Your issue has been reported.

As I answered in the initial reply, I export MIDI data recorded in one Sequencer as a SMF (.mid). This *guarantees* the documented data is transferred properly. I never would attempt to sync and transfer multiple Tracks in a realtime transfer.

That said, I understand that you want to know why there is a change in the position of your data when you manually transfer data to a computer using the normal synchronization methods; that I cannot answer, because I never would try to transfer data in that manner;

However, if that is how you desire to do it, it should not be too much trouble to ask that you supply the info on your setup; it's impossible to know or even guess about the actual cause.

MIDI is serial... means one event after another. In a Normal MIDI record situation your are recording but one channel at a time to one track at a time.
When you 'transfer' an entire sequence in real time, you are creating an unusual situation, where all data on all channels is simultaneously attempting to travel the same system. This is why the question about your methodology are important (at least on this side of this question).

In normal procedures (best practices) you would avoid this issue entirely by documenting the data in a file.

Is the issue a time stamping issue or is it a loss of synchronization? That will have to be determined. I can't answer that. And critical to that determination will be as much detail as you can provide.

You are so focused on your complaint, I guess, that you don't see any reason to provide your basic information to get clarification... I can't help you there. But I have enough information now to send a query, thanks. Hopefully now you are willing to wait...

Meantime: you will just need to Save The SMF and avoid the issue all together. (It's quicker, it's accurate)

When I receive a reply, to your query, I will pass it along.

Many thanks, Bad Mister! We very much hope that Yamaha will get a trouble tiicket for this incident and it will be resolved. Or he will say that the solution is impossible. Sorry if we were not polite. It's because of our bad English. Sorry again.

It turns out that the problem of the synchronization of MIDI events in Yamaha is quite old. Here is an article from the 2005 SOundOnSound magazine:

"One disturbing thing came to light whilst constructing a test multitimbral piece in Sonar — the Motif Rack exhibited some conspicuous timing problems when playing a simple six-part tune. Initially, I thought it was due to the Motif being sluggish to respond over its MIDI ports, but the problem persisted, even when I banished MIDI in favour of the USB port. Loading the same song into SQ01 produced the same timing errors. In order to confirm there was no problem within the sequencers, I copied the drum part to an adjacent track, and routed it to a completely different synth module. This played the drum part back perfectly in time — while the duplicate Motif Rack drum part stumbled conspicuously, and was noticeably late in triggering. In order to get some idea of exactly how much, I recorded an audio snippet of the twin drum parts and inspected the waveform at a high zoom factor. The results showed an average discrepancy of around 1800 samples — ie. 40 milliseconds at 44.1kHz sample rate (the sequence was at 86bpm). Mindful of the fact that sequencers prioritise the scanning of tracks in numerical order (the Motif drums were on track 10, according to their default MIDI channel) I moved both drum tracks to the top of the track list. The problem was slightly ameliorated, but there was still a distinct flamming between the part from the Motif and the one from the other module, more than would otherwise be expected from 'normal' MIDI-timing discrepancies. This time, the delay measured on average 400 samples, which translates into around nine milliseconds — better, but still perceptable. This is a potentially serious problem for a multitimbral synth."

http://www.soundonsound.com/sos/jun03/articles/yamahamotifrack.as p">Link