Slaving the Deck Vs. Chasing Tape

[jpmorris]

A lot of the issues with making the DAW slave are going to depend on the setup and also the architecture of the DSP kernel. At the end of the day, changing a pointer in memory is going to take less time to achieve than varying the speed of a capstan attached to a heavy, belt-driven flywheel.

Changing things DSP-side is going to disrupt things with regards to the wordclock, but that should be a momentary glitch which probably isn't going to be noticeable. If it's something that persists from thereon, I'd say there's a design flaw in the audio kernel (like SONAR v1, which was wholly useless for any kind of sync).

Another factor is whether the DAW is trying to maintain sync with other digital devices as well. If you're doing a three-way lock between the tape deck and two digital devices which are linked via a wordclock, then yes, it's going to be messy.
For the simple case of a DAW and a tape deck, jitter shouldn't be an issue since there's only one DSP clock involved.
At the end of the day, I figure you're going to have a choice between a momentary pitch shift on the tape side and a momentary blip on the digital side.

I haven't done any professional audio coding, but I have written a few simple mixing kernels in my time so I have some vague idea of what's liable to happen internally.

As for the ATS500, it doesn't have MIDI. You should still be able to use it to sync with a DAW, though. You'll just have to stripe one of the DAW tracks and send that to ATS500 to do a code-only lock.

 

[Chilljam]

You guys have gone a bit off topic here. It's a forum on 'home recording' there isn't necessarily a RIGHT or WRONG in this field... 500 years experience doesn't always make you RIGHT or WRONG, lots of people do things differently and manage to achieve great things.

 

[Beck]

You guys have gone a bit off topic here. It's a forum on 'home recording' there isn't necessarily a RIGHT or WRONG in this field... 500 years experience doesn't always make you RIGHT or WRONG, lots of people do things differently and manage to achieve great things. You guys are just starting to abuse each other so cool off.

You know what? You're right! I just logged back in to delete my scathing reply to Cory. I believe we're having what they call, "A forum moment!"

I confess I'm in a pissy mood tonight, and really should have slept first before I responded, whether I felt offended or justified or not. I agree my tone was unkind. I'd appreciate it Chilljam if you would delete your quote of my post, as I came back here just now to delete what I wrote.

At any rate let it be known I'm feeling under the weather and I really didn't mean to come down on Cory like that. So I want to apologize to him and anyone else that had the misfortune of reading it.

And Cory if it doesn't get deleted don't take it to heart... it was completely uncalled for and not really the way I feel about you... at least not when I've slept. Forget it... really. We had an argument… that’s all and I’m over it.

 

[The Great Cobb]

You are all impressive to me. I dont have anywhere near the level of patience to do any of these things. In order to do any of this stuff it would take me hundreds of hours and recording even simple demos would be insanely difficult and time consuming for me. It would be impossible for me to make creative decisions on the fly with systems as complex as these. I envy your ability to be so patient.

 

[sweetbeats]

You know what? You're right! I just logged back in to delete my scathing reply to Cory. I believe we're having what they call, "A forum moment!"

I confess I'm in a pissy mood tonight, and really should have slept first before I responded, whether I felt offended or justified or not. I agree my tone was unkind. I'd appreciate it Chilljam if you would delete your quote of my post, as I came back here just now to delete what I wrote.

At any rate let it be known I'm feeling under the weather and I really didn't mean to come down on Cory like that. So I want to apologize to him and anyone else that had the misfortune of reading it.

And Cory if it doesn't get deleted don't take it to heart... it was completely uncalled for and not really the way I feel about you... at least not when I've slept. Forget it... really. We had an argument… that’s all and I’m over it.

Tim, I'm good with it, and I even read what you wrote.

I should have put "IMO" instead of "IMHO" when I expressed my thought of your "disservice" to readers. It wasn't humble, and saying so was patronistic and incongruent with the respect that I do have for you.

I also apologize for things I said that were clearly digs at you namely insinsuating you have a chip on your shoulder, and for taking your statements and expanding them into generalizations.

I clearly took offense to things you said that I felt were directed at me. You bet I have gotten much of my information off the web, but I believe the VAST majority is from this forum and I'm selective because there does seem to be consistency here and valuable experience. I don't wanna be a self-proclaimed web-troll expert and I apologize for times that I'm sure I come off that way. I DO feel that I have some valuable experience of my own under my belt. I've been performing and recording for 20 years both for myself and for others...and no not "professionally", but it is something I'm passionate about and therefore occupies a significant amount of time in activity and education. Not "more" or "better" but maybe somewhat unique and specific to me and that's what I try to share. My experience with sync and my reasons for liking what I do and how I do it and why I do it is one such area. But I really feel I've tried to be clear that I don't see my way as "right" except within the scope of a specific view of the audio, and I've also tried to be clear that I'm splitting hairs.

