MJW

A more confusing but sometimes useful alternative is my Object Pruner plugin, which conditionally removes objects according to their sizes compared to the largest object. The default setting leaves only the largest object.

So sorry! It's a wonderful job, lynxster4! The roundness of the cushions is depicted so well.

Congratulations are certainly in order to @lynxster4, @Pixey, and @Disk4mat! An exceptionally impressive collection of of entries. It was a very close contest, though Pixey's blue couch is my favorite. I especially like its sheen. (Obviously I spent too much time sitting on the sofa instead of rendering a sofa.)

If you could post and example or two of type of the images you want to recolor, I believe it would help. Answering that kind of question in the abstract is difficult. Also, have you tried Hue/Saturation or BoltBait's Hue/Saturation+? Those are what I often find produce the most natural looking color changes.

Perhaps if you linked to the tutorial you're referring to I would understand what you're doing and could provide some advice. As it is, I don't know even how you're making the box you refer to. Please provide more details. Before I can tell you what you're doing wrong, I need to understand what you're doing. (Though I speak only for myself, I suspect others have the same problem.)

Like BoltBait says. If you're not writing when you expect to write, the two most common causes are you've got a selection somewhere, or you've got the the wrong layer as the active layer.

If those are the areas to be filled, it's even simpler than using the Magic Wand. Just add a layer below, put into it whatever you want in those regions, then Merge the layers. The areas are transparent, after all.

I suggest you try BoltBait's Hue/Saturation+, which will let you conditionally modify color. Try setting the From Hue to 315 and the To Hue to 33. Then set the Saturation Adjustment to -100. (That's what worked for the small region you showed. You may need to play around with the settings to get the best result.) You may want to first loosely select the region around the hair to prevent other features from being recolored. EDIT: A useful trick is to put the recolored image in a layer above the the original image, then use a soft eraser to remove or reduce unwanted changes by letting the original image show through. Remember that reducing the Primary Color's alpha can be used to reduce the erasing strength, giving you more subtle control. EDIT 2: I'll also add that a variation of the two-layer idea can be useful for recoloring. For the current example, recolor the whole image to gray by making it black and while. Put it in a layer below the original image. Now use the eraser to erase the hair in the original layer, letting the gray hair show through. I find it much easier to control the results when erasing then when using selections.

I'd strongly advise against that approach. Why do it the wrong way, then try to patch it up, when the right way is about as easy? But to each his or her own.

Please don't do that, unless you want jaggy text.

Assuming (as you seem to say it is) that the text in on a transparent background, the easiest method is to use Red ochre's Object2colour plugin. The Hue/Saturation trick is even trickier for white text, since you must first add some color before it works. It can be done without plugins by first using Brightness/Contrast (or the Lightness adjustment in Hue/Saturation) to make the text gray, then running the Sepia adjustment to color it, followed by Hue/Saturation (perhaps twice to get full saturation). But why go to so much trouble when a plugin can do it in one easy step? I'll mention one other method: create a new layer, fill it with color, then use BoltBait's Paste Alpha plugin to transfer the text's alpha into the new layer. Why use such a roundabout approach? you may ask. The reason is, you can fill the new layer with more than just a solid color. For instance, you could have a gradient instead, or something even fancier. If you made the mistake of not using a transparent background, but the background is black, you can use one of the color-to-transparent plugins, such as my Color Clearer, to convert the background to transparent. That will also work with any fully-saturated-color background. However, if all you want to do is change white-on-black to color-on-black, just fill a new layer with color, put it under the text layer, and set the text layer's blend mode to Multiply. Then merge the layer down. EDIT: I forgot to mention you can use BoltBait's new Apply Mask plugin in place of Paste Alpha to make the transfer-text-alpha method even easier.

I agree with the original comment that it would be a useful option, since it corresponds to that standard mathematical convention for graphs. I sometimes wish that computer graphics hadn't chosen the y-down coordinate system. Of course, that's a ship that sailed long ago.

