Performance Differences Between LineTo() and Polyline()

• Microsoft Windows Software Development Kit (SDK) for Windows, versions 3.0 and 3.1

SUMMARY

When an application must draw a large number of lines, the Polyline() function will execute 1.1 to 1.5 times faster than the equivalent calls to the LineTo() function on video hardware that does not have a graphics coprocessor (such as EGA and VGA). On hardware with a graphics coprocessor (such as an 8514/A), Polyline() can execute up to 12 times faster than LineTo().

In general, Polyline() is faster than LineTo(). This is magnified by drawing a very large number of lines. With a small number of lines, the difference in duration of the calls is negligible compared to the time required to draw the lines.

MORE INFORMATION

LineTo() and Polyline() both use the same internal code to draw lines, but with slightly different setup procedures.

Polyline() calls GlobalAlloc() and copies all points into that memory; as a result, there is some overhead. A pointer to the memory is passed to an internal function that draws all the lines.

LineTo() essentially builds one point and calls the same internal function as Polyline(). LineTo() does not allocate additional memory.

Polyline() is much more efficient for a large number of lines, since an application makes a single call to Windows and setup is only performed once. A large number of calls to LineTo() introduces significant overhead to make each call and repeatedly execute the setup code in LineTo().