Jump to content


Photo

More Shooting Data


  • Please log in to reply
39 replies to this topic

#1 Troy

Troy

    What... is the air-speed velocity of an unladen swallow?

  • Members
  • PipPipPipPip
  • 896 posts
  • Gender:Male
  • Location:Oklahoma City

Posted 09 June 2012 - 10:38 PM

Methodology:
-We first took a random sampling of two different sets of paintballs, and measured their weights.... when you have a $10k scale at your disposal, why the hell would you use anything else :rolleyes:? All balls were handled with powder free nitrile gloves.
Posted Image
-We measured the ambient air temp directly using a mercury thermometer (a digital thermometer was available, but mercury is more accurate than the cheap thermometer I had), and we got air pressure, wet bulb outside temperature and humidity from the OK mesonet.
-We mounted a 2011 Dye Nt with a Lurker .678 barrel in a craftsman table vice
Posted Image
-each ball was measured along it's seam and perpendicular to it's seam with a pair of digital calipers. We started off measuring 2 on the seam and on perpendicular to it, but the measurements across the seam were pretty close to exaclty the same, so we bumped it down to 2 measurements.
-I used a freak kit to verify that the Karnage paint was able to be blown through a .681 bore
-velocity data was collected on a chrony alpha
-our target was a piece of peg board with holes every 1 inch
-shots were taken inside a garage, to negate any wind effects
-My friend, Curtis, measured the paintballs and recorded the velocity, while I called out the shot locations. I was blind to any of the velocity readings and the individual seam measurements.
-Paintballs were loaded one at a time, by hand. Curtis wore a powder free nitrile glove to eleminate any chance that his sweat or oils would contaminate the balls.
-We used a laser pointer in the bore of the barrel to mark a point of reference to measure paintball impact locations. No attempt was made to account for drop over distance, our plan was to find the statistical center for each grouping of shots.
-We shot at a range of 10 feet, 20 feet and 30 feet.
-At 30 feet we started to gradually turn down the marker's velocity a quarter turn at a time, until it no longer cycled (at a little less then 250fps).
-We did not try to correct speed discrepancies between different kinds of paint. For example, at the 10 foot range, the HPR pressure was the same for both the Karnage and the Valken paint (you can notice a big velocity difference between the two paints)
-Towards the end, we ended up shooting my Bob Long Alias at whatever it was cronoed at, just to compare the NT's accuracy at similar speeds.

No more talk... here's the shooting data (note, this is raw data, I have not found the statistical centers of this data, yet)
Paintball Weights
Here is a power point Curtis made crunching some of the data.

And here is a picture of Dr McMurtrey, himself:
Posted Image

Edited by Troy, 11 June 2012 - 08:37 AM.

\m/

#2 Suit

Suit

    Member

  • Members
  • PipPip
  • 142 posts

Posted 09 June 2012 - 10:49 PM

The effort you put into this defiantly shows, very nice work.

#3 cockerpunk

cockerpunk

    All the Dudes

  • Moderator
  • PipPipPipPipPipPipPipPip
  • 9,115 posts
  • Gender:Male


Posted 09 June 2012 - 11:52 PM

The effort you put into this defiantly shows, very nice work.


agreed!

what is your hypothesis?
The ultimate truth in paintball is that the interaction between the gun and the player is far and away the largest factor in accuracy, consistency, and reliability.

And yes, Gordon is the sexiest manifestation of "to the front."


#4 Troy

Troy

    What... is the air-speed velocity of an unladen swallow?

  • Members
  • PipPipPipPip
  • 896 posts
  • Gender:Male
  • Location:Oklahoma City

Posted 10 June 2012 - 06:56 AM


The effort you put into this defiantly shows, very nice work.


agreed!

what is your hypothesis?


Basically, I didn't believe that velocity is correlated to accuracy (although it clearly was in rntlee's test), so I wanted to test paintballs in an under bored situation as close to the end of the barrel as possible to account for as many barrel induced effects as possible.

I also wanted to set up a system that measured as many other potential variables as possible, like ball roundness, so we could see if that was correlated to accuracy as well. The biggest criticism I had with rntlee's data set were that he didn't describe his system sufficiently, it was my goal to avoid that problem.

I didn't have a whole lot of expectations going into this test, but my goal was to get a dataset that was big enough, and to control as many variables as possible, so we could develop hypotheses, then test them against my data, and see if anything falls out.

I think I might do another set at 30 feet with dressed seams. I think it would be interesting to see if the accuracy discrepancy I see between the two different paints could be entirely accounted for by different seams. It already looks like that the paint that is the least round is the most accurate, which is counter intuitive, so i would like to figure out why. I'll also mention that by visually looking at these balls, it looks like they are both perfectly round, only when you get them in the calipers can you tell that they really aren't.

Edited by Troy, 10 June 2012 - 07:16 AM.

\m/

#5 brycelarson

brycelarson

    Show me the Data!

