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
MOTU 1248, 8M, 16A Thunderbolt interface Audio Interfaces
Old 17th January 2017
  #2941
Gear Addict
 

no condescending explaining necessary. read what i wrote. it basically says exactly that, so why repeat it back to me? also, i wrote that the 24ai does NOT have the same quality converters as the 16A, hence my comment. jeez...

Quote:
Originally Posted by msmucr View Post
Excuse me, but I've explained before, such rigs with its local and expanded channels aren't comparable to some "single box" system like HD24, Radar or so.. in this case there isn't any inherent latency associated with transport between those boxes, so there's noting to compensate for.
Similarly you also don't have any offset among channels within single MOTU interface..
MOTU also have 24AI and 24AO, it will behave exactly as your HD24 from 2001 with regards to exact channel alignment .

Michal
Old 17th January 2017
  #2942
Gear Addict
Quote:
Originally Posted by d. gauss View Post
1) the 8A is supposed to have exactly the same "everything" (converters, etc.) as the 16A, just less of it.

2) MOTU could have made things easier for everyone if they just made a 24 ch A/D with the same conversion as the 16A/1248, but they don't. the one they do make uses lesser quality chips.

linking devices together (that you have to manually offset) to record a 10 piece band is no fun and plain stupid in 2017. makes me really miss my alesis hd24 (2001??)... 24 in/out with a hiccup ever.
I have been looking for a review that specifically addresses the issue of using multiple units in an AVB network. AVB is the unique selling point of the MOTU line and I think people need to understand how it really performs.

I don't think it would be difficult for MOTU to automatically detect and align their own devices (including the network switch) considering network latency is fixed at 30 samples one way.
A drop down menu similar to how one chooses the Clock Master could be provided where you pick which interface is the main unit to align to. This would automatically delay the inputs and outputs of the main unit connected to the DAW and other networked interfaces to the latency of worst networked interface.

Maybe this is also a feature that can be included as part of the AVB standard where connected devices report their latencies to the network and adjust accordingly automatically so that all the inputs and outputs on the network are sample accurate.
Old 17th January 2017
  #2943
Well the problem is just that you have ADAT ports on most of the AVB Interfaces. I think it will be difficult to compensate for different converters, most manufacture's won't even publish the latency caused by their converters. And with the 112D it will be a different topic with AES/EBU in combination with MADI.

This is no fault of the avb network, you have to deal with this problem in every situation, where different converters are used to record the same signals in the same room. If you ever used a digital setup with more than one converter, you have the same issue.

But it would be very nice, if you could align these offsets in the mixer.
Old 18th January 2017
  #2944
Maybe it has already been asked and answered, my search did not find it, but has anyone used any of the MOTU Thunderbolt devices with a new MacBook Pro with Thunderbolt 3 ports? Does it work with the USB-C/Thunderbolt 3 to Thunderbolt adapter?
Old 20th January 2017
  #2945
Here for the gear
 
safetyfirst's Avatar
 

Since I switched to the new and supposedly improved driver I got huge problems. I even switched to macOS Sierra to make sure everything is on the latest. But the problem still exists, as following: Let's say I reboot my MacPro 5,1, turn on my Motu 1248 connected over USB, start working in Logic, everything works fine but as soon as I switch to Safari and listen to some Youtube and then go back to Logic, Audio stops working and I have to reboot the device. Same goes for switching to After Effects. It seems as soon as the device has to deal with a different audio capable program it struggles. I had quite a few issues like this over the two years I'm owning the 1248 and 16A.

Anybody else experiencing something like this? Maybe it could be a hardware problem on my side.

Hardware: MacPro 5,1, 12 Core 2,93GHz, 32 GB RAM, macOS Sierra 10.12.2, Logic 10.2, Ableton Live 9.7

Thank you!
Old 20th January 2017
  #2946
Gear Addict
 

Quote:
Originally Posted by safetyfirst View Post
I had quite a few issues like this over the two years I'm owning the 1248 and 16A.

