avatar

burns46824

Gold Finger

Posts: 270 Member Since: 07/02/2011

Lead

Apr 3 17 6:09 PM

Tags : :

My setup is analog.  My Soundcraft Ghost console has mute automation, which I'd really like to start using in order to get my noise floor down without having to sacrifice audio quality with gates.  I think the only way to go about this is to stripe track 24 with SMPTE and feed that into the console SMPTE input.  (You can also use MIDI timecode, but I'm pretty sure that's only viable with a DAW setup.)

Anyway, I found this SMPTE WAV generator website and I was wondering what frame rate, bit depth, and start time I should use:

http://elteesee.pehrhovey.net/

I figure on striping the entire reel.

Also, how hot do I need to record SMPTE?  I want to record as low as possible so I don't get crosstalk on channel 23, making it unusable.  I'm recording at 30 IPS.

Thanks.
Quote    Reply   
avatar

burns46824

Gold Finger

Posts: 270 Member Since:07/02/2011

#2 [url]

Apr 3 17 6:34 PM

Hehe, oh man, if I really have to sacrifice TWO channels for this, I don't know if I want to bother...

Even at -10 VU, you can hear it on 23?

Quote    Reply   
avatar

Mike Rivers

Aqua Marine

Posts: 2,513 Member Since:13/10/2012

#3 [url]

Apr 3 17 6:49 PM

burns46824 wrote:
My setup is analog.  My Soundcraft Ghost console has mute automation, which I'd really like to start using in order to get my noise floor down without having to sacrifice audio quality with gates.  I think the only way to go about this is to stripe track 24 with SMPTE and feed that into the console SMPTE input.  (You can also use MIDI timecode, but I'm pretty sure that's only viable with a DAW setup.)

Anyway, I found this SMPTE WAV generator website and I was wondering what frame rate, bit depth, and start time I should use:

Also, how hot do I need to record SMPTE?  I want to record as low as possible so I don't get crosstalk on channel 23, making it unusable.  I'm recording at 30 IPS.

 


There are converters between SMPTE and MTC (in both directions) but since you have to record something on analog tape, SMPTE is the way to go. If this project doesn't have have to sync to video, keep it simple and use 30 frames/second. I see that program generates a WAV file which you'll then have to play from a DAW (or a media player, maybe even a phone) to get audio. Or you could burn a CD from it. I see that program gives you a choice getting your time code as an 8- or 16-bit WAV file. Might as well use 16 bits. It might make for cleaner audio, and that's what counts. Disk space is cheap, and you can put an hour's worth on a CD.

For your application, start time is really arbitrary. By starting the time code at 58 minutes (the generator program's default), you can have some slop and start the actual program at 1 hour even. But it might be less confusing to just start it at zero.

Generally -10 dB on the VU meter is about as high as you should go, but try it and see what your console needs to reliably read it and record at a lower level than -10 if you can. And if you're using noise reduction, bypass it on your time code track. And you really shouldn't count on using the adjacent track unless you have to. Most people feel reasonably safe putting bass on the track next to the time code because you can roll off some of the high end (where the SMPTE hash is most noticeable) and the bass will still sound OK. You don't want to put something on the adjacent track that's hot and has a lot of high end or transients or crosstalk from that might interfere with your time code.

You should experiment a bit before you do anything important.



For a good time, call mikeriversaudio.wordpress.com

Quote    Reply   
avatar

chance

Aqua Marine

Posts: 2,665 Member Since:30/01/2011

#4 [url]

Apr 3 17 8:43 PM

In my past I have found the outer tracks of tape sometimes gives drop outs. What are you syncing to? Is your analog machine set or able to chase TC? Either way you need a master source. There are drop and nondrop Frames. 24fps for film, 29.97 fps for audio. I believe you want to sync to your "mute?. Does your mute generate smpte so your tape machine can chase it? (or vise versa) My console has a dedicated processor for mute automation, but I never used it

Quote    Reply   
avatar

burns46824

Gold Finger

Posts: 270 Member Since:07/02/2011

#5 [url]

Apr 3 17 9:51 PM

mikerivers wrote:

burns46824 wrote:
My setup is analog.  My Soundcraft Ghost console has mute automation, which I'd really like to start using in order to get my noise floor down without having to sacrifice audio quality with gates.  I think the only way to go about this is to stripe track 24 with SMPTE and feed that into the console SMPTE input.  (You can also use MIDI timecode, but I'm pretty sure that's only viable with a DAW setup.)

Anyway, I found this SMPTE WAV generator website and I was wondering what frame rate, bit depth, and start time I should use:

Also, how hot do I need to record SMPTE?  I want to record as low as possible so I don't get crosstalk on channel 23, making it unusable.  I'm recording at 30 IPS.


 


There are converters between SMPTE and MTC (in both directions) but since you have to record something on analog tape, SMPTE is the way to go. If this project doesn't have have to sync to video, keep it simple and use 30 frames/second. I see that program gives you a choice getting your time code as an 8- or 16-bit WAV file. Might as well use 16 bits. It might make for cleaner audio, and that's what counts. Disk space is cheap, and you can put an hour's worth on a CD.

