| T O P I C R E V I E W |
| Coolwind |
Posted - 20/01/2012 : 23:04:11
http://www.bware.it/LK8000/LK8000-23U-UPGRADE.ZIP
Lots of fixes for pilots using imperial or nautical distances.
Apologies for not checking these things 10 months ago!!
Zoom is now back ok. Please read the full list of things changed.
Bruno is still working on Optimized routes for PGs, Ulli is still working on Sideview for airspaces and sonar, me kalman and mat are here and there on the code.
Version 2.3u January 20, 2012
-------------------------------
NEW
. Support for CUP files with no header line
. Nearest airspace Sideview rescaled dynamically while approaching an airspace
. Reset topology zoom levels in configuration to defaults.
This is needed because old values in previous profiles were saved as km, not as meters.
Changing the name of variable for profile does the job. Old variables are not read, and new variables will be saved.
CHANGES
. Faster zoom in/out with full topology and airspace rendering.
. Nearest airspace Sideview fixes
Deactivate ballast buttons in basic config when polar has no ballast
Work on analysis pages:
Polar page
Added Glider Name
Bigger Font
Color adjustments for inverted colors
Airspace sideview
Sky shading in inverted colors mode
Color adjustments for inverted colors
Added current view name in title
Nearest AS
added too far warning
Baro page
Sky shading in inverted colors mode
500m scaling if meter altitudes
Color adjustments for inverted colors
Climb page
Sky shading in inverted colors mode
0.5 m scaling if m/s climb
Color adjustments for inverted colors
Task
Bigger font
adjusted colors
. Topology configuration setting steps lowered from 0.5 to 0.1.
This is needed to grant compatibility while changing distance units.
FIXES
. Fix: Flying below or above NOFLY ASP zones causes the yellow warning to disappear
http://www.postfrontal.com/forum/topic.asp?TOPIC_ID=5868
. Distance & ETE to Start : use Optimized Wpt when PGOptimize=ON
Change Optimized Wp to Correct Side of SG for Exit SG
. Fix flickering startup problem, workaround.
Since 2.2, selecting profile on startup would show for a while the first left button in a wrong position.
Use Correct Finish raduis in PGOptimize
. Fix two critical memory leaks (LoadSplash and CreateProgress)
. Fix bug in Notepad and Logbook view: long notes would crash the software.
Now we split a long note into two or more notes with the same title.
Current note sizes are: max 300 notes per file, each note max 200 characters per title, and 5k in size,
and max 300 lines per single note.
. Fix Max Height Gain should not be reset after task start or restart.
It is reset only on free flight detection after takeoff.
. fix PG Optimized Wp with Exit Start
. Use Optimized Wp for NextETE if PGOtimize=ON
. Fix Topology thresholds
There was a big problem derived from Kalman's 407af5370979b3aaeaa2d8e01240ffb126988e3b commit.
Many part of the software rely on absolute mapscale values for deciding what to print and how.
For example, topology items. The problem was that if we decide to print small towns at 5km mapzoom,
with the new system we shall be printing small towns at.. 5nm mapzoom as well. This is resulting
in totally messy map if using any distance unit other than metrics.
. Fix wrong thresholds for moving map painting when using non-metric distances (nautical or statue miles):
- max zoom level for shading terrain
- max zoom level for colored snail trail
- max zoom level for drawing thermal circles
-
. Fix RealScale for Activemap nearestwaypoint: on non-metric distance units selected, it was hard to select a waypoint
on the moving map clicking on it, not enough tolerance!
. Fix RealScale for Heading (track) line, appearing wrongly also at low zoom levels when non-metric distance units selected.
The heading track line, if activated, should only appear within a reasonable zoom level.
. Fix RealScale for WaypointInRange: for non metric distance units, more waypoints than necessary were processed for calculations.
. Fix RealScale for Waypoint printing on the map.
For non-metric distance units, waypoints were not correctly printed depending on map scale.
. Fix problem with non-metric distance unit, setting configurable topology thresholds incorrectly.
This was resulting in quite wrong rendering of topology!! Selecting for example 5 for small cities,
would show small cities up to 5km in the original intentions, but really it was also showing them
up to 5nm , if nautical miles were in use.
Now the topology configuration must show corrected scaled values for topology, and convert them
between different distance units.
. Fix problem in topology settings, also in 2.2, not saving decimal values .
|
| 15 L A T E S T R E P L I E S (Newest First) |
| eric.carden |
Posted - 16/07/2012 : 18:39:02
Great. Thanks.
Eric |
| Coolwind |
Posted - 16/07/2012 : 15:24:53
Hi, the fix is now merged in the 3.0d which will be released in a couple of days, max.
