MOTUNATION (formerly UnicorNation) is an independent community for discussing Digital Performer and other MOTU audio software and hardware. It is not affiliated with MOTU.
monkey man wrote:Now waiting to see if Apple offers a more traditional form-factor in the next Mac Pro (I figure something's due now). If it does, then perhaps I'll take the AVB plunge; in that case the only obstacles would be DHP and all the peripheral nonsense, what with shifting' all my drives and so on out of the box and redesigning my beautiful wooden rack...
I have similar issues/concerns. But I doubt we will see any sort of return to a traditional form factor on Mac Pros in the near future. Never say never though.
I hear you, Billy. I'll never say never either; the only folks we'll end up hurting by doing so will / could be ourselves. Sounds like a lousy deal, really.
mikehalloran wrote:
f you're angling that USB3 could "replace" the PCIe cards to extend the life of the AudioWire line (something I'm still hoping for), I agree that it could. I'd be wary of the switch from the 424 card to USB though:
Ahhh... that wasn't where I was going. I was referring to PCIe cards for USB 3 to get you some of that speed you are looking for.
Thank you, Mike. It's not speed I'm looking for. Rather, it's the ability to "link" or "marry" multiple interfaces so that I can monitor all of them from a single unit. The AudioWire system has always offered this. The newer ones, because I'd need at least 3 or 4 of them, would require TB to do so as you can only link 2 via USB. Further to that, only USB 2 is offered, so the I/O count is more or less what one could expect from a single unit only. I addressed this here:
Monkey Nuts wrote:The 424 card allows all I/O to fly around between interfaces at near-zero latency. This is how I use CueMix (to monitor all MIDI hardware outs) ... because the latter (PCI-424 card) seemed to "directly wire" all that I/O, hence my use of the term "marry".
The AVB network allows units to function as a single functional unit, just slightly more routing to configure. The downside is that's only going to happen with TB (I stop short of adding AVB direct, not having looked into it - has it's own limitations).
I don't know enough to say why USB3 was rejected for AVB use. Perhaps there's a technical limitation with it's use for AVB linking.
USB2 leaves a gap between full blown AVB/TB track counts and less than single interface track counts. There are countless people wanting full 32i/o over USB2, which seems like a real oversight. I'd have elected for USB3 for that reason alone. Is TB coming on line for PC users? I won't guess. I haven't paid attention to PC/direct AVB possibilities, at a glance looks like a 'no'.
The newest AVB interface (112D) is a router giving MADI, AES/EBU, ADAT, TB, and USB2 options.
It's all more expensive than the old PCIe line to implement, also sounds a lot better.
Doug Williams
Electromagnetic Radiation Recorders
The Martha Bassett Show broadcast mixer
Tape Op issue 73
DP 11.31
Studio M1 Max OS12.7.3
MOTU 16A and Monitor 8
M1 Pro MBP for remotes and editing
Then there's USB Type C, which my lil' brother reckons is on the way. Can't help but wonder if MOTU would possibly update the line when an alternative to TB takes hold.
I'm guessing, as you alluded to, Doug, that USB3 had limitations that made it unviable (probably timing issues, no?), but I still can't help but wonder if this would've been a problem in the simple 2-unit setups that USB2 use is limited to. The full channel counts would've been there, theoretically.
You'd think the next-gen USB flavour would be able to accommodate an AVB setup. If, unlike in the case of TB, this new-fangled protocol could be added to non-TB Macs (and PCs) via a PCI card, folks such as myself would be all over it! It's almost as if MOTU jumped in a little too early. No harm done; sales are there in the meantime, but a willingness to update the line to keep pace with impending alternative connection protocols must, IMHO, manifest in the interest of the long-term viability of the line.
As I said, I'll be all over AVB the second I can do it sans TB. Taking the plunge on my current Mac now and upgrading the CPUs; the wait for Apple and / or MOTU to conjure up something that'll rescue me became unbearable months ago. In light of this, as much as it pains me, I'll have to live with the "comparatively-sub-standard" conversion of the 24I/Os until alternatives to TB "emerge".
Geez, that hurts. I mean, it really, really hurts.
I've recently picked up a bunch of cheap 24i/os from you upgraders, and am presently pretty happy...but if I could just connect the 2nd port on my Mac Pro to an AVB-ready switch to get the audio streams that way, it should have plenty of bandwidth and be able to manage low latency without the latency inconsistencies that show up with USB2.
I'd like to hear from someone who's actually tried it first, though...
Wohali, firstly the only way to use ethernet is to install an AVB-capable card (PCIe). The standard ports on pre-TB macs won't cut it, and IIRC anything prior to Yosemite won't either.
The latency, whilst I (and IIRC Doug) can't remember the exact figures, is comparatively huge when placed next to either your system (I call it AudioWire™) or the new TB-connected AVB one. I'm talking unworkably huge, perhaps even half a second!
Yes, the tables on MOTU's site suggest it's fairly low... ridiculously low, in fact. That's over the network, from an interface's input, through MOTU's AVB box to another unit's outs. This has nothing to do with connection to, say, a Mac via ethernet. Therein lies the catch. This is why I'm waiting for news on this front. If an e-net connection can approximate a TB one, even if it's not quite as zippy, I'll be a very happy camper and many folks who've not TB-capable Macs and PCs will be able to jump onboard too.
