It's beginning to look like the iphone gives better/more consistent results than the BB, presumably due to the difference in sound processing on the BB as mentioned on their website.
Break speed? Theres an app 4 that....
- Thread starter Thunderball
- Start date
It's beginning to look like the iphone gives better/more consistent results than the BB, presumably due to the difference in sound processing on the BB as mentioned on their website.
Follow us on Twitter... We'll be posting updates about it there. twitter.com/breakspeed.
Think of it as "Voice recognition" on steroids. We're a lot more accurate and reliable than VR, but once in a while it will mis-read a break shot. When this happens, you can still get your break speed. You can edit the waveform and find the points yourself. Take a look at the waveform for a few successful breaks (or look at our screenshot on the website) and you'll get an idea how to pick the right points. Make sure you zoom in to get an accurate result.
Very rarely will you ever come across a break speed that you just can't get, even manually. I've only seen this when somebody breaks at the exact same time you do and the breaks overlap.
More specifically, this is an issue with devices that use Auto Gain Control. The BB devices have this and it cannot be disabled.
Many Windows Mobile devices allow you to enable/disable this and the default is off (good.) However, I know of one device that has it on and cannot be disabled.
It looks like the Android devices do not have AGC, but in the future, a manufacturer could produce an AGC device, so we're hoping this doesn't happen.
The iPhone devices are all the same, and (so far) none of them have AGC (and are not likely to ever have it.)
We're working on ways to improve the break recognition that completely ignores the amplitude decimation that comes from AGC, which could put this entire issue to rest.
So I remembered to bring my headphones with a mic (I ordered the mini mic but it hasn't arrived yet) to the pool hall with me today. I'm kinda sick so my speed were a little low. My average was only 23.78 with my high being 24.91. I've hit 27.56 on the gun in Vegas so hopefully once I'm feeling better I be able to get above that. I can already see this app getting a ton of use from me and my friends.
If you're going to compare Break Speed with a radar gun, there are some things you should know. You may think I'm biased, but I'm going to stick to facts and be completely fair.
Also, most pool players that have ever had their speed tested with a gun should probably follow along with this, even if they don't use Break Speed because it can change the way you think about the gun.
Most people know that RADAR works by timing the difference between how long it takes for signals to be reflected back to the gun.
Now, consider standing on the side of the road pointing a gun straight across the road, waiting for a car to drive through the radar gun's beam. As a car passes through the beam, the car is not getting any closer or further from the gun (it's going completely sideways through the beam), so the signals bouncing off the car back to the gun all bounce back with the same round-trip time. To the gun, the car will appear to be traveling at 0MPH.
Ask any police officer, and they will confirm this - the most accurate result is when they're traveling directly toward or away from you. Any angle you introduce works in the driver's favor (it produces a slower rate of speed.)
When they time a break shot, you can't get the cue ball coming directly at the gun. There will always be some angle because the rack is in the way. When we had our speed tested at the booth in Vegas, they held the gun directly over the head ball of the rack and pointed it at roughly a 45-degree angle down. The cue ball would enter the beam about a foot from the rack.
So if the gun was held at a 45-degree angle, wouldn't it produce a speed at roughly half the correct speed? Not exactly half, but close enough for this discussion. Yet the gun produced reasonable results. How?
I don't know this for sure, but I can imagine that this means that they probably calibrated it with a special tuning fork (which is how you calibrate a radar gun.) This could ~double the speeds produced by the gun to correct for the angle. This will have a negative effect on the accuracy of the result, but there is actually more to it than that.
We now have a human being holding a gun at the end of an extended arm. Watching closely, I saw as much as a 15-degree difference in the way she was holding the gun each time she would record a break speed. This is where the real error is going to come from.
To be fair, Break Speed as a similar problem, in that we rely on you (a human) to set the cue ball position accurately within the application and on the table. This is why we recommend, if you are working on your break and you need absolute accuracy, to break from the spot or from a carefully marked intersection of the diamonds on the table. This way, you can get the most accuracy in placing the cue-ball within the application and on the table. Check our FAQ for more details. With this, you should be able to get extremely accurate results (hundredths of a mile per hour.)
