The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
Clocking multiple interfaces - Lavry Presonus Alesis
Old 1 week ago
  #1
Gear Nut
 

Thread Starter
Clocking multiple interfaces - Lavry Presonus Alesis

Hi

I've only really used single interfaces in the past but now I'm combining the following:


Presonus lightpipe ADAT
Lavry AD11
Lavry DA10
Alesis HD24XR
Motu 896 or Behringer ADA etc for 8 more outputs
*EDIT - forgot! Mac Pro digital output


So I wanted to ask opinions on clocking. Get an external clock? Use one of these as master?

much appreciated

Last edited by AKJM; 1 week ago at 04:45 PM..
Old 1 week ago
  #2
Gear Maniac
 
WKWizard's Avatar
I would use one of the Lavry units. They are well regarded and should have the best clocking on your list.
Old 1 week ago
  #3
Whichever way you do it, some of the units will be compromised - so do it the way that you think sounds best.

Normal practice would be to treat the best AD as master when tracking, and the best DA when mixing ITB. If you're capturing back to the same system (ie your DAW you're using both AD and DA so again go with what you think gets the best results.

I'd only get an external clock for features.
Old 1 week ago
  #4
Lives for gear
 

The bigger problem you have trying to use all that together is that you only can use one main interface with the computer at a time and the Presonus unit just has ADAT I/O and the Lavry units I/O AES or Spdif. If you want to keep the Lavry's then dump the Presonus and pick up a used RME Raydat if you have a PCIe slot to use in your computer. That will give you 32 channels of ADAT I/O and at the same time 2 channels of Spdif via RCA breakout cable to the Lavry's. The RME's Totalmix software mixer will make it easy to route in and out 34 channels from the various converters (2 Lavry units / HD24XR / Behringer). Rule of thumb is that your best AD converter should be Master clock and other units slaves. You will be able to clock the devices then either from embedded clocking within the ADAT and Spdif signals or via separate wordclock cables. If going the wordclock route skip daisy chaining and instead use "T" connectors. Threads here and on the net will inform you about setup with "T" connectors.
Old 1 week ago
  #5
Gear Nut
 

Thread Starter
Quote:
Originally Posted by Bassmankr View Post
The bigger problem you have trying to use all that together is that you only can use one main interface with the computer at a time and the Presonus unit just has ADAT I/O and the Lavry units I/O AES or Spdif. If you want to keep the Lavry's then dump the Presonus and pick up a used RME Raydat if you have a PCIe slot to use in your computer. That will give you 32 channels of ADAT I/O and at the same time 2 channels of Spdif via RCA breakout cable to the Lavry's. The RME's Totalmix software mixer will make it easy to route in and out 34 channels from the various converters (2 Lavry units / HD24XR / Behringer). Rule of thumb is that your best AD converter should be Master clock and other units slaves. You will be able to clock the devices then either from embedded clocking within the ADAT and Spdif signals or via separate wordclock cables. If going the wordclock route skip daisy chaining and instead use "T" connectors. Threads here and on the net will inform you about setup with "T" connectors.
yeah I intend to get a raydat ... just waiting for a well priced used one to come up.

Multiple interfaces is ok with the osx aggregate device thing - I'm actually also going to be using the mac pro digital output as well!

I think I'll go with the lavry as master until I get a raydat, then re-evaluate
Old 1 week ago
  #6
Gear Nut
 

Thread Starter
Quote:
Originally Posted by psycho_monkey View Post
Whichever way you do it, some of the units will be compromised - so do it the way that you think sounds best.

Normal practice would be to treat the best AD as master when tracking, and the best DA when mixing ITB. If you're capturing back to the same system (ie your DAW you're using both AD and DA so again go with what you think gets the best results.

I'd only get an external clock for features.
I missed out the mac pro too

I'm using the presonus to mix OTB on a console - then back into lavry - then back out other lavry for alternate monitor signal after software plugins.

So...

firewire - presonus - alesis hd24xr - console - lavry AD11 - logic plugins - mac pro digital out - lavry DA10

As other poster said - a raydat would streamline a lot.

