Datavideo NVS-25 Quick Start Manual page 21

• P-frames: Predicted (P) frames are encoded from a "predicted" picture based
on the closest preceding I- or P-frame.
B-frames: Bi-directionally (B) predicted frames are encoded based on an
•
interpolation from I- and P-frames that come before and after them.
The Group of Pictures (GOP) structure specifies the order in which intra- and inter-
frames are arranged. All three frame types do not have to be used in a pattern. The
typical GOP structure is IBBPBBP. The I-frame is used to predict the first P-frame and
these two frames are also used to predict the first and the second B-frame. The second
P-frame is predicted using the first P-frame and they join to predict the third and fourth
B-frames.The relationship between I-frame, P-frame, and B-frame is illustrated
diagrammatically below.
I
The GOP structure is often referred by the GOP size, which is defined as a distance
between two full
IBBPBBPBBPBBIBBPB..., the GOP size is equal to 12 (length between two I frames).
The NVS-25 offers the user three types of GOP structure: IBBP, IPPP and IBP. Note:
Since B-frame is predicted based on I and P frames, a GOP cannot end with a B-frame
because if B-frame is used as a reference for making further prediction, a growing
propagation error will result.
GOP pattern with longer GOP length encodes video very efficiently. Shorter GOP
lengths usually work better with video that has quick movements, but they do not
compress the data rate as much. On the NVS-25, there are 19 GOP sizes available for
the user, ranging from 6 to 60 with an increment of 3 between each level.
B B P
images. For example,
H.264 Network Video Streaming Server
B B
in a sequence
NVS-25
P
with pattern
21