No announcement yet.

Short fade-in prevents gapless looping of recordings

  • Filter
  • Time
  • Show
Clear All
new posts

  • Short fade-in prevents gapless looping of recordings

    I noticed a short fade in on my blackbox in recorded samples.
    Please see the picture attached. It shows a short increase in volume for the first 128 samples.

    This is an undesired effect, which prevents seemless looping. That is because on every new start of the sample the decrease in volume aka the fade-in is audible.

    For this demonstration constant white noise was fed to the bb and recorded at different volume levels. This effect is not dependent on input volume. For the picture, high volume was used to have a good visual representation of the effect.

    Attached Files

  • #2
    On my blackbox, the recorded audio is not in time with the original source.
    This leads to unwanted audio effects that you can hear in the attached file.
    Up to 4 seconds there is the original sound, which is recorded directly through the black box and then played back immediately.
    From second 4, you hear the out-of-sync playback of the original and the recording.

    This behavior is unwanted in two ways. First, it causes undesirable audio effects when original sources and recordings are used in parallel, as is the case with live performances. Second, it degrades timing, which is especially noticeable when recording multiple parts that all have a short delay that adds up as the number of recordings increases.


    • #3
      Your issue looks similar to mine:


      • #4
        Hi Lobosque,
        thanks for pointing this out!
        After watching your video I tend to agree!

        Nice groove btw, that is the original signal (not so much the blackbox recording)


        • #5
          The last days I tested with the latest stable fm of 1.7. The same effects occur (unvoluntary fade-in and fade-out on recorded clips).
          Here are some pictures showing the waveform in an audio editor.

          picture 1: close up of fade-in. For approx. 5ms and a delta of ~11dB.
          picture 2: fade-in at the beginning, and fade out at the end of the previous clip
          picture 3: overview fade-in / out, happens with every repetition cycle of the loop

          I have also uploaded a video documenting my process at wetransfer:

          Aaron could you please take a look at the material and explain what is going on from your perspective? Is my unit faulty or have I misunderstood something?
          Last edited by upwardspiral; 11-25-2021, 03:47 AM.


          • #6
            When I trigger a one shot kick drum one bar loop when using the seq, then resample that, the new sample will have its start (the transient, the 1st ms or so) cut off, so doesn’t loop correctly. (Using Rec Quant set to 1 bar).... operating system is v2

            Is this normal?


            • #7
              Hi nosebleed,

              this sounds familiar to me, so I replicated your steps on my machine. And let me tell you I encountered the same. It seems to go even deeper than that. I have documented my findings in this thread:


              There is (on top of a short delay) a fade-in for recorded audio on my unit as well as a fade-out. Feel free to tell me if that is the same that you experience. So far I have no solution for it.


              • #8

                yes, this seems like the same issue, looks like a 'feature' to stop clicks in loops but has this side effect to it, would be a good idea to be able to turn off this 'feature' sometimes when not needed and a smooth unfaded loop is needed. Hopefully they look into this issue, I have messaged them regarding it but no response.


                • #9

                  I just tested with the newest firmware (2.0.5) and the issue has not been fixed and remains. A bummer.


                  • #10
                    With the current firmware (2.1.0), the faulty loops have still not been fixed.


                    • #11
                      With the current firmware (2.1.5), the faulty loops and recordning behaviour have still not been fixed.


                      • #12
                        under the assumption that the above is a feature (aka intended behaviour) and not a bug, maybe you should file a feature request in the wishlist?


                        • #13
                          Originally posted by Georges View Post
                          under the assumption that the above is a feature (aka intended behaviour) and not a bug, maybe you should file a feature request in the wishlist?
                          Imprinting a small fadein directly into the .wav is a bug, no-one could ever want this _forced_ change of the original signal

                          The whole problem is well known since the dawn of digital samplers and this is how to deal with it:
                          Either the ENV Attack-Release is used in combo to shadow clicks at the of non-zero-start samples (or a dedicated crossfade algo), or a manual edit of the sample
                          But no pro-grade sampler would enforce that into the file, thus: it's a bug

                          I assume 1010 blackbox would run out of cpu-cycles when offering a loop-crossfade, but the ENV Attack-Release-Combo should be enough anyways