ProSoundWeb Community

Sound Reinforcement - Forums for Live Sound Professionals - Your Displayed Name Must Be Your Real Full Name To Post In The Live Sound Forums => LAB: The Classic Live Audio Board => Topic started by: Markus Bangerter on September 08, 2016, 09:47:26 AM

Title: what Digital Mixers to recommend
Post by: Markus Bangerter on September 08, 2016, 09:47:26 AM
hi all

I'm in process of replacing my two Midas analogue mixing consoles with a Digital Console.

today, i'm running already some Dante-enablend equipment (Lab.Gruppen PLMs, Lake Controller LM26s, Focusrite RedNets units, Dante Virtual Soundcard, ...)
therefore most of my racks are equipped with networking gear (Cisco Switches). Usually I'm running two independent fiber-optical based networks between Stage and FOH; Dante Dual redundancy

i had already a look at this consoles:
- Midas PRO 1
- Midas M32
- Yamaha CL / QL
- Digico S21
- Soundcraft Si Impact

regarding Dante, Yamaha CL/QL provides the best integration.
Midas uses AES50 connecting to I/O interfaces, so i would not be able to benefit from the existing network infrastructure.

what Console would you recommend ?
Title: Re: what Digial Mixers to recommend
Post by: Erik Jerde on September 08, 2016, 09:59:56 AM
It would probably be helpful to know what you're replacing and what you actually need the consoles to do.  From your post it reads like you just need them to do Dante.  I doubt that's really the case.
Title: Re: what Digial Mixers to recommend
Post by: Yoel Farkas on September 08, 2016, 10:03:32 AM
hi all

I'm in process of replacing my two Midas analogue mixing consoles with a Digital Console.

today, i'm running already some Dante-enablend equipment (Lab.Gruppen PLMs, Lake Controller LM26s, Focusrite RedNets units, Dante Virtual Soundcard, ...)
therefore most of my racks are equipped with networking gear (Cisco Switches). Usually I'm running two independent fiber-optical based networks between Stage and FOH; Dante Dual redundancy

i had already a look at this consoles:
- Midas PRO 1
- Midas M32
- Yamaha CL / QL
- Digico S21
- Soundcraft Si Impact

regarding Dante, Yamaha CL/QL provides the best integration.
Midas uses AES50 connecting to I/O interfaces, so i would not be able to benefit from the existing network infrastructure.

what Console would you recommend ?
This topic has been discussed on many threads here. the answer is as usual "it depends". depends how many channels I/O you need. do you like to have a tablet to control wireless? 
they are all good consoles. all have benefits over the other.
the Si impact will be the most affordable.

I would vote for the Digico S21 (or the new S31) this is my preference. but all are good.
Title: Re: what Digial Mixers to recommend
Post by: Markus Bangerter on September 08, 2016, 10:34:42 AM
This topic has been discussed on many threads here. the answer is as usual "it depends". depends how many channels I/O you need. do you like to have a tablet to control wireless? 
they are all good consoles. all have benefits over the other.
the Si impact will be the most affordable.

I would vote for the Digico S21 (or the new S31) this is my preference. but all are good.

thanks for feedback so far. Yes i know "it depends"

i would like to replace a Midas Verona 320

when focusing on using my existing (Ethernet-based) network; then i have to go the full Dante route which means Yamaha CL/QL series (Stageboxes running on Dante)
All other vendors do have either MADI or AES50 which is not compatible with Ethernet as transport protocol.

When focusing on running Dante just for I/O on consoles, then nearly every Console listed above can be used.

so my primary question is; Would you recommend Dante for interconnecting Stageboxes and Mixers ?
The advantage of using Midas PRO 1 would be 96kHz sampling-rate, while all others use "only" 48kHz
Title: Re: what Digial Mixers to recommend
Post by: Ray Aberle on September 08, 2016, 10:39:34 AM
The advantage of using Midas PRO 1 would be 96kHz sampling-rate, while all others use "only" 48kHz
... can you hear the difference between 96kHz and 48kHz??

So the other "usual" question that comes up with a gear purchase opinion-- how much are you budgeting for this? M32 will run you $5k; a CL5 will be north of $20K.

And then, how many physical faders do you need on the control surface? CL1 will give you 8+8 and a CL5 will give you 24+8 -- both are good boards, but do you need to see a larger number of faders at any one time?

Also, are you just buying this for yourself as an awesome new toy, or are there other people's concerns (as in, a venue, or dealing with touring artist riders and visiting band engineers) that you have to take into consideration?

-Ray
Title: Re: what Digial Mixers to recommend
Post by: RaySoly on September 08, 2016, 10:41:53 AM
thanks for feedback so far. Yes i know "it depends"

i would like to replace a Midas Verona 320

when focusing on using my existing (Ethernet-based) network; then i have to go the full Dante route which means Yamaha CL/QL series (Stageboxes running on Dante)
All other vendors do have either MADI or AES50 which is not compatible with Ethernet as transport protocol.

When focusing on running Dante just for I/O on consoles, then nearly every Console listed above can be used.

so my primary question is; Would you recommend Dante for interconnecting Stageboxes and Mixers ?
The advantage of using Midas PRO 1 would be 96kHz sampling-rate, while all others use "only" 48kHz


A&H Dlive http://www.allen-heath.com/dlive-intro/, runs at 96khz and offers dante on the console and mixracks….
Title: Re: what Digial Mixers to recommend
Post by: Yoel Farkas on September 08, 2016, 10:54:07 AM
so my primary question is; Would you recommend Dante for interconnecting Stageboxes and Mixers ?
The advantage of using Midas PRO 1 would be 96kHz sampling-rate, while all others use "only" 48kHz
The Digico S21 runs at 96Khz
All of the consoles have options for Dante cards.
Title: Re: what Digial Mixers to recommend
Post by: Erik Jerde on September 08, 2016, 10:55:09 AM
In the original post your said you want to replace two mixers.  If that's the case do you want two new mixers that can gain share with one stage rack?
Title: Re: what Digial Mixers to recommend
Post by: Markus Bangerter on September 08, 2016, 11:31:12 AM
In the original post your said you want to replace two mixers.  If that's the case do you want two new mixers that can gain share with one stage rack?

That could be a use case, but rather rare.
Title: Re: what Digial Mixers to recommend
Post by: Tim McCulloch on September 08, 2016, 02:47:58 PM
IMHO the hardest desk to set up and operate is the Midas Pro series, followed by DigiCo SD; I haven't used an S21 so can't offer any insight.  Once you're up and going they aren't particularly difficult but the challenge is getting to that point.

If you're spending someone else's money (arts centre, public auditorium, university) I suggest Yamaha CL5/Rio.

If you're spending your own funds I suggest either the M32 or S21.
Title: Re: what Digial Mixers to recommend
Post by: Kevin McDonough on September 08, 2016, 05:34:28 PM
IMHO the hardest desk to set up and operate is the Midas Pro series, followed by DigiCo SD; I haven't used an S21 so can't offer any insight.  Once you're up and going they aren't particularly difficult but the challenge is getting to that point.

If you're spending someone else's money (arts centre, public auditorium, university) I suggest Yamaha CL5/Rio.

If you're spending your own funds I suggest either the M32 or S21.


hey

yeah, lots of questions as with everyone else lol. Mainly, what do your customers ask for most?  Do you have to satisfy riders? or is it for your own use.

I own a pro1 and use bigger pro2s as well as expression and digico SD range regularly.

I actually find the Pro series quickest to set up now that I have my head around how it works and have a good workflow, the block patching is easiest once you get used to the screen they use.

An expression is a good "analogue replacement" desk, it's very straightforward and fairly cost effective, but nowhere near as flexible as midas pro, or even M32.

Sound wise Digico tips the scale and just pips everyone to the post, very closely followed by the Midas pros.

I don't think focusing on the protocol that connects the stage box to the desk to much is helpful. If it's Dante then its Dante, or if it's something else then it may need a little change in wiring but will be suitable for the purpose.

However if you do need dante to feed your final outputs from the desk into your PA or monitor system, then a lot of desks come with card slots that a dante card can be fitted into if needed to give you outputs into the PA.

And while there may only be a minimal audio difference between 48k and 96k, running at 96k also has the advantage that it significantly reduces latency through the desk, which is an important consideration especially when used for monitors or in-ears.