I guess for now I'll use a lavry - I'm guessing the AD11 will have superior clocking over older DA10...?
Old 1 week ago
  #7
Gear Nut
 

Thread Starter
Quote:
Originally Posted by Bassmankr View Post
The bigger problem you have trying to use all that together is that you only can use one main interface with the computer at a time and the Presonus unit just has ADAT I/O and the Lavry units I/O AES or Spdif. If you want to keep the Lavry's then dump the Presonus and pick up a used RME Raydat if you have a PCIe slot to use in your computer. That will give you 32 channels of ADAT I/O and at the same time 2 channels of Spdif via RCA breakout cable to the Lavry's. The RME's Totalmix software mixer will make it easy to route in and out 34 channels from the various converters (2 Lavry units / HD24XR / Behringer). Rule of thumb is that your best AD converter should be Master clock and other units slaves. You will be able to clock the devices then either from embedded clocking within the ADAT and Spdif signals or via separate wordclock cables. If going the wordclock route skip daisy chaining and instead use "T" connectors. Threads here and on the net will inform you about setup with "T" connectors.
yeah i have the t connectors and cables - if I understand correctly, you mean avoid using the word clock IN and OUT on devices - and instead just use the INs chaining with the t connectors..?
Old 1 week ago
  #8
Gear Nut
 

Thread Starter
Quote:
Originally Posted by psycho_monkey View Post
Whichever way you do it, some of the units will be compromised - so do it the way that you think sounds best.

Normal practice would be to treat the best AD as master when tracking, and the best DA when mixing ITB. If you're capturing back to the same system (ie your DAW you're using both AD and DA so again go with what you think gets the best results.

I'd only get an external clock for features.
sorry for multiple replies - I don't post much - not used to it!

I was under the impression the clocking was to avoid really obvious pops and clicks etc - I wanted to know so I can setup and hopefully not rewire everything later.... but is there actually perceptible sound character difference?
Old 1 week ago
  #9
Lives for gear
 

Yes that is what I meant, avoid routing wordclock through devices to the next device in a daisy chain fashion and instead use "T" connectors and route just to the device's wordclock in.

Different clocking setups can result in different sonics. Given it's not that hard to experiment with different clock routing, use your ears. It's a rule of thumb that your best A to D converter serve as Master, not writen in stone for every setup.
Old 1 week ago
  #10
Lives for gear
I agree with this advice. Whichever solution you choose, you should not use daisy chaining.

When confronted with a similar problem a few years back I bought a master clock and used it, but none of my converters were high end units so the master clock was probably a better choice.

Quote:
Originally Posted by Bassmankr View Post
Yes that is what I meant, avoid routing wordclock through devices to the next device in a daisy chain fashion and instead use "T" connectors and route just to the device's wordclock in.

Different clocking setups can result in different sonics. Given it's not that hard to experiment with different clock routing, use your ears. It's a rule of thumb that your best A to D converter serve as Master, not writen in stone for every setup.
Old 1 week ago
  #11
Gear Nut
 

Thread Starter
ok well that opened a can of worms!

It was pretty tricky to get everything to a point where I could just switch clocks while playing a track... what with settings on the units - Apple Audio/Midi aggregate device - and also logic itself. The BNC connector thing is so simple when you know how - but I was terminating an output rather than a "T" on the input... even though that's what you guys explained!

Anyway - I'm pretty surprised clocking has always flown under my radar considering the sound difference. And looking back this is all really easy to do - just a little confusing to understand at forst.
The Alesis is easily the worst - just sounds limited bandwidth in comparison - lacks full lows and sheeny highs. Right now I rate the presonus and lavry equally with a touch more openness and bass on the lavry - but I need to do more testing. My bias is with the lavry purely based on cost etc and so far it rings true.

THANKS EVERYONE - I'm very happy!
Old 1 week ago
  #12
Lives for gear
Quote:
Originally Posted by Bassmankr View Post
The bigger problem you have trying to use all that together is that you only can use one main interface with the computer at a time
True with a PC but he has a MAC where aggregation of multiple devices is common. However, that's usually with only 2 devices and once you attempt 3 or 4 or more.....well, good luck. So yeah, clocking syncing comes back into the picture.
Old 1 week ago
  #13
Quote:
Originally Posted by Alndln View Post
True with a PC but he has a MAC where aggregation of multiple devices is common. However, that's usually with only 2 devices and once you attempt 3 or 4 or more.....well, good luck. So yeah, clocking syncing comes back into the picture.
The aggregate driver is a "get me through" solution. You shouldn't base a system around it.
Old 1 week ago
  #14
Gear Nut
 

Thread Starter
Quote:
Originally Posted by psycho_monkey View Post
The aggregate driver is a "get me through" solution. You shouldn't base a system around it.
Please explain... it seems totally stable
Old 1 week ago
  #15
Quote:
Originally Posted by AKJM View Post
Please explain... it seems totally stable
It's just putting another layer of code in the system. Your throughout latency will be worse. You'll lose the option of low latency monitoring through your interface's cue mixer. Delay compensation for hardware inserts may become inexact. You'll lose tech support. It's more of a "get you through" solution.

