MOTU 24I/O System to MOTU AVB Upgrade on 2012 Mac Pro

[monkey man]

Been following the BeerGutz thread for 6 months but feel none the wiser for it. I don’t have the confidence or patience to post there and so have been hoping someone would ask similar-enough questions to mine, but so far no dice.

The requirement is:

I need to be able to replace my MOTU 3 x 24I/O system with an equivalent MOTU AVB one (I blame the new converters and nothing else for this). CueMix currently allows me to monitor all my hardware MIDI gear (over 60 inputs) for near-zero latency overdubbing and MIDI tracking. It also allows me to bring drums, bass and the rest of the input signals into my little Mackie Onyx on 3 discreet (mono and stereo) channels, thus providing a convenient means by which I can quickly balance the most useful (aforementioned) categories for said overdubbing. Crank the drums and bass for bass tracking, the drums for drums and so on.

The questions are:

1) Is the AVB system capable of providing all these inputs (plus guitar, bass, vox etc.) to me for monitoring the way CueMix does? It appears from the literature I’ve seen that each unit sports a 48ch mixer, but as I mentioned, I need upwards of 60 channels to be routed to the Mackie in several groups (along with DP’s main outs, of course, also to a discreet pair which occupies a stereo channel on the Mackie). I don’t care about all the fancy effects on offer; volume and pan per input will suffice.

2) Is it even possible to run this system on my 2012 Mac Pro? I saved for 6 years and upgraded to the last desktop (non-trash can model), buying in at 2.4GHz with a view to a future CPU boost. HCMarkus’ efforts to keep us up to date in this regard have helped this all come to a head for me now as I try to balance waiting for an announcement from Apple (on where it’s going with the Mac Pro range) with keeping and upgrading my Mac Pro (before the X5680s run out/become expensive on the s/hand market), and of course the MOTU AVB upgrade.

3) If not, is there a way to add thunderbolt to it?

4) If TB simply cannot be accommodated on this machine, would an AVB-capable ethernet card allow the performance I require?

There you go. This stuff has been the subject of many a nightmare (and daystallion) over a period of many months. I’ve desperately wanted to rejoin my Unicorn buddies and ask y’all what my options are, but as many of you are probably aware, confidence, energy and indeed health have been at a premium orders of magnitude beyond that which I’ve been able to afford. As I said, it’s all come to a head and I feel compelled to consult the most eminently intelligent, kind, understanding, honest and reliable folks I know, so it’s off to BeeGutz and I’ll catch y’all ‘round the traps.

NOT!

Thank you all so very much from the bottom, middle and upper regions of my heart. Heck, I’m already grateful and you haven’t even uttered a neigh yet.

Ready… set… NEIGH!!!

[EMRR]

Short answers from my phone:

Look at AVB sampling rate limitations and OS requirements. You can't get all rates currently on Yosemite. AVB Direct connection was giving higher latency initially, don't know if it's improved.

Each avb interface has a 48 track mixer, and they should cascade together. Only question there revolves around how direct avb connection works with multiple interfaces, seem to be some limitations there from what I'm reading.

Monitoring is still driving me nuts, mainly because I mix and monitor in analog. Direct hardware play through no longer exists, and I never had need of cuemix because of it. If you monitor from a separate software mixer, you should be able to adapt easily. In my case I have to manually switch a lot of settings that were previously automatic. Full band tracking punches are one setting, overdubs are another, etc. 5 months in I haven't totally adapted. I shouldn't have to manage 2 software mixers (avb and dp) along with an analog mixer, I didn't previously.

You can't add thunderbolt.

I'm definitely not going back to the sound of the 24i/o after getting used to this. I still have a 2408mk3 connected via ADAT for extra channels, and it is noticeably muddy and indistinct in comparison. The change is worth the headaches so long as you sort through the land mines, and I imagine it's the same or worse with any 3rd party interface.

The GS thread is still the single best source of info, regardless of the clueless parts. I don't see much here and MOTU participates there.

[monkey man]

You can't add thunderbolt.

OK… that really hurts. I must therefore fully explore the AVB-via-ENet card option first; a trash can Mac will break the bank, my patience and mess with my custom wooden rack setup (special bay for Mac Pro with door etc). It is therefore an absolutely last resort.

Look at AVB sampling rate limitations and OS requirements. You can't get all rates currently on Yosemite.

