Headphone Cue Mix in a digital environment

[Kingston]

But in some places, he might have to play 0.2 second ahead so his contribution arrives at the singers at the right time.[1]  And he may hear the choir 0.5 sec after what he plays and what he plays at a different time too!  Incredible.

But where then should I sit in the church to get the best combined timing!??

Another way too often recurring version of this is when a band plays in absolute perfect groove (whether live or studio) and the great guitarist is taking into account his time delay of the amp 10 meters away. In comes the recorder-person and sticks a single close mic to the amp and all of us listeners are left complaining what a crap timing this band has.

 

[ricardo]

But in some places, he might have to play 0.2 second ahead so his contribution arrives at the singers at the right time.[1]  And he may hear the choir 0.5 sec after what he plays and what he plays at a different time too!  Incredible.

But where then should I sit in the church to get the best combined timing!??

The nave or choir is usually best.  But in a very large cathedrals like Yorkminister, there are many places where the sound is strange.  The cathedra (bishops chair) is good in Yorkminster.  8)

 

[mulletchuck]

When doing punch-ins, you need to switch the monitoring source from PB to direct.

uh, no you don't...  At least not in Logic.  Punching In with Logic is simply specifying where the track will start/stop recording.  you're still monitoring thru the CueMix/Maestro so you don't have to do anything different to the project/app configuration.    No audio input monitoring (except for effects) in Logic ever takes place, be it punching in or recording. 

 

[abbey road d enfer]

When doing punch-ins, you need to switch the monitoring source from PB to direct.

uh, no you don't...  At least not in Logic.  Punching In with Logic is simply specifying where the track will start/stop recording.  you're still monitoring thru the CueMix/Maestro so you don't have to do anything different to the project/app configuration.    No audio input monitoring (except for effects) in Logic ever takes place, be it punching in or recording.

When you do that in Logic, or almost any other DAW, you monitor the delayed input (conversion times + buffer time), so it is not latency-free. If you want latency-free monitoring, it has to be switched in the analog domain, which is sometimes cumbersome.

 

[mulletchuck]

Like I said, monitoring happens directly thru Maestro/CueMix.  logic isn't doing the monitoring.  Maestro Latency is ~0.5ms.  I had to ask Apogee directly for that spec, cuz they don't list it anywhere.    I asked them "what is the Maestro Low-Latency Mixer latency using a symphony I/O running via USB"

 

[Rochey]

Like I said, monitoring happens directly thru Maestro/CueMix.  logic isn't doing the monitoring.  Maestro Latency is ~0.5ms.  I had to ask Apogee directly for that spec, cuz they don't list it anywhere.    I asked them "what is the Maestro Low-Latency Mixer latency using a symphony I/O running via USB"

The likelyhood is that 0.5mS is the converter latency, plus FPGA mixing,plus DAC output path.

/R

 

[abbey road d enfer]

Like I said, monitoring happens directly thru Maestro/CueMix.  logic isn't doing the monitoring.  Maestro Latency is ~0.5ms.  I had to ask Apogee directly for that spec, cuz they don't list it anywhere.    I asked them "what is the Maestro Low-Latency Mixer latency using a symphony I/O running via USB"

You don't understand what I said. Monitoring in a DAW has latency due to buffers. Monitoring without latency can be done only in analog. Almost latency-free monitoring can be done with the help of a dedicated hardware, such as your MOTU soundcard.
You don't monitor in Logic, you monitor in a DSP app named Cuemix (although Logic issues commands to Cuemix). If you had a card without DSP, or a card with DSP but without the app, you would be monitoring with Logic and you would experience latency.

 

[mulletchuck]

lol

I don't need to have any DAW software running to monitor the audio inputs of my interface.  That's what I was trying to say.    I don't know about the symphony, but i know the MOTU i had would act as a stand-alone mixer/monitor station.  Completely disconnected from the computer.   

Logic doesn't talk to CueMix, but it does talk to the MOTU driver.  CueMix is not the GUI for the MOTU Driver, but it does allow you to assign names to the channels.  The only thing Logic "takes" from cuemix are the user-defined channel names.  CueMix allows you to define the names of the channels (which are reflected in the driver), and in Logic, you can specify if these channel names are taken from CoreAudio's naming, or from the driver's naming of the channels.  That's it as far as Logic talking to CueMix or Maestro, tho. 

 

[abbey road d enfer]

OK, I never pretended to be a specialist of Logic and MOTU, but now, when you do a punch-in, who (or what) does the "tape/direct" switching? Do you have to go into Cuemix and switch the relevant tracks from PB to input? Pretty cumbersome, isn't it ? In PTHD, the switching is done by the DSP's on the cards, which receive a command from PT indicating which tracks are in Record. I assumed any integrated hardware + DAW would act the same...

