MJW
-
Posts
2,863 -
Joined
-
Last visited
-
Days Won
70
Posts posted by MJW
-
-
Though I'm not completely sure what you want to do, I made a plugin with the mellifluous name "Unselected Rectangle Keeper" that may perhaps help.
-
1
-
-
1 hour ago, bellaxx331 said:
I've saved the PDN successfully semi-transparent before I just forgot how to do it again for my other work.
You probably used actual transparency (the alpha channel) instead of trying to use a blend mode. There's absolutely no way to save transparency in a PNG file by using blend modes.
-
1
-
-
On 7/20/2024 at 6:03 AM, bellaxx331 said:
I'm using multiply blend mode to make my image semi-transparent as the bottom layer.
What exactly does that mean? The Multiply blend mode, or any other blend mode, doesn't make things transparent. Transparency is controlled by the alpha channel. If the bottom layer is opaque, flattening the image will result in an image without any transparency, no matter what the blend mode. In fact, changing the blend mode of the lowest layer has no effect if the layer is opaque, since there's nothing to blend to. (I'm pretty sure that's also true for non-opaque lowest layers, but I don't want to bother thinking it through.)
UPDATE: Perhaps I know what you mean, but I suspect you expect the layer blending to work differently than it does. If you have a light blue layer with Multiply mode, it's sort of like it's transparent -- like tinted glass. If you put a layer below it, it shows through, shaded light blue. If you have a layer above the Multiply layer, and merge the two layers, the resulting layer will have the Multiply blend mode. But other than that, the blending is the same as would occur if the lower layer had the Normal blend mode. You may think merging layers should result in a layer that blends with the layers below in a way that allows the image to be unchanged from what it was before merging them; but unfortunately, that's not possible. The only thing that's preserved is the lower layer's blending mode. (If you flatten an image, the single resulting layer's blend mode is always Normal.)
-
1
-
-
17 hours ago, tex editing noob said:
If I could find a way make the ALBD file use the greyscale channel from the ALP file I would be able to erase things to make them transparent. Or maybe there is a better way?
I'm not exactly sure what you mean. One file can't use another file for something. Files are just data. If the game program that uses the files takes both the (Red, Green, Blue, Inverse Metallic) file, and the grayscale file, and uses the grayscale file as alpha, you're all set: just make a grayscale file with the shades of gray set to the alpha values you want. If that's not what the program does, I don't see what you can do.
If the game program does take both files and uses them in the right way, you can use the alpha-handling plugins in BoltBait's plugin pack to help move around and edit the alpha channel. If you get to this point and are confused about the workflow to use, post a question about it.
(Forgive me if I'm misinterpreting what you're asking.)
UPDATE: I shouldn't say you're all set. You'll need to read and save files in the correct formats. Knowing little about the ALBD and ALP file formats, I don't know how easy or difficult that will be.
-
Maybe I'm missing something, but it seems like the game has chosen to co-opt the alpha channel for some other function besides transparency, so there's nothing left to use for transparency. I can't see how you can get around that.
-
Perhaps:
UPDATE: Don't know why the image shows up so weird. it doesn't look anything like that in the actual thread.
UPDATE 2: An example:
The example reminds me that the particular distortion of a trapezoid can be achieved by tilting the text with Rotate/Zoom. Perhaps preceded by stretching it vertically to compensate for the foreshortening from tilting. The perspective distortion it produces may or may not be what you want. There are also plugins, such as Distort This!, that let one do that type of thing.
-
Thanks, brittuk. I wish I could have provided a solution that would help solve your problem.
-
I'm almost certain there isn't, and with the current way PDN works, I doubt there will ever be. Adobe After Effects has text layers that maintain the text as text. PDN doesn't. As soon as text is committed (for instance, by clicking Finish), it's converted to pixels, and no longer has any separate identity as text.
Additionally, I don't believe plugins can create layers. (It's possible that functionality was added recently, but I don't think so.) If a plugin could create layers, it would be possible to write a plugin that could separate each object (connected non-transparent pixels surrounded by transparency) into a separate layer. That would sort of do what you want for text on a separate layer, but it would fail to work properly for letters that touched, and for the dots on i's and j's.
UPDATE: I realize about every other word in my comment is separate. I just couldn't come up with any synonyms.
-
1
-
-
On 5/22/2024 at 6:03 PM, BlastOfBN said:
However, if I recall correctly, an AI model would need to connect to the Internet, which is where AI models get their knowledge and information from, to ultimately render an image.
That's not how it works, at least not for the most widely used AI image generator, Stable Diffusion. All the knowledge and information is contained in files called checkpoints (aka, models), which are typically between about 2GB to 6GB in size. Once the checkpoints are downloaded, no Internet connection is needed to use them.
-
4 hours ago, lynxster4 said:
@CybrRyno You are much better off going with @Ego Eram Reputo advice. In Creative Text Pro you choose a color or texture by creating one (like your dark red color) and copying it to the clipboard. Then choose the Texture tab and in the drop-down menu select Apply Clipboard image. But you will always get a black-and-white bevel. There is no color wheel for choosing your own colors. I'm sure with some tweaking, you will get it right.
Ego Eram Reputo's method is certainly a good one, but there is a fairly easy way of using Creative Text Pro while still preserving the original color. It does require an additional plugin, Split Color and Brightness.
The method is:
- Duplicate the layer.
