Change Note 3.04.3
==================

- Fixed bug in bottom segment gas needs calculation.

- In the Tech version, max. ppO2 limits can now be adjusted up to 2.0 bar.
  Caution: know what you are doing! Recommended max settings are 1.4 bar for
  work gases and 1.6 bar for deco gases.

- CCR mode diluent ppO2 versus setpoint check can be disabled now.

- Fine tuning of the better gas cue behavior.

- Improved smoothing of compass display.

- Attention and warning thresholds for gas density put into the options
  table (adjustment via external software only, no mapping in OSTC menu
  as of now).

- Fixed bug in surface mode firmware version output (relevant to Sport only).


Change Note 3.04.2
==================

- Fine tuning of the deco calculations for extended stops and normal mode.


Change Note 3.03.6 - .8
=======================

- Various internal code improvements.


Change Note 3.03.5
==================

- Renamed OC gas type 'normal' to 'work' (gas for the working phase of the
  dive). Along with this, it received a down-arrow symbol to depict it as
  a gas for use at depth, similar to the '=' symbol that was introduced some
  time ago to visually depict deco gases (imagine the horizontal lines as
  deco stops).

  Set up gases as follows:

  Single gas dive:
  set gas to use as 'first' and all other gases to 'disabled'.

  Gas-change, extended no-stop dive:
  set gas to start with to 'first', all other gases to be used later in the dive
  to 'work' and the unused gases to 'disabled'.

  Deco-dive with one gas for descent & bottom phase, and deco gases:
  set bottom gas to 'first', deco gases to 'deco' and the unused to 'disabled'.

  Deco-dive with a deco gas used as travel gas on ascent, bottom, and deco gases:
  configure the deco gas twice: once as travel gas and once as deco dive. Set the
  same change depths for both instances and put the travel gas before the deco
  gas in the gas list (i.e: gas 1: travel, gas 2: work (bottom) gas, gases 3-5:
  deco gases). Set the travel gas to 'first', the bottom gas to 'work', and the
  deco gases to 'deco'. Take care to adjust the pressure available on the deco gas
  which is used for travel gas, too, to the amount of gas that is planned to remain
  after having drawn the travel portion from it!

- When ppO2 is above normal threshold, but still inside the deco max threshold, and
  the dive has not yet progressed to doing the deco stops, the ppO2 warning has been
  degraded to an attention, i.e. associated messages appear in yellow now instead of
  red as before.

- In the simulator, the +5 minutes function can now be triggered in succession without
  needing to reopen the options menu each time.

- The French language package has been upgraded.

- Setting the change depth of the first gas to zero caused some unintended behavior.
  This is captured now. Anyhow, since firmware 3.2 it is required to set proper change
  depths for gases and diluents, including the ones set as 'first'!

- The deco calculator now automatically picks the best gas suitable for the set bottom
  depth. If no suitable gas is found (i.e. configured), calculation will be aborted with
  an error message. By this change it is possible to calculate dives that use a travel
  gas (set as 'first') and a bottom gas (set as 'work'), as the deco calculator will now
  pick the proper bottom gas for calculating the bottom segment.
  Note: the deco calculator does not calculate the descent phase but actually spends all
  of the set bottom time at bottom depth. While this puts the deco calculation results
  on the safe side, beware that the needs on the travel gas will not show up in the gas
  needs calculation results.

- OSTC TR: The timeouts for attentions and warnings in connection with discontinued
  reception of transmitter data have been relaxed a bit:
  - For up to 2 minutes without updating the last value received will be retained.
  - After 2 minutes without reception of any updates the pressure value will be marked
    as outdated.
  - After 4 minutes without reception of any updates a transmitter will be declared lost.

- OSTC TR: pressure logging has been implemented. Logged pressure data can be retrieved
  by downloading the logbook data using external logbook software.

- Descent speed in the simulator has been increased from 19 to 24 meters per minute.
  This speed is not configurable, nor the simulator's ascent speed of 9 meters/min!

- In dive mode, the battery voltage will be shown when reaching 30%, and a warning will
  be issued when the percentage reaches 25% with 1.5 Volt batteries or 15% with LiIon
  batteries.


Change Note 3.03.4
==================

- More compact code for TR firmware and compass.

- Slightly more compact code for deco engine.

- Improvement of code for CNS calculation.

- Added a hysteresis to the dive start / dive end depth threshold.

- Bugfix in non-helium compile variant.


Change Note 3.03.3
==================

- Added an option "extended stops" (named "Gas Change/...before 1.Stop" in Deco Setup
  -> Deco Parameters -> more). If set to YES, a gas switch will be planned immediately
  on the change depth of the gas, even if a stop on that depth is not required by deco
  constraints. When set to NO, the gas switch will be planned to be done on the next
  regular deco stop.
  With extended stops enabled and a setting of "Gas Change (time):+0'" the switch will
  be planned with no actual halt at the change depth (i.e. "switch on the fly"). With
  change time set to > 0', a dedicated stop will be placed at the change depth for the
  given number of minutes. This gas change stop may not be needed by deco constraints,
  but it will influence the following deco stop depths and times as it introduces some
  added time at depth.


