The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
Has Anyone Heard The Focusrite AM2. . . Mixers (Digital)
Old 9th March 2017
  #1
Lives for gear
 
tourtelot's Avatar
 

Thread Starter
Has Anyone Heard The Focusrite AM2. . .

Dante headphone amp/monitor controller?

Looks like a nice Dante add-on if it sounds good. Can't find any reviews even if it's been out a year.

D.
Old 9th March 2017
  #2
Lives for gear
 
Yannick's Avatar
 

I wonder as well, and it is dirt cheap. If you substract the audinate fee and dual dante interface, it is basically free ???

Seems like a decent da and headphone amp, and it avoids a local switch for monitoring.
Old 10th March 2017
  #3
Lives for gear
 
tourtelot's Avatar
 

Thread Starter
Yes, but how does it sound?

D.
Old 12th March 2017
  #4
Gear Maniac
 

It's a neat little interface. I'd buy one or two if I could find a use for them.

Might buy one anyway.
Old 12th March 2017
  #5
Edit - I just ordered one. Should be here this week. I will test and use it this weekend and then, Doug, if you want, I'll loan it to you.

It looks like a good deal for 4 amu*.

*amu=audio monetary unit = $100 US
not to be confused with AMU=Aviation Monetary Unit = $1000 US.
Old 12th March 2017
  #6
Lives for gear
 
tourtelot's Avatar
 

Thread Starter
Quote:
Originally Posted by TMetzinger View Post

*amu=audio monetary unit = $100 US
not to be confused with AMU=Aviation Monetary Unit = $1000 US.
Awesome! Hahaha!

D.
Old 17th March 2017
  #7
OK got it in tonight. I'll write a full review after using it on location, but here's the short version.

Plugged it into a POE port on my switch and it came right up (1 Gbps). Dante controller found it right away - no need to install any Focusrite utilities. It became the master clock too. Reports that it's based on the Ultimo X2 chipset, so I'd expect similar good things from the Amphenol and other Ultimo-based products.

I fed test 440 Hz -20dBFS signal to it and my RDU-LB4P at the same time, and then fed my monitor controller/metering unit from each of them on a different pair of inputs. With the Line Out set to +4 dBU on each and the volume knob up on the AM2 - they matched perfectly. Mono signal read the same voltage on both outputs throughout the entire digital trim range. So technically, the device works exactly as advertised.

How does it SOUND? I put a couple of tracks I know well (Joe Jackson's "You Can't Get What You Want" and Donald Fagen's "The Goodbye Look") through it from Reaper at both 48K and 96K (yes, it syncs up to DVS at 96K with no issues), and it sounded great to me, which means, it sounded like NOTHING. Sounded "correct". Volume at max wasn't bad, wasn't ear-splitting either.

It's got a handy 3/8" mounting inset in the bottom if you want to mount it on a mic stand or accessory.

One glitch - the supplied AC adapter comes with multiple different plug-plates, and the US plug plate is not right - it's 1mm narrower than the others, and that means that when you slide it on to the adapter, it is not secure. I've notified Focusrite via email, and we'll see what they say.

Doug, PM me with shipping info, and I'll send it your way on Monday for an evaluation if you like.

Last edited by TMetzinger; 17th March 2017 at 03:24 AM..
Old 22nd March 2017
  #8
A more formal review. Doug, make note of the fact that the unit comes with a 12 dB pad on the headphone output set by default. Easily fixed with the Rednet software.
Attached Files
File Type: pdf RedNet AM2 Review.pdf (169.6 KB, 85 views)
Old 22nd March 2017
  #9
Lives for gear
 
tourtelot's Avatar
 

Thread Starter
Thanks for the work put in!

D.
Old 26th March 2017
  #10
Gear Nut
 

I have one and it's excellent – doesn't sound as good as my Avocet did, but it's a lot cheaper. Beware that owing to the Dante chipset used you can't get latency under 1ms, which makes it unusable in some monitoring scenarios. The more expensive KLANG:quelle can by comparison go down to 150us (this is Dante latency, it excludes the D/A latency – from memory it's about 200us at 96k/150us).
Old 26th March 2017
  #11
Lives for gear
 
tourtelot's Avatar
 

Thread Starter
Quote:
Originally Posted by lukehatpadl View Post
Beware that owing to the Dante chipset used you can't get latency under 1ms, which makes it unusable in some monitoring scenarios.
Hmm. No reason to doubt you but this seems odd since Dante latency usually relates to number of switches and is typically safe at around 250uSec. Why do you think it's so high? Empirically tested? Focusrite seems to make a selling point out of the AM2 being "perfect" for in-studio monitoring.

