Cellular forums Home > Archive > GPS > September 2006 > GPS Trackmaker multiple bugs??









You are viewing an archived Text-only version of the thread. To view this thread in it's original format and/or if you want to reply to this thread please [click here]

 

Author GPS Trackmaker multiple bugs??
Mike Scott

2006-09-25, 4:33 am

Has anyone else noticed the following bugs? I'm using OSGB coordinates,
1936 datum (I hope the latter's the correct thing - I'm very hazy here!)

1. Grid numbering is wrong. The eastings grid labelling runs
47000 48000 49000 60000 61000 .... 68000 69000 60000 61000 ....
^^^^^^^^^^ ^^^^^^^^^^
although the indicator at the top of the page gets it right.

2. There seems to be a major problem with background (image) map
calibration when the map crosses zone boundaries. The map is positioned
wrongly in non-obvious ways.

3. I sometimes get a 'runtime error 6; overflow' message box and the
program crashes out; these seems associated with panning across the map
but depends on the scale set - I can't offer more detail.

I'm not sure if #2 and #3 are related to the difference between 'british
grid' and 'true british grid' (again, I'm rather hazy as to why this
distinction should be needed).

(This is 13.0.163, free version)

--
Please use the corrected version of the address below for replies.
Replies to the header address will be junked, as will mail from
various domains listed at www.scottsonline.org.uk
Mike Scott Harlow Essex England.(unet -a-t- scottsonline.org.uk)
David Lee

2006-09-25, 4:33 am


"Mike Scott" <usenet.11@spam.stopper.scottsonline.org.uk> wrote in message
news:tAMRg.33246$cx.13899@newsfe1-gui.ntli.net...
> Has anyone else noticed the following bugs? I'm using OSGB coordinates,
> 1936 datum (I hope the latter's the correct thing - I'm very hazy here!)
>
> 1. Grid numbering is wrong. The eastings grid labelling runs
> 47000 48000 49000 60000 61000 .... 68000 69000 60000 61000 ....
> ^^^^^^^^^^ ^^^^^^^^^^
> although the indicator at the top of the page gets it right.
>
> 2. There seems to be a major problem with background (image) map
> calibration when the map crosses zone boundaries. The map is positioned
> wrongly in non-obvious ways.
>
> 3. I sometimes get a 'runtime error 6; overflow' message box and the
> program crashes out; these seems associated with panning across the map
> but depends on the scale set - I can't offer more detail.
>
> I'm not sure if #2 and #3 are related to the difference between 'british
> grid' and 'true british grid' (again, I'm rather hazy as to why this
> distinction should be needed).
>
> (This is 13.0.163, free version)


I've rather given up on GPSTrackmaker because it's so flaky! I run my Win
XP computer without Administrator privileges for general use and GPSTM just
can't cope with this. For instance printing appears to be totally
impossible unless you run as an administrator!

However re your problems...

There is a huge problem with the implementation of "British National Grid"
in GPSTM. The UK is mapped in a single continuous transverse mercator zone
and the lettered 100km square "zones" are merely a convenience.
Unfortunately GPSTM treats them as though they are separate UTM zones, with
different datum parameters. The trick is to select "British Uniform Grid",
which uses the absolute, completely numeric, grid reference system and
correctly treats the UK as a single zone.

As I understand it, in "Normal" mode the screen is linear in Lat/Lon so grid
lines will be curved, due to meridian convergance. The advantage is that
more than one zone can be displayed at a time but rectangular map images
will not accurately overlay the grid system. With "True Grid" selected the
screen is linear in the selected grid system but one of the disadvantages is
that adjacent URM zones cannot be displayed at the same time. Because UK OS
"zones" are treated as UTM zones, this means that the program can't cope
with more than one "zone" at a time so that you can't use calibration points
lying within more than one lettered "zone". You must convert the positions
of the calibration points to fully numeric coordinates and calibrate using
"British Uniform Grid" in "True Grid" mode.

I can't comment on the gridline labelling anomaly or the crashes I'm afraid.

David


David Lee

2006-09-25, 7:33 am

Excuse top posting but multiple level in-line replies get very confusing!

The alternative freebie program is Ozi Explorer - however the design
philosophy is different. The free version of Ozi is a crippled version of
the main retail product so that you will probably need to pay for it unless
your needs are very simple. GPSTM, on the other hand, is fully functional
and the pro version adds special non-essential features (such as spreadsheet
compatibility etc) which most people can easily live without - in fact the
author actually tells you (at least he did last time I looked) that you
probably shouldn't buy the pro version.

However, I haven't replaced GPSTM but simply changed the way I work to avoid
using it. I used to upload tracks to my GPS but now I simply upload
waypoints and rely on the topography from Dave Storey's free UK contour maps
(available from the Scottish Mountaineering Club). I also used to use GPSTM
to add tracks and waypoints to maps for ecological surveys but because it is
now such a b****** to use I have given up and do all the work manually using
a CAD package. One of the other horrors of GPSTM is that it is impossible
to format the output on a page - no previewer and PDF creator won't work
with it - so you have no idea exactly what area is going to appear on the
paper and you can easily waste reams of paper, so I would have had to copy
and paste into another package for output in any case.

