Select to expand quote

Macroscien said..

sailquik said..

Macroscien said..
I am still pleasently surprised how our GPS accurate are. Let assume following scenario.
We are going 40 ktn.
so 74.08 km/h
or 20.577 m/s !!
so during 1 second we do already 20 meters !!!

Let assume that our GPS has 5 meters precision to locate us on the globe. so for two measurements the possible margin of error could be 10 meters.
Translate it to speed now.

So our speed could be as high as 59ktn or as low as 20 ktn !!!

OK our GPS is running now 10Hz so distance shrink to 2.057 meters during one tick 1/10 of second but location precision remains the same - 5 meters !!

I don't even want to think what error margin could be.

OK lets claim that our GPS has 1 meter accuracy not 5.

So our speed measured 40 ktn may vary now
from 43.887 knots to 36.112 knots

This is the truth and the rest is up smart software and statistics.

The point I am trying to make that our measurements could be as bad as 3 full knots in worst case scenario but we fight here for every 0.001 knots !

BTW I love to be corrected because it mean i learned something new, so be kind and teach me something.
...
I know.
We trust our devices manufactures but the same may apply to Space and rocket science.
It you multiply error and software cleaning procedure to such cosmic distances it is not surprise that
rockets and landers on celestial bodies are crashing more often then success. Small error like ours here mean that you just landed your craft 1 meter below Moon surface successfully.

Well done Macro. You just very clearly described why it was such a huge improvement when we switched from using positional data to calculate our speeds, to Doppler derived data. We went from a potential error range of knots, to an error range of very small fractions of a Knot.

Positional accuracy is very important for the GPS guidance of lawnmowers and tractors.

We use the Doppler derived speeds which are many orders of magnitude better for measuring velocity. Speed Over Ground (SOG)

I recommend that you get up to speed on speed measurement and start with reading this paper:

nujournal.net/wp-content/uploads/sites/2/2021/02/SDOP4aa1.pdf

Thank you
Because I finally realized how our GPS works.
( Correct me if I am wrong)
We have indeed two different devices all in one box. One system is designed to establish our geographical position
and the second is simply speedometer.

Now to work out our current speed we could use both.

1. But the first will employ our primary school knowledge.
We learn our positions , work out distance and divide distance by time.
Done.

2.The second require our high school knowledge, and has nothing to do with our position on the map at all.
We just measure shift in frequency signal from satellites and we could work out our speed perfectly.
It seems that second method is more accurate for our needs.

So first will apply to all basic GPS and second to GPS with Doppler.

Now it would not be Marco post if I could not come with some invention or improvement !

I imagine that in the near future we will be using our GPS mostly at night !!!
Why?!
Because at night we could stream from our satellite not radio signal but laser generated light.
Light of passing satellite at night will be clearly visible to our optical sensors.
( Yep- astronomers already complain on night sky pollution by Elons Starlink satellites that are not even emitting any lights itself)
Due to much higher frequency of light versus radio waves accuracy of our measurements will be now thousands or million times better !
Done !

But Macro possibly will not last so long to see his invention implemented for our watersports GPS and I don't like personally to sail at night

Have you been drinking flood water again Macro

Select to expand quote

Macroscien said..>>>>
Due to much higher frequency of light versus radio waves accuracy of our measurements will be now thousands or million times better !
Done !

But Macro possibly will not last so long to see his invention implemented for our watersports GPS and I don't like personally to sail at night

Sorry macro, I think the limiting factor for accuracy isn't carrier frequency, but atmospheric distortion. this will affect light and radio wave similarly.

That's how RTK system improves accuracy, by measuring atmospheric distortion and cancelling it out.

Select to expand quote

decrepit said..

Macroscien said..>>>>
Due to much higher frequency of light versus radio waves accuracy of our measurements will be now thousands or million times better !
Done !

But Macro possibly will not last so long to see his invention implemented for our watersports GPS and I don't like personally to sail at night

Sorry macro, I think the limiting factor for accuracy isn't carrier frequency, but atmospheric distortion. this will affect light and radio wave similarly.

That's how RTK system improves accuracy, by measuring atmospheric distortion and cancelling it out.

So lets think . Equip each satellite with two light sources some space apart or use different length. Entalg photons? To work out those distortion. Still better then just one antenna on our GPS. But I think that light will win or even X ray with be next? Bandwidth is the key to ultimate accuracy.Ok I read your post below RTK. This is about position accuracy.
But according to my speculation before it is Doppler that give us accurate speed measurement not position logging.
We are not interested in exact positioning , maybe beside alpha. To make sure you did not miss entry gates by 5 cm.

Select to expand quote

Mr Keen said..

Macroscien said..

sailquik said..

Macroscien said..
I am still pleasently surprised how our GPS accurate are. Let assume following scenario.
We are going 40 ktn.
so 74.08 km/h
or 20.577 m/s !!
so during 1 second we do already 20 meters !!!

Let assume that our GPS has 5 meters precision to locate us on the globe. so for two measurements the possible margin of error could be 10 meters.
Translate it to speed now.

So our speed could be as high as 59ktn or as low as 20 ktn !!!

OK our GPS is running now 10Hz so distance shrink to 2.057 meters during one tick 1/10 of second but location precision remains the same - 5 meters !!

I don't even want to think what error margin could be.

OK lets claim that our GPS has 1 meter accuracy not 5.

