Purpose: I want to remodel a part of "Beatrich"(Farmware).

Yuji Tanaka188 views3 posts
  • Yuji Tanaka

    スクリーンショット 2019-09-06 14.02.10.jpg

    Hi
    I am a big fan of Purmut8.
    I have begun to study a script of Purmut8.

    Purpose: I want to remodel a part of "Beatrich"(Firmware).

    In the situation using with "SYNC" mode.
    I want to up minimum of Clock Frequency on a little (above 30kHz)
    The reason wants to improve sound quality a little more.
    Do you know the method?

  • Magnus Lidström

    Hi Yuji!

    Glad you like Permut8! I do too. :)

    Unfortunately this is not easy. Permut8 has a delay line memory of 64k 12-bit samples * 2 for stereo and when you run in synced mode its sample rate is adjusted so that these 64k samples align with your song position according to the time scale you choose. This also means that the Clock Frequency also depends on the song tempo.

    I can think of two (rather complicated) workarounds to allow for higher sample-rate If I were to create an alternative version of the Beatrick firmware:

    1) Firmwares have an additional 64k 32-bit words memory (apart from the delay-line memory) that could be used to store every other sample, allowing twice the rate. The tricky part is that some of this 64k memory is already used for other firmware data. (But one could pack two samples in each 32-bit word.)

    or perhaps easier:

    2) If mono sound is enough, one channel in the delay line memory could be used to store every other sample instead of storing stereo data.

    Regardless of which solution, increasing the sample-rate like this wouldn't be transparent to the end-user. You would need to put Permut8 in twice the synced rate, e.g. for 4 bars you would put it in 2/1. The unwanted side-effect to this would be that Permut8 wouldn't automatically align to the correct beat in the song. It would play in sync, but could be 2 beats off if you jump to a new position in your song.

  • Yuji Tanaka

    Hi Magnus

    Thank you for useful advice.
    "64k 12-bit samples * 2 for stereo" is rules in Permut8.
    It does not touch it in Firmware.

    Firmwares have an additional 64k 32-bit words memory
    (apart from the delay-line memory)
    that could be used to store every other sample, allowing twice the rate.
    The tricky part is that some of this 64k memory is already used for other firmware data.

    This is very important indication.
    I was interested there.
    What my understanding does not extend to is sad.

    If mono sound is enough, one channel in the delay line memory could be used to store every other sample instead of storing stereo data.

    I want to use "Beatrich" in stereo Hi-Fi if possible.
    The combination of "Beatrich" and Permut8 is superior to every Stutter Effect Plugin.

    (But one could pack two samples in each 32-bit word.)

    Mr.Magnus.
    I want to certainly ask you.
    Do you have the plan making Hi Fi version of "Beatrich"?

You need to be to post a reply

Sign In / Sign Up


First time here? Just enter your current email and sign up.
×
Facebook sign in no longer available. Use the same email to set password and access your account. If you need help, contact us.