Loop size limitiations?


    May 02 2006 | 5:01 pm
    Will radial's loop channels handle ~6 minute AIFF tracks? Is the audio engine robust enough for this or is it better suited for short loops? For example, if I am riding the pitch control to match up a loop in real time, will the audio engine be able to keep up with the larger file size, or does it try to recalculate the pitch of the entire loop buffer each time it changes? I don't have a working copy to try it out with, but does anyone use Radial for DJing? It looks ideal for the job if it has fast response times and fine grained control.

    • May 02 2006 | 5:43 pm
      On May 2, 2006, at 12:01 PM, synthique wrote:
      > > Will radial's loop channels handle ~6 minute AIFF tracks?
      The radiaL application loads loop material into internal memory. So, the amount you have in your machine will obviously be a limiting factor.
      > Is the audio engine robust enough for this or is it better suited > for short loops? For example, if I am riding the pitch control to > match up a loop in real time, will the audio engine be able to keep > up with the larger file size, or does it try to recalculate the > pitch of the entire loop buffer each time it changes?
      The changes you make via UI or external controllers are changing the rate at which things are played back from a buffer (or buffers). Another potential problem is that when working with longer loops, one uses the same floating-point "range" to map to the total length of a buffer. The shorter the buffer, the better the granularity.
      > I don't have a working copy to try it out with, but does anyone > use Radial for DJing? It looks ideal for the job if it has fast > response times and fine grained control.
      It works fine for me, but I'm probably not doing something you'd consider DJing [working with live improvisors and buckets of shorter loop material].
      David Harrington (I think that's his name, anyway) has a new disk out on the pfMENTUM label that uses radiaL as one tool among many in quite an interesting way. I just got a radio station promo in the mail this week, so it may not be out yet. Keep an eye on http://www.pfMENTUM.com. You ought to be watching out for their stuff, anyway - it's a very interesting and diverse label.
    • May 02 2006 | 6:13 pm
      Thanks Gregory. I'll get it running and see how it behaves. The granularity seems like a problem. It sounds like the range is calculated as a percentage of the loop duration, which would make getting used to the pitch control difficult. Is the source code Radial available? I have not read the license, not sure if these are open source or not. It should be pretty simple to create an arbitrary pitch bend based on a constant instead of having the range determined by the buffer size. Thanks
    • May 02 2006 | 6:32 pm
      On May 2, 2006, at 1:13 PM, synthique wrote:
      > > Thanks Gregory. I'll get it running and see how it behaves.
      Good. I always think that personal experience is best when it comes to determining a good fit between tools and work.
      > The granularity seems like a problem. It sounds like the range is > calculated as a percentage of the loop duration,
      I was thinking more about placement within the buffer. Consider that you've got N possible values between 0. and 1.0, and those are used to represent 100 ms and 10 minutes. Each unit of X is mapped to quite a different range of values with respect to X+ N (the next increment).
      > which would make getting used to the pitch control difficult.
      If you're thinking about pitch control in terms of something like a turntable, I don't think that you're going to find a precise analog to what you're describing.
      > Is the source code Radial available?
      No, it isn't.