So our speed measured 40 ktn may vary now
from 43.887 knots to 36.112 knots

This is the truth and the rest is up smart software and statistics.

The point I am trying to make that our measurements could be as bad as 3 full knots in worst case scenario but we fight here for every 0.001 knots !

BTW I love to be corrected because it mean i learned something new, so be kind and teach me something.
...
I know.
We trust our devices manufactures but the same may apply to Space and rocket science.
It you multiply error and software cleaning procedure to such cosmic distances it is not surprise that
rockets and landers on celestial bodies are crashing more often then success. Small error like ours here mean that you just landed your craft 1 meter below Moon surface successfully.

Well done Macro. You just very clearly described why it was such a huge improvement when we switched from using positional data to calculate our speeds, to Doppler derived data. We went from a potential error range of knots, to an error range of very small fractions of a Knot.

Positional accuracy is very important for the GPS guidance of lawnmowers and tractors.

We use the Doppler derived speeds which are many orders of magnitude better for measuring velocity. Speed Over Ground (SOG)

I recommend that you get up to speed on speed measurement and start with reading this paper:

nujournal.net/wp-content/uploads/sites/2/2021/02/SDOP4aa1.pdf

Thank you
Because I finally realized how our GPS works.
( Correct me if I am wrong)
We have indeed two different devices all in one box. One system is designed to establish our geographical position
and the second is simply speedometer.

Now to work out our current speed we could use both.

1. But the first will employ our primary school knowledge.
We learn our positions , work out distance and divide distance by time.
Done.

2.The second require our high school knowledge, and has nothing to do with our position on the map at all.
We just measure shift in frequency signal from satellites and we could work out our speed perfectly.
It seems that second method is more accurate for our needs.

So first will apply to all basic GPS and second to GPS with Doppler.

Now it would not be Marco post if I could not come with some invention or improvement !

I imagine that in the near future we will be using our GPS mostly at night !!!
Why?!
Because at night we could stream from our satellite not radio signal but laser generated light.
Light of passing satellite at night will be clearly visible to our optical sensors.
( Yep- astronomers already complain on night sky pollution by Elons Starlink satellites that are not even emitting any lights itself)
Due to much higher frequency of light versus radio waves accuracy of our measurements will be now thousands or million times better !
Done !

But Macro possibly will not last so long to see his invention implemented for our watersports GPS and I don't like personally to sail at night

Have you been drinking flood water again Macro

Yep. Plenty recently More to come. Next cyclone just behind the corner.

Bring it, certainly is entertaining

Select to expand quote

Macroscien said..

Apollo mission landed on the Moon perfectly with Armstrong at the helm overriding all computers on board.
50 years later the most recent most advanced Moon landers keep crashing because unable calculate distance correctly.
Have you see recent Japanese Moon landender landing on his head?

No ,not entirely correct . NASA just wanted to land on the moon and had a general location of the landing spot. It was up to the pilot to fine tune the exact spot.
Japan tested the tech to do it without a pilot or GPS and land at the exact location nominated. It worked except for the overshoot on the height which caused the bounce. Got to compare the same thing as both where very different projects and have zero relationship to each other other than the moon.

We need improvent accuracy GPS for our everyday life.
You want to self park your vehicle within cm not meters.
Or hide a stash of money in the jar underground in the garden.
I am reading about newest Mate 70 Chinese smart phone and nobody say how good GPS going to be?! Better cameras or bigger screens that is all they do worry about. For us all the best could be new smartphone with improved GPS not another knock of DIY watch. I am happy even to invest $35 for additional plastic waterproof bag for new smartphone.
New smartphone with Doppler could solve all our problem we worry about for years here.Maybe is all windsurfers and other watersport users send request to Samsung phone manufacturer and ask for Doppler?
Like millions of customers action asking manufacturer to do something they need? Global petition by all GPSTC and others?

Select to expand quote

Macroscien said..

Apollo mission landed on the Moon perfectly with Armstrong at the helm overriding all computers on board.
50 years later the most recent most advanced Moon landers keep crashing because unable calculate distance correctly.
Have you see recent Japanese Moon landender landing on his head?

Armstrong didn't override the computer, he provided input.

Select to expand quote

Macroscien said..
We need improvent accuracy GPS for our everyday life.
You want to self park your vehicle within cm not meters.
Or hide a stash of money in the jar underground in the garden.
I am reading about newest Mate 70 Chinese smart phone and nobody say how good GPS going to be?! Better cameras or bigger screens that is all they do worry about. For us all the best could be new smartphone with improved GPS not another knock of DIY watch. I am happy even to invest $35 for additional plastic waterproof bag for new smartphone.
New smartphone with Doppler could solve all our problem we worry about for years here.Maybe is all windsurfers and other watersport users send request to Samsung phone manufacturer and ask for Doppler?
Like millions of customers action asking manufacturer to do something they need? Global petition by all GPSTC and others?

You do realise doppler is already on smartphones, in addition to doppler speed accuracy estimates? I have an app that already records the data for post processing.

I love it when common practice is questioned. But the answer to the original question is actually very simple:

The precision use to report numbers should always be at least as high as the accuracy (and preferably higher).

We know that the current u-blox based units have an accuracy of 2 digits , with errors in the range of 0.05 being typical for 10 seconds. For longer categories, the estimated accuracy is even better - it's typically 3 digits (e.g. 0.003) for 1 hour. For the longer GPSTC categories, speeds absolutely should be reported with 3 digits, since reporting with only 2 digits would actually reduce accuracy.