The two representations of UK OS grid references are identical with the
exception of the first digits of the Easting and Northing so the datum is
the same in both. The full national grid reference is uniquely defined in
metres E & N from the reference point (the false origin). This is unwieldy
for local use so that the grid system used on paper maps is based on 100km
squares. The 100km square is defined by a two letter code and the numerical
part of the grid reference repeats in each of the squares. For example, the
six digit grid reference of my home is 765 472. However to uniquely define
this position nationally you must also state which 100km square I live in
and you can either quote this as SO 765 472 or as 3765 2472. Thus the SW
corner of Grid Square SO is 300km east and 200km north of the false origin.
Similarly the point SV 765 472 would lie in the most south westerly 100km
square, corresponding to 0765 0472, and representing an area of the Atlantic
Ocean just north of the Scilly Isles.

In order to treat OS national grid zones correctly GPSTM would require code
specific to the UK in order to correctly cope with the alphabetic notation.
The compromise is to treat them as though they were individual UTM zones,
which means that as far as the program is concerned they all have different
datums and so cannot be displayed together. I would have hoped that the
author would have felt it worth the effort to do the job properly and fully
support the UK National Grid, but that was his decision and, unless you have
paid for the pro version you can't really complain!

I did send a bug report to the author a while ago (the error message
actually told me to!) and he did reply explaining that it was an issue
requiring installation with Administrator privileges. There are now so many
issues of this nature that frankly I just can't be bothered to write them
all up and send in another bug report!

David

"Mike Scott" <usenet.11@spam.stopper.scottsonline.org.uk> wrote in message
news:l7ORg.43326$G72.19492@newsfe3-gui.ntli.net...
> David Lee wrote:
> ...
>
> An interesting observation. Are there any other freebie products out
> there with similar functionality?
>
>
> Now that makes some sort of sense, thank you. Since the "BUG" (!!! :-) )
> and the OS grid references only seem to differ in their printed
> representation (I hope - am I missing something? I've written a perl
> script to do the conversion as I understand it to be) one can only wonder
> why they're handled so differently in Trackmaker.
>
> Presumably both use the OSGB 1936 datum rather than WGS84??
>
>
> Maybe I ought to report those to the author, although he doesn't seem to
> encourage contact by the 'freebie' users, understandably.
>
> --
> Please use the corrected version of the address below for replies.
> Replies to the header address will be junked, as will mail from
> various domains listed at www.scottsonline.org.uk
> Mike Scott Harlow Essex England.(unet -a-t- scottsonline.org.uk)



Mike Scott

2006-09-25, 10:33 am

David Lee wrote:
> Excuse top posting but multiple level in-line replies get very confusing!


I know; some people are picky - I take the view that I do what's (IMO)
clearest.

Thanks again for the comments; I'll take a look at the maps you suggest,
as I've not seen these before.

I'll think a bit about mailing the gtm author! Hardly difficult to
convert between the systems - I think it's a couple of dozen lines of
perl, and that's just a crude first cut.


--
Please use the corrected version of the address below for replies.
Replies to the header address will be junked, as will mail from
various domains listed at www.scottsonline.org.uk
Mike Scott Harlow Essex England.(unet -a-t- scottsonline.org.uk)
David Lee

2006-09-25, 10:33 am

Mike Scott wrote...
> Thanks again for the comments; I'll take a look at the maps you suggest,
> as I've not seen these before.


Assuming you mean the contour maps then you'll find them at
www.smc.org.uk/contour_maps2.htm. However they are Garmin .img format and
so only for use in Garmin receivers and MapSource. If you have a Garmin
then they are great - the maps are "transparent" so you can load them along
with Garmin MapSource products and they appear as overlays. Also they were
derived from NASA SRTM data and so have no copyright issues. If you have
any problems or questions then post them here since the author also goes by
the name of "GSV Three Minds in a Can" and is a regular and very helpful
contributor to this forum.

David


Holger Issle

2006-09-25, 12:33 pm

On Mon, 25 Sep 2006 10:24:17 GMT, Mike Scott wrote:

> An interesting observation. Are there any other freebie products out
> there with similar functionality?


Not free, but there are 3 mapping packages for the PC: Ozi Explorer,
Fugawi and TTQV. The latter is designed in Germany and features very
good support. It is available as a 25 day test version of their
website (www.ttqv.de). Give it a try, the program is available in
english language. The license model is at various layers... you may
only need the lowest level.
--

Ciao,
Holger (GUS-KOTAL, GUS#1100)

90-92 Honda CB400 10 Mm | 93-95 Yamaha TDM 850 26 Mm
95-97 KTM 620 LC4 13 Mm | seit 97 BMW R1100GS 50 Mm (Die Renndrecksau!)

cu @ http://www.issle.de
LinkBot





Other Archives: Real Estate forum archive | Web Design archive | Software support archive | PC Hardware reviews archive | Medical topics archive

Copyright 2004 - 2008 cellphonetopics.com