MJW

I'll try to officially release it soon. Unfortunately, I haven't seen co-developer @Red ochre around the PDN forum for quite a long time, but I'm pretty sure he would have no objection.

@VolvobmT24, CodeLab isn't needed to make the plugin work. You can just download and unzip the DLL. I included the code, so if someone wants to, they can paste it into CodeLab and build the plugin, but that's not what most people would do. I think you need to use a newer version of Paint.Net.

From what I can find with a web search, PSA files aren't image files. They seem to be catalogs of images, used by Photoshop Album and Photoshop Elements to organize collections of images into albums. If that's the case, I doubt there's any PDN support for them. I suppose a FileType plugin might be able to open a menu, allowing the user to select which image to open from the files in the catalogue, but I'm not sure anyone would spend the time to write it. EDIT: Even if such a plugin could be written, I think it might run into problems with the filename association. (I'm speaking about a subject I know little about, since I've never written a FileType plugin.)

alex_bb, you aren't the only one who doesn't like this behavior. Rick Brewster, the developer of PDN, is considering adding an option to disable the automatic layer-switching.

@welshblue, to use the Texture View Skewer: Before running the Texture Shader on a height map: Duplicate the height map layer (to save it for later use). Run the Texture Shader on the top layer. Copy the unshaded height map to the clipboard. Run the Texture View Skewer on the shaded image, making sure the Use Clipboard for Texture Map option is selected. For computational reasons, the Texture View Skewer always moves the viewpoint to the left. If, as is commonly the case, you want the viewpoint to be downward, then before copying the height map to the clipboard, rotate the entire image clockwise, and after running the Texture View Skewer, rotate it back counterclockwise. If you have multiple shaded layers, copy the merged shaded image into the clipboard, then paste it into a new layer or image. Disable the shaded layers, and enable the height map layers. Copy the merged height-map image into the clipboard. Run the Texture View Skewer on the merged shaded image. (The point being, you need to run the View Skewer on the merged shaded image, using the merged height maps.) I always run Transparent to Transparent Black before running any plugin that might be affected by the height of transparent pixels. The biggest problem with the Texture View Skewer is that it doesn't foreshorten or perspectively converge the skewed image, so that has to be done as a post-processing step. I'm trying to come up with a method to combine that into the plugin. (The current the Texture View Skewer is something of a hack, but I've used it a lot.)

Thank you very much, welshblue! If I can find time, I'll think about writing a tut for it. It's actually mostly just various applications of the Texture Shader and Texture Merger, but I think it might be useful to explain the procedures step by step, since the Texture Merger is such a complex and confusing plugin. The feature I think turned out best is the ferrule (the metal part that holds the eraser). It actually wasn't too complicated. I made a rectangle of the correct size, then added the little notches in the profile to make the ridges. I duplicated the layer and flipped in both vertically, then horizontally, to make the notches on each side and both ends: I then ran the Texture Object Rounder. I think I ran a few iterations of the Texture Smoother to smooth it out, then used rectangular selections to trim the very ends a tiny bit (followed by Transparent to Transparent Black). I generated a vertical grid, blurred it slightly, and mapped it horizontally in onto the center of the rounded ferrule. I made a texture map consisting of three white-on-black blurred dots, then used the Texture Merger to subtract them to produce the indentations on each end. The shading was done using the Texture Shader in Reflection Map mode. The off-angle view was produced using my beta plugin, Texture View Skewer. That's a plugin I would love to make into a released plugin, but can't quite figure out how to make it work the way I'd like it to work. (I actually made the ferrule slightly too long, and resized in the final shaded image using Move Selected Pixels. Very kludgy! I intend to fix it in original height maps, but was afraid I might not get around to it before the submission deadline.)

I'm not sure if this is pertinent, but if you want to select the opaque elliptical region, first select the transparent region with the Magic Wand, then invert the selection. (BTW, I like that image.)

If you could find and link to some examples that are similar to what you want to do, it would help a lot. I'm not sure what you mean by "I'm trying to make a shape look like gold." That could cover quite a few different techniques. If you haven't read through it already, you might take a look at the Metallize plugin thread. For Metallize, you need to start with an image where the color brightness varies (preferably smoothly) across across the object. Metallize modifies the color so the object looks reflective, and therefore metallic. In the plugin thread, dpy's leaf example shows the results.

I'm so sorry! I had a feeling I forgot to do something. It's there now.