For your application, start time is really arbitrary. By starting the time code at 58 minutes (the generator program's default), you can have some slop and start the actual program at 1 hour even. But it might be less confusing to just start it at zero.

Generally -10 dB on the VU meter is about as high as you should go, but try it and see what your console needs to reliably read it and record at a lower level than -10 if you can. And if you're using noise reduction, bypass it on your time code track. And you really shouldn't count on using the adjacent track unless you have to.

Really good advice, thanks.  So I might be able to record even lower than -10 VU?  That's good to know.

In reply to Chance, the idea is to send the SMPTE from the tape into the console in order to automate channel mutes so that I can get my noise floor really low.

Quote    Reply   
avatar

chance

Aqua Marine

Posts: 2,665 Member Since:30/01/2011

#8 [url]

Apr 3 17 10:40 PM

burns46824 wrote:

In reply to Chance, the idea is to send the SMPTE from the tape into the console in order to automate channel mutes so that I can get my noise floor really low.

How are you striping the tape with the SMPTE? What is the TC source?
and yes as Terry noted the smpte noise crosstalk is expected. On 16trk 2" not as much

Last Edited By: chance Apr 3 17 10:44 PM. Edited 1 time.

Quote    Reply   
avatar

chance

Aqua Marine

Posts: 2,665 Member Since:30/01/2011

#9 [url]

Apr 3 17 10:57 PM

I used to use an MTS-1000 SMPTE generator linked to an MSR-16 tape machine for remote video projects. I don't have the deck anymore but I "might" still have the MTS-1000. You no doubt know that your tape machine will be the slave and do the chasing

Quote    Reply   
avatar

queef bag

Platinum Blonde

Posts: 1,275 Member Since:24/02/2011

#10 [url]

Apr 3 17 11:00 PM

ya could record bass next to time code, if it was a nice round tone... and roll off highs in the mix to get rid of cross talk.

but try and record tambourine on the back beat next to it and watch the time code reader loose it on 2 and 4.

jf

Quote    Reply   
avatar

injured ear

Silverado

Posts: 79 Member Since:20/02/2012

#11 [url]

Apr 3 17 11:00 PM

What's your tape machine? Some are more forgiving of crosstalk/bleed than others.
A lot of folks are saying that you can't use track 23. I disagree, but I don't recommend printing your lead vocal comp on 23. (though I have witnessed it more than once on major label releases)
Generally we used to record the click track to 23 as it was the most gate-able/mute-able signal.
Be careful of recording loud signals to 23 as the crosstalk to 24 can interrupt clean reading of the timecode, but if you're only recording SMPTE to drive your automation system, you can probably afford a dropout here and there. Maybe run a highpass on channel 24 before it hits the timecode reader.

All of this reminds me of why I don't miss analog tape.

Greg Thompson

www.injured-ear.com

Quote    Reply   
avatar

technologyworks

Gold Finger

Posts: 661 Member Since:21/02/2011

#12 [url]

Apr 4 17 4:48 AM

Does your desk have a display for the SMPTE it is receiving?
You could start at -10, and feeding the desk from the repro head try reducing the record level to see what level becomes unreliable. It's just mute automation, not keeping 2 multitracks in sync.
Do you have a patchbay? I always used to take the track 24 output straight from the recorder's XLR to it's destination to keep the SMPTE signal away from the audio patchbay.
I bet you could pick up an old SMTPE generator/reader for peanuts on Craig's list or evilbay. it might take some of the guesswork out of setting up.

Peter

Peter Beckmann

Technologyworks Mastering 

Quote    Reply   
avatar

Mike Rivers

Aqua Marine

Posts: 2,513 Member Since:13/10/2012

#13 [url]

Apr 4 17 5:56 AM

chance wrote:
I used to use an MTS-1000 SMPTE generator linked to an MSR-16 tape machine for remote video projects. I don't have the deck anymore but I "might" still have the MTS-1000. You no doubt know that your tape machine will be the slave and do the chasing


What he's synchronizing is the console's automation system with the tape position, so the tape machine will be the master.

As for the time code source, he found a web site that generates a WAV file of time code, so I assume that his plan is to get the WAV file, then work out a way to play it into the tape deck so he can record a time code track on tape.



For a good time, call mikeriversaudio.wordpress.com

Quote    Reply   
avatar

extrememixing

Platinum Blonde

Posts: 2,319 Member Since:02/02/2011

#14 [url]

Apr 4 17 9:28 AM

I am so happy that I have nothing to add to this thread! I don't miss smpte, or its bleed to adjacent tracks at all. Or chasing noise and gating things. Or those first couple of hours of a mix writing the mutes. To each his own.

Steve

Quote    Reply   
avatar

burns46824

Gold Finger

Posts: 270 Member Since:07/02/2011

#15 [url]

Apr 4 17 11:40 AM

Thank you guys, again, for all the info. This is really helpful. I really enjoy using the old technology, actually. I recently used a pilot tone to sync a LinnDrum to my tape machine, for example.