In some circumstances it'll do the job, I just wouldn't base a system around it.
Old 1 week ago
  #16
Lives for gear
 

Quote:
Originally Posted by AKJM View Post
Please explain... it seems totally stable
I agree with PM.. it's typically kind of band-aid solution, because it will bring several inherent difficulties
- it's another software driver layer between the interface driver and DAW
- when different device drivers are used, not all of them necessarily use the same internal buffering scheme for audio streaming to/from the device, so this "encapsulating" layer has to adapt to that and adjust DAW IOs according to slowest device in the group.
- requires, all devices in aggregation group are externally synchronized (WC, digital I/O), otherwise there has to be one device set as group clock master and other devices has to use either continuous re-sampling or discarding of samples to overcome that

All that should be handled in transparent manner, however first two points are sometimes very problematic with regards to good low latency performance, efficiency during audio streaming and sometimes also reliability.
That's why, I'd think about that in more some serious scenarios, only when same underlying driver will be used. Some interface vendors officially supports Core Audio aggregation (RME, Metric Halo.. or AFAIK also Presonus with their FW interfaces), when it's used with several their devices, because they already know internal buffering scheme and performance characteristics of the common driver, know it responds well to buffer size changes from aggregation layer and tested that during development.

What's IME most problematic, is when someone wants to use completely different drivers in aggregation group (say USB device, old Firewire interface and built-in Mac IO).

The third point is logical, because there's not possible to synchronize device from its driver to align it with other stream, so all devices has to be already in-sync before aggregation.

In your particular case, it should be likely possible to sync everything via WC (except of Lavry DA, which gets synced on its digital input).. so from that point, it is probably manageable.

However, as I've said.. it's still rather band-aid solution, when lot of I/Os are needed from some old hardware to record some event with relaxed DAW buffers.. but similarly like PM mentioned, I wouldn't count with software aggregation as with primary method of working or put some significant money there, when buying new gear.
DAWs generally performs best and most reliable, when "talking" just to single device.

YMMV..

Michal
Old 1 week ago
  #17
Yeah...just like that!

Well said

Quote:
Originally Posted by msmucr View Post
I agree with PM.. it's typically kind of band-aid solution, because it will bring several inherent difficulties
- it's another software driver layer between the interface driver and DAW
- when different device drivers are used, not all of them necessarily use the same internal buffering scheme for audio streaming to/from the device, so this "encapsulating" layer has to adapt to that and adjust DAW IOs according to slowest device in the group.
- requires, all devices in aggregation group are externally synchronized (WC, digital I/O), otherwise there has to be one device set as group clock master and other devices has to use either continuous re-sampling or discarding of samples to overcome that

All that should be handled in transparent manner, however first two points are sometimes very problematic with regards to good low latency performance, efficiency during audio streaming and sometimes also reliability.
That's why, I'd think about that in more some serious scenarios, only when same underlying driver will be used. Some interface vendors officially supports Core Audio aggregation (RME, Metric Halo.. or AFAIK also Presonus with their FW interfaces), when it's used with several their devices, because they already know internal buffering scheme and performance characteristics of the common driver, know it responds well to buffer size changes from aggregation layer and tested that during development.

What's IME most problematic, is when someone wants to use completely different drivers in aggregation group (say USB device, old Firewire interface and built-in Mac IO).

The third point is logical, because there's not possible to synchronize device from its driver to align it with other stream, so all devices has to be already in-sync before aggregation.

In your particular case, it should be likely possible to sync everything via WC (except of Lavry DA, which gets synced on its digital input).. so from that point, it is probably manageable.

However, as I've said.. it's still rather band-aid solution, when lot of I/Os are needed from some old hardware to record some event with relaxed DAW buffers.. but similarly like PM mentioned, I wouldn't count with software aggregation as with primary method of working or put some significant money there, when buying new gear.
DAWs generally performs best and most reliable, when "talking" just to single device.

YMMV..

Michal
Old 1 week ago
  #18
Gear Nut
 

Thread Starter
Quote:
Originally Posted by psycho_monkey View Post
It's just putting another layer of code in the system. Your throughout latency will be worse. You'll lose the option of low latency monitoring through your interface's cue mixer. Delay compensation for hardware inserts may become inexact. You'll lose tech support. It's more of a "get you through" solution.

In some circumstances it'll do the job, I just wouldn't base a system around it.

