Sirisian

Do people really get confused over that? I think people would just ignore it until they need it like they do in fireworks and other programs. Does Rick support power-user features?

Thanks for replying. Thought this was just going to drop off the page. Yes. His argument was to redo the save as dialog which is ridiculous. I'm saying that that it should use the tried and tested separate export feature found in other image editors. The only comment of Rick's that scares me is: But yeah it seems like Rick has the best intentions for the UI in mind, so it's probably just a matter of time. The project is closed source, so I can't refute that, but if he's abstracted the code then he already has all the code to save all the formats and to flatten layers. Actually he has the code to everything already that would be required by the export feature. If anything it shouldn't be more than "duplicating" what the save as feature does with the intention of creating a tag of the current working file as another file and continuing to work on the main file. How much work that would involve is debatable, but I'm sure Rick knows.

I saw this post that talked about export since I was looking for a way to quickly save as png without flattening. I work using the pdn format but I save to png 32 bit. It would be nice to have an Export feature in the file menu separate from the save options. Basically just hit export and it brings up the normal dialogs for saving but just saves the image leaving me with the pdn as I continue working. I'm sure you've realized that not everyone wants to work in png and destroy their layers just because they need to upload for the web really quick. At the moment (as you know) when working on a pdn you have to do save as, select png from the file types and hit save, hit okay on the dialog, click flatten, now you have a png file with a new name if you changed it, then hit ctrl+z to get your layers back, press ctrl+s to save and since it has layers it pops up to save as pdn, rename back to old name and save. It's a lot of steps. So an alternative would be click Export and if you've exported before it already has the name and file type you want filled in ready to overwrite your last export then click save have that dialog pop up for the bit type and click okay and you're done. It would be even simpler if you merged the save dialog into the image setting dialog then you could hit Export and hit an export button and be done in two clicks. A lot more work though at a minor usability gain.

I like to do pixel art for fun, and I was wondering if there could be a pixel mode added or a toggle. It would only be a few changes hopefully. I noticed an older post on this same idea. When you select the rectangle tool and zoom in. Click and it creates a 2x2 rectangle but if you look at the bottom right it says 1x1. It's probably measuring from wherever the mouse cursor is at. Would it be possible to make it say the pixel dimensions instead so a 20x20 rectangle says 20x20 pixels and not 19x19? Same thing for the other shapes. Also could you snap the mouse to the center of the pixels when the user clicks to create a line? Without this feature it's awkward for drawing pixel perfect reproducible lines unless you manually click the center of the pixel when starting and stopping drawing the line.

Would you care to explain why? (I'm actually curious how interface designers think). Do you believe that it would hurt the usability of the program, or think it's not worth implementing? Have you gotten negative feedback about a similar feature?

Actually if implemented correctly (I happen to be a programmer) would cost very little. Also my friend was a bit confused with that "every operation" statement. Say you have selected a piece of text and want to edit it. You simply flatten it and it turns into an image layer. If you choose a brush or a pencil tool and draw while you have an image layer selected it works on that layer. Like I said if implemented correctly the cost is trivial. (By the way all of these ideas are taken from Adobe Fireworks). What you are describing is a vector drawing program. Perhaps, you need Inkscape instead of Paint.NET. Not really. What I described is just another way to handle the current layer design. Currently everything you draw whether it's text/pencil/brushes happens on the same layer. I'm suggesting that lines and curves remain editable existing in their own layer until flattened into images and that "image" layers are created when one tries to draw while a non-image layer is selected. This system is just a slight variation of the current design as you can see since after you draw a line it just loses it's editable abilities and gets flattened into the image layer. By keeping it editable as a separate layer it would make the program much more flexible by not forcing the user to make permanent changes until they're ready.

Already suggested: viewtopic.php?f=12&t=21749 Ah cool a ribbon interface sounds like it'll work. Not as specialized as just docking the windows cleanly into the program, but I can see how a ribbon would work. (I already resized the window and put the windows outside the program, it's actually why I pointed out the docking thing since it's kind of awkward since I prefer having programs fullscreen when working). ...but you didn't architect Paint.NET. No, that is correct. I'm just using the program. I was referring to the previous posters comment which said that it would "decrease performance". I wasn't sure where he got this information, unless the lead programmer for paint.net has mentioned it in the past? Personally I find it hard to manage layers in Fireworks for that reason. Every time you use, say, the brush tool, it creates a new layer and if you just want to use it to make a quick touch you now have some layers to flatten or manage. But it doesn't. Using an image layer, for instance a picture, and applying the brush to it works on the current image layer. This is the intuitive operation and doesn't create a new layer. I explained this in my previous post to clarify.(If you have fireworks you can test this by drawing with the pencil tool then with the same layer selected drawing with the brush. They are on the same "image" layer and no extra layers are created).

Actually if implemented correctly (I happen to be a programmer) would cost very little. Also my friend was a bit confused with that "every operation" statement. Say you have selected a piece of text and want to edit it. You simply flatten it and it turns into an image layer. If you choose a brush or a pencil tool and draw while you have an image layer selected it works on that layer. Like I said if implemented correctly the cost is trivial. (By the way all of these ideas are taken from Adobe Fireworks).

Not sure if this has been mentioned but it would make things much easier if every operation created a new layer. So placing text would make a new text layer, a line a new line layer, etc. Also keep all of the objects as shapes in their own layer until they are flatted would make things easier. Separate the line from the curve so they are distinct objects. This would allow someone to place a line then move the endpoints around without curving the object or having to start from scratch. Also until text is "flattened" to an image layer keep it as an object in its own layer. This would allow someone to go back and move around the text later or double click and edit it. The pencil tool should add pixels to the current selected layer if it's an image layer. If it's not then it should create a new layer. As mentioned a user could select and flatten layers into image layers. In order to fully use the layers one should be able shift click them or shift+up/down to multi-select them. Then allow the user to flatten the layers or drag single or groups up and down in depth. Also windows. I noticed the history/layers/etc can snap to the edges. They are in the way. It would be nice to dock them into the program. Place the tools on the left side, and the primary/secondary color pickers on the left side blow the tools as two color pickers. The layers and history can go on the right side. These changes would greatly increase the usability of the program and make it on par with current image editors.