What bob said!
Break speed? Theres an app 4 that....
- Thread starter Thunderball
- Start date
What bob said!
We're making good progress. Most of the difficult stuff is in and working (you can capture breaks, view/pan/zoom waveforms, etc.) However, we've run into a speed-bump with the EVO. There are some specific things that HTC has done in their version of the 2.1 OS that are causing havoc.
These are specific to the EVO and not other devices running 2.1. I'm in contact with them, and they are aware of the issue and claim there is no way around it (it's a side-effect of a new "feature" they added.) I, however, am not quite so easily dissuaded.
They (HTC) really need to release the source to their Android kernel modifications for this device - they've released kernel source for all other devices and it's required as part of the open source license!
I'm staying productive, though, continuing to make great progress on the app (I'm up until 5am every night, and I have a day job) while I wait for various research threads into this problem to bear fruit.
That's all I got. Sorry if it's not what you were hoping for. Best guess... 2 weeks (maybe 3) from this point.
If you don't like my answer, please contact HTC support and request a kernel source drop for the EVO (code-named "Supersonic").
These are specific to the EVO and not other devices running 2.1. I'm in contact with them, and they are aware of the issue and claim there is no way around it (it's a side-effect of a new "feature" they added.) I, however, am not quite so easily dissuaded.
They (HTC) really need to release the source to their Android kernel modifications for this device - they've released kernel source for all other devices and it's required as part of the open source license!
I'm staying productive, though, continuing to make great progress on the app (I'm up until 5am every night, and I have a day job) while I wait for various research threads into this problem to bear fruit.
That's all I got. Sorry if it's not what you were hoping for. Best guess... 2 weeks (maybe 3) from this point.
If you don't like my answer, please contact HTC support and request a kernel source drop for the EVO (code-named "Supersonic").
Let me start by saying that I own an EVO and I'm not about to have a version of Break Speed for my personal use that is inferior to any other version out there.
So tonight I didn't just attack the problem, I launched a full-out assault on it...
Android is open-source, which means I was able to download and browse the source code for the entire operating system (it's ~3GB of code.) This means that nothing can hide from you - if there's a problem, you'll at least be able to understand it and (if you're lucky), maybe discover a work-around.
I found the solution.
With this speed-bump is now behind me, it's full-steam ahead (which is good, because the stuff I have leftover is the fun stuff.)
Now I just need to email HTC back and tell them that there is a solution. Apparently, this is a bit of a discovery - there are others having the same problem. I should also post notes to them in the various forums to help them through this problem, too.
So tonight I didn't just attack the problem, I launched a full-out assault on it...
Android is open-source, which means I was able to download and browse the source code for the entire operating system (it's ~3GB of code.) This means that nothing can hide from you - if there's a problem, you'll at least be able to understand it and (if you're lucky), maybe discover a work-around.
I found the solution.
With this speed-bump is now behind me, it's full-steam ahead (which is good, because the stuff I have leftover is the fun stuff.)
Now I just need to email HTC back and tell them that there is a solution. Apparently, this is a bit of a discovery - there are others having the same problem. I should also post notes to them in the various forums to help them through this problem, too.
Can you tell us in layman terms what the problem was and your solution? I also have the Evo and want to know how fundamentally different Android is with Sense.
Thanks!
The problem was largely internal with the way they were delivering audio data. It turns out that HTC took advantage of a new feature of the 2.1 operating system.
The documentation for Android development is sketchy at best, but there's so much community support out there, that it really doesn't matter. However, not a lot of people are doing voice-recognition-like work like we are, and so our problems are sometimes harder to find solutions for. Once I browsed the source code, I was able to get a better understanding of what the documentation was trying to tell me about this new feature - they literally devoted 10 words of documentation to a very complex feature, leaving out the details that I needed.
Again, I'm actually not unhappy - the lack of documentation is so often made up by better community support than documentation could hope to offer, and availability of the source code is far more information that the documentation could ever achieve.
Don't let this post scare you about the EVO. It's the absolute best Android device on the market. Android is a very clean design running on a solid core operating system and HTC didn't screw it up. And although I haven't seen the code for Sense, it behaves as though it also has a very clean design behind it as well.
i finally remembered to use this app. just a few minutes ago
anyone who dosent buy it shouldnt own a cue. Go bowling or something.
