External sync drifting when syncing to tape

Cubase
21

That may be true, but I think Cubase does all its calculations such as bar lines, tempo maps, etc, by dividing up the wordclock. If an external master decides to run at a different speed, Cubase (the program) is not likely to glitch its playback to accomodate an out of sync MTC address, so that incoming address needs to also supply the wordclock that Cubase has been following.

Bottom line, even if you’re not syncing audio, only MIDI, Cubase is still at the same disadvantage.

In my experience, Cubase is much happier if it’s the MTC master, but that puts the tape machine in the same position of having to vary it’s motor speed to match the incoming MTC. -By using a synchronizer via “phase Locked Loop” (PLL) to convert MTC to vary the capstan rpm to chase the rolling addresses.

22

Yes, you pointed that out before….I’m just reiterating that in rock solid no-drift sync…”shoulds” will waste a lot of time once you hear with your own ears that a “should” isn’t working.

In a healthy chain (here), Cubendo slaves/chases/located flawlessly when switched to mtc on the panel….when..all else in the sync chain is operating as it is here.

Good luck!

23

Except when it doesn’t :slight_smile:

I believe Nuendo 3 was the first sb product I tried years ago and was thrilled when I experimented and found what worked for slaving it to my tape machines, linndrums, keyboard room midi kbds, FostexDa90s/TascamDa88s etc. Rock solid sync across all the dozens of midi devices being most important at that time.

24

Works fine either way here….although….

the op doesn’t appear to have motor synchronizers as options. He’d have to own an ats500/mts1000/AdamsSmith etc (which I own) to use with the msr16 acc2 ports as additional tach etc control….to slave the motors….which when I do….dtp is still the master brain talking to the motor synchronizers.

In his scenario, getting a tight Cubendo-as-slave will be his only option…which is do-able.

25

I have got a fostex 4030 which i got incredibly cheaply with a very ropey R8 always wondered if I could use that on the Tascam.

Incidentally , I learned one more thing tonight if i hit the stop button on Cubase as it’s fake syncing, there is only a momentary stop, fraction of a second, but the position/clock resets to the correct time as displayed on the JLC no doubt to drift out again over the next 10 minutes

26

4030 was interesting (as a lot of Fostex engineers came from Teac/tascam). Similar sorta/kinda to the tascam ats500 approach, but those 4030 guys really wanted to keep that thing in use with fostex-only reel to reels imo at the time. I don’t know how reliable their internal eproms are over time….for sure, failed eprom chips in Tascam mts1000s are why those things are always sold on ebay “not working”….just no way to repair something like that :slight_smile:

27

Ok well, there have been some developments.

I was hunting around for another DAW to try this on then suddenly realised I’d installed a well known one beginning with the letter R when I was trying to diagnose my firewire issues.

Guess what - it worked perfectly, tracked the JLCooper pps-100 Midi-SMPTE synchronizer on both LTC and Midi Timecode - did each for 10 minutes, no loss of frames the SMPTE readout on R***** matched and followed the readout on the JLC.

Also, yesterday I took my JLCooper box to the local studio I work in and tried it with their PC - It’s running Windows 10 and Cubase 10. Guess what - the drift was back.

So I’m still scratching my head here and it could still in theory be user error (cos let’s face it - I am the common denominator), but that’s now Cubase 6, Cubase 10 and Cubase 14 on Windows 7, windows 10 and Windows 11 and this is not synching properly, yet appears to do so with another DAW.

Question for DosWasbest; you said it worked fine for you both ways - what PC/Mac/windows/OS/linux config do you have. I’m deciding whether I can bother technical support with this now.

best

Mark

28

“Both ways” meaning I can sync Cubendo as slaves to the 24trk tape machines….or slave the tape machines to the Cubendos.

At the computers-only level, I run 2-4 synchronized pc machines at a time, Windows 11 Pro (although I can dial in a couple of old winXp and 7 machines when I have reason).

Cubase 3,4,5,6,7,8,9,10,11,12,13,14 (depending on what I’m doing between testing or recording)…Nuendo 14,13,12,11,10.

When I run the tape machines as chaselock slaves to the computers, their motor synchronizers handle the tach etc sync. I own a bunch of ats500s for that as they historically have been great little workhorses for mci, studer, 3m, tascam etc machines…even compared to adams smith stuff imo.

When running Cubendos as slaves, tape machine synchronizers are powered off as no need for them….just smpte from a mahine and dtp resolving everything.

I work exclusively at 48k across all my systems, always have, always will :slight_smile:

I usually run the Cubendos as the slaves in practice as the instantaneous lockups work better for me in the constant short-distance ff/rew I do with the tape machines during transfers.

I’m generally only synchronizing the entire shebang at the same time when I’m transferring zillions of old project slave reels (46trk stuff etc) to the daws for remixes. If I’m purely recording from scratch, I generally am on tape-only for overdubs as I do boatloads of tape bouncing and am generally baking in a predetermined kind of sound.

Every audio signal anywhere is routed through the consoles completely.

No Macs. I dumped all my Macs about 20 years ago.

One point of advantage in owning so many varying types of synchronizers is I can instantly dial in varying combinations (dtp, ats500s pps100 and more) to quickly get to a solution..which means I haven’t encountered sync challenges for decades.

In your case, you’ve sorta blown a couple of weeks just futzing with the jl cooper box only..and a lot of headscratching….which would drive me bonkers :slight_smile:

Likely more info than you asked for, but this may give a good indication that rock solid sync is a reality for many….and has been for eons.

29

Futzing is my middle name. I just didn’t want to spend more money when I wasn’t certain what the problem was. I’ll certainly look at the DTP, though like so much of the stuff - it’s legacy hardware and rare as hens teeth this week, all with it’s own potential problems. I’ve also spent a lot of money on Cubase itself, so really wanted to be sure that there wasn’t something wrong with it and/or something I was doing wrong. Not sure I have the answer yet, but I really appreciate all the time and effort that folks have put in to answering my query -so thankyou.

best

Mark

30

One most important point that DOSWasBest didn’t address in this most recent summation- I suspect that the DTP is feeding wordclock to Cubase. Some hardware or software has to be responsible for counting the “ticks” in the two clocks.

31

Yes, I’m assuming that the DTP is the master and the audio interface is clocked to it and thus Cubase is too.

32

Yes, exactly. Dtp does feed master wc to all devices that have it here.

The op mentioned at some point not seeing the need for wc in the equation…but it’s….imo….critical in any rock solid setup …esp to best avoid inexplicable sync glitching for any scenario.

As an aside, one very fun/interesting use of the presence of wc in the scenario (with dtp and a couple of other devices I have)…is that…with the Cubendos slaving in their mtc transport sync settings….I can alter the clock frequency leaving dtp with up/down potentiometers…..meaning….I can instantly “varispeed” the Cubendo transports in realtime just as I can with the tape machine motors.

This is also allowed to occur because the audio interfaces themselves are slaving to the dtp wc and instantly obey the realtime changes….pitch and speed instantly going up or down to whatever I want. Essentially changing the overall system sample rate on the fly to make this possible. No plugin….no complicated adjusts.

This works the same as the op varispeed +/- on his msr16….however for Cubendo instead of the tape recorder.