Bílý vrch
icon TrekBuddy
www.trekbuddy.net
Outdoor companion.
  • internal / bluetooth / simulator GPS
  • offline raster maps
  • smart GPX / raw NMEA logs
  • waypoints and simple navigation
  • custom views
  • MIDP and Symbian phones
  • Blackberry
  • Android
Visit wiki to see all features, guides and howtos. Project tracker.

Partners:    (Polish/Polski)(Polski) Compass mapy      (Polish/Polski)(Polski) Galileos mapy      (Polish/Polski)(Polski) CartoMedia      (Czech/Èesky)(Èesky) Eaglesoft trasy      (Polish/Polski)(Polski) ExpressMap     

 FAQFAQ   SearchSearch   RegisterRegister 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 
240x320 5-panel HECL-CMS for Flying
Goto page Previous  1, 2, 3, 4, 5, 6  Next
 
Post new topic   Reply to topic    TrekBuddy Forum Index -> Themes
View previous topic :: View next topic  
Author Message
nattila



Joined: 21 Sep 2009
Posts: 29

PostPosted: Fri Sep 25, 2009 3:54 pm    Post subject: Re: wind test Reply with quote

thomast wrote:
second reason can be that the timebase is wrong,i sent you my debug cms.(as zip) when you use it in 5hz mode, shows the last line the time in ms.it should change in 200ms steps.did it?

Hello Tomas,
Although I can't use your CMS because my bug, I love it. Great job!
I made some tests with your debugger. The timecode changes in 200ms steps when it counts. Sometimes doesn't. It freezes down but the CMS still working. Sad Somtimes only two 0 0 are here.

Honestly I don't need 5hz frame rate. Don't you have a quick custom hack to slow down my frame rate? Would it possible modify air.hcl somehow?

thanks.
Back to top
View user's profile Send private message
troubadixx



Joined: 07 Jul 2009
Posts: 4

PostPosted: Fri Sep 25, 2009 10:14 pm    Post subject: Re: 240x320 5-panel HECL-CMS for Flying Reply with quote

thomast wrote:
Hi troubadixx,
looks nice, but you need good eyes Shocked

Maybe we should call it eagle-eye-view...


Hallo Thomas,

I finally got around testing TrekBuddy and your ABSOLUTELY GREAT CMS in the air.
(Please forgive my noisy enthusiasm Very Happy )
I have now 20+ hours of airtime (paragliding) with it - and I'm loving it!
Please note, that I was testing the version as of end of August.

I was very positively surprised by the "eagly-eye-view" in practice: Mounting my Nokia 6300 on my front rescue container, I have it quite closely and quite directly in my standard line of sight. With this setup, I can quite effortlessly (and without special eagle eyes Wink) have all relevant information available on one screen at one glance.

One issue re/ navigation, UI related:
The sliding compass requires indeed a bit more of an "eagle eye". That's not so much of a problem when just checking the heading. In practice, though, it takes a while to figure out the difference between heading and bearing - i.e. where to turn. Arrow I'm excited to see that your latest update addresses just this issue with the DIR2WP arrow! Very Happy

Two downsides with my specific setup, device related:
* Battery lifetime on my mobile phone is not sufficient for really long flights (Nokia 6300 gave me up to 6 hours battery life in Spain in high temperature conditions - I expect this to be considerably less in spring in the Alps)
* Wearing gloves, trying to change settings (loading a waypoint, changing setting - or merely changing the CMS view) is very cumbersome, with thick gloves it's next to impossible

And two fundamental reasons why I'm still using my Garmin (eTrex Vista HCx):
* Possibility to show map with tracklog and some key infos (e.g. vario and speed) on one screen
* Routable (vector) maps for the way home after XC
Unfortunately I doubt that TrekBuddy will offer these possibilities anytime soon...

However, I still love this project - and appreciate the effort you're putting in!!!

Werde wohl nicht mehr weiter zum Fliegen kommen diese Saison. Ergo werde ich höchstens einen Trockentest Deiner neuen Version schaffen...
Melde mich, falls mir was auf- oder einfällt.

Viele begeisterte Grüße,
Troubadixx.
Back to top
View user's profile Send private message
thomast



Joined: 16 Jan 2009
Posts: 101

PostPosted: Sat Sep 26, 2009 9:35 am    Post subject: Re: wind test Reply with quote

rod wrote:

I think these parameters (dir and speed wind) should be adjusted at takeoff. What happens if the flight takes long hours and travels a great distance, it is necessary to adjust these parameters again?
And, this theme can be used on sailplanes or airplanes (from the standpoint of Math Model)?
RSG.


yes wind you can set before start,so s2f calculator works from the beginning correct.otherwise you must wait until the cms detects the windconditions.later you can overwrite this predefined values just by pressing 2 in the main cms.