Perhaps I'm misunderstanding what you mean, but the destination buffer is the usual place for a plugin to write.

At least to me, it isn't clear what you mean by "filling" the shape (or what it means to "fit" the Union Jack into the shape). Perhaps you could explain more fully or show an example of a result that's similar to what you're after. I'm afraid your included image doesn't make it clearer to me.

Probably the best way is the one Red ochre already suggested : use Object2colour. However, if you're one of those people who for whatever reason want avoid plugins, an alternative might be to originally create the logo in fully-saturated red (or green or blue), then use Hue/Saturation to change the color to whatever you want. You can, of course, use selections if you want to change different letters to different colors.

Congratulations to @Pixey and @lynxster4. Few, but very well done entries. I really intended to enter, myself. I had what I considered to be a clever concept, but couldn't figure out how to do one vital aspect of it. Very frustrating! (I was trying for a rug. Maybe I should have aimed for a "carper" instead. Just kidding around, Pixey!)

If it's not too complicated (or private) to quickly explain, perhaps you could tell us what you're ultimately trying to achieve. I ask because, depending on what you want to do, there may be a better method than using the Magic Wand.

Another way is to use the History to Undo everything. Just click on the topmost entry in the History window.

Another method, if radial lines are what are wanted, is to use Red Ochre's Gears plugin. Run the plugin (in the Iterative Lines subfolder), and set Tooth height, Shaft hole radius, and Collar Rim width to zero. Set Number of spokes to 32.

Now that the Drop Shadow plugin works differently, I think it would be best the follow @Lance McKnight's advice to use Trail instead. Something like: Create the original shape as red Duplicate the layer Change the top layer to Black and White. Run Trail on the lower layer, with the angle set to straight down (270°). You'll probably want to uncheck Fade Out. Apply the Conditional Hue & Saturation color tricks to the lower layer. Adjust the Brightness/Contrast of the top layer. Merge the layers. Though it sounds a bit complicated, I think in practice it may actually be simpler than the original method. EDIT: Thanks to @TrevorOutlaw for reminding me about Trail's Fade Out option, which should be unchecked to produce the original result. Also, from what I can see, if the original object has a smooth (non-jaggy) profile, and the edge produced by Trail will likewise be smooth; at least if the High Quality option is checked. EDIT 2: For a smoother, feathered bottom edge, first run Trail with Fade Out unchecked, then run it again with Fade Out checked, and the Distance set to around 2 or 3.

PNG is a lossless format, so I don't think it makes sense to have adjustments for dithering or transparency. EDIT: I didn't consider that there are options for saving at lower than 8-bits-per-color-component resolution. So as Emily Litella, used to say, "Never mind!"

For reasons which I do not know (though I expect they are good ones), when a plugin is called, a rendering pass is often started, then aborted, and another pass is begun. Perhaps that's what's happening. (By a rendering pass, I mean an attempt to render the full image -- the thing that happens each time a control is changed.)

They can write to the same variable, but since the variable is shared between them, and not individual to each thread, they'll write on top of each other. The value will be whatever was written last. If it were a local variable, rather than a class-level variable, each thread would have its own copy. But, of course, the variable would be lost as soon as the particular Render call exited.

I'm pretty certain that's wrong. Your attempt at computing the selection size doesn't work with multi-processing, since multiple Render routine calls are writing to the same variable. What you're seeing is, I believe, a result of the selected canvas being processed as a single ROI in the first instance, and two ROIs in the second. If what you say were true, then running a plugin on a partially visible image wouldn't change the pixels in the non-visible region. You can easily confirm that the entire selected region -- visible or not -- is processed. (I hypothesize that when part of the image isn't visible, PDN processes the visible area first to give the user a quicker visible response.)

I'm not quite sure what that means (or how the method of computing the rectangle size affects it), but if you read the tutorials, you'll see how Render is called. In summary, the selected region is divided into disjoint rectangles, and Render is called for each of those rectangles.