oh sorry marc, I confused the things a tiny bit, christoph kemper is the CEO :roll:

Mad: this seems to be a bad area for Cubase/TI, try moving your whole project forward 8 bars. Avoid bars 1 to 9 when ever working in Cubase/Virus. The same sort of external midi clock errors occur with virus b even.

There are quite alot of people reporting this issue on the Cubase forum, here?s one of the threads. Sadly there hasn?t been any response from Steinberg on the issue for a long while :(

Why not ? At least Nord2 and Nord3 have a rock solid timing !

I have the same problem with my Virus C. I just start from bar 5 or 9 and that solves it. Must be a problem in Cubase :?

has anyone bothered yet to compare the midi event stream output of say.. logic and cubase on the playback of the same midi stream?

shouldnt be too hard to make a full event dump with timestamps with something like midi-ox and a few loopback drivers

my guess is that the timecode is already buffered some sec in advance and not properly removed when you jump back to trackstart... (which COULD be partially caused by a mididriver that plays it too safe.. buffering further ahead = better chance of actually scheduling the even on time)

I have been suffering with Arp timing problems with cubase sx3 since I got the TI the month it was released. I have contacted Access support who suggested some tweaks to my PCI latency. Tried them all with no avail! Problem still remains. I'm sure its got nothing to do with PCI latency, in this case as there is no pops or clicks or anything other than the Arps out of time on mixdown. I have tried just about everything to fix this problem on my own setup.

MADSTATION 8)
Do me a favor and try this see if it helps any. Start you arrangement in cubase from bar 33. Let us know if it makes any difference?
8O