- Set the top layer's blend mode to Multiply.
- Apply Split Color and Brightness with its default settings (Display = Color, Blending Mode = Multiply) to the top layer.
- Run Creative Text Pro on the lower layer.
- Merge the two layers.
-
I think it's because all the shading on the object is provided by Creative Text Pro, which includes the color. The last tab allows for Color Adjustment, so that may do what you want. Other tricks include copying the original flat-colored version to another layer, and using blend modes (such as Multiply) to combine the colored version with the shaded version.
-
9 hours ago, _takechiya said:
I didn't realize I had to tick the alpha inversion box upon pasting.
I probably should have mentioned the various options in my comment. Other useful options besides Invert are the choice between using the map's alpha or gray, and the blending method. The Multiply and Minimum methods allow the alpha map to be applied to a layer that already has areas of transparency which shouldn't be replaced by the map's values.
-
1
-
-
There are two plugins that do that in BoltBait's plugin pack, Apply Mask and Paste Alpha. Apply Alpha Mask uses a map from another layer; Paste Alpha uses a map from the clipboard.
UPDATE: I corrected the name of the first plugin, which is Apply Mask, not Apply Alpha Mask, as I originally typed.
-
I suggest also trying BoltBait's Creative Text Pro. It's been more or less replaced by Text Fun Factory, but it has a valuable feature that seems to be absent from new version. In the first tab there's a checkbox called Apply to existing object instead of text which allows all the neat text shading effects to be applied to objects. You'd probably have to reshade both the text and non-text to get a consistent look, but I think you could get something close to the original, and perhaps even fancier..
-
2 hours ago, Ego Eram Reputo said:
With Difference, I got some mountain shadowing to show through at the cost of a darkening of the source image.
I don't think that's the image saramello (the OP) is concerned with. I think s/he wants to transform one of the small images in the head post to be like the other (or something like that). Tactilis just introduced that map image to show what the original two height map images represented. I doubt the second small image is a normal height map, so I'm not sure it can be done without a lot of effort.
UPDATE: Perhaps I'm just confused about what you're doing or demonstrating, Ego Eram Reputo. If you're using a blown-up version of the second small image to modify the map in the way the OP wants to do, I'd say that's very sensible, though it would seem to depend on the blurriness resulting from greatly magnifying the second height-map image. Actually, this whole thread leaves me rather confused.
-
2 hours ago, Tactilis said:
It's a map of Spain & Portugal:
Sure, but what are the images supposed to show about the relationship between the height mapping methods?
UPDATE: If, by chance, the first is the black and white version, and the second is the other version, the problem appears to go beyond remapping the colors. In the mountainous region of the second image, there are dark specks of color scattered upon a white background. That would seem to make no sense for a true height map, which should be relatively continuous. Maybe the non-white dots represent points where the height has been sampled; but in any case, I don't believe it's just a matter of remapping colors.
-
Seems to me you probably need to describe more precisely the transformation you want to make. Doesn't seem to me it's just changing one color scheme match another, whatever exactly that means.
UPDATE: To perhaps be clearer, it's relatively easy to guess how a black and white image represents height, but without knowing how the other format represents height, it's difficult to know how to transform one to the other.
Also, I'm not sure what those unlabeled images are supposed to show.
-
-
3 hours ago, Lemonade said:
While pasted part is still selected, with Move tool (M) grab one of the "knobs" to move around.
Hold Shift while moving to keep proportions of the element.Good idea.
-
You need to do that using a different layer. Layers provide a consistent way of handling that sort of thing, rather than having a myriad of ad-hoc solutions. There are likely plugins to flip the contents of rectangular selections.
-
EDITED: Sorry. This was an attempt to edit the above comment. Instead, I must have accidentally replied to it. My excuse is I was working on something else at the time, and unlike most people, I can't walk and chew gum at the same time.
-
1
-
-
6 hours ago, HenryH said:
1. When people talk about [Blurring an image], they usually mean Gaussian Blur?
2. What does [True Blur] do? -- (is the name appropriate?) -- is there another feature(plugin) with the name [True ....] ?
Gaussian blurs are more or less the standard blur, and most likely the one people have in mind when they talk about blurring, without any other qualification.
True Blur is Ed Havey's attempt to produce an improved version of Gaussian Blur. Keep in mind that the built-in Gaussian Blur has been improved since True Blur was written, so it may include many of True Blur's features. For instance, the current Gaussian Blur allows fractional radii, while the original did not.
-
2
-
-
3 hours ago, HenryH said:
Has this been implemented? --> (Unweighted - Averaging Blur)
Perhaps I'm mistaken, but I believe that's what the Bokeh Blur does. In a simpler form (without the gamma adjustments), it was previously caller Unfocus, or something like that.
-
1
-
-
Antialiasing for Magic Wand selections would be very desirable. Unfortunately, the Magic Mand only selects whole pixels, as Rick Brewster explained in a 2017 comment.
Crop to smallest enclosing rectangle?
in Paint.NET Discussion and Questions
Posted
Hmm. I just tried it in 5.0.13, and it worked for me. Are you sure you used it correctly? It's not the most intuitive plugin. You need to select the background with the Magic Wand, then run the plugin. It only erases the area outside the bounding box of foreground object, it doesn't crop to it, since at least at the time it was written, plugins couldn't crop (I think that's still the case).