So, no firewire unit then? WTF???

[Stuart Stuart]

I know it's been extensively discussed, but I'm simply amazed by the decision not to break out of the pci dead end. It's getting harder and harder to find a motherboard with more than a couple of PCI slots, and the more you use UAD plugs, the more you want to use them.

And now they're releasing the Neve stuff. Oh geez - this is gonna cause me some heartache ...

UA - please explain what the thinking here - are you EVER planning to release a firewire (or better yet, solid USB) design? You'd sell a truckload - surely the manufacturing cost must be reasonable, and (with my limited technical understanding), all the hard work has been done coding the plugs in the first place. TC seem to be cranking these products out at a rate that is embrassing you guys.

I've got two UAD cards, but I'd buy a firewire or USB device first day it's available. I think most other folks posting here would back me up. There's a god damn goldmine sitting there my friends ... dig in!!!

 

[1176]

Not to say \"it sucks\" entirely or that it is not convenient, but firewire is weak compared to PCI. I have the powercore and the powercore firewire, and the the PCI version can run many more plugs, regardless of TC's marketing. I doubt a USB device would work at all for UAD. The bandwidth of PCI is much more robust. I'm afraid a firewire UAD box would be very disappointing as far as plug-in counts.

 

[NuSkoolTone]

This is nearly cross posting, but I'll chime in. I'd be OK with 1 or 2 PCI-e UAD cards with 4x the current processing over a FW box. But let's get on with it though.....

What will probably happen is we will be looking at magma chasis solutions again. This time all PCI-e and up to eight cards. It would be nice if they can make a case that is a SLIM form factor to do this. I JUST BOUGHT a new rack for my daw, and I'd rather NOT put a 4 space nearly EMPTY BOX in it.


IF they Made a rack like this: (Open top view)

--------------
:Card:Card:
:Card:Card:
:Card:Card:
--------------
EDIT:Apparently I can't count. Edited the diagram.


I think they could fit at least 6 in ONE rack space(Albeit a fairly deep one). All they would need is power, ventilation, and the bridge circuitry plus a connector. The left rails is where there would be 1x(4x?) connectors. Then one Host card in the DAW just like with Magma.

These cards don't need a lot of room, and with minimal cooling I think they would be fine.

Now 6 is an odd number to use I suppose, but it would go with my two current PCI UADs quite well dontchya think?

 

[Tony Ostinato]

I think they were waiting for this:

http://www.sonicstate.com/news/shownews ... wsid=2945#

so that they could come out with this:

\"What product would you ask UA to build for your own personal studio? Would this be something that other people would want too?
Okay, now you're talking. What I would like is a high-resolution real-time effects box with analog and digital I/O; something like a unit that houses 12 UAD-1 processing cards that would allow me to use these effects in real time on multiple channels-let's say 16-with the ability to upgrade at will. Along with a really sexy programmable remote. I'm talking all types of plug-in running on this box (the ones that only UA can produce with there incredible emulation skills this I know others would want because now they have box that they can take on the road for live and then go back to the studio with it to mix a record.\"

which is from this:

http://www.uaudio.com/webzine/2006/april/index6.html


and is all based upon joe bryan saying in an interview early this year that firewire was \"just becoming doable\".


We wont know about it till its out, these guys have superb security. nobody knew about the pci-e model until they announced it, thats amazing nowadays.

 

[daverich]

Firewire is great, but like RME did with the fireface you need to tweak it to ignore all the bullshit and just do the job (put simply heh)

Kind regards

Dave Rich

 

[Miser]

How large is the UA`s current mpact chip stock?I think this is relevant question.
I remember that UA bought the whole stock years ago,but I`m not sure.
Maybe they try to sell away their existing stock by using it on UAD-1e card
and by promoting old UAD-1 cards (UAD 100 promotion),but this is just speculation.Would the new chip architecture mean changes to algorithms?
Fred`s dream system is wonderful with its open architecture,but
limited chip stock would seriously undermine the continuity and development of mpact chip based systems.
Has UA found a new chip developing partner?Dice II would work as a controller and routing chip,but for running their algos UA needs a different chip.