Really, at the end of the day it's a question only you can answer.  You need to consider what works in your market, what price points you can get each one for (including stage boxes and infrastructure) and so what will generate a good return, what will be in demand in your area, what you can cross rent and hire more of if needed (or what fills a gap in your local market if you can already hire other consoles at good rates when needed), and combine that with the technical considerations and advice given here, and see what will be best to grow your company and be viable.


k




Title: Re: what Digial Mixers to recommend
Post by: Don T. Williams on September 08, 2016, 06:04:34 PM
+1 for Kevin's post.  Once I understood the patching on the Pro Series I also understood that there are a lot of things you can do with it's patching that can't be done with lessor consoles.

There are a lot of good choices.  The best choice is what works best for you!
Title: Re: what Digial Mixers to recommend
Post by: Markus Bangerter on September 08, 2016, 06:09:54 PM
ok, thanks so far.

let's assume if will go the Yamaha route and start with a CL/QL. i have already some Focusrite RedNet4 and MP8R units.
is there any issue with using them as "Stage Racks" instead of the RiOs ?
Title: Re: what Digial Mixers to recommend
Post by: Tim McCulloch on September 08, 2016, 06:17:09 PM
ok, thanks so far.

let's assume if will go the Yamaha route and start with a CL/QL. i have already some Focusrite RedNet4 and MP8R units.
is there any issue with using them as "Stage Racks" instead of the RiOs ?

IIRC, you cannot directly control preamp gain from the CL consoles; you have to use RedNet Control on the computer of your choice.

If your interest in Dante is about trying to save money on i/o, I think you're looking at this all backwards.
Title: Re: what Digial Mixers to recommend
Post by: Markus Bangerter on September 08, 2016, 06:23:04 PM
IIRC, you cannot directly control preamp gain from the CL consoles; you have to use RedNet Control on the computer of your choice.

If your interest in Dante is about trying to save money on i/o, I think you're looking at this all backwards.

according to Focusrite:
- Remote-controllable Gain, Phantom Power and High-Pass Filter from Yamaha CL and QL Series Consoles.

 
Title: Re: what Digial Mixers to recommend
Post by: brian maddox on September 08, 2016, 08:06:39 PM
according to Focusrite:
- Remote-controllable Gain, Phantom Power and High-Pass Filter from Yamaha CL and QL Series Consoles.

 

Yeah, they recently added that capability.  Haven't had a chance to try it in the field, but I've been told it's totally seamless.
Title: Re: what Digial Mixers to recommend
Post by: Tim McCulloch on September 08, 2016, 09:03:55 PM
according to Focusrite:
- Remote-controllable Gain, Phantom Power and High-Pass Filter from Yamaha CL and QL Series Consoles.

 

Thanks, Brian M...  I didn't think control from the console was possible but I haven't had a reason to look recently.

Markus, whether or not it's a good idea is up to you, your uses and your clients.  If it were me, I'd use all Yamaha i/o so that visiting engineer's saved show files will work without issue.  If it is only you and your staff that operate the mixers then choice of i/o probably won't matter much.

How much RedNet stuff do you already have?  How many inputs/outputs do you need to accommodate?
Title: Re: what Digial Mixers to recommend
Post by: Markus Bangerter on September 09, 2016, 05:39:58 AM
Thanks, Brian M...  I didn't think control from the console was possible but I haven't had a reason to look recently.

Markus, whether or not it's a good idea is up to you, your uses and your clients.  If it were me, I'd use all Yamaha i/o so that visiting engineer's saved show files will work without issue.  If it is only you and your staff that operate the mixers then choice of i/o probably won't matter much.

How much RedNet stuff do you already have?  How many inputs/outputs do you need to accommodate?
Have 3 x RedNet 4, 2 x RedNet 2, 1 x RedNet 1. I was thinking of using the RedNet 4s a "Stage Racks" in a initial phase. maybe later on add a Yamaha RiO to the setup. This would help to protect investments already done to RedNet.

The RedNet units are primarily used to transport audio over long distances (25-300m) on larger areas; Dante Break-In/Break-Out to feed some smaller speakers.

Title: Re: what Digital Mixers to recommend
Post by: Kyle Van Sandt on September 09, 2016, 12:57:33 PM
Still having issues here with your venue programming.  At this point it does not really matter what your interconnect is, there are ways to make any console work with anything.  You might have some hoops to jump through but there ya go.  I would also really try to stay away from non-standard setups like non-matched pre-amps.  That will send visiting engineers running for a rental desk quickly. 

I went completely Midas in my 3 spaces.  Pro2's in the big hall, M32 in the smaller hall, and X32 out our outdoor stage.  We primary produce live music, so we wanted the Midas sound... and got it.  Pro2's are great desks but not the best tool for all shows.  For corporate work the QL is the choice.  For Theatre Digico is the desk. 


So... once again... content?  budget? guest engineers?  riders?  input size?.........
Title: Re: what Digital Mixers to recommend
Post by: Aisle 6 on September 12, 2016, 06:31:15 PM
As everyone has already mentioned, there are some real questions that need to be answered to direct you more accurately.

content?  budget? guest engineers?  riders?  input size?.........

However, I have become a huge fan of the new Allen & Heath d-live and think that it ticks the best part of all or most boxes.

* Easiest console to use by a huge margin.
* Sounds as good, if not better than almost anything currently on the market. (I know, bring the hate, but use one first for a few shows and then get back to me)
* 96k operation (I believe that this hales with the sound not just because the sample rate being higher should sound better, but the higher sample rate lowers the latencies, which is one of the current issues with digital boards)
* Great implementation of scene management that is easily up to theatre spec.
* Most engineers could easily operate one without ever seeing one before.
* Offline editor looks and feels almost exactly like the surface and can also be an online controller for the system
* I am beginning to see it on riders
* It is as easily at home in a theatre install as on a touring rock show.
* It is built tough and can take a serious knock
* It is super affordable
* It has a huge I/O count

...it sounds incredible. Sonically I would put it head to head with a Digico or a Midas.


Title: Re: what Digital Mixers to recommend
Post by: Cailen Waddell on September 12, 2016, 07:07:01 PM
IMHO the hardest desk to set up and operate is the Midas Pro series, followed by DigiCo SD; I haven't used an S21 so can't offer any insight.  Once you're up and going they aren't particularly difficult but the challenge is getting to that point.

If you're spending someone else's money (arts centre, public auditorium, university) I suggest Yamaha CL5/Rio.

If you're spending your own funds I suggest either the M32 or S21.

Ditto

Our municipal theaters are all Yamaha CL, most of audio guys have their own personal M/X 32 variants for their own small shows...

We are really happy with the CL series consoles.


Sent from my iPhone using Tapatalk
Title: Re: what Digital Mixers to recommend
Post by: Scott Holtzman on September 13, 2016, 03:02:37 AM
As everyone has already mentioned, there are some real questions that need to be answered to direct you more accurately.

content?  budget? guest engineers?  riders?  input size?.........

However, I have become a huge fan of the new Allen & Heath d-live and think that it ticks the best part of all or most boxes.

* Easiest console to use by a huge margin.
* Sounds as good, if not better than almost anything currently on the market. (I know, bring the hate, but use one first for a few shows and then get back to me)
* 96k operation (I believe that this hales with the sound not just because the sample rate being higher should sound better, but the higher sample rate lowers the latencies, which is one of the current issues with digital boards)
* Great implementation of scene management that is easily up to theatre spec.
* Most engineers could easily operate one without ever seeing one before.
* Offline editor looks and feels almost exactly like the surface and can also be an online controller for the system
* I am beginning to see it on riders
* It is as easily at home in a theatre install as on a touring rock show.
* It is built tough and can take a serious knock
* It is super affordable
* It has a huge I/O count

...it sounds incredible. Sonically I would put it head to head with a Digico or a Midas.




Quick swerve Scott, do most of the statements also hold true for the GLD?  I like the Digico Sx1 but I think it is too big a step for tech's that spend most of their time behind x/m32's. 

The GLD looks more approachable and I hope the ease of use and quality travel downward in the line though IIRC the GLD is 48Khz only.

Title: Re: what Digial Mixers to recommend
Post by: Speed Daemon on September 13, 2016, 10:52:48 PM
And while there may only be a minimal audio difference between 48k and 96k, running at 96k also has the advantage that it significantly reduces latency through the desk, which is an important consideration especially when used for monitors or in-ears.
Hi Kevin, could you expound on the latency thing for me?  It seems counterintuitive, in my mind more bits means more latency. What am I missing?

At my age, I'm not going to hear any difference in HF content that a 96kHz sample rate provides, so that's moot for me anyway. I was told that sound reinforcement consoles don't use the highest sample rates and bit width because it's easier to move less data. I surmise that the more costly boards with microprocessor controlled FPGA logic are more immune to this, but at a greater cost.