  • Moderator
  • PipPipPipPipPipPipPipPip
  • 8,588 posts
  • Gender:Male


Posted 10 June 2012 - 08:37 AM

nicely executed test.

I realize that it's the hardest part but distance is your friend when looking for differences. Once you start crunching the numbers we'll see what happens. Any chance you can stretch it out to at least 50 feet?

#6 Troy

Troy

    What... is the air-speed velocity of an unladen swallow?

  • Members
  • PipPipPipPip
  • 896 posts
  • Gender:Male
  • Location:Oklahoma City

Posted 10 June 2012 - 02:14 PM

nicely executed test.

I realize that it's the hardest part but distance is your friend when looking for differences. Once you start crunching the numbers we'll see what happens. Any chance you can stretch it out to at least 50 feet?


I might be able to stretch it out to like 37 feet (and 30 was a nice, even, number), but 50 feet is out of the question in my garage. Furthermore, I think that most of my eliminations occur right at about the 30 foot mark, so I think that length models reality.

Here is some number crunching...


The only variable that we found that had a significant correlation to accuracy was... believe it or not, the off seam measurement. There wasn't even any significant correlation between the ratio of seam diameter to off seam verses accuracy (believe it or not).

Edited by Troy, 10 June 2012 - 02:14 PM.

\m/

#7 supertux1

supertux1

    Member

  • Members
  • PipPip
  • 159 posts
  • Gender:Male
  • Location:Madison, WI

Posted 10 June 2012 - 02:17 PM

This data is great! I chose to do a statistical analysis on the paint measurements, since you have so many samples of each!

I calculated three things: the SD for the first diameter, the SD for the second diameter and I converted all measurements to inches.
I added a new data set: the absolute difference between the first diameter and the second diameter for each paintball, and calculated the SD of that.

Here's what I found:

SD D1 SD D2 SD ABS((D1-D2))
Valken: .008in, .006in, .009in
Karnage: .009in, .007in, .011in

I calculated a fourth and fifth data point, the average diameters of each and converted them to inches:

AVG D1, AVG D2
Valken: .667in, .679in
Karnage: .645in, .671in

General Conclusions:

1. Paintballs, no matter what brand, will always have one diameter with a greater variance than the other.
2. Higher quality paint will minimize this variance.
3. From ball to ball, higher quality paint will offer the least variances in both diameters.
4. Valken is bigger than Karnage.
5. Valken Field Paint is more consistently sized and 'rounder' from ball to ball than Karnage.
6. These differences are statistically slim and may not even be relevant in real world performance variations.

People say that the secret to accuracy is to use a good paint (consistently round, consistently sized) AND a good paint to barrel match, either underboring or overboring depending on the desire for efficiency.

Now, what does that mean in terms of real world performance? Do these fractions of an inch make any difference?

Let's test my conclusion about Valken being the best and see what the difference is between Valken and Karnage:

I'll start with the 30 foot tests

Bob Long Alias, 30 foot, @ 280FPS:

SD X, SD Y, SD FPS
Valken: 1.239in, 2.210in, 11.090fps
Karnage: 1.089in, 1.190in, 7.627fps

Interesting results for the Valken. An increase in variance in velocity leads to an increase in variance of Y hit position. No big surprise there.

Dye NT11, 30 foot, @ 280FPS:

SD X, SD Y, SD FPS
Valken: 1.732in, 1.651in, 10.421fps
Karnage: 0.786in, 1.164in, 5.67fps

Interesting results for the Valken again, same as before.

What's going on here? The more statistically round and consistently round paintball is performing relatively badly!
(But neither set is objectively badly at all, I would gladly accept any of these results for field play!)

Maybe the Karnage, despite it being less perfect than the Valken, is a better match for the barrel?

Let's compare the average diameters of both paintballs:


AVG D1, AVG D2
Valken: .667in, .679in
Karnage: .645in, .671in

Again, we see the difference in D1 vs D2 diameters for the Valken vs the Karnage.
Looking at the difference between both diameters, Valken still being closer to round than Karnage is. (.012in vs .026in difference)

Notice that the .678 diameter of the Lurker barrel falls between the average of D1 and D2 for the Valken and above both averages for the Karnage?

Maybe the Lurker is the perfect 'blow through' match, and therefore the least accurate, when matched with Valken?

Another thing to note is that both of these are REALLY small paints and the .678 isn't an underbore.

So perhaps the overall conclusion is this, sort of what we already knew:

1. The size of the paint relative to the size of the bore it is shot through is more important than the roundness.

#8 Egomaniacal

Egomaniacal

    Member

  • Members
  • PipPip
  • 139 posts
  • Gender:Male
  • Location:Maryland

Posted 10 June 2012 - 02:56 PM

Posted Image

My car got wrecked last night (it was parked, I'm fine) so I'm a bit busy, but I'll try to look at this later tonight. Lookin' good!
eiπ = − 1

#9 Troy

Troy

    What... is the air-speed velocity of an unladen swallow?