I've released Version 2.3.0. I added a new mapping method called Gradient Map (Equirectangular). This mapping mode is similar to the reflection mapping mode, but instead of using the reflection vector, it uses the gradient vector. It probably has fewer uses then reflection mapping, but provides an extra option. When used with blurry clipboard images, it can simulate diffuse light sources. EDIT: I probably should mention that "gradient" refers to the slope of the height-map surface, not to a color gradient. I considered calling it Normal Map, but that also has two possible meanings.

Yes, I've used TechnoRobbo's random-fill plugin that does that. For the small-sized sprites that approach works very well.

I believe what Eli means is that the photos would all be in the same layer, arranged in an array. That would certainly be possible, but as Eli suggests, somewhat inconvenient and impractical.

It's definitely true that plugins can only modify or even "see" the active layer. There's no way for plugins to access the data on any other layer. Plugins can also use the clipboard, but that wouldn't help for a median plugin, which needs at least three simultaneous images. A plugin could be written to do the median from file images (as the example shown appears to do). I don't think a file-based version seems useful enough to be worth the considerable effort it would take to write, but maybe someone else might.

I think both the radio and the tennis racket theme fall within the original intent. The difficulty with the tennis racket theme is all in the details. Among other things, it's really hard to produce the rounded cross-section of the frame, and the way the string holes are evenly arranged around the oval head adds considerable complexity. I gave it a little thought, and couldn't come up with a good approach. (As I mentioned before, Pixey's entry was beyond impressive.) I expect the radio theme will attract quite a few entries. Radios are more straight-forwardly geometric. I have some ideas already, and hope to submit an entry. The only question I might have about the theme in regard to the original intent of OOTF is that it allows such a wide variety of entries. I'm not sure that's really a criticism, though, since I'm not sure to what degree the competition should be about producing the same thing. There's an advantage to limited themes, which allow more direct comparisons -- so it's not comparing apples to oranges (which is why "apple" is a better theme than "fruit.") However, I think it would be challenging to come up with enough limited-variation themes. It would pretty much confine the themes to natural objects, since most man-made objects come in many different forms.

Oh, I see what the confusion was! I should have read the original comment more carefully (the story of my life). Two weeks would certainly be too long to keep the poll open.

I think the question ("Is there any special reason the poll for 'Valentine’s Day' is open for two weeks, rather than the normal one week?") assumes an incorrect premise that the normal SOTW entry period is one week. As Pixey mentioned, around the middle of last year it was changed to two weeks, in order to attract more entries. All the contests since then have lasted two weeks. Even before it was officially changed, the entry period period was often -- more often than not, I believe -- extended when there were too few entries. EDIT: Obviously I need a course in reading comprehension. I skipped right over the word "poll."

I think it was a good choice, just really difficult. Pixey's entry is amazingly good. I wish I could enter the competitions, and I'm sure I will in the future, but I've been distracted recently by non-PDN matters, and just have't been able to find the time.

I know essentially nothing about lightmap textures , but it seems to me that a program that creates lightmap textures has to be aware of the relationship between the objects' UV coordinates and vertex coordinates. That doesn't seem easy to do with PDN, where images are just collections of pixels. I think you'd have more luck using a 3D shading program. (If you phrase your question as more "Here's what I have; here's what I want to do with it," you're likely to get a more informative response.)

A bit more complex, but I think you could create the path in PreRender. That would require you dispose of it in an OnDispose routine instead of relying on a "using" block. EDIT: I haven't tried this yet, so I don't know if there might be problems accessing path from multiple threads.

I think you might want to move the following code to PreRender: try { myFont = new FontFamily(font.Name); } catch { myFont = new FontFamily("Arial"); } Naturally, myFont would need to be made global.

Also, in the Visual Studio project properties, as the last item in the list of frameworks, there's an entry "Install other frameworks..." As I recall, it just opens in your browser about the same thing that toe_head2001 linked to. I also recall it was kind of confusing, because there are two things I could do, and I wasn't sure which was the right one to install the framework.

@dipstick, I really like the reflecting hearts, but you might want to try the Texture Smoother. It does a pretty good job of smoothing out the waterlines caused by the low resolution of gray-scale height maps.

I agree that using selections can result in jaggy edges. Often there's an alternative method that works better. What are you trying to do by selecting the transparent background? Do you have some specific task in mind? If you want to erase the corresponding area in another layer, try BoltBait's Paste Alpha plugin.

Those are some very impressive umbrellas, @Pixey, @MadJik, and @lynxster4! I wish I could have entered, but I didn't have enough time to even come close to the quality of winning entries. Thanks to @toe_head2001 for hosting.