Getting a Tape Machine this weekend, let's talk Otari

Moooon River...

It's true!!
Just about everyone on the Film Crew...

Grips/Set Decorator/Truckers etc.
Almost all of them smoked. (Per the head Union Trucker)
Chris
P.S. I did hope to advance from the Local Lung Union... To become a Hand model. But I'd have to cough up more Dues!
(Hey I gotta throw in somethin' )

I have an MX-5050. Perfect operational shape, though looks-wise it's not the prettiest. Works for my purposes, though.

I said it another thread, but subtle vs. dramatic is, I suppose, a relative argument. I don't think the effect is subtle at all and no, my machine works fine - was worked on by a veteran tech in NYC who's seen hundreds of machines over 40+ years. We're talking about not having to de-ess things, dip upper mids or add 30-60 hz, no need to cut the fat from electronic instruments and so on - is that subtle? That's up to you I guess.

Once again, I'm not someone who thinks highly of A-B methodology because it relates very little to actual working, akin to listening for hi-hats when hearing the whole mix.

The issue with that approach will be tape machine drift. Without the deck locked to a synchronizer, and then the synchronizer resolved to black burst it will "drift". How much, how little? It depends. Enough to care, enough to be unusable? Depends. Even the voltage of the AC can affect the speed of the deck. So can heat.  Or the phase of the moon.  Analog tape machines do not play nice like digital machines do. So - if I was in your shoes -

I'd feed a drum machine or external click to both DAW and Tape and record both.  Use a side stick or something with a very clear transient for easy visibility. Then feed the analog recorded click and record it back into the DAW.  Then line up the first "clicks" to each other. Then play back the DAW and see how much "drift" there is over time.  Do you hear phasing, flanging, actual delay between the two? It's likely there will be some degree of this - the longer the recording, the more likely.

When doing film/ video, we often had 15+ minutes section, and the drift could be substantial - causing lip sync to look like a bad Japanese dubbed film. If you're only doing 3 minute songs, perhaps it will not be so egregious. At any rate, record the max time you might ever use. Then..

As mentioned, line up click #1. Go 30 sec in and see if there is any drift. Then 1 min in. Then 1:30, etc. all the way down to the end. If there is zero drift, then you are the luckiest guy around. IF there is drift - don't freak out. It's 100% normal.

Go to your daw. Figure out a formula of how many samples you will "pull up or slip back" the DAW click every 15, 30 or however many seconds. For example, if it's drifting 10 samples every 30 seconds, make an edit every 30 seconds in the TAPE to DAW track and slip the 1st edit 10 samples, the second 20 samples, the third 30 samples, etc.. Some elements like drums will be more critical. Some things like a string pad might be OK all the way through. Its all going to depend on the amount of drift.

Keep in mind, the amount of drift CAN vary from day to day depending on the electrical voltage and how regulated your deck is. The RIGHT way to do this is to lock and resolve your deck with a synchronizer and blackburst, and you may be able to be completely stable with zero drift.  That approach is several levels up the analog food chain and requires $$$, special cabling and decks, and a lot of experience/knowledge of a tech who knows what they are doing.

Good luck.

PS - even in the above scenario, drummers or producers or engineers who are EXTREMELY sensitive to timing issues might start to feel uneasy as the chunk of audio you selected (15 sec, 30 sec, 1 minute, etc.) comes towards the end of its length - only to feel "OK" again, as your edit brings things back into perspective.

A way to deal with this is to figure out how many samples out you can "feel" - either consciously, or subconsciously, and make your edits for that length of time. This could necessitate 20+ edits in some pieces. It's tedious. And your level of rhythmic sensitivity is going to be the key factor as to how complicated its going to be. I had one very famous producer I worked with who could easily feel 10 samples. This approach with Tape would have been a nightmare scenario for his work flow.

I believe you, you’ve been doing this a LOT longer than I have. I’m still confused though. If I’m recording off the repro head in real time, everything’s going to align over the long run, no matter how long you record. I can understand “micro” blips in timing, but it’s not like those are going to build up over time and physically offset the track like it would if you recorded to tape and then tried to play it back after (that I can understand).

Not only that, but you’re describing issues that would end up happening even if you took the digital equation out of this and were just working off of tape the whole time. But, as with most things in life, I’ve found I just have to try it and see what happens most of the time 😂. I just feel like I’m missing something on a fundamental level here, as this “should” work out fine as long as I’m recording off repro, in real time for all of this.