I can live with 44.1kHz as it’s all I’ve used thus far. I’d planned on sticking with it for the time being anyway, especially given the much higher converter quality of the new units.

AVB Direct connection was giving higher latency initially, don't know if it's improved.

I saw that. Half a second or something ridiculous. If I’m only monitoring from the Mackie desk using the (MOTU) inputs routed directly to the outputs, I’d hope this wouldn’t come into play; I assume the figure is for an interface->computer->interface trip. Hopefully the distribution from multiple units’ ins to just one unit’s outs for monitoring wouldn’t incur this unreasonable wrath of Dr. Time…

This aspect could be the make-or-break one for me as a direct AVB->computer (ENet card?) connection would therefore mean I don’t have to upgrade to Darth Vader’s Helmet. This, along with the following one, that is:

… they should cascade together. Only question there revolves around how direct avb connection works with multiple interfaces, seem to be some limitations there from what I'm reading.

Same here. Some sort of limitation on the number of AVB streams (each comprising 8 channels of audio). I’m a complete dunce when it comes to networks / networking. It’s like reading music; I can’t seem to get mah lil’ ol’ brain around it for some reason.

Monitoring is still driving me nuts, mainly because I mix and monitor in analog.

Well, I’ll be “monitoring in analog” but mixing ITB. I only require what CueMix gave me.

Direct hardware play through no longer exists, and I never had need of cuemix because of it.

This confuses me. Weren’t my various CueMix output groups using DHM? Groups of physical ins were routed directly to physical outs which were plugged into the Mackie, providing near-zero-latency monitoring. I always thought this was DHM, but I admit I never did figure out how this interacted with DP’s so-called DHM option in its monitoring-setup dialogue, hence my confusion as to the terminology.

If my basic requirement of input-to-output routing for “realtime” monitoring (outlined in the OP and a no-brainer using CueMix right now) can’t be replicated, then it would appear to be an epic fail and a giant leap backwards from my beloved MOTU. I’m sweatin’ bullets on this one.

If you monitor from a separate software mixer, you should be able to adapt easily. In my case I have to manually switch a lot of settings that were previously automatic. Full band tracking punches are one setting, overdubs are another, etc. 5 months in I haven't totally adapted. I shouldn't have to manage 2 software mixers (avb and dp) along with an analog mixer, I didn't previously.

I monitor only with the Mackie, so does this mean I won’t be able to adapt easily? Heck, I don’t want to have to adapt to anything at this stage; my entire modicum operetta’s been carefully planned for years. I ditched my O2R and OTB mixing for it, after all.

Doug, you’re a gem. Thank you for your time shared thus far and any effort you go to to help me through this. Even the MOTU wholesaler here in Jungleville (Aus) is confused. Further to this, I believe thanks and congratulations are in order for all the effort you've gone to to share the MOTU love over at BeerGutz, patiently copping the flack, sharing your null-test results and dealing with the many queries directed at you. Bravo, mate, bravo!

[HCMarkus]

I have been following allowing at BeerGutz as well MM… Thunderbolt will almost certainly never happen* on pre-6,1 Mac Pros, leaving USB or AVB (with an AVB-compatible Ethernet card installed) asa the only two options for the newest MOTU hardware.

Judging from Mr. Miller's comments, USB is currently the best way to get the new interfaces connected to legacy Macs; the Ethernet drivers are pretty new and have more latency. But the drivers are still being refined, so Ethernet does hold promise. I don't recall what the max i/o is…

If I recall correctly, USB currently allows max of 24 (or was it 28?) simultaneous i/o. For my uses, this would be plenty; as I only use more than a few simultaneous analog ins when tracking bands with drums, and can't see ever needing more than 20 simultaneous inputs. I am very interested in the A16… one sweet interface!

One thought: if your system is currently built around the MOTU 424 card, using 2408s digital i/o with external conversion is another option, but probably more expensive.

*I imagine someone could develop a faux-TB interface that mimicked TB sans video and connected via PCIe… but my imagination has been known to play tricks on me.

[EMRR]

Thanks for the kind words.

I don’t know how much power you need, but I’ve found the top end iMac to have more than enough kick for my needs. The Mini seems to have taken a step backward in processing power, but I see plenty of reports of the previous top model handling tons of processing with no hiccups. My top load is probably around 30 audio tracks, average 3 plugs per, no MIDI, no VI’s, all at 88.2 or 96. My G5 would handle that load at low sample rates with no strain, handle it with strain at high rates.

