Tech Disc Accuracy

[Sheep]

Tech Disc Errors

New BB stuff hates some links and messes up media linsk>
So no preview.
Guys getting random 95mph pulls on his tech disc. hahahaha.

A lot of interesting information still coming out as people get these and explore it.

We look at stokleys throw with it the other day, he did a casual 360 and somehow got that HUGE number. When that was realistically a 60ish MPH throw.

Anyways, keep chasing data with tech disc.

He's not wrong either in the comments, it does reward poor form. Because you start chasing data instead of chasing technique.

 

[BenjiHeywood]

I'm going to defend techdisc a bit. It can throw out silly numbers once in a while, if you do something funky, and it's possible to game it if you know how. But if used properly i find it very accurate.

You can get daft numbers if you don't let the disc be stationary at some point in the 6 seconds before release. That's repeatable behaviour, and is only a problem if you want to do something like elephant walk with a techdisc, or if you are too dumb to understand that 95mph throws indicate that you've done something weird rather than suddenly found the secret sauce. If you do a normal pause to line up the shot, then a normal walk up, you'll get very accurate results IME.

 

[Sheep]

I'm going to defend techdisc a bit. It can throw out silly numbers once in a while, if you do something funky, and it's possible to game it if you know how. But if used properly i find it very accurate.

You can get daft numbers if you don't let the disc be stationary at some point in the 6 seconds before release. That's repeatable behaviour, and is only a problem if you want to do something like elephant walk with a techdisc, or if you are too dumb to understand that 95mph throws indicate that you've done something weird rather than suddenly found the secret sauce. If you do a normal pause to line up the shot, then a normal walk up, you'll get very accurate results IME.

Yeah, that's kind of a coding issue.

The disc is measuring way to far long before the throw vs parsing data from a key point.

So thats essentially why stokley was able to get 86mph on it, he was swinging it around a lot prior to casually throwing it in the net. But then of course got online to brag to everyone about the data.


IT's just an example of why we have to be really careful with these tools, how we use them and the trust in the data were getting from them.

It's a measuring device, but we also have to look at what were measuring, the knowledge of the person doing the measuring.

Either way, I'm mainly seeing people chase data with it. But also as that guy said in there, the disc really is finicky depending on how you swing, which really at that point makes it harder to trust a lot of the data from it with what you said as well. We know we can game the data. So, how accurate is the correct data were getting then if were able to trick the system so easily?

 

[Nick Carroll]

One thing to think about: TechDisc will update software over time and fix bugs. Also future models may have newer hardware, better sensors, etc. One critique I have is they need to have a public changelog so we can see what bugs are fixed, if there are new features, etc. (maybe there is one and I just can't find it)

 

[BenjiHeywood]

The disc is measuring way to far long before the throw vs parsing data from a key point.

The opposite - it's sometimes not recording for long enough.

All i know for sure is that it needs to be stationary to calibrate. Here's what i imagine is happening - a bit of guesswork but i bet it's not far off.

When the disc is stationary, the combined acceleration measured by all the accelerometers is just going to equal gravity (9.81m/s/s). So if it measures exactly that acceleration for some short period of time, it can assume that it's stationary, and the exact measurements on the various accelerometers, at their different angles (i.e. the individual measurements which we combined to make 9.81) can be used to calculate its orientation.

[COLOR=var(--text)]Any acceleration it feels from that point on can then be used to calculate the movement from that stationary point.

With modern accelerometers, i see no reason it can't be extremely accurate. The only issue is if the uploaded data doesn't contain a stationary point. (It records data all the time, but only uploads the 6 seconds or so up to release, so if the acceleration is never 9.81 in that period then it won't have much of a clue what 'stationary' is and you'll get weird results.)

You could try to game it by (e.g.) walking very smoothly backwards at 5mph, so that it thinks it's stationary when it's actually doing minus 5mph. Your throws will register 5mph faster. The acceleration it feels when moving at constant velocity is the same as it feels when stationary - vis Newton or Einstein - which would be a problem if it was easy for a human to move it at constant velocity, but to do so accurately enough to fool the very sensitive accelerometers is going to be bloody difficult. You're going to wobble about a bit.

So in practice, the only problem is if it's never stationary in the 6 seconds and the software starts trying to find a stationary point that doesn't exist. I'd prefer it to throw up an error message in that situation, but actually it does give some funky results.

But none of those funny results has any impact on its accuracy in 'normal' use.

I think it would be better if they explained how it worked a bit, so that people could understand what not to do with it. I guess they have their reasons not to.

So - you're right that it can do funky things, and i have big doubts about stokeley's 86mph. People will misuse it, sure. But i wouldn't throw the baby out with the bathwater and assume it's not normally accurate. I think it's extremely accurate (in measuring, not the simulator obviously) for normal throwing, and very useful. Chasing data isn't silly in itself, it's only silly to chase the occasional glitchy data. (But as you say, people are pretty dumb sometimes.)[/COLOR]

 

[Sheep]

The opposite - it's sometimes not recording for long enough.

All i know for sure is that it needs to be stationary to calibrate. Here's what i imagine is happening - a bit of guesswork but i bet it's not far off.

When the disc is stationary, the combined acceleration measured by all the accelerometers is just going to equal gravity (9.81m/s/s). So if it measures exactly that acceleration for some short period of time, it can assume that it's stationary, and the exact measurements on the various accelerometers, at their different angles (i.e. the individual measurements which we combined to make 9.81) can be used to calculate its orientation.

[COLOR=var(--text)]Any acceleration it feels from that point on can then be used to calculate the movement from that stationary point.

With modern accelerometers, i see no reason it can't be extremely accurate. The only issue is if the uploaded data doesn't contain a stationary point. (It records data all the time, but only uploads the 6 seconds or so up to release, so if the acceleration is never 9.81 in that period then it won't have much of a clue what 'stationary' is and you'll get weird results.)

You could try to game it by (e.g.) walking very smoothly backwards at 5mph, so that it thinks it's stationary when it's actually doing minus 5mph. Your throws will register 5mph faster. The acceleration it feels when moving at constant velocity is the same as it feels when stationary - vis Newton or Einstein - which would be a problem if it was easy for a human to move it at constant velocity, but to do so accurately enough to fool the very sensitive accelerometers is going to be bloody difficult. You're going to wobble about a bit.

So in practice, the only problem is if it's never stationary in the 6 seconds and the software starts trying to find a stationary point that doesn't exist. I'd prefer it to throw up an error message in that situation, but actually it does give some funky results.

But none of those funny results has any impact on its accuracy in 'normal' use.

I think it would be better if they explained how it worked a bit, so that people could understand what not to do with it. I guess they have their reasons not to.

So - you're right that it can do funky things, and i have big doubts about stokeley's 86mph. People will misuse it, sure. But i wouldn't throw the baby out with the bathwater and assume it's not normally accurate. I think it's extremely accurate (in measuring, not the simulator obviously) for normal throwing, and very useful. Chasing data isn't silly in itself, it's only silly to chase the occasional glitchy data. (But as you say, people are pretty dumb sometimes.)[/COLOR]

I think if I had the option to give a coding suggestion I would do it like this.

Thats why I explain it more as a coding issue, vs an actual measurement device issue as well.

But we know there is a peak point in the swing, which is basically the hit point.

We can use that extreem data there to measure backwards and then forwards.

So, we take .5seconds before the hit and use that data to start the baseline.
Because it should know what 0 is from calibration. "should"

But it seems more like its taking gathered data and then adding on top, which is why you can game it. But if we calibrate it at "still" it should know what not moving is.

A lot of these older devices, especially the ones that were in our phones, required calibration to make them read accurately a long time ago. I don't see any reason that would be any different now days. You used to have to flip your phone around in certian ways to let the measuring devices get a baseline read to figure out whats going on.

I dont' know if tech disc has any of those steps, it doesn't seem like it. Just grip and rip.
So if we can teach it what 0 is. And what the angles are by calibrating it, we should be able to not game that data anymore.

I'm not going to spend 450 dollars to figure it out, so someone that has one would have to chime in on that.
But I do know a few people who would really like to look at their code to figure out whats going on, but its not an open API from what I understand.

The idea is great.
I wont change my opinion on usage of it being a measuring tool to use occasionally, not every throw.

I did see a guy using one of the other brand ones where you just stick it on the bottom of your disc the other day and he really liked it. But I can see that data getting skewed if its not centered well enough.

 

[Nick Carroll]

I have both the GameProofer now, in addition to the Tech Disc. I just attached the GameProofer tag to a disc so I'll be testing it soon, but the design of the GP is more minimalistic, with no ports or buttons on it, and requires wireless charging, and a magnet to act as a button. This is why it's a little cheaper, but the advantage to this design is that it's less prone to water, dirt, or anything else getting inside it, and it might be more durable. So you might be more inclined to throw a GP in a field for testing, compared to a TD.

That said from what I've read online and heard anecdotally from others, the GP is less accurate right now, but like the TD it should receive software updates over time. I may try both of them in a field this weekend.

 

[Sewer bill]

I think the results can be wonky but it could be checked if someone cared to without needing to get beyond the user interface.

To it's credit, independent of whether it's accurate or not as a product it is impressively durable. So far I have seen a car hit a pole hit a roof hit and several into the ground. It still works.

 

[jupiterboy]

Aren't most of these gyro chips self calibrating in some way? Like, they can get off and then they have a way of resetting themselves? Like when you have to calibrate a phone compass or astronomy app?

Getting at the idea that they may get wonky and self correct, and if you get date while they are wonky...

 

[Sheep]

Aren't most of these gyro chips self calibrating in some way? Like, they can get off and then they have a way of resetting themselves? Like when you have to calibrate a phone compass or astronomy app?

Getting at the idea that they may get wonky and self correct, and if you get date while they are wonky...

You still have to put that stuff in calibration mode. it can't self calibrate if it doesn't know it needs to. Baseline is everything with sensors.

I think the response if you were to ask would be "they are calibrated from the factory" and I know to much about sensors to know that it doesn't take much to knock them out of wack being that I spent 5 years going around calibrating scales and other sensors on machines.

 

[jupiterboy]

You still have to put that stuff in calibration mode. it can't self calibrate if it doesn't know it needs to. Baseline is everything with sensors.

I think the response if you were to ask would be "they are calibrated from the factory" and I know to much about sensors to know that it doesn't take much to knock them out of wack being that I spent 5 years going around calibrating scales and other sensors on machines.

I can totally see into the future when the app has you "move the disc in a figure 8" for 30 seconds to recal. That will make the current model obsolete, lol.

 

[CoachChris]

Tech Disc Errors

New BB stuff hates some links and messes up media linsk>
So no preview.
Guys getting random 95mph pulls on his tech disc. hahahaha.

A lot of interesting information still coming out as people get these and explore it.

We look at stokleys throw with it the other day, he did a casual 360 and somehow got that HUGE number. When that was realistically a 60ish MPH throw.

Anyways, keep chasing data with tech disc.

He's not wrong either in the comments, it does reward poor form. Because you start chasing data instead of chasing technique.

Stokely threw many throws, no anomaly, consistently was in upper 70s with his FH. Big arm, big lever.

 

[CoachChris]

.. but like the TD it should receive software updates over time. I may try both of them in a field this weekend.

I have been told that an update later on this year will add GPS flight data instead of just geolocation.

 

[419HondaGuy]

I reached out to them last year to see if they could install their tech./sensor on a Hero disc dog disc for me. I gave the specs/weight/dimensions of the discs my dog and I utilize for long distance, Taffy supersonic 215 andHalo Superstar 235. They said they could do it, and that superstar 235 would be best choice due to it's weight being 135-140g versus supersonic at 120-125g.

Curious if the disc being so much ligher than the golf discs they use if that would affect the accuracy of the readings?

 

[CoachChris]

I reached out to them last year to see if they could install their tech./sensor on a Hero disc dog disc for me. I gave the specs/weight/dimensions of the discs my dog and I utilize for long distance,

419, is there a governing body for K9 distance, ie rules, records, etc? Or is it just some posting world record throw claims from football stadiums?

 

[419HondaGuy]

419, is there a governing body for K9 distance, ie rules, records, etc? Or is it just some posting world record throw claims from football stadiums?

World record would be Guniess Record, which has one for longest catch to dog, MTA to a dog, then few other gimmicky ones.
Each organization like quadruped, Hero Cup, Skyhoundz XD, updog's far out - all have their own rules / records ect.

 

[disc-golf-neil]

Of course there are bound to be some defective tech discs or some occasional bad readings from time to time on a tech disc. I've had mine drop 15 mph on my throws where I could feel they were the same effort and after a restart it was resolved.

Why do you think tech disc stats for various pros are becoming common knowledge among pros and pros are using it as reference points / comparisons? If it was regularly wildly inaccurate, instead of only occasionally, don't you think the pro community, who have been exposed to radar guns, would be more dismissive of the tech disc stats?

Instead, you see things like many pros aware of who has the most elite spin, and discussing and hypothesizing how they achieve it. E.g., Gannon Buhr reaching 1600's spin while many most are around 1300-1500 and very few get up to 1600's.

Also, it's starting to reveal how people wrongly assume high spin in some cases as the reason for some people getting surprisingly far distances. A good example is Albert Tamm who has fairly low spin for his speed. Jomez commentary regularly has stated "he puts so much spin on the disc" as a reason for his effortless power. When in reality, it seems like it's actually his incredibly easy access to high speeds that look slow and effortless that result in further than expected distances because it looked like he threw slow but then it went far.

@Sheep you should get a tech disc and compare it to a radar gun plus photogrammetry (Nick Krush made a tutorial vid on this) since you seem pretty invested on this topic.

 

[Sheep]

Of course there are bound to be some defective tech discs or some occasional bad readings from time to time on a tech disc. I've had mine drop 15 mph on my throws where I could feel they were the same effort and after a restart it was resolved.

Why do you think tech disc stats for various pros are becoming common knowledge among pros and pros are using it as reference points / comparisons? If it was regularly wildly inaccurate, instead of only occasionally, don't you think the pro community, who have been exposed to radar guns, would be more dismissive of the tech disc stats?

Instead, you see things like many pros aware of who has the most elite spin, and discussing and hypothesizing how they achieve it. E.g., Gannon Buhr reaching 1600's spin while many most are around 1300-1500 and very few get up to 1600's.

Also, it's starting to reveal how people wrongly assume high spin in some cases as the reason for some people getting surprisingly far distances. A good example is Albert Tamm who has fairly low spin for his speed. Jomez commentary regularly has stated "he puts so much spin on the disc" as a reason for his effortless power. When in reality, it seems like it's actually his incredibly easy access to high speeds that look slow and effortless that result in further than expected distances because it looked like he threw slow but then it went far.

@Sheep you should get a tech disc and compare it to a radar gun plus photogrammetry (Nick Krush made a tutorial vid on this) since you seem pretty invested on this topic.

I'm sitting here with a multi thousand dollar high speed camera. I'm not investing in a tech disc. hahaha.

I don't personally agree with most of the radar gun stuff either or using them to improve. I've seen radar guns give horribly bad information as well.

