Jonas Beckeman

Nice to see you're interested! So, here's an update: http://flowlabs.net/ Main new features: * Filters * Adjustment/effects layers * Blur/Sharpen/Smudge brushes * Less crash-prone, especially when switching between documents There's a change log on the site. The number of rainy days the next month is the governing factor for how much makes it into the August update. So, set your weather widgets to Uddevalla...

PaintLab CTP! Download @ http://www.openblackbox.net/PaintLab/default.aspx It's an early prototype, so I fully expect it to break in some situations - never mind that. But if you can't even get it to run (even though you have .NET 2.0 installed), please leave me a note! BTW - it automatically loads an example PSD each time you open a new document, it's supposed to at this stage (quicker debugging).

I think what you're asking is "how does the concept of layers work", right? Answer: You just keep each layer's bitmap separate in memory, and composite them together when you want to present it to the user. Some ways to composite are described in this article: http://www.pegtop.net/delphi/articles/b ... /intro.htm If you have trouble understanding the terminology on that site, you should probably pick up a book on basic raster graphics theory/programming.

A follow-up on my monologue here: I finally decided to write a little paint app, in order to test my signal processing engine OpenBlackBox. It's nothing more than a proof of concept, but I think it has a few interesting points: * All processing is done in a network of nodes - from filters, layers, adjustment layers / effects down to the brushes. * GPU codepath (although broken ATM) * Multi-space color editor (HSB/HSL/HWB/RGB/XYZ/L*a* and gradient editor * Multicolor brush designer * Preparing to incorporate Maxim Shemanarev's fantastic anti-grain geometry library for excellent vector graphics performance and quality. If anyone's interested in trying it out, I can make it available for download after the weekend (don't have access to my server right now).

Got a first version of my signal processing framework up and running: http://www.openblackbox.net Nothing much to see yet, but the foundation is solid. I've gathered quite a lot of GPU shaders that I'll add over time, and it shouldn't be too long before I have the functionality of most Photoshop filters in there. Except for those complex Artistic ones. And the Sketch category. And the Brush Strokes. Well, some of the filters anyway! It should be easy to use OBB from Paint.NET as a fast processing/compositing engine (or even easier, make a Filter plugin) - and that way both projects would benefit from shaders, filters, procedural texture generators that are written by either project's team or their communities. Anyone up for it on this side?

> Remove the user interface part and let the developer implement the user interface part. That's my line of thinking as well. I designed a painting tool in that sad old Macromedia Director, where the UI was just another component. I had plans to create a version of it that was a children's paint program with a game style interface. The animals you encounter and make friends with in the adventure part of the game appears as tools and helpers, fairys as particle effect "brushes" etc. I'll be porting it to C# one of these days, after I'm done with all the other interesting projects...

I've looked into it a bit, and unfortunately it's not a simple task. However, such an adapter would be beneficial to many .NET projects, including Microsoft's new Expression suite, so maybe MS could be talked into funding such an undertaking... Many of those Photoshop plugins aren't that complicated to mimic these days though - there's lots of open source code for hundreds of advanced image manipulation tasks. ImageJ plugins can be a gold mine, and since they're Java it's easy to translate to C#. I think it would be interesting to start a community which builds a library of good .NET image manipulation routines, with a common interface that make them usable in any .NET host. I'm working on something like that, but before I present it I want to implement GPU accelerated versions of a few of the filters so the project seems a little sexier...

Yup, GDI+ isn't very good. Take a look at AGG instead: http://antigrain.com/ http://www.creativedocs.net/devs/agg

Hello team, I'm working on a couple of open source projects which have some similarities with Paint.NET, in that they do C# image processing. http://www.codeproject.com/csharp/endogine.asp http://www.endogine.com/Tools/Paint/index.asp and I just started on a framework similar to http://www.spiralgraphics.biz/gen2tour/index.htm (but mine will be a lot more powerful, of course!) I suspect that you're spending time researching the some of the things I am, so I'm wondering if we might have anything to learn from each other. My current technical investigations include: Flash/Photoshop file parsers Photoshop 8bf filter adapter XAML support High-performance C++ convolution core (like Intel's IPP) HLSL versions of all heavy algorithms Are you planning any of this? If so, it would be sad if we were doing duplicate work. Let me know!