Qlab redundancy - Dante Switching - Yamaha

[Peter Rice]

Hi All

This may seem a little off topic as it's really a Yamaha Question but one I think other Qlab users will have come up against

I’m trying to set up a redundant Qlab system using 2 playback machines, a Yamaha QL/CL5, Internal Dante and 2 x MY16-AUD.

The show we are running has 32 Qlab channels so I have set up a system whereby I can have the Main Machine on the internal Dante and the back up machine spread across the 2 Slots. In order to minimise channel count across both machines I would like to create an input patch change so I only use 32 channel inputs, which I can assign to a User Define key for mid show switching

Inputs for the 2 machines work on their relevant input sources and can bee seen on the channel inputs, but as the CL/QL series don’t have Input Patch libraries like former Yamaha desks it makes it tricky to switch between the two.

The only other way I thought was to use the Dante Input Patch Library to point to the relevant input stream, save each in it’s own library and select the change of memory on a user defined key. 

When the Dante input is patched to Internal Dante all works well, but when the Dante inputs is patched to the Slot inputs I don’t get input (even though if the slot is patched directly to the channel input all works if fine)

I’ve spoken to Yamaha support engineer today (in Germany) and explained what I was trying to do and he said he didn’t think it could be achieve in this way. He proposed doing something more convoluted using ‘Focus’ to change input patches in unused memories, but I feel like others have achieved this in a more simplistic manner.

It’s not possible to both machines just sat across the 64 Dante Inputs 33-64 are used for other input sources

All Qlab, Dante and Yamaha Software and Firmware are at the latest current versions

Any help/thought/experience on steps I am missing or alternative ways to achieve this appreciated

Thanks

Peter

[Dominic Bilkey]

Hi Pete

Make the desk the Dante controller and then Recall a Dante patch by assigning that command to a UDK. 

Used it on panto and works like a charm!

Leaves you with a "blip" in the sound but very workable. 

D

--
Contact support anytime: sup...@figure53.com
Follow Figure 53 on Twitter: http://twitter.com/Figure53
---
You received this message because you are subscribed to the Google Groups "QLab" group.
To unsubscribe from this group and stop receiving emails from it, send an email to qlab+uns...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/qlab/6af4b589-d992-456b-8446-8894342a632c%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Sam Kusnetz]

Hi Peter

I agree with Dominic; make the Yahama the Dante controller and use Dante input libraries to switch between the main and backup Mac.

Furthermore, if you add a Dante-compatible network switch between the Macs and the console, then you can ditch the input cards entirely.

I’ve got this exact setup running on two off-Broadway shows right now and it works a charm.

Best

Sam

—

Sam Kusnetz | Figure 53

s...@figure53.com

To view this discussion on the web visit https://groups.google.com/d/msgid/qlab/D1C99331-D6DC-4519-B59C-F0BBCD1E486E%40gmail.com.

[jerem...@jjlee.com]

Something i’ve done before is utilized the INSERT patch point for redundancy.  Patch the main QLab input to the channel itself, and patch the redundant one to the Return of the Insert on the channel.  Normally the Insert would remain disengaged, but when you need to switch over, engage the Insert.

You can link multiple channels on the CL5 in regards to Insert On/Off, and assign one of the channels to a user defined key.

As long as you don’t need the insert point for anything else, it works great!  I’ve also done this for backup mics in a pinch.  You just have to make sure where in the signal flow you’re putting the insert point on that channel.

-- 

Jeremy Lee

    Sound Designer - USA 829

        http://www.jjlee.com

[Peter Rice]

Hi Dom

Yes this is how I'm implementing it at the moment. The issue is when the Dante Input Patch switches to the Slot Inputs the audio doesn't patch through. I can see the slot inputs, they patch and they assign correctly, but I have no audio. When it is patched to the Internal Dante it works fine...

If I patch the slot input directly to a channel input there is audio so I know signal is getting to that point...

[Peter Rice]

Hi Sam

 

Furthermore, if you add a Dante-compatible network switch between the Macs and the console, then you can ditch the input cards entirely.

Yes I have a Dante-compatible network switch but want to use both the internal Dante and the Slot cards to offer independent redundancy options and maximise my Dante inputs

[Peter Rice]

Hi Jeremy

Thanks, I had this option in mind but just wanted to see if there were a 'clean' way to just recall a different patch and leave all else in the signal chain untempered. 

Sure I can find a way to make it work as I want but if not I have this and a couple of other option in my back pocket...

Thanks

[Sam Kusnetz]

Furthermore, if you add a Dante-compatible network switch between the Macs and the console, then you can ditch the input cards entirely.

Yes I have a Dante-compatible network switch but want to use both the internal Dante and the Slot cards to offer independent redundancy options and maximise my Dante inputs

Hi Peter

Could you tell me more about this, please? To the best of my knowledge, the number of Dante inputs has nothing to do with whether cards are installed.

Also, I don’t see how there’s any added redundancy offered by using a card slot rather than the console’s built in Dante primary and secondary connections.

If you know more about this than I do, I’d love to learn!

Sam

[Dominic Bilkey]

Hi Pete

Why does it need to switch to the slot?

A Dante repatch repatches machine A's stream to the onboard. No slot required. 

This is a Dante repatch to the onboard not a Yamaha input patch recall. 

Or am I missing something?

To view this discussion on the web visit https://groups.google.com/d/msgid/qlab/71ddd643-9067-4e37-8de5-d370fbc9d421%40googlegroups.com.

[Dan Miele]

Could you tell me more about this, please? To the best of my knowledge, the number of Dante inputs has nothing to do with whether cards are installed.

