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 30th December 2016
  #2881
Gear Addict
 

anybody using more than one unit? if so, have timing problems? i just bought an 8A to expand my 16A. unfortunately, the two units together are not sample accurate as i would have hoped/expected them to be.
i.e. multed source signal from external preamp routed to input 1 on 16A and input 1 on 8A and then recorded = 8A recording being 23 samples behind the 16A. not good.

for reference. macbook pro i7, yosemite. thunderbolt to 16A, cat5 from 16A to 8A. 8A slaved via AVb to 16A. recorded into reaper 5.30
Old 30th December 2016
  #2882
Lives for gear
 
loopy's Avatar
 

Quote:
Originally Posted by jlaws View Post
As a result of your post, DAW Plus, I tried 64/16 and amazingly it performs much better than 64/48 (which was what i had tried before) in my DAWs. It can handle a fairly full mix at that buffer setting, which is quite amazing.
Based on DAW Plus and you I just did the same thing with my MOTU Ultralite AVB, 64 samples and 16 on the safety buffer and it's runs even better than 64/64 which is usually what I run.

I just finished playing around with Synthogy Ivory, full bore, all options turned on and playing some complicated music with all 10 fingers going and lot's of pedal action.
Was noodling for about 30 minutes straight and not a single click, or pop with Reaper reporting 1.8ms / 1.8 ms for total RTL of 3.6ms.

Ivory when really played tends to bring out the worst in driver performance
even though on my system it's on an SSD.

That is amazing performance.

FWIW on my system, Reaper's RTL is usually within 0.3 or 0.4 ms total RTL as compared to measuring with Oblique.

Funny thing is I never bothered to test 16 safety buffer because 48 gives a few pops and clicks under stress so I settled on 64.

Now 16 is my new friend
Old 30th December 2016
  #2883
Quote:
Originally Posted by d. gauss View Post
anybody using more than one unit? if so, have timing problems? i just bought an 8A to expand my 16A. unfortunately, the two units together are not sample accurate as i would have hoped/expected them to be.
i.e. multed source signal from external preamp routed to input 1 on 16A and input 1 on 8A and then recorded = 8A recording being 23 samples behind the 16A. not good.

for reference. macbook pro i7, yosemite. thunderbolt to 16A, cat5 from 16A to 8A. 8A slaved via AVb to 16A. recorded into reaper 5.30
I will be interested to see the outcome of this, as I own a 16A and was coveting the 8A.

have you contacted support?

cheers

Wiz
Old 30th December 2016
  #2884
tft
Gear Nut
 

Quote:
Originally Posted by d. gauss View Post
anybody using more than one unit? if so, have timing problems? i just bought an 8A to expand my 16A. unfortunately, the two units together are not sample accurate as i would have hoped/expected them to be.
i.e. multed source signal from external preamp routed to input 1 on 16A and input 1 on 8A and then recorded = 8A recording being 23 samples behind the 16A. not good.

for reference. macbook pro i7, yosemite. thunderbolt to 16A, cat5 from 16A to 8A. 8A slaved via AVb to 16A. recorded into reaper 5.30
two different devices connected in series or parallel (of any brand) will probably never be "in time" like you expected.
the important part is that they are "in sync"! so the the timing difference is constant all the time.
measure it (like you did) and then compensate for it after recording in your daw.
that's the technically correct way to get them "in time".

the other question is, if 23 samples is of any concern if it is different sources apart from each other. 1 ms is about 30 cm distance (sound wave in air) and 23 samples is about 0,5 ms in 44,1 khz, so that equates to 15 cm distance-change. now imagine two microphones in the same room, recording two different instruments a few meters from each other ... you won't hear any difference between compensated or not i guess.
that said, always connect "multi-mic'd" single instruments via the same interface and you should be safe (or compensate, if you have to use different interfaces for that).
Old 30th December 2016
  #2885
Gear Addict
 

23 samples is a big concern here for what we do and what we need.

do you have an intimate knowledge of the AVB network protocol spec, as well vast experience troubleshooting large scale, multi-room audio networked facilities? i.e. "AVB provides its own network-wide time base for better-than-sample-accurate phase lock across all connected devices."