  • Members
  • PipPipPipPip
  • 896 posts
  • Gender:Male
  • Location:Oklahoma City

Posted 10 June 2012 - 03:15 PM

General Conclusions:

1. Paintballs, no matter what brand, will always have one diameter with a greater variance than the other.
2. Higher quality paint will minimize this variance.
3. From ball to ball, higher quality paint will offer the least variances in both diameters.
4. Valken is bigger than Karnage.
5. Valken Field Paint is more consistently sized and 'rounder' from ball to ball than Karnage.
6. These differences are statistically slim and may not even be relevant in real world performance variations.

People say that the secret to accuracy is to use a good paint (consistently round, consistently sized) AND a good paint to barrel match, either underboring or overboring depending on the desire for efficiency.

Now, what does that mean in terms of real world performance? Do these fractions of an inch make any difference?

Let's test my conclusion about Valken being the best and see what the difference is between Valken and Karnage:

I'll start with the 30 foot tests

Bob Long Alias, 30 foot, @ 280FPS:

SD X, SD Y, SD FPS
Valken: 1.239in, 2.210in, 11.090fps
Karnage: 1.089in, 1.190in, 7.627fps

Interesting results for the Valken. An increase in variance in velocity leads to an increase in variance of Y hit position. No big surprise there.

Dye NT11, 30 foot, @ 280FPS:

SD X, SD Y, SD FPS
Valken: 1.732in, 1.651in, 10.421fps
Karnage: 0.786in, 1.164in, 5.67fps

Interesting results for the Valken again, same as before.

What's going on here? The more statistically round and consistently round paintball is performing relatively badly!
(But neither set is objectively badly at all, I would gladly accept any of these results for field play!)


You probably started writing this post after I posted some of my number crunching crap...

There was a relative drop in accuracy from Valken to Karnage, BUT that difference wasn't significantly correlated to the roudness of the paintball. However, it was significantly correlated to the offseam dimension.

Maybe the Karnage, despite it being less perfect than the Valken, is a better match for the barrel?


This is a possibility. The valken is big enough to where it's additional size COULD have messed with the marker if most of our data came from a poppet marker... but that was one of the reasons we wanted to use a spoolie instead of a poppet based marker for most of our data. It should be less effected by the pressure in the barral.

The valken paint will not go through the .682 SS freak bore that I have. The Karnage paint will blow through there (with a good amount of effort). The valken free falls most of the way through a .687 SS freak bore (which is my next biggest size)... so I expect it is around .685ish.

Maybe the Lurker is the perfect 'blow through' match, and therefore the least accurate, when matched with Valken?


You have to REALLY blow HARD to get the Karnage to get through the lurker barrel. You have to blow the barrel like a 10 cent hooker (and cover the porting up) to get the Valken to clear the barrel. Because of your post, I did find something out... I'm fairly positive that the lurker barrel is at a .685 after the control bore. While it isn't anywhere close to a blow through match in the control bore, I bet it is a blow through match after it. Can't test that very well though, because of the porting.

AVG D1, AVG D2
Valken: .667in, .679in
Karnage: .645in, .671in


There has got to be something wrong here... either with your conversion, or my instrument. Those numbers do not illustrate how much bigger both of the balls are when compared to the Lurker barrel.

So perhaps the overall conclusion is this, sort of what we already knew:

1. The size of the paint relative to the size of the bore it is shot through is more important than the roundness.


The data seems to back up a slightly modified version of that statement:

The minimum dimension of the paint in this case is more important than the roundness.

(I didn't test several different bores, so I don't think that we have the data to say that the minimum dimension relative to the bore is important... if we could correlate several different paint sizes with several different bores, THEN I think that theory could get off the ground)

One other thing that I will mention is that the Karnage paint shell felt more "traditional," like the soft gelatin shell I am used to. The valken had a more, slick, plastic feeling shell. I'm not sure if that effected the shot pattern, The karnage could have deformed better and made a better seal (note how much faster the Karnage was then the Valken).

Edited by Troy, 10 June 2012 - 03:21 PM.

\m/

#10 supertux1

supertux1

    Member

  • Members
  • PipPip
  • 159 posts
  • Gender:Male
  • Location:Madison, WI

Posted 10 June 2012 - 04:10 PM

You probably started writing this post after I posted some of my number crunching crap...

There was a relative drop in accuracy from Valken to Karnage, BUT that difference wasn't significantly correlated to the roudness of the paintball. However, it was significantly correlated to the offseam dimension.


Right, I found that the 'roundness' didn't impact the results as much as the overall size. Did you find that the difference was related to the absolute offseam size or the variation (SD) between offseam sizes in one brand of paint?


Maybe the Karnage, despite it being less perfect than the Valken, is a better match for the barrel?


This is a possibility. The valken is big enough to where it's additional size COULD have messed with the marker if most of our data came from a poppet marker... but that was one of the reasons we wanted to use a spoolie instead of a poppet based marker for most of our data. It should be less effected by the pressure in the barral.