Anyway, I appreciate and accept your apology very much...and yeah I do know that scripture.

The "professional" method has always been to slave the tape machine.

No, that's incorrect. Synchronization of electronic devices such as synths, drum machines and computers in music recording, whether using SMPTE, Roland's proprietary pre-MIDI standard, or MIDI ALWAYS had the analog deck as master. I was there from teenage working in TV A/V production and music recording studios as it evolved. I watched Synchronization grow up to what we have today and I've always employed it from pre MIDI days in my own personal studios as well, and still do. In video production using SMPTE, the ATR multitrack was slaved to the VTR, but no, not in music recording. And by the way synchronization of two analog machines was always a pain in the ass! It was never ideal and was despised by everyone I knew in the recording community.

BTW, that was extremely informative.

A lot of the issues with making the DAW slave are going to depend on the setup and also the architecture of the DSP kernel. At the end of the day, changing a pointer in memory is going to take less time to achieve than varying the speed of a capstan attached to a heavy, belt-driven flywheel.

Changing things DSP-side is going to disrupt things with regards to the wordclock, but that should be a momentary glitch which probably isn't going to be noticeable. If it's something that persists from thereon, I'd say there's a design flaw in the audio kernel (like SONAR v1, which was wholly useless for any kind of sync).

Another factor is whether the DAW is trying to maintain sync with other digital devices as well. If you're doing a three-way lock between the tape deck and two digital devices which are linked via a wordclock, then yes, it's going to be messy.
For the simple case of a DAW and a tape deck, jitter shouldn't be an issue since there's only one DSP clock involved.
At the end of the day, I figure you're going to have a choice between a momentary pitch shift on the tape side and a momentary blip on the digital side.

Yeah I suppose it is important to note that I have at least 3 digital devices in additon to the Micro Lynx. My setup isn't just a sync box interfacing with two nodes, though in effect it sees the digital as one node as it just listens to the MTC coming off the MIDI interface. I have the audio clock generator card for the Micro Lynx and may at some point try using THAT as the wordclock master for the digital system as well as the machine control for the tape deck.

Good stuff everybody.

I think for anybody curious about sync, this would be a good thread to read. At the very least it presents what is in my opinion not a clear-cut simple issue. It is complex, but in most cases I figure those complexities can exist unseen in the background or are not really issues at all.

 

[miroslav]

It's hard to get a clear picture if one way is necessarily *better* than the other.
It may come down to each person's rig and what options they have available.

The one point that sticks in my head is that whatever "issues" are bing introduced by the "irregularity" of the tape deck's mechanical nature...
...never really goes away regardless of which device issues the "go" command.

At the end of the day...if you can't really hear it or see it...
...it might not be worth worrying much about it...?

 

[themaddog]

As we've seen, there are two schools of thought of the subject:

I'm of the school of thought that it's better to have the tape machine as the master. But, I'd like to add a little bit of what I know:

1- According to my RADAR manual, it's better to have the DAW chase raw SMPTE rather than MTC (my RADAR can chase either). I'll have to look at the manual again to see the precise language, but if I remember it correctly, it says the SMPTE is more sample accurate. It makes little difference to me, because without having to use MTC, all I do is stripe a SMPTE track onto my Tascam MS-16 and the RADAR will interpret the SMPTE track directly, without needing a SMTPE/MTC syncbox. Perhaps they meant, when syncing a tape machine to this DAW, that taking a syncbox out of the equation means a more efficient sync. Again, this is not my area of expertise, and if anybody is curious as to the actual wording, it's in the RADAR V manual, available online.

2- In regards to the fluctuation off the tape machine, this is a real issue, but I've found it's no big deal and I haven't heard any audio artifacts. When my RADAR chases the tape machine, it'll display how much it needs to fluctuate the speed. The most I've ever seen it fluctuate is + or - .2%.

3- Beck pointed out that if you use the DAW as the master, the tape machine will sometimes struggle to catch up, and this can screw up the recording due to the slight pitch changes on the analog end. This is true, and I have experienced it -- the one time I synced two tape machines. I would think that this would be less likely to occur, however, when a tape machine is slaving to a DAW because of how stable DAW's tend to be.

Just my thoughts! We're not only lucky to have the benefit Beck's years of experience, but it's also great to re-examine the standard methods from time to time, such as Sweetbeats has. I'll be willing to bet both of these recordists are satisfied with how they do things in their studios. I'd also be willing to bet they both record some pretty kickass music.

-MD

 

[sweetbeats]

...if you use the DAW as the master, the tape machine will sometimes struggle to catch up, and this can screw up the recording due to the slight pitch changes on the analog end. This is true, and I have experienced it -- the one time I synced two tape machines. I would think that this would be less likely to occur, however, when a tape machine is slaving to a DAW because of how stable DAW's tend to be.

All excellent points and well stated. I think a lot of times it IS dependent on what gear you are working with.

I want to comment on your point above. In my setup with the Micro Lynx slaving the BR-20T to Cubase through a Yamaha i88x (has the MIDI I/O) and the wordclock master is the Presonus Digimax FS (i88x is slaved to that via ADAT Lightpipe) I can run through a full reel of tape never seeing more than +/- 1 subframe error...and just a flicker at that. This is a combination of the Micro Lynx and the BR-20T transport, but it is a VERY effective lock, and like I said earlier can chase to a new timeline location and resolve on the fly in typically about 1 second once the Micro Lynx puts the BR-20T into PLAY, and not mroe than 2 seconds. And scrubbing on the fly to that new location appears similar (in transport behavior) to when I am operating the BR-20T outside of a sync relationship and am using the onboard autolocator to search to a cue point.

I got the Micro Lynx for a VERY reasonable price as well as a parts system and I got the BR-20T for relatively cheap too (about $550 for all that stuff), otherwise I couldnt have afforded it...and I had to modify an ATX computer power supply to power the system...AND make up the interface cable...AND learn how to use the thing. There were headaches for sure and I needed lots of help but hopefully you all see it is effective, but not everybody, and in fact most, can't afford the hardware or can grab those deals. And I wanted to be able to use the DAW interface/controller (Yamaha 01X) to control the system AND I have multiple digital components as well. Pianodano is in the same boat that way only way deeper...the DAW, Tascam DM-24 mixer, 3 DA...78's? Keyboards/sequencers. I think it is fair to say it may be a mess if he tried to slave the whole system to his MS-16.

I guess what I'm saying is I hink it is of paramount importance to ascertain whether or not you have a NEED to set the tape deck as a slave. I think there are situations that warrant it. Maybe even if you want to do it because "its there" like...ummm...somebody.

 

[dementedchord]

ok so each to his own... i can on ly say it was reall obvious back when i had 2 adats... one org and one xt... if the orig chased the xt it was WAY slower to sync... it having the lesser of the two transports... a quick lock being optimal for me...

 

[pianodano]

What a thread.

I began my solo home recordings with a Tascam 38. It wasn't long thereafter that the Commodore 64 and the necessary interfaces and rudimentary sequencing software became available and was widely advertised in mags such a Recording. At the time all that was possible for syncing on limited budgets and setups was FSK. It was a nighmare. Granted it would work ok to maintain lock to tape. But because it was dumb fsk and had no positional address information, you had to start lock at the top of the signal/ song. So many of us that used sequencers at that time learned to do a temp recording of the sequence data to a single guide track on tape which would then leave 6 tracks available for actual audio. We/ I would finish recording whatever audio was going on tape with planning and foresight as to bounces and then go back to using the sequnce data during the mix or when all the audio was finished. It was a long and tedious process and was never the best way to work.

But SMPTE was becoming available. So in passing, let me mention the EXTREME difficulties in using SMPTE for music sequencers vis a vi tempo maps. Suffice to say that that is the stuff for NASA engineers to ponder.

As soon as Roland developed their flagship MC500 dedicated hardware sequencer, at long last I as a keyboardist had the possibilty of using crash free sequencer. So I quickly plunked down the CC and paid $1800.00 for it. I still have it, but now only use it for live work when a PC crash is not an option. At the time it came out, pc's as we now know them, were still a thing of the future.

Unfortunately, the MC500 was only 4 tracks in late 86 but the MK II 8 track software was released in the summer of 87 for those (like me) that waited for the MK II hardware version initially. That wonderful sequencer was a joy to use, you could merge tracks, was crash free as advertised and was as fast to work with as anything you can imagine. There was still a major drawback though. It could only print dumb FSK to tape. Much later, we learned it could read smart FSK, but again in the beginning, there was nothing available at a reasonable price that could print smart fsk to tape.

Midi time code was a couple years off yet. Did I mention the difficulty in working with tempo maps using SMPTE ?

So even with those wonderful advances in midi technology and rudimentary simplified syncing, there were still severe constraints placed on the multitalented artist that was working alone and with limited budget.