Thanks.

D.
Old 26th March 2017
  #12
Quote:
Originally Posted by tourtelot View Post
Hmm. No reason to doubt you but this seems odd since Dante latency usually relates to number of switches and is typically safe at around 250uSec. Why do you think it's so high? Empirically tested? Focusrite seems to make a selling point out of the AM2 being "perfect" for in-studio monitoring.

Thanks.

D.
I've tested. Dante ships the data to the endpoint as quickly as it can. The latency setting for a device is a "target" value, and between any subscriber/receiver pair the greater of the two settings is used for measuring and reporting "late" packets.

Here's an example - I've got my AM2, with a Latency value of 1 ms, subscribing to my 100 Mbps Radio Design labs interface on the same switch, which is set to it's minimum value of 2 ms.

Note that the reported latency are all 812 microseconds. With other devices, I've seen reported latency lower.

Sometime this week I'll ship audio from a device to both the RDL and the AM2 at the same time, and record the analog outputs to four channels in a daw and we'll see what sort of delays exist between the outputs.
Old 26th March 2017
  #13
Lives for gear
 
tourtelot's Avatar
 

Thread Starter
Okay. Thanks both.

D.
Old 27th March 2017
  #14
So, here's what I did.

I used Reaper on one computer to send out a 1 KHz tone with an automated mute envelope so it would send ON for 5 seconds, then OFF for 5 seconds, and repeat. The signal was sent to a single DANTE transmitter on that computer.

I then had three devices subscribe to that Transmitter:

Radio Design Labs RDL-LBU-4P output channel 1, patched right back to input channel 3.
Focusrite AM2 output channel 1, patched via analog cable to the RDL input channel 1
X32 Rack, with the input running into one channel and an analog output driven from that channel at the input tap point, patched via analog cable to the RDL input channel 2.

Then on a second computer, I recorded four signals via Dante:
1 - a direct subscription from the first computers output channel (so no DA/AD here)
2 - RDU Channel 1 (DA/AD from Rednet AM2)
3 - RDU Channel 2 (DA/AD from X32)
4 - RDU Channel 3 (DA/AD from RDU output 1)

Everything was connected to a single switch - the RDU unit runs at 100 Mbps and everything else ran at 1 Gbps.

And here's a screenshot showing the relative signals:


The direct signal starts at sample 3687011 (our baseline for the simple ethernet transport).
The AM2 signal starts at sample 3687221 (a delay of 210 samples at 48K or 4.375 ms)
The X32 signal starts at sample 3687265 (a delay of 254 samples at 48K or 5.292 ms)
The RDU signal starts at sample 3688943 (a delay of 1932 samples at 48K or 40.25 ms)

Here's the tail of the signal - it looks to me like the "delay" on the RDU (Track 4) is that the output amplifier ramps up rather than kicking on immediately since the tail is much more closely aligned.


My initial takeaway is that there are indeed differences between the A/D and Dante latency of different devices. Not sure yet what it means in practical terms. Clearly the latency is many orders of magnitude larger than an all-analog chain, but I don't know how it would compare to using two different MADI interfaces, for instance.

I also want to do some more testing - let's find out if the same signal sent to two outputs channels on the same device will show any difference in phase, for instance. Same for the same signal sent to two inputs on a single device.

EDIT: I did this testing - the same signal captured on multiple channels of the same device, or sent to multiple channels on the same device, is perfectly in phase. So now I really want to test what would happen with two devices of the same type, since they should have identical performance and they're all on the same clock. Then I could start inserting switch hops and see what happens.

For now, at least, I think I'd be careful when recording or mixing from multiple disparate DANTE A/D devices as differing latencies might give you phase/delay issues.

Last edited by TMetzinger; 27th March 2017 at 02:13 AM..
Old 27th March 2017
  #15
Lives for gear
 
tourtelot's Avatar
 

Thread Starter
Isn't Dante supposed to make all Dante chip latency the same ITB? I believe the other latency that is occurring is in the AD/DA conversion.

