of your in-camera settings (wb, sharpening, etc) have been applied, but the underlying (untouched) data has been discarded.
Were you to shoot in raw, you would have saved the untouched data, along with the in-camera settings, but those setting would have been saved as parameters rather than as a fait accomplis. Working in raw you can change those paramaters, and perhaps correct some shooting errors committed at the source.
For instance, last Saturday I was shooting a concert, and switching between available light (ISO 1600, incandescant wb) and SB800 (ISO 200-400, flash wb) throughout the evening. a few times I would switch from one
mode to the other, but forget to change my wb.
Working in raw it was a trivial exercise to open those affected images and reset the wb to the one more appropriate to how I shot the image. Then I could (will) go back to my normal (minimal) workflow and see what else needs to be done to those images, but as you can see, I was able to easily correct a fundamental error with no loss of any quality.
Had I shot in jpg, the error correction would not have been as trivial a task, and any act of saving the data may have resulted in a further loss of data and/or image quality.
Thanks again Gary for your patience and my further questions.
My pleasure.
Posted: Sun Oct 29, 2006 11:31 am
).
