stb/tests/oversample/README.md

95 lines
4.5 KiB
Markdown
Raw Normal View History

2014-12-07 10:17:44 +03:00
# Font character oversampling for rendering from atlas textures
TL,DR: Run oversample.exe on a windows machine to see the
2014-12-07 10:22:14 +03:00
benefits of oversampling. It will try to use arial.ttf from the
Windows font directory unless you type the name of a .ttf file as
a command-line argument.
2014-12-07 10:17:44 +03:00
2014-12-07 10:28:04 +03:00
## Benefits of oversampling
2014-12-07 10:28:28 +03:00
Oversampling is a mechanism for improving subpixel rendering of characters.
2014-12-07 10:28:04 +03:00
Improving subpixel has a few benefits:
2014-12-07 10:31:12 +03:00
* With horizontal-oversampling, text can remain sharper while still being sub-pixel positioned for better kerning
2014-12-07 10:28:04 +03:00
* Horizontally-oversampled text significantly reduces aliasing when text animates horizontally
* Vertically-oversampled text significantly reduces aliasing when text animates vertically
* Text oversampled in both directions significantly reduces aliasing when text rotates
## What text oversampling is
2014-12-07 10:17:44 +03:00
A common strategy for rendering text is to cache character bitmaps
and reuse them. For hinted characters, every instance of a given
character is always identical, so this works fine. However, stb_truetype
doesn't do hinting.
For anti-aliased characters, you can actually position the characters
with subpixel precision, and get different bitmaps based on that positioning
if you re-render the vector data.
However, if you simply cache a single version of the bitmap and
draw it at different subpixel positions with a GPU, you will get
either the exact same result (if you use point-sampling on the
texture) or linear filtering. Linear filtering will cause a sub-pixel
2014-12-07 10:19:25 +03:00
positioned bitmap to blur further, causing a visible de-sharpening
2014-12-07 10:17:44 +03:00
of the character. (And, since the character wasn't hinted, it was
2014-12-07 10:19:25 +03:00
already blurrier than a hinted one would be, and now it gets even
2014-12-07 10:17:44 +03:00
more blurry.)
You can avoid this by caching multiple variants of a character which
were rendered independently from the vector data. For example, you
might cache 3 versions of a char, at 0, 1/3, and 2/3rds of a pixel
horizontal offset, and always require characters to fall on integer
positions vertically.
When creating a texture atlas for use on GPUs, which support bilinear
2014-12-07 10:19:25 +03:00
filtering, there is a better approach than caching several independent
2014-12-07 10:17:44 +03:00
positions, which is to allow lerping between the versions to allow
finer subpixel positioning. You can achieve these by interleaving
each of the cached bitmaps, but this turns out to be mathematically
equivalent to a simpler operation: oversampling and prefiltering the
characters.
So, setting oversampling of 2x2 in stb_truetype is equivalent to caching
each character in 4 different variations, 1 for each subpixel position
in a 2x2 set.
2014-12-07 10:21:11 +03:00
2014-12-07 10:28:04 +03:00
An advantage of this formulation is that no changes are required to
2014-12-07 10:21:11 +03:00
the rendering code; the exact same quad-rendering code works, it just
uses different texture coordinates. (Note this does potentially increase
texture bandwidth for text rendering since we end up minifying the texture
without using mipmapping, but you probably are not going to be fill-bound
by your text rendering.)
2014-12-07 10:28:04 +03:00
2014-12-07 10:46:40 +03:00
## What about gamma?
Gamma-correction for fonts just doesn't work. This doesn't seem to make
much sense -- it's physically correct, it simulates what we'd see if you
shrunk a font down really far, right?
But you can play with it in the oversample.exe app. If you turn it on,
white-on-black fonts become too thick (i.e. they become too bright), and
2014-12-08 22:42:47 +03:00
black-on-white fonts become too thin (i.e. they are insufficiently dark). There is
2014-12-07 10:46:40 +03:00
no way to adjust the font's inherent thickness (i.e. by switching to
bold) to fix this for both; making the font thicker will make white
text worse, and making the font thinner will make black text worse.
2014-12-07 10:49:29 +03:00
Obviously you could use different fonts for light and dark cases, but
2014-12-08 22:42:47 +03:00
this doesn't seem like a very good way for fonts to work.
2014-12-07 10:46:40 +03:00
Multiple people who have experimented with this independently (me,
Fabian Giesen,and Maxim Shemanarev of Anti-Grain Geometry) have all
2014-12-07 10:48:10 +03:00
concluded that correct gamma-correction does not produce the best
2014-12-07 10:50:34 +03:00
results for fonts. Font rendering just generally looks better without
gamma correction (or possibly with some arbitrary power stuck in
2014-12-07 10:46:40 +03:00
there, but it's not really correcting for gamma at that point). Maybe
this is in part a product of how we're used to fonts being on screens
which has changed how we expect them to look (e.g. perhaps hinting
oversharpens them and prevents the real-world thinning you'd see in
a black-on-white text).
2014-12-07 10:50:34 +03:00
(AGG link on text rendering, including mention of gamma:
http://www.antigrain.com/research/font_rasterization/ )
2014-12-07 10:46:40 +03:00
Nevertheless, even if you turn on gamma-correction, you will find that
oversampling still helps in many cases for small fonts.