You're so hung up on whether or not it can be done when the proof may be commercially available early next year. Instead of being so adamant about it being impossible, why don't you wait a few months before you end up with egg on your face.
Pocket Pro
- Thread starter Ringer
- Start date
- Status
You're so hung up on whether or not it can be done when the proof may be commercially available early next year. Instead of being so adamant about it being impossible, why don't you wait a few months before you end up with egg on your face.
Frans@France
New
Could you explain to me why you are so adamant that a gauge connected to the shaft at grip height can reliable measure the angle of the club face which is around 30-40inch away from that same gauge while the shaft is being bend below the connection point of that same gauge due to the forces of the swing?
S
SteveT
Guest
I would like to know what algorithmic assumptions were made to back-calculate the data for the Pocket Pro ..!!!!
Last edited:
Perhaps if the Pocket pro is put on in a certain way (likely aligned with leading edge) and in certain vertical position (like straight up and down), then the device would know what the face angle was doing and would know the clubs's orientation in relation to the earth's gravity. I guess I doubt it can measure toe droop and other impact deviations.
Frans@France
New
birly-shirly
New
This is essentially how I assumed it would work. Except that I think Frans is right in that what the sensor can detect will just be the differences between one reference point and another. That's a limitation in one sense, but if you can design software to correctly interpret start-up and impact, then you wouldn't actually need to worry too much about the initial alignment of the tool.
A square clubface at address would be a prerequisite - although I suppose you could calibrate for a consistent degree of open or closed alignment. But I think the alignment part of that would be quite challenging...
More useful might be the possibility of using a visibly straight ballflight to calibrate the data handling to account for the effects of shaft flex.
I still don't see that this is fundamentally flawed as an idea. Frans seems to think it can't be done for full shots. I would guess that the biggest hurdle would be sampling rate. Did the video claim a sampling rate of 1kHz? How much clubface rotation in 1/1000 sec? Less than 0.5*?
Obviously the manufacturers need to show that they can make this work. I'm in no rush to claim that they can't.
Frans@France
New
So we have a machine with the problem of the placement on the shaft and the problem of not measuring drop and lead. Those two problems will make sure that both clubhead speed will be inaccurate and face angle will be inaccurate right from the start.
Then we also have the calibration or 0-point placement. In order to make sure that the 0-point is 100% correct at address we need to calibrate the gauge to the LIE of the club AND to the line perpendicular to the face.
But those with a small lie/loft machine will have no problem doing that, for irons that is
Those with a high quality loft/lie machine will also able to calibrate it on the driver which is maybe square or 0.5degree closed or open.So we start swinging......the sensor will NOT (repeat) NOT measure it's position is space in x,y,z coordinates but will for each sample measure the acceleration. So for each sample within a certain time frame you get the m/s2 data.
Next we take the m/s2 and the sample frequency. Using all samples and some nice math we can then ESTIMATE the displacement in x,y,z.
Now please read the following again : "These quantities suffere from substantial integration drift" or with other words : it will only be an estimate and nothing more.
---
Not trying to be negative about those entrepreneurs just trying to protect the consumer.
Last edited:
S
SteveT
Guest
Hey .. maybe the gizmo uses polar coordinates? ... Just asking .... out of ignorance ...
Frans@France
New
Hey .. maybe the gizmo uses polar coordinates? ... Just asking .... out of ignorance ...
LOL
S
SteveT
Guest
Polar coordinates were fully developed by Euler, a German-Swiss ... whereas the Cartesian system refers to the French mathematician and philosopher René Descartes (who used the name Cartesius in Latin).
birly-shirly
New
So we have a machine with the problem of the placement on the shaft and the problem of not measuring drop and lead. Those two problems will make sure that both clubhead speed will be inaccurate and face angle will be inaccurate right from the start.
Then we also have the calibration or 0-point placement. In order to make sure that the 0-point is 100% correct at address we need to calibrate the gauge to the LIE of the club AND to the line perpendicular to the face.
But those with a small lie/loft machine will have no problem doing that, for irons that is
So we start swinging......the sensor will NOT (repeat) NOT measure it's position is space in x,y,z coordinates but will for each sample measure the acceleration. So for each sample within a certain time frame you get the m/s2 data.
Next we take the m/s2 and the sample frequency. Using all samples and some nice math we can then ESTIMATE the displacement in x,y,z.
Now please read the following again : "These quantities suffere from substantial integration drift" or with other words : it will only be an estimate and nothing more.
---
Not trying to be negative about those entrepreneurs just trying to protect the consumer.
OK. So the question is how accurate the "estimate" can be.
