EternalNY1

Thanks everyone, I appreciate the feedback. I'm going to grab those plugins (which I somehow apparently missed). Eli - That example is actually very close to what I need, if I can put those on layers and blend out the seams ... that's where I was going with "dissapate", not sure how else to explain it.

Yep, that's the closest I've gotten actually, but I can't get it without the seams even with a blur effect (either on that layer or on another layer with varying opacity). It's not a big deal and it's hard to explain, but unfortunately I have run into this situation more than once ... I basically want some sort of "dissipate" filter where it would slowly alpha the colors out over distance, yet not make it look overly stretched. Edit: I've done far better than this attached image, it's just showing the defaults with the Trail plugin. I'll probably just have to go with a close-in fade.

I did get there. I apologize for my poor description, but essentially I need to center an image like that but on a 4x horizontal canvas, while keeping the majority of the subject in focus. It looks better if the edges are stretched (into infinity and beyond?) instead of just a rapid fade-out. I tried applying stretch filters + transparency ... just to try to get hints of color from the edges to pull further out before they dissipate. Thank you for the assistance though ... that's the route I will go if I can't get it more "artistic" or whatever you would call it.

I've tried that approach, including additional transparent layers, but I can't either get rid of the seam or make it fade over enough distance so that it isn't obvious. In my example of a photo cropped at the wrists with arms outstretched, I would like to keep the subject in full focus but fade out the edges in a way that makes it seamless. I even tried blurs like the zoom blur, overlayed at various transparencies to try to "pull" the edge of the photo out into the expanded canvas, but it always ends up either too drastic or with a hard seam line where the blur meets the photo. I write code for a living, don't edit photos ... but I must be missing something obvious. Edit: I attached an example I just got from Google Images ... I need to make the cut on these sorts of things less obvious. At least in my current cases the background matches the expanded canvas background.

I am somewhat of a "power user" with Paint.net (10+ years, lots of plugins, lots of layers) ... so feel free to throw at me some complexity. If you have a photo that is cut off at the edges, what is the most efficient way to "fade out" this image into a larger, solid-color expanded canvas? For example, if you have a person in an image who is holding their arms out, but the photo crops out their hands at the wrist ... how can you make this look presentable if you have to double the canvas size horizontally onto a solid white background? I'm thinking of a smooth fade without artifacts. I've tried this with a variety of layering techniques and plugin effects (blurs, etc) and I can't get a seamless transition from photo to solid color. Thanks!

Yep, over the canvas. I've literally been using the software since 2004 when it came out of WSU, and sadly on top of that I'm a senior C# dev (what it's written in). It's odd, CTRL+A works to select, CTRL+D does nothing, but CTRL+Z to undo works fine. I will try to run some tools, it must be something on my side maybe hooking the key command (never good). I only noticed it in Paint.net.

Odd. As a software developer, here's the easiest repro steps ever: File -> New Image CTRL+A (expected result -> select all / actual result -> select all) pass CTRL + D to deselect (expected result -> deselect all / actual result -> nothing) fail No idea why.

Apologizes if this has been brought up before (I looked) ... Does CTRL+D no longer deselect the currently selected region? It's driving me nuts, maybe I just didn't notice until now. But If I select a region (lasso, area tool, whatever) and want to remove the selection, CTRL+D no longer does the trick. CTRL+I will stil invert it (thankfully), but it seems ESC is now required? Edit: ESC is not deselecting either.

Were you saving these files in the Paint.Net installation folder?? If you are in Windows 10, you can read this ... if not you are going to need a raw undelete tool from the drive like DiskDigger or something similar. Not a Paint.net issue.

I promise I won't derail this thread into something completely random, and this is actually on topic now! There is one glaring problem with this statement regarding the Chrome change. That quote above is only possible if the software allows the user to change the mappings. If the mouse software can allow me to say "X does Y", a browser can, but Google WONT. That is a different story. And yes I realize one is a driver and the other is a browser but, in the end, we're talking about what is best for the end-user, because they keep the lights on. As to what that has to do with this? Thanks again to Rick for making it a user option!

I think they are either rolling out as a A/B test phase, or they are only putting it on certain platforms? New people are joining in the fun every day though ... the message on their product forum has 384 posts (rants) and almost 12,500 views. https://productforums.google.com/forum/#!topic/chrome/D4VMOXTxPwg;context-place=forum/chrome I got it when I updated to 52.0.2743.82m.