The mathmodel was created from "Helmut Reichmann" a Sailplanepilot Champion.
So it work also for Sailplanes ,when the Polare is set to the Correct one.The Polare of a ask21 is already in the script.

For motorplanes it makes no sense to have the s2f stuff...

Have a nice Weekend,
Thomas
Back to top
View user's profile Send private message
gdpno69



Joined: 03 Sep 2007
Posts: 27

PostPosted: Sat Sep 26, 2009 9:53 am    Post subject: Testing Reply with quote

Hi Thomas,
just downloaded and simulating (unfortunately bad wheather today Sad ).
Climb goes out one digit and invades speed.
Wind arrow indicates wind direction but 180-degree rotated.
Cannot find any cms about gforce. What gforce are measuring exactly?
Back to top
View user's profile Send private message
thomast



Joined: 16 Jan 2009
Posts: 101

PostPosted: Sat Sep 26, 2009 10:03 am    Post subject: Re: 240x320 5-panel HECL-CMS for Flying Reply with quote

Hallo Troubadixx,
thanks for youre detailed review.

Quote:
One issue re/ navigation, UI related:
The sliding compass requires indeed a bit more of an "eagle eye". That's not so much of a problem when just checking the heading. In practice, though, it takes a while to figure out the difference between heading and bearing - i.e. where to turn. Arrow I'm excited to see that your latest update addresses just this issue with the DIR2WP arrow! Very Happy

i have the same Problem to see that with my old Eyes...

Quote:
Two downsides with my specific setup, device related:
* Battery lifetime on my mobile phone is not sufficient for really long flights (Nokia 6300 gave me up to 6 hours battery life in Spain in high temperature conditions - I expect this to be considerably less in spring in the Alps)

you fly 6 Hours? SPAM! Good Condition.
The new Version of the CMS has a reduced CPU Usage,so the power consumption should be a bit less.but 6 hours batterietime is not so bad.


Quote:
* Wearing gloves, trying to change settings (loading a waypoint, changing setting - or merely changing the CMS view) is very cumbersome, with thick gloves it's next to impossible

Yes it rearly complicate.It helps a bit when you create a keymap.txt
in the recources dir. i dont need key key * and 0.so i override them with key2(set a value) and key 6 (change cms).On my Nokia:
42=54
48=50

Quote:
* Routable (vector) maps for the way home after XC

Googlemapsmobil did this for me.i have a flatrate(simyo).
On Symbian, Google latitude is a nice to have.

mit der guten Thermik ists wohl bald vorbei.
Aber Soaren mach ja auch Spass.


Happy Landings,
Thomas
Back to top
View user's profile Send private message
thomast



Joined: 16 Jan 2009
Posts: 101

PostPosted: Sat Sep 26, 2009 10:37 am    Post subject: Re: wind test Reply with quote

nattila wrote:

Honestly I don't need 5hz frame rate. Don't you have a quick custom hack to slow down my frame rate? Would it possible modify air.hcl somehow?
thanks.


could be that the s2f calculation 5x sec. needs to much time.
one time per sec on my new mobile need about 20%

The User "gdpno69" has made a more intelligent algo.
but i have at the moment no time to implement that.

but if you dont need s2f Funktion you can make this Hack:
just rename the Files:
cms.air15.xml and cms.air16.xml
to
cms.air75.xml and cms.air76.xml


Thomas
Back to top
View user's profile Send private message
gdpno69



Joined: 03 Sep 2007
Posts: 27

PostPosted: Sat Sep 26, 2009 1:12 pm    Post subject: Re: wind test Reply with quote

thomast wrote:
nattila wrote:

Honestly I don't need 5hz frame rate. Don't you have a quick custom hack to slow down my frame rate? Would it possible modify air.hcl somehow?
thanks.


The User "gdpno69" has made a more intelligent algo.
but i have at the moment no time to implement that.

Thomas


Thanks Thomas.
@nattila. Hope I will be able to share with you all my version in 1-2 days. Still testing a bit with my previous flights logs an algorithm to detect circles before wind calculation, without FIFOs. Not sure whether this is better or worse...

