18th April 2012, 08:12 AM
I don't have Photoshop but I have recently been discovering layers using The Gimp software. As an example one technique was labelled as a 'Contrast layer'. This was achieved by duplicating the layer then desaturating and inverting the duplicated layer. Funnily enough to sharpen up the whole thing the last part of the technique was to apply a 'Gaussian blur' to this, to sharpen the whole image. The overlay option was then chosen to bring the two layers together.
Having done the whole process it did seem to overall bring a much crisper and sharper feel to the image.
So, I guess that leads into my question which is, is the layering technique an addition to what I would call normal processing i.e. adjusting black, white points etc. Or is what is achieved by layering something that can be done in the normal process?
Cheers for now
Last edited by oldgreygary; 18th April 2012 at 08:19 AM.
18th April 2012, 11:15 AM
I've always considered this 'unsharp mask' processing as part of the sequence (but I'm quite prepared to be in a minority here...)
As a general rule I have two versions of anything I keep - the original and a 'modified' copy, the latter containing any mods (USM, noise, white points) etc.
24th April 2012, 01:57 AM
Layers are a good addition to "standard" processing techniques inculding levels and color ajustments. The particular process you described is available as a filter in GIMP called "unsharp mask" along with many other operations in the filter menu. In most "standard" fixes you will not need to do any custom layers work, as all of the operations are available. If you are doing any in depth processing you will need to use layers to augument the prebuilt operatiors.
24th April 2012, 06:24 PM
For me, Gary, layers and masks are an absolute essential for editing; as I have so often mentioned previously.
Have I already given you this link? http://www.ronbigelow.com/articles/l...1/layers-1.htm
Gets a bit involved, and is based on Photoshop operation but Gimp will be similar.
Editing with Adjustment Layers is, as far as I'm concerned, the only way to work. But the term adjustment layer often refers to a number of techniques.
Basically, creating a duplicate layer of the background then editing that layer, sometimes with various blend mode changes, and using a mask which can be selectively edited to limit the adjustment to just the required areas.
Using the best sizes and opacity/softness of your brushes is something to be learned, but fairly simple - mostly. And trying to remember whether you are using a Reveal or Hide mask, and which 'colour' of brush works to edit that particular mask still catches me out on occasions.
I don't know if Gimp has actual 'Adjustment Layers' as an option but this, very simply, means adding a 'blank' layer above the background image then editing the layer (eg with Curves) so everything below that point in the layers stack becomes changed.
And one of the great advantages of Adjustment Layers is that they don't significantly increase the total image size; unlike actual duplicate layers of the background.
Another thing I find handy about editing with layers is that you aren't making permanent changes, until you combine the layers, so it is easy to delete the layer if you do something wrong. And, by turning the 'view' of the layers on and off you can easily see the before and after result of your adjustments.
Once you start getting into the many options of layers/masks you will find many 'strange' options for editing; as you already mentioned.
So, I would say layers/masks are certainly worth learning. And keep coming back to ask more questions. Many of us keep learning new things all the time.
24th April 2012, 08:57 PM
Agreed. I just couldn't live without them.
Originally Posted by Geoff F
25th April 2012, 08:46 AM
Just to add some further thoughts....
I guess where my confusion comes in is that the products seem to crossover. Adobe ACR, Photoshop, Lightroom, Elements. For instance ACR/Elements/Lightroom all have the ability to process RAW images. I think they all are billed as being non-destructive to images and in effect you can do similar things to a process like layers. e.g. you can change sharpening parameters without affecting other changes. So, its layers without the layers so to speak. Lightroom seems to offer a powerful method to process RAW and do a fair amount of editing which made me wonder why would I then need to do work with layers?
Cheers for now
25th April 2012, 09:01 AM
Originally Posted by oldgreygary
Photoshop comes with 3 main parts: Bridge for handling a folder full of images (think "image browser on steroids") - ACR for processing RAW images - and Photoshop "proper" for more advanced image editing (layers etc).
Lightroom - in terms of image handling & manipulation is - roughly - functionally equivalent to Bridge & ACR rolled into 1 program. Also, Lightroom and ACR use the same RAW processing engine - and neither of them "do layers". Photoshop Elements is simply a stripped down version of Photoshop (complete with a "mini-Bridge" and a "mini-ACR").