But many developments where taking place for the pro studios using SMPTE. API Mag Link, Quad Eight, 3M to name a few where making signifcant inroads into reliable systems for synchcronizing 2 or more machines. And Compumix had by then actually developed their portable automated mixdown system. But it was all very expensive. And they all used SMPTE.

As I got deeper into studying sychronization, I learned it never had been possible to interlock 2 machines using hysteresis synchronous or servo motors without extreme complexity. Those methods use capstan rotational speed controls which are always inaccurate because of inherent tape slippage/ via the pinch roller. Extremely rudementary and simplistic methods used phase lock to keep two machines in phase using stripped tracks but there was no address information. You would mark the the 2 tapes on the two machines with a crayon, line up the mark at the heads on both machines and try to hit start on both at the exact same time. They would stay in phase ok, but might not be at the exact same place in the song at the same time. What a system. But that's all that was available at one time.

Chapter 2.

Moving on in development and this important to understand and consider for those that might decide to varispeed a DAW .. unless extremely accurate tempo is not critical to you - (But then again on the other hand it better be with a DAW chasing) - - strictly speaking of tape and two machines, it is not necessary that both tape speeds be constant, only that the speed of both tapes must be equal at all times. Drats. So if tempo is critical, which it should be to a musician or maybe he/she didn't master the requsite internal clock training via the metronome he should have practiced with for a couple years, there must be a comparator that references the tone coming off the master machine and then a resolver to control the speed of the master machine to keep the tone printed on the master sync track at the exact same frequency as the reference tone so that you can have the master machine maintaining accurate and constant speed. In my mind and based on all that mumbo jumbo alone, score 1 point for using the DAW as the master. You don't have to worry about master tape slippage anymore.

But anyhow, only when tape slippage is accounted for should you think about syncing something to a master tape tone. Or you could do what some say to do .... which is -fergit all about it and hope for the best as in maybe nobody will notice.

For my part, as I began to integrate the great sounding 8mm digital audio machines that were becoming affordably available in the 90's, the fabulous digital mixers that Tascam was finally making affordable for us musicians and the amazing advances in computing power, disc storage and sample libraries that was rapidly coming to market, I saw that at long last, I might not any longer be limited by track counts. But I still loved the sound of the wider format tape machines, notably, MCI. But 2" tape costs where just a bit too much considering the amount of tape I used experimentally. So I went back to the good old MS-16 which is a great sounding machine in it's own right.

Fortunately for us, Tascam incorporated the capablility of syncing that machine and many of its siblings in their inards. But that technology was made for syncing to machines. They even made a sychronizer for SYNCING TWO or more MACHINES. Sweetbeats can tell you all about it.

Unfortunately, there was nothing that was cheap available that would work without fail, that you could rely on when you had paying clients, that could natively interpet and process the later developed midi time code data that DAW software engineers decided to output when you wanted to sync tape to DAW for controlling and syncing a ATR. That's really like trying to integrate two different technologies. So I guess in their infinite wisdom they said something like, hey, let us put a couple wigets in the interface that says something like - do you want to "trigger and freewheel OR full chase lock" and then we can have a box that pops up that says "WAITING FOR MIDI SYNC" and wala . . . that all there is to it according to the published data.

So according to what we have all been told, we know that every daw will sync to incoming TC.

Yepper. Me being as optimistic as I always try to be when using new technology, bought all kinds of gizmo boxes trying to make that idea work.

Unfortunately, that simplistic solution didn't work well in my setup because. . . . . . . . .


To start with - The sheer complexity of all that synchronization techno mumbo jumbo was overwhelming in the beginning phase to a simple musician such as myself and as I invested in and became more and more involved in high quality sample libraries, electronic drumsets, farm machines running samples over a network being handled by the host, dozens and dozens of midi ports and the inherent timing problems that are plainly evident to anyone that has ever attempted to build up a complex yet maintain at the same time (pun intended) timing accuracies on a reliable system and has maybe actually studied these problems in depth, will have learned this basic fact. The windows os sucks big time when processing midi data while simultaneously processing audio. It became more and more evident to me that there had to be a better way because the DAW software didn't always exactly work as advertised. Fact is, it often does not. Many of those dang software "experts" said my system was to way too complex or that I simply couldn't expect the software to not have bugs. More like buggy comes to mind.

