Problem Syncing Slow TSR-8 to DAW

S

sherm_jonathan

New member
Some of you may recall me from some threads ago asking how to sync my DAW to a TSR-8. Well, I did purchase a PPS-100 and sync the two. Problem is, I recorded 4 pieces of music in a session, each about 8 minutes long, on the synced TSR-8 and DAW. However, upon playback, the TSR-8 remained slightly slower, though measuredly so, so that at the end of each piece, the TSR-8 is about a second or so behind the DAW. Does this send me back to the drawing board, or is there a way to remedy this? Muchos Thanks
 
sherm_jonathan said:
Some of you may recall me from some threads ago asking how to sync my DAW to a TSR-8. Well, I did purchase a PPS-100 and sync the two. Problem is, I recorded 4 pieces of music in a session, each about 8 minutes long, on the synced TSR-8 and DAW. However, upon playback, the TSR-8 remained slightly slower, though measuredly so, so that at the end of each piece, the TSR-8 is about a second or so behind the DAW. Does this send me back to the drawing board, or is there a way to remedy this? Muchos Thanks
Click to expand...

Yeah, I remember that, but I don't recall what your DAW is. That will make a difference in how to proceed.
 
I have had this happen once. Something went wrong inside the sequencer, basically. I ditched it for another one and it never happened again. Not much help for you, though :(
 
Beck, your memory is good. The DAW is Cubase LE 4 on Macbook Pro.

JP, That's a bummer. But the PPS was acting a little funny, just on the display it would start scrambling information. I'll give it another chance, but will eye it suspiciously from now on.
 
  • So LE will for sure sync to external MTC? I'm using Studio 4 and it does...just checking if the lighter version does as well.
  • What timecode reference did you use? 25? 30? 30 non-drop?
  • What are you using for the MIDI interface for the MacBook/DAW?
  • Cables good both audio and MIDI? If possible, try swapping them out one at a time to rule out connectivity as an issue. I ask this because of your comment about the display.
  • What level did you stripe the SMPTE code at on track 8? Should be about -10.
  • dbx was OFF on track 8 when you striped the SMPTE code right?
 
Thanks Sweetbeats and Beck, I've got some leads now to try and fix this and report back. My interface is a Presonus Firepod. Suspicious?
 
The Firepod has MIDI i/o? I have a Digimax FS...no MIDI on that one and I can't recall what the Firepod's got...nothing wrong with it tho'...there's a ton of them out there...affordable good quality...I really like my Digimax...
 
Hi There,

I seem to remember in the original thread that I said:


Be careful, mtc is not accurate enough to lock machines, you will find it will drift over the length of the song, mtc is fine for running midi devices but not for locking machines.

You need smpte time code on the TSR, then you need a device that reads the smpte and generates word clock. Then you can lock the computer (cubase) to the word clock as there will be an option somewhere for external clock.

Always use the analog machine as master

I think you have time drift due to midi clock not being accurate enough.

Cheers

Alan.
 
Alan, I do recall that. If I have to start over, I'll do it that way. But first, perhaps there's other reasons. To answer Mr. Sweetbeats, I am using the 30 timecode reference, but I didn't know I had a choice. Is one more appropriate than the other? The firepod does have a midi connector, but I am using junky cables because I figured quality didn't matter. Also, I didn't go -10 but +10 on the time stripe (I missed the - sign). The dbx was off. Haha, how'd I do?
 
Just to start over, give me the exact set up you have, what is master, what is striped, which interfaces are plugged into what, sound card model. etc etc

Sorry to start over but it's kind of spread across 2 threads, I will then look up info on everything you are using and see if I can help out.

When you say the TSR is slower, is the TSR the slave? Are you talking about timing drift, the song slowly goes out of sync? If you start the song near the end, is it now in time?

Many years ago at the start of the computer recording thing I did sync checks between my analog machine (MSR16S) and my computer by recording a kick drum from a drum machine on both the analog and computer and seeing how long it took to go out of sync. I had a midi sync box smpte in MTC out, can't remember the name it's in a box somewhere. The set up was not up the scratch. But then I got a Tascam MX2424 which has smpte in and inbuilt work clock, perfect sync.

Cheers

Alan.