One of the reasons I went with Midas was that the AES50 interface used point-to-point topology and layer 2 Ethernet signaling. While less flexible, it looked like a simple way to eliminate and latency with store and forward switching that the layer 3 and 4 protocols must contend with. I can still buy a card to interoperate with MADI and Dante IIRC.

I looked at the S21, but couldn't find much technical detail about communication with the stage boxes. Can anyone point me to a good reference on this? I may be using DiGiCo mixers in the future, and want to come prepared.  Thanks!
Title: Re: what Digial Mixers to recommend
Post by: Tim McCulloch on September 13, 2016, 11:12:44 PM
Hi Kevin, could you expound on the latency thing for me?  It seems counterintuitive, in my mind more bits means more latency. What am I missing?

At my age, I'm not going to hear any difference in HF content that a 96kHz sample rate provides, so that's moot for me anyway. I was told that sound reinforcement consoles don't use the highest sample rates and bit width because it's easier to move less data. I surmise that the more costly boards with microprocessor controlled FPGA logic are more immune to this, but at a greater cost.

One of the reasons I went with Midas was that the AES50 interface used point-to-point topology and layer 2 Ethernet signaling. While less flexible, it looked like a simple way to eliminate and latency with store and forward switching that the layer 3 and 4 protocols must contend with. I can still buy a card to interoperate with MADI and Dante IIRC.

I looked at the S21, but couldn't find much technical detail about communication with the stage boxes. Can anyone point me to a good reference on this? I may be using DiGiCo mixers in the future, and want to come prepared.  Thanks!

Faster processing, Bill.  A clock cycle at 96kHz is half the time of a 48kHz clock cycle.  Since nothing i/o related can happen faster than the clock speed (bit in, bit out) a higher sampling rate means less latency.
Title: Re: what Digital Mixers to recommend
Post by: George Dougherty on September 13, 2016, 11:13:26 PM
Hi Kevin, could you expound on the latency thing for me?  It seems counterintuitive, in my mind more bits means more latency. What am I missing?
Sample-rate, not bits.  You're talking slices of time with sample rate.  Bit-rate is how many bits make up one of those slices of time.  48,000 slices vs 96,000 slices = half the time accorded to an individual sample though they're typically addressed in sample buffers that give the actual minimum latency through a system.  Handoffs and additional processing steps in a system will also add their own latency.
Title: Re: what Digital Mixers to recommend
Post by: Doug.Jane on September 13, 2016, 11:21:51 PM
Sample-rate, not bits.  You're talking slices of time with sample rate.  Bit-rate is how many bits make up one of those slices of time.  48,000 slices vs 96,000 slices = half the time accorded to an individual sample though they're typically addressed in sample buffers that give the actual minimum latency through a system.  Handoffs and additional processing steps in a system will also add their own latency.
Latency is not really related to sampling rate. Latency is the time taken for the signal to be processed by the DSP. The DSP takes a number of sample rate cycles to do its processing, but its speed is determined by its clock, which is running much faster than the sample rate, typically 10s of MHz.
Title: Re: what Digial Mixers to recommend
Post by: Speed Daemon on September 13, 2016, 11:53:07 PM
Faster processing, Bill.  A clock cycle at 96kHz is half the time of a 48kHz clock cycle.  Since nothing i/o related can happen faster than the clock speed (bit in, bit out) a higher sampling rate means less latency.
Isn't the #1 producer of latency the CPU and other logic that processes the signal?  I mean, sure you can have a device that gives a 1 channel output that's synchronous with one input.  But once you start summing multiple inputs, add effects etc., the time it takes to do the calculations is going to make the output a lot later than the input.  That's the nature of the beast.
Title: Re: what Digital Mixers to recommend
Post by: Speed Daemon on September 14, 2016, 12:08:28 AM
Sample-rate, not bits.  You're talking slices of time with sample rate.  Bit-rate is how many bits make up one of those slices of time.  48,000 slices vs 96,000 slices = half the time accorded to an individual sample though they're typically addressed in sample buffers that give the actual minimum latency through a system.  Handoffs and additional processing steps in a system will also add their own latency.
That's my understanding, that the processing is what drives latency.  We can put garbage in and out at any given rate as long as it's not processed.  But start doing math on it, and that's where latency happens, is it not?
Title: Re: what Digital Mixers to recommend
Post by: Scott Holtzman on September 14, 2016, 12:09:20 AM
Isn't the #1 producer of latency the CPU and other logic that processes the signal?  I mean, sure you can have a device that gives a 1 channel output that's synchronous with one input.  But once you start summing multiple inputs, add effects etc., the time it takes to do the calculations is going to make the output a lot later than the input.  That's the nature of the beast.


I can't imagine anything is done in a floating point CPU.  The CPU coordinates activity between the DSP's. 

Digital summing is mathematically very low overhead.  I have not looked at a DSP chip in years but just looking at what you can do with a mini-dsp project box I can imagine that any type of filter or dynamic processing is all a matter of accessing libraries.

I would not be surprised if some day you see console software abstracted from the I/O later.  Imagine picking out your dynamics from Waves or another vendor, your favorite I/O box and a human interface of your choice.  To me this is the future.

The interesting thing about the Digico is some folks have written about it's interface being a little latent. Perhaps the presentation layer is underpowered even though it has some awesome DSP horsepower.

The there is the Rivage, looking at it online is like porn for sound guys, it is simply elegant, Yamaha really knocked it out of the park.

I want to again restate that the first vendor who makes the input board some type of removable box so the console can be traditional or use it with a digital connection will have a huge win.  Behringer/Midas, Digico...you listening?

Title: Re: what Digital Mixers to recommend
Post by: Speed Daemon on September 14, 2016, 12:12:41 AM
Latency is not really related to sampling rate. Latency is the time taken for the signal to be processed by the DSP. The DSP takes a number of sample rate cycles to do its processing, but its speed is determined by its clock, which is running much faster than the sample rate, typically 10s of MHz.
That's my understanding too.  So it only follows that if you give a DSP more samples per second, that increases the load on the DSP, not decreases. Thanks!
Title: Re: what Digital Mixers to recommend
Post by: Speed Daemon on September 14, 2016, 12:59:48 AM
I can't imagine anything is done in a floating point CPU.  The CPU coordinates activity between the DSP's.
I don't know where floating-point came into the discussion, but I'm old enough to remember when the FPU was an option to most computer CPUs. If you're talking about DSP as a noun and not a verb, the DSP unit is just a specialized CPU. A DSP unit may be able to do more complex functions per clock cycle (with the internal bus clock not being the same as the sample rate), but it's still a computer of sorts. The thing that interests me is that some vendors are hard-coding certain algorithms into FPGAs, thus eliminating CPU/DSP cycles.

Quote
I would not be surprised if some day you see console software abstracted from the I/O later.  Imagine picking out your dynamics from Waves or another vendor, your favorite I/O box and a human interface of your choice.  To me this is the future.
That seems to be the state of the art now. I read an article on ProSoundWeb about a FOH mixer who's using a pair of WAVES servers to be a virtual effects rack.  After working in IT for some 20 years, I'm not sure how I feel about that from a "points of failure" standpoint, but people are doing it.  There's even a video HOWTO for doing it with an M32.

Quote
The interesting thing about the Digico is some folks have written about it's interface being a little latent. Perhaps the presentation layer is underpowered even though it has some awesome DSP horsepower.
That's the nature of the beast. I have some TiVo DVRs that can record 6 channels simultaneously, but the thing I really miss is that instant feedback when I press a button. They could have done a number of things to make the UI the priority, but if it means a dropped frame or a price tag I couldn't afford, then what's the point. I hear the S21 has a bit of UI lag, but look at the price.

Quote
I want to again restate that the first vendor who makes the input board some type of removable box so the console can be traditional or use it with a digital connection will have a huge win.  Behringer/Midas, Digico...you listening?
Ever since I tried (and failed) to build a programmable effects box out of a 68000-based computer, I've been having my own pie-in-the-sky dreams about how a digital PA should work. I figured that doing all the processing on-stage and making the "mixer" just a control surface was ideal. I'm seeing that happen now. But when it comes to modularity I'm disappointed. I would like to be able to choose my own mic preamp/input module, and be able to use it with any processing module. And every powered speaker with DSP should take a digital feed. At the very least every brand of digital snake should at least have a card to enable to use it with any other brand of digital mixer. Time will tell.

Title: Re: what Digital Mixers to recommend
Post by: Tim McCulloch on September 14, 2016, 01:09:48 AM
That's my understanding too.  So it only follows that if you give a DSP more samples per second, that increases the load on the DSP, not decreases. Thanks!