I only hit 3 balls all day, I been in a slump and been playing 1P, so I took today off pool, I saw the app in my phone and decided to see what I could come up with. i was cold, stiff hadnt hit a ball nothing and was over 22 all 3 times using my player. I was amazed I was over 15-my break blows.
I'm going to have fun playing with this.
great app.
Best
eric
anyone who dosent buy it shouldnt own a cue. Go bowling or something.
I only hit 3 balls all day, I been in a slump and been playing 1P, so I took today off pool, I saw the app in my phone and decided to see what I could come up with. i was cold, stiff hadnt hit a ball nothing and was over 22 all 3 times using my player. I was amazed I was over 15-my break blows.
I'm going to have fun playing with this.
great app.
Best
eric
Last edited:
This comment absolutely made my day! Thank you!
I've been up since 7am working on it and it is now 4am. I'm ready for some horizontal-ness.
Android update
The Android app is nearing "feature complete" stage. All the hard stuff is done. Aside from the cue-ball placement interface (the bit with the pool table), the rest is all tiny stuff (storing your preferences, demo version, etc.) We also need to do a lot of testing before we release it and we have full-time jobs, so this will probably stretch out for a bit. No set date yet, but my original estimates are looking pretty accurate.
If you've seen the iPhone version, then you know exactly how the Android version will work. We're modeling the Android after the interface we designed for the iPhone version, so they look and work as close to the same as possible. The exceptions are for standard Android interface items (example: the iPhone has tabs at the bottom, but the Android uses the pop-up menu.) Otherwise, they are hard to distinguish from one another.
Should be worth the wait!
A bit about the iPhone version
I spent most of the day today putting out an updated iPhone version that supports iOS 4. This is the first time I've had a chance to work on it since I submitted it on 5/10. I really didn't want to take time away from the Android development, but Break Speed doesn't currently work on iOS 4 (they broke some stuff that I was using.) Since the iPhone app has to go through a review process, I had to get it in for review ASAP in order to (hopefully) make the June 21st deadline for when iOS 4 hits the streets. While I was at it, I added a (free) demo version to the App Store.
New demo versions!
For the iPhone/Android demo versions, we can't do what most demo versions of apps do (prevent you from saving your data, etc.) We do one thing: provide you with a break speed. How do we provide a demo so you can evaluate the software and build confidence that it works on your device, yet still entice users to pay for it in order to fund development?
Craig and I argued and fought about what the right solution was. We talked about it on the phone, in the car, at the pool hall, in email, had an interesting bout over the tin-can-network and even snuck into a near-by high-school so we could pass notes. In the end, here's what we came up with... (can you tell the Ambien is kicking in?)
We elected to go with an "estimated speed". The speeds are presented such that they give you a 5MPH range, with your speed falling somewhere (randomly) within that range. We do not modify the actual detection, so you can edit the waveform and view the results to make sure that it accurately detected the break, but you just won't have the exact speed. For that, you have to fork out some cash and help us prove to our wives that this isn't some big waste of time.
Curious why that you thought this was so bad? All mine measured high 21s and I finally got one to climb into the 22s. Felt like I was hitting pretty good each time and netted solid results on the table.
You made it seem like getting 4 to a rail was a problem (ok ...not quite lol)..but if this your everyday speed, I 'd say exceptable....but why did you think it wasn't? I mean 22 is a pretty dam good crack imo.
Yes, our minimum target is 1.5.
First, thanks Paul for the Windows Mobile fix for my phone, worked great.
My high was a bit over 20mph, but was only 3 attempts.
You thinking about doing a data collection from all of the break apps with registered user names so we can go to your website and check progress, comparisons? I figure if you quit your day job you can do it. Thanks.
My high was a bit over 20mph, but was only 3 attempts.
You thinking about doing a data collection from all of the break apps with registered user names so we can go to your website and check progress, comparisons? I figure if you quit your day job you can do it. Thanks.
You thinking about doing a data collection from all of the break apps with registered user names so we can go to your website and check progress, comparisons? I figure if you quit your day job you can do it. Thanks.
Yep. Not sure when this is coming, but it's in the plans.
You could do what GPS did for a while -- provide randomly degraded accuracy to the non-paying user. That's sort of what you did with bands.
