Using High Speed Video to check for illegal modes and electronics cheating...

[Simon]

The following videos come from an idea I had to use my high speed video camera to verify the guns firing in relation to the intentions of the person pulling the trigger.

This would show, ramping modes, cheating modes, electronic bounce, and mechanical bounce if they add extra shots above and beyond the intentions of the person pulling the trigger.

The firing intentions are ascertained by how many times the player moves his fingers back and forth.

First gun is an 09 Bob Long Victory. It's actually his own personal gun (and one of the first prototypes) that he sent me to test out.


http://www.youtube.com/watch?v=CFNJtBNXfgY

Second gun is an 07 Ego from Planet Eclipse. A great gun, that still shoots extremely well even after years of abuse, and use in our reball test facility.


http://www.youtube.com/watch?v=OGBDgtXCaOo

[Simon]

Here's a full video of Rusty Glaze, A professional paintball player, shooting his gun.


http://www.youtube.com/watch?v=ECS1pDGkXRc

[JonoTwisT]

these are always the coolest vids to see.
I hold paintballs and handle them almost daily and just because they feel solid
you can tend to forget they are gelatinous in structure. I want a high speed
camera now to do my own testing. ;)

[Simon]

The high speed video camera is probably one of the most used, and most useful tools in our workshop.

There are cheaper and cheaper options becoming available.

I am becoming more and more tempted to pick up one of the Casio's EXILIM EX-FH20 high-speed digital SLR's or it's brethren so I can take it to more events and film more stuff without worrying about damaging a $20,000 camera...

[Simon]

Here's the original video with some footage of Rusty and my thoughts on what it shows.


http://www.youtube.com/watch?v=dr-L3WzTEDw

[63VDub]

My only thought is how would you analyze it in the game? have the output piped to a monitor and have a tech watch it real time? Or, if post game, would deductions be taken retroactively? I like it, I'm just wondering about implementing it.

[JonoTwisT]

20k. ouch. thats 2 years of my gross salary before uncle sam gets his mittens on anything.

[Simon]

My only thought is how would you analyze it in the game? have the output piped to a monitor and have a tech watch it real time? Or, if post game, would deductions be taken retroactively? I like it, I'm just wondering about implementing it.

That's the big question. To do the post processing takes a little while. That would be needed to show it definitely, but you can tell really quickly if someone is using a full auto break out mode. Quickly enough that you could pull the player from a field and have a one-4-one or whatever punishment was deemed necessary.

My biggest thing is to not penalize innocent players, but to stop the cheating.

Some of the verification would take a little time, but it would be definitive.


20k. ouch. thats 2 years of my gross salary before uncle sam gets his mittens on anything.

And that's why I am not intending to take that camera to the harsh enviroment of a paintball field just yet.

The Casio does look like a cheaper and better option for doing this. I just haven't had time to look into one in more detail.

The camera I have is probably overkill. It's capable of 16,000 frames per second. You only probably need 300+ to do this decently. I do most of my filming for this kind of thing at 1,000 frames per second, so the Casio would suffice...

[Simon]

I am bringing this up again as Chuck Hendsch again asked me about doing this to help enforce semi only at the NPPL events.

I need to nudge Gordon and see if he can replicate this with his camera so I can decide if I want to splurge on one or not...

[cockerpunk]

heres the last time i filmed trigger pulls with my camera -

http://www.youtube.com/watch?v=L5_4oLnTJK0

its about half way though. at 1200 fps i can easily capture the pulls themselves. the azodin KP pumping is actually at 300 fps, which may or may not be fast enough to catch trigger events (its plenty fast for pumps, so it should work). the 600 setting i know i can get both the trigger and fingers in enough detail, and the gun firing.

acutally i lied - http://www.youtube.com/watch?v=P2tsXsx6IQA

thats me shooting my karni full body shot at 600 fps.

works perfect.


getting that kind of view would sure be hard, but it does work pretty well. the longest part of verification would actually be having to watch the footage and get a good sample before making judgment.

[Jack Wood]

I watched some of the HB webcast, and there was no way some of those guns were in semi! Naughty Dogs looked most dubious.....

[Simon]

You could have warned me you weren't wearing any clothes. *shivers* :D

It does look like it would work.

Now to find out how cheap I can pick up one of those cameras...

[Simon]

I watched some of the HB webcast, and there was no way some of those guns were in semi!

Indeed. Some guns seemed dodgy and some were definitely a bit bouncy.

Off the break is always interesting too. It's amazing the roar of some teams guns that gets a lot less as they make it to the corners and first bunkers...

