The following glossary of terms, may be a useful reference for articles on FontBlog
Aliasing: Sometimes called “the jaggies” or stair-stepping. It is an artifact of displaying round or diagonal features on a rectilinear grid like a display screen.
Anti-aliasing: Techniques intending to minimize the impact of aliasing, including gray-scale rendering and ClearType. Traditional anti-aliasing techniques use signal-processing methods of sampling and coverage. Many anti-aliasing techniques tradeoff Luminance Edge Contrast (blurriness) for spatial accuracy (position correctness), especially on low-resolution display grids. Anti-aliasing can create blurriness when the size of features you are displaying are similar or smaller than the size of the pixel grid on the display (e.g. if the stem-size in a glyph is equal-to or smaller than one-pixel).
Font Smoothing: A specific technique of anti-aliasing used by Microsoft Windows starting with Windows 95. It uses a combination of grid-fitting and oversampling to create a sharper screen image than traditional anti-aliasing techniques. Grid-fitting has the side-effect of increasing Luminance Edge contrast.
ClearType: A Microsoft specific technique that uses sub-pixel rendering to improve the quality of scan-conversion. The technique is based on breaking down a display pixel into its physical sub-components known as sub-pixels—in most cases these are red, green, and blue display elements. Because of the physical nature of sub-pixels, they may have a directional component that may improve quality in one axis (e.g. horizontally) while being neutral along the other axis (e.g. vertically). At small sizes and display resolutions, this asymmetry is not noticeable, but at larger sizes it becomes visible.
Symmetric Smoothing: A technique that improves the non-symmetric aspects of ClearType rendering, but at the cost of reducing luminance Edge Contrast. Luminance edge contrast increases as the PPem size increases. The art of enabling symmetric smoothing is to enable it at a PPEm size where there is a minimal loss of luminance edge contrast while maximizing shape integrity. DirectWrite and GDI both now support symmetric rendering.
Hinting: A generic term used in the font industry for improving font quality at low resolutions. It was originally used to describe a set of hints to the scan-conversion routine to create more consistent raster images.
Grid-fitting: A specific technique used for improving scan-conversion that involves “snapping” the vertical and horizontal components of the vector outline to align with the target size (PPEm) pixel-grid.
Instructing Fonts: A TrueType specific term that involves using instructions in the TrueType interpreter to implement hinting and/or grid-fitting.
PPEm: Pixels per Em. A resolution independent method of representing font heights, or Ems, in the raster image. For a given display resolution, font point size can be converted to pixels per em with the following formula:
PPEm (in pixels/em) = (Size in points/em) * ((device resolution in pixels/inch) / (72 points/inch)). For example, 12 points/em on a 300 pixel/inch printer would be 50 pixels/em.
Overscaling: Scaling the font outline data by a large amount above the target size, often 2x, 4x, or even 16x
Illustration of 4x Overscaled outline
Downsampling: Converts the overscaled pixels to the original target pixel size by applying levels of gray to the pixel based on the percent coverage.
Luminance Edge Contrast: Is the maximal light intensity change over a given distance for a feature on a pixel display. If a color changes on a pixel element from the brightest white to the darkest black on the adjacent pixel element, then the luminance edge contrast is high. If a pixel element changes from a gray to a minimally darker gray in the adjacent pixel element, then the luminance edge contrast is very low. If the transition is from maximal white to medium gray then to minimal black, the luminance edge contrast will depend on how physically close those transitions occur. If the display resolution is high, the luminance edge contrast will be higher than if the display resolution is low.
Display Resolution: Is the density of pixels on a display surface. The higher the density of pixel resolution with respect to the resolution of the human eye, the more accurately the object can be represented. Typically display resolution is measured in either pixels per inch (PPI) or dots per inch (DPI) (which are equivalent). A more accurate unit of measurement with respect to the human visual system is pixels per degree, which takes into effect the distance the viewer is from the display.
This post first appeared on MSDN Blogs | Get The Latest Information, Insights, Announcements, And News From Microsoft Experts And Developers In The MSDN Blogs., please read the originial post: here