diff options
| author |  Rick Walsh <rickmwalsh@gmail.com> | 2017-10-29 17:47:09 +1100 |
|---|---|---|
| committer |  Robert C. Helling <helling@atdotde.de> | 2017-10-30 21:48:15 +0100 |
| commit | f2535995f0dcf2c6541525fcb09f86f138a2c186 (patch) | |
| tree | b0f8d00f8353f7e99b492540c023334155ee17a1 | |
| parent | eb62ced8a185927ca1c6328e54a84b2f684dfcd1 (diff) | |
| download | subsurface-f2535995f0dcf2c6541525fcb09f86f138a2c186.tar.gz | |
VPM-B ceiling outside of planner: over-estimate deco_time on first iteration
This means the iterations converge from an over-estimate, consistent with
planning VPM-B dives
Signed-off-by: Rick Walsh <rickmwalsh@gmail.com>
| -rw-r--r-- | core/profile.c | 6 |
1 files changed, 4 insertions, 2 deletions
diff --git a/core/profile.c b/core/profile.c index e96febffb..d62caab93 100644 --- a/core/profile.c +++ b/core/profile.c @@ -1028,9 +1028,11 @@ void calculate_deco_information(struct dive *dive, struct divecomputer *dc, stru if (first_iteration) { nuclear_regeneration(t1); vpmb_start_gradient(); - /* For CVA calculations, start by guessing deco time = dive time remaining */ + /* For CVA calculations, deco time = dive time remaining is a good guess, + but we want to over-estimate deco_time for the first iteration so it + converges correctly, so add 30min*/ if (!in_planner()) - deco_time = pi->maxtime - t1; + deco_time = pi->maxtime - t1 + 1800; vpmb_next_gradient(deco_time, surface_pressure / 1000.0); } } |