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: Jeff Lelko on October 20, 2019, 06:02:29 PM
-
Greetings - quick question about AES I/O - is it worth using at the professional level? My guess would be “yes” as it significantly reduces the number of analog to digital and digital to analog conversions, but I’d like to hear input on reliability for professional applications before converting.
My typical rig for audio playback is something like (Computer or Solid State Media Player ) to (dLive DM32) to (dbx DriveRack VENU360) to (amps and speakers). I count 5 conversions in that signal chain. Going with the AES I/O which is present on all my sources, DriveRack, and dLive via the AES3 option card, I’d only have one. I’ve never used AES I/O before, so I’m asking is this bulletproof enough to rely on versus the trusty analog? The cables would be proper data cables and the cable lengths would all be under 1 meter. All input is appreciated. Thanks!
-
There are as you are pointing out, advantages to using aes/ebu outs and ins, reduced latency probably the biggest advantage. There are a number of empirical observations that suggest even as many as 10-15 AD/DA conversions have little or none in terms of detrimental effect on sound quality, but why not purify the audio path as much as possible? A large number of powersoft and other top amp manufacturers who include DSP, have AES/EBU, arguably possible to have only one AD/DA from mic to speaker! 😁
-
A rule of thumb I read a while back - AES3 for 2 channels of audio, Dante for anything more than 2 channels.
AES3 can be thought of as the professional version of S/PDIF.
-
Thanks Roland and Caleb! I agree that the numerous conversions have no real detrimental impact to the audio - especially in the live production environment. I just noticed that basically all my FOH gear has this option and questioned if it’s worth using. I just don’t want to take a hit on reliability in any way. One other advantage of AES is not chewing through inputs on my DM32. The single AES3 option card would handle all the I/O I need!
I agree with you on the Dante argument too Caleb. My only issue with that is none of my FOH equipment has Dante built-in. I know A&H makes a dLive card for that too, but my media players and DSP would be left out, and my playback computer currently doesn’t use Dante. I think Dante would push the “worth it” question to the “not”, but AES is only the cost of a single card and some cables. Thanks again!
-
For my main system I use the the AES out of my Allen Heath QuPac into a DBX360,
works and sounds great.
Each one of my amp racks has it's own DSP in the rack, when I
need to drive two racks I use the Whirlwind SP1X2 AES splitter.
-
The are a number of analog to Dante, or AES or USB to Dante converters out there, the Audinate branded AVIO line being just one with example.
-
Also be aware that AES3 can cause extra latency compared to analog.
-
Short answer: Yes.
Caveats:
- Each device needs to either: A) Be running at the same sample rate as the device it is obtaining signal from; OR B) Be able to sample rate convert the incoming signal.
- You may see some increased latency over the current analog connections, but this may be insignificant in your application. Or it may not - this is for you to decide.
-
Short answer: Yes.
Caveats:
- Each device needs to either: A) Be running at the same sample rate as the device it is obtaining signal from; OR B) Be able to sample rate convert the incoming signal.
- You may see some increased latency over the current analog connections, but this may be insignificant in your application. Or it may not - this is for you to decide.
In the case with my QuPac and DBX360 there is less latency using the AES to drive the mains, keep that in mind if you use an analog drive to run subs on an aux.
The 360 can be set to clock to the incoming AES or sample rate convert, I have mine set to clock to the incoming AES.
-
One interesting choice in using AES interconnects in your signal chain is that it allow you to decide which device will be doing your conversion. I still use BSS366T crossovers with my Vertecs because the BSS has a very good sounding D to A conversion.
-
One other subtle advantage of digital interconnects is that most are designed to fail to silent. Something kills the console with the amps still on? Analogue gives you a boom (and a reputation), digital just stops passing audio.
-Russ
-
One other subtle advantage of digital interconnects is that most are designed to fail to silent. Something kills the console with the amps still on? Analogue gives you a boom (and a reputation), digital just stops passing audio.