 

[mulletchuck]

Nope.  Like I said.  You can disable software monitoring in logic preferences, and use only your hardware monitoring.  Again, "Punch In" only specifies where recording will start and stop for a particular track.    It has nothing to do with monitoring.  You don't have to deal with that cumbersome procedure that you're talking about when you don't have a DAW that requires integrated hardware.

For me, this is how I work:

Logic Software monitoring is turned off.  in Maestro, whatever channel my mic is plugged into is unmuted in Maestro's Mixer page.  I simply hit record in logic, and whoever is recording on that mic hears the signal from maestro, not Logic, at ~0.5ms delay. 

However, if I were to have logic's Software Monitoring turned ON, I would hear two sets of the audio input, one from Maestro with almost-zero latency and then logic's monitoring, which is tied to buffer size.   

There are tons of hardware interfaces out there now which support zero-latency.  MOTU, Apogee, Metric Halo, Alesis iO26, probably RME.  mAudio did when I used their Delta 66 card and breakout box back in 2007.  All of these boxes have built-in audio pass-thru from hardware inputs to hardware outputs.  you don't need a DAW to listen to their inputs at near-0 latency. 

 

[abbey road d enfer]

Nope.  Like I said.  You can disable software monitoring in logic preferences, and use only your hardware monitoring.  Again, "Punch In" only specifies where recording will start and stop for a particular track.    It has nothing to do with monitoring.  You don't have to deal with that cumbersome procedure that you're talking about when you don't have a DAW that requires integrated hardware.

For me, this is how I work:

Logic Software monitoring is turned off.  in Maestro, whatever channel my mic is plugged into is unmuted in Maestro's Mixer page.  I simply hit record in logic, and whoever is recording on that mic hears the signal from maestro, not Logic, at ~0.5ms delay. 

However, if I were to have logic's Software Monitoring turned ON, I would hear two sets of the audio input, one from Maestro with almost-zero latency and then logic's monitoring, which is tied to buffer size.   

There are tons of hardware interfaces out there now which support zero-latency.  MOTU, Apogee, Metric Halo, Alesis iO26, probably RME.  mAudio did when I used their Delta 66 card and breakout box back in 2007.  All of these boxes have built-in audio pass-thru from hardware inputs to hardware outputs.  you don't need a DAW to listen to their inputs at near-0 latency.

All this is perfectly understood.
Meantime I've checked the MOTU website; what I see is that Cuemix is not capable of providing "tape machine type" monitoring; your live tracks are always on in the cue/monitor sends (as well as the the PB, which comes from the DAW and thus is switched off when Record is active, or switches to input with buffer latency - it's an option in Samplitude, I guess it is also in Logic). Most of my clients (me too) want this "tape machine type" monitoring, where they hear themselves only when in Record, not before, not after.
Some hardware+DAW combination offer this possibility, by having the DAW issuing a command that tells the DSP switcher to switch from PB to input (I know PT does, Tascam's X48 was meant to do it, but I think it never worked satisfactorily).

 

[0dbfs]

Part of the asio spec allows the dsp/HW mixer to be controlled by the daw. Steinberg refer to it as direct monitoring. In addition to what abbey is describing it also controls panning and levels of the dsp/HW mixer. That is if the HW supports that functionality. You don't have to use this capability. You can use HW monitoring only or SW monitoring only. In An integrated system like pt they make the HW and SW so they can work out those comms/ctrl details between HW and SW.

Ever check out ssl on a pcie card? Looks pretty cool. I wonder which daws can control that? How about uad's mixer on a card/box?

When I used cubendo/motu, the so-called direct monitoring never worked as advertised so there was no auto input switching or headphone input signal mixing from the daw using this asio mixer in a chip control protocol. It was cumbersome to use the ctrl app (cue mix). But possible. logic probably has this too and it may work with certain hardware.

Pthd does integrate this nicely which makes working large tracking sessions much better. Of course you can also run plugs on the pthd mixer on a chip. Which do also incur various latencies depending on the plug.

Analog console monitoring has some of the same but different issues. Obviously latency is not one of the problems but switching (auto) between input and and playback is an issue.

Tape machines have auto/manual input switching which is a standard and it makes recording better in most cases. You have always had the ability to send to headphones from the input side or the return side. Return side provides the auto input switching which is generally expected.

Anyway, cubase had config prefs which allowed the record/input behavior to work with any of the monitoring methods.

Does logic provide the same flexibility wrt options and perhaps direct monitoring style mixer-on-chip control via asio ctrl specs?

Cheers,
jb

 

[mulletchuck]

Most of my clients (me too) want this "tape machine type" monitoring, where they hear themselves only when in Record, not before, not after.

ah, gotcha.  That's how logic's Punch-In works, if you have software monitoring turned on.  but, it's not zero latency, as explained earlier. 

