On checking the new version of GPSResults using Dan’s best track at Pt Walter (he wore 3 gps units) it gave a surprisingly lower value than earlier assessment with Realspeed which apparently was on trackpoint rather than Doppler setting. Anyway as the top 3 all used Navi units with Doppler I was able to analysis their tracks with both programs and the averaged 2 sec results rearranged the top of the leader board. I think in future, as it is getting close at the top it is time to get a bit more accurate and I request that those that have a top ten speed send me the track and I’ll put the averaged 2 second Doppler as the max speed for this table.
*Speeds are Display on Garmin
And 2 sec Doppler on Navi/GT-11
** Trackpoint ranking
Rider.....GPS Display*...Location.......Date………..Trackpoint 2 sec GPS speed
Slowboat......41.30.......Pt Walter......30.07.2007..**……Dan………..…….. 41.75 knots
Dan.............41.15.......Pt Walter......26.01.2008..**……Slowboat…..…….41.72 knots
25...............40.98.......Hardie’s.........30.12.2007
Yoyo............40.09.......Pt Walter.......26.03.2007
Elmo............39.8....... Hardie’s..........30.12.2007
Claude.........38.9.........Hardie’s..........30.12.2007
Snides8........38.7........Hardie’s..........30.12.2007
Robbie..........38.6........Pelican Pt.......31.12.2005
Hardie..........37.7.........Hardie’s..........04.01.2008
AdamL..........37.55………Hardies.........26.01.2008
TimS............37.5........Hardie’s..........30.12.2007
ChrisA..........37.3.........Woodies.........31.12.2007
Bender.........37.3.........Hardie’s..........27.10.2007
JohnN...........37.2........Hardie’s..........30.12.2007
MattL…………..37.0……….Pt Walter………..26.01.2008
Izymeister.....36.6.........Melville..........07.08.2006
Jesper..........36.0.........Melville...........03.10.2006
Pointman…....35.91.......Melville …..….…14.01.2008
Somepunk….…35.8……….Lake P Bar.…….30.07.2007
Sam……………..35.7……....Hardies…………..04.01.2008
Nebbian…….…35.65………Hardies……….….30.12.2007
Decrepit........35.3........Hardie’s..........30.12.2007
Lao shi……….…35.2…….…..Pt Walter....….26.01.2008
BigDog………...34.9………..Melville…………..26.01.2008
Pugwash.......34.2.........Melville..........29.04.2007
Magna……...…34.0……....Pelican Pt…......??.??.2007
Graceman.....33.94.......Woodies..........21.12.2007
Bigdog…………33.81………Melville…………..05.01.2008
4XL.............33.8.........Pt Walter........09.03.2007
Pauly………....33.5…….....G Bay………......17.10.2007
RichardS…..…33.0…….....Mel??.............06.08.2007
Whitey…………36.9…………Pt Walter………..03.02.2008
Bugs……….....32.9…….....Hardie’s….......23.06.2007
Nelson……..…32.9…………Melville…………..26.01.2008
Man of………..32.0………...Hardies………….30.12.2007
Oli2…………....31.95…......??????...........21.07.2007
Mineral.........31.7.........Hardie’s.........08.12.2007
Liz(f)............31.56..........????............31.12.2007
Gadget……...…31.5……....Safety Bay….…06.01.2008
Lenny...........31.4.........Hardie’s.........01.07.2007
Bean…………….31.4……….…Hardie’s…………08.12.2007
AndrewN……...31.1…….....????????........08.09.2007
SimonF…….....30.6….…....Mandurah.......21.07.2007
Windrider………29.9…..…….Melville………..26.01.2008
TonyC………....29.65…......Hardie’s…...….05.01.2008
Luff………...…..29.3……......?????............18.10.2007
DavidS…….....29.2…….......????.............26.08.2007
Grumps…….....29.2……......????............17.10.2007
ChrisP…………..29.2………….Hardies…………..26.01.2008
CraigM………....29.1…..….…?????? ….......20.`0.2007
PaulK............28.6.........Melville.........15.03.2007
MoiraD(f).......28.4.........Mandurah......15.09.2007
Daddycool......28.3.........Melville..........27.03.2007
Mickey(f).......27.8.........Gnaraloo.......18.10.2007
ClintR…………...27.78……….????????........06.01.2008
MintoxGT….....27.67……...Melville??......14.01.2008
Binny(f).........27.3.........Mandurah......03.12.2007
Jet…………………27.2…………Safety Bay…….??????2007
GrahamF……....26.8…….....??????.........12.08.2007
WilliamD…….....26.4……......Mandurah…..21.07.2007
TraceyP(f)…....26.36……....Woodies….…19.11.2007
Propercarter....25.0………...Melville….…..26.08.2007
Natalie(f)...…..21.5………....Woodies.…....03.12.2007
Coala Bear…….15.6………….Melville…………26.01.2008
your just trying to annoy severne's marketing team arent you lol
"New speed record for the Swan River in Perth - West Australia, by Severne team rider KA 18 Dan Engdal on the new production Code Red 6.7"
-on the front page for severne.... (for now)
Damn, there goes my chance of "borrowing" a CR from the Severne guys. :-))
Actually it is a bit more complicated than I had hoped. For example Dan had 3 gps units. Do you average the results or take the best? One had 41.24 on the dial that I saw the other 45.65 as Dan still had his SP tracks on it. Like a lot of new owners of the Navi he was still trying to figure it out as he states himself on GPSTC " Not Doppler data - next lot will be when I figure it out"
One other problem is that the 3 different programs for analysis give different results and also are much more favourable for Trackpoint compared to Doppler.