Hey Sam,

If I'm remembering correctly, adding a MY16-AUD does increase the number of Dante channels the console can utilize. I believe this has to do with the actual Audinate chips used in the devices. Both the CL/QL5 probably use a Dante Brooklyn chip, which is 64x64. The MY16-AUD most likely uses a Dante Broadway chip, which is 16x16.

By adding MY16-AUD cards, you're actually increasing the number of Dante receivers available. The console then treats these inputs as just another audio source, as it would with any other type of MY card. This is why you could use the MY16-AUD with an LS9 or M7, which do not natively support Dante.

Cheers,

Dan

[Sam Kusnetz]

Dan

Aha! That’s very interesting to learn. Thank you.

Sam

—

Sam Kusnetz | Figure 53

s...@figure53.com

--
Contact support anytime: sup...@figure53.com
Follow Figure 53 on Twitter: http://twitter.com/Figure53
---
You received this message because you are subscribed to the Google Groups "QLab" group.
To unsubscribe from this group and stop receiving emails from it, send an email to qlab+uns...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/qlab/c932d363-0d9c-432d-bba4-283ffa960932%40googlegroups.com.

[jerem...@jjlee.com]

This is the cleanest that I know of.  If you want to get down and dirty, you could reverse engineer the SysEx spec and send MIDI to the console to specifically change patches. But that would likely be coming from QLab. And wouldn’t work if the machine went south...

-- 

Jeremy Lee

    Sound Designer - USA 829

        http://www.jjlee.com

To view this discussion on the web visit https://groups.google.com/d/msgid/qlab/5e3f6915-4b91-4411-a8b8-8d9ad296b456%40googlegroups.com.

[jerem...@jjlee.com]

Absolutely true. Even though the CL5 has 72 mono channels and 8 stereo channels, if you want access to more than 64 channels via Dante, it requires cards.  I personally think it was rather foolish of Yamaha not to simply have 2x Brooklyn cards internally.  And that it can’t do 96k is kind of bizarre, given that the RIO boxes are fully capable.  Maybe they’ll make a CLPro with a Dante HC chip, lots of channels, and the ability to have a high sample rate.

-- 

Jeremy Lee

    Sound Designer - USA 829

        http://www.jjlee.com

To view this discussion on the web visit https://groups.google.com/d/msgid/qlab/CAOOK5AZ-4%2BtOATLfKOemffDf1hz0aZaMMjLS9Rufb27gnJcrWg%40mail.gmail.com.

[mackerr]

I would suggest the solution Yamaha recommended. Create 2 scenes outside of the range of scenes you are using for your show. Set the focus of those 2 scenes to input patch only, and remove input patch from the focus of all other scenes. Set up your primary Dante patch and save it in your first patch scene, and do the same for your bu patch in your second patch scene. Assign them to UDK buttons and you can switch between input patches without affecting any other scenes or parameters.

Using the slot inputs also backs up your patch cable and the Dante chip as well.

AFAIK you cannot use the insert ins from Dante on a CL, although I think you can on a QL.

Mac

[John Huntington]

On Thursday, May 18, 2017 at 12:06:03 PM UTC-4, Peter Rice wrote:

... The issue is when the Dante Input Patch switches to the Slot Inputs the audio doesn't patch through. I can see the slot inputs, they patch and they assign correctly, but I have no audio. When it is patched to the Internal Dante it works fine...

If I patch the slot input directly to a channel input there is audio so I know signal is getting to that point...

Stupid question--do you have the cards set to "slave to external word clock"?   That would explain the first issue you list, but the second is baffling.

John 

www.controlgeek.net

[John Huntington]

On Thursday, May 18, 2017 at 9:31:31 PM UTC-4, mackerr wrote:

Using the slot inputs also backs up your patch cable and the Dante chip as well.

I agree and one of the reasons I like this approach.  (although so far I haven't seen an Audinate chip go bad).

Also Ethernet is supposed to be transformer isolated so this setup should even give you good electrical isolation.

John

 

[Peter Rice]

If I'm remembering correctly, adding a MY16-AUD does increase the number of Dante channels the console can utilize. I believe this has to do with the actual Audinate chips used in the devices. Both the CL/QL5 probably use a Dante Brooklyn chip, which is 64x64. The MY16-AUD most likely uses a Dante Broadway chip, which is 16x16.

By adding MY16-AUD cards, you're actually increasing the number of Dante receivers available. The console then treats these inputs as just another audio source, as it would with any other type of MY card. This is why you could use the MY16-AUD with an LS9 or M7, which do not natively support Dante.

Cheers,

Dan

Exactly this, it then gives me another 32 Dante inputs. But as some pointed out if I do the patching in Dante Input Patch on the QL/CL for DVS this is less relevant as I can just point this Dante sockets to the other machine in an alternative patch.

Stupid question--do you have the cards set to "slave to external word clock"?   That would explain the first issue you list, but the second is baffling

Everything is set to clock from Dante with the desk as the Preferred Master

Hi Pete

Why does it need to switch to the slot?

A Dante repatch repatches machine A's stream to the onboard. No slot required. 

This is a Dante repatch to the onboard not a Yamaha input patch recall. 

Or am I missing something?

I've not explained the thinking behind my goal clearly enough. Qlab is flowing on DVS thought the Primary Dante route. If that fails and we go to Dante Secondary but we lose all our DVS/Qlab. By running the second machine directly into the Slots (via a small switch to access both card simultaneously) I can bypass the regular Dante network and provide redundancy not only against Main machine failure, but also against Dante Primary failure. DVS is great, as is Dante Primary/Secondary switching but the two don't go together as far as I understand...

Thanks for your responses, it seems there isn't a simple way to achieve what I want so I will look down the route using 'Focus' on the desk for now. If only the input patch library still existed (and I never thought I'd say that!)