For the 2 second category, where accuracy is more likely to be in the single-digit range (0.1 to 0.2 knots), it does seem questionable to report 3 digits. Statistically speaking, a speed of 29.997 +- 0.1 knots is not significantly different from a speed of 30.0 +- 0.1 knots. But reporting it as 29.997 knots actually gives a little bit more information: the chance that your team mate did hit 30 knots is slightly less than 50%. If the number was reported with 1 or 2 digit precision, and therefore rounded up to 30.00 knots, that would state that the probability that he hit 30 knots was exactly 50%. If he wants to be reasonably sure that he really got 30 knots over 2 seconds, he needs to get a result of at least 30.1 knots (or, more accurately, 30 knots plus the reported +/- error estimate for the 2 seconds).

Select to expand quote

Roo said..

Macroscien said..
We need improvent accuracy GPS for our everyday life.
You want to self park your vehicle within cm not meters.
Or hide a stash of money in the jar underground in the garden.
I am reading about newest Mate 70 Chinese smart phone and nobody say how good GPS going to be?! Better cameras or bigger screens that is all they do worry about. For us all the best could be new smartphone with improved GPS not another knock of DIY watch. I am happy even to invest $35 for additional plastic waterproof bag for new smartphone.
New smartphone with Doppler could solve all our problem we worry about for years here.Maybe is all windsurfers and other watersport users send request to Samsung phone manufacturer and ask for Doppler?
Like millions of customers action asking manufacturer to do something they need? Global petition by all GPSTC and others?

You do realise doppler is already on smartphones, in addition to doppler speed accuracy estimates? I have an app that already records the data for post processing.

So do we have the same precision and accuracy on our or any of smartphones as our Motion GPS?
I still have only older Samsung smartphone that hardly can catch up on motorway to find nearest intersection and directions sometimes too late after you just passed in the wrong.

Select to expand quote

boardsurfr said..
I love it when common practice is questioned. But the answer to the original question is actually very simple:

The precision use to report numbers should always be at least as high as the accuracy (and preferably higher).

We know that the current u-blox based units have an accuracy of 2 digits , with errors in the range of 0.05 being typical for 10 seconds. For longer categories, the estimated accuracy is even better - it's typically 3 digits (e.g. 0.003) for 1 hour. For the longer GPSTC categories, speeds absolutely should be reported with 3 digits, since reporting with only 2 digits would actually reduce accuracy.

For the 2 second category, where accuracy is more likely to be in the single-digit range (0.1 to 0.2 knots), it does seem questionable to report 3 digits. Statistically speaking, a speed of 29.997 +- 0.1 knots is not significantly different from a speed of 30.0 +- 0.1 knots. But reporting it as 29.997 knots actually gives a little bit more information: the chance that your team mate did hit 30 knots is slightly less than 50%. If the number was reported with 1 or 2 digit precision, and therefore rounded up to 30.00 knots, that would state that the probability that he hit 30 knots was exactly 50%. If he wants to be reasonably sure that he really got 30 knots over 2 seconds, he needs to get a result of at least 30.1 knots (or, more accurately, 30 knots plus the reported +/- error estimate for the 2 seconds).

Lovely.
But I need to still make some calculation to prove it or disprove.
For example you will be happy to claim victory because your 1 hour was 24.003 and your mate did only 24.000.
But it could be quite unfair is your victory is due to margin error and actually your mate did better and now lost due to technicality or our system.
This errors accumulate over time too.

I observe this quite often, we feel good beating somebody by 0.001 but is it really true?

Select to expand quote

Macroscien said..

Roo said..

Macroscien said..
We need improvent accuracy GPS for our everyday life.
You want to self park your vehicle within cm not meters.
Or hide a stash of money in the jar underground in the garden.
I am reading about newest Mate 70 Chinese smart phone and nobody say how good GPS going to be?! Better cameras or bigger screens that is all they do worry about. For us all the best could be new smartphone with improved GPS not another knock of DIY watch. I am happy even to invest $35 for additional plastic waterproof bag for new smartphone.
New smartphone with Doppler could solve all our problem we worry about for years here.Maybe is all windsurfers and other watersport users send request to Samsung phone manufacturer and ask for Doppler?
Like millions of customers action asking manufacturer to do something they need? Global petition by all GPSTC and others?

You do realise doppler is already on smartphones, in addition to doppler speed accuracy estimates? I have an app that already records the data for post processing.

So do we have the same precision and accuracy on our or any of smartphones as our Motion GPS?
I still have only older Samsung smartphone that hardly can catch up on motorway to find nearest intersection and directions sometimes too late after you just passed in the wrong.

No, we do NOT! Not by a very long way!!

The smartphone have a huge disadvantage because they use a very tiny chip antenna. They are not capable of comparable accuracy from that alone.

Add to that they the GPS chip they use may or may not be in the same class.

The vast majority of smartphones don't give the Doppler error estimate.In fact, I have not found or heard of one that does. If there is one, please inform us. I have not been able to get this out of the Umidigi A7 Pro phone I have, but it does give some more data about the satellites it uses, and it is Multi Constellation so it gets a lot of sats.

Roo, if you have an app that can extract and record the Doppler Error data, why not make it available?