 

[robi]

This is nearly cross posting, but I'll chime in. I'd be OK with 1 or 2 PCI-e UAD cards with 4x the current processing over a FW box. But let's get on with it though.....

What will probably happen is we will be looking at magma chasis solutions again. This time all PCI-e and up to eight cards. It would be nice if they can make a case that is a SLIM form factor to do this. I JUST BOUGHT a new rack for my daw, and I'd rather NOT put a 4 space nearly EMPTY BOX in it.


IF they Made a rack like this: (Open top view)

--------------
:Card:Card:
:Card:Card:
:Card:Card:
--------------
EDIT:Apparently I can't count. Edited the diagram.


I think they could fit at least 6 in ONE rack space(Albeit a fairly deep one). All they would need is power, ventilation, and the bridge circuitry plus a connector. The left rails is where there would be 1x(4x?) connectors. Then one Host card in the DAW just like with Magma.

These cards don't need a lot of room, and with minimal cooling I think they would be fine.

Now 6 is an odd number to use I suppose, but it would go with my two current PCI UADs quite well dontchya think?

You can buy the magma parts and stick it in a selfmade case, any design possible, maybe you could even sell some.

 

[Eric Dahlberg]

The fact that 8 cards can now be used may support the notion that firewire is coming soon, seeing as it's almost a pointless feature to add at this point.

 

[Dan Duskin]

Not to say "it sucks" entirely or that it is not convenient, but firewire is weak compared to PCI. I have the powercore and the powercore firewire, and the the PCI version can run many more plugs, regardless of TC's marketing.

You much have a really bad firewire chipset.

At 6ms system latency...
I can load a UAD-1 to about 85-90% before pops and clicks.
But I can load all 4 DSP's on my powercore firewire to 100% without a single pop or click. And generally speaking, I can get more plugins on the powercore than than two UAD-1's. Plus, the latency of the powercore is HALF that of the UAD-1. And on top of this I can load 4 more powercores on the same firewire bus (soon more with a driver update). So... is firewire really bad? I don't think so... somehow TC is making it kick the shit out of uad-1 on pci!

 

[Tony Ostinato]

aaaannd after encountering and dealing with firewire limitations while working with powercore firewire they went on to developing this:


http://www.sonicstate.com/news/shownews ... wsid=2945#


but im not hiding that a firewire thing is what i want to happen and that could be clouding my theory.

 

[Dan Duskin]

I've had problems with my powercore firewire, and problems with my uad-1 pci. I love both of them!

My point is that even though I've had an equal amount of issues with both, the one thing that has been consistant is better \"performance\" (speed) from the powercore firewire.

again...
poco fw @ 6ms latency can load all 4 dsps 100%
uad-1 @ 6ms can only go to about 85-90%
powercore latency is half that of the uad-1
THE WINNER OF SPEED AND PERFORMANCE = Poco FW

And lets not forget... it's 1394a... if UA made a 1394b unit it would be even better!

People can talk theory and bus speed as much as they want... but in the real world this is what we have.

 

[bob humid]

Firewire is great, but like RME did with the fireface you need to tweak it to ignore all the bullshit and just do the job (put simply heh)

Kind regards

Dave Rich

you didnt read the above links, didn´t you? DICE II is FireWire & DSP on a chip.. dedicated technology for audio purposes. makes very much sense..

in the past FireWire was a desaster

robert

 

[Paul Woodlock]

Not to say "it sucks" entirely or that it is not convenient, but firewire is weak compared to PCI. I have the powercore and the powercore firewire, and the the PCI version can run many more plugs, regardless of TC's marketing.

You much have a really bad firewire chipset.

At 6ms system latency...
I can load a UAD-1 to about 85-90% before pops and clicks.
But I can load all 4 DSP's on my powercore firewire to 100% without a single pop or click. And generally speaking, I can get more plugins on the powercore than than two UAD-1's. Plus, the latency of the powercore is HALF that of the UAD-1. And on top of this I can load 4 more powercores on the same firewire bus (soon more with a driver update). So... is firewire really bad? I don't think so... somehow TC is making it kick the shit out of uad-1 on pci!

