Emu XL-7 & MP-7 User's Group

In response to numerous comments about how tight the sequencer timing
is on the XL/MP, here are some concrete timings. Executive summary:
the timing accuracy varies from superb to unusable, depending on
circumstances.

The first set of drift timings are using a track pattern of 16th
notes, running at 125bpm. There is *no* MIDI output - all tracks
are set to "int". (oh: OS version 1.32.)

* Empty pattern, apart from the test track: drift of around 2msec.

* Factory pattern (Arctic Drift, in fact), tracks 1-15 muted, track
  16 as test track: drift of around 3msec.

* Factory pattern, tracks 1-15 running with volume=0, track 16 as
  test track: drift routinely at 10msec, worst case around 24msec.

* Factory pattern, test track on track 1, other tracks running with
  volume=0: drift around 2msec.

SUMMARY: best-case scenario has sequencer timing within 2msec, which
is pretty hot. Data in other tracks doesn't influence performance if
the tracks are muted. (I have a feeling this wasn't true of OS 1.31,
but could be wrong.) With lots of tracks running, the timing of
high-numbered tracks is pretty dreadful, but low-numbered tracks are
prioritised and tight. So: keep the timing-critical tracks near the
top (top 6 if possible; from T8/T9 onwards the slop is audible).

A second set of timings tests the arpeggiator: this time, the sequence
is four quarter-notes (300 ticks long), but the preset repeats at 1/16
notes.

* Test pattern only: drift < 2msec.

* Test pattern as track 16, others muted: < 2msec.

* Test pattern as track 16, others running volume=0: routinely 8msec,
  often around 15msec.

* Test pattern as track 1, others running volume=0: routinely 8msec,
  often around 15msec.

SUMMARY: in ideal circumstances (sparse data), the arpeggiator has
slightly better timing than the sequencer, but it degrades equally
badly when other tracks are active, and -- curiously -- lower-numbered
tracks are apparently *NOT* prioritised, so there's no clear way of
improving critical arpeggiator timing in a busy pattern.

I've not done any timing tests of the internal clock-based modulation
architecture; I guess that would be next. I can live with the
sequencer timing since it's clear how to keep it tight; the
degradation of the arpeggiator timing is worrying. (I've also had more
serious problems with arpeggiator drift which I plan to track down.)

Interestingly, turning MIDI on for the tracks ("ext"/"both") doesn't
seem to make much difference. The XL-7 seems to have
bandwidth/performance issues internally, regardless of the external
MIDI traffic.

A note on the timing procedure: I just let the XL-7 freewheel,
recording the test patterns into Digital Performer (also freewheeling,
using a Korg OasysPCI as clock source). Each measurement is done with
a single 30-second recording of the ticks displayed on two tracks, one
of them offset by visual inspection by a whole number of
16th-notes. The drift can be seen visually by comparing events on the
two tracks, and the sample offsets can be read off by inspection (and
then divided by 44.1).

Any comments?

-- 

  nick rothwell -- composition, systems, performance -- http://www.cassiel.com