EDIT: I just noticed that you mentioned an AVB-capable switch. Might work into a standard e-net port. Dunno, as I know nothing about networks. The latency limitation, at the very least, will remain 'though, as will the Yosemite one, I think...
Some hard data will be good to know. My background is in low-level programming, including network design. OSX and UNIX are definitely capable of sub-millisecond processing of Ethernet packets, so there shouldn't be an OS or hardware limitation here.
You're right that for best performance, you'd need a NIC capable of 802.1Qav (AVB), and that the standard ports on a Mac wouldn't speak that without special support. Assuming drivers are available, Newegg sells an Intel AVB-capable card (the I210-T1) for about $70...cheaper than a PCIe-424 for sure
The alternative is that one of the MOTU devices learns how to stream non-AVB Ethernet to a Mac or PC, which I gather is what they've done with the latest firmware/driver revisions. It's disappointing that this introduces ~500ms of latency!
Not meaning to thread hijack, but With all the additional headaches OSX 10.10 is giving me, this AVB switchover may be what forces me onto Windows after ~27 years of running Performer/DP on a Mac...if native AVB streaming support to a PC can be sorted out somehow.
Joan, if anyone feels your pain, it is I! Sorry to hear that. You're not hijacking at all IMHO; it's entirely relevant.
Yes, a PCIe AVB-capable card, connected to the MOTU router which accommodates the AVB interfaces would be perfect for me. I'd only have to ditch the 424 card and 24I/Os and replace all with the MOTU gear and PCIe card. This is what I'm hoping will become feasible.
Still living in hope. I wonder - could you wait too as I've chosen to do?
Yup, I'll probably hold off for a bit longer. The 24I/Os still work fine, as do the other interfaces, and they're getting cheaper on eBay. I might even try my hand at improving the DACs and clock sources in them. We'll see.
Hopefully we'll be rewarded for our MOTU loyalty in the long run; I have to assume we're not alone. There must be a shedload of folks out there sans TB who simply don't have the means to overhaul their systems to the extent required when upgrading to Darth Vader's helmet...
New firmware update (1.2.0) shipping today for all AVB interfaces. This update, as promised, supports higher channel counts via USB. On both OS X and Windows, we now support up to 64 channels over USB at 44.1/48khz and up to 32 channels at 88.1/96khz. As always, let me know if you run into any issues and I'll do my best to help sort it out.
Now, if the 64 channels could be asymmetrically assigned, I could ditch a small amount of gear in order to reduce my input-count requirements somewhat, and use, say, 56 of 'em as inputs and the rest as monitoring-out stems to my desk.
I'd be able to dive into the AVB thing, completely replacing the AudioWire system.
So close I can smell it... and yet still so far away! Doh and double doh! Aarrgghh!
Hang on... I thought this 64 channels was a total I/O count and that one was limited to 32 in each direction, an update (via a more efficient driver) from the previous limitation of 24 ins and 24 outs.
According to the MOTU site, it's 64 both ways:
Q: Can I use MOTU AVB interfaces on my older Mac that is running OS 10.6 or 10.7 and does not have Thunderbolt ports?
A: Yes (although this is not officially supported. Official support requires 10.8 or later.) USB can be used to connect a Mac to the AVB system for up to 64 channels of input and 64 channels of output (at 44.1 or 48 kHz). Because MOTU AVB interfaces are USB audio class compliant, no driver installation is necessary under Mac OS X 10.6 or 10.7. As soon as you plug in the USB cable and switch on the interface, the Mac "sees" all the audio channels and makes them available to your DAW or other Mac audio software. Web app control (for access to device settings, the on-board mixer, etc.) can be accomplished through a direct Ethernet connection from the network port on the interface to the Mac's network port. The Mac accesses the web app (which is being served directly from the interface itself) through the network connection.
Please, tell me I'm not dreaming. Sure, it wouldn't be the 96I/O I'm set up for, but with judicious parsing of my MIDI outfit, I think I could make 64I/O work. It'd be a squeeze, but still possible.
My only fear would be ditching the ol' faithful AudioWire system for... uugghh... USB2.
Is this the same as USB 3.1? If so, it is on its way. Apple announced the first Mac to support it recently.
Actually, Apple has little to do with it. It's Intel. But they are finally releasing the chip sets necessary. Don't let the ".1" fool you. USB 3.1 is a huge upgrade in the spec.
DP 11.31; 828mkII FW, micro lite, M4, MTP/AV USB Firmware 2.0.1 2023 Mac Studio M2 8TB, 192GB RAM, OS Sonoma 14.4.1, USB4 8TB external, M-Audio AIR 192|14, Mackie ProFxv3 6/10/12; 2012 MBPs Catalina, Mojave IK-NI-Izotope-PSP-Garritan-Antares, LogicPro X, Finale 27.4, Dorico 5.2, Notion 6, Overture 5, TwistedWave, DSP-Q 5, SmartScore64 Pro, Toast 20 Pro
