1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

USB 2.0 Audio Interfaces

Discussion in 'Studio Techniques' started by Jay Asher, Apr 3, 2009.

  1. Jay Asher

    Jay Asher Senior member

    For years I have advised people, perhaps erroneously, that USB was inferior to Firewire for audio interfaces because it was not as robust a protocol and because it apparently transfers info in "packets", which is supposedly less good for audio.

    However, RME is about to release a USB 2 interface which they make some pretty impressive claims for.

    Whaddya think, folks?

  3. Eddie Sullivan

    Eddie Sullivan Senior member

    From the Wiki on USB

    USB compared with FireWire

    USB was originally seen as a complement to FireWire (IEEE 1394), which was designed as a high-speed serial bus which could efficiently interconnect peripherals such as hard disks, audio interfaces, and video equipment. USB originally operated at a far lower data rate and used much simpler hardware, and was suitable for small peripherals such as keyboards and mice.
    The most significant technical differences between FireWire and USB include the following:
    USB networks use a tiered-star topology, while FireWire networks use a tree topology.
    USB 1.0, 1.1 and 2.0 use a "speak-when-spoken-to" protocol; peripherals cannot communicate with the host unless the host specifically requests communication. USB 3.0 is planned to allow for device-initiated communications towards the host (see USB 3.0 below). A FireWire device can communicate with any other node at any time, subject to network conditions.
    A USB network relies on a single host at the top of the tree to control the network. In a FireWire network, any capable node can control the network.
    USB runs with a 5 V power line, while Firewire can supply up to 30 V.
    These and other differences reflect the differing design goals of the two buses: USB was designed for simplicity and low cost, while FireWire was designed for high performance, particularly in time-sensitive applications such as audio and video. Although similar in theoretical maximum transfer rate, FireWire 400 tends to have the performance edge over USB 2.0 Hi-Speed in real-world uses, especially in high-bandwidth use such as external hard-drives.[26][27][28][29] The newer FireWire 800 standard is twice as fast as FireWire 400 and outperforms USB 2.0 Hi-Speed both theoretically and practically.[30] The chipset and drivers used to implement USB and Firewire have a crucial impact on how much of the bandwidth prescribed by the specification is achieved in the real world, along with compatibility with peripherals.[31] Audio peripherals in particular are affected by the USB driver implementation.[citation needed]
    Initially, cost was significant in USB being more widespread than FireWire. Over time, USB benefited from network effect.[citation needed]
  4. Hm, I don't really think it's about what's on paper.
    Thing is, USB has gone a long way for Macs. One of the reasons why Mac users always recommended FW over USB has been that up until the event of the IntelMacs, USB has been performing notoriously bad, whereas with most Windows PCs all was fine and dandy already. Apparently Apple didn't give much of a damn about the quality of their USB chipsets. I have seen G4/5 Macs losing connections to (quality!) USB drives without any reasons, sometimes not even accepting them at all, etc.
    But as said, it doesn't seem to be any problem anymore. I have a FW audio interface (M-Audio) and some USB ones (NIs RK2 and Kore, plus an IKMM Stealthplug), the performance figures are almost identical both in terms of lowest possible latencies and CPU consumption.

    Having said that, I still find the number of simultaneous streams RME wants to achieve over USB (18 in each direction...) quite astonishing. I could as well imagine that, when in fact using any number of channels getting close to the max., there will be a more or less noticeable CPU hit, especially when running it at extremely low latencies.
    Defenitely something worth watching, IMO. If this really works as well as they say, it'd be immensely useful for some people I guess.

    - Sascha
  5. Eddie Sullivan

    Eddie Sullivan Senior member

    USB vs. Firewire

    The excerpt from the Wiki was really a response to reference in the original post about USB traveling in 'packets' rather than some sort of complete answer or conclusion on my part- I probably should have made that clear. I'm also surprised that a USB 2.0 interface can claim that kind of performance. I've always suspected that Firewire is somehow better for audio than USB, but honestly could never articulate it well, except that I know Apple and Sony marketed Firewire as a specific standard for high definition transfer of audio and video into computers, and that USB's original intended implementation was for control devices and printers and keyboard vaccums etc...