Edit, Sorry Roo, I read your post again and realised you are talking about post processing. I assume that is done on the positional data? I have never heard of post processing to get a better solution on Doppler data.

Select to expand quote

Macroscien said..
Lovely.
But I need to still make some calculation to prove it or disprove.
For example you will be happy to claim victory because your 1 hour was 24.003 and your mate did only 24.000.
But it could be quite unfair is your victory is due to margin error and actually your mate did better and now lost due to technicality or our system.
This errors accumulate over time too.

I observe this quite often, we feel good beating somebody by 0.001 but is it really true?

Boardsurfr already answered that question very well. Read it again, carefully.

The error of our one hour runs with a Motion Mini actually does approach 0.003Kt. This is also the resolution of the data in the file so the lowest reported error we can see in that data. So if your mate beats your 1 hour by 0.003Kt, there is at least a 98% probability that he really did beat you.

Errors in Doppler speed calculations do not accumulate in the Ublox data. They reduce, or cancel out with more data. The larger the time of the run, the smaller the overall error.

And it is also true that the error margin is greater for the shorter distances like 10 seconds, as Boardsurfr already explained. So then, if your mate beat you by only 0.01Kt, the probability that he actually did beat you is lower. So you can argue with him all day if you like, or you can just shrug and try harder next time.

Some interesting calculations for you Macro:

The difference between you going 30 knots, and your mate going 30.1 Kts, is that he will be approximately 5cm further ahead of you every second. 30Kts = 15.43333m/s. 30.1 Kts = 15.48478m/s

So over the 10 seconds he should beat you by about 50cm. That is probably significant, but still it would be very hard to see, even if you were close and side by side.

At 30 knots it takes almost exactly 120 seconds to do a Nautical Mile. If your mate beats you by 0.003Kts, he will be 0.1852m ahead of you.

That's 1852mm, which is exactly 1mm per meter faster than you!

It's up you you to decide how devastated you should feel about that.

Select to expand quote

sailquik said..
Some interesting calculations for you Macro:

The difference between you going 30 knots, and your mate going 30.1 Kts, is that he will be approximately 5cm further ahead of you every second. 30Kts = 15.43333m/s. 30.1 Kts = 15.48478m/s

So over the 10 seconds he should beat you by about 50cm. That is probably significant, but still it would be very hard to see, even if you were close and side by side.

At 30 knots it takes almost exactly 120 seconds to do a Nautical Mile. If your mate beats you by 0.003Kts, he will be 0.1852m ahead of you.

That's 1852mm, which is exactly 1mm per meter faster than you!

It's up you you to decide how devastated you should feel about that.

I love that science factual approach. So i will post soon some proposal experiment fir us to make. I am on way to farm so be later. So far so good and no sailing on my Hilux. Roads suppose to be flooded !

Select to expand quote

sailquik said..

Macroscien said..

Roo said..

Macroscien said..
We need improvent accuracy GPS for our everyday life.
You want to self park your vehicle within cm not meters.
Or hide a stash of money in the jar underground in the garden.
I am reading about newest Mate 70 Chinese smart phone and nobody say how good GPS going to be?! Better cameras or bigger screens that is all they do worry about. For us all the best could be new smartphone with improved GPS not another knock of DIY watch. I am happy even to invest $35 for additional plastic waterproof bag for new smartphone.
New smartphone with Doppler could solve all our problem we worry about for years here.Maybe is all windsurfers and other watersport users send request to Samsung phone manufacturer and ask for Doppler?
Like millions of customers action asking manufacturer to do something they need? Global petition by all GPSTC and others?

You do realise doppler is already on smartphones, in addition to doppler speed accuracy estimates? I have an app that already records the data for post processing.

So do we have the same precision and accuracy on our or any of smartphones as our Motion GPS?
I still have only older Samsung smartphone that hardly can catch up on motorway to find nearest intersection and directions sometimes too late after you just passed in the wrong.

No, we do NOT! Not by a very long way!!

The smartphone have a huge disadvantage because they use a very tiny chip antenna. They are not capable of comparable accuracy from that alone.

Add to that they the GPS chip they use may or may not be in the same class.

The vast majority of smartphones don't give the Doppler error estimate.In fact, I have not found or heard of one that does. If there is one, please inform us. I have not been able to get this out of the Umidigi A7 Pro phone I have, but it does give some more data about the satellites it uses, and it is Multi Constellation so it gets a lot of sats.

Roo, if you have an app that can extract and record the Doppler Error data, why not make it available?

Edit, Sorry Roo, I read your post again and realised you are talking about post processing. I assume that is done on the positional data? I have never heard of post processing to get a better solution on Doppler data.

Daffy,

pretty much most phones with Android 9 and above have the doppler S. Acc in the raw data. It can be saved very easily. I'm having one of our software guys write an app to record and analyse it, essentially an update to GPSLogit. The latest Android 14 phones have an incredible amount of raw data being produced from the current GPS chipsets, quite a few of the higher end ones use the U-Blox.