The CPU operates at speeds much greater than the sampling rate, as does any off-CPU DSP.  At some point there is an output buffer and final D/A conversion (at the sampling rate).

While we're measuring i/o sampling in kHz, the CPU and DSP are running in the mHz.  DSP can be paralleled to reduce processing latency, but it is generally correct that more processing increases the DSP latency (think:  dynamics, EQ or harmonics processing, from shorter to longer).  Processing creates the *variables* in amount of latency; the amount of latency caused solely by the i/o sample rate is fixed.

Getting stuff in, through and out faster is important mostly for singers using IEMs (the comb filtering resulting from bone conductivity and a digitally mixed IEM signal) and for percussionists and others that are time-sensitive.  Lower total latency helps both types of users; for singers it raises the frequency at which the combing begins.
Title: Re: what Digital Mixers to recommend
Post by: Scott Holtzman on September 14, 2016, 01:41:50 AM
I don't know where floating-point came into the discussion, but I'm old enough to remember when the FPU was an option to most computer CPUs. If you're talking about DSP as a noun and not a verb, the DSP unit is just a specialized CPU. A DSP unit may be able to do more complex functions per clock cycle (with the internal bus clock not being the same as the sample rate), but it's still a computer of sorts. The thing that interests me is that some vendors are hard-coding certain algorithms into FPGAs, thus eliminating CPU/DSP cycles.


When I use the term Floating Point I am talking about doing processing in code on a general purpose CPU that would make use of the floating point instructions but in reality I am referring to the CPU intensive processing in the way recording software works on a PC.  For all intents and purposes the FPU is part of the ALU (arithmatic logic unit) in any modern processing design.

An FPGA is functioning as a dedicated DSP when it is programmed for that function. 

As Tim pointed out insertion of dynamics, eq and effects all have an effect on latency.

For the record I am old enough to have started on 8 bit CPU's and I remember my first audio processing code to decode DTMF in a generic Z-80 simply by sampling the data on a port at regular intervals.  Tone generation same deal.  You used interrupts to service the decoder/generator.  Even on an 8 bit 4Mhz CPU there was still a ton of cycles in between polling the port to take care of other tasks.

Title: Re: what Digital Mixers to recommend
Post by: Speed Daemon on September 14, 2016, 03:10:27 AM
When I use the term Floating Point I am talking about doing processing in code on a general purpose CPU that would make use of the floating point instructions but in reality I am referring to the CPU intensive processing in the way recording software works on a PC.  For all intents and purposes the FPU is part of the ALU (arithmatic logic unit) in any modern processing design.

An FPGA is functioning as a dedicated DSP when it is programmed for that function. 

As Tim pointed out insertion of dynamics, eq and effects all have an effect on latency.

For the record I am old enough to have started on 8 bit CPU's and I remember my first audio processing code to decode DTMF in a generic Z-80 simply by sampling the data on a port at regular intervals.  Tone generation same deal.  You used interrupts to service the decoder/generator.  Even on an 8 bit 4Mhz CPU there was still a ton of cycles in between polling the port to take care of other tasks.
The sad irony is that I can't escape computer science, even in my leisure activities. I'm trying to do this for fun.

One thing that I've noticed about embedded processors in just about every application is that the bean counters always have them using the cheapest crap that they can get away with. That's not a bad thing when you can get so much functionality from a $2000 mixer (I can't remember what my first Soundcraft Series 1 cost, but it had to be more than that). It does make me wonder if this too will pass, or if featuritis will gobble up processing power as it has done elsewhere.
Title: Re: what Digital Mixers to recommend
Post by: George Dougherty on September 15, 2016, 12:39:57 AM
Latency is not really related to sampling rate. Latency is the time taken for the signal to be processed by the DSP. The DSP takes a number of sample rate cycles to do its processing, but its speed is determined by its clock, which is running much faster than the sample rate, typically 10s of MHz.

Actually, unless DSP based hardware works drastically different from computer audio processing, samples are collected into buffer sets rather than individual samples.  The speed of the DSP determines the amount of work that can be done on a sample buffer set, but the size of the set and the number of sets held in memory during processing and routing determines the latency. 
The speed of the DSP will also impact how small that buffer size can be, but it still comes down to a finite set of samples that represent fractions of a second. 96K with modern DSP allows for lower latency times because the fraction of a second being represented is half that of 48K.  You can still make the two equal though.  Sixteen 48K samples are equal to thirty two 96K samples.
Title: Re: what Digital Mixers to recommend
Post by: George Dougherty on September 15, 2016, 12:53:48 AM
That's my understanding, that the processing is what drives latency.  We can put garbage in and out at any given rate as long as it's not processed.  But start doing math on it, and that's where latency happens, is it not?

Any time it's handled and passed off digitally you add latency.  There's latency in the chips that take an input stream and pass it off to another step, such as throwing it onto a network audio stream like DANTE, or sending it out an AES/EBU port.  It's a bit of a combination on the math side though and yes it takes time to do latency and the time you need to complete it will determine how many samples you can work on at one time.  All things being equal though, 96K gives you smaller slices of time to work with and thus the potential for lower latency since in many cases the number of samples dealt with is fixed.  Many pieces of audio hardware have lower latency at higher sample rates because n samples at 96K is half the time of n samples at 48K.
Title: Re: what Digital Mixers to recommend
Post by: George Dougherty on September 15, 2016, 01:03:45 AM
If you want to look at a possible modular type approach, something like AMP might make that a reality.  The mixer logic is handled in a computer and there's now a Dante PCIe based option where you can utilize whatever Preamps you'd like.  Apparently the Yamaha TIO boxes are now usable in channel counts up to 128 (not being limited by the TF console hardware) and AMP will remote control the preamps.

ampmix.net is the home of the project.  Similar concept to SAC but much better implemented.

Title: Re: what Digital Mixers to recommend
Post by: Speed Daemon on September 15, 2016, 01:33:10 AM
Any time it's handled and passed off digitally you add latency.  There's latency in the chips that take an input stream and pass it off to another step, such as throwing it onto a network audio stream like DANTE, or sending it out an AES/EBU port.  It's a bit of a combination on the math side though and yes it takes time to do latency and the time you need to complete it will determine how many samples you can work on at one time.  All things being equal though, 96K gives you smaller slices of time to work with and thus the potential for lower latency since in many cases the number of samples dealt with is fixed.
The problem there is that the number of samples is not fixed, it's doubled compared to 48 kHz.  All things being equal, that means twice the work for the processing engine, not half.

The latency here is not the time between samples, it's the time it takes to perform all of the mathematical functions on all of the inputs, and put it back out. This is one reason why early digital PA consoles don't all operate at 192 kHz by 24 bits, because more input means more work, and more work adds latency and/or cost.

Eventually technology will progress to the point that Hi-Rez sample rates will be commonplace even for live sound mixing, but it will be to brag about having bigger numbers, not some imagined technical benefit. If you're doing a FOH mix and sending your inputs to a recorder, then Hi-Rez is desirable. But if you're just doing a FOH mix, you gain nothing by upping the sample rate above 48 kHz.
Title: Re: what Digital Mixers to recommend
Post by: Jeff Simpson on September 15, 2016, 04:32:37 AM
The problem there is that the number of samples is not fixed, it's doubled compared to 48 kHz.  All things being equal, that means twice the work for the processing engine, not half.

When using dedicated DSP (as opposed to an x86 type platform that I don't know much about), no matter how many channels or what the sampling rate, any particular processing block must complete in one sample period. If this was not the case, the processing block would end up with a bigger and bigger backlog of samples to process, with no chance of ever catching up. You could try to implement some sort of interleaved processing, where two DSPs would alternate taking two sample periods to process every alternate sample. However, since this would involve doubling the number of DSPs, it is much easier to deploy this extra hardware by halving the number of channels each DSP has to deal with, and going back to having everything take one sample period. For example, let's say my DSP can cope with processing 4 input channels at 48 kHz; if I want to move to 96 kHz, the easiest approach is to add another DSP, and have each process only 2 channels.

With this in mind, I have a hard time imagining a system where doubling the sample rate does not halve the time taken by a processing element. My experience is that system latency is dominated by conversion time, and by moving audio between things, be it between DSPs and other chips, or between units connected via a digital snake.
Title: Re: what Digital Mixers to recommend
Post by: Speed Daemon on September 15, 2016, 05:44:47 AM
With this in mind, I have a hard time imagining a system where doubling the sample rate does not halve the time taken by a processing element.
Yes, this seems to be a prevalent misunderstanding, thinking that the output of a digital mixer is isosynchronous with the input, just like with analog. While that would be very nice indeed, there are a variety of technical realities that prevent that from ever happening with serialized pulse code data.