AVB latency is only high for audio through ethernet to computer, not across AVB to another interface. If you connect computer via TB, it’s great.

Sub-mixing multiple interface mixers should be as simple as treating one as a group within another. I haven't done this yet, but appears straightforward. Master or group outputs from one AVB stream into input channels of another. Routing has actually been eye opening to me:

interface inputs
computer inputs
computer outputs
interface outputs
digital mixer inputs
digital mixer outputs
AVB stream inputs and outputs


I’d say the assumption in the new system is that one will work ITB only, driving monitor mixes from the AVB mixer, and there’s no thought towards analog monitoring. The AVB mixers are somewhat like what Cuemix gave you, as far as I can tell. I never opened Cuemix, so my new reality is driving me nuts. I always tracked with direct hardware play through enabled, and used the analog mixer for monitor mixes. DP did input/playback switching for me. That's gone. To track with low/no latency, you either manually route (within the AVB app) interface inputs to interface outputs in parallel to computer inputs, or you route them to the AVB mixer and build monitor mixes there. If you route interface input to output for tracking, it does not automatically switch for playback, you have to do it manually from the routing window. If you monitor off of DP throughput, you have system latency in between. If you spend your time mirroring your analog mix with your AVB mix, you can get 'sort of' the same thing for both input and playback, but not really. I am used to using analog faders and pans for monitor mixes, while building DP processing in stages during playback moments. These days I’m doing less of that because I’m working on two different mixes or routing paths at all times. I find myself having to think through routing changes a lot these days. It’s lost efficiency for me, and I still haven’t sorted ‘best practice’ for my various needs. I can’t exactly build a universal template because I work with ever changing outside artists needs, not a set internal flow.

Here’s an example. I currently have a live band with 16 inputs tracking, live vocals. The singer has multiple mic choices for different sonic effects, some with analog distortion and reverb on the way in. He might do multiple crossing layers right after a master pass, before the next master pass, and I can quickly have 9-12 vocal tracks to chase in playback monitoring along with 3 inputs to monitor. Then they might do another full band master pass, which might require monitoring for a full band punch if they blow it in the last quarter of the song, so I have to be able to back up and hit play so they can jump in immediately before the dynamic and tempo is lost, and hear the correct blend of input and playback at the right moments, without latency. It’s a lot more to think through; it used to be nothing to think about at all.

I'm rambling, hope that adds something. Yes, you could use a 2408/PCI-424 or two to get everything in and out from an AVB interface set AND still have direct hardware play through. 2408mk3’s seem to be plentiful in the $300-350 range last I looked. I did test the 16A along those lines initially, and it worked fine.

Yes, Direct Hardware Playthrough is my #1 feature request for AVB, with no mention of looking into it so far.

[mikehalloran]

*I imagine someone could develop a faux-TB interface that mimicked TB sans video and connected via PCIe… but my imagination has been known to play tricks on me.

I wouldn't hold my breath. Intel controls the Thunderbolt spec and hasn't announced any chips that could make it happen.

[monkey man]

Thank you very much Sir Markus, Doug and Mike.

My head hurts, but your "rambling" is in fact greatly appreciated, Doug. Sounds like a freakin' nightmare, mate, and of course, bands take this stuff for granted, all the while blissfully unaware of the miracles you have to pull off for their comfort. I feel for you, man.

I'm still none the wiser, really. How 'bout this: If I could hook 3 or 4 units up via AVB (to each other), connecting the "final" one via USB to the 'puter, will I:

1) Be able to spit the live MIDI-gear inputs out to, say, 3 stereo pairs of that "final" unit for real-time monitoring from the Mackie (what I do now with the 24I/Os and CueMix)?

2) Be able to see all inputs for direct printing in DP when the time comes (from all units, not just the "final" one used for monitoring)?

The reason I ask is 'cause instead of "rendering" all MIDI-gear tracks at once as I can do now, I s'pose I could live with doing them in groups due to the 24 I/O constraints of the USB connection. It'd take a whole lot longer (multiple passes instead of one), and defeat the very idea of my hardware-instead-of-VI decision (bit late to change now), but as a means to buy into those converters, well, I guess you guys would understand.