Roo

Here's what my phone produces:
Raw,utcTimeMillis,TimeNanos,LeapSecond,TimeUncertaintyNanos,FullBiasNanos,BiasNanos,BiasUncertaintyNanos,DriftNanosPerSecond,DriftUncertaintyNanosPerSecond,HardwareClockDiscontinuityCount,Svid,TimeOffsetNanos,State,ReceivedSvTimeNanos,ReceivedSvTimeUncertaintyNanos,Cn0DbHz,PseudorangeRateMetersPerSecond,PseudorangeRateUncertaintyMetersPerSecond,AccumulatedDeltaRangeState,AccumulatedDeltaRangeMeters,AccumulatedDeltaRangeUncertaintyMeters,CarrierFrequencyHz,CarrierCycles,CarrierPhase,CarrierPhaseUncertainty,MultipathIndicator,SnrInDb,ConstellationType,AgcDb,BasebandCn0DbHz,FullInterSignalBiasNanos,FullInterSignalBiasUncertaintyNanos,SatelliteInterSignalBiasNanos,SatelliteInterSignalBiasUncertaintyNanos,CodeType,ChipsetElapsedRealtimeNanos
#
# UncalAccel,utcTimeMillis,elapsedRealtimeNanos,UncalAccelXMps2,UncalAccelYMps2,UncalAccelZMps2,BiasXMps2,BiasYMps2,BiasZMps2
#
# Accel,utcTimeMillis,elapsedRealtimeNanos,AccelXMps2,AccelYMps2,AccelZMps2
#
# UncalGyro,utcTimeMillis,elapsedRealtimeNanos,UncalGyroXRadPerSec,UncalGyroYRadPerSec,UncalGyroZRadPerSec,DriftXRadPerSec,DriftYRadPerSec,DriftZRadPerSec
#
# Gyro,utcTimeMillis,elapsedRealtimeNanos,GyroXRadPerSec,GyroYRadPerSec,GyroZRadPerSec
#
# UncalMag,utcTimeMillis,elapsedRealtimeNanos,UncalMagXMicroT,UncalMagYMicroT,UncalMagZMicroT,BiasXMicroT,BiasYMicroT,BiasZMicroT
#
# Mag,utcTimeMillis,elapsedRealtimeNanos,MagXMicroT,MagYMicroT,MagZMicroT
#
# OrientationDeg,utcTimeMillis,elapsedRealtimeNanos,yawDeg,rollDeg,pitchDeg
#
# Fix,Provider,LatitudeDegrees,LongitudeDegrees,AltitudeMeters,SpeedMps,AccuracyMeters,BearingDegrees,UnixTimeMillis,SpeedAccuracyMps,BearingAccuracyDegrees,elapsedRealtimeNanos,VerticalAccuracyMeters,MockLocation,NumberOfUsedSignals,VerticalSpeedAccuracyMps,SolutionType
#
# Nav,Svid,Type,Status,MessageId,Sub-messageId,Data(Bytes)
#
# Status,UnixTimeMillis,SignalCount,SignalIndex,ConstellationType,Svid,CarrierFrequencyHz,Cn0DbHz,AzimuthDegrees,ElevationDegrees,UsedInFix,HasAlmanacData,HasEphemerisData,BasebandCn0DbHz
#
# Agc,utcTimeMillis,TimeNanos,LeapSecond,TimeUncertaintyNanos,FullBiasNanos,BiasNanos,BiasUncertaintyNanos,DriftNanosPerSecond,DriftUncertaintyNanosPerSecond,HardwareClockDiscontinuityCount,AgcDb,CarrierFrequencyHz,ConstellationType
#

I propose the following experiment :
Let's call it Macro Test 1.
Let's take two GPSes high quality and frequency. Like a Motion?
Now 1 GPS will be attached to our belly, torso or arm, as you like.
The 2 GPS we strap to the top of our mast.
Now we could start sailing as usual and compare results coming from those two units
MacroTest 1a
slide modification with one GPS on the belly and a second on top of the front of our board riding descent chop at a decent speed.
Again. Compare results
Macro Test 1b
One GPS on top of the mast and a second on the beak of our board again
Compare results

The theory behind experiment and concept.
Our precise GPS should exactly record :
a) path or length of the travel, say distance
b) speed/velocity at time increments

So in ideal conditions We/ our GPS travels in a straight line with constant speed. Nothing unusual happens. Nothing to worry about.
But the top of our mast in gusty, choppy condition will not be traveling at exact straight line but vibrating, oscillating, and shaking terribly.
So if we imagine a path in air , that is nothing like a straight line.
If our precise GPS is not able to record all this length of the complicated path. than summary distance should be greater than those going straight line on our belly.
Time passed is the same on both units but one recorded longer distance. So eventual speed should be higher on top of the mast.
Also one on the front of the board should record longer distances due to vertical chop movements. In extremes MacroTest 1c one GPS is on top of mast and second on board. If we are lucky those two differences may add up.

Since our GPS consists actually of two parts or systems: one recording distance/ location and another speed only by calculating Doppler
discrepancies in two GPS readings could vary, depending on which system was employed. Position or Doppler.

We could speculate what additional path GPS may record and if it even matters.
Let say that the mast top keeps oscillating/ bent 0.5 m twice a second ( 2Hz) .


So we add 0.5 x 2 x 2= 2 meters distance to our measurements.
or more if circle 2PiR _ up to 3.14 meters ! I am going to perform the following experiment once I manage to convince somebody to lend me 1 Motion GSP. Then I could post the results here.
You are free to make those experiments and comment results, there is no copyright on common sense here.

