Hi, are GPS logs meant to be recorded at 2 or 3 second intervals? I thought I read somewhere it was three, but see 2 sec max speed is what's used on the gps speed sailing website. If someone in the know could clarify that would be great. Thanks.

Sorry fail to understand your question, Who which or what, are you talking about here when you say "meant"?

The GPS Team Challenge accepts data logged at 1sec, and won't accept data logged any slower. Our current preferred devices record at 0.2s (5hz) and coming some time soon are devices that record at 10hz and 18hz

The required GPS logging setting when posting windsurfing related results - is to set them 1-sec interval on devices that only run that fast - aka 1Hz. If you have a new device that logs faster - say 5Hz (0.2 seconds) - then you can choose whether to
a) log more data possibly not having enough storage space for your 24hr sessions and running out of battery
b) store less data possibly extending battery capacity.

You shouldn't need to log at 2-seconds for any device available in the last 10 years.

Matt,
Are you confusing (or do you mean) that the 2 sec peak speed is based on 3 1sec log points i.e. just say I had a 2 sec peak of 45.00 knots() it would be based on the following best consecutive log points (hi-lighted) 1second apart - 1)44.8kn, 2)44.90kn, 3)45.10kn, 4)45.00kn, 5)44.8kn.

Mathew / Decrepit - Please correct me if this is not the case

Select to expand quote

mathew said..
The required GPS logging setting when posting windsurfing related results - is to set them 1-sec interval on devices that only run that fast - aka 1Hz. If you have a new device that logs faster - say 5Hz (0.2 seconds) - then you can choose whether to
a) log more data possibly not having enough storage space for your 24hr sessions and running out of battery
b) store less data possibly extending battery capacity.

You shouldn't need to log at 2-seconds for any device available in the last 10 years.

Matt, Larko and I recently just used our GW60's together. They were fully charged, turned on at the same time, mine logging at 1hz, Larko's at 5hz. And guess what, the both ran out of battery at the same time!! I know Mike has done tests where he seems to think logging at 5hz uses more battery but our side by side comparo seems to refute that.

Select to expand quote

sick_em_rex said..

mathew said..
The required GPS logging setting when posting windsurfing related results - is to set them 1-sec interval on devices that only run that fast - aka 1Hz. If you have a new device that logs faster - say 5Hz (0.2 seconds) - then you can choose whether to
a) log more data possibly not having enough storage space for your 24hr sessions and running out of battery
b) store less data possibly extending battery capacity.

You shouldn't need to log at 2-seconds for any device available in the last 10 years.

Matt, Larko and I recently just used our GW60's together. They were fully charged, turned on at the same time, mine logging at 1hz, Larko's at 5hz. And guess what, the both ran out of battery at the same time!! I know Mike has done tests where he seems to think logging at 5hz uses more battery but our side by side comparo seems to refute that.

Thanks Rex, locosys confirmed a while back that there wouldn't be any difference. The GW60 still samples at 5hz, then averages out the 5 samples to record at 1hz, so all 1hz does is save memory, and unlike the GW52, the GW60 has enough to last more than 24hrs, so 1hz doesn't have any advantages. except make a smaller file to download and process.

Select to expand quote

sausage said..
Matt,
Are you confusing (or do you mean) that the 2 sec peak speed is based on 3 1sec log points i.e. just say I had a 2 sec peak of 45.00 knots() it would be based on the following best consecutive log points (hi-lighted) 1second apart - 1)44.8kn, 2)44.90kn, 3)45.10kn, 4)45.00kn, 5)44.8kn.

Mathew / Decrepit - Please correct me if this is not the case

Brian, depends what program you're using and what method you choose for integration. The default realspeed method is as you say based on 3 samples, but I think KA72, GPSResults and perhaps GPSARpro only use 2 samples.
Seems there's a bit of confusion about the GT31 1sec result, is it the average of 1s or is it a single short sample.
Definitely, the GWs should only need two points averaged, if they are set to 1hz
I don't have the maths to understand what they all mean, but I think "cubic spline" takes the extra sample, and rectangular doesn't, but I could be corrected here.

There are two common settings in software to work out a 2 second (or 'N' Second) 'average'.
The oldest way is 'Rectangular' which uses the average of two points @ 1Hz (2 sec average), or 10 points @ 5Hz (20 points at 10Hz etc).
The newer way devised by the maths gurus, and thought by them to be a better close fit to reality, uses 'Triangular' integration, specifically Cubic Spline. This is a mathematical method to find a best fit curve over multiple points with no info in between the points. This method uses the extra point in each calculation. i.e. 3 for 2 seconds @1Hz, 11 for 10 seconds @1Hz, etc

The 10 second and 5 x 10 second results from GPS-Results and GPSAR-Pro are default 'Triangular' (cubic spline).
Older versions of RealSpeed has Triangular by default, but a later versions (V1.929) has a drop down window to select either 'Triangular' or 'Rectangular'. Mal was working on applying that drop down choice to each individual division when development stopped. The latest circulation of RealSpeed (V2 - which can cope with 5Hz and 10Hz data and correctly calculate Alphas with those Hz rates) has the drop down boxes in both the options (overall) and in the division edit windows, but they are not activated yet. V1.929 has it activated in the OPTION menu.