The valken paint will not go through the .682 SS freak bore that I have. The Karnage paint will blow through there (with a good amount of effort). The valken free falls most of the way through a .687 SS freak bore (which is my next biggest size)... so I expect it is around .685ish.


Maybe I am not being precise enough with unit conversions or you squeezed the balls a little (lol) when measuring them. I took your measurements in mm and divided them by 25.4 to come up with inches.
(FWIW: I regularly shoot valken redemption through a .682 freak, it blows through my .684 and rolls through the .687 and occasionally breaks in the .679)
In any case, the exact diameters may be off, but how you measured it and how I converted it, the Valken is consistently larger.


AVG D1, AVG D2
Valken: .667in, .679in
Karnage: .645in, .671in


There has got to be something wrong here... either with your conversion, or my instrument. Those numbers do not illustrate how much bigger both of the balls are when compared to the Lurker barrel.


Yeah I was thinking that the measured sizes were a little too small, but as long as you measured them consistently with the same method and device, some conclusions can be drawn when comparing various brands.

So perhaps the overall conclusion is this, sort of what we already knew:

1. The size of the paint relative to the size of the bore it is shot through is more important than the roundness.


The data seems to back up a slightly modified version of that statement:

The minimum dimension of the paint in this case is more important than the roundness.

(I didn't test several different bores, so I don't think that we have the data to say that the minimum dimension relative to the bore is important... if we could correlate several different paint sizes with several different bores, THEN I think that theory could get off the ground)

One other thing that I will mention is that the Karnage paint shell felt more "traditional," like the soft gelatin shell I am used to. The valken had a more, slick, plastic feeling shell. I'm not sure if that effected the shot pattern, The karnage could have deformed better and made a better seal (note how much faster the Karnage was then the Valken).


I would agree with that, a relatively small minimum dimension would affect the seal. So do you think there are other variables here we're not accounting for, such as the friction and malleability of the shell?

#11 Troy

Troy

    What... is the air-speed velocity of an unladen swallow?

  • Members
  • PipPipPipPip
  • 896 posts
  • Gender:Male
  • Location:Oklahoma City

Posted 10 June 2012 - 06:24 PM

I would agree with that, a relatively small minimum dimension would affect the seal. So do you think there are other variables here we're not accounting for, such as the friction and malleability of the shell?


The only other thing I didn't account for was the difference in the seam. Rntlee has shown that seams significantly effect the accuracy of a paintball... these seams feel the same to me. But I may not be able to perceive what could be a significant difference between the two different kinds of paint.
\m/

#12 Law

Law

    Newbie

  • Members
  • Pip
  • 32 posts

Posted 10 June 2012 - 07:43 PM

I always enjoy reading test data in this forum! If you haven't already, you might want to confirm the bore of the lurker.

#13 Spellfire

Spellfire

    Newbie

  • Members
  • Pip
  • 9 posts

Posted 10 June 2012 - 07:55 PM

Well we do have a microscope that we can use to measure the average "bump" size in the seam...

This is Curtis BTW

#14 Spellfire

Spellfire

    Newbie

  • Members
  • Pip
  • 9 posts

Posted 10 June 2012 - 07:58 PM

Although we didn't have a good way of measuring it, The Karnage is noticeably softer than the Valken paint

#15 Troy

Troy

    What... is the air-speed velocity of an unladen swallow?

  • Members
  • PipPipPipPip
  • 896 posts
  • Gender:Male
  • Location:Oklahoma City

Posted 10 June 2012 - 08:11 PM

Well we do have a microscope that we can use to measure the average "bump" size in the seam...

This is Curtis BTW


That's Dr. CURTIS!

Maybe after D-Day we can look at the seams

Edited by Troy, 10 June 2012 - 08:15 PM.

\m/

#16 Spellfire

Spellfire

    Newbie

  • Members
  • Pip
  • 9 posts

Posted 10 June 2012 - 08:30 PM


Well we do have a microscope that we can use to measure the average "bump" size in the seam...

This is Curtis BTW


That's Dr. CURTIS!

Maybe after D-Day we can look at the seams


Hell yes!

#17 Eskimo

Eskimo

    Member #: 666 ( Seriously )

  • Members
  • PipPipPipPipPipPipPip
  • 6,110 posts
  • Gender:Male
  • Location:Whitby

Posted 11 June 2012 - 07:13 AM

So the Karnage, which is "SLIGHTLY" Oval shaped, with a larger Diameter On Seam then Off seam (did I read that right?) Tends to have better groupings at larger distences.

and the NT seems more consistent then the Bob long.

Citrus Connection
Proudly Partnered With: PBL, TechPB, Pinokio Hoppers, HK·Army


#18 brycelarson

brycelarson

    Show me the Data!