-Russ
Something else I appreciate with digital interconnects is the ease of maintaining headroom. To properly calibrate an analog signal path you have to pay attention to the clip point on each device in the path; in digital it’s all just 0dBFS and you can move on to the next part of the job.
-
Something else I appreciate with digital interconnects is the ease of maintaining headroom. To properly calibrate an analog signal path you have to pay attention to the clip point on each device in the path; in digital it’s all just 0dBFS and you can move on to the next part of the job.
Ehhh...sort of.
But if your signal chain is anywhere close to 0dBFS, you're already to hot to begin with.
-
Thanks for all the input so far - the consensus definitely seems to say AES is worth a try. I’m curious about the latency comments though. My (uneducated) thought is that AES would be no worse if not better than analog in terms of latency due to the many A-D and D-A conversions taking place when patching together via analog. In a 100% analog system I’d agree, but since each component of my system is digital until the amplifier, wouldn’t staying digital only help with any latency issues? Thanks again!
-
Thanks for all the input so far - the consensus definitely seems to say AES is worth a try. I’m curious about the latency comments though. My (uneducated) thought is that AES would be no worse if not better than analog in terms of latency due to the many A-D and D-A conversions taking place when patching together via analog. In a 100% analog system I’d agree, but since each component of my system is digital until the amplifier, wouldn’t staying digital only help with any latency issues? Thanks again!
Some devices have more AES3 latency because they use embedded *sample rate conversion* to match the internal processing clock rate. In the original Crown I-Tech the input-to-output latency for AES as almost double the time for an analog input. Not every device behaves this way but you'll need to look at the numbers just as you would for multiple AD/DA conversions in any time sensitive application.
-
Ehhh...sort of.
But if your signal chain is anywhere close to 0dBFS, you're already to hot to begin with.
Depends what your application is. I’ve got a live webstream going out right now; if it’s peaks weren’t fairly close to 0dBFS there is no way it would be loud enough.
-
Some devices have more AES3 latency because they use embedded *sample rate conversion* to match the internal processing clock rate. In the original Crown I-Tech the input-to-output latency for AES as almost double the time for an analog input. Not every device behaves this way but you'll need to look at the numbers just as you would for multiple AD/DA conversions in any time sensitive application.
Thanks for the clarification Tim. Definitely something interesting that I didn’t think of. I’ll have to run an experiment to see where my gear ends up latency-wise when running AES versus analog. Thanks again!
-
Thanks for the clarification Tim. Definitely something interesting that I didn’t think of. I’ll have to run an experiment to see where my gear ends up latency-wise when running AES versus analog. Thanks again!
Hi Jeff-
A 100% analog rig will be almost the speed of light (for our purposes). Add *anything* digital to the signal path and the clock starts running. If none of your digital devices need SRC to work the digital through put is about as minimal as it can get but still takes time.
Using AES 3 means you don't go through multiple AD/DA conversions which will eventually degrade the audio and take more time. You'll preserve signal quality and hopefully have less time penalty.
-
Thanks again Tim. I agree that a 100% analog system would function with the least amount of latency. Since all the devices in my signal chain prior to the amps are digital, I’m just surprised that staying digital could potentially induce more latency than going through all the AD/DA conversions. I can see where the added delays can factor in though. I guess there’s only one way to find out for sure!
-
Thanks again Tim. I agree that a 100% analog system would function with the least amount of latency. Since all the devices in my signal chain prior to the amps are digital, I’m just surprised that staying digital could potentially induce more latency than going through all the AD/DA conversions. I can see where the added delays can factor in though. I guess there’s only one way to find out for sure!
In an all digital system without SRC, the number will be lower than using the analog inputs of the digital devices. The point I'm making is that every digital device adds *some* latency, end to end, and using the analog inputs will create more latency and eventual signal degradation from multiple AD/DA conversions.