When it comes to sample rates, it's simple: all else being the same, doubling the sample rate does double the rate of "garbage in" and therefore doubles the load to the part that processes the input to produce an output.  Double is always double and never half.  (If we want to get into the finer points of A/D conversion we could have quite an argument, but none of the consoles mentioned aren't doing anything unusual there.)

Yes, you can throw more processing hardware at the doubled bit rate, be it a microcontroller system or hard-wired logic. But the result is that the buyer is paying more for the hardware needed to handle more input. If cost is no object, we all would have the state of the art, of course.  But in the spirit of the original question, I still don't see any real-world benefit in doubling the sample rate for sound reinforcement.

I'd expect the golden eared crowd to claim sonic advantages of higher sample rates, but this is the first time I've ever seen someone say that more samples means less latency. That does not compute. Again, if I'm missing something, please help me out.
Title: Re: what Digital Mixers to recommend
Post by: Aisle 6 on September 15, 2016, 05:56:06 AM
Quick swerve Scott, do most of the statements also hold true for the GLD?  I like the Digico Sx1 but I think it is too big a step for tech's that spend most of their time behind x/m32's. 

The GLD looks more approachable and I hope the ease of use and quality travel downward in the line though IIRC the GLD is 48Khz only.

The GLD is also a great little board and is super flexible and so simple to use...however, it definitely sits high in it's price point sonically. What I mean is that I would choose it over many other boards in that price bracket, but it is no where near the d-live.
I own both and am happy with the GLD and would choose it over an X32 or M32
Title: Re: what Digital Mixers to recommend
Post by: Aisle 6 on September 15, 2016, 06:12:52 AM
Yes, this seems to be a prevalent misunderstanding, thinking that the output of a digital mixer is isosynchronous with the input, just like with analog. While that would be very nice indeed, there are a variety of technical realities that prevent that from ever happening with serialized pulse code data.

When it comes to sample rates, it's simple: all else being the same, doubling the sample rate does double the rate of "garbage in" and therefore doubles the load to the part that processes the input to produce an output.  Double is always double and never half.  (If we want to get into the finer points of A/D conversion we could have quite an argument, but none of the consoles mentioned aren't doing anything unusual there.)

Yes, you can throw more processing hardware at the doubled bit rate, be it a microcontroller system or hard-wired logic. But the result is that the buyer is paying more for the hardware needed to handle more input. If cost is no object, we all would have the state of the art, of course.  But in the spirit of the original question, I still don't see any real-world benefit in doubling the sample rate for sound reinforcement.

I'd expect the golden eared crowd to claim sonic advantages of higher sample rates, but this is the first time I've ever seen someone say that more samples means less latency. That does not compute. Again, if I'm missing something, please help me out.

I am not a technician of the calibre that can explain such things, however I do recall reading several specification with regard to DAW's. Specifically a document that went along with Pro Tools. It stated latencies when recording and these latencies were always halved with the doubling of the sample rate.

All I am trying to say with regard to the low latencies of the d-live, is that I do not buy into the enhanced audio argument of 96K as there are many and varied elements that effect the outcomes of live sound. I am making a statement that the factor of time does have an impact on the result and that a lower latency can only assist with improving this. The d-live has one of, if not the lowest latencies of a live digital mixing console and I attribute my experience with it in part to this fact.
Title: Re: what Digital Mixers to recommend
Post by: Jeff Simpson on September 15, 2016, 07:16:02 AM
Yes, you can throw more processing hardware at the doubled bit rate, be it a microcontroller system or hard-wired logic. But the result is that the buyer is paying more for the hardware needed to handle more input. If cost is no object, we all would have the state of the art, of course.  But in the spirit of the original question, I still don't see any real-world benefit in doubling the sample rate for sound reinforcement.

My point was that the latency is halved for 'processing' tasks, giving an overall latency reduction that could be seen as a real world benefit in some scenarios. I was really intending to counter some posts stating that doubling the sample rate does not affect the latency of processing tasks. Perhaps this was slightly off topic, sorry.

I will of course agree that all else being equal, the cost of the hardware will increase if you double the sample rate.
Title: Re: what Digital Mixers to recommend
Post by: Speed Daemon on September 15, 2016, 08:18:01 AM
I am not a technician of the calibre that can explain such things, however I do recall reading several specification with regard to DAW's. Specifically a document that went along with Pro Tools. It stated latencies when recording and these latencies were always halved with the doubling of the sample rate.

All I am trying to say with regard to the low latencies of the d-live, is that I do not buy into the enhanced audio argument of 96K as there are many and varied elements that effect the outcomes of live sound. I am making a statement that the factor of time does have an impact on the result and that a lower latency can only assist with improving this. The d-live has one of, if not the lowest latencies of a live digital mixing console and I attribute my experience with it in part to this fact.
Hi Scott, I'd really have to see the Pro Tools document to see what it's about. All I can say now is that Pro Tools is computer software for recording and editing, which makes it a different beast from a digital mixer.

The thing that we're talking about is very much an electrical engineering matter, so if you want to fully understand it, you'll need to read some very thick books, I'm afraid. With that understood, I can discuss the basic reasons why the A&H DLive system has an advertised < 0.7 ms latency.

The first thing is that the DLive makes extensive use of FPGA logic. A general purpose computer CPU relies on a number of basic instructions, and does them over and over to do complex calculations, whereas a FPGA has the entire complex calculation programmed into it, so it can perform a complex operation very quickly. The downside is that it can't change algorithms on the fly, as a computer can. As CPU, GPU and DSP chips get faster and faster, they might make the FPGA obsolete, but for now it's the king for specialized tasks like this. So using FPGAs to speed up number crunching is one big factor that gives the DLive (and the S21 IIRC) a performance edge over CPU/DSP-based mixers. An edge that comes at a price.

The other thing about the DLive is that all of the brains reside in the stage box. Some digital mixers have the brain and control surface in the same box (just like an analog mixer), whereas the DLive is a bit like a control surface for a DAW, except instead of a computer, the control surface runs the mixer that's in the stage box. This eliminates all of the potential bottlenecks of a digital snake, as the mix happens right onstage. So you don't need to know the finer points of the ISO/OSI layers and why they're important, the differences between Dante and AES50 etc.

There's no doubt that the DLive system is pretty sweet. I would have looked at it myself, but it's more than I wanted to spend for my own system. I'll definitely keep it in mind if I start making money in this town.  :)

That's not to say that other digital mixers are rubbish, though. You still have some degree of latency, and if you pile on the special effects, that latency can grow. It's worth noting that most loudspeaker management systems are digital, and add some delay (latency) of their own. Same thing with class D amplifiers with DSP built-in. Like it or not, modern live sound involves a lot of time delays at every stage.

I sorta feel bad for the weekend warrior who plays small clubs, because new pure analog gear is starting to get rare. You almost have to have a degree in electrical engineering and/or computer science to really know how the new digital gear works. It's just blind luck that I happened to study all the right things in college, and got jobs that exercised my mind in those areas. I keep a copy of "Principals of Digital Audio" by Ken Pohlmann as a reference. It's not an easy read, but provides a lot of the fundamentals that every digital mixer owner should have some working knowledge of.

Have fun with your DLive! If you ever bring it to Albuquerque, I'd love to see it in action.
Title: Re: what Digital Mixers to recommend
Post by: Scott Helmke on September 15, 2016, 08:50:26 AM
Processing of digital signals is pretty quick. It's conversion from/to analog that adds the most latency in most mixers.

Just look at the specs of an average device - analog in to analog out is at least two milliseconds.  Digital in to digital out is usually less than half a millisecond.
Title: Re: what Digital Mixers to recommend
Post by: Speed Daemon on September 15, 2016, 09:55:33 AM
My point was that the latency is halved for 'processing' tasks, giving an overall latency reduction that could be seen as a real world benefit in some scenarios. I was really intending to counter some posts stating that doubling the sample rate does not affect the latency of processing tasks. Perhaps this was slightly off topic, sorry.

I will of course agree that all else being equal, the cost of the hardware will increase if you double the sample rate.
That's really the bottom line in this context. If you're choosing a mixer or DAW for recording music, you want to have the most resolution that you can afford. Same thing if you have a blank check from a successful musician who wants "the best" when going on tour. And of course the hardware companies need some reason to get their customers to keep buying new products whether they need them or not.