@Thomas. Have you seen my previous post? Are you sure you have uploaded the right version?
Back to top
View user's profile Send private message
rod



Joined: 13 Jul 2009
Posts: 26

PostPosted: Mon Sep 28, 2009 2:24 pm    Post subject: Re: wind test Reply with quote

thomast wrote:

yes wind you can set before start,so s2f calculator works from the beginning correct.otherwise you must wait until the cms detects the windconditions.later you can overwrite this predefined values just by pressing 2 in the main cms.

The mathmodel was created from "Helmut Reichmann" a Sailplanepilot Champion.
So it work also for Sailplanes ,when the Polare is set to the Correct one.The Polare of a ask21 is already in the script.

For motorplanes it makes no sense to have the s2f stuff...

Have a nice Weekend,
Thomas


Hi Thomast:
Thank you very much for your answer and for your excellent theme. Very Happy

Greetings,

RSG.
Back to top
View user's profile Send private message
gdpno69



Joined: 03 Sep 2007
Posts: 27

PostPosted: Mon Sep 28, 2009 5:32 pm    Post subject: CMS for flying rel 1.2 Reply with quote

Version 1.2, released on 22.11.2009, is an improvement of version 1.1.
Functional
Version 1.2 adds a small graphical representation of real direction compared to apparent direction, due to wind influence. This gadget may be useful when visually targeting a goal, in presence of strong wind and in absence of close reference points which allow you to detect your drift.
Technical
Computations are made not more frequently than 1 s, so the CMS should work with GPS receivers updating more frequently.
The algorithm now uses much less computing CPU (the same as version 1.0): TB misses atan computation, so I implemented in HECL the bit-burst algorithm from Paul Zimmermann, whose paper is in internet, and is included in the doc directory, along with its test in Excel format. For the precision that I need, the algorithm is converging in a few (not more than 5-6) steps and is very very fast.



Screenshot0017.jpg
 Description:
New settings for the implementation of the MacCready theory.
 Filesize:  26.1 KB
 Viewed:  19312 Time(s)

Screenshot0017.jpg



Screenshot0018.jpg
 Description:
The new view with some statistics.
 Filesize:  20.57 KB
 Viewed:  19312 Time(s)

Screenshot0018.jpg



Screenshot0019.jpg
 Description:
Main screen
 Filesize:  38.73 KB
 Viewed:  18961 Time(s)

Screenshot0019.jpg



AirCMS_GDP_rel1.2.zip
 Description:
New documentation and code to unzip and insert into ui-profiles folder.

Download
 Filename:  AirCMS_GDP_rel1.2.zip
 Filesize:  258.99 KB
 Downloaded:  445 Time(s)



Last edited by gdpno69 on Sun Nov 22, 2009 2:11 pm; edited 2 times in total
Back to top
View user's profile Send private message
rod



Joined: 13 Jul 2009
Posts: 26

PostPosted: Mon Sep 28, 2009 7:23 pm    Post subject: Re: CMS for flying Reply with quote

gdpno69 wrote:
Version 1.0, released on 28.09.2009, inherits most of code from the version that thomast relased on 25.08.09, except for the following changes. Feel free to merge it with Thomas other version to enhance functionality.
A big thank you to kruch, who implemented my desires, and thomast, who made big part of the initial work and gave me many hints in private emails.
Version is still under test. Please report any bugs.
Functional
Completely rewritten wind algoritm, now computes wind only when in circles (it should be more precise)
Completely rewritten S2F algorithm, now allows to set expected vertical speed of next thermal, in order to implement MacCready theory (the rings that someone used to put on their circular variometer)
Modified CMS, to include graphical representation of wind, relative to flyer course, and a more precise glide ratio
Inserted additional polar curves for paragliders.
Complete documentation in Word format (created with OpenOffice).
Polar curve computation in Excel format (OpenOffice).
Technical
Algorithms are now a bit faster and precise, as do not use anymore queues, nor iterations. The only queue to compute average values (e.g. climb) is implemented as a circular buffer and does not need copies anymore.
With light off (but radio and bluetooth circuits on), in NMEA simulation mode, measured consumption on N95 8Gb is about 0,17W and about 20% CPU, while the latest algorithm released by thomast (as of 28.09.2009) is about 0,22 W and 25% CPU. Unfortunately, battery duration is not considerably higher when you turn light on to watch display, as consumption arrives at 0,38W. In this case, estimated duration is about 12 hr. This estimation should change if you communicate with BT antenna, or use internal receiver (very energy hungry), or turn off GSM circuits.