I agree with Bob,
Round off the number. So on the demo version it would show 18, whereas the upgrade version would be more precise 18.512 mph. Strip out the other features, such as changing the break spot, data saves, etc.
Limit the number of uses, say 3-5 attempts, anything beyond that put in a time delay (1 or more hours) until it resets. Another idea is put a mandatory iAd commercial starting at 1 minute. So, starting with 4th-6th attempt has 1 minute iAd. This keeps the incentive to upgrade, while gives the casual and curious exposure to your app.
Round off the number. So on the demo version it would show 18, whereas the upgrade version would be more precise 18.512 mph. Strip out the other features, such as changing the break spot, data saves, etc.
Limit the number of uses, say 3-5 attempts, anything beyond that put in a time delay (1 or more hours) until it resets. Another idea is put a mandatory iAd commercial starting at 1 minute. So, starting with 4th-6th attempt has 1 minute iAd. This keeps the incentive to upgrade, while gives the casual and curious exposure to your app.
Last edited:
Break Speed on Android - Beta
All,
The Android version has reached "release candidate" stage. What that means is that *we* think it's done.
But we need a few dedicated beta testers to prove to us that we're wrong (or that we're right!)
The version that will be used for Beta will be the Demo version, which provides estimated (ranged) break speeds.
We're not looking for any/every person we can find. In fact, we're looking for a pretty small group of people that meet specific criteria:
We want people that are technical enough to help diagnose problems, should they arise. We're looking for a good variety of devices and OS revisions (if we get too many of a specific device, we'll stop accepting beta testers for that device.) We're looking for people to act quickly... you are the last step before we place the application on the Android Market. We only expect the beta to last a couple days, so we'll want your reaction and feedback as quickly as possible. Ideally, you'll have the opportunity in the next day or two to make real-world use the software in a pool hall or wherever you normally play. Finally, we're looking for people willing to get involved. This means that if you become a beta tester, you have to "cross your heart and swear to die" that you'll give us some feedback (good, bad or indifferent.) If it works as advertised and you can't find any problems, just tell us.
If you want to get involved, stop by the website and hit the support link and email me (Paul Nettle) directly from the link on that page. With your email, tell me which device and OS revision you have and if it's rooted (people root devices, so we need to test this.) Also include a note about the next time you'll be at a pool hall.
NOTE: Google's Android Market doesn't allow for promotional codes or even the ability to 'gift' the application to anybody. So please understand that we won't be able to provide you with a promotional copy after the beta - you will still have to pay for it. :angry:
All,
The Android version has reached "release candidate" stage. What that means is that *we* think it's done.
But we need a few dedicated beta testers to prove to us that we're wrong (or that we're right!)
The version that will be used for Beta will be the Demo version, which provides estimated (ranged) break speeds.
We're not looking for any/every person we can find. In fact, we're looking for a pretty small group of people that meet specific criteria:
We want people that are technical enough to help diagnose problems, should they arise. We're looking for a good variety of devices and OS revisions (if we get too many of a specific device, we'll stop accepting beta testers for that device.) We're looking for people to act quickly... you are the last step before we place the application on the Android Market. We only expect the beta to last a couple days, so we'll want your reaction and feedback as quickly as possible. Ideally, you'll have the opportunity in the next day or two to make real-world use the software in a pool hall or wherever you normally play. Finally, we're looking for people willing to get involved. This means that if you become a beta tester, you have to "cross your heart and swear to die" that you'll give us some feedback (good, bad or indifferent.) If it works as advertised and you can't find any problems, just tell us.
If you want to get involved, stop by the website and hit the support link and email me (Paul Nettle) directly from the link on that page. With your email, tell me which device and OS revision you have and if it's rooted (people root devices, so we need to test this.) Also include a note about the next time you'll be at a pool hall.
NOTE: Google's Android Market doesn't allow for promotional codes or even the ability to 'gift' the application to anybody. So please understand that we won't be able to provide you with a promotional copy after the beta - you will still have to pay for it. :angry:
Crud I wish my contract with Alltel was up already cuz I would've been a pretty darn good candidate as I have it on my Windows Mobile HTC Touch Pro right now so I could run them both and compare results.
Bump from page two.
Paul, pm sent. I hope its received in the spirit it was written..just trying to help you do well.
Paul, pm sent. I hope its received in the spirit it was written..just trying to help you do well.