That's honestly really interesting. As a long-time software developer (well before when Firefox was named Firebird!) I had no idea those keys were even mapped to that. It's honestly always been CTRL+TAB to move to the next for me, really quick with the left hand and three leftmost fingers. I also had no idea you could do CTRL+SHIFT+TAB ... I'm usually lazy and just keep CTRL+TABing until I get back to it. But I do use CTRL+SHIFT+T numerous times a day to re-open the one I just accidentally closed! Anyways, I don't work on this project so none of this matters anyway The problem is that this is sort of the "standard" across software projects where you have anything resembling tabs now. So much so that when you type "keys to move tabs" into Google, or anything that resembles that, you get: But I will say, again, Google's ridiculous decision to remove backspace to "go back" from Chrome, and on top of that mapping it to the abominable ALT+Left? That one definitely requires both hands, removing one from the mouse for no reason. So now I just right-click and then left-click "Back" on the context menu. Grrr....

Thank you! Going the correct route, opposite of what Google is doing with the "you can all forget about the backspace key". I am fully aware "more options" can often lead to a development nightmare, but in this case, I think it's required. I realize the benefits of the feature, but the upsides (for me, using it daily) is outweighed by far by the downsides.

I will admit, I have learned the benefits of this feature on certain occasions. Today I had to select a large area of a vertical image when it was already zoomed in. I lassoed across the bottom, then up, and let it scroll all the way to the top. So that was actually pretty awesome, and I see more now why this was requested. However, it definitely needs tweaking ... there are more situations where I'm at a loss as to what I'm looking at after doing something, or why scrollbars exist when everything is currently in view.

I realize all this, but opening an empty instance if Paint.Net, and pasting an image, I guarantee the user expects the canvas to fit the image. The reason is that the user never said they wanted a canvas. The X,Y dimensions of a default canvas don't matter in this case, because the user wants to paste a brand new image into the empty application. Instead, they get the image, plus a bunch of needless whitespace, because the application forced this on them due to a default canvas being stretched in proportion. I don't know ... it just feels very unnatural to me. Just suggesting something that would possibly help out new users, and make me not have to reach for CTRL+ALT+V (it's awkward, even after being a developer for over 20 years ... left pinky on CTRL, middle finger on ALT, index finger crossing over to V ... and I have to look to make sure). Or, I could paste like I do everywhere else with CTRL+V and have my image pasted and the canvas sized to fit. Like I said, I love this program, using since '04 ... I just donated money. Just some suggestions to help keep improving.

True, I do use that, because I have to. Wouldn't it be better when you paste an image with CTRL+V, to just have the dialog offer this option? What would you say 99.999% of users are going to do when they have an image on the clipboard and go into Paint.Net? Hit CTRL+V to paste it in there, or hit an unknown keyboard command (or hunt around in menus)? It just feels like a UX issue that could use attention.

When pasting an image into Paint.Net, you are only given two options: Expand the canvas Keep it the same size The problem I'm having is with "Expand Canvas". It will fit the image, but if you have a long vertical image, but the canvas is wider by default, you end up with white space. Can we add a simple third option here, "Fit To Image"? It at least keeps the image area selected, so I can just "Crop" ... but why the extra step? I included an example where of "Expand Canvas" where it requires the cropping, but I feel shouldn't if that option could be added. On paste (where we could use a third option, "Fit to Image"): The resulting, needless white-space:

Or, just hold down the control key and roll back with the mousewheel (which you most likely have a hand on already) and see the part that was obscured. Otherwise, this new "feature" lets you handle stuff under floating windows, which I can't see a use for. And if you move off the image surface itself, then it will scroll so you can't even see what you were working on at all! Here's a simple example of the issue: Let's say I had a section in the top right corner I needed to lasso-select. It used to be extremely intuitive, just start off-canvas, swing around, and stop: If you do this nice and quick like usual, and your cursor hits any side of the screen, here's what you get: This is useless. Why would I ever want it to scroll into this off-image area where the screen is ... well, its nothing. At least it was nice enough to show me a small bit of the area I had selected.

Any time I've ever needed to do something like this, I do what I do in every other program ... I zoom out with the selection kept. Generally with CTRL+mousewheel out. Same thing I do to set the zoom level in the web browser, the font size in Visual Studio, etc. That doesn't make any sense either. I get annoyed with the docked layers, colors, toolbox too ... but you know what I do? Zoom out of the image the same was mentioned above. It's extremely simple, keeps your selection, brings it back into view. And you can just do the opposite to get back in close ... move the mouse cursor over an area, CTRL+mousewheel in. This whole "overscroll" thing is a huge productivity killer so far and a huge annoyance. People who do precision graphical design work use other powerhouse programs like ... Photoshop? I use Paint.net for professional, precision graphic design also. That's not what they pay me for, but it's always a part of my job and I've gotten good at it with PDN and not having to have them budget for PS. It does what I need it to do. And I'm talking .PDN files with 20+ layers, complex overlays with different blend types, working with "curves", etc. So I'd say I'm a power-user of Paint.Net, but see no benefit to this feature. What I do "do", oddly enough, is write software, and C# in .Net is a big part of that (what Paint.Net is written in). This can be fixed. Make it an option setting, or if that's not comfortable, hide it from the users (advanced settings, executable flag, etc). This is not a monumental task given what has already been done to make this software, it just needs a little refactoring to "overscroll or don't overscroll". I don't want to go elsewhere, because Paint.net is awesome ... but this feature has got to go. Or make it configurable, please!