  • Moderator
  • PipPipPipPipPipPipPipPip
  • 8,588 posts
  • Gender:Male


Posted 11 June 2012 - 07:51 AM

I might be able to stretch it out to like 37 feet (and 30 was a nice, even, number), but 50 feet is out of the question in my garage. Furthermore, I think that most of my eliminations occur right at about the 30 foot mark, so I think that length models reality.

Here is some number crunching...



I would take 37 over 30. Simply because we're more likley to see statistically significant differences if they're present.

There are 2 problems I see. 1. The link to your google doc isn't working for me. 2. There's no beer in the photo of your testing.

#19 Spellfire

Spellfire

    Newbie

  • Members
  • Pip
  • 9 posts

Posted 11 June 2012 - 09:42 AM


I might be able to stretch it out to like 37 feet (and 30 was a nice, even, number), but 50 feet is out of the question in my garage. Furthermore, I think that most of my eliminations occur right at about the 30 foot mark, so I think that length models reality.

Here is some number crunching...



I would take 37 over 30. Simply because we're more likley to see statistically significant differences if they're present.

There are 2 problems I see. 1. The link to your google doc isn't working for me. 2. There's no beer in the photo of your testing.


lol... can't argue with the lack of beer. Troy should have fixed the link to the powerpoint. On the first slide you can see that even at 20ft there was a significant difference in the mean accuracy as well as at 30ft. The difference between the means at both 20 and 30 feet is only about an inch, so i expect that increasing the range by 10 feet would, at best, slightly exaggerate the difference. Not so much between the markers, so we may find a difference between markers at >30ft. The data was parametric so the test in this case was a one-way ANOVA with a Turkey post-hoc test.

#20 cockerpunk

cockerpunk

    All the Dudes

  • Moderator
  • PipPipPipPipPipPipPipPip
  • 9,115 posts
  • Gender:Male


Posted 11 June 2012 - 09:51 AM

i really wonder if you can get any really good information from 30 feet. i don't disagree that lots of people get shot at 30 feet, but even at 50 feet we didn't see a large spread in shot patterns. at 30 feet your talking about a 2-4 inch pattern.
The ultimate truth in paintball is that the interaction between the gun and the player is far and away the largest factor in accuracy, consistency, and reliability.

And yes, Gordon is the sexiest manifestation of "to the front."


#21 Troy

Troy

    What... is the air-speed velocity of an unladen swallow?

  • Members
  • PipPipPipPip
  • 896 posts
  • Gender:Male
  • Location:Oklahoma City

Posted 11 June 2012 - 10:11 AM

i really wonder if you can get any really good information from 30 feet. i don't disagree that lots of people get shot at 30 feet, but even at 50 feet we didn't see a large spread in shot patterns. at 30 feet your talking about a 2-4 inch pattern.


Meh, the data speaks for itself. If you have a complaint see the ANOVA.

P.S. post number 777 bitches!

Edited by Troy, 11 June 2012 - 10:15 AM.

\m/

#22 Spellfire

Spellfire

    Newbie

  • Members
  • Pip
  • 9 posts

Posted 11 June 2012 - 10:15 AM

i really wonder if you can get any really good information from 30 feet. i don't disagree that lots of people get shot at 30 feet, but even at 50 feet we didn't see a large spread in shot patterns. at 30 feet your talking about a 2-4 inch pattern.


Well it's not a question of good data, it's just what is important to you. We did see differences in accuracy at 20 and 30 ft in paint (see the powerpoint), however the difference was only an inch. There is no doubt that if you increase the distance the spread will be greater. So at 50 ft the difference in mean accuracy may be 3in or so. Really the best part of this dataset is that there is a significant difference in accuracy between paint even at close ranges, so identifying the difference between paint will identify one of the biggest contributing factor for inaccuracies. Of the properites we measured in the balls there was two main differences between the paint: 1. The Karnage was, on average, smaller on the off-seam diameter than the Valken paint. 2. The Karnage paint was about 100mg lighter than the valken paint. Like most things, there are multiple factors that can influence accuracy, I would say that small off seam diameter and lighter mass are at least two of these factors.

#23 cockerpunk

cockerpunk

    All the Dudes

  • Moderator
  • PipPipPipPipPipPipPipPip
  • 9,115 posts
  • Gender:Male


Posted 11 June 2012 - 10:57 AM


i really wonder if you can get any really good information from 30 feet. i don't disagree that lots of people get shot at 30 feet, but even at 50 feet we didn't see a large spread in shot patterns. at 30 feet your talking about a 2-4 inch pattern.


Meh, the data speaks for itself. If you have a complaint see the ANOVA.

P.S. post number 777 bitches!


i did, but i find at some point when your statistical significance level is on the same order of magnitude as your measurements, and indeed the ball size itself ... i don't find it as compelling.

im not saying i disagree with your conclusions or the test data or anything, just saying the short range makes it less compelling.


i really wonder if you can get any really good information from 30 feet. i don't disagree that lots of people get shot at 30 feet, but even at 50 feet we didn't see a large spread in shot patterns. at 30 feet your talking about a 2-4 inch pattern.