Quote:
Originally Posted by tft View Post
two different devices connected in series or parallel (of any brand) will probably never be "in time" like you expected.
the important part is that they are "in sync"! so the the timing difference is constant all the time.
measure it (like you did) and then compensate for it after recording in your daw.
that's the technically correct way to get them "in time".

the other question is, if 23 samples is of any concern if it is different sources apart from each other. 1 ms is about 30 cm distance (sound wave in air) and 23 samples is about 0,5 ms in 44,1 khz, so that equates to 15 cm distance-change. now imagine two microphones in the same room, recording two different instruments a few meters from each other ... you won't hear any difference between compensated or not i guess.
that said, always connect "multi-mic'd" single instruments via the same interface and you should be safe (or compensate, if you have to use different interfaces for that).
Old 30th December 2016
  #2886
Quote:
Originally Posted by d. gauss View Post
anybody using more than one unit? if so, have timing problems? i just bought an 8A to expand my 16A. unfortunately, the two units together are not sample accurate as i would have hoped/expected them to be.
i.e. multed source signal from external preamp routed to input 1 on 16A and input 1 on 8A and then recorded = 8A recording being 23 samples behind the 16A. not good.

for reference. macbook pro i7, yosemite. thunderbolt to 16A, cat5 from 16A to 8A. 8A slaved via AVb to 16A. recorded into reaper 5.30
That seems like a wrong offset compensation, as the latency itself should be known by the system. Or the compensation is lacking. I would report this to the MOTU team. @mrmiller

Last edited by DAW PLUS; 30th December 2016 at 12:29 PM..
Old 30th December 2016
  #2887
Lives for gear
 

Quote:
Originally Posted by d. gauss View Post
23 samples is a big concern here for what we do and what we need.

do you have an intimate knowledge of the AVB network protocol spec, as well vast experience troubleshooting large scale, multi-room audio networked facilities? i.e. "AVB provides its own network-wide time base for better-than-sample-accurate phase lock across all connected devices."
You're mixing two different things, first is phase lock - eg. devices aren't drifting (that's absolutely crucial) and second thing is delay compensation.
What you're experiencing, is likely the second problem, local analog I/Os at 16A aren't delay compensated to AVB streams coming from 8A.
As Leon previously said, if that delay is known, then it should be technically possible to automatically compensate for that. Eg. all local analog I/Os will be delayed and aligned by AVB stream with highest latency.

If that's supported by firmware at 16A, I don't know, maybe MOTU support guys or mrmiller will expand about that.

Anyway, it's always possible to manually compensate for that in a DAW. Sometimes that's only way how to do that in setups with various converters (each has different delay.. it can be due to conversion chips, internal processing whatever), if you require perfectly aligned takes.
As you're using Reaper, this can be easily managed by adding Delay/time_adjustment JSFX to tracks from 16A and add additional 23 samples there.

Michal
Old 30th December 2016
  #2888
Gear Addict
 

Quote:
Originally Posted by msmucr View Post
You're mixing two different things, first is phase lock - eg. devices aren't drifting (that's absolutely crucial) and second thing is delay compensation.
What you're experiencing, is likely the second problem, local analog I/Os at 16A aren't delay compensated to AVB streams coming from 8A.
As Leon previously said, if that delay is known, then it should be technically possible to automatically compensate for that. Eg. all local analog I/Os will be delayed and aligned by AVB stream with highest latency.

If that's supported by firmware at 16A, I don't know, maybe MOTU support guys or mrmiller will expand about that.

Anyway, it's always possible to manually compensate for that in a DAW. Sometimes that's only way how to do that in setups with various converters (each has different delay.. it can be due to conversion chips, internal processing whatever), if you require perfectly aligned takes.
As you're using Reaper, this can be easily managed by adding Delay/time_adjustment JSFX to tracks from 16A and add additional 23 samples there.

Michal
i require perfectly aligned takes of 24 to 32 tracks at a time with no after the fact futzing. in this day and age, that should be expected, and if two linked devices cannot reliably provide that, then they should not be on the market until they can. ancient tech like RADAR and HD24 can do this with ease.

so far motu has not responded on the subject.

i suggest some of you take a look here (a more detailed explanation of the motu problems written by someone else) to see what i am talking about in further detail:
MOTU's New Leaps with Lowering Latency in Music Industry News and Views Forum
Old 30th December 2016
  #2889
Lives for gear
 

Quote:
Originally Posted by d. gauss View Post
i require perfectly aligned takes of 24 to 32 tracks at a time with no after the fact futzing. in this day and age, that should be expected, and if two linked devices cannot reliably provide that, then they should not be on the market until they can. ancient tech like RADAR and HD24 can do this with ease.
I see.
I'm just try to explain you technically, what's going on, when using several converters and each is connected in a different way, so each "path" has different latency. As I've said, it should be possible to compensate that internally by delaying I/Os from the first box.

Ancient or not - HD24 or RADAR is a different story, because it's complete system, where all analog I/Os are part of the one box. If one is using multiple HD24s just as an ADAT converters, then it also different topology, because all the HD24s has the same delay for a give sample rate, so naturally it won't be affected by this problem.

Quote:
so far motu has not responded on the subject.
Hopefully they'll address that or at least reply to you.

Quote:
i suggest some of you take a look here (a more detailed explanation of the motu problems written by someone else) to see what i am talking about in further detail:
MOTU's New Leaps with Lowering Latency in Music Industry News and Views Forum
O.K. it was apparent from your previous description, what problem it is.
There are two uncompensated delays at his post.
First one is pretty normal for most of native systems. Any DAW rely on information about input and output delay, which is reported by audio interface driver. That is being used for compensation of roundtrip delay by automatic adjustment of recorded items position.
Typically, if one requires perfect alignment of source and recorded takes, then it is necessary to manually adjust that for the particular setup, even if one interface without any expansion is being used.
At Reaper, you have those offset fields at Preferences/Audio/Recording at the bottom of the settings pane.

Second delay, is the one, which you've mentioned - eg. different latency of AVB and local analog I/Os. But apparently that one is consistent and the same for the second and third AVB connected box.

As I've said, as a temporary solution, you can use that time_adjustment plugin. Possibly, you can try to insert that into tracks from 16A as an input FX, that way, effect should be directly applied to the recorded takes.
I can imagine, if that will be working, you can prepare such input track with all settings and this delay inserted, it can be saved as a Track Template (right click at TCP at Reaper).

Michal
Old 30th December 2016
  #2890
Gear Addict
 

Quote:
Originally Posted by msmucr View Post
I see.
But apparently that one is consistent and the same for the second and third AVB connected box.
no it is not. please reread the linked post. <<All of my tests have shown random and variable subsample offsets, making it impossible to use the 16A in a parallel processing configuration. >>
Old 30th December 2016
  #2891
Lives for gear
 

Quote:
Originally Posted by d. gauss View Post
no it is not. please reread the linked post. <<All of my tests have shown random and variable subsample offsets, making it impossible to use the 16A in a parallel processing configuration. >>
You're true, I've just guessed that by the first picture with misaligned sine wave takes.. there is one offset for the TB connected 16A roundtrip and the second additional offset to another units connected via AVB.
Anyway, that varying subsample offsets according to the clocking configuration, which he reported are much worse issue IMO. I'm just curious, if that will be better, if he would use wordclock from main unit instead of AVB to carry the clocks.

Anyway, you've said, your measured offset is 23 samples and not varying. So workaround could be applied maybe.
And as I've mentioned, maybe MOTU support will reply after holidays.

Michal
Old 30th December 2016
  #2892
Gear Addict
 

Quote:
Originally Posted by msmucr View Post
I've just guessed that by the first picture with misaligned sine wave takes..

Michal
if we could refrain from guessing, and someone who actually has more than one motu AVB unit please comment, it would be much more helpful. thanks.
Old 30th December 2016
  #2893
Quote:
Originally Posted by d. gauss View Post
anybody using more than one unit? if so, have timing problems? i just bought an 8A to expand my 16A. unfortunately, the two units together are not sample accurate as i would have hoped/expected them to be.
I have more than one unit and never experienced this problem. I use ADAT converters in my setup too and the avb devices are in perfect sync, even if you send signals over a 50 m ethernet cable. Did you select one device as main clock?
Old 30th December 2016
  #2894
Lives for gear
 

Quote:
Originally Posted by d. gauss View Post
if we could refrain from guessing, and someone who actually has more than one motu AVB unit please comment, it would be much more helpful. thanks.
o.k. sorry for trying to be helpful. Good luck with your rig.

Michal
Old 30th December 2016
  #2895
Gear Addict
 

Quote:
Originally Posted by Bjoern Bojahr View Post
I have more than one unit and never experienced this problem. I use ADAT converters in my setup too and the avb devices are in perfect sync, even if you send signals over a 50 m ethernet cable. Did you select one device as main clock?
yes, 16A is master device. 8A following it. MOTU has responded. in short, according to them, the analog inputs on the 16A and the AVB signals from the slaved 8A all have to be routed through the MOTU mixer rather than just assigned to computer ins. i.e., assigned to the "post FX out" tabs in the mixer app. weird, non-intuitive (i never use any input FX processing), but if that's what they want, so be it. HOWEVER, doing so now still results in sync being off. this result was confirmed by MOTU.

Last edited by d. gauss; 31st December 2016 at 08:54 PM..
Old 31st December 2016
  #2896
Gear Addict
 

So you are saying a multed signal sent simultaneously to an input on each of your different avb connected devices and each recorded on a separate track results in a total null when phase flipped against each other? You can confirm this?

After numerous tests, my 8A consistently arrives 23 samples later than the 16A when slaved to it. If the avb mixer had a sample delay in it, all would be golden.

Even more interesting/frustrating is that while the relationship between the 2 devices seems to stay steady at a 23 sample difference, the overall RTL in relation to the source signal varies after reboots. i.e. a 48 sample manual offset in the DAW to achieve a null between the source signal and the 16A is no longer valid after a system reboot. Can be anywhere from 48 to 52 so far. (meanwhile, the 8A remains consistently 23 samples out of sync behind that number). Not good if you plan to do parallel outboard processing recalls.


Quote:
Originally Posted by Bjoern Bojahr View Post
I have more than one unit and never experienced this problem. I use ADAT converters in my setup too and the avb devices are in perfect sync, even if you send signals over a 50 m ethernet cable. Did you select one device as main clock?
Old 1st January 2017
  #2897
Quote:
Originally Posted by d. gauss View Post
After numerous tests, my 8A consistently arrives 23 samples later than the 16A when slaved to it. If the avb mixer had a sample delay in it, all would be golden.
That would really be nice. A sample delay would be a cool addition to the avb-mixer!

Currently I use the small ultralite as my main interface (it's clock master) and connect a 8pre to the ADAT input at the back of the interface. I used the ultralite mic-input as reference for my measurements at 44.1 kHz.

The 8pre is directly connected to the interface (no avb used, just ADAT), and I get a constant sample delay of 37 samples / 0,83 ms compared to the internal mic-input of the ultralite. Whenever I use a StageB16 with avb-connection to the ultralite this one has a constant sample delay of 25 samples / 0,56 ms compared to the internal mic-input of the ultralite.

These delay values remains the same, even if I use the PostFX-Routing in the mixer or disconnect avb and use only the adat input.

I thought, this delay is something you always need to adjust in any digital system with different a/d converters (or even different board-revisions of the same converters...). I had the same "issues" years ago with the combination of 3 different adat-converters in my old setup. There was always a small amount of latency between them, so I think this is no problem caused by the avb network, it might just be a difference between the converters? The preamps of the 8pre, ultralite and stage b16 differ in their technical specs and sound, so their converters might also be different designs?
Old 2nd January 2017
  #2898
Lives for gear
 
jlaws's Avatar
Sample delay would be great. If anything it would allow time alignment for subs within the mixer, too.
Old 2nd January 2017
  #2899
Here for the gear
 

Hey everyone, I'm on windows 7 and I'm experiencing real issues with my 1248. I have constant sound drop out, With Reaper, Pro Tools and even Chrome and windows multimedia.

I've talked to MOTU people for a while but they don't answer my messages anymore.

Is somebody has experienced the same problems/has a solution?

Right now, my 1248 is useless. I'm thinking of buying a Mac...
Old 2nd January 2017
  #2900
Quote:
Originally Posted by pifpaf View Post
Hey everyone, I'm on windows 7 and I'm experiencing real issues with my 1248. I have constant sound drop out, With Reaper, Pro Tools and even Chrome and windows multimedia.
AFAIK there is an issue with Chrome.
But what are your system specs? USB or TB connection? Anti Virus active? Did you tweak your system according to the AVID suggestions?

Quote:
I've talked to MOTU people for a while but they don't answer my messages anymore.
Well, we just passed the holidays, so I imagine their hotline was not occupied for a while.

Quote:
Is somebody has experienced the same problems/has a solution?
No, actually the 1248 offers the best results performance wise atm from all the interfaces I've tested. Try a buffer setting of 64/16.

Quote:
Right now, my 1248 is useless. I'm thinking of buying a Mac...
That will probably fix the issue. So does buying a good Windows system. Or fixing the configuration issue you seem to have now.
Old 2nd January 2017
  #2901
Here for the gear
 

Hello,

My anti virus is on , I didn't follow the avid suggestions but everything was fine with my previous SC a Roland Octacapture as well with my friends MERGING Horus and RME UC, so I will incriminate MOTU drivers before my computer.

Now that the holidays are over I hope to get more help from MOTU.

Quote:
No, actually the 1248 offers the best results performance wise atm from all the interfaces I've tested. Try a buffer setting of 64/16.
The 1248 is an excellent SC but there is still problem related to the windows drivers, not my computer. You may haven't experienced troubles, I doesn't mean that nobody has experienced troubles. I've tried a lot of different settings. Anyway, all settings should work without drop outs. If the buffer is too low, you hear cracks in the sound but it should still work. And 128 is not too low.

Quote:
That will probably fix the issue. So does buying a good Windows system. Or fixing the configuration issue you seem to have now.
I have a good windows system. Like I said i've tried 3 other brands before without experiencing any of these problems.
Old 2nd January 2017
  #2902
Lives for gear
 
Hokut's Avatar
 

Hi,
every time I set Logic project to 96Khz and the 1248 + 16A automatically change to 96Khz I hear a high pitched sound coming out of my speakers.
Any idea what this could be? It's like a high pitched hum of some sort
Old 2nd January 2017
  #2903
Quote:
Originally Posted by pifpaf View Post
Hello,
My anti virus is on , I didn't follow the avid suggestions but everything was fine with my previous SC a Roland Octacapture as well with my friends MERGING Horus and RME UC, so I will incriminate MOTU drivers before my computer.
Understandibly. However, it could be that the Antivirus interferes with the MOTU contrary to the other interfaces, as it is arbitrary towards different applications and processes.

Quote:
The 1248 is an excellent SC but there is still problem related to the windows drivers, not my computer. You may haven't experienced troubles, I doesn't mean that nobody has experienced troubles. I've tried a lot of different settings. Anyway, all settings should work without drop outs. If the buffer is too low, you hear cracks in the sound but it should still work. And 128 is not too low.
Well, the fact that it runs really well here (with certain buffer settings, not all) shows that the drivers are capable of running well and are not bad by default. So there must be another factor involved why they are underperforming on your system.
Quote:
I have a good windows system. Like I said i've tried 3 other brands before without experiencing any of these problems.
When you state you have a "constant sound drop out", does it mean no sound at all or dropouts in the stream, i.e. crackles interfering the sound continuously?
What buffer settings do you use?
The weird thing is that Reaper and PT use the ASIO driver and Chrome and Windows Multimedia use the WDM driver, which are completely unrelated. So I actually suspect it is the controller they are attached to, which might be a compatibility issue as it runs with other USB interfaces.
Have you tried a USB3 port instead of USB2, or vice versa?
Old 2nd January 2017
  #2904
Here for the gear
 

Thanks for answering. Unfortunately I have only usb 3 on my computer. I have tried different usb without success. The sounds totally cuts un till I close the current program then I can hear a little tick in the speakers and the sound is coming back. I mean gonna try to disable the AV. I'm leaving on holidays. I let you know next week.
Old 3rd January 2017
  #2905
Gear Nut
Quote:
Originally Posted by Hokut View Post
Hi,
every time I set Logic project to 96Khz and the 1248 + 16A automatically change to 96Khz I hear a high pitched sound coming out of my speakers.
Any idea what this could be? It's like a high pitched hum of some sort
Do you have any plugins running?
Old 3rd January 2017
  #2906
Gear Nut
Quote:
Originally Posted by pifpaf View Post
Hey everyone, I'm on windows 7 and I'm experiencing real issues with my 1248. I have constant sound drop out, With Reaper, Pro Tools and even Chrome and windows multimedia.

I've talked to MOTU people for a while but they don't answer my messages anymore.

Is somebody has experienced the same problems/has a solution?

Right now, my 1248 is useless. I'm thinking of buying a Mac...
Hey!

For the record with windows 10 my Motu 1248 and 24ao work great (aside from with Bitwig where I get dropouts which I assume is related to bitwig since reaper works great) ! so maybe you can upgrade to 10 if your hardware supports it?

I do sometimes have dropouts with Chrome. These are normally fixed by disconnecting (while playing audio on chrome) and reconnecting the thunderbolt or usb port.

Also I don't think this applies to USB but with thunderbolt you must make sure it set to output 1-2 and not 1-24 in the sound settings. Furthermore, you should make sure the buffer settings are not too low/ too high on the motu web app, depending on your harware
Old 3rd January 2017
  #2907
Lives for gear
 
Hokut's Avatar
 

Quote:
Originally Posted by hellofishy View Post
Do you have any plugins running?
no, empty project. Was setting up a new project to work at 96Khz and got the high pitch sound coming from the speakers (Adam A7s)
Old 4th January 2017
  #2908
Here for the gear
Quote:
Originally Posted by pifpaf View Post
Hey everyone, I'm on windows 7 and I'm experiencing real issues with my 1248. I have constant sound drop out, With Reaper, Pro Tools and even Chrome and windows multimedia.

I've talked to MOTU people for a while but they don't answer my messages anymore.

Is somebody has experienced the same problems/has a solution?

Right now, my 1248 is useless. I'm thinking of buying a Mac...
I have had the same issue.
I have tried the new driver on a Windows 7 Dell Vostro Laptop, which I then upgraded to W10, a 2013 MBP with W10, and a 2012 Mac Mini with W10 and it hasn't worked on any of them. I also tried to use Thunderbolt with W10 on Bootcamp and the system wouldn't even boot.

It works perfectly under OSX, but just not on Windows.

Uninstalling the driver and installing the older (V1.6?) seems to fix the problem.
Its worth noting that V1.6 also installs an extra MOTU USB application within Windows which is used to change the buffer settings... The buffer/latency settings in the AVB control panel (in the web browser) have no effect when running the older driver.

The latency isn't as good under Windows as OSX but I'm guessing thats down to USB vs Thunderbolt and my underpowered Windows Laptop, but since downgrading to the older driver everything seems to be stable.
Old 4th January 2017
  #2909
Here for the gear
 

Do you know where to find this old driver? Could you PM it to me?
Old 4th January 2017
  #2910
Lives for gear
 
loopy's Avatar
 

Quote:
Originally Posted by pifpaf View Post
Do you know where to find this old driver? Could you PM it to me?
Here is where the all the drivers, including older versions can be downloaded.

MOTU.com - Driver Download Links and Change Log - MOTU Hardware
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