The Cut-off Time and The Number of Passes at The Split Point
Due to various special cases mentioned above, when calculating a participant result, it is necessary to filter the raw data of the participant obtained by the device. This corresponds to steps 5 and 6 in the participant's calculation process. Generally, there are two filtering methods: one based on the cut-off time and the other based on the number of passes.
The cut-off time is an attribute of the split point, which refers to the fact that the split point only accepts participant data within a specified time interval. Data outside this time interval is excluded from the split point.
The number of passes is also an attribute of the split point, which refers to the participant's data when passing the timing point corresponding to the split for the nth time.
1. Data Filtering Based on Cut-off Times
Each split point can have cut-off time set. Data that falls outside these times is definitely unnecessary. Data within the cut-off time may not necessarily be required. There may be overlaps between the cut-off times of different split points. A split point's cut-off time may have only an upper limit or a lower limit set, or neither limit may be set. As shown in Figure 5.1.
Figure 5.1: Cut-off time and cut-off times overlap
If a split point does not have cut-off time set, the cut-off time is determined as follows:
Lower limit: The later of the closest preceding split point's lower limit for cut-off time and the earliest gun start time for the contest. As shown in Figure 5.2.
Figure 5.2: Lower limit setting for cut-off time
Upper limit: The upper limit of the closest subsequent split point's cut-off time. If no subsequent split points have an upper limit for cut-off time, the upper limit will be a distant future time. As shown in Figure 5.3.
Figure 5.3: Upper limit setting for cut-off time
Using this method, even if a split point does not have cut-off time set, a cut-off time can still be obtained.
2. Data Filtering Based on Pass Count
Pass count and cut-off time are independent concepts. The calculation of pass count is determined by the order of multiple split points corresponding to the same timing point, and does not require manual specification. It is unrelated to whether the split points have cut-off times.
The concept of pass count exists only when multiple split points share the same timing point.
When multiple split points share the same timing point, the pass count for each split point is sequentially 1, 2, 3, and so on.
A defect in using pass count to determine data and split affiliation is that a data loss anomaly can lead to incorrect data for subsequent split points, expanding the range and impact of the error.
Assuming there are three split points, A, B, and C, all sharing a common timing point, then: if data from A is lost, it will cause A to incorrectly use data from B, B to incorrectly use data from A, and C to have no data, which means one error turns into three errors. If data from B is lost, it will cause B to incorrectly use data from C, and C to have no data, turning one error into two errors.
For these split points, if there is no overlap between the cut-off times, although there are pass counts, the pass counts are not necessary, and data can be directly filtered using the cut-off times.
When there is an overlap in the cut-off times between split points, the required data cannot be directly filtered by the cut-off times, and only the pass counts can be referenced to filter the data.