The point is though Dan is that the next generation of UAD cards will no doubt have a lot more DSP capability. meaning more channels can be sent to and from the card.

If we're perhaps looking at a 4 x card ( which is about right for typical processing evolution ) then Firewire ain't really upto to it. Firewire also has the software overheads you don't get with PCIe

Maybe it's the 'external' benefits that make firewire look attractive? I think this would be better served By UA making PCIe cards in the future and also providing a consumer cost magma type box to house them in.

Best of both worlds!

 

[Spacey]

I still dont se why they just make a pcie card like say an old scsi card interface type of thing, with a cable coming off the card into a rack full with old uad-1's.

You wouldn't have to worry about the bandwidth of firewire as it would be like sticking 8 cards on one pcie slot card.

 

[Dan Duskin]

[quote="Dan Duskin":ny7p8fcs]

Not to say "it sucks" entirely or that it is not convenient, but firewire is weak compared to PCI. I have the powercore and the powercore firewire, and the the PCI version can run many more plugs, regardless of TC's marketing.

You much have a really bad firewire chipset.

At 6ms system latency...
I can load a UAD-1 to about 85-90% before pops and clicks.
But I can load all 4 DSP's on my powercore firewire to 100% without a single pop or click. And generally speaking, I can get more plugins on the powercore than than two UAD-1's. Plus, the latency of the powercore is HALF that of the UAD-1. And on top of this I can load 4 more powercores on the same firewire bus (soon more with a driver update). So... is firewire really bad? I don't think so... somehow TC is making it kick the shit out of uad-1 on pci!

The point is though Dan is that the next generation of UAD cards will no doubt have a lot more DSP capability. meaning more channels can be sent to and from the card.

If we're perhaps looking at a 4 x card ( which is about right for typical processing evolution ) then Firewire ain't really upto to it. Firewire also has the software overheads you don't get with PCIe

Maybe it's the 'external' benefits that make firewire look attractive? I think this would be better served By UA making PCIe cards in the future and also providing a consumer cost magma type box to house them in.

Best of both worlds![/quote:ny7p8fcs]

Sure FW is up to it. It's up to the task right now with 1394a. If UA is fussy and can't get it right with that virtually endless bandwidth (more than any dsp would ever need) they could go over the top and make it 1394b (firewire 800).

 

[RichR]

Sure FW is up to it. It's up to the task right now with 1394a. If UA is fussy and can't get it right with that virtually endless bandwidth (more than any dsp would ever need) they could go over the top and make it 1394b (firewire 800).

Are you taking higher sample rates into account?

 

[Blatboy]

Just out of curiosity, how would firewire compare to doing it over an ethernet connection like the APA units that Waves is now peddling?

 

[Dan Duskin]

The APA uses TCP/IP packets. It's fast enough... the problem is that it's not consistently fast (so I hear)... and that the latency fluctuates (yikes!).

 

[Blatboy]

Interesting.... I need to do more research on the APA... damn that box is $$$ though.

Re: future versions of the UAD: While I can see that PCI could indeed be faster, and less, I donno, firewire than firewire...heh heh heh... guys like me that have the AMD8131 G5 issue would buy up a firewire UAD in a New York second. Hell, I'd buy two, even if it was the same freakin' card. Until that (or some other solution) comes out, I may have to use other plugs for my larger sessions. (Not the end of the world for me, for sure...) That's certainly cheaper (and makes more sense) than buying a new G5 just for the fix.

 

[RichR]

Sure FW is up to it. It's up to the task right now with 1394a. If UA is fussy and can't get it right with that virtually endless bandwidth (more than any dsp would ever need) they could go over the top and make it 1394b (firewire 800).

Are you taking higher sample rates into account?

Every time I mention "higher sample rates", I'm treated like the invisible man even though it has a direct relationship with the future of the UAD.

I wonder why?