Back in the bad old days of analog to digital conversion, there was a legitimate need for higher sample rates, namely to allow the "brick wall" filter to be moved far away from the audible range. But with modern delta sigma conversion, that problem has been solved. So unless you're recording dog whistles, you can downsample to 48 kHz and no human ear will notice. In the digital domain you can lose the inaudible part of the audio spectrum, and use the extra capacity for more channels.

So if cost is a factor, I'd put I/O sample rates low on my list of priorities. I'd rather spend that money on better FOH speakers, more monitors or whatever it is that I really need. In my particular case I chose a popular mixer design that a lot of people use, so that there's a good chance that another band's FOH mixer can jump right in, and so I can rent my system out to make some extra money. Everyone's situation is different, of course.
Title: Re: what Digital Mixers to recommend
Post by: Kevin McDonough on September 15, 2016, 10:16:16 AM
hey

edit: I had to leave this half typed for a while as some things came up, and then some more replies were added before I actually finished and posted it.

LOL.  Unfortunately I'm not a computer scientist so I'm not totally sure of the exact mechanisms. But upping the sample rate definitely does lower the latency  ;D  It's confirmed by all the desk manufacturers on their spec sheets and things. So it's not in question, just the mechanism of what actually causes it.

What I would guess is happening is that, as people have suggested, it's to do with the shorter time of the samples.

Latency isn't just one thing, it's made of two parts. The first is the AD conversion, which is always the big one  (and DA again at the end), and then after that the second is the calculations and conversions of those samples as gain is added, compression, gates, effects, etc etc. I.e. the processing time.

Some of these steps may be able to work on a single sample by sample basis, and some may need to chunk a couple of samples together into a processing block before it can do the sum. But whether it needs 1 or 2 or 4 or however many samples to work, these will always be double as quick as the same number of samples at 48KHz.

So the first part of the latency, just waiting on the samples coming in, is halved due to doubling the sampling rate.

The thing to remember is the sample rate is time fixed - i.e. a 96khz signal always takes a specific amount of time. 

But the time taken to do the calculations on that sample isn't fixed, so that can be faster or slower depending on the processor. The processing doesn't have to be in real time, or even out of order or different samples/chunks of samples in parallel, as long as it's all stitched together again in the right order afterwards.

So of course as people have pointed out: in order to double the sample rate like this, it is double the data rate, and so much greater hardware is needed to achieve the processing. Faster ram and processors, more DSP chips to run through the calculations etc etc. It'll cost more, but is definitely doable.

While its hard to nail down exactly what causes the differences (preamps will play a part, quality of algorithms and effects etc), it is a consideration that  Digico and Midas are generally agreed to be the best sounding desks around, (closely followed it now seems by D-live), and the fact that all are 96khz desks (or potentially 192Khz for an SD7) could be part of the difference. Either by an audio difference, or just by lower latency making it seem more 'real' and immediate.

So the faster sample rate speeds things up for one part of the latency, and the bigger better processors that would go along with it will speed things up for the other part of the latency.

Of course, what most big desks do now a days is used a fixed latency system to keep everything sample synchronous and stop any phasing issues. For each section of processing, they work out what the maximum processing latency can be, and they delay all the other signals by the same amount. When they're all stitched together again at the auz or master bus, this means they stay in perfect time/phase.  So the desk will have a fixed latency time, and anything that only has minimal processing and isn't actually slowed by that amount will be delayed to match.


Of course, whether all this is worth the money for all the extra processing etc then becomes up to the user. For some it will be, and I'd suggest that for monitors and especially In-Ears it's a big consideration. For others not.


k
Title: Re: what Digital Mixers to recommend
Post by: Speed Daemon on September 15, 2016, 10:23:27 AM
Processing of digital signals is pretty quick. It's conversion from/to analog that adds the most latency in most mixers.

Just look at the specs of an average device - analog in to analog out is at least two milliseconds.  Digital in to digital out is usually less than half a millisecond.
Scott, could you give a specific example? I have to admit that I've never measured the delay in any DAC or ADC. For home use I wish I could find a good DAC with a generous FIFO to eliminate the time base errors (jitter) inherent in spinning disc media.
Title: Re: what Digital Mixers to recommend
Post by: Speed Daemon on September 15, 2016, 10:29:22 AM
LOL.  Unfortunately I'm not a computer scientist so I'm not totally sure of the exact mechanisms. But upping the sample rate definitely does lower the latency  ;D  It's confirmed by all the desk manufacturers on their spec sheets and things. So it's not in question, just the mechanism of what actually causes it.
Kevin, could you point me to an online manual that says this? My own system doesn't have the option to double the sample rate.
Title: Re: what Digital Mixers to recommend
Post by: Josh Millward on September 15, 2016, 11:47:27 AM
Scott, could you give a specific example? I have to admit that I've never measured the delay in any DAC or ADC. For home use I wish I could find a good DAC with a generous FIFO to eliminate the time base errors (jitter) inherent in spinning disc media.

I have measured this before and he is exactly correct.

It takes a certain number of samples (that is to say, a certain amount of time) for the ADC and DAC to do its thing. When you double the clock rate, it can now do its thing twice as fast, ergo it takes half the amount of time. The samples are going through the machine twice as fast.

A number of years ago I was measuring a MediaMatrix NION with Smaart. When I would double the sample rate (from 48kHz to 96kHz) the total latency through the machine was just about exactly halved. I did not repeat the test at 24kHz sample rate because I did not think of it at the time.

The really interesting thing about this though was that I measured both sample rates with no DSP processing and with the NION stuffed full of DSP processing. The variation in throughput latency between no DSP and lots of DSP was below the error range of Smaart. In other words, using the NION, I measured almost exactly the same throughput latency irrespective of how much DSP processing I was doing. That means the actual DSP processing is extremely fast and the ADC and DAC are extremely slow in comparison as they totally swamp the measurement.

It is also important to note that when switching from 48kHz to 96kHz, because you are doing the work twice as fast and you have a fixed amount of DSP resources available, you can now only do half as much work.
Title: Re: what Digital Mixers to recommend
Post by: Roland Clarke on September 15, 2016, 12:16:05 PM
It's been explained to me that it's to do with the buffering.  Assuming that your sample rate is low enough not to really cause any problem for your processing (therefore having negligible impact on your latency), at 96khz you are filling your buffer up, or emptying it twice as fast as at 48khz.  Given that processing happens exceedingly fast these days the buffer latency is going to be exactly half the time at double the rate.  I would expect almost any desk spec where they give the latency spec to bare this out.  😊
Title: Re: what Digital Mixers to recommend
Post by: Kevin McDonough on September 15, 2016, 02:37:43 PM
Kevin, could you point me to an online manual that says this? My own system doesn't have the option to double the sample rate.

Off the top of my head I don't know of any specific manuals, you'd need to search through. Problem is that most live desks are fixed to one or the other, as you say I don't know of many that are switchable.

However here a few bits that would possibly satisfy you in terms of confirmation:

Quote
James Gordon, managing director of Digico, is on video/record many times talking about 96KHz sample rates, and how it allows digico desks to achieve such low latencies. In particular (as an example of a desk that switched) when the SD9 received its 2013 processing/software upgrade and moved from 48KHz to 96KHz he sited one of the major advantages and reasons for this is the reduce in latency. I'm sure if you found latency figures from before and after this, with the desk hardware and processing chips remaining the same etc, it would confirm. He also mentions their low latency on several other videos introducing new desks such as when the SD10 was launched, and during software upgrades of other desks.


Taken from https://us.novationmusic.com/answerbase/latency-explained (https://us.novationmusic.com/answerbase/latency-explained)
Quote
1. Sample frequency: Buffer size remains fixed no matter what the sample frequency is so the higher the sample frequency, the quicker audio data will pass through the buffer and the lower the latency will be. As an example, audio data will pass though a buffer twice as fast when using 96KHz sample rate as opposed to 48KHz. Most sequencers and sound cards give the option of selecting which sample rate is to be used.


2. Buffer size: The smaller the buffer, the less time it will take for audio data to pass through it, therefore the lower the latency will be. Making the buffer smaller comes at a price however. With a smaller buffer there is less overhead for delays in processing and so the CPU needs to work harder for smaller buffer sizes to ensure that any delays are kept within the time allowed by the buffer.

taken from http://forum.cakewalk.com/How-Big-a-Difference-from-48khz-to-96khz-m2123831.aspx (http://forum.cakewalk.com/How-Big-a-Difference-from-48khz-to-96khz-m2123831.aspx)
Quote
Using unreal (but simple) figures.

Assume a buffer size of 100 samples.
If you have a sample rate of 100Hz (100 samples per second), your latency will be 1 second.
i.e. it takes 1 second for the first sample to make it's way through the 100 sample buffer.
OK?