I fear the USB connection might render the AVB connectivity between MOTU units and attendant benefits useless. I suspect my question points to my ignorance. Sorry if this is out of place to say, but why, oh why couldn't / can't the Gods of MOTU allow an all-singing, all-dancing, new-fangled system to simply do what its (many) old systems did, without:

1) The end user having to acquire a degree in networking. I'm being facetious not just because I'm a monkey; networking or the very idea of it sends my lil' ol' brain into a spin.

2) Said user having to jump through all the hoops and tackle the massive reconfiguration (and expense) involved in "upgrading" to Darth Vader's bloody helmet?

Honestly, this head-scratchin', wonderin' and waitin' game of the past 6 months has seen me reach the end of my tether, and I feel I'm about to crack. Help! LOL

[EMRR]

Partway through first coffee, eyes partially open. Let's see what I can muddle. See if you can find relevant bits in the manuals in case I am wrong.

USB won't pass 3-4 units worth of inputs and outputs. Without reading back, I'm recalling more than 24-32 channels to pass, which is the limit of USB if I'm seeing it right.

You could run USB from each unit and lock the AVB units together via word clock cables, in that case you'd have to run them (I think) as an aggregate system and any use of the AVB mix app would be an instance per device, rather than a unified AVB mixer. I think you could treat them as sub-mixers satisfactorily, but it's more to manage.

You will be able to route all inputs to DP, that's the easy part. Any interface input mults easily to computer input, interface output, and AVB mixer. You should be able to submix as you describe with no problem.

The tricky part bombing me is that you can't get input and playback to auto-switch to the same interface output. If the AVB mixer had 32 sends or groups, and I had a spare 32 interface outputs to feed to 32 extra console channels, with a bit more manual operation I could manage something close to what 'direct hardware play through' did automatically. On that note I agree it's poor planning in any arena when new products drop foundational features of the products they replace, with no warnings easily found. In this case it ignores the way any full featured analog studio has always run. It defines the target user as entirely ITB without describing it as so. As I've said elsewhere, we may have just had it better than anyone else, as I've not found much clarity on the way other interface systems are functioning in this regard.

A problem for me with the AVB routing and mix app is the layout. Even with a 27" screen and unnecessary sections collapsed, I find routing channels I need to see quickly off screen from one another, with scrolling required on a frequent basis. It takes 3-5 seconds for the app to change from mixer to routing, so it's not quick to confirm routings and get back. You could open multiple tabs and switch tabs constantly. To keep track of everything properly, you have to spend time naming things. Take the time of naming a full set of channels in DP and double it, with double checking added in. These things scream '2nd engineer/intern', not sole operator. Or maybe they shout 'three 27" monitors'! I do still have a control room window I need to see through. The AVB app is an additional required layer beyond what I had to do in the past, without much (any?) benefit added. There are certainly cases where one might never need to change anything in the AVB app, just build a preset and leave it closed, satisfactorily switching anything needed from DP.

I feel certain there's a solution I have not sorted yet. I have thought several times I'd found one, and the 'gotcha' has appeared almost immediately. I hope to feel like an idiot once I figure it out, rather than never find a path. I am fighting years of ingrained PCIe system habit. At this point I'm very happy when a band wants to track without headphones, and has no need to monitoring until overdub time.

If I were doing it again I might have strongly considered updating my PCIe system computer, losing the 24i/o, adding another (used market) 2408mk3, and patching AVB interfaces through the 2408's to retain 'direct hardware play through'. That previous sentence suggests it would be a simple code addition to add DHP to the AVB systems, but I really don't know what would be required. I still would not roll back to the old PCIe sound, now that I'm here.

[monkey man]

Thank you again, Doug. I love reading these "thinking out 'loud" rants of yours; I learn / re-learn so much, man.

Dang. No inter-interface communication when "main" one is connected to a computer via USB. That sucks. Bummer. There goes my near-last big hope.

I also thought of the discreet USB connection per interface. Haven't got my head around that one yet. For starters, I've no PCI slots left due to USB expansion cards as things stand already, and am still technically a few ports short. Oh... I just realised that if I ditch the 424 card a USB one could replace it. That'd mean every expansion slot'd be filled with USB ports! Sounds... risky. LOL I don't know. Somehow replacing the trusty 424 Audio Wire system with USB fills me with dread. I started on OS9 with the original 324, so you understand.