Despite somewhat of an overlap, it's reasonably accurate to say that "Photoshop starts where LR & ACR stop". Most folks probably don't appreciate just how powerful Photoshop is -- it's also true to say that many photographers don't need that kind of power (nor do they want to make the time/money investment in using it); for them Lightroom does all they need (and I might all that they could do pretty much the same things in ACR + Bridge, but the catch is you can't buy ACR + Bridge without buying Photoshop).
25th April 2012, 09:34 AM
This looks like the perfect place to ask about whether or not I'm using Layers correctly for a desired effect. I've recently started to experiment with combining two images, the original with adjustments in ACR and a B&W copy with added contrast, tone, etc. What I've been doing is opening the original image in Photoshop and double clicking on the layer to turn it into a regular layer and choosing color from the drop down menu in the Layers palette. Then I open my B&W copy as a separate file and change it to a regular layer. I use the move tool to combine the B&W layer with the original image and change the B&W layer to one of the choices from the drop down menu inside the Layers palette (depending on what effect seems to work best) and then combine down (I can't remember the exact name of the command in the Layers menu. I don't usually mess with the opacity on any of the layers so I'm not sure if I'm doing this right. I then sharpen with USM and really like the results I've been getting but I'd like to know if this is the best way to achieve this effect?
25th April 2012, 10:23 AM
I would absolutely hate to be without layers, in particular the range of tools when I use an Adjustment Layer. Since I run my cameras at low settings I need to adjust every file I shoot. I am using Paint Shop Pro, started with v.7AE and currently using X4.
If you have not got adjustment layers then I suggest you should get yourself an editing programme that has them. The adjustment layer is an additional layer above the image layer and everything you change with it is in the AL and can be toggled on and off to compare what you have done with what the original is. The original file is untouched . You can try different things on different ALs and toggle them on and off to compare adjustments. But selecting an area the adjustment will only apply to that area and if the photo needs different amounts of say darkening/lightening you can do this by adjusting at different levels. Whatever you set the adjustment layer to do can be completely removed by painting an area with black and if you stray over the edge of an area you can replace the effect by painting white over your error. There are 254 other shades between black and white for partial adjustments. Say a steam loco against bright sky, you could need a different treatment for the sky, the general body of the loco and the 'motion', wheels and gear. It can all be done from the one file probably without the need for bracketted exposures, so long as the sky is not burnt out.
The origin of the unsharp mask I think was back in the early days of photography when it was found that making a duplicate image and blurring it made the combined image sharper .... but when you have the unsharp mask tool I don't see much point in doing it the old way. PSP also has "High Pass Sharpening" which is less likely to produce white borders around dark areas bordering lighter areas ... the sign of over sharpening with 'unsharp mask'.
EDIT .. FrankMi gives a good description of variable density AL in 'Masking" thread.
Last edited by jcuknz; 25th April 2012 at 10:57 AM.
25th April 2012, 10:24 AM
Originally Posted by Sponge
Not quite sure what you're referring to here.
25th April 2012, 10:41 AM
Further ti Ian Darkslide
I make a copy of every file I take into my archive folder on a second hard drive and then move the camera file into an appropriate folder in my 'working HD'* ... this empties the camera card with SD cards, some junk is left with CF cards. This means that whatever I do with the working copy I have back-up in the archive folder. Files that get worked up are later saved to the archive folder and unused files deleted. Thouigh with the huge HD capacities we have these days I'm not sure if that is really needed. Practice dates from when I only had 80Gb HD :-) and bought a 120Gb HD as back-up. Now the back-up is a 500Gb and an external HD to my computer. Connected by USB.
* I am using Windows XP and hope it continues for ever since the process has been changed with later OS.
25th April 2012, 11:21 AM
Originally Posted by Colin Southern
I was referring to the part in the palette that starts out with "normal" and is grayed out until I double click on the layer and change it from background to a regular layer (not sure if that's the correct term). I'm honestly not sure what I'm doing at this point, I just experimented and found that by doing this (double clicking) I'm able to work on what was the background layer, mostly since I want to combine or merge the background layer, original image in color (which I change from "normal" to "color"), with the B&W copy (which I'll change from "normal" to "soft light" for example). Do I even need to change the background layer to a regular layer and change it from "normal" to "color" or am I just adding in an unnecessary step? I guess I'm just trying to figure out the best way to blend the 2 layers (is there a way to do this better as an adjustment layer as I see Geoff mentioned?) and whether leaving the opacity at 100% on both layers before merging down makes sense.
