Thursday, June 28, 2012

Ebook issues

The ebook server is now active again but work is still being done. You are able to unlock documents etc. The software is having an update done but this is more on the admin side, so it shouldnt affect everyone. If it does, it just means you'd need to download the document again later. I'll let everyone know if this is the case.

Thursday, June 21, 2012

Online wheel analysis software

The issue with the software not dealing with 1500+ spins is still not yet resolved. The two obvious solutions are:

1. Dont use so many spins (you dont need that many anyway)
2. Upgrade server (but this means even more expense for me but only for the few players it affects)

The way things are looking, it appears the software will need to write cache files like virtual memory, then access them as it does analysis bit by bit. This is not a simple task and will take at least 2 weeks to program and test. Given that I stopped taking new players and I am not changing everyone monthly fees to access the server, and the issue only affects the minority of players, I'm looking for a more viable solution. For now though, simply don't use as many spins. Keep it to no more than 300-400 per direction. Either way the programmers are still looking for a simpler solution.

Once it is solved whatever way, we'll start work on the module that does the last major automation part which will make it as foolproof as possible, minus some few minor bits. After that, then we'll integrate the mobile phones (system phone).

Wednesday, June 20, 2012

Documents

Still you may not be able to access any of the ebooks/documents. The server change is in the final stage, and everyone will have document access again within 48 hours. When you will have access depends on your ISP as they propagate the changes.

Monday, June 18, 2012

Analysis Software

There was a hacking attempt on the analysis software. I expected it would happen eventually, which is why all scripts are run on a remote server. The server that was attacked only holds the front, and the actual location is well protected. Paranoia is sometimes justified, and does pay off. But the user database was flooded with garbage, which may take a while to fix. So there may be a service interruption for a day or so. As for the auto analysis issue with 1500+ spins, programemrs are still working on it. I just spoke to them and will get another update tomorrow.

Ebooks and Documents

Most documents will be locked automatically because we're in the middle of changing ebook servers. The whole process takes about a week and its almost done, but it also depends on your internet service provider. So for now and for the next few days, many people wont be able to open any documents. At least when the move is done, there wont be further issues.

Thursday, June 14, 2012

Hybrid V3 Video

The progress video of the hybrid v3 is now available at http://www.genuinewinner.com/forum/index.php/topic,401.0.html

The timings from repeating the same video are below:



This is from a standard PC and still the timings are almost all within 10ms accuracy. In fact they are mostly 5ms accuracy, but the few discrepancies are because of the computer's background processes. On the actual hybrid hardware, there wont be background processes that interfere. It is basically a PC stripped of background processes so that all it has to do is get accurate timings. Either way, the difference between 5-10 or even 15ms isnt significant

The progress video shows:
  • The test interface
  • Wheel location tracking (tracking where the wheel is when the camera is moving).
  • Ball timings: I repeat the spin a few times so you can see the accuracy of timings, when the camera is moving more it would in typical conditions

Again these are the hardest parts done.

Players for this computer will be asked to arrange a meeting with me for training in about a month or so - just whenever the completed package is fully tested and ready to be used. I'll keep you updated.

Ebook documents

The recent problem with the books was the server, not software. It is still having issues so I've started the server transfer process. Unfortunately it will take at least a few days, but at least it will solve the problem. I'll post an announcement when it is done.

Wednesday, June 13, 2012

Hybrid V3 video

Just a note I was out today so couldnt do the V3 hybrid video, but am doing it tomorrow. Again the hard parts are all done, and its working very well. Now its more just tidying up, last features, bug fixing etc.

Instructions/ebooks

The recent issues with the ebook software were due to the server being down, not a software error. The server is ok now, so everyone should be able to open the documents. If the server has issues again, I'll just change hosts.

Tuesday, June 12, 2012

Ebook access

There is a problem with the ebook authentication process. Keep in mind it is new so is bound to have a few bugs, especially after a recent update. We're aware of the problem and fixing it asap. I'll post an update once the issue is corrected. But for now, you may not be able to access any documents (only for day or so).