I wonder how much longer I should wait for clarity on my routing issues, MOTU's USB and AVB driver / app tweaks, Apple's desktop direction (new model/s) and of course the CPU upgrade which (I hope) should still be great value for my machine. If I sit on this Mac, it's looking more and more like I'll have to stick with the 24I/Os, which kinda defeats years (7 or so) of saving for my monitor upgrade (done) and mic / preamp one too (doing). Sure, a better mic and preamp, along with monitors (I was speaker-less in my music room for about 6 years!) would surely improve my "product", but I seriously went / am going the extra mile with these upgrades, hence the fact that I (and I'm sure you too) consider the new MOTU system a serious no-brainer and fully justifiable. Oh, and the AVB setup I envisaged would've removed the Mackie as an additional input layer too, as I route direct guitar, bass, and mic pre inputs via its direct outs (DB24 connector?) to a 24I/O. A 1248 as the "main" interface could've replaced it for these inputs, although I'll bet the Mackie's DI inputs would've been better (I've often waxed lyrical about them).

Confused as ever. Where's Magic D when you need him? It used to be that I only had to wish for it and he'd appear. He speaks, I hang on every letter and walk away filled with knowledge, love and happiness. LOL I laugh, but it's true.

Please, feel free to vent any time in this thread, Doug; I'll be watching it like a hawk, along with that lengthy and frustrating GS one.

[monkey man]

At this stage, unless I can integrate, say, a 1248 via USB easily into my system, retaining the 24I/Os for CueMix and MIDI gear and using the new unit for guitar, bass and vox input as well as main mix output (oh, and I'd have to retain the Mackie as a unifying component), it's looking like an (to me at least) Epic Fail. Without DHP this will be impossible anyway, although MOTU has suggested it may well be added in a future (firmware?) update. I cannot say how bummed I am; I wish I could stop shakin' my head about this.

I've spent 7 years upgrading my input and output audio paths, from mic to preamp to monitoring; the only thing missing will be the AD / DA stuff in the middle.

Bottom line:
Unless the upgrade can be achieved using my (new, for God's sake) 2012 Mac Pro, MOTU's successfully dangled an unattainable banana in front of this monkey's face. Unbelievable. It's cost me 7 months, and an awful lotta sleep. This is not the sort of thing I've come to expect of my beloved MOTU. Retro compatibility and support has thus far, IMHO, been one of this excellent company's mainstays / strengths, and I've come to rely on this continuing to be the case.

As I mentioned to Mark in his Peluso thread, I can only hope that the cumulative effect of summing all my MIDI hardware inputs with my (being upgraded) input path doesn't introduce much in the way of muddiness, and especially phase anomalies. I'm so sensitive to phasing it's ridiculous. In my last audio dip course (practicals done at Metropolis in Melbourne), I heard obvious phasing on overheads whilst standing far over to one side of the control room, when none of the other students, including the uber-veteran freelance engineer conducting the session (50 odd years of experience), who were all centrally located, could hear a thing. "You're dreaming", I was told a somewhat biblical three times after I kept insisting it was obvious to me. Finally the engineer brought up the phase display (probably to try to shut me up) on the mega (SSL) console, and lo and behold, a decent amount of comb-filtering was evident. I wasn't questioned on this issue from that point forward. I mention this only in the hope that it'll help to convey how much I'd appreciate being able to ditch the 424 system for these awesomely-spec'ed new units. I can't help but wonder how much business MOTU is missing out on as a result of Apple's stubborn refusal to allow TB-upgrading of its legacy machines. It's not as if a 2012 Mac Pro is by any means a retro machine anyway.

Rant over. I don't know what else can be said.

PS: Apple to the rescue with either a conventional-form-factor Mac with TB?
A third party (or Apple) to offer an upgrade path to TB on its older machines?

Unless I'm missing something, these (probably naive) two possibilities are my only hope.

[monkey man]

Well, I do know what else can be said:

After feeling bad for saying less than good things about my favourite company yesterday, the cold light of a new day demands that I give up some of that, perhaps unfair, ground that I snatched.

MOTU can't help that Apple won't allow me to upgrade my machine to TB. Let's be clear about that; MOTU is simply keeping up with Apple's hardware developments, as it has always done. Its AVB line of interfaces and its brilliant use of the protocol, exceeding its latency limitations and so on is IMHO a game changer, especially when one considers the asking prices of the various units. Brilliant stuff.