If you now double your sampling rate to 200Hz, (200 samples per second)can you see how the samples will make their way twice as fast through the buffer? Now the latency (the time it takes to pass through the buffer) is now 0.5 second.

OK, scale these up to 4800kHz, a 1000 sample buffer and we have:-

1000 / 48000 = 20.8ms latency.

Double your sample rate to 96000 and your latency is:-

1000 / 96000 = 10.4ms.

And the all of these various snipits of conversation from a thread on https://www.reddit.com/r/audioengineering/comments/3c2ugm/people_claim_that_96khz_has_less_latency_than/ (https://www.reddit.com/r/audioengineering/comments/3c2ugm/people_claim_that_96khz_has_less_latency_than/)
Quote
It's best to think about it as each step in the process taking a fixed number of samples, including the plug-ins. If the encoding takes 32 samples, then it's still 32 samples regardless of sample rate, but it's going to happen just over twice as fast at 96kHz than 44.1kHz. Those slices are a lot "thinner" at 96kHz.
So yeah latency is way lower all the way through from input to output, including through plugins because the samples are just processed way faster.
The downside is that it puts double the load on the computer as there's now 96,000 samples per second instead of 44,100. And your file sizes are just over double as well, so it's eating up your HD space faster. If it wasn't for these expensive downsides, everyone would just be recording at 96kHz, or 192kHz and enjoying lower latencies all the time.

Quote
I'm a software engineer who's tinkered with writing VST effects.
From my experience, the latency is generally not so much a factor of the effort of the work done, but the fact the plugin(s) need working space. E.g. each plugin needs to build up a block of audio data before getting to work on it. Doing one sample at a time is not efficient from a processing point of view and you also need blocks of data for certain functions (e.g. Fourier transforms).
Now, if its just an efficiency thing, and the processing window size does not need to be relative to the frequency (e.g. plugin with a look-ahead measured in seconds), then at the higher rates you get the advantage of that buffer being available quicker and then pushed out quicker.
To explain this, consider a software monitoring scenario with plugins on a single track between the in and the out. In reality, the software actually 'reacts' to the output asking for data, but I'm going to ignore that because I can't work out how that would help the explanation.
In that example, the system reads from interface input until it has something like 256 samples. The System passes it to the first plugin which does its thing and hands the processed block on. The system repeats this for each plugin. Finally at the last step the block is handed back and the system writes the block back to the interface for playback.
The input and output operations are 'real time' operations. They are constrained by the sample rate and can't be accelerated by a faster CPU.
The bit in the middle with the processing is surprisingly fast unless you're using some crazy complex plugins. Basically it usually takes a negligible amount of time on a modern CPU to process the whole chain compared to the input and output which have to be at the sample rate e.g. 44.1khz.
So, in a simplistic view, at any given time, there's a buffer you're filling at the input, a buffer you're processing through the plugins and a buffer that is being played out of the output.
The higher the sample rate the quicker those buffers get through the input and output.
Plugins don't really affect latency (unless they require that the buffer is larger for reasons of needing more working space). If the time taken to process 256 samples through all the plugs is longer than the time it takes to play the previous 256 bytes out of the output you get dropouts instead of latency increases.

Quote
basically your daw processes audio in chunks, the size of these chunks is defined by your buffer size. the time difference between input and output is 2* your buffer size, so if you increse the sample rate you buffer becomes shorter in time while staying the same length in samples. yes using higher sample rates uses more cpu power but it is lower latency

Quote
I agree with the others that a 64 sample buffer is played in less time at a higher sample rate. However, the latency vs. sample rate discussion assumes your processing power is sufficient in both scenarios.
At 44.1khz, your processors must be able to process 44.1k samples through everything (plugins, etc.) in one second.
At 96khz, that same processing power has to process 96k samples through the same tasks in one second. More than twice the computation has to occur in the same period of time.
Looking at it a different way, if you run a 64 sample buffer at 44.1k, you're refreshing that buffer every 1.45ms. That is 689 buffer refreshes per second. That same buffer is refreshing every 0.67ms at 96k. That is 1500 buffer refreshes per second. So yes, you may theoretically reduce latency but you require more power to do it.


These were all just found with a quick google, I'm sure there would be plenty more if you wanted to research further.  The consensus seems to be that upping the sample rate lowers latency as the processing buffer is now filled up, processed, and emptied again more quickly. This of course assumes that you have the hardware and processing power to cope with it and not cause drop outs. And also explains why a lot of older desks or recording systems could switch between different sample rates, but would often have less channels or less processing available as it went higher, to maintain enough DSP to handle the increased workload.


Hope that helps,


Kev
Title: Re: what Digital Mixers to recommend
Post by: Kevin McDonough on September 15, 2016, 02:46:32 PM
Haha,

however I feel our discussion, while interesting and informative (having to look into the reasons for this in a little more detail has certainly led me to learn something new), is drifting away from the OP's original question  ;D

While it is arguable whether 96Khz may be an issue in audio terms, if  he would like lower latency (especially if his desk will be used for monitoring and IEMs), it is another factor to add into the advice he was already given for his "which desk" question.

Would be happy to continue the latency chat in more detail in a separate thread if people felt they wanted to?

K
Title: Re: what Digital Mixers to recommend
Post by: Scott Helmke on September 15, 2016, 02:49:58 PM
Scott, could you give a specific example? I have to admit that I've never measured the delay in any DAC or ADC. For home use I wish I could find a good DAC with a generous FIFO to eliminate the time base errors (jitter) inherent in spinning disc media.

I spent a few minutes this morning trying to find an example, but oddly enough couldn't before I needed to head off to work. But it's pretty well known that conversion is the bulk of the latency, and there's really nothing that can be done about it.  It takes time to see what the waveform actually is, I guess.

I wasn't aware that time base errors were inherent in spinning disks. Does that mean that CD-ROMs with software are inherently inaccurate too?  Numbers on a disk are just numbers on a disk.
Title: Re: what Digital Mixers to recommend
Post by: Kevin McDonough on September 15, 2016, 02:59:53 PM
I wasn't aware that time base errors were inherent in spinning disks. Does that mean that CD-ROMs with software are inherently inaccurate too?  Numbers on a disk are just numbers on a disk.

Yeah they're fairly common. Initially when CD-roms were created I'm sure things were made to super tight tolerances and it was of course only reading at 1x speed, so I'm sure was pretty stable. But as they upped the read and write speeds to things like 48x, and I'm sure as they became cheaper and more widely available, jitter and errors started to creep in as they weren't reading things quite as accurately as before.

Similarly with spinning hard disks, as they're crammed more and more data onto platters to get bigger sized disks, the "pockets" of data become ever smaller and more difficult to read, and so again errors and problems get created.

Computers have sophisticated error correction algorithms that they use as they read optical disks and hard drives to be able to try and compensate for this.

k

Title: Re: what Digital Mixers to recommend
Post by: Aisle 6 on September 15, 2016, 05:55:42 PM

The first thing is that the DLive makes extensive use of FPGA logic. A general purpose computer CPU relies on a number of basic instructions, and does them over and over to do complex calculations, whereas a FPGA has the entire complex calculation programmed into it, so it can perform a complex operation very quickly. The downside is that it can't change algorithms on the fly, as a computer can. As CPU, GPU and DSP chips get faster and faster, they might make the FPGA obsolete, but for now it's the king for specialized tasks like this. So using FPGAs to speed up number crunching is one big factor that gives the DLive (and the S21 IIRC) a performance edge over CPU/DSP-based mixers. An edge that comes at a price.

The other thing about the DLive is that all of the brains reside in the stage box. Some digital mixers have the brain and control surface in the same box (just like an analog mixer), whereas the DLive is a bit like a control surface for a DAW, except instead of a computer, the control surface runs the mixer that's in the stage box. This eliminates all of the potential bottlenecks of a digital snake, as the mix happens right onstage. So you don't need to know the finer points of the ISO/OSI layers and why they're important, the differences between Dante and AES50 etc.

There's no doubt that the DLive system is pretty sweet. I would have looked at it myself, but it's more than I wanted to spend for my own system. I'll definitely keep it in mind if I start making money in this town.  :)

Hi Bill.
Thanks for all the info, I was aware of the basic layman concepts behind the d-live. I was fortunate enough to be flown overseas to see the prototype of this console well before it was announced. Obviously at this level, I was privy to the function...as I say, at least in layman speak.

I also think that one of the biggest advantages is that there is no third party OS. A&H have written their code straight to the chipset. No windows or linux.