Well it's not a question of good data, it's just what is important to you. We did see differences in accuracy at 20 and 30 ft in paint (see the powerpoint), however the difference was only an inch. There is no doubt that if you increase the distance the spread will be greater. So at 50 ft the difference in mean accuracy may be 3in or so. Really the best part of this dataset is that there is a significant difference in accuracy between paint even at close ranges, so identifying the difference between paint will identify one of the biggest contributing factor for inaccuracies. Of the properites we measured in the balls there was two main differences between the paint: 1. The Karnage was, on average, smaller on the off-seam diameter than the Valken paint. 2. The Karnage paint was about 100mg lighter than the valken paint. Like most things, there are multiple factors that can influence accuracy, I would say that small off seam diameter and lighter mass are at least two of these factors.



all good points!

i think based on rntlees data and our work with the dimpled balls shows, seam size or overall unequal aerodynamic properties is the real meat of most inaccuracy.
The ultimate truth in paintball is that the interaction between the gun and the player is far and away the largest factor in accuracy, consistency, and reliability.

And yes, Gordon is the sexiest manifestation of "to the front."


#24 Jaccen

Jaccen

    Sophomore Member

  • Members
  • PipPipPip
  • 562 posts
  • Gender:Male
  • Location:SW Ontario

Posted 11 June 2012 - 12:41 PM

This was a lot of work, you better f'in like it


I f'in like it. Thanks for providing the data.

#25 Spellfire

Spellfire

    Newbie

  • Members
  • Pip
  • 9 posts

Posted 11 June 2012 - 01:35 PM



i really wonder if you can get any really good information from 30 feet. i don't disagree that lots of people get shot at 30 feet, but even at 50 feet we didn't see a large spread in shot patterns. at 30 feet your talking about a 2-4 inch pattern.


Well it's not a question of good data, it's just what is important to you. We did see differences in accuracy at 20 and 30 ft in paint (see the powerpoint), however the difference was only an inch. There is no doubt that if you increase the distance the spread will be greater. So at 50 ft the difference in mean accuracy may be 3in or so. Really the best part of this dataset is that there is a significant difference in accuracy between paint even at close ranges, so identifying the difference between paint will identify one of the biggest contributing factor for inaccuracies. Of the properites we measured in the balls there was two main differences between the paint: 1. The Karnage was, on average, smaller on the off-seam diameter than the Valken paint. 2. The Karnage paint was about 100mg lighter than the valken paint. Like most things, there are multiple factors that can influence accuracy, I would say that small off seam diameter and lighter mass are at least two of these factors.



all good points!

i think based on rntlees data and our work with the dimpled balls shows, seam size or overall unequal aerodynamic properties is the real meat of most inaccuracy.


I'll go with that, we plan to address the seam issue in this data set by measuring the seam bump for both paint. It's a bit counter-intuitive but it seams that unequal dimensions or overall unequal aerodynamic properties increase the accuracy.

#26 cockerpunk

cockerpunk

    All the Dudes

  • Moderator
  • PipPipPipPipPipPipPipPip
  • 9,115 posts
  • Gender:Male


Posted 11 June 2012 - 01:54 PM




i really wonder if you can get any really good information from 30 feet. i don't disagree that lots of people get shot at 30 feet, but even at 50 feet we didn't see a large spread in shot patterns. at 30 feet your talking about a 2-4 inch pattern.


Well it's not a question of good data, it's just what is important to you. We did see differences in accuracy at 20 and 30 ft in paint (see the powerpoint), however the difference was only an inch. There is no doubt that if you increase the distance the spread will be greater. So at 50 ft the difference in mean accuracy may be 3in or so. Really the best part of this dataset is that there is a significant difference in accuracy between paint even at close ranges, so identifying the difference between paint will identify one of the biggest contributing factor for inaccuracies. Of the properites we measured in the balls there was two main differences between the paint: 1. The Karnage was, on average, smaller on the off-seam diameter than the Valken paint. 2. The Karnage paint was about 100mg lighter than the valken paint. Like most things, there are multiple factors that can influence accuracy, I would say that small off seam diameter and lighter mass are at least two of these factors.



all good points!

i think based on rntlees data and our work with the dimpled balls shows, seam size or overall unequal aerodynamic properties is the real meat of most inaccuracy.


I'll go with that, we plan to address the seam issue in this data set by measuring the seam bump for both paint. It's a bit counter-intuitive but it seams that unequal dimensions or overall unequal aerodynamic properties increase the accuracy.


i would assume it would actually have a pretty small effect. there sphereicity could probably be off by quite a lot before the fundamental aerodynamics would be seriously effected.a fat 5-25 mil sharp seam though i would think would be a MAJOR problem. toss a couple hundred RPM spin on there (typical) and you get what you get.

Edited by cockerpunk, 11 June 2012 - 01:55 PM.