MOTU can't help the limitations of USB either, so we cannot reasonably expect the sorts of I/O counts the AVB system offers with TB from it.

An ethernet connection for AVB to the computer, whilst promising, appears to have its inherent limitations as well. MOTU can't help that.

As an audio system for installations, studios, live use and so on where channel counts can approach stratospheric numbers, MOTU's AVB solutions appear to be just the ticket. My beef is that the lowly small studio owner, home-based artist or hobbyist (I'm talking, say, 96 channels of input or less) cannot buy into this system without the requisite Mac Pro upgrade, along with having to shift his / her peripherals out of the box to 3rd-party chassis, expanders and the like. Then there's having to deal with a new way of doing things (networking), along with its attendant surprises and clients.

The solution? Easy. MOTU either:

1) Upgrades the converters offered on its (I think now legacy) AudioWire (PCI-based) interfaces and returns them to market. This would involve pretty much zero R&D time / expenditure, as the units have been designed, tried and tested no end and are rock solid. CueMix, DHP and so on would function just as we've become used to, and love, for that matter. Heck, even the existing drivers would work.

2) MOTU releases AudioWire versions of the current new crop of AVB interfaces. The extra DSP power of the new line could be reduced as well, allowing CueMix to function as it already does. Further, for those who wish it, that additional DSP could be offered as higher-powered equivalents of this new AudioWire line. CueMix could be tweaked to accommodate both scenarios, allowing a greater number of realtime FX for the latter units, obviously.

Both options I've suggested would:

1) Provide seamless upgrade paths for all those currently happy with their I/O counts using AudioWire. This, in my estimation, would constitute the majority of existing MOTU PCI users, both PC and Mac-based.

2) Represent great value for money for MOTU - current users would upgrade in a heartbeat (myself included) and, as suggested, the R&D time would be negligible.

3) Exploit further the trusty, rock-solid AudioWire system which so many of us value highly and which has proved itself over roughly 15 years. In fact, I'd go so far as to say that it's been the only system I've run which has been 100% trouble-free. I'd trust it in the most mission-critical scenario I could possibly imagine, and then some. It ain't broke. It's never been broke. It "don't" need fixing.

There. I feel better now. I'll watch MOTU as a hawk does a rat. I don't think the suggestion is unreasonable. I know it would mean mucho more $ for my beloved MOTU. I also know that the Gods of MOTU are a super-savvy bunch of awesome people; they're no fools. Gods of MOTU, I'm praying to you...

[mikehalloran]

MOTU can't help the limitations of USB either, so we cannot reasonably expect the sorts of I/O counts the AVB system offers with TB from it.

True but can you factor USB 3 into the equation? PCIe cards have been out for awhile now.

[monkey man]

Well Mike, MOTU only offers USB2 on the AVB line.

If 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:

1) The PCIe card currently marries all one's interfaces elegantly for up to 4 units, whereas USB3 connections would require PCI 4-port USB3 cards, Macs that offer sufficient USB3 ports,TB hubs or whatever, for the equivalent.
2) I have a PCI 4-port USB3 card (for external backup), and it doesn't eject disks properly and also has had various issues with DU. I've had to use the stock USB2 ports to format drives and so on, so as things stand, whereas I'd stake my life on the 424's reliability, I wouldn't bet a rotten banana on USB3.
3) 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). To achieve this using USB3 would, if I'm reading this correctly, require additional CPU power to "simulate" this feature of the PCI card because the latter seemed to "directly wire" all that I/O, hence my use of the term "marry".

I s'pose there'd be no harm in simply trusting MOTU to make whichever route it might offer work. History tells us that this should be the case.

Where I'm coming from with the AudioWire system is just that if it ain't broke, and it's never been broke, all current functionality at the very least, should be retained in future incarnations (not including AVB, as, I think, we're discussing one of my proposed alternatives to AVB).

Thank you for chiming in, man. I'm still sweatin' on this whole shamozzle. 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...

Oh, what I meant to say, Mike, was that I very much appreciate hearing any thoughts you have on this. I feared the thread would disappear from the short-term radar into the ether. Thank you!

[mikehalloran]

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.

Well Mike, MOTU only offers USB2 on the AVB line.

I wasn't aware of that bottleneck.

[billf]

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.