shep@datacube.UUCP (10/11/85)
The problem of using "tape sync" out of the RX-11, or for that matter any Yamaha/Roland/Etc. product using RX-11-like tape-sync, is that the analogue signal used is quite susceptible to impulse noise. It will work fine as long as no "extra" pulses get in there. The input circuitry for these "sync" pulses is just a high-gain comparator that is divided down as required. If noise gets in, the count is perturbed. When you use your Fostex in the way you describe, crosstalk leakage, from both the electronics and the tape, cause impulse noise on the sync track and perturb the count. This is -not- a fault with the Fostex deck, as a Tascam Portastudio also exhibits the problem. In fact, if you use this sync format on a 2", 24-track, and lay down a hot snare drum or power chord on a adjacent channel, it too will cock up. There are at least two solutions: - Build an adaptive phase locked loop for this sync signal. I did this in the pre-MIDI days of the late 1970's. The PLL locks onto the recorded rate and rejects pulses outside the window. This worked fine for many albums using the early Roland drum machines. The sync format hasn't changed! (Listen to Brian Briggs "Brian Damage", Bearsville Records; all percussion on that album was done this way). - Use "real" time-code. Like SMPTE. SMPTE code really does work. Plus you can start anywhere and lock-up. Some companies make widgets that read/write SMPTE time-code and convert it to MIDI. (Roland Sync Box SBX80 ?) Inherent in the implementation are phaselock circuits that improve noise immunity. Shep Siegel UUCP: ihnp4!datacube!shep Datacube Inc.; 4 Dearborn Rd.; Peabody, Ma. 01960; 617-535-6644