How much shaking of your GPS could add speed to your results ( if any at all ?) I don't know. This is the subject of this experiment to tell us all.
But Burrum water current is like 3 ktn ? which may contribute nicely to our pleasant resultsWe also could speculate that wave rider testing Macroeffect will record even greater distortion/ improvements in TOP GPS readings/ recordings

Select to expand quote

Roo said..

Daffy,

pretty much most phones with Android 9 and above have the doppler S. Acc in the raw data. It can be saved very easily. I'm having one of our software guys write an app to record and analyse it, essentially an update to GPSLogit. The latest Android 14 phones have an incredible amount of raw data being produced from the current GPS chipsets, quite a few of the higher end ones use the U-Blox.

Roo

Here's what my phone produces:
Raw,utcTimeMillis,TimeNanos,LeapSecond,TimeUncertaintyNanos,FullBiasNanos,BiasNanos,BiasUncertaintyNanos,DriftNanosPerSecond,DriftUncertaintyNanosPerSecond,HardwareClockDiscontinuityCount,Svid,TimeOffsetNanos,State,ReceivedSvTimeNanos,ReceivedSvTimeUncertaintyNanos,Cn0DbHz,PseudorangeRateMetersPerSecond,PseudorangeRateUncertaintyMetersPerSecond,AccumulatedDeltaRangeState,AccumulatedDeltaRangeMeters,AccumulatedDeltaRangeUncertaintyMeters,CarrierFrequencyHz,CarrierCycles,CarrierPhase,CarrierPhaseUncertainty,MultipathIndicator,SnrInDb,ConstellationType,AgcDb,BasebandCn0DbHz,FullInterSignalBiasNanos,FullInterSignalBiasUncertaintyNanos,SatelliteInterSignalBiasNanos,SatelliteInterSignalBiasUncertaintyNanos,CodeType,ChipsetElapsedRealtimeNanos
#
# UncalAccel,utcTimeMillis,elapsedRealtimeNanos,UncalAccelXMps2,UncalAccelYMps2,UncalAccelZMps2,BiasXMps2,BiasYMps2,BiasZMps2
#
# Accel,utcTimeMillis,elapsedRealtimeNanos,AccelXMps2,AccelYMps2,AccelZMps2
#
# UncalGyro,utcTimeMillis,elapsedRealtimeNanos,UncalGyroXRadPerSec,UncalGyroYRadPerSec,UncalGyroZRadPerSec,DriftXRadPerSec,DriftYRadPerSec,DriftZRadPerSec
#
# Gyro,utcTimeMillis,elapsedRealtimeNanos,GyroXRadPerSec,GyroYRadPerSec,GyroZRadPerSec
#
# UncalMag,utcTimeMillis,elapsedRealtimeNanos,UncalMagXMicroT,UncalMagYMicroT,UncalMagZMicroT,BiasXMicroT,BiasYMicroT,BiasZMicroT
#
# Mag,utcTimeMillis,elapsedRealtimeNanos,MagXMicroT,MagYMicroT,MagZMicroT
#
# OrientationDeg,utcTimeMillis,elapsedRealtimeNanos,yawDeg,rollDeg,pitchDeg
#
# Fix,Provider,LatitudeDegrees,LongitudeDegrees,AltitudeMeters,SpeedMps,AccuracyMeters,BearingDegrees,UnixTimeMillis,SpeedAccuracyMps,BearingAccuracyDegrees,elapsedRealtimeNanos,VerticalAccuracyMeters,MockLocation,NumberOfUsedSignals,VerticalSpeedAccuracyMps,SolutionType
#
# Nav,Svid,Type,Status,MessageId,Sub-messageId,Data(Bytes)
#
# Status,UnixTimeMillis,SignalCount,SignalIndex,ConstellationType,Svid,CarrierFrequencyHz,Cn0DbHz,AzimuthDegrees,ElevationDegrees,UsedInFix,HasAlmanacData,HasEphemerisData,BasebandCn0DbHz
#
# Agc,utcTimeMillis,TimeNanos,LeapSecond,TimeUncertaintyNanos,FullBiasNanos,BiasNanos,BiasUncertaintyNanos,DriftNanosPerSecond,DriftUncertaintyNanosPerSecond,HardwareClockDiscontinuityCount,AgcDb,CarrierFrequencyHz,ConstellationType
#

That amazing Roo.

Do you know if that Umidigi A7 Pro I have can produce all that too?

An update like GPS-Logit that can capture the Doppler error would be fantastic. It would be verity interesting to see the quality of the data from the phone antenna.

Got to give it to you Roo. You are always way out there exploring the boundaries.

Select to expand quote

Macroscien said..
I propose the following experiment :
Let's call it Macro Test 1...............

I think I know pretty much what you will find from experiments already done, and data we have already collected, much of which has been reported and seen in these forums.

But go for it Macro, and come back and tell us what you find.

Select to expand quote

sailquik said..
The smartphone have a huge disadvantage because they use a very tiny chip antenna. They are not capable of comparable accuracy from that alone.

It is certainly true that the chip antenna gives a poorer signal than the antennas we usually use. I've seen this with a u-blox chip, too, which I ended up returning because of poor data quality.

The stuff Roo posts is nothing new. He's actually not showing any data, he just lists what data an Android program can get from a GPS chips if the chip and phone firmware provide these data. It's a generic API that supports all kinds of different chips, and all kinds of data that a chip might provide. Roo has made similar claims years ago. It's a pretty trivial exercise to write a program that actually records the data. On (at least) the vast majority of cheap phones, the relevant field (speed accuracy) will likely be zeros.