[cockerpunk]

well to be fair, the only time firepower is a commanding asset is off the break, so naturally, fast shooters will shoot the fastest then.

that being said, this is paintball. if someone can figure out a way to cheat, they will. something the woodsball leagues have not figured out. im talking to you tom cole!



the biggest trouble i see, is that reviewing even like 10 seconds of footage at 300 fps, takes 100 seconds. and at 600 fps, that takes 200 seconds. in Xball, that kind of time just wont be fast enough, unless you penalize retroactively. in 7 man, you might have enough time, but its certainly questionable. and if your taking 2 and 4 looks at it to be sure - games long over.

i still think directional mics and sound provides a far faster and easier method to check for gun speed.


EDIT - im always amazed looking at that video how much like an open bolt a gun a cocker with a slow loader really is. the block spends more time back then it does forward.

lol

[Janek]

The problem with utilising microphones and sound is that anyone with half a brain can who writes marker software can write an algorythm that will make ramping undetectable. Besides using a high-speed camera to watch trigger pulls vs shots on players suspect of cheating, the only other fool proof method would be a sensor combo mounted on the marker to monitor trigger activations and paintballs going out the barrel.

[Simon]

Gordon, We aren't looking to check the guns for speed.

We are looking to check for illegal modes and added shots.

The other issues with sound is that you can't tell if a gun is adding shots or not because there is nothing to relate it against. All it tells you is the time between shots.

Janek is absolutely correct. I've shown people two guns, one that was at a consistent rof but slower average BPS, and one that was a higher average BPS but that "stuttered"... without a question everyone thought the consistent gun was shooting at a higher ROF. It is easy (and has been done) to integrate that stuttering effect into a cheating gun.

You also can't tell the ROF at the breakouts with directional mic's because there are so many guns in the area the mic is trained on, so you can pick up multiple guns firing. There is no certainty to the results under those conditions so no way you can penalize.

The directional mics are really only good for telling ROF. In a league where ROF is supposedly unlimited, they don't make a lot of difference or give a lot of help to enforce the rules.

If we can prove without a doubt a cheating gun was used, then it doesn't matter if the penalty wasn't given in game. It can be given after, and made so severe, no-one will push the rules anymore.

[cockerpunk]

if you don't think time is important, then it should be perfect.counting pulls to shots fired takes time, and the footage itself takes a long time to review. i don't know what kind of penalty your thinking of assessing, becuase you wont be able to pull a 1 for 1 or something, it would have to be like -30 points or something.

[Janek]

Make the penalty heavy enough so that once you dish it out for the first time no one will try again... Ban the gun, eject the player from the tournament and don't allow for him to be replaced...aka Man-Down until the very end. And be very vocal about it during and after the tournament. I guess you won't have to do it more than once.

[coyut]

sounds good to me :)

[Doobie]

I'm guessing Chuck doesn't want to cap the semi? I am sure there is a physical limit to how fast someone can pull. Has to be up around 16 or so. If it was capped at that, and you used the methods already in place to detect illegal guns in conjunction with the sound timers, it might get more of the "cheating" under control. In the NCPA, we are capped at 12.5 ramp. A gun shooting too fast is obvious. It is pulled after the point, tested on a timer for verification, and then the penalty is assesed after the point in the form of putting a player in the box. (Major i think) That format alows that to happen.
NPPL would have to have a retro-active penalty. Remember the heartbreaker chrono tests AFTER a game BITD? As a team Captain, what would you do to a player who had an illegal gun if it cost you the game? :)

[weerez935]

on a gun with a microswitch (I don't think opti ones could be monitored) it seems like it would be easy to mount a small lepel uni directional microphone in the trigger to hear the microswitch click. and another on the barrel. These mics wouldn't have to be high quality because you don't want them to pick up sounds from further away. and they are much cheaper than 20k cameras.

The problem there would be if the mic picked up the cycling of the gun as opposed to the actual shot.

[Simon]

It's a good idea, but anything you have to install on every gun is either going to be a PITA to implement and keep working, or open to tampering by the players. Anything you can do remotely without changing or interfering with the players is a lot easier, and may ultimately end up cheaper anyway. I don't know the cost that virtue went to to build their RF based monitoring system, but it's not cheap, and the number of units they have out in the pro guns probably cost thousands too. Then the PITA they are to install and maintain has to be considered also.

[weerez935]

also the possiblility of breaking or falling off.

Would it be easier to only allow them a hopper that feeds 12 bps?
I guess that would not solve everything for nppl, but wouldn't you have to have a high speed on everyone? lol 10 20k$ cameras

[Simon]

A hopper that is regulated won't work. That same hopper on two different guns will give totally different rates of fire. Possibly 8bps on one gun, or even 14 on another depending on what type of loader you are using and how it works internally.

The only way to control rof accurately and fairly is at the gun.

You wouldn't need a high speed camera for every player. You'd only need one for the whole event. You can put it in a special box like road speed camera's in the UK and then have dummy boxes, but I doubt you even need that.

Players don't know they are being filmed when playing. You just need to catch a few people, publish the video, make the penalty not worth the risk, and it should stop everyone from intentionally cheating. :)

[weerez935]

True I guess I didn't think of that