Thanks for any advice as I must admit I'm just experimenting but it's still not exactly clear what I'm doing! =)
25th April 2012, 11:42 AM
Originally Posted by Sponge
The "double clicking" bit just unlocks the base layer.
Easier way to do what you're doing is ...
- Open both images in Photoshop
- Click on File -> Scripts -> Load Files onto Stack -> (click "add open files")
- Then just change the blending mode of the top layer.
25th April 2012, 11:49 AM
Thank you, that's what I was trying to figure out, the quickest and best way to stack and merge the images and whether I needed to do anything to the background.
25th April 2012, 01:11 PM
I quite agree with the others, as layers (and the associated layer masks) are the two most important tools in my workflow. The only real impact of layers is storage (RAM and disk space), and when all is said and done, those are relatively inexpensive in the scheme of things.
I never edit the original, and in fact the first thing that I do in my workflow is to duplicate the bottom layer. This automatically gets Photoshop to save any file with the psd extension, so the original is safe. I try to use adjustment layers and layer masks wherever possible in my workflow, so if something goes awry, I don't have a lot of work to repeat. Cloning and other tools that let you set the tool to use "current and below" on an empty layer are great features.
I guess PS is stuck with some legacy code and tools where work has to be done on an image layer. I get a bit out of my comfort zone whenever I do something that actually changes an image layer and if I have to use a tool that does, I always do that on a duplicate layer. I have my undo history at a high setting so that these operations can be undone easily.
I like layer masks for very much the same reason; non-destructive work and it’s easy to tweak the mask with the paintbrush tool.
25th April 2012, 05:46 PM
Gary. Simple answer. For what you are trying to do with your photos (ref Project 52 etc) you do need layers.
They may initially cause a few headaches - but you won't regret these methods in the long term.
26th April 2012, 11:46 PM
Since I don't use PS any more and even when I did I never saw the point of this practice which appears to be PS Mantra.
Originally Posted by GrumpyDiver
The image on your screen which you are working on is already a copy of the file sitting back in the folder. You have not brought it out like a book from a bookshelf to be damgaed or destroyed. If you delete it the original remains back in the folder, and in my case a further duplicate copy is in my archive Hard Drive, but that is another story [above].
The key is to always use "save as" rather than "save". With 'Save As' you are prompted for a file name and/or suffix type if you want to save what you have done and alter the name slightly. Save replaces the original file and is rarely used in my 'workflow' from before I found photography on my computer.
For me "Z" is the easiest two finger as one addition to the file name followed by "Y" "X" for subsequent versions. The shift key and Z being next to each other on my keyboard :-)
Funny Story ...it has just dawned on me that my irritation with Photoshop stems from way back when I only had 6.5Mb hard drive [ maybe it was 6.5Gb :-) ] and PS kept on saving copies of my files when I was paranoid about not using my scarce memory space :-) It wasn't Photoshop but a simpler Adobe editor I had then.
27th April 2012, 12:43 AM
They're two different approaches, and cover different scopes. Duplicating the original layer keeps a copy of the pixels as they were before one starts pushing and pulling them; "save as" saves the file at intermediate steps ... they're different. Although Photoshop won't touch the original file, one may not want to go back to the original file (and possibly lose hours of work) - nor may one want to go back to a previously saved version for the same reason. By duplicating the base layer - and keeping edits on either adjustment layers or just one of the layers with image data on it, one has the option to recover from an "undesired edit" without throwing away everything that they've done since (perhaps a good example would be to fix some small area of accidental damage). It's also a good way to be able to compare original and edited versions at the click of a botton.
Originally Posted by jcuknz
The first HDD for a PC that I'm aware of was for the IBM PC XT - (and was 10mb) - before that we had to make do with the twin 5.25" 360k FDDs - yep - I was there too
27th April 2012, 02:07 AM
Gee that ages you, as I started with a 10MB IBM and the upgrade was 20MB. Programs ran off a 5.25 floppy disk as well1
27th April 2012, 06:24 AM
Ha - more than you think ... before the IBM PC I had an Osborne 1 ... and before that a Sinclair ZX80 (that I used to do assembly language programming on) ... and before THAT I was programming on HP41c calculators ... and before THAT the Texas Instruments TI58 (not TI58c, darn it).
Originally Posted by Ken MT
I was involved in the PC inductry since before there was a PC industry! (remember DOS 1.0 - CP/M - Topview - Windows 2.0 ... seen em all!) Not to mention WordStar - SuperCalc - dBase II - Lotus123 ... oh gawd, I AM old