Additional error from radar guns comes from guns that are not properly tuned. Not only do you have to use a tuning fork, but that fork must be struck on a certain kind of surface. For example, the police are taught to strike it on the heel of their shoe. Furthermore, the placement of the radar gun so close to the subject being timed will introduce additional inaccuracy.
So although nobody has yet to complain that the speeds coming from Break Speed don't match the speeds they get from a radar gun, if you have found that to be the case, now you know why.
I did try to do this, but it was not obvious by looking at the waveform, even zoomed in. It was very odd. Is it possible to export the waveform to load into a 'real' audio editing software package?
If you're using a WM or BB version, yes. The iPhone version doesn't support saving/loading breaks yet. If you save the file on a BB or WM version, then you can email it to me (hit the support page on our website and look for my name - there is an email link directly to me.)
I didn't bother saving it at the time. But I will next time it happens.
Another note of the way this app calculates speed vs the radar gun:
This application give you the average speed from the time the cue ball was struck, until the time the cueball hit the rack.
A radar gun (I believe) will give you the speed as the cueball is in its detection range. So if the radar gun was aimed close to the starting position of the cueball, the speed reported would be higher than if the radar gun was aimed close to the rack position. This is because the cueball is constantly slowing down as its moving due to the friction between the cloth and the cueball.
This, along with the inconsistencies mentioned by the app's creator due to human error when using the radar gun, make this method of speed detection much better.
This application give you the average speed from the time the cue ball was struck, until the time the cueball hit the rack.
A radar gun (I believe) will give you the speed as the cueball is in its detection range. So if the radar gun was aimed close to the starting position of the cueball, the speed reported would be higher than if the radar gun was aimed close to the rack position. This is because the cueball is constantly slowing down as its moving due to the friction between the cloth and the cueball.
This, along with the inconsistencies mentioned by the app's creator due to human error when using the radar gun, make this method of speed detection much better.
I have a lg tritan but would love to have this app is are lg going to be something that will get this app when you have time.
Honestly, this was way too long for me to read right now cause I'm tired. I read a little of it and I think I got what you were saying. I wasn't complaining that the speed I was getting from Break Speed was different than what I saw on the gun. I pretty much knew it was going to be different because I'm sick and if I would have hit 27+ I would have fallen over dead. lol. I'm honestly kinda shocked I hit over 24 being sick. Hopefully in the next few days I'll be feeling better and I can get back over 27.
BlackBerry Break Speed
Hey Guys,
Please send your BlackBerry issues to c-support@mybreakspeed.com . I have a BB myself and that is why I wrote the BB version. Below are some points that I hope help explain what is going on.....
--- NOTE - THIS IS ALL SPECIFIC TO THE BLACKBERRY VERSION! ---
AUTO DETECTION OF THE BREAK (aka sometimes I get weird results)
As Paul has mentioned before the BlackBerry has an Auto Gain issue so the data we get from the microphone is modified before we ever get it.
If you get strange results from the auto-detection please check the HELP SCREEN for instructions on using the tools provided to find break speeds manually. I wish it weren't necessary but the reality is that the BlackBerry device sometimes requires you to manually find your own break. I added tools to the BB version to try and make this easier for you (for example BB is the only device where you can play back the audio and that is to help you find the break manually). Use the tools and you can find your break in nearly any situation (playing pool next to a live band is the only time I couldn't).
If you work with the software a bit you can get quite good at finding the break speed manually. When you first start it seems difficult but it doesn't take long to start seeing trends in what is going on (most of the time when auto-detection gives you a weird number it found part of the break and you can use what it found to home in on the right data).
SUPPORTED BB MODELS
Someone posted that they could not load our software on an 8330 device. I have PMd that person to try and get details so I can help them. My lowest end test device is a BlackBerry Pearl 8130 and it runs like a champ (this is a dinky little candy bar style phone). I did upgrade the OS on that device to version 4.5. If you are running an older OS I recommend you upgrade to 4.5 at a minimum (on the options screen click about to see what version you are running). I don't know of any reason why the software would not work on an 8330. BB has at least 20 different models in use and many I have only been able to try in a simulator (it isn't practical for me to buy one of each of them). Please use the trial download to make sure it works on your device.
