Picture to People font engine can render fonts modifying them like you are used to: bold, italic and both.
As you can see, the font used for this example is not so simple as that ones from the first example. My font renderer can draw harder typefaces.
For now, I see no use to underline or strikethrough styles, so I will not implement them. I prefer implement more useful styles and transformations you can't find in other font engines.
Finally I can show some results about text drawing using only Picture to People software.
The text you see below was totally made with my own font rendering engine. For this, I have used two canonical fonts: "Arial" and "Times New Roman".
Realize the perfectly anti-aliased filling with no "blurred" appearance. I created a completely new robust algorithm for that with several important advantages. I tell more about that in future.
Despite its very professional quality, I will improve this library to very beyond a plain font engine.
I have just created a group to Picture to People project.
There you can:
- tell about your experience using the project;As you can see, it will work like a mix of forum, rss, ... So It can be worthy if you join. It's here: http://groups.google.com/group/picturetopeople.
For me, the last use of this new channel is the more important one. Now I have an "official" way to receive asks for new effects and new tools from Picture to People users around the world.
I'm facing well the hard challenge of have my own font engine. Some of my very low level libraries are been improved to achieve this goal.
Some of the main problems to someone trying to make this:
- TTF file format is weird and complex;
- font hinting is a very hard subject and trying auto-hint fonts is even worse;
- some fonts are very complex;
- there must be special care to draw fonts in small size; it can need proper algorithms;
- you should have a very complete, robust, and fast rasterizer behind you.
Some of my pieces of code are being stressed, but I'm glad about that. In the end I'm getting a general improvement in several low level libraries, what will help me in other developments.
In time: be patient. I will show you screenshots soon.
Now I'm working in an engine to render texts from scratch (from TTF fonts) with arbitrary vector based transformations.
Believe or not, it's a VERY hard task. The reason is not the transformations by themselves; the development of a complete font engine is a very complex goal.
In next posts, I will explain superficially why typography is a hard subject (despite the normal problems concerning professional Computer Graphics).
This library is one of the last low level tools I will need to create to use in a new and ambitious high level library. For now, let's keep it like a surprise.
Here I show you the last "Real Pencil" example for a while. Can you see like the delicate strokes follow the photo structure? It's one of the more important features of this renderer.
As a complete project, this tool uses libraries that even I took hundreds of hours to develop. Anyway, it was a worthy effort, because the users feedback is very positive: they say it really can generate "human style" drawings.
I'm already working hard in several other Picture to People sub-projects. If everything keeps going well, there are some surprises coming.
Subscribe in a reader