I'm still tickled, after all the difficulties that you just listed yourself, that you were LOL'ing when I suggested that developing the software might be difficult...
Frans@France
New
Depends on sample rate, accuracy of the acceleration and swingspeed of the hands
You missed the point, the software is easy, the formula is (depends on the programming language) maybe 10 lines long. It's the hardware that makes that the values are estimates. Sample rate, accuracy and the fact that the displacement is measured using acceleration is all hardware related.
birly-shirly
New
Sorry. Maybe I should have posted my 2 sentences in 2 separate posts.
Obviously you are correct as to what is being actually measured.
I'm not pretending to know what the algorithm would look like. I just imagine that there could be a lot of work solving ancillary problems - like interpreting take away and impact as events, or perhaps making some provision for people to use the HW with a variety of clubs, lengths, lies.
A couple of curiosity points. Do you have any thoughts as to WHY they went with a sensor position jammed up by the grip? Would that kind of hardware make a noticeable difference to swingweight if it was nearer the head?
Would the results be much more accurate if they put sensors in two places - one near the grip and one nearer the head?
Obviously you are correct as to what is being actually measured.
I'm not pretending to know what the algorithm would look like. I just imagine that there could be a lot of work solving ancillary problems - like interpreting take away and impact as events, or perhaps making some provision for people to use the HW with a variety of clubs, lengths, lies.
A couple of curiosity points. Do you have any thoughts as to WHY they went with a sensor position jammed up by the grip? Would that kind of hardware make a noticeable difference to swingweight if it was nearer the head?
Would the results be much more accurate if they put sensors in two places - one near the grip and one nearer the head?
Frans@France
New
adding that functionality to the software is just that....work.....
Good remark!
4/5 gram in the grip will change the sw with 1 point. At the head however (woods) the sw will change 4 points with those same 4/5 gram! [2 points with an iron]
Together with a clip, battery, bluetooth etc I guess it will be more then 25 gram...so It will act as a counterweight (placed near the grip) to the club and as such might cause swing changes to certain golfers.
Shot Limit
New
If it was lower on the club it might not be able to handle the impact.
I read a post somewhere else by someone in on this project (supposedly) in response to the question of whether is could account for toe droop / lead deflection
"Our most recent advancements allow us to graphically display the deflection of the club head during the swing - i.e. the distance from the club head to where it would be if the shaft wasn't bent. To a close approximation, this distance should be proportional to the torque the hands exert on the club. Also, the frequency that we can measure is the resonant frequency that the club would vibrate at if you clamped the grip in a vice and struck the club head. This provides us with the constant of proportionality between club head deflection and torque.
We are also able to display the angular velocity (or angular acceleration) during the swing. This would actually be easier - it's something we measure directly and it doesn't depend on the physical parameters of the club.
All this and more coming very soon."
He also said a video comparing the results from trackman and the pocketpro is coming soon.
I'm not taking a position as to whether or not this device is an acceptable replacement for tm/flightscope, just wanted to share.
"Our most recent advancements allow us to graphically display the deflection of the club head during the swing - i.e. the distance from the club head to where it would be if the shaft wasn't bent. To a close approximation, this distance should be proportional to the torque the hands exert on the club. Also, the frequency that we can measure is the resonant frequency that the club would vibrate at if you clamped the grip in a vice and struck the club head. This provides us with the constant of proportionality between club head deflection and torque.
We are also able to display the angular velocity (or angular acceleration) during the swing. This would actually be easier - it's something we measure directly and it doesn't depend on the physical parameters of the club.
All this and more coming very soon."
He also said a video comparing the results from trackman and the pocketpro is coming soon.
I'm not taking a position as to whether or not this device is an acceptable replacement for tm/flightscope, just wanted to share.
Dustin Smale
New
Updates
I've been following these guys since I heard about them here and they have posted the results of their "Pocket Pro Analysis" vs. "Video Analysis" on their website.
http://www.pocketpro.org/images/results/bigresults.jpg
This doesn't really mean anything to me except that the graphs are almost perfectly symmetrical. This was released at the PGA show and I was wondering what these graphs meant, and if they hold any merit.
I know its not a Trackman comparison, but hopefully that will be explored in the future.
I've been following these guys since I heard about them here and they have posted the results of their "Pocket Pro Analysis" vs. "Video Analysis" on their website.
http://www.pocketpro.org/images/results/bigresults.jpg
This doesn't really mean anything to me except that the graphs are almost perfectly symmetrical. This was released at the PGA show and I was wondering what these graphs meant, and if they hold any merit.
I know its not a Trackman comparison, but hopefully that will be explored in the future.
- Status