INSTALLATION ON BB
BB install tools don't always work as designed. Sometimes when I load software it says "successful" but the software has a generic icon and does not function. On the BB I went with a green theme so the icon should be green and it should say MPH in it. If you don't see that icon after installing then your install tools failed and you have to uninstall and try again. In my experience this happens at least 1 in 10 tries and I have had it happen to me twice in a row a couple of times (but never 3x in a row).
USER VIDEO
I am working on a user video I hope to complete over memorial day weekend. It will show how I use the software and the process for manually finding your break speed. Please check mybreakspeed.com next week for that video.
If you have problems please contact me at c-support@mybreakspeed.com . I have a day job outside of the Break Speed project but I still respond to those emails several times a day and I will do everything I can think of to help you.
Regards,
Craig York
c-support@mybreakspeed.com
TOP Break Speed: 24.2MPH
Hey Guys,
Please send your BlackBerry issues to c-support@mybreakspeed.com . I have a BB myself and that is why I wrote the BB version. Below are some points that I hope help explain what is going on.....
--- NOTE - THIS IS ALL SPECIFIC TO THE BLACKBERRY VERSION! ---
AUTO DETECTION OF THE BREAK (aka sometimes I get weird results)
As Paul has mentioned before the BlackBerry has an Auto Gain issue so the data we get from the microphone is modified before we ever get it.
If you get strange results from the auto-detection please check the HELP SCREEN for instructions on using the tools provided to find break speeds manually. I wish it weren't necessary but the reality is that the BlackBerry device sometimes requires you to manually find your own break. I added tools to the BB version to try and make this easier for you (for example BB is the only device where you can play back the audio and that is to help you find the break manually). Use the tools and you can find your break in nearly any situation (playing pool next to a live band is the only time I couldn't).
If you work with the software a bit you can get quite good at finding the break speed manually. When you first start it seems difficult but it doesn't take long to start seeing trends in what is going on (most of the time when auto-detection gives you a weird number it found part of the break and you can use what it found to home in on the right data).
SUPPORTED BB MODELS
Someone posted that they could not load our software on an 8330 device. I have PMd that person to try and get details so I can help them. My lowest end test device is a BlackBerry Pearl 8130 and it runs like a champ (this is a dinky little candy bar style phone). I did upgrade the OS on that device to version 4.5. If you are running an older OS I recommend you upgrade to 4.5 at a minimum (on the options screen click about to see what version you are running). I don't know of any reason why the software would not work on an 8330. BB has at least 20 different models in use and many I have only been able to try in a simulator (it isn't practical for me to buy one of each of them). Please use the trial download to make sure it works on your device.
INSTALLATION ON BB
BB install tools don't always work as designed. Sometimes when I load software it says "successful" but the software has a generic icon and does not function. On the BB I went with a green theme so the icon should be green and it should say MPH in it. If you don't see that icon after installing then your install tools failed and you have to uninstall and try again. In my experience this happens at least 1 in 10 tries and I have had it happen to me twice in a row a couple of times (but never 3x in a row).
USER VIDEO
I am working on a user video I hope to complete over memorial day weekend. It will show how I use the software and the process for manually finding your break speed. Please check mybreakspeed.com next week for that video.
If you have problems please contact me at c-support@mybreakspeed.com . I have a day job outside of the Break Speed project but I still respond to those emails several times a day and I will do everything I can think of to help you.
Regards,
Craig York
c-support@mybreakspeed.com
TOP Break Speed: 24.2MPH
Ok, here is my review and suggestions:
The Positives:
Overall a very nice app. It really helps drive consitency in your break, versus just the feel of it how hard I think I hit it. You may have had a bad rack, etc and start messing with speed/force instead of sticking with what works because of a bad rack. The thing I have really noticed is with solid racks in 9 ball how well I can control the table when I am actually hitting the ball the same each time.
The Suggestions:
As far as true accuracy (i.e. 22.78 mph is really 22.78 mph), that is VERY debatable. It takes a very small amount of change in cue ball placement during configuration to change the calculated mph by several miles. (i.e. a cue ball width is equal to 3 mph pretty easily).
A side effect/related problem is the fact that the 7' size is based on 39" x 78" nose to nose dimension. I am not sure why this is since most tables I am familiar with are actually 40"x80" nose to nose.
These two items together become a big deal if you are trying to get a real number out of the system (basically your configuration is off by a cue ball width of error on the spots alone compared to reality just by not having the proper size table to start with).
