search

mxmxmx / temps_utile-

teensy 3.2 trigger generator
Other
146 stars 52 forks source link

Inconsistent offset in mult/divide #47

Closed joodi3k closed 5 years ago

joodi3k commented 5 years ago

A channel's behavior is inconsistent when adding offset to divide. I've tested it with no global divide and under 48ppqn. Please see the attached videos as the behavior is so erratic that it defies easy description.

The first video is with no global divide, clock is being provided by a metropolis. The second is the same but I'm adjusting offset on channel one instead of two. The third is with 48ppqn being provided by a beat step pro.

https://www.dropbox.com/s/hysvjxpbmwnbui0/IMG_2467.MOV?dl=0

https://www.dropbox.com/s/menuacn9d1vdb5w/IMG_2478.MOV?dl=0

https://www.dropbox.com/s/jzkq2lrqfmxez9w/IMG_2479.MOV?dl=0

In all three cases the metropolis is providing reset every 4 bars.

mxmxmx commented 5 years ago

so in the absence of easy description, i'm assuming you're saying that:

  1. this issue is about the latest firmware version, 1.3beta: https://github.com/mxmxmx/temps_utile-/tree/1.3b

  2. i'll be able to reproduce a/the problem simply by doing the following:

    • set divisor to some value > 1
    • set phase parameter to some value > 0
joodi3k commented 5 years ago

Actually I’m on v1.2.3. I’ll update and see. My understanding of 1.3b was that it does address divider issues, but not the one I’m experiencing.

To your second item, maybe. Some offset values appear to work. Others change the divisor. Some seem to be the incorrect offset. And still others cause the channel to freeze. None of these inconsistencies are terribly “consistent” in that they don’t seem to always come into play at particular offsets. Except at 50%. That seems to always lock up. Even resyncing the TU doesn’t help.

Reset has nothing to do with it in that the reset does not appear to fix any of the errant behavior. Which is to say if reset caused everything to work as intended again, I’d be fine.

While we’re on the subject, I see a couple of other wrong-seeming behaviors in my 1u TU. In LFSR mode, probability doesn’t change the chance of bits flipping unless I have taps established. And even then it’s kind of hard to tell. Haven’t delved into this further because I keep running into the divider issue.

Also, the save and load behavior does not reflect what’s in the manual. I appear to only have one save slot that loads on power up. Which is really all I need, but now that we’re in the weeds it does seem worth noting.

Thanks

Sent from my iPhone

On Mar 6, 2019, at 11:53 PM, mxmxmx notifications@github.com<mailto:notifications@github.com> wrote:

so in the absence of easy description, i'm assuming you're saying that:

  1. this issue is about the latest firmware version, 1.3beta: https://github.com/mxmxmx/temps_utile-/tree/1.3b

  2. i'll be able to reproduce a/the problem simply by doing the following:

    • set divisor to some value > 1
    • set phase parameter to some value > 0

  3. the reset stuff has nothing to do with it?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHubhttps://github.com/mxmxmx/temps_utile-/issues/47#issuecomment-470395411, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AuEIznCUwKoGYUHwiqSyF5DsmaAlEKInks5vUKlXgaJpZM4biJlj.

joodi3k commented 5 years ago

Duh. Latest firmware seems to have cleared up the clock issue, at least with basic testing of the 50% offset. Load/save menu looks correct now. Not sure which version Shay at Plumb loaded mine with when he built it. Or if it just was borked when it got flashed. Thanks for the help.

On Mar 6, 2019, at 11:53 PM, mxmxmx notifications@github.com<mailto:notifications@github.com> wrote:

so in the absence of easy description, i'm assuming you're saying that:

  1. this issue is about the latest firmware version, 1.3beta: https://github.com/mxmxmx/temps_utile-/tree/1.3b

  2. i'll be able to reproduce a/the problem simply by doing the following:

    • set divisor to some value > 1
    • set phase parameter to some value > 0

  3. the reset stuff has nothing to do with it?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHubhttps://github.com/mxmxmx/temps_utile-/issues/47#issuecomment-470395411, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AuEIznCUwKoGYUHwiqSyF5DsmaAlEKInks5vUKlXgaJpZM4biJlj.

Jay Reynolds Tour Manager Asleep At The Wheel 513-348-7712 jreynolds.bluesky@gmail.commailto:jreynolds.bluesky@gmail.com

mxmxmx commented 5 years ago

ok, let me know if/when i can close this issue.

re manual: that was updated to reflect v1.3, so you wouldn't have seen the new load/save menu (with 4 slots, not just one)

joodi3k commented 5 years ago

You can close it. Thanks again.

Sent from my iPhone

On Mar 9, 2019, at 7:28 AM, mxmxmx notifications@github.com<mailto:notifications@github.com> wrote:

ok, let me know if/when i can close this issue.

re manual: that was updated to reflect v1.3, so you wouldn't have seen the new load/save menu (with 4 slots, not just one)

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHubhttps://github.com/mxmxmx/temps_utile-/issues/47#issuecomment-471176958, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AuEIzmZ2knIC_oKjp0o4TsChyddRKe18ks5vU7bfgaJpZM4biJlj.

mxmxmx commented 5 years ago

perfect, thanks!