In Trackpoint Dan does indeed comes out on top by 5/1000 Knot on the 1 second peak and 2/100 of a knot on 2 sec with 41.747 knots to Chris' 41.721. But that was his fastest gps. If you average Dan’s 2 Navis then Chris comes out on top!
So I guess the Severne guys are right on that score. The fastest gps speed on the Swan is Dan's 41.747 when using GPSResults in trackpoint mode on 2 second.
Another problem is different gps units. Originally the Garmin Foretrex/Geko were the gps units most had and it's display was significantly more accurate than its trackpoint. Hence my use of it in the WA Speed Table. But its display Doppler speed is averaged over several seconds. Clearly the Navi has an advantage here as it updates every second. To keep things balanced between different gps units I used the 2 second value of the Navi as the most appropriate equivalent speed.
Finally, we are in a change over period. Some people are using Trackpoint analysis and some Doppler. It appears Doppler is more accurate and it is the intention of the gps-ss organization to go this way even though it does give less flattering numbers. I think it is important to be consistent so in future will use Doppler also for the WA S T but I understand there will be some confusion initially.
Elmo your Geko display is also valid as it is Doppler.
can of worms... the "5 second filter" on the garmin display was pushed around in the early days to explain the apparent delay in appearance of the result on the screen, and the difference between trackpoints and display. This difference is because the garmin display uses doppler data like the navi records, and not the inflated grid-affected trackpoints.
So in reality there is very little difference between the garmin display and the navi display. I've compared both and they are always very similar (no bias toward garmin or navi). And the navi display comes directly from the doppler data sequence which is recorded. So a 2sec Navi doppler result is probably more conservative than the garmin display speed... (the navi exhibits a similar lag on the speed readout to the garmin- indicating a similar level of filtering is likely applied).
Chris , looking at the NMEA sentences put out by the Garmin Foretrex and reading technical articals about the algorithims of different Chipsets it was mention in several places that the Foretrex updates every 2 seconds (which begs the question why you can change the recording to 1 second.)
Surprisingly, the trackpoint between Realspeed and GpsResults was almost identical (1/1000 diff) but the doppler was quite different between the 2 programs ( eg 41.326 & 41.276 = ave 41.302) . I would have thought it would be the other way around.
""So I guess the Severne guys are right on that score. The fastest gps speed on the Swan is Dan's 41.747 when using GPSResults in trackpoint mode on 2 second.""
Session by slowboat:
Date : Monday, March 26, 2007
Spot : Point Walter, WA, Australia
Board : Carbon Art Speed 44
Sail : KA Sails Koncept 5.0
Fin : Custom Custom 23 Speed 23
GPS type :
Average speed : 39.60 knots (40.0 39.8 39.5 39.4 39.3)
Max. GPS (display) : 41.3 knots
Max. 2 sec. (software) : 42.5 knots
100 m run : 42.4 knots
250 m run : 39.7 knots
500 m run : 38.3 knots
Distance : 60.0 km
Windspeed : 22 knots
Windgusts : 28 knots
now thats an oldish session, but im guessing for trackpoint 2sec that gives it to chris??..
anyway bring on full doppler rankings!! Far less confusing!!
My experience with a large amount of side by side testing with Foretrex and GT-11 is that the display max speeds are very, very similar. Most of the time they are within less than 0.1 knots and if anything there is a tendency for the Foretrex to more often show a higher max speed but rarely more than 0.15 knots.
I have always been under the impression the Foretrex updates the display speed every second. I watch it a lot when sailing and have never noticed 2 sec update.
So I would say if you rank on the max. display speed you are using a 1 second or instant speed. For a more reliable ranking I would go for a 10 second ranking and use Doppler. Of course this does not work with the Fortrex saved data so as long as you realise that the error margin of the display is in the order of 0.1 seconds, or a little more, and this makes it unreliable for rankings when the differences are less than 0.2, it seems like a reasonable way of doing a FUN ranking for mixed GPS types.