Of course it is also in the current 3.1 branch, as all of the fixes in 3.0.
|
| mazuk |
Posted - 13/07/2012 : 15:41:55
Ok, thank you Eric. |
| eric.carden |
Posted - 13/07/2012 : 15:22:55
Hi Mazuk,
I pushed the fix March 4, and Paolo merged it the next day. It's been in the LK master branch (and all beta versions) since then - it just didn't make it into the May 18 3.0c release.
Eric |
| mazuk |
Posted - 13/07/2012 : 10:12:06
Eric,
are you going to push the trail fix also to your fork?
ciao
Mazuk |
| eric.carden |
Posted - 12/07/2012 : 18:44:26
Oops. I apparently forgot to report back about the trail drift bug I fixed (a fix that I see isn't in v3.0c). On 3/29/12, I finally created my first PNA EXE, and I tested the fix using Condor. I've been using the fix in all flights (Condor and real) since then, and I've accrued probably 30 flight hours (mostly Condor) with the fix with it always working as expected.
Eric |
| Coolwind |
Posted - 05/03/2012 : 01:19:49
Got the reversible fix, lovely because there is no problem to do testing with it!
merging later this evening |
| eric.carden |
Posted - 05/03/2012 : 00:54:31
I dug around, found what I believe is the problem, and just issued a pull request for changes to LKDrawTrail.cpp to fix the problem. The code was/is trying to figure out whether a drifted point will be on screen without first figuring where that point should be drawn. It was using an overly-conservative (to avoid trying to show an off-screen point) approach, and with high winds and a very zoomed-in view, points are being omitted that should be shown. Be gentle, as I'm new to this Git/Github stuff. :-)
I haven't figured out how to set up my development environment to create PNA EXEs, so I haven't been able to test the fix. I'll be happy to test it, if you can provide me a PNA EXE including the changes I made.
Thanks,
Eric
P.S. - I know you're busy with documentation. This issue isn't urgent. |
| Coolwind |
Posted - 25/01/2012 : 16:21:57
Ok It wont be fixed in this case, not by me at least, for 3.0. |
| eric.carden |
Posted - 25/01/2012 : 15:58:56
I did some testing last night, and I found that the problem is the same in v2.2e as in v2.3u. The problem's there in "glider" mode, too, not just "Paraglider/HangGlider" mode. I tested wind speeds of 4, 8, 12, 16, 20, & 24 kts and saw the problem in all cases except the 4 kt case. The problem may have been more severe with increasing wind speed. Then I turned trail drift off (I prefer it on) and re-tested (only at 24 kts), and there wasn't a problem. So it seems to be a problem with trail drift that manifests itself more and more with increasing wind speed.
Eric |
| AlphaLima |
Posted - 25/01/2012 : 09:07:13
Just as an idea,
have you checked Trail drift ON/OFF in the Config 3 dialog?
|
| Coolwind |
Posted - 24/01/2012 : 22:54:07
New issue. However you should first be able to replicate it on 2.3, and then check that in 2.2 it did work correctly.
|
| eric.carden |
Posted - 24/01/2012 : 22:29:16
I haven't had time to test this more fully yet (maybe tonight), but I wanted to report it in case it's new. While using LK 2.3u with Condor with 24 kt (44 km/h) winds, the snail trail misbehaved in circling zoom.
For what it's worth, I was flying a sailplane (of course) but had LK set up like for my (real-world) hang glider. Except for the high winds, I've done this plenty of times before without having a problem, so I'm guessing the high winds (I've never set them this high in Condor before) are a key part of the scenario.
The problem was that the trail (1) wasn't as long as usual and (2) would vanish at intermediate points (i.e., not at the oldest points first). So I only got about one full circle of trail - usually with a piece of the circle missing, which looks a lot like zero lift (very thin line) unless I look very closely.
I'm not sure whether older versions would have had the same behavior (I can try if desired), because I've never quite used LK in this scenario before. My biggest real-life cross-country flight, though, was with winds this strong aloft, so it's important to me that this work in my hang glider with winds this strong.
If this is a new issue and you'd like more details, let me know what details would be most helpful, and I'll provide them.
Eric |
| davesalmon |
Posted - 22/01/2012 : 22:44:50
OK, thanks. |
| Coolwind |
Posted - 22/01/2012 : 21:54:01
I try to explain better, Dave. There was a problem indeed, now fixed, that prevented your DAT file to be loaded.
However, when you say that you use that DAT and not the other CUP, I can tell you that the only difference for LK is the name of the waypoints. In DAT the name is the one that fits your need for the first 3 letters, in CUP it does not.
This has nothing to do with LK! We are NOT using the 3 letters in DAT at the beginning of each line, this is what I am telling you.
In any case, now it is fixes so you can load again those waypoints (2.3v).
|
|
|