I don't know the answer to your question, 0dbfs.  Logic is mac only, so it doesn't deal with ASIO.  it deals with CoreAudio, which doesn't have anything to do with controlling hardware devices.  it merely reads packets from the input stream based on your selected channel. 

This is the best way I can explain how audio drivers work on mac.  it's a super simple input/output stream-based device that you can use to route audio between applications.  For example, I can play back a session in Logic, send the session output to SoundFlower, Open Skype, set skype's audio input to SoundFlower 1-2, and whoever i'm skyping with will hear the output of Logic at full resolution, as opposed to listening to the sound of Logic playing thru my speakers, and being picked up by the built-in mic on my machine.  If i want them to hear what i'm saying, I simply input-monitor an audio track in Logic with a mic connected  and set the track's output to SoundFlower.    In fact, all I do is create an Aggregate device in logic that comprises of my interface, the machine's built-in inputs, and soundflower.  So, I can send the built-in mic signal, logic's output stream, and anything plugged into my interface's inputs as a stereo stream to another application's audio input.    It's pretty interesting.  you'll notice that soundflower's buffer is fixed at 512samples.

  http://code.google.com/p/soundflower/source/browse/tags/release-1.4.2/Source/SoundflowerEngine.cpp

 

[Autophase]

not sure if i have fully understood your needs here, but cubase has an incredibly flexible sub mix function called "Studio Mixer"

 

[skipwave]

For a stripped-down, lightweight mobile multi-tracking rig, I decided to try this cheap and cheerful "More Me" solution: http://www.sweetwater.com/store/detail/MA400/

I'm sure the amp won't have enough poop for the loud drummers' cans, so a Gyraf phones-amp will handle that.

The monitor input will be fed by a Samson SM10 line mixer, which wisely incorporates a couple of microphone preamps that I'll use for talk/listen. The mixer will be fed by a Focusrite Octopre's line outs, which function as direct-thru from its line-ins while it's A/D provides lightpipe to the laptop Protools rig.

All this because I couldn't find a suitable rackmount line mixer with enough aux sends to feed a Furman HR box. Not to mention I hate slinging CAT5.

 

[ricardo]

Not to mention I hate slinging CAT5.

Can you explain that?  Are you using CAT5 for audio ?  Is that UTPs or STPs?

 

[abbey road d enfer]

Not to mention I hate slinging CAT5.

Can you explain that?  Are you using CAT5 for audio ?  Is that UTPs or STPs?

The Furman HR cue system uses two CAT5 cables to transport 6 channels of audio (4 mono + 1 stereo) and DC supply. I use 60ft of non-shielded non-twisted pair cable without problems.

EDIT: I use cables similar to those:
http://www.amazon.com/Cables-Unlimited-UTP-1800-50W-UltraFlat-Patch/dp/B0007QQJX4
They are twisted-pair indeed.
I was worried the section would limit the power capability of the system, but at the moment, I've been able to run up to six stations without any problem.

 

[0dbfs]

Cat5 works great for audio. Analog and digital. Terminates easily to Xlr, trs, and rj45. It's cheap to abandon in temp setups and robust enough for permenant setups. Once you have some Xlr dongles you can get whatever you need pretty much anywhere like home depot. For monitoring stations you can get 0v, V+, L, and R on one cable that fans out on each end.

Avioms and such are convenient too and once you've got some cat5 trunks terminated into patch panels it's pretty easy to put audio, kvm, Internet,Aviom,  whatever on a cable.

Not adequate for mic audio.

Cheers,
Jb

Cheers,
jb

 

[skipwave]

Avioms and such are convenient too and once you've got some cat5 trunks terminated into patch panels it's pretty easy to put audio, kvm, Internet,Aviom,  whatever on a cable.

That's the key right there. Taking the time during construction to bring the CAT5 out to well-placed patch panels is absolutely critical to making it a quick and effective solution during sessions. The environment I used it in did not, and consequently those CAT5 pairs were always daisy chained between the satellite boxes making a web of trip-lines around the live room. Then the CAT5 gets stepped on enough to become intermittent.

Poor implementation is why I've spent more time than I care to admit troubleshooting cue systems based around the Furman and Aviom systems (on client session time). They are usually setup in a way that makes it impossible to patch around . I don't fault the manufacturers or the technology, it's user error.

Does any manufacturer provide CAT5 in a snake to a stage box? That'd make it a more workable solution for a mobile setup.

Which ever solution you choose, it's hard to overstate how important being able to quickly give the client a good sounding headphone mix is to having a successful session.

 

[ricardo]

Does any manufacturer provide CAT5 in a snake to a stage box? That'd make it a more workable solution for a mobile setup.

http://www.etslan.com/pdfdocs/Audio/PA200_Series_InstaSnake.pdfShielded CAT5, 5E or 6.  Simple & effective