The ultimate truth in paintball is that the interaction between the gun and the player is far and away the largest factor in accuracy, consistency, and reliability.

And yes, Gordon is the sexiest manifestation of "to the front."


#27 Snipez4664

Snipez4664

    Junior Member

  • Members
  • PipPipPipPip
  • 785 posts
  • Gender:Male

Posted 11 June 2012 - 02:03 PM

So the question is, do seams create spin, amplify spin's effect, or both.


and, for each of the 2 factors, to what degree? The NT did shoot a slightly better pattern than the alias, which surprised me.
Posted Image

This post brought to you by: Lurker
Owner/Operator/Lead Engineer - Lurker Paintball
Check out our products at Lurker Paintball

#28 cockerpunk

cockerpunk

    All the Dudes

  • Moderator
  • PipPipPipPipPipPipPipPip
  • 9,115 posts
  • Gender:Male


Posted 11 June 2012 - 02:39 PM

So the question is, do seams create spin, amplify spin's effect, or both.


and, for each of the 2 factors, to what degree? The NT did shoot a slightly better pattern than the alias, which surprised me.


i don't think the spin is amplified by the seams, i think the seams simply effect the vortex formation and separation, and due to spin, they are dynamic. this leads to a non-normal distribution in the random walk.
The ultimate truth in paintball is that the interaction between the gun and the player is far and away the largest factor in accuracy, consistency, and reliability.

And yes, Gordon is the sexiest manifestation of "to the front."


#29 Snipez4664

Snipez4664

    Junior Member

  • Members
  • PipPipPipPip
  • 785 posts
  • Gender:Male

Posted 11 June 2012 - 03:18 PM

Totally disagree, seam position has a strong effect on the effectiveness of backspin systems
Posted Image

This post brought to you by: Lurker
Owner/Operator/Lead Engineer - Lurker Paintball
Check out our products at Lurker Paintball

#30 brycelarson

brycelarson

    Show me the Data!

  • Moderator
  • PipPipPipPipPipPipPipPip
  • 8,588 posts
  • Gender:Male


Posted 11 June 2012 - 03:45 PM

any ideas on how to quantify ball surface? you know, slickness, gloss, whatever we decide to call it.

#31 Snipez4664

Snipez4664

    Junior Member

  • Members
  • PipPipPipPip
  • 785 posts
  • Gender:Male

Posted 11 June 2012 - 03:58 PM

Would be a pain, but measured CoF would be best. Easiest way to do it is probably to use a barrel with an underbored and a squeegee with weights or something.
Posted Image

This post brought to you by: Lurker
Owner/Operator/Lead Engineer - Lurker Paintball
Check out our products at Lurker Paintball

#32 Troy

Troy

    What... is the air-speed velocity of an unladen swallow?

  • Members
  • PipPipPipPip
  • 896 posts
  • Gender:Male
  • Location:Oklahoma City

Posted 11 June 2012 - 04:45 PM

Would be a pain, but measured CoF would be best. Easiest way to do it is probably to use a barrel with an underbored and a squeegee with weights or something.


I think that would be too dependent on paintball size.

Popping the paintball and testing the shell directly may be the best option.

Maybe melt it down into a cube or something and test the deformation of a certain, standard, weight.

Edited by Troy, 11 June 2012 - 04:59 PM.

\m/

#33 cockerpunk

cockerpunk

    All the Dudes

  • Moderator
  • PipPipPipPipPipPipPipPip
  • 9,115 posts
  • Gender:Male


Posted 11 June 2012 - 06:29 PM

Totally disagree, seam position has a strong effect on the effectiveness of backspin systems


yes, in the gun, i thought you meant the airflow
The ultimate truth in paintball is that the interaction between the gun and the player is far and away the largest factor in accuracy, consistency, and reliability.

And yes, Gordon is the sexiest manifestation of "to the front."


#34 Spellfire

Spellfire

    Newbie

  • Members
  • Pip
  • 9 posts

Posted 11 June 2012 - 06:40 PM





i really wonder if you can get any really good information from 30 feet. i don't disagree that lots of people get shot at 30 feet, but even at 50 feet we didn't see a large spread in shot patterns. at 30 feet your talking about a 2-4 inch pattern.


Well it's not a question of good data, it's just what is important to you. We did see differences in accuracy at 20 and 30 ft in paint (see the powerpoint), however the difference was only an inch. There is no doubt that if you increase the distance the spread will be greater. So at 50 ft the difference in mean accuracy may be 3in or so. Really the best part of this dataset is that there is a significant difference in accuracy between paint even at close ranges, so identifying the difference between paint will identify one of the biggest contributing factor for inaccuracies. Of the properites we measured in the balls there was two main differences between the paint: 1. The Karnage was, on average, smaller on the off-seam diameter than the Valken paint. 2. The Karnage paint was about 100mg lighter than the valken paint. Like most things, there are multiple factors that can influence accuracy, I would say that small off seam diameter and lighter mass are at least two of these factors.



all good points!

i think based on rntlees data and our work with the dimpled balls shows, seam size or overall unequal aerodynamic properties is the real meat of most inaccuracy.