P.S I have a firebox as well but I don't sync using it, only use it for location recording and stereo mixdown from analog. They are a good piece of gear.
P.SS Frame rate in smpte does not really matter as long as the sending and receiving gear is set to the same rate, I use 24 Frames/sec Australian Standard for film.

Even more P.S I did some homework on the firebox, it will accept external clock soure via the spdif input, so if the problem can't be solved without a smpte to wordclock generator at least you can plug one into the system.

Alan.
 
Last edited:
Wait...he striped the deck with MTC?? Yah...that won't work. But no...he said he used 30fps frame rate, so it must have been SMPTE...:confused:

Frame rate in smpte does not really matter as long as the sending and receiving gear is set to the same rate, I use 24 Frames/sec Australian Standard for film.
Click to expand...

When I asked the question about which frame rate was used I wis thinking of possible drift if the stripe was 30fps and the sync box was set to sync to 30fps non-drop which I think would cause lock issues and drift. Witzendoz said it...frame rate doesn't matter as long as you use the same rate for your striped track and have the sync box set to read that same rate. I think it is often automatically sensed by the sync box anyway.

Rule of thumb with which I'm familar regarding the level to strip at is -7 ~ -10dBV...its mainly a bleed issue. You'll definitely pick up bleed from the stripe on track 8 at +10dBV over on track 7...could maybe cause issues for the sync box too? SMPTE timecode at +10dBV is pretty powerful for something that has been designed to read it at maybe -10dBV ~ 0dBV.

Also, if your audio cable running from track 8 to the sync box, and the MIDI cable going from the sync box to the Firepod are trashed, I'd swap them out...who knows if you are having intermittent connectivity or poor signal quality and that's going to wreak havoc with the sync box.

Can't recall if the PPS-100 has a LOCK indicator on it. I'm assuming it does...is that light on steady or is it lighting intermittently at all?
 
Alan, thanks for the homework, that would be a good option. The TSR is the master and I striped track 8 with SMPTE. I run this back through my tascam 320b and into the firepod to Cubase, though I guess I could skip the mixer and send it straight to the firepod. The song slowly goes out of sync from the beginning, so even if I stop and skip to the middle, it is out of sync by that point. However, the beginning at least starts in sync before slowly drifting. I'll have to check on the other stuff, but that is the interesting part, the drift appears to be in the recording and not the playback?
 
Where is the PPS-100 in the signal chain?

Ideally your timecode signal chain should be:

atr track 8 output --> PPS-100 SMPTE timecode input --> PPS-100 MIDI out --> Firepod MIDI in --> Cubase set to sync to incoming MTC

If you are trying to run the SMPTE audio output from track 8 to an audio input on the Firepod and have Cubase lock up to that it won't work. Cubase LE will not sync to incoming SMPTE timecode and that's what the JL Cooper box does for you...it converts the SMPTE code to MIDI timecode which Cubase LE will sync to. Does that make sense?

If I'm understanding your present setup right that's why it is drifting, because you aren't locked.

Wordclock doesn't have anything to do with this...the SMPTE code is essentially your clock reference since the TSR-8 is the master, but very few DAW applications have the feature that allows it to sync direct to SMPTE code.

Be careful, mtc is not accurate enough to lock machines, you will find it will drift over the length of the song, mtc is fine for running midi devices but not for locking machines
Click to expand...

Actually its too accurate...the frame rate and pulse type cannot be accurately captured by the analog tape medium, but is great for locking an analog deck to a MIDI device as long as you have the proverbial "box" (the sync box) that goes in between...an interpreter of sorts. The deck speaks SMPTE, the DAW listens to MTC. The PPS-100 is your interpreter so that when track 8 speaks SMPTE the DAW hears MTC.
 
Last edited:
sherm_jonathan said:
The TSR is the master and I striped track 8 with SMPTE. I run this back through my tascam 320b and into the firepod to Cubase, though I guess I could skip the mixer and send it straight to the firepod.
Click to expand...
What sweetbeats says about the PPS-100 apparently not being in the signal chain should be your first port of call, but as a rule it's a bad idea to run the SMPTE through the mixer. If you've touched the EQ or got the effects loop in you'll be in for a fair bit of head-scratching.

Is the PPS-100 in the chain? If not, are you sure that your version of Cubase can decode SMPTE in realtime? I've heard Reaper can, though I haven't tried it.