I’ve never used Clasp. My impression was it was a Synchronizer ITB? Like those old synchronizer’s Micro Lynx Timeline.. with their own master remote controls.. slaves ALL the Machines/video to that synchronizer. Everything gets a SMPTE stripe on an unused channel.. later machines I think even had a tiny head just for SMPTE? So you didn’t have to use up a channel. I never got to use that stuff, born a year too late, except I did slave my 1972 a80 to a synchronizer for the modernized capstan servo.

In your situation, since you aren’t tracking, rewinding, then dumping- you are just capturing… it depends on the machine. The older vibey stuff could possibly drift, as the take up reel get heavier the speed can slow and suddenly your measured gap is wider. For stuff after the crystal clocking was invented, probably will stay true even as tape reel weight transfers from reel to reel. (Why I used a syncro for the a80) Even then- it was a huge engineering challenge up to the very end of the era to handle a fragile tape and heavy weight transfers while staying at an exact speed. This is why fear of robots taking over doesn’t worry me much.. they’ll just break down.

Your machine is probably new enough not to worry about timing too much. Also it’s 1/4” so the reels aren’t that heavy. My 1/4” mci doesn’t give me too much trouble, if it does I know it. And LPR35 is one tape, since it’s lightweight, can mess up the speed. I have to press the torque limiter button on the MCI, and splice onto some junk reels so that both reels are weighted down somewhat equally.

Your method seems good. I usually use plugins with time in negative samples to monitor, then at the end “select all” move all the tracks at once. If I screw up, they have a timestamp so I can try again. The time stamps will usually all line up relative together anyway. When bouncing pre-recorded stuff to/from, Any tape hiccups/wow can just punch in right over the problem area, works perfect

Brad McGowan did this very thing years ago, and even put out a video about it on YouTube. It was basically a poor man's CLASP, and worked on the same principle before CLASP even existed. If it's a real time transfer, as it's being played/recorded, I don't see the issue with any timing or alignment concerns, other than maybe not doing phase coherent stuff like drums in multiple passes.  IOW, bounce the drums down to a stereo track and transfer that.  Brad had/has an 8 track, so that made it a little easier to not have to bounce down so much, but he was doing the record and bounce off the repro head real time, just like is being described in this thread.

Yeah at the very least you want a 4 track, one track for time code and possibly a guard track. 8 tracks is ideal, radar and PT lock pretty decent to tape.Even when the MTP was current they locked better than a Adam Smith Zeta that cost 10 times as much.

You had to order it that way, as you needed an external box as well as a slightly different head assembly (I think the time code printed on a center track).

I still don't have a Calibration Tape here, so it's a lot of back and forth to get things where I think they need to be. I had the machine calibrated "correctly" yesterday and ended up tweaking a bit by ear to bias/lean things more towards the head bump and to roll off the highs a bit. I think I went WAY overboard, as I tried drums and vocals and could notice I was loosing a lot of info starting around 4k-6k, pretty steep drop off. That's what you get when you think you can make it "better" by ear  . It's all good, this is part of the fun for me and the learning process. I went too far yesterday.

I decided to go the other way this morning and see what happened when I calibrated it a bit flatter up towards 10K, which technically made the low end head bump area pretty flat on the meters, but it sounds a bit thinner to me listening on guitar. I was starting to think that the Otari was just going to be fairly dark and beefy, which was fine with me. I listened back to some files a few people had run for me through their MTR's that were calibrated correctly by an expert, and all of them were fairly flat with just a slightly rounded off high end and a nice head bump down low. I figured I better shoot the other direction just to make sure this was just me dicking around with the calibration as opposed to the machine having an issue. I'll link another guitar pass below where I got the machine about as bright as I could while seeming pretty balanced accordingly in the other frequency areas. Surprisingly, you can get these machines pretty damn bright!

Obviously too much, and I'm not a huge fan of the sound, but this is just experimenting. I'm sure I'll continue to screw things up going back forth for the next few days and eventually land on some happy middle ground. I don't want it this bright and thin, I also don't want it as dark and thumpy as I had it yesterday...


drive.google.com/drive/folders/1v-9jXfkmnY2QfhEd3lM11RSYUvWLkqsv?usp=sharing