Woo3 and Xensr compared

[deniska]

Just get a woo...u Will start from 10m easy

funny that if you look at some older threads from woo1 days.. most people were complaining about how they go above other's kites but the stupid woo would show 9 meters...

[kiterocky]

Nobody Will know the reality...until a reality test...let me know when they do

[fluidity]

I've suggested a likely cause of error before.
No degrees here, no top 20 android apps, no genius friends to advise on other's errors.
However, the woo chips have to work out an average water height level reference if they are going to record a jump height from lake/river/sea level. If the average water level reference is calculated by the chip manufacturer (who cares very little about woos and a whole lot more about selling gaming console sensors and airbag crash sensors) then that average height calculation is going to be averaged over a period of time regardless of whether some of it was in the air.
Then the woo software writers get all the high level maths tools to analyse the "facts"as reported by the factory chip maths coprocessor and the zero height references are distorted by the jump periods influencing the averages.
So the woo software writer is working with bad data and every jump looses height due to the woo thinking it's already higher at the moment it starts.
Clear to everybody?

No!

Wo bases on accelerometer, not some pressure sensor or other shit which needs water level reference. You dont need to know the water level.

No! You missed my point wrogu.
The accelerometer monitoring maths processor chip doesn't know the absolute water level but it still knows when it is at rest (1gravity at a particular vector) and by averaging it should calculate wave peaks and average water level relative position. On a jump it can start by measuring when vibrations from motion in water cease but height from average water level is not same as height jumped off the wave. The factory maths chip is not a woo chip, it's a semiconductor manufacturer's chip. They have no significant interest in Woo sales, Woo is a niche market product unlike car crash sensors and gaming console sensors. So the algorithm to determine "at rest height"(I've renamed it from average water level so you don't go off on a tangent) is confused because it doesn't understand that previous recent jumps aren't part of the at rest height. Remember this is an accelerometer chip with rotational and axial vector directions and likely compass also. The maths chip never knows absolute positions but it can't work out a height other than from leaving a wave without the "at rest height" reference. This has to be averaged. To average it has to include multiple wave cycles to reduce start and stop sample time influence from non zero end samples. If woo designed the factory chip they would need to specifically look for wave peaks and troughs and then determine the at rest height but they don't, they will be analysing data that is most likely already tainted by previous recent jump influences.

[deniska]

If woo designed the factory chip they would need to specifically look for wave peaks and troughs and then determine the at rest height but they don't, they will be analysing data that is most likely already tainted by previous recent jump influences.

How do you come up with stuff like this?
This sounds like it was written by AI bot...

[fluidity]

If woo designed the factory chip they would need to specifically look for wave peaks and troughs and then determine the at rest height but they don't, they will be analysing data that is most likely already tainted by previous recent jump influences.

How do you come up with stuff like this?
This sounds like it was written by AI bot...

No, I ran a business for 4 years doing embedded programming, high level programming and circuit board design.

[CaliRider]

The accelerometer monitoring maths processor chip doesn't know the absolute water level but it still knows when it is at rest (1gravity at a particular vector) and by averaging it should calculate wave peaks and average water level relative position. The factory maths chip is not a woo chip, it's a semiconductor manufacturer's chip. The maths chip never knows absolute positions but it can't work out a height other than from leaving a wave without the "at rest height" reference. If woo designed the factory chip they would need to specifically look for wave peaks and troughs and then determine the at rest height but they don't, they will be analysing data that is most likely already tainted by previous recent jump influences.

my friends and i are wetting our pants reading this......granted we've had a few beers but hahahahahahaha this is killing us. ""maths chip""??? seriously wut are you even talking about?? it is really difficult to type because we are laughing soooo hard .. I'm not a math or nerd guy and even i know that this is all done by the software written for the computer chip in the woo or pique or xenser or whatever. like this website, its all programming. theres no kiteboarding discussion board chip. hahahaahahahahaha its all code dudes. theres no factory chip for kieboard jumps. holy crap you make me feel like a genius and you claim to be a nerd!!! sorry but this is too funny. probably the only gold in this 10 page thread.
someone get leo in here to explain it. actually that would be pretty funny since everyone nerdy that i know says woo didnt build the jump software code but bought it from someone else and thats why their devices are crap from version to version. oh man another round of beers coming. later dudez

[Pemba]

Well finally an explanation. It's all code ! It's all programming ! Yeah, another round of beers sounds appropriate. Say Hi to your friends from me.