Can you explain why this been requested so much? That's not a joke, I honestly don't understand the purpose. I have been using this software almost every day since 2004. However, I am not a graphics designer, I am a software developer. So maybe I'm not doing what others are doing. Why would you want to ever have it scroll into an area you can't see? If using a brush, you'll overwrite something you didn't mean to. If using the lasso, you'll end up selecting something off-screen you didn't mean to. That's all I can see, so I'm missing something here.

You aren't running the latest Chrome. Believe it or not, the Chromium issue tracker shows this "feature" initially proposed in 2012, and the last comment is 4 hours ago. You can see more of the fun over here. Anyway, anything about what I was actually commenting about? I had this frustration again today. The lasso tool has become essentially unusable in many cases. There are often complex images where I need to select out a certain portion. I used to be able to start this process by making a rough outline of the area to be cut out, and then invert it and delete it. Let's say the portion of the image I wanted to remove happens to be in the upper-right side, but is too complex to use the magic wand on. I would lasso around it, starting at one corner and swinging around to the other. This is impossible now, as a quick wrap-around an image, if the lasso hits the edge of the screen, will now quickly scroll into oblivion, leaving me staring at a sea of gray. I know I can recenter with a keyboard shortcut but I am lost as to why this is now necessary. It actually shows more gray blank canvas area than it does image, and to make it worse, it won't even keep the selected area as the "stop point". So where my image was, I now have a mostly gray screen (a part of it may be over somewhere in the corner), and I can't even see the full section what I just selected (where I would have if it didn't scroll). I'm not even sure why this new feature exists? What problem was this trying to solve? It's a great piece of software, but as a developer, I'd revert this, push a build, add a configuration option, and send out another update.

I love this software, and having been using it since it was first released (As a software developer I have worked with .Net since beta in early 2000 ... this was one of the first great examples of what you can do). I apologize that I have not read this entire thread. I do see others mentioning this, but I am just going to add how glaringly noticeable (in a bad way) this odd new scrolling feature is. No feature has ever irritated me in a Paint.Net update as much as this one. It involves one or both of the following changes: New: The canvas may now be scrolled past the edge of the image Improved: The rate of auto-scrolling, which triggers when the mouse is at the canvas edge and a button is being held down, has been improved and is based on time instead of frames I don't know which one of these is responsible, but it now feels so completely unintuitive to maneuver around an image, because it doesn't "stop" at the edges like it used to. I am not sure exactly how I am noticing it so much. I think it may be that tools such as the Lasso tool now allow you to pull it off of the actual canvas and then it will start scrolling into the "outside the image" space if you continue off-screen, instead of stopping. I don't see a point to this, as we don't have layers with x/y coordinates that can be outside of the sublayer boundaries. Meaning a layer "above and next to" the lower layer. It's hard to explain, but I noticed it immediately, which led me to the patch notes, which led me to this comment. I'm not sure why you are able to zoom so far out on an image while it is still now showing scrollbars? I can zoom out of a large image to the point where it is 10x10 pixels wide on my screen, but the horizontal scrollbar and vertical scrollbar are showing padding? Where can I scroll to if the entire image is in view and taking up only 100 pixels on a 1920 x 1080 monitor? Odd how this is the second software question I had to ask today, after Google decided that the backspace key is no longer valid to go "back" in the Chrome browser! I find both of these very unnatural, but the Paint.Net one seems like a bug. With Chrome, they are just digging in their heels I guess because it causes people form data loss. At the very least, give us a configuration option (Google seems to giving a big "nope" to that one, but they are trying to force a major new standard for browsers). Anyway, this is odd to me:

Is it possible to crop a layer? I am working with a very large image with many layers ... I was hoping I could crop the layers down from the main canvas size to save memory and CPU. Thanks!

Clearly that makes no sense ... it isn't an increase in size, it's just larger? I am just confused on how an image can become larger than (original image size + new Paint.Net metadata), ever. If (original image + Paint.Net JPEG algorithm + metadata) results in a larger file than (original image + Paint.Net metadata), then what is the advantage of the algorithm? Why not just skip that step and stay with an image that 1) is higher quality (because its unaltered) and 2) has a smaller file size?

Paint.net "metadata" aside (is that part of the JPEG standard?)... I had a 100k JPEG go to 150k after cropping a large portion of it. Under no circumstances should a compression algorithm return an image that has more bytes after "compressing" it. At worst, it should add (original + metadata), meaning 0% compression. A 50k increase to a 100k file is ridiculous. I'm still waiting for an answer that addresses that.