In recreational mode, we keep adding time at the last depth
until an ascent does violate the ceiling. Then we roll back
the last added time step and record the ascent. The test for
the ceiling violated was before adding the time so if it alreay
failed the first time we tried to unroll a time step that was
never added which resulted in a small kink in the pressure graph.
This patch corrects this logic by changin a while to a do {} while.
Furthermore, it removes the computation of deco state during the
final ascent since that is not used anywhere later.
Signed-off-by: Robert C. Helling <helling@atdotde.de>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This reverts commit 3d8e5b638a.
Calculating the next gradient should be based on the tissue loading at the end
of the previous iteration, so it was wrong to restore the deco state first.
This has a tiny affect on the calculated profile, and makes one of the tests
fail.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This looks like possibly a false positive in the Coverity scan, but we can
always assume that the first point of the dive plan has been entered by
the user.
Coverity CID 1325285
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
for a 'struct gasmix' initialization, the 'permille' value
from franction_t should have it's own braces.
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
We do not want to remember max_bottom_ceiling_pressure from the previous planned
dive - it makes the Boyle's law compensation incorrect if the previous planned
dive was a deeper deep-to-shallow multi-level dive.
E.g. Plan these dives (without applying this patch) with VPM-B nominal
conservatism:
Plan 30 m for 20 min (total run time = 28 min)
Now plan 100 m for 20 min, followed by 70 m @ 23 min and 70 m at 30 min
Re-plan 30 m for 20 min (total run time = 50 min)
With this change, the re-planned dive run time is 28 min, as it should be.
We probably don't have to reset first_ceiling_pressure too, but it's cleaner if
we do.
Signed-off-by: Rick Walsh <rickmwalsh@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The next gradient calculation at the start of the CVA loop should be performed
using the tissue loading before starting the iterations, rather than the loading
at the end of the previous iteration.
Signed-off-by: Rick Walsh <rickmwalsh@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Unify spelling of this name in strings shown to the user, as commented
in mailing list. Internal coding use untouched.
Signed-off-by: Salvador Cuñat <salvador.cunat@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
There is an unfortunate convention that the sample structure contains a setpoint
at its end rather than during its duration which causes the expression
sample[-1].setpoint in several places. This adds another one for the planner
to use the correct setpoint during the manually entered leg of the dive.
Signed-off-by: Robert C. Helling <helling@atdotde.de>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The Boyle's law compensation compares the ambient pressure to a baseline value,
and adjusts the theory bubble radius accordingly. Currently we use the ceiling
at the last user-entered waypoint (the start of the decompression phase) as the
baseline value. However, in a deep to shallow multi-level dive, decompression
can start earlier, and taking a shallower ceiling leads to a more aggressive
ascent. This is particularly noticeable if the user enters stops during ascent.
With this commit, we take the baseline ambient pressure for Boyle's law
compensation as the deeper of the:
(1) Ceiling prior to ascending during a user-entered portion of the dive, and
(2) Ceiling at the start of the last user-entered waypoint.
This makes the calculated profile more conservative for some deep to shallow
multi-level dives.
Signed-off-by: Rick Walsh <rickmwalsh@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
So far, add_segment() returned a tissue tolerance (i.e. ceiling)
computed just in its return statement. This tissue_tolerance
needed to be dragged around until it was needed or be dropped
if not needed at all.
As for VPM-B, this ceiling computation is a bit expensive, this patch
calls the computation function tissue_tolerance_calc() when the
value is actually needed and not before.
This changes the signature of some functions.
Signed-off-by: Robert C. Helling <helling@atdotde.de>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
and not some time before and store the result in a global variable.
This stores only the bottom gradients and computes the Boyle compensation
when computing the allowed ambient pressure.
As the Boyle compensation needs a reference ambient pressure, to find the ceiling,
we have to iterate this computation until the reference pressure is close enough to
the ceiling.
Signed-off-by: Robert C. Helling <helling@atdotde.de>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
We now do the first_ceiling_pressure calculation at the start of the CVA loop.
We don't need to do it before.
Signed-off-by: Rick Walsh <rickmwalsh@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
We should run clear_deco as early as we can. We should calculate the nuclear
regeneration and start gradient after calculating tissue tolerance. We don't
need to redo nuclear regeration and start gradient after that.
Signed-off-by: Rick Walsh <rickmwalsh@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
By calculating crushing pressure after the manually entered phase, we were doing
it wrong for multi-level dives.
Signed-off-by: Rick Walsh <rickmwalsh@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The pressure for the Boyle compensation is of the first ceiling,
i.e. the ceiling seen from the bottom rather than the first
stop.
Signed-off-by: Robert C. Helling <helling@atdotde.de>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Otherwise, the results of the calculations tend to be rather
random and irreproducible...
Signed-off-by: Robert C. Helling <helling@atdotde.de>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The Boyle's law compensation depends on first_stop_pressure. To produce
profiles similar to other VPM-B implementations, we should calculate it as the
ceiling before starting the ascent.
Commit 159c9eb2c1, Compare ceiling to next stop
rather than try to ascent for VPM-B, changed (VPM-B) to consider the current
ceiling rather than an incremental ascent between one stop level and the next.
However, the initial ascent generally steps through several stop levels, so
first_stop_pressure was still not calculated as the ceiling prior to commencing
the calculated ascent.
