Table 1.
Comparison of computational time consuming under different NLTE settings.
| Case | Lyman-α | Lyman-β | Δτmin | Cost [μs per grid point per core per time step] |
|---|---|---|---|---|
| 1 | off | off | - | 38.70 |
| 2 | on | off | 10−5 | 42.14 |
| 3 | on | on | 10−5 | 45.35 |
| 4 | on | off | 10−8 | 42.72 |
| 5 | on | on | 10−8 | 46.88 |
Notes. Five simulation cases, all starting from the same initial conditions described in Sect. 3.1, are compared. In Case 1, the NLTE RT module was not activated, while it was enabled in Cases 2–5. Radiative transfer was solved for Lyman-α only in Cases 2 and 4, and for both Lyman-α and Lyman-β in Cases 3 and 5. In Cases 2 and 3, we restricted RT to the lower atmosphere by adopting a large value of Δτmin, whereas in Cases 4 and 5 we solved it for both the lower atmosphere and the optically thin corona using a smaller Δτmin. Each simulation was run for 1000 time steps, and the computational time was measured over time steps 100–1000. The first 100 time steps were excluded to allow for atmospheric relaxation.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.