Select to expand quote

Macroscien said..
I propose the following experiment :
Now 1 GPS will be attached to our belly, torso or arm, as you like.
The 2 GPS we strap to the top of our mast.

As Daffy said, this and similar experiments have been done multiple times. I have tried a mast mount, about a meter above the boom, which was quite excellent at showing how much the mast flexes, but total crap with respect to speed accuracy, since point-to-point speeds differed by pretty large percentages from mast movements.
Anyone who uses two boom-mounted ESP32 loggers does a similar experiment every time. You can usually tell which GPS was mounted on port and which on starboard by looking at the alphas - one of the units typically travels a larger path. You can pretty much always see a difference during the sail flip. On the track data, you can often make out which GPS was on which side - but not always, since sometimes the position errors between the units differ enough to flip their apparent position. In a 10-second run with enough curve in it, you may also be able to see which units was on the inside of the curve (the slower one) - although the differences will probably not be statistically significant.

Select to expand quote

boardsurfr said..

sailquik said..
The smartphone have a huge disadvantage because they use a very tiny chip antenna. They are not capable of comparable accuracy from that alone.

It is certainly true that the chip antenna gives a poorer signal than the antennas we usually use. I've seen this with a u-blox chip, too, which I ended up returning because of poor data quality.

The stuff Roo posts is nothing new. He's actually not showing any data, he just lists what data an Android program can get from a GPS chips if the chip and phone firmware provide these data. It's a generic API that supports all kinds of different chips, and all kinds of data that a chip might provide. Roo has made similar claims years ago. It's a pretty trivial exercise to write a program that actually records the data. On (at least) the vast majority of cheap phones, the relevant field (speed accuracy) will likely be zeros.

Macroscien said..
I propose the following experiment :
Now 1 GPS will be attached to our belly, torso or arm, as you like.
The 2 GPS we strap to the top of our mast.

As Daffy said, this and similar experiments have been done multiple times. I have tried a mast mount, about a meter above the boom, which was quite excellent at showing how much the mast flexes, but total crap with respect to speed accuracy, since point-to-point speeds differed by pretty large percentages from mast movements.
Anyone who uses two boom-mounted ESP32 loggers does a similar experiment every time. You can usually tell which GPS was mounted on port and which on starboard by looking at the alphas - one of the units typically travels a larger path. You can pretty much always see a difference during the sail flip. On the track data, you can often make out which GPS was on which side - but not always, since sometimes the position errors between the units differ enough to flip their apparent position. In a 10-second run with enough curve in it, you may also be able to see which units was on the inside of the curve (the slower one) - although the differences will probably not be statistically significant.

Interesting.
I did not read everything on SB. So what could the difference on Aphla
1) One gps mounted on the body or boom mast clamp
2) second GPS mounted at the outhaul boom end
We could suspect the sail flip is quite a dramatic event. Interesting how much it affects alpha results.

At our precision 0.001 could be significant

Here a typical plot from 4 units (ESP-GPS, 5 Hz, Speed over ground), 2 boom mounting and 2 head mounting. You can clearly see which units are the on the boom, and which are on the head !
I always use 4 units in this configuration, and the head units always gives the highest numbers ! Differences are int the range of 0.02 knots.

Select to expand quote

sailquik said..
Some interesting calculations for you Macro:

The difference between you going 30 knots, and your mate going 30.1 Kts, is that he will be approximately 5cm further ahead of you every second. 30Kts = 15.43333m/s. 30.1 Kts = 15.48478m/s

So over the 10 seconds he should beat you by about 50cm. That is probably significant, but still it would be very hard to see, even if you were close and side by side.

At 30 knots it takes almost exactly 120 seconds to do a Nautical Mile. If your mate beats you by 0.003Kts, he will be 0.1852m ahead of you.

That's 1852mm, which is exactly 1mm per meter faster than you!

It's up you you to decide how devastated you should feel about that.

I really love and love those calculations and numbers you provided.
Because it is something real we coukd refer to.
The we coukd dispute them . Then we coukd come to common agreement.
So if I question something don't be offended because my intention is to understand and if there is sny room for improvement , lets do so.
Or just let's us know the limits.

Select to expand quote

rp6conrad said..
Here a typical plot from 4 units (ESP-GPS, 5 Hz, Speed over ground), 2 boom mounting and 2 head mounting. You can clearly see which units are the on the boom, and which are on the head !
I always use 4 units in this configuration, and the head units always gives the highest numbers ! Differences are int the range of 0.02 knots.

Lovely!
This is exactly what I was thinking!!
With this only difference that I intend to post recording onto ka72 to give everybody chance to download file for own analysis.
So i plan to do domevtest, upload into ka72 and provide us clear link here.
I love excell to make all my further analysis.
Because I could instantly as excellent for interpretation I want.

Select to expand quote

sailquik said..

Macroscien said..

Roo said..

Macroscien said..
We need improvent accuracy GPS for our everyday life.
You want to self park your vehicle within cm not meters.
Or hide a stash of money in the jar underground in the garden.
I am reading about newest Mate 70 Chinese smart phone and nobody say how good GPS going to be?! Better cameras or bigger screens that is all they do worry about. For us all the best could be new smartphone with improved GPS not another knock of DIY watch. I am happy even to invest $35 for additional plastic waterproof bag for new smartphone.
New smartphone with Doppler could solve all our problem we worry about for years here.Maybe is all windsurfers and other watersport users send request to Samsung phone manufacturer and ask for Doppler?
Like millions of customers action asking manufacturer to do something they need? Global petition by all GPSTC and others?