Change Note 3.03.2
==================

- Complete rework of the alternative dive mode screen layout.

- Various code improvements.



Change Note 3.03.1
==================

- Various code improvements.

- Included RX firmware update into tech version.

- Modularization of the code to allow to compile in or not various features.


Change Note 3.02.1
==================

- Reworked code for ascent and deco stops calculation. Does produce different results
  compared to the code used in 3.00.3! With this implementation of an GF-controlled
  deco schedule it becomes visible when deco stops disappear due to a high GF slope
  (difference between GF low and GF high) combined with a moderate depth of the first
  deco stop.

- When switching between GF and aGF, the depth of the first stop (determined by GF low)
  will be recalculated now.

- Improved calculation of the supersaturation.

- New tissue graphics option (compile option as of now): shows absolute tissue pressures
  (N2 + He added) and saturation per tissue. In the saturation bargraphs, the green line
  marks GF low position, the yellow line GF high and the red one 100%.

- Numerous improvements around dive loop and deco engine start-up.

- speed_fastest now available again on new OSTC Sport, will adjust to 32 MHz.

Memory status: 2.252 bytes free with single language (English), RX-functions & screendump


Change Note 3.00.3c
===================

- GF low can not be set higher than GF high any more as such a setting would produce
  an illogical deco schedule.

- In case the compass is not calibrated, the heading is now displayed as "---°"
  instead of ".00°".


Change Note 3.00.3b
===================

- In the surface tissue graphics, tissues whose supersaturation exceed 100% are shown
  in red color now.

- If any tissue is beyond 100%, no desaturation and no-fly times will be shown anymore.

- After use of the dive simulator and the deco calculator the interim tissue desaturation
  is catched up now.

- Internal: prohibit selection of fastest speed on OSTC Sport hardware (uses flag for
            screen type 2 while there is no dedicated flag for Sport hardware itself).

Memory status: 2.627 bytes free with single language (English), RX-functions, screendump
                                     and deco engine performance measurement compiled in

Change Note 3.00.3
==================

- Added support for display type 2.

- Added Deco Engine scheduling performance measurement (accessible via debug output).

Memory status: 2.952 bytes free with single language (English), RX-functions and screendump.


Change Note 3.00.2
==================

- The Deco Engine has been reworked to now run for up to 700 ms each second (configurable,
  800 ms may be feasible also), split into two compute chunks per second. This increases
  reactivity to buttons presses and compass display, while boosting deco calculation turn-
  around times significantly.

Memory status: 3.760 bytes free with single language (English), RX-functions and screendump.


Change Note 3.00.1
==================

- The simulator now uses the current ambient pressure as surface pressure instead of a fixed
  1.000 hPa as done before. So in mountain diving, when being at the dive site the simulator
  will show the same NDL times and deco schedules as will be shown in a real dive.

- When exiting the surface menu, the return to the normal surface screen has been sped up
  and the display of the power-up logo screen is skipped now also.

- When submerged, the unit will now proceed from surface mode display to dive mode display
  without intermediate display of the power-up logos.

- A cue is given when it is time to go up to th next deco stop by highlighting the next stop
  data until ascent is done.

- When being shallower than the indicated stop depth, i.e. because the deco needs to be
  accelerated, the reduced ambient pressure is figured in into the deco calculations such
  that the displayed stop time, TTS and gas needs do immediately reflect the reduced depth. 

- When set to imperial units, all depth will be shown in feet now not only in dive mode but
  also throughout the whole surface menus.

- When in a leap year, the 29th of February is now correctly handled when adjusting time and
  date on exactly that day.

- Bugfix: with the option 'always show ppO2' set, in CCR and pSCR mode a "Dil:x.xx" ppO2
  message was sometimes shown twice, once as an info and additionally as an attention or
  warning. This has been fixed now: in case of an active attention or warning the secondary
  info message gets suppressed.

- Bugfix: the Simulator +5 min function had a punch-through effect on the real world last
  dive time (displayed surface interval).

- Tissue desaturation is now also computed while in surface and menu modes.


Some more internal improvements:

- Handling of data transfer into and out of the interrupt service routine has been reworked.

- Memory placement of key variables has been reworked to reduce the number of bank switches.

- In case option check reveals no errors, an unnecessary write-back to the EEPROM is suppressed
  to speed-up restart and reduce wear.

- Large portion of the code have been inspected / reviewed.

- Further improvements in code quality, formatting and commenting (done while code review).