Signed-off-by: Rick Walsh <rickmwalsh@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
otherwise VPM-B planned profiles seem to violate the ceiling. This needs
the first_stop_pressure to be available also in the profile, so I made
it global in planner.c
Important lesson: If you want to use deco_allowed_depth on a tissue_tolerance
that comes from a VPM-B planned dive, you have to call boyles_law() before
add_segment()!
Signed-off-by: Robert C. Helling <helling@atdotde.de>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Since a8ce8, that made deco_allowed_depth work for VPM-B as well, this
function became obsolete but was reintroduced by one of Jan's latest patches.
Signed-off-by: Robert C. Helling <helling@atdotde.de>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
For VPM-B, to stay within the reference implementation, to decide if we
need to stop we check if the current ceiling is above the next stop depth
rather than trying to ascent and check if we violate a ceiling. This
leads to more conservative profiles.
Signed-off-by: Robert C. Helling <helling@atdotde.de>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Previously we were calculating the compensation only on the deco
stops, gas change stops appearing before the first deco stop were
ommited.
Signed-off-by: Jan Darowski <jan.darowski@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Usually, we try to ascent to the next stop and check if we break the
ceiling while doing that. This patch adds a preference value to rather
check if the ceiling is above the next stop before attempting to ascent.
The difference if off-gasing during the ascent is taken into account.
Logically, it does not sound like it could be relevant to ignore that
off-gasing but it leads to more conservative schedules and it seems
the original Fortran VPM-B implementation does just this. So one could
argue it is part of that model (if it makes sense or not), so we should
at least give users the possibility to turn this on.
Maybe we should even make this the default for VPM-B.
This patch just addes the code to have the value in the preferences and
the planner to act accordingly. There is no UI for it, yet. To test, you
have to set it in the code. There could be a later patch with a UI if people
like to have it.
Signed-off-by: Robert C. Helling <helling@atdotde.de>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This patch makes deco_allowed_depth() work both for Buehlmann as well as
VPM-B (as long as the VPM-B internal variable total_gradient[] is valid).
As a bonus, in VPM-B mode, in the planner, the ceilings are VPM-B ceilings
and not Buehlmann GF.
Signed-off-by: Robert C. Helling <helling@atdotde.de>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Now, we calculate the volume of free gas not only based on the deco
time but also time on the surface, needed for the full desaturation.
Signed-off-by: Jan Darowski <jan.darowski@gmail.com>
The default stack size on Windows per thread is 1024kb.
Using the heap prevents a stack overflow in add_plan_to_notes().
The alternative is to tell the linker to use N bytes of stack:
-Wl,--stack,N
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This fixes a bug where we would not do any gas changes at all if at the end of
bottom time we were deeper than the MOD of the bottom mix. Instead we would
constantly try to switch to that gas and find we are already breathing it.
A test case would be a dive with air and EAN50 to more than 66m. That would
have never switched to EAN50.
Signed-off-by: Robert C. Helling <helling@atdotde.de>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The purpose of parts of the planner code is not not always obvious. This is
especially true of the conditional statements in the add_plan_to_notes function.
Annotate the code with comments to explain the logic.
Signed-off-by: Rick Walsh <rickmwalsh@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
In commit 156ad42a3c ("snprintf is happier if it has an explicit string
literal format argument") I got a little too aggressive making sure there
are string literal format strings...
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
If display transitions in deco isn't selected, then we shouldn't show the
transition segment between two deco stops, even if there is a gas change. We
should still show the ascent segment up to the first deco stop. The
(gaschange_after && !isascent) condition is necessary to pick up backgas breaks.
An example plan is shown below.
Previously:
depth duration runtime gas
40m 1min 1min air
40m 34min 35min
21m 2min 37min <--meaningful ascent segment
21m 1min 38min EAN50
18m 1min 39min
15m 3min 42min
12m 4min 46min
9m 5min 51min
6m 0min 51min <--unnecessary ascent segment
6m 13min 64min oxygen
6m 6min 70min air
6m 2min 72min oxygen
0m 1min 73min
depth duration runtime gas
40m 1min 1min air
40m 34min 35min
21m 2min 37min <--meaningful ascent segment
21m 1min 38min EAN50
18m 1min 39min
15m 3min 42min
12m 4min 46min
9m 5min 51min
6m 13min 64min oxygen
6m 6min 70min air
6m 2min 72min oxygen
0m 1min 73min
Signed-off-by: Rick Walsh <rickmwalsh@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
When the last stop at 6m/20ft option is selected, replace the 3m/10ft stop with
zero depth, rather than doubling up on the 6m/20ft stop. This removes the need
to differentiate between 6m (=6000mm) and 20ft (=6096mm) and saves calling a
helper function. It does not alter the calculated profile at all.
Signed-off-by: Rick Walsh <rickmwalsh@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Previously we used strncat to output VPM mode without correctly defining the
length of the string, and didn't do anything for recreational mode. This
resulted in the output being junk recycled from the previous temp string.
We could use strncat if the string length were defined, but using snprintf will
make it simpler to include the VPM conservatism when that has been implemented.
Signed-off-by: Rick Walsh <rickmwalsh@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Update the disclamer text to reflect which algorithm where used.
[Dirk Hohndel: cleaned up coding style & replaced snprintf with strncat]
Signed-off-by: Joakim Bygdell <j.bygdell@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Make precedence of && over || explicit.
Explicitly convert between char * and unsigned char *.
Don't assign potentially negative return code to an unsigend variable.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
