I am a noob in PS, so there’s one thing that is really bugging me. Let’s say that I want to have a semi-transparent overlay over a map, for example there are a few countries and I want to ,highlight" just one of them…
I select the brush, set the opacity at 50% and everything is ok, but…since I am drawing a lot of shapes on the maps, I can’t do it all at once in a single stroke and I have to eventually begin drawing with new strokes and to pass with the brush over a part of a area that I already passed with. The problem is that when I do that, the opacity is not 50% anymore, but double (in the area where the first and second stroke are ,collide"), so in this case I get 100% opacity…
I don’t know if this will work, but if it’s like GIMP, there is a “hardness” setting for the paintbrush. The default is 50%. If the hardness is less than 100% and you paint over a spot, it gets darker. If you set the hardness to 100%, overpainting makes it no darker. So setting the hardness to 100% (and perhaps a smaller opacity) might work.
If my understanding of the problem is correct, the solution is not to set the opacity of the paintbrush, but to leave it at 100% but do the painting in a new layer, then set the layer opacity to 50% or whatever is appropriate. This is somewhat what Onomatopoeia was suggesting but I don’t see why you’d need a separate layer for each shape. In fact with a single layer, you can tweak the transparency, hue, and brightness to whatever looks best, and it will all be uniform.
In general it is easier to use a new layer for each shape. This allows you the ability to tweak it as much as you like, without affecting anything else.
Yep. The first thing for a noob to learn is to USE the tools PS gives you, and quit trying to do it within the limitations of older, simpler tools. This is one of the things that layers are specifically for.
Create a map-shape as perfectly as needed on its own layer, and you can do ANYTHING to the underlying image with it, completely independently of the original image and any other shape-layers. No need to try and juggle all those differences in a single pass as one element.
Agreed, although it sounds like he’s trying to highlight certain shapes all in exactly the same way, hence a single layer is sufficient. But if not, or if in doubt, separate layers are the way to go, and they can always be merged or linked later.
A layer lets you do anything safely and easily without damaging or changing the original layer. Using a layer takes away a lot of the need to have delicate control. You can paint and erase and lasso and magic-wand over and over until you have your selection exactly right, and always undo or fix any mistakes with no consequences.
Perhaps there was some confusion. When I said “a single layer”, I thought it was clear I meant a single additional layer on top of the original background image! I’ve been working with Photoshop layers for years, they’re a wonderfully powerful feature.
Create a second layer, with all “100%” transparent colour (i.e. black)
Paint that layer with a different colour, say, white.
Make this layer 50% transparent.
Now the combined picture should be whitish (“greyed out”?) where you painted the layer, original colour intensity where you didn’t.
The layer should look like a “mask” of white with a hole where the emphasized country is.
Rather than a a brush, I’d probably use the magic wand to select the shape of the country, make a new layer and use paint bucket to fill in the shape on the new layer. Do that for every country you want. Then just reduce opacity on new layer. You can also use layer tools to change color, and a stroke or drop shadow to help them stand out.
Have you considered using Illustrator and vector maps? They are easily found online and can be simpler to work with (click on a country/county/state/whatever and change its color individually).
Of course, that doesn’t work so well when your original layer is something you want to keep. But even then, you could use Photoshop and the Pen tool to trace a vector “shape” on top of the raster background, and adjust that shape individually.
But otherwise, basically what wolfpup said: Make ONE additional layer, paint the whole thing with a 100% opacity brush, and then adjust the layer opacity.
Would there be a benefit to using a vector overlay for a raster-based image? The learning curve for Illustrator is a lot steeper than PhotoShop. if the map already has borders, magic wand will find them.
If you’re planning on doing any graphics in the future, I highly recommend learning Illustrator. And I wouldn’t say it’s a lot steeper; depends on how deep you go with Photoshop.
Vectors aren’t really that hard to learn, and make a lot of sense for things that are naturally shapes (like areas). The main advantage in this case is that it’d let you work with geographical entities more naturally, being able to select each state/country and manipulate it at well without affecting any of the other ones. Think of a PowerPoint with a rectangle behind some text, serving as a background. You can just select that rectangle and change its color or size at will. Now imagine a screenshot of that same setup, which is now a raster. If you wanted to change the rectangle, you’d have to somehow select it independently of the text, enlarge it or flood fill it, and then hope you didn’t miss any of the insides of "o"s or "e"s, etc.
If you can find the original area as a pre-made vector map (you should very easily be able to, if it’s a modern area you’re interested in), it is a lot easier to select that area directly than to start with a raster and try to use the magic wand to manually define your own areas.
A similar workflow is technically possible using raster layers, one for each area. But if you just wand it and don’t make new layers, after you deselect that region you have to wand it again to select it to manipulate it. If you wand two adjacent regions and do something too similar to them, the wand might have trouble distinguishing them apart next time. You can save each wanded area as a saved selection or a new layer to get around that, but at that point the vectors are probably less work anyway.
Also, the wand doesn’t work too well with JPEG compression unless you nail the tolerance exactly right, and even then noise in the image can cause weird effects. Tracing it by hand, whether raster (lasso) or vector (making a shape) would probably look a bit nicer.
So if you can find a vector of the original map, it’s a lot easier to work with. If you have to define the shapes yourself, then it’s more or less the same time investment either way (learning curve aside).
And if you ever want to really get into more serious mapping, vectors are important because:
Free geographical definitions (in the form of shape files) are vector: that means things like geographical boundaries, roads, bodies of water, etc. (There will also be raster files for things like aerial photography and satellite data)
Once you have the basic geography of your map done, working with a vector export in Illustrator and/or InDesign will be a lot cleaner than trying to transform and resample a raster in Photoshop. You can style all the highways one way, all the rivers another way, or change the fills of geographical areas easily, or even just blow the map up for printing on large-format printer. Doing any of those things with a raster output is much harder and won’t look as good. Vectors allow cleanly defined shapes/lines/points independent of display/print resolution and separate from other shapes/lines/points on any layer.