Anybody else experiencing something like this? Maybe it could be a hardware problem on my side.
yup, you are not alone. i was stable with a single 16A until i bought an 8A to expand and updated the driver. now things are very wonky. :(
(still on yosemite though)

btw you state that you have 1248 & 16A. are they sample accurate to each other via loopback? my 16A/8A combo is NOT. very frustrating.
Old 20th January 2017
  #2947
Here for the gear
 
safetyfirst's Avatar
 

Quote:
Originally Posted by d. gauss View Post
yup, you are not alone. i was stable with a single 16A until i bought an 8A to expand and updated the driver. now things are very wonky. :(
(still on yosemite though)

btw you state that you have 1248 & 16A. are they sample accurate to each other via loopback? my 16A/8A combo is NOT. very frustrating.
Thanks for your reply. No, unfortunately here also not sample accurate. Really love the AVB concept and channel count I can have with the two units but with the driver behaving like this it's very frustrating and this is since Mavericks almost the same. In the beginning I had to reboot the device because the web control panel didn't connect any more. Now it's when switching between applications. Also sometimes it mixes in a clock error on the connected 16A. Definitely considering to switch.

Is Mr. Miller still on here? Would be great to get some insights if there is fix coming.
Old 21st January 2017
  #2948
Gear Maniac
 
not like this's Avatar
 

What is the frequency response of the pres in the 1248? How would they compare to the new Slate VSR-8 interface which boasts flat pres for use with modeling software?
Old 22nd January 2017
  #2949
1248, 16A and Stage B16 use CS5381 chips, with 0.0002 percent distortion. But the device is a complete different design than slate vrs.

It integrates with everything, works with UAD and large sessions at higher cpu load, has low latency even over usb2 on pc, gives you a great sounding hardware mixer with headphone and live mix distribution per interface and allows every possible and expandable configuration over network by adding other avb devices, Adat, Madi, aes/ebu... and you can use "real" mics with their remote controlled preamps. Even ribbons or low output dynamics sound really great on the motu pres.

Different route than the vrs8... It's really a very open system.
Old 22nd January 2017
  #2950
Gear Maniac
 
not like this's Avatar
 

For clarity, I own a 1248 and I'm curious about using the new ML-2 pencil mic with the 1248's mic pre.

I suppose I can just try it when they ship!
Old 22nd January 2017
  #2951
Quote:
Originally Posted by safetyfirst View Post
Since I switched to the new and supposedly improved driver I got huge problems. I even switched to macOS Sierra to make sure everything is on the latest. But the problem still exists, as following: Let's say I reboot my MacPro 5,1, turn on my Motu 1248 connected over USB, start working in Logic, everything works fine but as soon as I switch to Safari and listen to some Youtube and then go back to Logic, Audio stops working and I have to reboot the device. Same goes for switching to After Effects. It seems as soon as the device has to deal with a different audio capable program it struggles. I had quite a few issues like this over the two years I'm owning the 1248 and 16A.

Anybody else experiencing something like this? Maybe it could be a hardware problem on my side.

Hardware: MacPro 5,1, 12 Core 2,93GHz, 32 GB RAM, macOS Sierra 10.12.2, Logic 10.2, Ableton Live 9.7

Thank you!

Next time it happens.,. Go to logic audio preferences and restart the driver and see how it goes.


Cheers

Wiz
Old 22nd January 2017
  #2952
Quote:
Originally Posted by not like this View Post
For clarity, I own a 1248 and I'm curious about using the new ML-2 pencil mic with the 1248's mic pre.

I suppose I can just try it when they ship!
I don't know if the preamp is that crucial to the simulation system, I assume the real challenge for them is to get their reference-mics to a constant quality level.

The problem with these mic simulations will be the software-only-monitoring. But at the price-point of the ML-2, I think it could be a fun and creative tool to use, even if the simulations might not outperform or play in the same league of their modelled originals.

It might even be, that the 1248 and ML-2 perhaps lead to a better sounding result, than their own combination of mic and preamp. Who knows?
Old 22nd January 2017
  #2953
Gear Nut
 
wilkinsi's Avatar
Since I only work at 44.1k, it would be nice to be able to make use of the extra ADAT ports (normally required for high level sample rates). I can't afford an ADAT/MADI converter to add to my 112D.
Old 22nd January 2017
  #2954
Gear Maniac
 
not like this's Avatar
 

The preamp and mic in the slate stuff is flat so you can accurately apply the simulated microphone and mic pre.

