13
0
Commit Graph

31 Commits

Author SHA1 Message Date
ffc9239de5 ControlList: replace code use to interpolate MIDI CC curves
Original code is of questionable historical provenance, and
was needlessly (it seems) complexity. New code is relatively
simple arithmetic linear interpolation.
2022-08-17 20:55:17 -06:00
b03e9b4c10 NOOP: add helpful explanatory comment 2022-08-12 09:21:23 -06:00
610b5dddbf ControlList: GUARD_POINT_DELTA needs to take time domain into account 2022-08-11 15:26:42 -06:00
c8feef51ab convert use of operator* for tim::line types with ::scale(ratio_t) 2022-05-27 12:47:44 -06:00
60e5b84d78 temporal: alternative solution to overflow in timeline operator*()
This uses boost::multiprecision::int512_t when multiplying and dividing by the numerator
and denominator of a ratio_t. 128 bits would be sufficient but for some reason, the boost
docs show the 512 bit variant being very slightly faster.

This is a better solution than using a double, which although it will prevent overflow
has fairly limited resolution.
2022-05-24 21:46:10 -06:00
Mads Kiilerich
c1c95f538f evoral: fix ControlList::_x_scale to avoid ratio overflow when adjusting fade length
Region fades would sometimes get in a mode with weird behaviour. They
would be drawn in 2d with crossing lines, mainly moving back and forth
horizontally - not as a function of time. It would sound as it looked.
The fade would sometimes jump around when resizing. It could be worked
around by resetting the fade shape. It turned out the problem could be
reproduced by making minute long fades.

This change fixes or works around the problem.

Back story:

timepos_t (in temporal/timeline.h) uses 62 bit for integer value and the
max value is thus 4611686018427387904 ~ 5e18. timepos_t counts
superclocks, where superclock_ticks_per_second is 56448000 ~ 6e7. It can
thus store up to 8e10 seconds - thousands of years.

ratio_t (in temporal/types.h) can represent fractions as 64 bit (signed)
numerator and denominator. timepos_t avoids floating point operations,
but has operator* with ratio_t. To avoid crazy loss of precision it will
multiply the superclock count with the numerator before dividing with
the denominator.

Audio region fade in and out uses a number of increasing timepos_t
values (in a ControlList) up to the length of fade. When dragging to
resize, these values are (in_x_scale) multiplied with the ratio_t of the
new and old fade length. The problem is that the 62 bits will overflow
if using fades more than sqrt(5e18) ~ 2e9 superclock ticks ~ 38 seconds.
It will overflow into the "beat" flag and (at 58 seconds) into the sign
bit. The timepos_t values in the fade will thus jump and can be negative
or change to count beats.

To work around that problem, this changeset just use floating point
values for scaling the timepos_t values. All scaled values are stored as
integer anyway, so it should not make any actual difference for this use
case. There might however be other uses of ControlList where it matters.

As an implementation detail of this "workaround" of using double, it
could perhaps also be nice to implement timepos_t operator* (or
operator*=) for double. But I'm not sure we want floating point support
in timepos_t.

An alternative (and better) solution would be to convince the fraction
multiplication to use 128 bits. It is essential to avoid overflow -
mainly in static analysis, alternatively as runtime checks or asserts.
2022-05-24 17:15:37 -06:00
cd332a2af0 when pasting a Range of automation, first add guard points,
so the automation data before and after this range is retained
2022-05-16 07:16:25 -05:00
3a174ff914 add support for 'positive' guard points in ControlList 2022-05-16 07:16:25 -05:00
c98561e95c when cutting or clearing an automation range, always add boundary points
* these guard points are necessary to retain the automation that existed
 before and after the selected range is removed
2022-05-16 07:16:24 -05:00
2b8b9a3a2f const for const-sake 2022-05-01 18:01:35 -06:00
d218dcb21b evoral: when adding points to a ControlList, coerce the time domain of the new point to the list time domain
This may not be the ideal solution, but for now it appears to be the best approach to preventing
points with different time domains with the same list
2022-05-01 17:50:51 -06:00
8657eba4c5 evoral: NOOP reorder and edit comments 2022-04-05 20:52:09 -06:00
8c3fad0133 evoral: fix implementation of rt_safe_earliest_event_linear_unlocked
Code within the method was using @param start_time rather than start, which is a modified
value required to generate the correct results.

This comment also contains some logical reordering, optimization and commenting
on this rather complex method.
2022-04-05 20:52:09 -06:00
256e6f97a2 evoral: change how we specify a zero min_x_delta to rt_safe_earliest_event_linear_unlocked
See comment in the code for more details.
2022-04-05 20:52:09 -06:00
176c41a485
Fix region-gain when region-start is trimmed
Region-gain (unlike other automation) is specific to the
region and independent of the source. Region::start() offset
does not apply. When region-start is trimmed the region's
envelope is modified (not just offset). The event-list is truncated.

Any audio-region envelope does (and must) have a point exactly
at the start and end of the region.

truncate_start() can thus calculate the earliest position of
valid events with the new length relative to the last event.

The mathematical operator for that is subtraction, not distance.
2022-03-17 23:46:55 +01:00
60be0c27a1
Use updated temporal API 1/2 2021-12-11 14:42:36 +01:00
940f3022c1 prevent addition of events to a ControlList whose time-domain does not match the ControlList's own time-domain 2021-08-30 15:15:14 -06:00
6d154c14b4 libevoral: add a new method to set the time domain that can be used from ::set_state(), when the event list is not (yet) empty 2021-08-13 12:51:34 -06:00
1a5e2aacff remove another NUTEMPO warning 2021-08-13 12:51:32 -06:00
779a04b0ce Evoral: alter ControlList::paste() to internally change time domain of pasted events 2021-08-13 12:51:31 -06:00
bed76798f3 manual fixups after rebase against master (general libs edition) 2021-08-13 12:51:31 -06:00
18fea5c7a9 changes required by fixing ambiguities in timepos_t/timecnt_t API (evoral edition) 2021-08-13 12:51:31 -06:00
def35cf043 mark BBT_Offset (double) constructor explicit to avoid implicit conversion in timeline expressions.
Also clean up the mess this had caused.
2021-08-13 12:51:30 -06:00
ebd8704f9d libevoral: tweaks related to timeline types based on libardour conversion 2021-08-13 12:51:29 -06:00
6b09642406 changes to compile against libtemporal and use of timepos_t/timecnt_t 2021-08-13 12:51:28 -06:00
b9cfb31205
Fix control-list editing (#8384)
This fixes segfaults as well as corrupt listes when copy/pasting
due to invalid iterators.

::mark_dirty() must be called with WriterLock, and
::rt_safe_earliest_event_linear_unlocked() must not be called
while _events is being modified. The Sequence iterator
(only user of that function) does not ensure this. Only the
sequence read-lock is taken.
2020-08-31 07:39:45 +02:00
65ecc1b40e
Do not interpolate away explicit MIDI automation points #8362 2020-08-15 01:57:01 +02:00
e4c56a0371
Remove unused API
These are longer be used since Seuqnce has a "force_discrete"
boolen that needs to be taken into account in addition to
user-configurable ControlList _interpolation mode.
2020-08-15 01:56:42 +02:00
650fd31332
Improve debug message 2020-06-12 05:08:40 +02:00
7d90ad4023
Fix bool-automation anchor
Typo sneaked in from ff2f93497...cc7de475f2
2019-11-13 16:53:46 +01:00
eaae38ba84 move evoral/src/* to evoral/ 2019-11-02 16:32:18 -06:00