Ok I see what you mean. I think I'll be ok for now. Latency isn't an issue as I generally track direct onto the Alesis - or I switch off the aggregate and use just the larvy AD. I monitor either direct out the desk or on a loopback through conversion so i'm ok without interface mixers. Hardware inserts are all outboard for me.... so yeah for now I think I'll avoid the issues you mention. But also I will eventually grab an RME ray dat and use just that for all the digital connections.thanks for your comments
Old 1 week ago
  #19
Gear Nut
 

Thread Starter
Quote:
Originally Posted by msmucr View Post
I agree with PM.. it's typically kind of band-aid solution, because it will bring several inherent difficulties
- it's another software driver layer between the interface driver and DAW
- when different device drivers are used, not all of them necessarily use the same internal buffering scheme for audio streaming to/from the device, so this "encapsulating" layer has to adapt to that and adjust DAW IOs according to slowest device in the group.
- requires, all devices in aggregation group are externally synchronized (WC, digital I/O), otherwise there has to be one device set as group clock master and other devices has to use either continuous re-sampling or discarding of samples to overcome that

All that should be handled in transparent manner, however first two points are sometimes very problematic with regards to good low latency performance, efficiency during audio streaming and sometimes also reliability.
That's why, I'd think about that in more some serious scenarios, only when same underlying driver will be used. Some interface vendors officially supports Core Audio aggregation (RME, Metric Halo.. or AFAIK also Presonus with their FW interfaces), when it's used with several their devices, because they already know internal buffering scheme and performance characteristics of the common driver, know it responds well to buffer size changes from aggregation layer and tested that during development.

What's IME most problematic, is when someone wants to use completely different drivers in aggregation group (say USB device, old Firewire interface and built-in Mac IO).

The third point is logical, because there's not possible to synchronize device from its driver to align it with other stream, so all devices has to be already in-sync before aggregation.

In your particular case, it should be likely possible to sync everything via WC (except of Lavry DA, which gets synced on its digital input).. so from that point, it is probably manageable.

However, as I've said.. it's still rather band-aid solution, when lot of I/Os are needed from some old hardware to record some event with relaxed DAW buffers.. but similarly like PM mentioned, I wouldn't count with software aggregation as with primary method of working or put some significant money there, when buying new gear.
DAWs generally performs best and most reliable, when "talking" just to single device.

YMMV..

Michal

thanks for the detail!

as I said above - ideally I will get a ray dat eventually and not bother with this.

You are correct though - it is very manageable with my setup. |I'm very happy for now
Old 1 week ago
  #20
Gear Nut
 

Thread Starter
Ok... so just for anyone else who tries this -

eventually I have run into problems. On a different project which had maybe more complex plugins I have found that it seems the clock "drifts"... after some time there are LOTS of pops and crackles. I have to reset the audio driver to fix it. It may just be the project because it has plugins I have never used before... but yeah I need a raydat!
Old 1 week ago
  #21
Gear Nut
 

Thread Starter
Quote:
Originally Posted by psycho_monkey View Post
It's just putting another layer of code in the system. Your throughout latency will be worse. You'll lose the option of low latency monitoring through your interface's cue mixer. Delay compensation for hardware inserts may become inexact. You'll lose tech support. It's more of a "get you through" solution.

In some circumstances it'll do the job, I just wouldn't base a system around it.
Hi again - I wanted to ask - you mentioned about hardware inserts - well I'm having trouble googling the correct terms for doing the opposite - i.e. using a vst plugin as an insert on my outboard mixing console... do you know if this is possible in a stable way? (probably not with my clocking situation!)
Old 1 week ago
  #22
Lives for gear
 

Exact procedure depends on used DAW software, but generally you'll simply add another track, set required physical input and outputs, which are patched to your desk, add some software inserts and enable input monitoring.

That should be pretty straightforward, only thing, which is up to your decision is, whether you like to record that wet track as well, when you start DAW transport. In most cases I'd say, it's not really necessary. Some DAWs has independent controls for input monitoring and rec arm (Cubase, Pro Tools), some has that linked together in default (like Reaper, where you need to have track armed to monitor live input, but you can disable recording via right click) in Pro Tools you can also create dedicated AUX track, which doesn't record audio at all.

Roundtrip latency of such effect channel depends on used working buffer size of course, but if you'll use it for some reverb or delays, it isn't necessarily a problem, because it will be perceived just as normal pre-delay.

Michal
Old 1 week ago
  #23
Gear Nut
 

Thread Starter
Quote:
Originally Posted by msmucr View Post
Exact procedure depends on used DAW software, but generally you'll simply add another track, set required physical input and outputs, which are patched to your desk, add some software inserts and enable input monitoring.