You do realise doppler is already on smartphones, in addition to doppler speed accuracy estimates? I have an app that already records the data for post processing.

So do we have the same precision and accuracy on our or any of smartphones as our Motion GPS?
I still have only older Samsung smartphone that hardly can catch up on motorway to find nearest intersection and directions sometimes too late after you just passed in the wrong.

No, we do NOT! Not by a very long way!!

The smartphone have a huge disadvantage because they use a very tiny chip antenna. They are not capable of comparable accuracy from that alone.

Add to that they the GPS chip they use may or may not be in the same class.

The vast majority of smartphones don't give the Doppler error estimate.In fact, I have not found or heard of one that does. If there is one, please inform us. I have not been able to get this out of the Umidigi A7 Pro phone I have, but it does give some more data about the satellites it uses, and it is Multi Constellation so it gets a lot of sats.

Roo, if you have an app that can extract and record the Doppler Error data, why not make it available?

Edit, Sorry Roo, I read your post again and realised you are talking about post processing. I assume that is done on the positional data? I have never heard of post processing to get a better solution on Doppler data.

I think that you underestimate power of smart phone manufacturer.
What we trying to achieve here and cost us 300 to $500 they could do for 5 to $10 on mass market production.
The problem is they don't know what we/ customer want?!
More pixels on camera? !
Bigger screen?@
Nonsense like that.
Customers want now very precise GPS!!
GPS on smart phone that I coukd use as tape measure!
If they give us 1cm accuracy carpenter coukd build whole house using smartphone and gps!
Even if now I need to carry gps smart phone gos in my camel back because is too big to fit on the wrist ,let's it be . Or we coukd always add smart watch on wrist and carry smart phone in backpack.

Macro. Phone manufaturers have been trying to add increased GPS accuracy to their smartphones for years, mainly for navigation. But there is not much demand for anything much more than for that. For that sort of accuracy poeple use different specialised devices that do those tasks SOOOOOO much better.

And there are numerous papers, published and available on the internet, where top science teams in universities and companies have shown how they are trying to get decimeter locational accuracy to a point where it might be practical for smartphones. They are mostly still a long way off from practical solutions. Do some searches. Do some study. Start by searching even this forum for posts about these sorts of things.

Get yourself educated by your own efforts, and then, when you have at least some sort of up to date understanding, come back with something that has not been already researched and done. It's somewhat rude to keep expecting people in this forum to be educating you from scratch all the time.

Do your experiments yourself and see if you learn anything we don't already have a pretty good handle on.

Daffy.
What is the declarer margin of error of our GPS devices?
1% ?
or better?
So if we have result 40 ktn in any discipline the margin of error will be
40 x 1% = 0.4 knots

But we declare the winner because of the 0.001 knots advantage.
In serious competition, the picture is taken on the finish line. Even dog racing.

So technically the loser could be a winner, just we decided so.

For the system to be 100% fair not just statistically probable the winner and loser should be declared when our sphere of errors separates completely for each race.
Sailor 1 .
40.000 knots
+ - 0.4 knots
= 39.960 to 40.040 knots

Sailor 2. result
40.060 knots
Obvious winner? wait a minute

+- 0.4 knot
= 40.020 to 40.080

So there is a chance/ probability that Sailor 1 was doing 40.040 ktn and the Sailor 2 only 40.020


we could start arguing that the declared margin of error or our GPS is smaller/ better.

So we shrink only the error spheres above but not eliminate methodology completely.

At sports events usually the results count not things like style or color of clothes, judges appreciation to one over another.

We could add a point of privilege and deduct 0.5 knots of Macro results because the commission don't like his style or dirty sail.give another sailor 0.5 ktn bonus for the latest board and sail colors.

To be absolutely fair we should declare sailor results even - if a declared error of margin of both sailors interferes. ( not separate completely)

I could see the possibility of a procedure to implement total fairness in compiling results that would require additional algorithms to round results according to GPS devices precision at the time. It may not be necessary for our GPSTC as no money changes hands or gambling going on on results.

But for dog racing declaration like that:
" Mate There is 80 % probability that my dog wins over your dog. Give me your $10,000 dollars now"
can get you stabbed in the back in the dark.

Now a bit of science

www.gpsworld.com/gnss-systemalgorithms-methodsinnovation-doppler-aided-positioning-11601/

Select to expand quote

rp6conrad said..
Here a typical plot from 4 units (ESP-GPS, 5 Hz, Speed over ground), 2 boom mounting and 2 head mounting. You can clearly see which units are the on the boom, and which are on the head !
I always use 4 units in this configuration, and the head units always gives the highest numbers ! Differences are int the range of 0.02 knots.

at 30 ktn this 0.02 knots quoted
equal to 0.66% percent accuracy - not far from my 1% example.
( or just right at 20 ktn)

If we have access to your raw data like KA72 submission we could download for further analysis.

I will use my GPSResults and move to Excel copied data.

Unless somebody has better tools to decrypt our GPS file into a spreadsheet.
For sure KA72 author Dylan has a bunch of those and use every day.