Just the issue alone of a DAW running two separate clocks drove me nuts for weeks on end. One for audio and one for midi ? I mean give me a frickin break. And the way the system polls for midi data when it might have the cycles available ? And no hope of MS ever fixing it? And even the enginners that designed the USB to midi interface code freely admitting that they could not foresee the inacuracies introduced by the os as being as crippling as it can sometimes be for those musicans with complex systems and extreme sensitivity to timing errors caused by midi latency and jitter? Not to mention audio latency to boot ? Trying to maintain sync and seeing midi jitter of 40 to 50 ms ? That couldn't be foreseen ?

I literally spent hundreds of hours chatting with amazingly brilliant minds over the now on this very thread belittled internet desparately trying to figure out EXACTLY how to really and realiably make all this crap work so that maybe I could finally get back to recording music and get the technology the heck out of the way. When I got really serious about it, I did not let money or the lack of it be the determining factor in the resulting system.

I used what I learned from many many learned people, much of it via the internt, some of whom have already left this planet, to put together a idiot proof system for myself, of which I am very proud and pleased with. And while I have also tried my very best to pass on what little I really have learned so as to be of a little help to other people that really do display a eager interest in engineering a foolproof way to synchronize but may only be in the beginning stages of putting together ultimately complex systems, I recognize that there are always those that have discovered another simpler, cheaper way. So to them, I say figure it all out for yerself and do whatever floats yerboat.

Fortunately for me, I have seen the technology progress from the beginning to where it is now and have had the benefit of discovering for myself what is and what ain't ready for prime time. Unfortunately for me, I have spent much of my life always trying it the wrong way first. It has been a lifetime process . . . . in my own case.

So again for anyone that wants to reinvent the wheel -go ahead.

And to anyone not using Protools that wants to get on with making music and have in place the solid foundation for building up a complex system of sychronized DAW, midi systems and ATR machines, get a Timeline Microlynx while they are still out there. The majors let them go for a song because Profools syncs with only a minimum investment in more of their proprietary crap. That was what Protools was built on. Syncing. Using PROTOOLS METHODS AND HARDWARE. They broke the back of the entire analog tape industry on that one premise.

























And that's all I've got to say about syncing.


Regards,

Danny

 

[miroslav]

I can see where the Timeline Microlynx would be beneficial in a multi-machine/DAW/MIDI type of sync environent...
...but what do you really gain with it if it's a simple deck/DAW sync setup AND if the deck already has a built-in SMPTE/MTC syncronizer that is doing what the Timeline Microlynx does...converting incoming/outgoing time code?

 

[miroslav]

I’ve been kicking this discussion around on another forum…and all of a sudden thanks to some suggestions from others…a whole new approach came to light that allows tape decks and DAWs to be used without ANY consideration for sync…kinda like the CLASP system minus all the neat electronic switching that it has (but not really a deal breaker).

Basically…just use the tape deck…AS A PLUG-IN…which is what we were talking about with the CLASP system.
You never need to sync up the DAW and the deck….never.

The suggestion was this:
You arm both the tape deck and the DAW simultaneously for record.
Hit REC on the deck and then also on the DAW…taking the output off the PB head of the deck into the DAW.
If you need more tape tracks…you simply “rewind” your DAW back to its “0” start point…hit REC on the deck and then also on the DAW, same as before.
You simply monitor all your playback off the DAW…not the deck.

Yeah…the tracks on your tape from pass to pass will never be in sync with each other…but who cares…the tracks are already in your DAW, that’s what you are monitoring. You get full tape deck use, without the need to sync.
By doing it this way...the DAW runs on its internal (or external Word) clock...and the tape deck runs on its internal clock. So by removing the need to sync them together, you actually minimize the Wow & Flutter since they are not pushing-n-pulling at each other trying to stay in sync.
Someone pointed that out to me...that syncronization actually introduces MORE Wow & Flutter than if you just let the deck run on it's own, internal clock...and of course, same goes for the DAW.
There is no Master & Slave relationship.

It’s a very interesting way to go….

Of course...if you GOTTA keep the multitrack tapes for future use, and all in sync with each other...that's a different story.
Currently I still hold on to and save my multitrack tapes...but I gotta say...I've wondered why I was bothering with that, since I already had them in the DAW and archived digitally 3-4 different ways. I never thought about this "no sync" way to record with a tape deck and DAW before, but I still often wondered what was my reason for saving the multitrack tapes once I dumped the into the DAW...???

Granted...some folks will really feel the need to save their tape tracks even after they dump-to-DAW, maybe for some possible future remix...etc...so of course, for them it's back to the sync methods, since without timecode stamps those tracks will be hard to use later...thought you know, even that's not too out of the question as long as you have maybe a click reference or similar that will aid you in lining up the tracks later in the DAW...but that's another story....