Hybrid V3

I received and tested the latest Hybrid V3 version, which has a few improvements. Most relevant is if a ball detection is missed, the software will look at data from a few seconds earlier, and can establish ball timings from this. This in turn allows it to more accurately predict when the ball passes a fixed point in real-time.

Put another way, if the ball decelerates with timings:

500ms
600ms
700ms
800ms
900ms
1000ms

Then the software knows when to expect the ball to come into view. This way it can better filter out video noise, so it knows what parts of the video are the ball. It is simple for us as humans to do this - to us we just know roughly when the ball will pass, so we know when to look for it. But coding a computer to do it is much more complex.

From tests I did, I did not miss one ball detection at all - not even one. The maximum deviation of timings I encountered was 15ms, which is still well below the 50ms we get from any manual clocking computer. But the average error was about 5ms. So it is about 5 times more accurate than manual clocking roulette computers.

The remaining work before it is available to players:
  • refine rotor detection algorithms (not difficult work - 3-5 days)
  • wireless transmission module (part that sends prediction to players - fairly straight-forward but tedious testing as we need to ensure it isnt easily detectable, about 1 week)
  • software security: to make it just about impossible to use the software without authorization (1 week)
  • General bug testing and refinements (1 week)

Rounded off to about a month before full completion/availability. It has all taken longer than expected, as any developmental project does. But we have the wheel tracking part done (finding wheel location and tracking it), and the ball timing algorithms are perfect now. So these are by far the two hardest parts done. I will do a video of these parts for players first chance I get.

I have prepared the hardware so once everything is ready, the package will be complete, and players can arrange to meet for training etc.

Thursday, June 7, 2012

Email responses and v3

While I've stopped sales, it has only meant I shifted focus to other things so I'm still as busy as before (for now). Things will start to calm down in a week or two.

So for now, I've lagged behind on emails and couldnt take calls today. I expect to be up to date on emails by end of tomorrow. Please dont resend emails - I'll respond soon. Realistically, in roughly a month I'll have a lot more spare time on my hands so I'll respond in less time to emails.

Also I have the latest hybrid v3 version for testing. Everything is going very well and I'll make the video for purchasers asap.

Friday, June 1, 2012

Hybrid V3

I expect to do the V3 video for purchasers some time next week (to show progress). Things are going very well, but we changed part of the program, which led to some bugs that set us back a week or so. But we're past them now.

Below are some comparative timings from the same spin on my home PC (4 different tests):



These are not at all optimized, and my PC runs many processes. The V3 hardware will be stripped of background processes so will be even more accurate, although even the above is overkill. Remember that manual clocking computers can only be accurate to within 50ms, and the above is much more accurate. The above timings are from a moving wheel in a video, to make things difficult. But I dont expect you'll dance in front of the wheel anyway.

With the ball at roughly 700ms, this is quite fast and still you can see the accuracy of ball timings is very high.

Registering ebooks

Just a note about registering the ebooks: we are doing some maintenance on the authentication server so you may get some errors when trying to register, so just try later if you have issues.

All documents

All documents registered before today have now been disabled, so you wont be able to open them. For the new documents, we are now using the software I had created for it, which is also the service I provide at www.freedocumentsecurity.com

I had this software created because no other software had the features needed. However, the only documents that have particularly sensitive information is the Uber documentation, but even if they leaked it wouldn't be a major issue. But still it is better to have the security than not.

To request access to any new documents, see http://www.genuinewinner.com/software.html



Computer shipping

Yesterday I received another 5 Uber phones from the technician, and today shipped a different 5 - you will have been notified if yours was sent. I have 7 Uber phones left for now and will ship them mid next week. If you dont receive your phone from the remaining 7, you definitely will about 2 weeks from now. But with the 7, I should have enough to ship everyone else who is waiting.