As well, by all accounts, 1mSec is a number far below anything that should affect, let's say sync in an overdub situation. 45mSec is like standing 45 feet away from your pre-recorded voice and trying to keep sync. Impossible, I'd say.

D.
Old 27th March 2017
  #16
Lives for gear
 
tourtelot's Avatar
 

Thread Starter
Attached Thumbnails
Has Anyone Heard The Focusrite AM2. . .-screen-shot-2017-03-26-6.30.49-pm.jpg  
Old 27th March 2017
  #17
Quote:
Originally Posted by tourtelot View Post
Isn't Dante supposed to make all Dante chip latency the same ITB? I believe the other latency that is occurring is in the AD/DA conversion.

As well, by all accounts, 1mSec is a number far below anything that should affect, let's say sync in an overdub situation. 45mSec is like standing 45 feet away from your pre-recorded voice and trying to keep sync. Impossible, I'd say.

D.
I agree that the latency measured at the head of my test was an artifact. The tail showed all three devices within one ms of each other.
Old 27th March 2017
  #18
Gear Nut
 

Trust me, I bought an AM2 thinking it would be useful for headphone monitoring and when I found out it was limited to 1ms or greater I was somewhat annoyed (I ended up keeping it though and buying a KLANG:quelle as well).

The Ultimo chipset used by the AM2 does not support latencies under 1ms, this is mentioned in the spec sheet:

https://www.audinate.com/sites/defau...aug-2016_0.pdf

The lower latencies are only possible with the Brooklyn chipset or the PCIe card. It may take some research to determine which chipset is used by the device you have in mind. (Having multiple switch hops also impacts the latency but you can't get around the chipset one.)

And, as I mentioned before, you also need to add the converter latencies to calculate the total round-trip latency. For monitoring whilst playing, I need it to be under 1ms, and the AM2 is not useful there.

Last edited by lukehatpadl; 27th March 2017 at 07:32 AM..
Old 29th April 2017
  #19
Lives for gear
 
tourtelot's Avatar
 

Thread Starter
Just ordered mine. I read on the manual that the 1/4" TRS jack is wired Tip = Right. Tim? It this correct? Almost every headphone I've ever had experience with is wired Tip = Left.

Easy enough to change if it is.

Thanks.

D.
Old 29th April 2017
  #20
Hang on. I suspect it's a misprint but I'll test the l/r balance.

Edit - when I pan a signal Left, it goes to the left ear in my headphones. So I suspect a typo.
Old 1st May 2017
  #21
Lives for gear
 
tourtelot's Avatar
 

Thread Starter
Thanks Tim. I just got confirmation from a tech at Focusrite that the version of the manual that I found does have a misprint. The device is wired correctly, Tip = Left.

D.
Old 6th May 2017
  #22
Lives for gear
 
tourtelot's Avatar
 

Thread Starter
Got my AM2 today. Super easy setup (OSX RedNet App) and pretty much plug and play. Nice and loud and, at least to my ears, no distortion at levels that I need to turn down quickly. Even my Benchmark runs out of steam at less volume. (Not that I normally listen at ear damaging levels, but it's nice to know that 5th gear works.)

So far, two thumbs up.

D.

Oh and left is left and right is right
Old 17th June 2017
  #23
LX3
Lives for gear
 
LX3's Avatar
 

Actually, I don't find that the headphone amp on the AM2 really goes loud enough. It's passable for in-ears, but hasn't got enough oomph to drive closed cans in loud situations.... especially when soloing a track that isn't peaking over -10dB (which is most of my tracks). If I'm using cans I spend a lot of time with the headphone volume at max, which tells the whole story. My DirectOut Ma2chbox goes a lot louder.

Something Focusrite could address in a mkII version I hope?

Its a shame that so many manufacturers underestimate how much power the headphone amps in pro audio kit need (The Marantz PMD661 being the classic case - makes the recorder very hard to use.)
Old 17th June 2017
  #24
Lives for gear
 
tourtelot's Avatar
 

Thread Starter
You probably know this from further up in the thread but output gain is selectable via the RedNet Control 2 app on your Mac or your PC. There are three settings.

I am not using my AM2 on a rock and roll stage, for sure, but I am pretty keen on loud and clean (if I need it) and my AM2, set at its highest output gain is plenty loud for me.

D.
Topic:
Post Reply

Welcome to the Gearslutz Pro Audio Community!

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


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

Advanced Search
Forum Jump
Forum Jump