View Full Version : what zoom level is enough?
DJ-S1
Apr-27-2005, 05:07 PM
Just wondering, when you folks are cleaning up an image what is the closest zoom level you use? IOW, if you can't see a defect (say a halo or some artifact) at 100% zoom, is that sufficient for printing? Or do you clean things up until you don't notice them at 150%? Or 200%? :scratch Or does it depend on the size of the print you expect to make?
XO-Studios
Apr-27-2005, 05:14 PM
Just wondering, when you folks are cleaning up an image what is the closest zoom level you use? IOW, if you can't see a defect (say a halo or some artifact) at 100% zoom, is that sufficient for printing? Or do you clean things up until you don't notice them at 150%? Or 200%? :scratch Or does it depend on the size of the print you expect to make?
Hmm I always do a final check by setting ZOOM PRINT SIZE (important to have monitor resolution size properly calibrated in PS) and double whatever zoom levels comes up in there.
XO,
cletus
Apr-28-2005, 05:18 AM
Never given much thought to this :scratch
When working on complex masks I tend to work at fairly high zoom levels (400% or more) and every once in a while drop back to 100 or 200% to check my progress. If the mask looks alright at 400% and at 100% I would bet that it will look alright on any print size.
mercphoto
Apr-28-2005, 07:20 AM
Just wondering, when you folks are cleaning up an image what is the closest zoom level you use?
Personally, I've never understood why anyone would go beyond 100%. At that point the software is inventing pixels to display on your screen. Why bother? What does it prove? Why does it help? And you're not going to see it on a print.
XO-Studios
Apr-28-2005, 12:31 PM
Personally, I've never understood why anyone would go beyond 100%. At that point the software is inventing pixels to display on your screen. Why bother? What does it prove? Why does it help? And you're not going to see it on a print.
From the PS CS helpfile
Magnifying and reducing the view
You can magnify or reduce your view using various methods. The window's title bar displays the zoom percentage (unless the window is too small for the display to fit), as does the status bar at the bottom of the window.
Note: The 100% view of an image displays an image as it will appear in a browser (based on the monitor resolution and the image resolution).
cletus
Apr-28-2005, 01:15 PM
Personally, I've never understood why anyone would go beyond 100%. At that point the software is inventing pixels to display on your screen. Why bother? What does it prove? Why does it help? And you're not going to see it on a print.
There are lots of times I find higher magnifications to be helpful... For example - when I'm working with a complex mask, by working at higher magnifications I don't have to be quite so steady handed with the paintbrush and other 'free-hand' tools. There are other times when viewing the image at 100% the pixels blend together making it difficult to see exactly what's going on. By working at higher magnifications the enlarged pixels don't blend, making some details easier to see.
wxwax
Apr-28-2005, 01:53 PM
For precise masks, I make the pixels reasonably large so that even my unsteady had can make accurate distinctions.
I've read that for sharpening, there's no benefit to greater than 100%, because that's the most you'll see in a print.
DJ-S1
Apr-28-2005, 02:30 PM
So to paraphrase, for masking zoom a lot and for other stuff more than 100% is overkill. So if a speck is visible at 250% but not at 100%, no need to clean it up. Sound right?
The reason I ask is that sometimes I find myself chasing stuff all around an image only to realize that I'm zoomed up pretty high. Then again, I've also seen some images here on the site in the past where someone will comment on how noisy it is, but I can't see it at 100%. If I blow it up, there it is clear as day. Or someone will say that an image was oversharpened because they can see artifacts, but they are really only apparent at high zooms. It's confusing, and I wasn't sure just how "clean" I needed to get an image.
So I'm just trying to figure it all out. Thanks for the conversation. I guess what I'm taking away from this is (for example) that I shouldn't run images through a noise program unless the noise can be seen at 100%.
wxwax
Apr-28-2005, 03:02 PM
So to paraphrase, for masking zoom a lot and for other stuff more than 100% is overkill. So if a speck is visible at 250% but not at 100%, no need to clean it up. Sound right?
The reason I ask is that sometimes I find myself chasing stuff all around an image only to realize that I'm zoomed up pretty high. Then again, I've also seen some images here on the site in the past where someone will comment on how noisy it is, but I can't see it at 100%. If I blow it up, there it is clear as day. Or someone will say that an image was oversharpened because they can see artifacts, but they are really only apparent at high zooms. It's confusing, and I wasn't sure just how "clean" I needed to get an image.
So I'm just trying to figure it all out. Thanks for the conversation. I guess what I'm taking away from this is (for example) that I shouldn't run images through a noise program unless the noise can be seen at 100%.
That's what I'm getting out of it. :nod
colourbox
Apr-30-2005, 12:02 AM
So to paraphrase, for masking zoom a lot and for other stuff more than 100% is overkill. So if a speck is visible at 250% but not at 100%, no need to clean it up. Sound right?... I guess what I'm taking away from this is (for example) that I shouldn't run images through a noise program unless the noise can be seen at 100%.
Pretty much. The special thing about 100% is that it is the view where one image pixel maps exactly to one monitor pixel. At lower magnifications, each monitor pixel is an average of more than one image pixel, which isn't accurate. At 200%, each image pixel just repeats on four monitor pixels, so there's nothing more to learn than at 100%. The only reason I zoom in past 100% is that I sit farther from the screen than most people.
vBulletin® v3.8.5, Copyright ©2000-2012, Jelsoft Enterprises Ltd.