In most setups, when MTC is correctly working the computer will automatically start playback when it gets the signal from the tape.
Bear in mind that there are two ways of synchronizing, though. Usually in the config options you get to choose between:

1. Trigger and freewheel
2. Full chase lock

The former starts the machines in sync and then allows them to drift apart, which is not what you want.
The latter should keep them locked together, with the computer compensating for variations in the tape playback speed.
 
Seems to me you’re not getting sync after the PPS-1 or your DAW receives the start command. You may have a situation where the PPS-100 is picking up the START signal and then losing sync with the tape machine.

This can happen with a read problem at the PPS-100 due to low or high level SMPTE strip to tape. In that case the PPS-100 will go into Flywheeling, AKA free run, or free wheel (as JP mentioned). That happens when the PPS-100 can no longer make sense of the SMPTE info.

The PPS-100 manual is not very well written if you have no other experience with SMPTE. I think it assumes (wrongly) that people will be coming with some sync background. I like the J.L. Cooper products… I have both the PPS-1 (version 2) and the PPS-100, but the manuals are a bit too brief for the uninitiated.

The manual also wrongly states that you should run the signal through the mixer to tape, and that you can leave noise reduction on. I wonder if this was a marketing call on their part (They knew better). A lot of cassette multitracks were coming out that didn’t have the ability to disable noise reduction on the sync track. You could get away with it sometimes, but it’s not considered good practice… not reliable. dbx and sync is an absolute no, no.

The record level should be about –7 dB going by the LED meter on the TSR-8 (That’s the level I use with the J.L Cooper units). I’m wondering where you’re getting the +10 figure. The TSR-8 meter only goes to +8. The best way to set it is to turn the switch on the back all the way to your left (If you're facing the front of the TSR-8 or think of it as all the way towards the input/out RCA jacks). Then use a small screw driver to adjust the level on the back of the TSR-8 until you have –7 on the LED meter for track 8.

Are you sure dbx was off on Track-8? You have to push the Record-Ready button for track-8 before you hit the Sync-Lock button to disengage dbx on track-8. Of course if your not using dbx at all this doesn’t apply.

Another problem could be hot levels on Track-7, which can bleed through to channel-8 and cause sync read problems.

I don’t know Cubase that well. I had Version 1.xx.something a few years ago. I don’t know what V4 can or can’t do.

Some DAWs are better at sync via MTC than others. By DAW (Digital Audio Work Station) I mean PC (or MAC), Audio/MIDI interface and recording software. Sync problems can occur with any of these components of your DAW, or the sync device itself.

I run a PC-based DAW, using an Echo Layla and an older version of Cool Edit Pro (which I find to be more stable and less cumbersome than newer versions.

ATR/SMPTE to DAW via MTC runs flawlessly with either of these configurations… perfect sync with either my PPS-1 or PPS-100 like you have.

One other thing… before continuing reinitializes the PPS-100 (just in case it’s the problem).

To initialize, turn off the power. Next, hold in RUN button and turn on the power.

This puts the PPS-100 into a self test mode. The red LED will flash and some codes will
appear in the display). After 5 seconds, turn off again.

:)
 
Beck said:
I run a PC-based DAW, using an Echo Layla and an older version of Cool Edit Pro (which I find to be more stable and less cumbersome than newer versions.
Click to expand...

Hey Beck,
Do you know of a way to set Cool Edit Pro to always wait for an incoming MIDI signal automatically?

I am able to get Cool Edit to sync, but I have to enable the MIDI trigger option each and everything I want to sync. So, I open up my session, enable the MIDI trigger, hit play on the tape machine, but as soon as I hit stop, when I want Cool Edit to sync again, I have to enable the trigger again.

Have you found a way around this? I use N-Track now, because I don't have that constant MIDI trigger problem, but I still like Cool Edit's interface better, probably because it's what I first learned on.

Thanks,
-MD
 
There is some good advice above regarding signal chain,

If it's going to work it has to be:

smpte track tape out, to, pps100 smpte in, pps100 midi out to firebox midi in. Then select sync to mtc (or what ever it says) in your software. Always leave the mixing console out of the signal chain.

There may also be something to turn on in the fire box set up pages / mixer page, I can't remember.