Verdict:
So overall I would suggest that if you only want to know how fast you are really hitting the cue ball, get a very accurate placement of the cue ball using head/foot spot combinations (obviously another 7' size would be very helpfull), but if you are just looking to get relative speeds from one attempt to another for consistency then the app is great for that regardless of the table size/accuracy of the cue ball starting position.
Other Suggestions:
I like the idea of a history as has been suggested, but really unless the measurements are based on a very specific set of conditions, the accuracy of comparing who is the fastest break speed really is pointless at best.
Hopefully this comes off as someone who likes the app and thinks its a good purchase, but at the same time sees some good revisions for version 2.
Best of luck.
The Positives:
Overall a very nice app. It really helps drive consitency in your break, versus just the feel of it how hard I think I hit it. You may have had a bad rack, etc and start messing with speed/force instead of sticking with what works because of a bad rack. The thing I have really noticed is with solid racks in 9 ball how well I can control the table when I am actually hitting the ball the same each time.
The Suggestions:
As far as true accuracy (i.e. 22.78 mph is really 22.78 mph), that is VERY debatable. It takes a very small amount of change in cue ball placement during configuration to change the calculated mph by several miles. (i.e. a cue ball width is equal to 3 mph pretty easily).
A side effect/related problem is the fact that the 7' size is based on 39" x 78" nose to nose dimension. I am not sure why this is since most tables I am familiar with are actually 40"x80" nose to nose.
These two items together become a big deal if you are trying to get a real number out of the system (basically your configuration is off by a cue ball width of error on the spots alone compared to reality just by not having the proper size table to start with).
Verdict:
So overall I would suggest that if you only want to know how fast you are really hitting the cue ball, get a very accurate placement of the cue ball using head/foot spot combinations (obviously another 7' size would be very helpfull), but if you are just looking to get relative speeds from one attempt to another for consistency then the app is great for that regardless of the table size/accuracy of the cue ball starting position.
Other Suggestions:
I like the idea of a history as has been suggested, but really unless the measurements are based on a very specific set of conditions, the accuracy of comparing who is the fastest break speed really is pointless at best.
Hopefully this comes off as someone who likes the app and thinks its a good purchase, but at the same time sees some good revisions for version 2.
Best of luck.
For the most part, your notes are correct. In general, though, what matters most about cue-ball position is the distance to the head ball. So if you're breaking from the head string (which is marked in the application) then you're likely to get a more accurate result because you're placing the cue ball with some reference. Even if your cue ball placements don't exactly match (horizontally along the head string), you will still get accurate results because the distance is far less affected from horizontal error than 'distance' error.
If you read our FAQ, I point out that if you want the ultimate accuracy, break from the spots, and make sure that your cue ball (and head ball) are properly centered on the spot. The Windows Mobile version comes with a button to "spot cue ball" - this is coming in the iPhone version.
Did you measure the table yourself? If so, which table is this? I ask because I spent a lot of time researching this and had very long conversations about the size of a 7-foot table (including non-standard sizes.)
In truth, there is very little information out there, and MOST of it is wrong. In fact, the table sizes on Diamond's website are incorrect. Also the regulation table sizes in the BCA rule book for the professional players is also wrong (I believe they were copied from the Diamond site.)
In the end, I left it up to the professionals and from an unlikely source: the slate manufacturer, Original Italian Slate. One of the 'professionals' has sold tables and has done table maintenance his whole life (he measures all kinds of tables regularly.) The second person is the guy that made my custom table (Art Ball) - he has been making tables for the last 20 years. When I told them what Diamond's site said the sizes of tables were, they both argued against those numbers vehemently. Then we had one measured. These findings were confirmed with OIS for all table sizes.
So if you actually measured the table and found that it is different from what the software says, please let me know!
On the WM version, you can actually specify custom table sizes, in one-inch increments. However, I decided to leave this out of the iPhone version because non-standard table sizes are very rare and it made the UI much harder to use. I will likely remove this from the WM version so that it can mirror the cleaner and simpler interface from the iPhone version.
No worries - all feedback (positive or negative) is good. If we screwed something up, we want to know about it so we can correct it.
If you have a chance, please go measure the table you're talking about, nose to nose (pull the tape tight!)