Just out of curiosity, how valid is it to talk about GPS speeds (in knots) with a precision of 3 (or possibly even 2) decimal places in time intervals of 2 seconds
After all, 0.001 knots is 0.00185 km/h… Or, sweet FA, as I see it… Using the top speeds under examination as an example:
Knots......km/h......m/s.........m/2s
41.302....76.409....21.225....42.449
41.154....76.135....21.149....42.297
41.747....77.232....21.453....42.907
41.721....77.184....21.440....42.880
I am not that flash at maths, but I make that a relative difference of 0.152 m or 0.027 m (That’s a MASSIVE 3 mm!!!)… And a GPS is how accurate
Surely you need to be covering at least 1000 m in 2 seconds or recording over a larger time interval for the second or third decimal place to be significant particularly when you consider the accuracy of the ruler (gps)!
A very good point Pugs.
When Yann Mathet first wrote GPSAR he only produced results to 0.1 Knot or 0.1 Kmh.
There is a certain logic in this because he knew the accuracy of the GPS data was not better than this anyway.
When RealSpeed and GPS-Results came along they produced results to .001 knots or .001 Kmh. We knew this was probably considerably more than the accuracy of the data but one beneficial side effect of this was that we could see that the calculations in the two programs agreed to .001 knot which was reassuring. (At my request Yann later produced a special version of GPSAR which produced results to the same resolution so I could compare the results from it with the other programs. The results were reassuringly close and the very small differences only reflected the slightly different mathematical model of the shape of the earth that GPSAR used).
Another benefit is that we now could resolve event rankings to two decimal places. This did (and does) not mean they were always correct but still it was the best we could do at the time and GPS-SS is only a (serious ) fun ranking, not a World Record ranking.
When Tom Chalko had the insight to steer us in the direction of the Doppler data for speed measurement the margin of error got pretty close to the two decimal place resolution (and arguably better in ideal conditions) so the results to 3 decimal places when using Doppler data are getting a lot more meaningful.
Error is more probable over short time measurements and decreases with the number of samples. This is why we consider the 2 second ranking, even when Doppler data is used, as carrying far less credibility than say a ten second ranking, and the 5 x 10 sec average is and even better tool for a more credible ranking.
Tim, on Chris' track for that date his Navi had a hiccup just prior to the peak. I suspect it might have caused it to overshoot a little on the trackpoint. If you look at the smaller peak to the right and the larger one to the left you will see there is a much closer alignment between the trackpoint and doppler curves.
of course Chris also has a foretrex to back up his peak but I'll stick to the Navi results..
I just put down the data the software analysing programs spit out. But you are right. It is overkill. I'll round up to 2 decimal places in future.
Fair enough... interesting to see how the track looks around the peak... kinda lyk it compensated for the small dip with the peak.... perhaps satellite trouble??.. that said there are a few other old tracks of chris's with 42ish 2secs.... though probably garmin?...
Spot : Lake Preston, Australia
Board : Carbon Art Speed 50
Sail : KA Sails Koncept 5.8
Fin : Tectonics F1 Falcon 30
GPS type :
Average speed : 38.84 knots (40.6 38.9 38.5 38.2 38.0)
Max. GPS (display) : 41.0 knots
Max. 2 sec. (software) : 41.8 knots ****(could still be a 41.75 rounded up....)
Date : Tuesday, August 15, 2006
Spot : Point Walter, WA, Australia
Board : Hammerboard Speed 44
Sail : KA Sails Koncept 5.8
Fin : Custom Chris L 20 speed 20
GPS type :
Average speed : 39.10 knots (41.0 39.2 38.9 38.5 37.9)
Max. GPS (display) : 41.0 knots
Max. 2 sec. (software) : 42.1 knots
Date : Tuesday, September 27, 2005
Spot : Point Walter, WA, Australia
Board : Hammerboard Speed/Slalom
Sail : Severne SSR 5.6
Fin : Custom Speed 27
GPS type :
Average speed : 38.03 knots (39.7 38.3 38.0 37.7 36.5)
Max. GPS (display) : 40.8 knots
Max. 2 sec. (software) : 42.5 knots
Tim, you just don't let go do you. The answer is Nup, nup and nup.
Lake Preston
Navi.....Doppler = 40.86 , trackpoint = 41.27
Garmin..Display = 41.00 , trackpoint = who cares (grid effect)
Hammerboard at Pt W 2006 Garmin Display = 41.0 , trackpoint =.. see above
Hammerboard at Pt W 2005 Garmin Display = 40.8 , trackpoint =.. see above
Casual user observations - not [quite] an addict!!
Having carried both garmin and navi (in the same aquapac) for the last couple of months my experience is that Garmin max display speed is usually very close [less that 0.1 knot] to the navi max display speed. The navi max display speed is always the top end [high and low] of the doppler 2 sec max with realspeed. So I conclude that Navi display is doppler?
I seem to reacll reading somewhere that the garmin display is doppler too? Is this correct?
Would seem to hold given the alignment of my observations over about 10 sessions.
maybe the display max apple was the most comparable apple of all?
yoyo,
just had a closer look at the top of the table and MY BAD!!... i was thinking that the trackpoints 2sec for chris/dan listed were 'fastest WA session on track points only'... and not simply the 2 sec from fastest display session... so thts why i posted those...
perhaps i shud take more of a less typing more sailing aproach.lol.....