Mangling of U.S. Contest Flight Logs?

[[email protected]]

I was recently asked to contribute some flight log analyses for an upcoming presentation by John Mittell for his club on his first-time competitor experience at the Seniors. I downloaded flight logs from the SSA site and started going through them in See You, and I noticed that thermals looked very funny. Instead of pentagons or hexagons or circles, I was seeing jagged back-and-forth lines. I looked at the fix times in SeeYou and discovered that fixes were 12 sec apart instead of 4 sec (or 1 sec for some ClearNav files), making thermals basically impossible to analyze, and causing See You to badly miscalculate average climb rates. When I looked directly at the IGC files with a text reader, I saw the following note embedded in the file:

LXXX
LXXX Note: This IGC file has been condensed by Winscore for Internet spectators.
LXXX Contact the contest organizers for the complete version of this file.
LXXX (The complete version was used for contest scoring calculations..)
LXXX

In this day and age of ubiquitous high-speed internet, this makes little sense, and compression to this degree makes them completely useless for their intended purpose - i.e. making them available to other interested parties for education and training purposes.

I'm not sure who to approach about this, but how do we get this policy of deliberate filicide (my new word for the day) reversed?

Frank (TA)

[Ron Gleason]

On Monday, 26 March 2012 19:02:01 UTC-6, wrote:
> I was recently asked to contribute some flight log analyses for an upcoming presentation by John Mittell for his club on his first-time competitor experience at the Seniors. I downloaded flight logs from the SSA site and started going through them in See You, and I noticed that thermals looked very funny. Instead of pentagons or hexagons or circles, I was seeing jagged back-and-forth lines. I looked at the fix times in SeeYou and discovered that fixes were 12 sec apart instead of 4 sec (or 1 sec for some ClearNav files), making thermals basically impossible to analyze, and causing See You to badly miscalculate average climb rates. When I looked directly at the IGC files with a text reader, I saw the following note embedded in the file:
>
> LXXX
> LXXX Note: This IGC file has been condensed by Winscore for Internet spectators.
> LXXX Contact the contest organizers for the complete version of this file.
> LXXX (The complete version was used for contest scoring calculations.)
> LXXX
>
> In this day and age of ubiquitous high-speed internet, this makes little sense, and compression to this degree makes them completely useless for their intended purpose - i.e. making them available to other interested parties for education and training purposes.
>
> I'm not sure who to approach about this, but how do we get this policy of deliberate filicide (my new word for the day) reversed?
>
> Frank (TA)

There is a parameter in WINSCORE that allows the scorer, when up loading IGC files, to compress them and to specify the minimum time between fixes.

The scorer can re-upload if they so desire, have a better internet connection and have the time. Last time I was at Seminole "ubiquitous high-speed" access was not available, rather they only have satellite based internet access. Many rural areas do not have the same access to the internet as you city folks.

You are jumping to conclusions and implicating hard working volunteers. There is not deliberate conspiracy issues occurring here. Take off your tainted, rose colored glasses and try to understand reality and stop beating on the volunteers that make competitions happen.

Ron Gleason

[Sean Fidler]

I agree file compression is silly in this case.

[[email protected]]

Ron,

Hey, thanks for the information, and the personal attack - I haven't gotten that kind of rise out of anyone since I since last year, when I was censored at your Logan mountain-survival contest for using the U word ;-).

Lets see: The zipped file size for March 16 was 655KB, which represents about 10-15 seconds at normal DSL rates (500-1000Kbs upload speed). With un-mangled files, the upload, done once per day, would have taken about 3 times longer, or about 45 seconds. Stop me when I go off track here, but I fail to see the advantage of saving 30 seconds (or even 3 minutes)/day while rendering ALL the files essentially useless for analysis.

The Seniors stopped using satellite-based internet connectivity several years ago, but we still have problems with slow connections there when everyone tries to get on at once. Many of us have started using the Ubiquitous (another U word!) USB cell modems - my Verizon USB modem cost me all of $50 for 1GB data with DSL-or-better upload/download rates. However, even on bad days at the Seniors, I'm not really sure how the difference between 15 seconds and 45 seconds of upload time on a once/day basis would be significant.

The scoring at this year's Seniors was done by Florin Alexandrescu. This post was not, and is not, any reflection on the superb job he did, especially for a first-time scorer. It is, however, a call to stop compressing IGC files to the point of unusability in an attempt to save an insignificant amount of internet upload time. I would be more than happy to use my own USB modem bandwidth for this purpose at any contest I attend - just ask ;-).

Frank (TA)



> There is a parameter in WINSCORE that allows the scorer, when up loading IGC files, to compress them and to specify the minimum time between fixes.
>
> The scorer can re-upload if they so desire, have a better internet connection and have the time. Last time I was at Seminole "ubiquitous high-speed" access was not available, rather they only have satellite based internet access. Many rural areas do not have the same access to the internet as you city folks.
>
> You are jumping to conclusions and implicating hard working volunteers. There is not deliberate conspiracy issues occurring here. Take off your tainted, rose colored glasses and try to understand reality and stop beating on the volunteers that make competitions happen.
>
> Ron Gleason

[Ron Gleason]

If you want a change made document it and submit it to the rules committee prior to October of each year.

The WIINSCORE Users Manual for 2011 can be found here http://www.gfbyars.com/winscore/Winscore%20User's%20Guide.pdf

Go to page 40 for the guidelines in the manual for uploading IGC files. Here is a paragraph from that section

Minimum time between fixes - Typically, flight data recorders save a fix at 4
second intervals. However, if the .zip file contained every fix that each pilot
made for the entire day, the .zip would be too large to send efficiently.
Therefore, the scorer can set a new time interval that is used to create the .zip. If the time interval entered here is larger than what the pilot used during his flight, then some of his fixes will be ignored when his flight is put into the .zip. Using a short interval will give more detail to the Internet viewer, but will increase the .zip size and using a longer interval will give less detail to the internet viewer, but will decrease the .zip size. Testing has show that a 15 second interval to be a good tradeoff between resolution and file size.

Whining on RAS does not necessarily get things changed, in my experience.

Ron

[Simon Taylor[_2_]]

There's a bit of confusion here. ZIP files are lossless, that is to
say that compressing files into ZIPs and then decompressing
them never alters the files.

On the other hand, Winscore has a function it describes as
'condensing' - looking at the code it seems a position fix is
discarded if less than n seconds have elapsed since the last fix,
and that's the feature that's causing the loss of information.

It's difficult to estimate what effect that will have on filesize. I
think many IGC loggers use two separate rates for position fixes
- slow during cruise and fast during climb. Processing these files
by filtering fixes by frequency alone will have a savage effect on
the circling position data and leave the cruise trace
comparatively untouched.

At 15:23 27 March 2012, wrote:
>Ron,
>
>Hey, thanks for the information, and the personal attack - I
haven't
>gotten=
> that kind of rise out of anyone since I since last year, when I
was
>censor=
>ed at your Logan mountain-survival contest for using the U
word ;-).
>
>Lets see: The zipped file size for March 16 was 655KB, which
represents
>ab=
>out 10-15 seconds at normal DSL rates (500-1000Kbs upload
speed). With
>un-=
>mangled files, the upload, done once per day, would have
taken about 3
>time=
>s longer, or about 45 seconds. Stop me when I go off track
here, but I
>fai=
>l to see the advantage of saving 30 seconds (or even 3
minutes)/day while
>r=
>endering ALL the files essentially useless for analysis.
>
>The Seniors stopped using satellite-based internet connectivity
several
>yea=
>rs ago, but we still have problems with slow connections there
when
>everyon=
>e tries to get on at once. Many of us have started using the
Ubiquitous
>(an=
>other U word!) USB cell modems - my Verizon USB modem
cost me all of $50
>fo=
>r 1GB data with DSL-or-better upload/download rates.
However, even on bad
>=
>days at the Seniors, I'm not really sure how the difference
between 15
>seco=
>nds and 45 seconds of upload time on a once/day basis would
be significant.
>
>The scoring at this year's Seniors was done by Florin
Alexandrescu. This
>po=
>st was not, and is not, any reflection on the superb job he did,
>especially=
> for a first-time scorer. It is, however, a call to stop
compressing IGC
>f=
>iles to the point of unusability in an attempt to save an
insignificant
>amo=
>unt of internet upload time. I would be more than happy to
use my own USB
>=
>modem bandwidth for this purpose at any contest I attend -
just ask ;-).
>
>Frank (TA)
>=20
>
>=20
>> There is a parameter in WINSCORE that allows the scorer,
when up loading
>=
>IGC files, to compress them and to specify the minimum time
between fixes.
>>=20
>> The scorer can re-upload if they so desire, have a better
internet
>connec=
>tion and have the time. Last time I was at Seminole
"ubiquitous
>high-speed=
>" access was not available, rather they only have satellite
based internet
>=
>access. Many rural areas do not have the same access to the
internet as
>yo=
>u city folks.
>>=20
>> You are jumping to conclusions and implicating hard working
volunteers.
>=
>There is not deliberate conspiracy issues occurring here. Take
off your
>ta=
>inted, rose colored glasses and try to understand reality and
stop beating
>=
>on the volunteers that make competitions happen.=20
>>=20
>> Ron Gleason
>

[soartech[_2_]]

> Whining on RAS does not necessarily get things changed, in my experience.
>
> Ron

No, but it's sometimes a good place to find out WHO to complain to. :-)

[Steve Leonard[_2_]]

I agree with you, Frank, in that I would like to see the whole flight,
as logged. And that there is likely little time saved by having
Winscore "reduce" the file sizes. But, how badly is the climb rate
mis-calculated by dropping some data points? Unless the climb is just
a turn or two, where the climb rate could have been mostly the zoom
entrance, are you really losing much? What matters is bottom to top,
unless you are trying to show "He shifted over a bit here, and look
how much faster he went up." In that case, can you tell the shift
from the drift? And if what you are intersted in is bottom to top, if
you have the end points, do you lose anything by not having the
middle? I can see value in showing "Thermal started to drop off here,
so you should have left and not taken the next three turns." But,
even at 15 seconds between samples, this would still be apparent.
Less points around the turn will also damp out the zooms and dives
some pilots make when thermalling, thus actually giving you a better
indication of actual climb rate.

Note to scorers: Please look at Winscore and set it up so the zip
file it creates will go ahead and keep more of the data points. Help
keep Frank happy. And a lot of the rest of us, too. :-) Hard drive
space is cheap. Help me fill mine with flight files.

Steve Leonard
ZS

[Greg Arnold]

On 3/27/2012 9:25 AM, Ron Gleason wrote:
> If you want a change made document it and submit it to the rules committee prior to October of each year.
>
> The WIINSCORE Users Manual for 2011 can be found here http://www.gfbyars.com/winscore/Winscore%20User's%20Guide.pdf
>
> Go to page 40 for the guidelines in the manual for uploading IGC files. Here is a paragraph from that section
>
> Minimum time between fixes - Typically, flight data recorders save a fix at 4
> second intervals. However, if the .zip file contained every fix that each pilot
> made for the entire day, the .zip would be too large to send efficiently.
> Therefore, the scorer can set a new time interval that is used to create the .zip. If the time interval entered here is larger than what the pilot used during his flight, then some of his fixes will be ignored when his flight is put into the .zip. Using a short interval will give more detail to the Internet viewer, but will increase the .zip size and using a longer interval will give less detail to the internet viewer, but will decrease the .zip size. Testing has show that a 15 second interval to be a good tradeoff between resolution and file size.
>
> Whining on RAS does not necessarily get things changed, in my experience.
>
> Ron



I don't think that language requires the scorer to use a longer time
interval. The language is "the scorer can set a new time interval". So
I don't see why you would go to the rules committee.

[Ron Gleason]

I just installed the 2012 version of WINSCORE.

The defaults I found for Flight Log Transfer are:

Condense IGC files - OFF

If I activated 'Condense IGC Files' then the default value for minimum time between fixes is 4 and the 'Include only the top contestants' value was 15

FWIW

Ron Gleason

[Guy Byars[_4_]]

>
> I'm not sure who to approach about this, but how do we get this policy of deliberate filicide (my new word for the day) reversed?
>
> Frank (TA)

Who to approach about this? Maybe a polite note to the author of
Winscore requesting the program defaults be changed might be a good
start. But of course, it is more satisflying to rant on
rec.aviation.soaring instead.

Also please include in your file size analyais the sizes of IGC files
from contestants who set their logging interval at 1 second, then turn
the logger on at the hotel room in the morning and then turn it off
just before bed at night. Several of those 1MB+ files will greatly
increase the upload times and was one of the reasons the *OPTION* to
reduce the upload files was provided to the scorers.

Guy Byars