So the pre is crucial to accurately replicating the original sound. I'm optimistic that you could make it sound decent without the correct pre, but if anyone has any numbers in the 1248 pres, I'd love to see them.

Quote:
Originally Posted by Bjoern Bojahr View Post
I don't know if the preamp is that crucial to the simulation system, I assume the real challenge for them is to get their reference-mics to a constant quality level.

The problem with these mic simulations will be the software-only-monitoring. But at the price-point of the ML-2, I think it could be a fun and creative tool to use, even if the simulations might not outperform or play in the same league of their modelled originals.

It might even be, that the 1248 and ML-2 perhaps lead to a better sounding result, than their own combination of mic and preamp. Who knows?
Old 22nd January 2017
  #2955
Here for the gear
 
safetyfirst's Avatar
 

Quote:
Originally Posted by Wiz_Oz View Post
Next time it happens.,. Go to logic audio preferences and restart the driver and see how it goes.


Cheers

Wiz
Yes, that works and this is how I'm doing it most of the times or rebooting the device within the control panel. But this is just not how it's supposed to be. Thanks though for the tip.
Old 22nd January 2017
  #2956
Quote:
Originally Posted by safetyfirst View Post
Yes, that works and this is how I'm doing it most of the times or rebooting the device within the control panel. But this is just not how it's supposed to be. Thanks though for the tip.
This used to happen to me.

I think its caused by the system sound, changing the sample rate of the 16A and the Logic Project is at a different sample rate, than the sound being played.

FWIW, at some point it went away for me.

It was a couple of driver revs back.

cheers

Wiz
Old 22nd January 2017
  #2957
Here for the gear
 
safetyfirst's Avatar
 

Quote:
Originally Posted by Wiz_Oz View Post
This used to happen to me.

I think its caused by the system sound, changing the sample rate of the 16A and the Logic Project is at a different sample rate, than the sound being played.

FWIW, at some point it went away for me.

It was a couple of driver revs back.

cheers

Wiz
Yes, I think so too, I always work in 48KHz in Logic so with the rest beeing 44 ... It went also away a few driver revs ago but since 2.0 it's back in full effect. Super annoying. Can't understand why there are so many hickups with those devices, feels alot like beeing a beta tester.
Old 23rd January 2017
  #2958
Lives for gear
 
jlaws's Avatar
Quote:
Originally Posted by not like this View Post
The preamp and mic in the slate stuff is flat so you can accurately apply the simulated microphone and mic pre.

So the pre is crucial to accurately replicating the original sound. I'm optimistic that you could make it sound decent without the correct pre, but if anyone has any numbers in the 1248 pres, I'd love to see them.
Iirc audiofanzine tested the pres on the 1248 and found them to be remarkably flat. This was on their French site, but it's easy enough to translate.
Old 23rd January 2017
  #2959
nms
Lives for gear
 
nms's Avatar
Quote:
Originally Posted by safetyfirst View Post
Since I switched to the new and supposedly improved driver I got huge problems. I even switched to macOS Sierra to make sure everything is on the latest. But the problem still exists, as following: Let's say I reboot my MacPro 5,1, turn on my Motu 1248 connected over USB, start working in Logic, everything works fine but as soon as I switch to Safari and listen to some Youtube and then go back to Logic, Audio stops working and I have to reboot the device. Same goes for switching to After Effects. It seems as soon as the device has to deal with a different audio capable program it struggles. I had quite a few issues like this over the two years I'm owning the 1248 and 16A.

Anybody else experiencing something like this? Maybe it could be a hardware problem on my side.
This is quite common with converters when you have sources on your computer trying to lock into different sample rates. You can only be locked into one sample rate at a time obviously. DAW usually takes preference. Ultimately if you don't want to deal with it you need to set your OS to use the same SR as you typically work in your DAW. Otherwise, you just have to learn what not to do and how to quickly fix it when conflicts mess it up. I work in Ableton and if my OS is at a different rate I just switch off the audio in the top right corner of Live before playing other sources. Of course I forget sometimes, but a quick power cycle on my interface (Lynx Hilo) solves it. Setting everything to the same SR will end your issues though.
Old 23rd January 2017
  #2960
Here for the gear
 