To be clear, I do not work for A&H, I do ave an account with them as well as other manufacturers who also send me new consoles to test in the field.
Title: Re: what Digital Mixers to recommend
Post by: Aisle 6 on September 15, 2016, 06:04:20 PM

LOL.  Unfortunately I'm not a computer scientist so I'm not totally sure of the exact mechanisms. But upping the sample rate definitely does lower the latency  ;D  It's confirmed by all the desk manufacturers on their spec sheets and things. So it's not in question, just the mechanism of what actually causes it.

This too was my understanding, thanks for some further intelligent clarification Kevin.



While its hard to nail down exactly what causes the differences (preamps will play a part, quality of algorithms and effects etc), it is a consideration that  Digico and Midas are generally agreed to be the best sounding desks around, (closely followed it now seems by D-live), and the fact that all are 96khz desks (or potentially 192Khz for an SD7) could be part of the difference. Either by an audio difference, or just by lower latency making it seem more 'real' and immediate.

I agree, it is very difficult to identify one element that effects the big picture. Although, I have been trying to subtly say without saying it, is that the d-live is not a runner up to the Midas, Digico sonics any longer. Use all of these boards all the time and folks will realise as time goes on that the d-live is actually up there with anything. Mind you I would also add Cadac to that list, but we would get even further off topic. : P

Of course, whether all this is worth the money for all the extra processing etc then becomes up to the user. For some it will be, and I'd suggest that for monitors and especially In-Ears it's a big consideration. For others not.
k
For the shows I mix, every minor advantage is welcomed never questioned.
Title: Re: what Digital Mixers to recommend
Post by: Peter Morris on September 15, 2016, 09:41:33 PM
While its hard to nail down exactly what causes the differences (preamps will play a part, quality of algorithms and effects etc), it is a consideration that  Digico and Midas are generally agreed to be the best sounding desks around, (closely followed it now seems by D-live), and the fact that all are 96khz desks (or potentially 192Khz for an SD7) could be part of the difference. Either by an audio difference, or just by lower latency making it seem more 'real' and immediate.


As someone who has owed both a dLive and a Pro2 ... I tend to think the dLive actually sounds better than the Midas  :o but at this level its hard to tell. There is something about the sound quality that everyone loves with dLive.  The quality is most noticeable when doing complex mixes with lots of inputs.

Its not the mic pres that everyone talks about but the internal signal path / processing.  The dlive uses variable (fixed) point processing.  Fixed point sounds better than the floating point that Madas and Digico use. The problem with the fixed point is running out of head room.

To ensure there are no head room problem I believed A&H have used a 72 bit signal path with a 96 bit accumulator.... or something like that (?) In other words you get the best of both worlds. FWIW SSL new consoles also use fixed point with a 64 bit signal path.  All of these consoles are 96kHz. 

In any case the dLive does sound very very good and the UI is a delight to use.
Title: Re: what Digital Mixers to recommend
Post by: Mac Kerr on September 15, 2016, 10:05:23 PM
Processing of digital signals is pretty quick. It's conversion from/to analog that adds the most latency in most mixers.

Just look at the specs of an average device - analog in to analog out is at least two milliseconds.  Digital in to digital out is usually less than half a millisecond.

This. A>D and D>A conversion is usually a fixed number of samples. It happens quicker when the samples fly by quicker. As processing blocks get added up they too can become significant, but are usually less than the conversion time.

Mac
Title: Re: what Digital Mixers to recommend
Post by: Stephen Kirby on September 15, 2016, 11:20:54 PM
I noted that most of the quotes Kevin listed come from the recording world.  Where plug in processors can "look ahead" and latency is basically the amount of offset cause by the most power hungry or combination of plug ins that the system offsets itself with to keep everything sample aligned.

Digital summing got a black eye as there was noticeable sound degradation the more tracks were being mixed together.  This engenders tons of outboard summing boxes where each channel would be run though a D/A summed with an analog mixer or even a simple resistive summing matrix, and back though an A/D into the 2 bus of the DAW.  The absence of digital summing was deemed better than the two extra conversions even though folks recognized that as less than optimal.  Eventually it was noticed that the digital summing wasn't so bad if the tracks were -12 to -18dB instead of nearly FSD (a legacy analog technique of maximizing s/n by tracking everything as near [or even over] clipping).  So now many folks have gone back to "in the box" mixing.  That sounds similar to me to the A&H concept of extra digital headroom.  And I think many digital live sound mixers are deliberately staying below FSD as people are still used to tickling the reds on analog mixers.  I've seen it opined here that some digital desks with "better" pres are basically configured with a bit more headroom and some sort of graceful clipping that tolerates the old analog techniques.
Title: Re: what Digital Mixers to recommend
Post by: Peter Morris on September 16, 2016, 12:56:55 AM
And I think many digital live sound mixers are deliberately staying below FSD as people are still used to tickling the reds on analog mixers.  I've seen it opined here that some digital desks with "better" pres are basically configured with a bit more headroom and some sort of graceful clipping that tolerates the old analog techniques.

Exactly!
Title: Re: what Digital Mixers to recommend
Post by: George Dougherty on September 16, 2016, 02:04:25 PM
I've seen it opined here that some digital desks with "better" pres are basically configured with a bit more headroom and some sort of graceful clipping that tolerates the old analog techniques.
It's not just an opinion. The Venue consoles have 6db or so of headroom between the meters clip point and FSD. It is an option you can disable if you'd rather work metering against true FSD but it was a little more friendly to those who learned to run inputs hot.
Title: Re: what Digital Mixers to recommend
Post by: Aisle 6 on September 16, 2016, 06:15:18 PM
It's not just an opinion. The Venue consoles have 6db or so of headroom between the meters clip point and FSD. It is an option you can disable if you'd rather work metering against true FSD but it was a little more friendly to those who learned to run inputs hot.

Hi George
Indeed, the Venue console preference of metering is excellent. You can actually adjust the metering scale quite a bit to be 0,1,3 or 6dB at full scale before clip. I think folks here are more talking about summing through the mix busses and that some of the consoles most favoured for their sound have very high head room and some are almost  infinite. Gotta love maths. : )

Title: Re: what Digital Mixers to recommend
Post by: George Dougherty on September 17, 2016, 09:58:01 AM
Hi George
Indeed, the Venue console preference of metering is excellent. You can actually adjust the metering scale quite a bit to be 0,1,3 or 6dB at full scale before clip. I think folks here are more talking about summing through the mix busses and that some of the consoles most favoured for their sound have very high head room and some are almost  infinite. Gotta love maths. : )



There is that too, though Stephen's comment sounded like he was talking a bit more about input stages.  It's very easy to build in plenty of headroom in the mix path, rather than not like some small popular non-motorized consoles originally did.
Title: Re: what Digital Mixers to recommend
Post by: Roland Clarke on September 18, 2016, 05:45:01 PM
I noted that most of the quotes Kevin listed come from the recording world.  Where plug in processors can "look ahead" and latency is basically the amount of offset cause by the most power hungry or combination of plug ins that the system offsets itself with to keep everything sample aligned.

Digital summing got a black eye as there was noticeable sound degradation the more tracks were being mixed together.  This engenders tons of outboard summing boxes where each channel would be run though a D/A summed with an analog mixer or even a simple resistive summing matrix, and back though an A/D into the 2 bus of the DAW.  The absence of digital summing was deemed better than the two extra conversions even though folks recognized that as less than optimal.  Eventually it was noticed that the digital summing wasn't so bad if the tracks were -12 to -18dB instead of nearly FSD (a legacy analog technique of maximizing s/n by tracking everything as near [or even over] clipping).  So now many folks have gone back to "in the box" mixing.  That sounds similar to me to the A&H concept of extra digital headroom.  And I think many digital live sound mixers are deliberately staying below FSD as people are still used to tickling the reds on analog mixers.  I've seen it opined here that some digital desks with "better" pres are basically configured with a bit more headroom and some sort of graceful clipping that tolerates the old analog techniques.

This is an "old wives tale".  Digital summation by its very nature is "perfect".  That's not to say that other factors may be at play, but summing boxes fall into the same category as external wordclock improving jitter, it's urban myth at best, in reality, hogwash.  Digital is capable of far more pristine sound than analogue, period.
Title: Re: what Digital Mixers to recommend
Post by: Jeff Simpson on September 19, 2016, 07:06:36 AM
Digital summation by its very nature is "perfect".

Fixed point addition is perfect, assuming you do not run out of headroom. Floating point addition can introduce errors; how big these errors are and how often they might occur depends on how many bits are used, but to say that all summing systems are perfect is not quite accurate.