Tech disc is going to be good when they get it together. But putting the level of faith you do or some others do in the information is miss-guided. That is the part you all are not willing to accept when I talk about it.

Data is good.
Chasing data is bad.

We can learn a lot from data, but you gotta have an understanding of everything else going on also.

Such as the comments you made about spin there.

How do you think simon throws glitches 400+ feet? No spin or lots of spin?
What does spin do to a disc?

I keep hearing over and over and over again people constantly say that "more spin makes a disc understable" which physics show us is incorrect. We can see this is a gyroscope.

But, the relative speed/spin/stability/other is the whole equation. Not just spin/speed.

You can take some data and then just pound that data into a round hole and say "look it fits, I'm right!" and.. That's just not always how it works.

The problem with the tech disc stuff overall as a whole at the moment is just not enough of them are out there. I'm not gonna spend the money on something like that, it will not make me any money. But spending some money on a bandsaw is going to make me money. Which one do you think i'm going to buy? A tool that will make me money the rest of my life, or a measuring toy for a game I play that I don't trust yet?

Maybe i'll get one some day. But I just have questions till then and observations.

 

[disc-golf-neil]

I'm sitting here with a multi thousand dollar high speed camera. I'm not investing in a tech disc. hahaha.

I don't personally agree with most of the radar gun stuff either or using them to improve. I've seen radar guns give horribly bad information as well.

Tech disc is going to be good when they get it together. But putting the level of faith you do or some others do in the information is miss-guided. That is the part you all are not willing to accept when I talk about it.

Data is good.
Chasing data is bad.

We can learn a lot from data, but you gotta have an understanding of everything else going on also.

Such as the comments you made about spin there.

How do you think simon throws glitches 400+ feet? No spin or lots of spin?
What does spin do to a disc?

I keep hearing over and over and over again people constantly say that "more spin makes a disc understable" which physics show us is incorrect. We can see this is a gyroscope.

But, the relative speed/spin/stability/other is the whole equation. Not just spin/speed.

You can take some data and then just pound that data into a round hole and say "look it fits, I'm right!" and.. That's just not always how it works.

The problem with the tech disc stuff overall as a whole at the moment is just not enough of them are out there. I'm not gonna spend the money on something like that, it will not make me any money. But spending some money on a bandsaw is going to make me money. Which one do you think i'm going to buy? A tool that will make me money the rest of my life, or a measuring toy for a game I play that I don't trust yet?

Maybe i'll get one some day. But I just have questions till then and observations.

I will accept the tech disc data I'm getting might be wrong as soon as someone does a careful test showing it, like what I said, where you ideally have 3 instruments so that at least two converge to be more sure that one is wrong.

So far, it's just a few anecdotes of a defective disc or some on and off erroneous throw readings, but the overall consensus, even among pros, seems to be it's accurate, so that's the best assumption until someone does a more careful test.

Some people even think it's inaccurate because they can't believe how wrong they are about their stats. For example, just yesterday, I had someone threw it that I met on the course who can throw over 400 semi often and mid 400's sometimes. I've never noticed huge air bounces from him, but he was consistently between 6-10 nose up, he didn't do a drastic swoop at the last moment to create an air bounce, he threw on plane with a lower launch angle that balanced out the nose angle to not typically produce a nose up swoop.

While watching him throw, I couldn't believe it at first, I was sitting 5 feet away to his side watching closely, I couldn't see him doing pronation or obvious nose up mechanics. He then tried turning the key for the first time and got some throws where the nose angle was cut in half (+5 nose) but I was still surprised it was +5, it looked like he turned the key really well. So I stepped in for a quick throw to see if it felt like the numbers were off for me. I turned the key hard and threw -7 with a 10 deg launch angle so it didn't seem to be erroneously reading higher than usual nose angles. This is just one anecdote, but it's to point out how it's still easy for me to be suspicious and investigate. I don't blindly trust it, but based on all the available evidence, it seems the most rational position for the moment is to assume it's pretty accurate until more systematic and consistently repeatable evidence comes out to disprove that. If that happens, I hope it happens ASAP because it would make them improve the system faster. A system like this, as it continues to iteratively improve, I think is likely to easily and quickly become the best source of consistently accurate stats since it is onboard, direct readings rather than external readings.