I am still not convinced that MTC will be accurate enough to lock a computer to an analog machine, I only use MTC to lock drum machines to Midi sequencing, however I am prepared to be proved wrong ha ha.

Cheers

Alan.
 
This post will probably cause all kinds of aggravation, dissension and derision and I appologize for that in advance, but I have to post it anyway. I am amazed to still see threads like this pop up on various forums because there has been so much discussion of this subject over the years. Perhaps some folks have been able to achieve some measures (pun intended) of success syncing a computer playing sequences to a tape recorder but I still believe that is not the norm nor would anyone with serious intentions continue to persue this method once they have witnessed, first hand, the problems and thought long and hard about the impracticality of the task.

To start with, A DAW requires the most accurate clock source available to work reliably and deliver high quality audio. A Big Ben comes to mind and the reason they cost $1500.00 is because they are a extremely accurate clock suitable for the highest quality low jitter clocking. Throw into the mix (pun intended) various plugs and sample libraries streaming from disk, which in and of themselves require delay compensation all while the operator is trying to maintain a latency of 5ms or less and your DAW has a daunting task indeed just keeping everything glued together doing what it was designed to do.

Now throw out everything just mentioned because anyone that asks a DAW to slave to a tape recorder with varying speed (it will), tape slippage (it does), or whatever, is doing just that. Imagine while synced, the DAW and recorder start to drift and the DAW is suddenly 15 frames behind and depending on the software error compensation tolerance, instantly the DAW must loose about 24000 samples at 96 K. Or vice versa, the recorder speeds up for a split second and now the DAW must throw in 24000 samples that doesn't exist. Contrary to popular belief, it is not as simple as slowing down or speeding up the audio playback of the DAW. For precision audio, the DAW must still reply on the most perfect clock you have instructed it to clock too if you really want high quality audio.

Conversely, it is the simplest thing in the world to vary the speed of a phase locked loop capstan motor and that is exactly what a professional sychronizer will do it's when constantly comparing incoming TC from the grouped devices, one of which is master (ALWAYS THE DAW) and the rest are slaves (ALWAYS THE ATR's) by varying the speed of the capstan. It will work perfectly in every respect with no artifacts at all and will positively maintain the system resolved and locked to subframe accuracy until you either run out of timecode on tape or press stop. A professional synchronizer is designed to seek and locate addresses with almost no head wear at all by calculating needed tac revolutions in relation to the last read address and it will only retract lifters when very nearly at the seek address with very little overshoot/undershoot to finally park at the offset address. And it will require no more that 2 to 3 seconds of preroll to pull up or pull down the atr and achieve lock.

This stuff was figured out and perfected years ago by some really brilliant people (rocket scientists probably) and all of the intricacies of doing it are available on the web, but it still seems that about every 3 of 4 months or so, somebody else is trying to reinvent the wheel.

Reliable synchronization, for sure is very complex subject and operation, but it can be made very simple and absolutely foolproof for the operator with the right approach, synchronizer and a recorder that can be varispeed controlled externally. Really, for myself, I learned what little I do know about it the hard way. And that was by doing it wrong every single way that I could think of using only my pea brain. But in my years of experimenting, the only time I have been able to reliably sync a sequncer to tape was when using Smart FSK printed to tape and driving a "DEDICATED" midi sequencer. And I do mean midi only.


Regards,

Danny
 
Last edited:
I've synced my DAW to tape using the old Windows Logic 5 & PTLE from version onwards 5 to my tape machine a Fostex E16 hundreds of times without any problems. Also used Nuendo for while.

I've used a Phillip Rees TS-1 for this task, which is very simple to operate

I've even recorded tracks back from the DAW to the tape machine which play back in time & I've done this for songs up to 7 / 8 mins without hearing any problems.

I don't really know if there might be problems with sync after 20mins or so because I've never tried recording material that long, perhaps if you were syncing & recording music to for film using VHS / video or similar you might have some trouble??

For me MTC / SMPTE works fine for recording, perhaps there are differences between how different DAW programs & sync interfaces respond to & implement MTC / SMPTE data which causes problems. Also some tape machines might have too many speed fluctuations to be reliable enough to sync properly
 
You must log in or register to reply here.
Back
Top