My multitrack is an MTR-90III with all the channel cards modified by Alan Garren (waltzingbear).  His mod sounds great.  I do not miss digital!

Quote    Reply   
avatar

MusicalRival

New Forum Friend

Posts: 3 Member Since:26/03/2017

#16 [url]

Apr 4 17 1:50 PM

I could never understand why SMTPE was Sooo hard to change on video. I could never get my DAT to sync with multiple sony cameras so I could sync them in post. I always had to write down the offset between the DAT and every camera and then calculate what every camera was shooting for any particulate time, so the sound would match that camera. SMTPE is just a control track of audio like a old phone modem. I should be able to dup the video with a new SMTPE.  

Quote    Reply   
avatar

Mike Rivers

Aqua Marine

Posts: 2,513 Member Since:13/10/2012

#17 [url]

Apr 4 17 2:43 PM

MusicalRival wrote:
I could never understand why SMTPE was Sooo hard to change on video. I could never get my DAT to sync with multiple sony cameras so I could sync them in post. I always had to write down the offset between the DAT and every camera and then calculate what every camera was shooting for any particulate time, so the sound would match that camera.
 


Were your DATs equpped for time code synchronization? There aren't many of them. I have a Sony PCM-7010 that can sync to time code. I've never used it, but when I read the manual about how to set it up (and why) my brain glazed over.

A time code DAT has a separate input for SMPTE time code and will synchronize the DAT time code to the SMPTE time. But without the time code input and no means to jam SMPTE time into the native DAT time, there's no way to synchronize them. Another thing is that the frame rate of a DAT os fixed so it doesn't always correspond to the frame rate of the video recorder. I really don't know how DAT internally deals the fractional frame rates (most video is 29.97 frames per second) where real time is 30 frames per second. It should be capable of correcting for drop-frame time code forma, which is something akin to leap year, allowing the time to get off 1 frame every 1/30th of a second, and correcting for it by going from frame 29 to frame 2 at the end of a minute, except that every 10 minutes instead of starting at frame 2 on the minute, it starts on frame 1 to get things back in line, more or less, for another 10 minutes. 

And I don't think it was invented on April 1.



For a good time, call mikeriversaudio.wordpress.com

Quote    Reply   
avatar

berolzheimer

Aqua Marine

Posts: 2,871 Member Since:29/01/2011

#18 [url]

Apr 6 17 3:06 PM

I often found back in the day that lower record levels worked better for a lot of readers- you might even try -20 as a starting point. That said, the lower the level of code, the more likely it is to be interfered with by crosstalk from the next channel. One thing I used to do to decrease timecode bleed onto the adjacent track is use noise reduction on that track (23, NOT 24), Dolby or DBX, even if I wasn't using it on the rest of the tracks.

Quote    Reply   
avatar

maarvold

Aqua Marine

Posts: 3,105 Member Since:23/01/2011

#19 [url]

Apr 8 17 1:15 PM

mikerivers wrote:

drop-frame time code forma, which is something akin to leap year, allowing the time to get off 1 frame every 1/30th of a second, and correcting for it by going from frame 29 to frame 2 at the end of a minute, except that every 10 minutes instead of starting at frame 2 on the minute, it starts on frame 1 to get things back in line, more or less, for another 10 minutes. 

And I don't think it was invented on April 1.

Drop frame accomplishes its goal, but seems like such a 'Rube Goldberg invention' way to do it.  

When I was working on episodic tv scores, slaving 2" tape using Lynx's and 3/4" BVU-type video decks, I used to like to make the video the slave.  I'll bet the dat scheme works in a similar way. In the 2" with BVU world, both machines are resolved to [29.97] black burst; When you hit play on the 2" (which is in "Ext" for its speed source) it begins to play and the Lynx compares its playback TC rate to the black burst generator and when they are in sync it is considered locked.  The video deck--which is slaving--behaves much the same way: the concept is that it slaves, location-wise, with the incoming SMPTE from the 2" and when the two machines are in sync, the Lynx releases the need to control the machine by the timecode and instead keeps the machine in sync with the black burst clock... this is called "Lock and Release".  If the [slaved] video machine's timecode drifts outside a predetermined window of satisfying the "Lock (to SMPTE numbers)" requirement--let's say 1 frame off for argument's sake--the Lynx comes out of "Release" and again uses the SMPTE numbers to lock the machines together, then rereleases the video to blackburst lock.  The main point of this is that--just like digital--the 2 sources are slaving to a master clock and remain in sync for extended periods of time.  FWIW.  

Quote    Reply   
avatar

waltzmastering

Platinum Blonde

Posts: 1,638 Member Since:02/02/2011

#20 [url]

Apr 9 17 1:05 PM

The price for the boxes needed to sync are super low now compared to where they used to be.
Had an Adam Smith Zeta 3 and remember them costing a few thousand.
then the Digidesign Sync, which worked good, but can be had used now for $100.
For mute automation, you might want to see if it will accept FSK or Midi Time Code as well. gl

Quote    Reply   
Add Reply

Quick Reply

bbcode help