I think I have less bias than most people. I haven't been playing long enough before the tech disc was available to develop deeply rooted beliefs and biases about my throw and throw stats in general. So it's easy for me to accept new data that contradicts previous data if it seems like it's using more accurate methods of acquisition and has reproducibility.

 

[Sheep]

I will accept the tech disc data I'm getting might be wrong as soon as someone does a careful test showing it, like what I said, where you ideally have 3 instruments so that at least two converge to be more sure that one is wrong.

So far, it's just a few anecdotes of a defective disc or some on and off erroneous throw readings, but the overall consensus, even among pros, seems to be it's accurate, so that's the best assumption until someone does a more careful test.

Some people even think it's inaccurate because they can't believe how wrong they are about their stats. For example, just yesterday, I had someone threw it that I met on the course who can throw over 400 semi often and mid 400's sometimes. I've never noticed huge air bounces from him, but he was consistently between 6-10 nose up, he didn't do a drastic swoop at the last moment to create an air bounce, he threw on plane with a lower launch angle that balanced out the nose angle to not typically produce a nose up swoop.

While watching him throw, I couldn't believe it at first, I was sitting 5 feet away to his side watching closely, I couldn't see him doing pronation or obvious nose up mechanics. He then tried turning the key for the first time and got some throws where the nose angle was cut in half (+5 nose) but I was still surprised it was +5, it looked like he turned the key really well. So I stepped in for a quick throw to see if it felt like the numbers were off for me. I turned the key hard and threw -7 with a 10 deg launch angle so it didn't seem to be erroneously reading higher than usual nose angles. This is just one anecdote, but it's to point out how it's still easy for me to be suspicious and investigate. I don't blindly trust it, but based on all the available evidence, it seems the most rational position for the moment is to assume it's pretty accurate until more systematic and consistently repeatable evidence comes out to disprove that. If that happens, I hope it happens ASAP because it would make them improve the system faster. A system like this, as it continues to iteratively improve, I think is likely to easily and quickly become the best source of consistently accurate stats since it is onboard, direct readings rather than external readings.

I think I have less bias than most people. I haven't been playing long enough before the tech disc was available to develop deeply rooted beliefs and biases about my throw and throw stats in general. So it's easy for me to accept new data that contradicts previous data if it seems like it's using more accurate methods of acquisition and has reproducibility.

I've talked to a few people now who notice the tech disc "can" be accurate, but you can cheat it. Forehands especially. To much odd movement before the throw messes it up.

So, that was where my question of calibration comes to. What/how are we calibrating it? Is it as simple as restarting the device? unacceptable then.

Granted, we can't easily calibrate a load, but we can calibrate 0 and a few other things. such as flipping the disc over. etc, sitting the disc still, so it knows what 0 is, and what level and 180 of that is.

So, since we know some odd movements before the throw could skew the data.
What does some of the random movements in the throw from pro players do to the data.

Like there are some players who throw with an open wrist position and dont actually curl the wrist in until the disc is in the pocket, then fling the wrist out.

So, some of what I'm gathering is the stuff reads pretty good, but the software interpretation is what gets messed up.

If I had one to mess with for a week, I'd have some tests I'd do to see so I could have a more accurate answer. But I get the feeling some of these weird things we see come from how the software is defining a baseline in the middle of the throw.

So, if the tech disc is seeing lets say -50 on the sensor in the middle of the throw, it perhaps might be resetting that -50 to 0 as the disc accelerates to 100 giving a reading of 150, vs actually 100.
Thats' what I"m thinking is happening on some of these really oddball readings people get. And that's all software.