safetyfirst's Avatar
 

Quote:
Originally Posted by nms View Post
This is quite common with converters when you have sources on your computer trying to lock into different sample rates. You can only be locked into one sample rate at a time obviously. DAW usually takes preference. Ultimately if you don't want to deal with it you need to set your OS to use the same SR as you typically work in your DAW. Otherwise, you just have to learn what not to do and how to quickly fix it when conflicts mess it up. I work in Ableton and if my OS is at a different rate I just switch off the audio in the top right corner of Live before playing other sources. Of course I forget sometimes, but a quick power cycle on my interface (Lynx Hilo) solves it. Setting everything to the same SR will end your issues though.
Thanks so much for your insight! Completely get that but I thought that the interface should be able to switch flawlessly between different cycles. At least it worked for a while without any problems. Thanks again!
Old 24th January 2017
  #2961
Gear Nut
Quote:
Originally Posted by d. gauss View Post
are they sample accurate to each other? have you done a loopback to see?
So I did a loopback test and found that they are not sample accurate. The output of the 24ao seems to be a bit delayed as compared to that of the 1248 when I use the 1248 as my primary interface with the 24ao connected with AVB
Old 25th January 2017
  #2962
Gear Maniac
 

Quote:
Originally Posted by hellofishy View Post
So I did a loopback test and found that they are not sample accurate. The output of the 24ao seems to be a bit delayed as compared to that of the 1248 when I use the 1248 as my primary interface with the 24ao connected with AVB
I think motu reps should be responding to these claims by now. AVB should be delay free or there should be an easy way of dealing with it.

I was planning to buy a 1248 but now I am holding off until further info
Old 25th January 2017
  #2963
Lives for gear
 

I agree, the motu reps need to comment on this and let us know if there is any plan to fix the delay between units when using avb. I own a 1248 and use it for a small live recording rig using 2 steinberg mr816s connected through ADAT for extra inputs, and was planning on eventually expanding this setup and having it be my main rig, I'm having second thoughts now. I also would like a comment from the motu reps that come on gearslutz on whether the ridiculous compressor auto gain will ever be able to be disabled in a future update, it makes the compressor completely unusable, it is the worst implementation I have ever used. I spoke to a rep at namm, he said that he had issues working with the compressor as well, but that the person that designed it thought that it worked great and loved the compressor design. He said he would pass my comments along, hopefully the person that designed the compressor eventually gets the hint how bad it is and that many(probably almost all) of us hate the auto gain implementation. The hardware on these boxes is solid, it would be a shame to have these interfaces fall out of favor because of a couple show stoppers that could be fixed if priorities were put in the right place.

Last edited by duncansound; 25th January 2017 at 07:19 AM..
Old 25th January 2017
  #2964
0.625 ms latency over AVB. It's in the brochure.
Old 25th January 2017
  #2965
Gear Addict
AVB has a maximum fixed network latency of 2ms. That means it should take no more than 2ms for audio to travel from the network output of one interface to the network input of another networked interface. MOTU have reduced this to 30 samples meaning the slowest it will ever be at 48kHz is 0.625ms. At 96kHz is 0.3125. At 192kHZ is 0.15625ms. So if you send audio from your DAW to a 1248 connected by either Thunderbolt or USB, then send audio from the 1248 to any other networked AVB interface, audio from the analog outputs of those two interfaces will never be aligned. If you want the outputs of in this case to be aligned audio coming out from the 1248 has to be delayed in the very least to account for the 30 samples of network latency.
Now let pretend you have 3 networked AVB interfaces as follows. A 1248 AVB in the control room connected to an AVB switch in one Vocal booth 100 meters away. In the same Vocal Booth there is also the new 624 connected. Then you have another AVB switch in a Live Room connected to the AVB switch in the Vocal booth. Let us say the Live Room is 200 meters from the Control Room and 100 meters from the Vocal Booth. There is the new 624 connected to the AVB Switch in the Live Room

1248 Thunderbolt DAW (Control Room)----100 meter CAT6 CABLE----->AVB Switch with a 624 in the same room (Vocal Booth)-----100 meters CAT6 CABLE----->AVB Switch with a 624 in the same room (Live Room)

Now audio is sent from the DAW computer to the 1248 via Thunderbolt. The 1248 sends an output stream to the network. Both the interfaces in the Vocal Booth and Live Room, will receive the stream from the 1248 at the same time despite being connected at different distances away and through different numbers of AVB Switches. The analog output of the two 624s will be aligned (sample accurate to each other), however they will be delayed from the analog output of the 1248. Supposing we swap one of the 624 with a 16A. They (624 & 16 A) will receive the AVB streams from the 1248 in the Control Room at the same time but will their analog outputs of that stream be aligned? If the convertors are the same they should.