Note that with 'Triangular' integration, 'sawtooth' is soothed out. Since it is common to get a 'sawtooth' appearance from any GPS data, Triangular integration is still the best method. That is, it will more commonly produce an average speed closer to actual reality.

It is ALWAYS better to set =/>5Hz GPS's to 5Hz or higher. The above statement explains why more data is better and more accurate.

Higher Hz will also be more accurate in that the software can work with more resolution to find the fastest part of a 2 second or 10 second speed curve.

Higher Hz will also enable a more accurate calculation of distance divisions like 100m, 250m, 500m and NM.

Both the GW52 and GW6o are limited far more by battery life than memory storage space. (both are not the best options for 24 hr sessions! - use your old GT-31 for that).

The GW60 has far more memory space than the GW52. More than enough for 24 hours of 5Hz recording, but you will need to stop to recharge it a few times for that. :-)

The GW52 can log 24 hours at 1Hz in memory space and can be connected to an external usb battery pack (all inside the waterproof housing/bag) to last the time. In theory, the same could be done with the GW60, but it would surely prove to be a bit more physically awkward.

Select to expand quote

sick_em_rex said..

mathew said..
The required GPS logging setting when posting windsurfing related results - is to set them 1-sec interval on devices that only run that fast - aka 1Hz. If you have a new device that logs faster - say 5Hz (0.2 seconds) - then you can choose whether to
a) log more data possibly not having enough storage space for your 24hr sessions and running out of battery
b) store less data possibly extending battery capacity.

You shouldn't need to log at 2-seconds for any device available in the last 10 years.

Matt, Larko and I recently just used our GW60's together. They were fully charged, turned on at the same time, mine logging at 1hz, Larko's at 5hz. And guess what, the both ran out of battery at the same time!! I know Mike has done tests where he seems to think logging at 5hz uses more battery but our side by side comparo seems to refute that.

:) I didn't actually say "GW60"... I also said "possibly extending battery capacity". Sailquik and I built a GPS-logger capable of logging from 1Hz to 16Hz -> the 65mA consumption was at 1Hz and the 79mA consumption was at 16Hz.

Your mileage may vary according to the device used - there is no conspiracy of false information here.

Select to expand quote

sausage said..
Matt,
Are you confusing (or do you mean) that the 2 sec peak speed is based on 3 1sec log points i.e. just say I had a 2 sec peak of 45.00 knots() it would be based on the following best consecutive log points (hi-lighted) 1second apart - 1)44.8kn, 2)44.90kn, 3)45.10kn, 4)45.00kn, 5)44.8kn.

Mathew / Decrepit - Please correct me if this is not the case

Just responding to the "confusing" bit... the question uses the words "gps logs at 2 or 3 second intervals"... which doesn't at all have anything at all to do with peak/average values for specific divisions, and only with logs and the interval.

Select to expand quote

mathew said..

sausage said..
Matt,
Are you confusing (or do you mean) that the 2 sec peak speed is based on 3 1sec log points i.e. just say I had a 2 sec peak of 45.00 knots() it would be based on the following best consecutive log points (hi-lighted) 1second apart - 1)44.8kn, 2)44.90kn, 3)45.10kn, 4)45.00kn, 5)44.8kn.

Mathew / Decrepit - Please correct me if this is not the case

Just responding to the "confusing" bit... the question uses the words "gps logs at 2 or 3 second intervals"... which doesn't at all have anything at all to do with peak/average values for specific divisions, and only with logs and the interval.

Mathew, not all people say exactly what they mean all the time. I think Brian was allowing a generous amount for that.

Correction of terminology. In my post above, 'Triangular' should be 'Trapezoidal'.

Seniors moment!

Select to expand quote

sausage said..
Matt,
Are you confusing (or do you mean) that the 2 sec peak speed is based on 3 1sec log points i.e. just say I had a 2 sec peak of 45.00 knots() it would be based on the following best consecutive log points (hi-lighted) 1second apart - 1)44.8kn, 2)44.90kn, 3)45.10kn, 4)45.00kn, 5)44.8kn.

Mathew / Decrepit - Please correct me if this is not the case

Yes, I was looking at that 2 sec peak stat, thinking that you'd need to set the GPS then to log at 2 sec intervals.

Sorry for the confusion... math and science aren't my strong point.

Select to expand quote

mattn said..

sausage said..
Matt,
Are you confusing (or do you mean) that the 2 sec peak speed is based on 3 1sec log points i.e. just say I had a 2 sec peak of 45.00 knots() it would be based on the following best consecutive log points (hi-lighted) 1second apart - 1)44.8kn, 2)44.90kn, 3)45.10kn, 4)45.00kn, 5)44.8kn.

Mathew / Decrepit - Please correct me if this is not the case

Yes, I was looking at that 2 sec peak stat, thinking that you'd need to set the GPS then to log at 2 sec intervals.

Sorry for the confusion... math and science aren't my strong point.

Cheers Mattn.

Understanding what people say and what they actually mean can come in handy occassionally, hey Mathew