    I do know that RME makes some really great sounding hardware, although I've heard anecdotal accounts of driver issues and some incompatibilities, I think RME has many loyal users and fans...

    I'm thrilled with my MOTU UltraLite... it's concentrated audio goodness is soooo satisfying...
  6. I just watched a video taken at Musikmesse, with Mathias Carstens (chief of RME) explaining the thing. He said that it would indeed be as fast as a PCI solution, given the same buffersize - so far USB and FW devices always added some internal "safety buffers" that were higher than what can be done with PCI, so the latter so far is still king when it comes to lowest latencies. But Carstens said that this would no more be the case and that the Fireface UC would even be faster than the FW model!
    It seems they really put some thought into this, you even need to boot the device for use with either OSX or Windows, apparently there's some noticeable differences in how the two deal with USB.
    And he finally said that it should be no problem of having it running at 32 samples buffersize constantly under OSX (14 are possible, but apparently it's just there to prove a point, he said that at these low settings the system would indeed be taxed too much).
    And well, the price will be exactly the same as for the FW model.

    That thing really starts to look interesting for me, given that I constantly miss a PCI (or PCMCIA) option on my Macbook.
    Now, if Apple would just get their act together, supplying a sufficient amount of USB ports on their mobile machines...

    - Sascha
  7. Eddie Sullivan

    Eddie Sullivan Senior member

    Mad interesting about the 'dual boot' nature of this interface : )
  8. Orren Merton

    Orren Merton Logic Samurai / Administrator Staff Member

    Very nice! I have to say, having dealt with Mr. Carstens when I had issues with my G5 and Fireface 800 years ago, he seemed like the sort of guy who when he understood something, he understood every aspect from every angle possible...if anyone could come up with a way to workaround the limitations of USB's packetizing nature, it would be he and his engineers!

    I have to say, I'm far more inclined to get a MacBook one day than I had been previously, knowing that the FF400 will be available.

    That has always been my biggest complaint with Apple laptops. I have a tiny little Linux EeePC that fits in the palm of my hand, and it has one more USB2 port than my MacBook Pro! I very much appreciate Apple's dedication to elegance in industrial design, but the limited USB ports are one area I think they take it too far.

  9. Ideed. My quite aging Samsung laptop (it's 4.5 years old) has 4 USB ports, all even throwing out a little more than the nominal bus-power, so I can connect tons of USB sticks, dongles, MIDI keyboards, mice and what not, without the need for any additional power supply, let alone a USB hub.
    With the Macbook (the last white model, 2.4GHz), I can still get along more or less ok-ish when cabling up my little mobile studio, but that's only because I'm using a FW interface. A setup of interface, MIDI keyboard and mouse would already not be possible anymore on the latest MBs, without using a hub.
    Now, the new RME thing won't run on buspower anyway, so you need an additional PSU in any case, but why I *must* have to deal with an active USB hub in addition is completely beyond me. Seriously, I don't think that apart from Apple, anybody would get away with supplying 2 USB port as the only outside hardware communication option in what is not exactly a cheap laptop (especially when you justify taking away the FW port saying that USB is more common, such as Steve Jobs did).
    And talking about elegance in industrial design, I find it way more elegant to connect, say, 3-4 USB devices straight to my machine instead of having to deal with whatever additional hubs. Saves space in your computer bag, doesn't require additional power outlets and there's no nasty additional cabling going on, either. And it's not exactly as if unused USB ports would do much harm to whatever design.

    - Sascha
  10. Alan Branch

    Alan Branch Logic Samurai

    HI Jay,
    I really like RME products they really do make some great quality stuff, if they say they can get that buffer that low then it sounds very good for a mobile recording i/o.
    I did see this the other day

    I thought it was amazing as it has everything in one box! controller, I/O, MIDI , mixer , effects and even a SD recorder (only stereo so far tho) Roland build quality looks really good too, and it works with a Mac & PC. If the sound quality is any good I think they will sell tons of these little things.
    Perfect for laptop recording, small budget studio etc.

Share This Page