[ShaveTheWhales]

Garmin watch works with this kite app for height..
Change is like a holiday

https://apps.garmin.com/en-NZ/apps/9c6b ... 2f19bd68#0

If it doesn't work properly - it's in the right forum post

[wrogu]

Garmin watch works with this kite app for height..
Change is like a holiday

https://apps.garmin.com/en-NZ/apps/9c6b ... 2f19bd68#0

If it doesn't work properly - it's in the right forum post

It uses watch's internal sensor, right? If so its not so good, as board is the best jumping indicator (best signal coming from landing and leaving water surface). Imho sensor placement is crucial.

As for woo hardware - it is prolly something like this:
https://www.invensense.com/products/mot ... ng/9-axis/
Plus a power section and a CPU, memory, I/O. SImple as that.

[nothing2seehere]

I've suggested a likely cause of error before.
No degrees here, no top 20 android apps, no genius friends to advise on other's errors.
However, the woo chips have to work out an average water height level reference if they are going to record a jump height from lake/river/sea level. If the average water level reference is calculated by the chip manufacturer (who cares very little about woos and a whole lot more about selling gaming console sensors and airbag crash sensors) then that average height calculation is going to be averaged over a period of time regardless of whether some of it was in the air.
Then the woo software writers get all the high level maths tools to analyse the "facts"as reported by the factory chip maths coprocessor and the zero height references are distorted by the jump periods influencing the averages.
So the woo software writer is working with bad data and every jump looses height due to the woo thinking it's already higher at the moment it starts.
Clear to everybody?

No!

Wo bases on accelerometer, not some pressure sensor or other shit which needs water level reference. You dont need to know the water level.

No! You missed my point wrogu.
The accelerometer monitoring maths processor chip doesn't know the absolute water level but it still knows when it is at rest (1gravity at a particular vector) and by averaging it should calculate wave peaks and average water level relative position. On a jump it can start by measuring when vibrations from motion in water cease but height from average water level is not same as height jumped off the wave. The factory maths chip is not a woo chip, it's a semiconductor manufacturer's chip. They have no significant interest in Woo sales, Woo is a niche market product unlike car crash sensors and gaming console sensors. So the algorithm to determine "at rest height"(I've renamed it from average water level so you don't go off on a tangent) is confused because it doesn't understand that previous recent jumps aren't part of the at rest height. Remember this is an accelerometer chip with rotational and axial vector directions and likely compass also. The maths chip never knows absolute positions but it can't work out a height other than from leaving a wave without the "at rest height" reference. This has to be averaged. To average it has to include multiple wave cycles to reduce start and stop sample time influence from non zero end samples. If woo designed the factory chip they would need to specifically look for wave peaks and troughs and then determine the at rest height but they don't, they will be analysing data that is most likely already tainted by previous recent jump influences.

I don't think it matters about the average height of the water because its applying kalman filters to the data from the inertia sensors in the device. The height should be calculated from the point of lift-off not from an average wave height. Anecdotally - I don't get a jump recorded if I ride over a big wave and pull my legs up at the top (1m minimum height recording ). If just detecting an upwards motion of the board was enough I would expect to get a jump detected.

Besides which, there was a load of fuss on here with people complaining about the accuracy with the guy jumping 16m on flat water. The average height there wouldn't have made any difference as the water was flat. I reckon he was just really good.

Sure you could cheat the software if you wanted to (throwing the board up/moving the sensor from the middle of the board to one end etc) but what would be the point? I think the Xensr rep acknowledged that with the iwatch software you could punch the air at your highest point to get a higher reading. It might be interesting if watch software moved into the android market place as you'd potentially see more people doing Captain Morgan impressions instead of thrusting the board up high.