5.03b4 track delay anomaly

[Synthbuilder]

Live master with Reason as rewire slave.

Set up two identical Redrums in Reason. Both playing the same pattern. One is driven by a Live midi channel, the other with Reason's own sequencer. The two should be in perfect sync and they mostly are with PDC on or off.

However, if you pre-delay the midi channel controlling Redrum in Live with the track delay setting nothing happens. One would expect the pre-delayed channel to be pre-delayed, but both Redrums continue to play in sync.

But for positive track delay settings the midi channel is suitably delayed the two redrums now play out of sync. So this bit is working.

However, now pre-delay any other midi channel that controls ANY other midi output [including the very same redrum that the other channel is controlling]. This affects the midi timing and the two redrums now fall out of sync. That is, any midi channel can control the Live driven redrum except for the one actually controlling redrum with its midiclip.

This same anomaly can also be recreated by predelaying any audio channel and master channel.

I should add that rewired inputs are still not in sync with any audio clips played within Live. The rewired inputs are always delayed by the latency amount. Therefore this bug does not allow for any manual compensation for this delay.

Tony

Win XP SP2, P4 3.0GHz, 1GB RAM

[Synthbuilder]

We still have this problem in 5.03b5.

To summarise any single negative delay in any track (audio or other midi) will affect all rewire midi outputs.