smaller always better? , .684 vs. .688

[cockerpunk]

I figured as much.

Do you have a string of 100 shots on a non-breaking-in-gun so that I can re-make my point?

Again, I'd rather do it with your data than mine.

in that same data set, after the first 150 or so shots, all the way through 10 cases is the data you are looking for.

For starters, you assume I do think we're safe. Let's... stop bringing side issues into this and just talk about standard deviation.

my point was to illustrate the absurdity of 20 shots not being enough to find meaning, as every single day at every paintball field in America, far fewer shots are used to determine if people will have the gift of sight after playing paintball. i don't think that is a side issue.

I haven't eyeballed the minimum number of shots needed to get a decent average.

Average and standard deviation are not the same thing. It is entirely possible to arrive at a decent average at some number of shots, but still not have enough shots to get a decent standard deviation.

yup, i agree

I am speaking to standard deviation directly.

Again, this point can probably be made if you point me at your data for a longer run of shots.

Just looking at your breaking-in-data which you don't want to accept as valid, but let's look at it anyways:

you miss understand me ... the changes in velocity in the break in data are all explained quite easily. after the first 150 shots, the gun ran like clockwork, and we have tons of chrono data on it showing that 20 shots is more then enough to determine statistics.

At shot 512, the STDEV was 7.91, and the average was like 248.

On the other end of the range, shot 447 had STDEV of 11.64, but the average... was 243.

So 25 shots might be good enough for average, but the STDEV varies so much as to not be useful.

you can see where that number comes from, that 267 on shot 442 dragging the st dev up. i wonder if you can tell why that is .... i know why ...

alright, we can argue about st dev all day, and im happy to do that. but lets establish then a common ground ... we both agree 20 shots is effective at establishing the average of a sample. yes? this means then we can agree from the data set about the efficiency of barrels, yes? we can agree that under boring is more efficient. cause that is half the battle right there.

alright ....

ill give you consistency is a tricky one, if you note in the first test data, two .687 freak inserts shot quite different consistencies ....

[GoatBoy]

i think you need to look at what that test was about ... the reason for the sliding average is because we were examining break-in. we were looking for how fast, if at all, the gun changes speeds when broken in. so you are comparing a brand new ... literally never fired gun changing speed when breaking in ... to a gun under normal operation, and then declaring that 20 shots is not enough. in one the gun is dynamic, in the other it is not. that makes no sense at all.

I figured as much.

Do you have a string of 100 shots on a non-breaking-in-gun so that I can re-make my point?

Again, I'd rather do it with your data than mine.

and again, for a safety check, if 20 shots doesn't give you a good average of the velocity, then we are all VERY VERY VERY unsafe right now. that means the 3 to 8 shots you fire over the chrono is not a good enough sample to establish the mean, and thus very unsafe ... by your definition of what is needed to compute an average. we should be shooting hundreds of shots over the chono to get a good average by your definition in this thread.

For starters, you assume I do think we're safe. Let's... stop bringing side issues into this and just talk about standard deviation.

I haven't eyeballed the minimum number of shots needed to get a decent average.

Average and standard deviation are not the same thing. It is entirely possible to arrive at a decent average at some number of shots, but still not have enough shots to get a decent standard deviation.

I am speaking to standard deviation directly.

Again, this point can probably be made if you point me at your data for a longer run of shots.

Just looking at your breaking-in-data which you don't want to accept as valid, but let's look at it anyways:

At shot 512, the STDEV was 7.91, and the average was like 248.

On the other end of the range, shot 447 had STDEV of 11.64, but the average... was 243.

So 25 shots might be good enough for average, but the STDEV varies so much as to not be useful.

[cockerpunk]

I have measured paintball diameters. I agree a single paintball can vary .020" or more in diameter depending on where the measurement is taken. I also agree orientation is important. If diameter can vary by .020" or more based on orientation, how can you quantify .003-.005" being a good underbore and .007-.009" to be too much of an underbore when even good paint, varying .020" in diameter for any single paintball, could meet either one of those conditions depending on how it orients in the breech/barrel?

I'm not saying you are right or wrong; I'm just saying the data is incomplete to convince me at this point.

I am sincerely enjoying this thread, and I mean for all of my responses to be respectful and mature. I sympathize, as it is often hard to be on the presentation side of data. I've been there many times. I simply want to discuss the data, not attack it.


-Nathan

what i am saying is the actual measurement of a paintballs size is not a very good determination for what bore size it is.

this is why i most often refer to paint in "blow test size" not actual physical size. because the blow test functional size is the only one that actually matters.

[nak81783]

its obvious you have not measured paintballs before, nor considered what paint to barrel match actually is.

1. the min to max diameter of a paintball, even a good one, can vary by as much as .020" inches. not kidding

2. the actual size of the paintball isn't the important part of its interaction with the barrel. the interaction with the barrel be summed up with two competing values ... the friction withe barrel, and the quality of the seal with the barrel.

now lets break that down:

friction: friction is obvious with the barrel, computed with the normal force (compression in this case) and the coefficient of friction. the coefficient of friction is largely irrelevant because it is a term in very computation, and is very similar between brands, so we can effectually ignore it. this leaves friction as essentially only related to the force of compression acting on it from the barrel. essentially, the amount the ball has to change to fit in the barrel. this is related to size, but also to orientation of the ball, and out of roundess of the ball.

seal quality: this is a function of the area gap between the footprint of the ball, and barrel. this is related to size, but also to orientation of the ball, and out of roundess of the ball.

so, the ratio of these two competing qualities is what determines the performance of a ball/barrel system.

so measuring the ball is largely a waste of time, unless you measure it in relationship to these two things. the blow test measures these two competing forces directly. because the human body cannot produce much beyond a few PSI, and the resualts are clear in terms of friction, this allows us to select the "blow test size" of a paintball. now the actual size as i said is essentially irrelevant, because again, the only thing that matters in this system is those two competing forces, as it is the only set of forces the ball ever sees in the barrel. and since we determine under vs overbore based on the blow test size, the conventions are all correct. blow test your paint, and then move down 3 to 5 thousandths for ideal performance.

I have measured paintball diameters. I agree a single paintball can vary .020" or more in diameter depending on where the measurement is taken. I also agree orientation is important. If diameter can vary by .020" or more based on orientation, how can you quantify .003-.005" being a good underbore and .007-.009" to be too much of an underbore when even good paint, varying .020" in diameter for any single paintball, could meet either one of those conditions depending on how it orients in the breech/barrel?

I'm not saying you are right or wrong; I'm just saying the data is incomplete to convince me at this point.

I am sincerely enjoying this thread, and I mean for all of my responses to be respectful and mature. I sympathize, as it is often hard to be on the presentation side of data. I've been there many times. I simply want to discuss the data, not attack it.


-Nathan

[cockerpunk]

Respectfully, I truly appreciate what you've done, but a blow test is too subjective for me, especially with the earlier disagreement of what a blow test fit is/means. With a min/max diameter, one could establish a worst and best case seal or underbore condition. If the data still trended as it does now, factoring in the best and worst case, it would be a more objective study.


-Nathan

its obvious you have not measured paintballs before, nor considered what paint to barrel match actually is.

1. the min to max diameter of a paintball, even a good one, can vary by as much as .020" inches. not kidding

2. the actual size of the paintball isn't the important part of its interaction with the barrel. the interaction with the barrel be summed up with two competing values ... the friction withe barrel, and the quality of the seal with the barrel.

now lets break that down:

friction: friction is obvious with the barrel, computed with the normal force (compression in this case) and the coefficient of friction. the coefficient of friction is largely irrelevant because it is a term in very computation, and is very similar between brands, so we can effectually ignore it. this leaves friction as essentially only related to the force of compression acting on it from the barrel. essentially, the amount the ball has to change to fit in the barrel. this is related to size, but also to orientation of the ball, and out of roundess of the ball.

seal quality: this is a function of the area gap between the footprint of the ball, and barrel. this is related to size, but also to orientation of the ball, and out of roundess of the ball.

so, the ratio of these two competing qualities is what determines the performance of a ball/barrel system.

so measuring the ball is largely a waste of time, unless you measure it in relationship to these two things. the blow test measures these two competing forces directly. because the human body cannot produce much beyond a few PSI, and the resualts are clear in terms of friction, this allows us to select the "blow test size" of a paintball. now the actual size as i said is essentially irrelevant, because again, the only thing that matters in this system is those two competing forces, as it is the only set of forces the ball ever sees in the barrel. and since we determine under vs overbore based on the blow test size, the conventions are all correct. blow test your paint, and then move down 3 to 5 thousandths for ideal performance.

[cockerpunk]

Yes and no. For the safety check they just do a quick check.

People should be maintaining their equipment and doing the longer run testing on their own time, not when there are 30 other people waiting behind them at the chrono station. Completely side issue.



If I'm reading your data correctly, that 7th column is a sliding window of standard deviation per the last 25 shots right?

Because if so, your data actually proves that 25 is not enough. Thanks for doing that. It's better to show you with your own data than with mine.

Let's just consider the last 100 shots as your earlier data is just too crazy to consider.

For the last 100 shots, the standard deviation is about 9.63.

If your sliding window standard deviation of 25 varies from 7 to nearly 12, that is way, way, way too much variance for these purposes.

If you redo that sliding window to 40 or 50, you'll see that the variance reported standard deviation is way less; small enough to be useful.

i think you need to look at what that test was about ... the reason for the sliding average is because we were examining break-in. we were looking for how fast, if at all, the gun changes speeds when broken in. so you are comparing a brand new ... literally never fired gun changing speed when breaking in ... to a gun under normal operation, and then declaring that 20 shots is not enough. in one the gun is dynamic, in the other it is not. that makes no sense at all.

and again, for a safety check, if 20 shots doesn't give you a good average of the velocity, then we are all VERY VERY VERY unsafe right now. that means the 3 to 8 shots you fire over the chrono is not a good enough sample to establish the mean, and thus very unsafe ... by your definition of what is needed to compute an average. we should be shooting hundreds of shots over the chono to get a good average by your definition in this thread.

[nak81783]

yes, we did not re-chrono the gun.

the size of the paintballs was established by blow test. min/max is not a very good measurement for paintball size in relationship to barrels. blowtest is actually ideal, because it measures directly the friction vs seal that defines the relationship between a ball and the barrel.

Respectfully, I truly appreciate what you've done, but a blow test is too subjective for me, especially with the earlier disagreement of what a blow test fit is/means. With a min/max diameter, one could establish a worst and best case seal or underbore condition. If the data still trended as it does now, factoring in the best and worst case, it would be a more objective study.


-Nathan

[GoatBoy]

they should be.

Yes and no. For the safety check they just do a quick check.

People should be maintaining their equipment and doing the longer run testing on their own time, not when there are 30 other people waiting behind them at the chrono station. Completely side issue.



If I'm reading your data correctly, that 7th column is a sliding window of standard deviation per the last 25 shots right?

Because if so, your data actually proves that 25 is not enough. Thanks for doing that. It's better to show you with your own data than with mine.

Let's just consider the last 100 shots as your earlier data is just too crazy to consider.

For the last 100 shots, the standard deviation is about 9.63.

If your sliding window standard deviation of 25 varies from 7 to nearly 12, that is way, way, way too much variance for these purposes.

If you redo that sliding window to 40 or 50, you'll see that the variance in standard deviation per window is way less; small enough to be useful.

[cockerpunk]

I believe we have all mentioned paintball inconsistency. I assume the conclusions this data is expected to show are higher velocities and lower standard deviations for underbored barrels for a constant regulator setting. If so, I would want to see the diameters of the paintballs used - perhaps min/max diameter of each paintball fired - to see the paint/barrel match.

yes, we did not re-chrono the gun.

the size of the paintballs was established by blow test. min/max is not a very good measurement for paintball size in relationship to barrels. blowtest is actually ideal, because it measures directly the friction vs seal that defines the relationship between a ball and the barrel.

[nak81783]

measurement sampling? how do you mean? of what?

I believe we have all mentioned paintball inconsistency. I assume the conclusions this data is expected to show are higher velocities and lower standard deviations for underbored barrels for a constant regulator setting. If so, I would want to see the diameters of the paintballs used - perhaps min/max diameter of each paintball fired - to see the paint/barrel match.

[cockerpunk]

First, let me apologize if my post came off directed at any single individual or if it came off less tactful than intended.

Concerning the data, may I ask for the measurement sampling of the paintballs used in this test?


-Nathan

measurement sampling? how do you mean? of what?

[cockerpunk]

OK, let me take a quick look at your break-in test.

People aren't looking for standard deviation when they do a safety chrono.

they should be.

[nak81783]

oldest test we did: https://spreadsheets.google.com/pub?...sMngwmAygIkHfQ

First, let me apologize if my post came off directed at any single individual or if it came off less tactful than intended.

Concerning the data, may I ask for the measurement sampling of the paintballs used in this test?


-Nathan

[GoatBoy]

we did in the break in test, yes.

n=20 is more then enough samples to establish a mean and standard deviation of a paintball setup. if n=20 isnt enough ... then every time you chrono, and you dont fire more then 20 balls over the chrono, then you are being terribly irresponsible!

OK, let me take a quick look at your break-in test.

People aren't looking for standard deviation when they do a safety chrono.

[cockerpunk]

I haven't sifted through all of your data, but have you ever done a run of 100 and chrono'd every shot?

we did in the break in test, yes.

n=20 is more then enough samples to establish a mean and standard deviation of a paintball setup. if n=20 isnt enough ... then every time you chrono, and you dont fire more then 20 balls over the chrono without adjusting, then you are being terribly irresponsible!