I'll go with that, we plan to address the seam issue in this data set by measuring the seam bump for both paint. It's a bit counter-intuitive but it seams that unequal dimensions or overall unequal aerodynamic properties increase the accuracy.


i would assume it would actually have a pretty small effect. there sphereicity could probably be off by quite a lot before the fundamental aerodynamics would be seriously effected.a fat 5-25 mil sharp seam though i would think would be a MAJOR problem. toss a couple hundred RPM spin on there (typical) and you get what you get.

Well, the data say that a 0.5 mm decrease in offseam diameter increases the accuracy by 1 inch which isn't that great. so while offseam diameter is a factor in accuracy it may not be the major factor. I do have to say that Karnage paint shell is more soft than the valken and it is more shiny... for whatever that is worth

#35 Troy

Troy

    What... is the air-speed velocity of an unladen swallow?

  • Members
  • PipPipPipPip
  • 896 posts
  • Gender:Male
  • Location:Oklahoma City

Posted 11 June 2012 - 07:04 PM

Well, the data say that a 0.5 mm decrease in offseam diameter increases the accuracy by 1 inch which isn't that great. so while offseam diameter is a factor in accuracy it may not be the major factor. I do have to say that Karnage paint shell is more soft than the valken and it is more shiny... for whatever that is worth


Got any ideas on how to quantify the difference?... besides dissolving it and running it through a HPLC.

Edited by Troy, 11 June 2012 - 07:04 PM.

\m/

#36 Spellfire

Spellfire

    Newbie

  • Members
  • Pip
  • 9 posts

Posted 11 June 2012 - 07:06 PM

any ideas on how to quantify ball surface? you know, slickness, gloss, whatever we decide to call it.


We plan on taking high magnification images of the seam and measure the length of the bump. I suppose we can attempt to measure reflectivity of white light as a measure of gloss. However we don't have the equipment to measure friction of the surface. I will say that the karnage paint has a shiny smooth surface while the valken pain that matte, rough surface. We should be able to use the confocal microscope to take an image of the shell of each paint to give a better handle on the differences of the surface characteristics (microscopic dimples) of the shell.

#37 Spellfire

Spellfire

    Newbie

  • Members
  • Pip
  • 9 posts

Posted 11 June 2012 - 07:10 PM

we can also measure shell thickness, I imagine the shell is thinner on the Karnage paint

#38 brycelarson

brycelarson

    Show me the Data!

  • Moderator
  • PipPipPipPipPipPipPipPip
  • 8,588 posts
  • Gender:Male


Posted 12 June 2012 - 07:18 AM

We plan on taking high magnification images of the seam and measure the length of the bump. I suppose we can attempt to measure reflectivity of white light as a measure of gloss. However we don't have the equipment to measure friction of the surface. I will say that the karnage paint has a shiny smooth surface while the valken pain that matte, rough surface. We should be able to use the confocal microscope to take an image of the shell of each paint to give a better handle on the differences of the surface characteristics (microscopic dimples) of the shell.


cool.

I just think it might be an important piece of information before we jump onto the roundness as impacting accuracy. The interaction of the ball surface to barrel might have something to do with the variation between the guns. I wonder what the breech pressure is in the two guns? I could see a higher breech pressure evening out ball surface variations somewhat. In other words - if the breech pressure is lower increases or decreases in the COF might create larger variations in consistency as well as possible changing ball rotation.

My suggestion would be to repeat one of the sample sets - the longer the better - with balls with seams dressed. See if the difference between the two balls closes. That data should help determine how much of the difference is seams. With that piece we can then move onto trying to pin down what's creating the differences.

Edited by brycelarson, 12 June 2012 - 07:38 AM.


#39 Troy

Troy

    What... is the air-speed velocity of an unladen swallow?

  • Members
  • PipPipPipPip
  • 896 posts
  • Gender:Male
  • Location:Oklahoma City

Posted 12 June 2012 - 09:00 AM

My suggestion would be to repeat one of the sample sets - the longer the better - with balls with seams dressed. See if the difference between the two balls closes. That data should help determine how much of the difference is seams. With that piece we can then move onto trying to pin down what's creating the differences.


For our next test, we'll put the marker right against the wall and measure the distance.

I really wanted to get a dual chrono setup so we could measure the delta between the two cronos and see if that correlated to accuracy... but sadly, another chrono (besides my own) never materialized. If we found a good correlation, then we would prove rather conclusively that effects during the ball's flight effect accuracy the most.
\m/

#40 Egomaniacal

Egomaniacal

    Member

  • Members
  • PipPip
  • 139 posts
  • Gender:Male
  • Location:Maryland

Posted 14 June 2012 - 06:06 PM

Did you get around to analyzing the Valken off-seam diameter vs. Accuracy?
eiπ = − 1




1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users