Regards
Enoch

Last edited by KimGitz; 25th January 2017 at 04:24 AM..
Old 25th January 2017
  #2966
Lives for gear
 

Audio travelling over Avb of course has an inherent delay, but it should be compensated for. In the documentation I have seen it mentioned that if you enable AVB streams on your interface, then the interfaces latency will increase by a small set amount, (I forgot the exact figure, probably 30 samples like mentioned above), so that audio on the local interface will be time aligned with the avb streams coming from other devices. If this is not the case I would like clarification from the motu reps, as the wording I read made it sound like the delay would be compensated for
Old 25th January 2017
  #2967
Gear Nut
 

If you use the same device like two Motu 8A's, would this be without the delay since the device is the same?
Old 25th January 2017
  #2968
Lives for gear
 

If you had one 8a locally connected through thunderbolt/usb, and one 8a connected to the other through the avb network, it sounds like under the current driver they would not be time aligned. Don't get me wrong, they SHOULD be time aligned, but it sounds like something is not working correctly in the current driver. I would think all the ad/da converters on all devices in the motu avb line would have the same latency, so that once the avb network latency is correctly compensated for in a new driver version, all inputs and outputs on all motu devices on an avb network will be time aligned to your local device. To clarify, I believe all devices on a motu avb network will already be syncronized/time aligned correctly now as far as the avb network itself is concerned, but the problem arises from having one local device that does not go through the avb network to communicate with the local usb/thunderbolt connected computer, it sounds like the current driver is not correctly compensating for this.

Last edited by duncansound; 25th January 2017 at 08:05 AM..
Old 25th January 2017
  #2969
Lives for gear
 
jlaws's Avatar
Yeah, I don't think anyone is complaining about the latency, since it's minimal. People are hoping (justifiably so) that the latency will be compensated for.
Old 25th January 2017
  #2970
Lives for gear
 

Enoch is correct.

Let's break it down..

AVB transport delay isn't the main problem with regards to alignment of multiple units. Similarly like other current networking protocols, this delay is fixed.
Generally AVB uses buffering and PTP presentation time to normalize transport latency from different endpoints.
Basically that pres. timestamp information, which is associated with every media packet, says when samples (in the case of audio transfers) were captured/generated in common time. Then there is fixed offset for presentation timestamps, which should be high enough to accommodate all the variations and forwarding delays in the network. Then target endpoint will hold this packet at its buffer, until the right pres. time is reached.
Standard uses 2 ms offset as default value for class A media packets, which was picked to cover 7 hops at 100Mbit network.. but of course, this offset can be shortened for the system with simpler topology and less hops.. MOTU have selected 625us as their default offset value. So media packets form all devices in AVB network should be aligned and presented at the same time at endpoint according to this offset.

However this is just one part of total latency, because it covers only AVB transport. You have to add also conversion and processing (if there is some DSP or routing) latency of the each particular source endpoint. If this additional delay will be the same for all devices (assuming that apply for most models from MOTU AVB lineup), then also audio from those network AVB devices will be aligned at main interface.

What seems to be so hot topic here, is alignment difference between all network devices and local analog I/Os at main interface, which naturally doesn't pass through network.
As I've commented before, although it might be nice to have an option for its compensation, it shouldn't be so limiting for normal use and to me it isn't as dramatic as it looks like. Unless one is spreading phase sensitive channel groups across multiple converters, which can be typically avoided, this likely won't be a problem.
Such offset is normal thing for any small setup with local I/Os and additional converter(s) hooked via ADAT, AES or say MADI.
Honestly as handy as it might be for I/O alignment, I haven't seen any audio interface with variable sample delays at its DSP mixer, except of MIOConsole by MH.

Michal
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