That should be pretty straightforward, only thing, which is up to your decision is, whether you like to record that wet track as well, when you start DAW transport. In most cases I'd say, it's not really necessary. Some DAWs has independent controls for input monitoring and rec arm (Cubase, Pro Tools), some has that linked together in default (like Reaper, where you need to have track armed to monitor live input, but you can disable recording via right click) in Pro Tools you can also create dedicated AUX track, which doesn't record audio at all.

Roundtrip latency of such effect channel depends on used working buffer size of course, but if you'll use it for some reverb or delays, it isn't necessarily a problem, because it will be perceived just as normal pre-delay.

Michal
thank you so much. This makes sense. Again - so simple -I suppose I thought I might need extra software

I'm in Logic - I'll have a play around. But I was hoping to use some buss compression so I will probably have to wait for better latency with a new Raydat
Old 1 week ago
  #24
Lives for gear
 

Quote:
Originally Posted by AKJM View Post
I'm in Logic - I'll have a play around. But I was hoping to use some buss compression so I will probably have to wait for better latency with a new Raydat
Hmm.. it can be bit tricky with bus compression.
If we're talking about final master bus compression, I'd say it's doable.. and you have several options, for example you can create one additional track, where you set input coming from master bus of your desk, add desired compressor there and set its output to internal bus in DAW. This track will be used just for that processing. Then you will add stereo recording track, where you'll set previously used internal bus as its input. There shouldn't be problem with that, if you would be listening to output of this track (there would be delay, but akin to playback head at tape machine).

If you mean using group bus compression just for some elements in a mix.. (like. drums, bg vocals etc. Not to mention some parallel stuff), roundtrip latency can be problematic regardless of its length. Because processed feed will be always delayed from the rest of mix before summing, it might cause some undesirable issues.

But you can certainly test it also without RayDat.

Michal
Old 1 week ago
  #25
Quote:
Originally Posted by AKJM View Post
thank you so much. This makes sense. Again - so simple -I suppose I thought I might need extra software

I'm in Logic - I'll have a play around. But I was hoping to use some buss compression so I will probably have to wait for better latency with a new Raydat
If you're doing this, I'd really only use it for send FX, where with an eg reverb or delay a few samples latency won't really matter.

As soon as you start using it for insert FX, or going in and out for busses, your latency is going to build up - and it's not the sort of thing that you can compensate for.

If you're planning on your mixbuss processing being ITB, I'd stay ITB and use hardware inserts. You're just asking for trouble otherwise. And forget trying to write automation in real time!
Old 1 week ago
  #26
Lives for gear
Quote:
Originally Posted by psycho_monkey View Post
The aggregate driver is a "get me through" solution. You shouldn't base a system around it.
I didn't say he should. I was just pointing out that it was possible on Mac unlike PC.
Old 6 days ago
  #27
Gear Nut
 

Thread Starter
Quote:
Originally Posted by psycho_monkey View Post
If you're doing this, I'd really only use it for send FX, where with an eg reverb or delay a few samples latency won't really matter.

As soon as you start using it for insert FX, or going in and out for busses, your latency is going to build up - and it's not the sort of thing that you can compensate for.

If you're planning on your mixbuss processing being ITB, I'd stay ITB and use hardware inserts. You're just asking for trouble otherwise. And forget trying to write automation in real time!
yep! it's not workable for tightness at all! It was just a thought really - I prefer hardware anyway but the extra choice would be nice.
Old 6 days ago
  #28
Gear Nut
 

Thread Starter
update to clocking....

The only true stable way I have found longterm and including my Lavry loopback monitoring....is to have the Presonus as master and connect hd24xr and lavry AD to it by word clock cable.

I could only do limited comparisons but there didn't seem much difference between Lavry and presonus clocking (assuming it was actually setup how it seemed - as there was a lot of trouble locking the clocks). The hd24xr clcoking was also not bad at all but I felt a difference in the lows that I personally didn't like - and I couldn't get the lavry stable this way more importantly.

Still want a raydat though!
Old 6 days ago
  #29
Quote:
Originally Posted by Alndln View Post
I didn't say he should. I was just pointing out that it was possible on Mac unlike PC.
Wasn't specifically referring to you, it was just a general statement.

Quote:
Originally Posted by AKJM View Post
yep! it's not workable for tightness at all! It was just a thought really - I prefer hardware anyway but the extra choice would be nice.
I used to do this sometimes when mixing on a desk, if the room didn't have amazing outboard. But mainly for delays.
Old 5 days ago
  #30
Gear Nut
 

Thread Starter
Quote:
Originally Posted by psycho_monkey View Post

I used to do this sometimes when mixing on a desk, if the room didn't have amazing outboard. But mainly for delays.

it's also a good route for sudden accidental john lennon phasey drums with compression misalignment!
Topic:
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump