.

re:  the US 20x20.
 
Be sure and check the sync with whatever you plan on using.  TASCAM can be finicky between the choices for sync you have.  "Automatic" can be anything but.

Caa2
Are people actually able to track 16 to 24 analog inputs at once, without glicthes, at 44.1kHz with a 64 sample buffer over USB?

When I do live recording of a band, I use highest available sample buffer and use through interface processing for EQ, compression and reverb.

Why risk it?

When tracking at home, I love all my VSTs and synths, so I go through sonar.

RME Babyface Pro
 
 
just bought one about a month ago.
 
i was going to get the motu avb,
but i got a deal on this one, and just went with it.
 
it sounds amazing to me.
 
 
and the thing that really blew my mind,
was that i hooked it up to my old win xp machine, without any software installed,
and it worked fine.
 
LOL
 
no sh!t.
 
it only has 2 analog preamp inputs, which is fine for me,
but has a ton of other connectivity.
 
the preamps are super neutral and true.
 
the control of the thing is complete, using the totalmix panel, that's really sweet.
 
the latency is just about nil.
 
the internal fx (delay and reverb) sound good enough to print.
 
i'm really impressed with it,
and the tracks i'm recording with it now, are easier to mix, easier to hear the freqs, i don't know, the captures are just so much more 'true' to my ears, it makes me realize just how much color my older interfaces were adding that i did not want.
 
YMMV

.

Caa2
Yes, I can explain.
 
I have been round tripping monitor mixes through SONAR using the actual FX I use in the final mix for nearly a decade. I will consider it a step backwards to revert to proxy mixing the monitor sends while tracking an ensemble.
 
In addition, I haven't had to set my buffer above 64bit for the past 4 years while using SONAR, which was the last time I upgraded my DAW CPU system. This makes monitoring while overdubbing seem just as familiar, so it will seem like a step backwards if I find I have to bump up the buffer size to get the newest gear to work well.
 
This is why I am asking if people actually do get to use all the throughput that the hardware, and its driver, purports to offer or if there is some sort of compromise that people are accepting and simply not speaking about.
 
Maybe you can explain; Do you actually track 16 to 24 channels of analog to digital conversion at the 64 sample buffer setting or is your first hand experience based upon using buffers that do not matter at all?
 
Thank you

OK. I understand. so far my experience from using relatively safe mid to large size buffers.
 
I could give it a try as it would make an interesting "load" test of my system now that I have just switched to the MOTUs, but I would need more info regarding what you usually do:
 
What sort of FX do you use during tracking (which kind? which brand? how many tracks?) how many submixes from these tracks? ...
 
but let's maybe stick the discussion to the thread that you opened on that subject ...
http://forum.cakewalk.com/MOTU-16A-on-a-Win7-x64-Machine-m3492494.aspx#3493114

.

Caa2
 I was tempted by the MOTU 16A from the moment it was announced but have been disappointed that Thunderbolt has taken so long to be adopted for Windows.
 I have been wondering if the USB option will provide me with the same performance I currently have with my 896HD set up.
 

MOTU is currently the only company to have true "PCIe via Thunderbolt" drivers currently available for Windows.
If you have all the following in place, you're good-to-go:

• Z170x or X99p motherboard that supports Thunderbolt-3 via USB-C port
• You need a Thunderbolt-3 (USC-C) to Thunderbolt adapter (~$84)
• You have to be running Win10

 
MOTU AVB's performance when connected via USB-2 is excellent (sub 5ms).
Thunderbolt performance in this case is PCIe level... so it'll allow slightly lower round-trip latency.
 
 

Caa2
I have been round tripping monitor mixes through SONAR using the actual FX I use in the final mix for nearly a decade. I will consider it a step backwards to revert to proxy mixing the monitor sends while tracking an ensemble.
 
In addition, I haven't had to set my buffer above 64bit for the past 4 years while using SONAR, which was the last time I upgraded my DAW CPU system. This makes monitoring while overdubbing seem just as familiar, so it will seem like a step backwards if I find I have to bump up the buffer size to get the newest gear to work well.
 
This is why I am asking if people actually do get to use all the throughput that the hardware, and its driver, purports to offer or if there is some sort of compromise that people are accepting and simply not speaking about.
 
Maybe you can explain; Do you actually track 16 to 24 channels of analog to digital conversion at the 64 sample buffer setting or is your first hand experience based upon using buffers that do not matter at all?
 

 

Caa2
 I often times have an instance of FabFilter Q2 EQ on each of the 16 tracks. A few tracks will have Fabfilter C2 compression. I use a couple of instances of Valhalla Room (one on a mains mix and maybe extras as effects on vocals or guitars.) I'll sometimes add a few modulation "effects" to tracks where the guitar or keyboard player wants them so I can record "dry" and tweak the "Effects" later. I run a few Fabfilter limiters on the outputs but make sure they are not set to look ahead mode. I do this all in SONAR. I transitioned from mixing monitors with analog hardware to doing it in SONAR. Although I have some experience with, and am aware of, CueMix I have never really had a need for it.
 
 In the past when I stress tested my system I simply hooked up 16 SM58s, turned on the radio, hit record and came back in an hour or so to check for drop outs.

OK - I just put it thru the test and it all worked fine.

I think I even got beyond your specs ... I created a quick MIDI arrangement, send it to multiple hardware modules all connected via analog inputs to the audio interfaces, sent 4 signals back out for re-amping and back in, plus 4 HP mixes, plus several mics in the room ...

INPUT via USB:
+ 2x stereo from hardware synth, feeding analog inputs @ MOTU #1
+ 4x stereo from 2nd hardware synth module, feeding analog inputs @ MOTU #2
+ 4x mono from re-amp via outboard gear
+ 4x mics recording sound from control room monitors via ADAT interface hooked to MOTU #1
=> total of used USB input channels: 20

OUTPUT via USB
+ 1x stereo to monitors
+ 4x stereo headphone mixes from Sonar buses
+ 4x mono output feeding outboard gear
=> total of used USB output channels: 14

FX in Sonar:
+ ProChannel EQ ................ across all tracks (17) and all buses (11) for low cut
+ ProChannel compressor ........ across all tracks (17) and all buses (11) just to test / increase load
+ FabFilter Pro-EQ2 ............ across all tracks (17) and all buses (11)
+ FabFilter Pro-C compressor ... across all tracks (17) and all buses (11) 

Adding the FabFilter Pro-L limiter is pushing it: I can strip 4 across buses with no problems, using 5 or 6 instances causes occasional slight crackling (disabling the ProChannel from the buses helps and lets me get to about 7 - 8 instances of Pro-L) ... but just occasional crackles, no audio drop outs or audio engine stops. when you push it, it crackles in the HP mix and when you back off with the plug-in use and you are fine.

I stopped the test after 40 min because I got tired of the cheesy tune I had programmed as endless 32 measures loop (simple groove clip stretched to eternity).

The recording was done @96 kHz, using 128 sample buffer + 48 samples host safety offset (which would be 64+24 buffer size @ 44.1 kHz), translating to a reported RTL of 3.7 ms (4.5 ms RTL measured).


extra features tested:
+ laptop hooked up to USB port of MOTU #2 => recorded some of the tracks for test (could have recorded all as a full backup), which yields a perfectly clocked sample accurate recording on a 2nd DAW


EDIT: 2x MOTU 1248 connected via LAN cable (AVB networking)

.