Hi Thomast and Gdpno69:
Very Impressive Very Happy .... Thank you all for the enormous interest and great professionalism.

Greetings,

RSG.
Back to top
View user's profile Send private message
nattila



Joined: 21 Sep 2009
Posts: 29

PostPosted: Wed Sep 30, 2009 8:32 am    Post subject: Re: wind test Reply with quote

thomast wrote:

just rename the Files:
cms.air15.xml and cms.air16.xml
to
cms.air75.xml and cms.air76.xml


Unfortunately it didn't help. I'm going to try it with a 1hz receiver unit.
Back to top
View user's profile Send private message
gdpno69



Joined: 03 Sep 2007
Posts: 27

PostPosted: Wed Sep 30, 2009 3:54 pm    Post subject: Re: wind test Reply with quote

nattila wrote:
thomast wrote:

just rename the Files:
cms.air15.xml and cms.air16.xml
to
cms.air75.xml and cms.air76.xml


Unfortunately it didn't help. I'm going to try it with a 1hz receiver unit.


Give a try to my version. However, even if it works, it could eat up 100% CPU. You'd better use 1Hz receiver.
Back to top
View user's profile Send private message
nattila



Joined: 21 Sep 2009
Posts: 29

PostPosted: Fri Oct 02, 2009 3:38 pm    Post subject: hcl Reply with quote

Hi, what kind of language was used to create air.hcl? I would try to modify the proc air::onLocationUpdated {} function just only run on every fifth signal. Could you help for me? I never give up. Smile
Back to top
View user's profile Send private message
gdpno69



Joined: 03 Sep 2007
Posts: 27

PostPosted: Fri Oct 02, 2009 4:05 pm    Post subject: Re: hcl Reply with quote

nattila wrote:
Hi, what kind of language was used to create air.hcl? I would try to modify the proc air::onLocationUpdated {} function just only run on every fifth signal. Could you help for me? I never give up. Smile

HECL scripts. If you are familiar with programming languages you will do it quickly. Surround it with a big 'if' where you check $cms::time. I cannot find remainder operator in HECL. Probably, you could do something like:
if {= [- $cms::time [* 1000 [/ $cms::time 1000]]] 0}
{<the whole function body>}
I have not tested this command, so I am not sure there is no syntax/semantic error Sad
Back to top
View user's profile Send private message
thomast



Joined: 16 Jan 2009
Posts: 101

PostPosted: Tue Oct 20, 2009 10:52 am    Post subject: Re: Testing Reply with quote

Hallo gdpno69,

gdpno69 wrote:
Hi Thomas,
Climb goes out one digit and invades speed.
Wind arrow indicates wind direction but 180-degree rotated.

in the code the script convert a float to 2 integers.
one contain the left part of the comma,
one the right part.and the rigth part now giving 2 digits(old 1digit).
i need this for the geforce theme.
in the air theme i modify the xml to "mask out" one digit again.

the wind arrow shows where the wind came from.
at south wind it shows to south.i prefer that, but i know its not clear.

i tested your version, looks nice specially the wind symbol.there its clear what is meant.

on my phone sometimes ldreq is sometimes Truncated.see screenshot.
the reason is that tb has a bug,that float values are shown a half digit to much to the left side.
i think its better to convert that float value to 2 integers.
my last version has two procs for this.
air::get_i () left part
air::get_d () right part


Quote:
in NMEA simulation mode, measured consumption on N95 8Gb is about 0,17W and about 20% CPU, while the latest algorithm released by thomast (as of 28.09.2009) is about 0,22 W and 25% CPU.

on my 6220c it takes 5% vs 6%.it is important that NEP is running in background otherwise itself creates 15% cpu load additionally...

Quote:
I am working to derive fast formulas to compute the difference of two vectors.Meanwhile, do you agree on my view?

i must think about it,but later.
at the moment i have some toothache Sad .so its not possible to think about such complicate things...

Happy Landing,
Thomas



Screenshot0106.png
 Description:
ld to long
 Filesize:  6.77 KB
 Viewed:  19428 Time(s)

Screenshot0106.png


Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic   Reply to topic    TrekBuddy Forum Index -> Themes All times are GMT
Goto page Previous  1, 2, 3, 4, 5, 6  Next
Page 3 of 6

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You cannot download files in this forum


Powered by phpBB © 2001, 2005 phpBB Group