I've noticed this problem multiple times and even before updating to the recent 1.10.73 firmware, but I held off writing this up until I could find a (hopefully) reliable reproducible "trigger condition" so I wouldn't be writing a problem report that just says, "Every now and then this weird thing happens." The steps below don't trigger the unexpected behavior every time, but it I've got it.
Steps to reproduce:
- Upload a track to the table called "MoveOut", which contains only two lines, one moving the ball to Rho 0 and the next moving it to Rho 1, i.e.:
- 0 0
- 0 1
- From Webcenter, download dbullardWebcenter's track called "clockworkSwirl5WithClipping" and create a new playlist containing only this track.
- Enable "Pause between tracks".
- Play MoveOut as a single track and let it complete, with the ball stopped at Rho 1.
- Put the table to sleep.
- Wake the table back up.
- Go to the playlist you created that contains only "clockworkSwirl5WithClipping" and press Shuffle to start playing it.
- If the bug is triggered, the ball will move to Rho 0 and then begin what will turn out to be an endless loop of drawing a "teardrop" shape. The "point" of the teardrop will be at Rho 1, with the curved "base" passing through Rho 0. The ball will continue traveling along these two arcs indefinitely until you choose to play something else.
Interestingly, the new firmware that shows the ETA on the track (very cool feature, by the way!) shows a bit of additional insight. When this bug is triggered, the ETA on the "clockworkSwirl5WithClipping" shows approximately 30 seconds, which is the time to complete one of these teardrop cycles. By the time the teardrop cycle completes, the progress bar has moved to the end, and then the app briefly shows that the table is homing, and then the cycle repeats, with both the ETA and progress bar reset, even though the track is still the same.
On a side note, there's sometimes odd behavior with the MoveOut track as well. Typically it just homes to Rho 0, then moves to Rho 1 in a straight line, as expected. But sometimes it will do that, and THEN move BACK to Rho 0 along an ARC, before moving out AGAIN to Rho 1 in a straight line. Not sure if that's a related issue.
Hopefully this is reproducible by the devs, and if so I'd be keen to know the underlying cause just out of pure technical curiosity. 🙂
Thanks John! This is exactly the detailed sort of reporting that will help us zero in on the bug. I've run into the teardrop endless loop before, but it was too sporadic to pin down. Will relay to our dev team!
I've found the bug and fixed it, it will be corrected in the next update. Thanks for the clear and helpful directions.
"endless teardrop" seems like the perfect name for a Sisyphus bug.
Also, I'm humbled by the details of that bug report. Kudos John!
"endless teardrop" seems like the perfect name for a Sisyphus bug.
Also, I'm humbled by the details of that bug report. Kudos John!
Thanks! I work in IT, and one frustration that's been consistent across several roles I've served in that field has been problem reports like, "It's not working." Sometimes even the problematic behavior isn't described, never mind what causes it. While I understand that the average user won't always know the value of providing a proper problem report or even what should go into one, that experience has caused me to try to write the kind of problem reports for others that I wish I received myself, i.e. ones that narrow down the problem condition as precisely as possible and that include clear steps that reliably reproduce the problem. I've found that if you can tell a technician/developer exactly how to trigger the problem rather than giving them a vague report and expecting them to spend time experimenting in the hopes of maybe stumbling on the issue, then the problem is more likely to be investigated and fixed more quickly -- as seems to have occurred here! 🙂
John, just so you know the depth of respect I have for your level of detail: I'm a software developer myself. So I get where you're coming from and why you do it. It must have taken a fair amount of work to figure out that repro. But it made a difference, and now it's fixed!
Feels a little weird about it being my track that showed this, never saw this behavior on my table.
Glad I could help ☺
We take your privacy seriously and will only use your personal information to provide you with the products and services you request from us. Please see our privacy policy for more details.