![]() |
![]() | |||||
|
|
|
|
|
|
| |
The 01x was the first commercial product based around mLAN connectivity. When the 01x was released in 2003 this Yamaha-initiated FireWire technology had received its fair share of column inches since it was first announced but how did mLAN work in practice? As with any technological development it takes time for people to catch up with the technology and vice versa. Any format that involves multiple devices with wide-ranging motivations and markets, not to mention the added complications of ever-changing computer platform standards and OSs, needs all the help it can get from the industry at large. Since the commercial industry at large comprises at least potentially competing companies it makes a lot of sense for manufacturers to utilize the brainpower within University science departments whose motives and interests are relatively unsullied by issues relating to the bottom line. One such University that Yamaha turned to when mLAN was still in development was Rhodes University in South Africa. 01xRay.com recently spoke with professor Richard Foss, the man who not only spearheaded the mLAN research team but who has also now formed a company to market the Matrix Mixer for 01x, the world's first independent (as in non-Yamaha) application for the 01x.
How did your association with Yamaha and mLAN begin? “I had been working with a company called Digital Harmony based in Seattle, doing FireWire work for them. I am a professor at Rhodes University and also manage a company called Networked Audio Solutions that performs contract work in areas related to my research. In 2001, when the contract work with Digital Harmony ended, I started looking around and thinking well, who else is doing FireWire and of course Yamaha was an obvious candidate. We approached them and at the start of 2002 I went to Japan, spoke with the guys at the mLAN Project, and they were very receptive to us working with them. We started with an initial project, which seemed to go very well, and that relationship has carried on for about five years now. We've done a number of projects for Yamaha – the research and development is slowing down now and the work has now moved more towards the synthesizer division, where they're now really using the results of that research and development.”
How and why did your work move from R&D for Yamaha to developing your own products? “We were appointed as an official mLAN Service Provider so that we could work with other companies utilizing mLAN technology. These companies then use mLAN in their own products. It's straightforward and does not cost anything to become an mLAN licensee. The mLAN Project has always promoted the opening of their technology. For example the IEC61883-6 specification, which is an International Specification was originally proposed by Yamaha and was then opened to the world. The Plural Node architecture, which is fundamental to mLAN version 2 was made into a 1394TA document when its viability had been tested. To simplify the use of the Plural Node Architecture, Yamaha proposed something called the Open Generic Transporter specification, which has been submitted as an AES specification candidate. So the thrust of the mLAN Project is really to get other companies to use the technology and to promote interoperation between various devices from these companies, so that we can all benefit from such interoperability.”
How common is it for manufacturers to make their technology so openly available? “I think mLAN is quite unusual and I've got a huge amount of gratitude and respect for the way in which Yamaha have approached the release of mLAN technology. They've released at the point that they feel the technology is mature and is capable of fulfilling the needs of the professional audio community. I think the mLAN project has been quite unusual in this regard.”
Unusual in that it's gone from proprietary to open or inherently unusual? “I think unusual in that the original focus was to create a means of interconnectivity that would be opened to the world, and that Yamaha would bear the cost of testing the technology and opening it at the time when it was mature. Of course the original goal as Yamaha perceived it in the early 1990s was that MIDI and the multitude of different digital audio specifications was not really conducive to interconnectivity. So they were looking for something that would work with a single cable that would use current high-speed transmission technology, and would allow for support of professional audio devices. They saw FireWire as a technology that had an appropriate mix of isochronous transmission that was very deterministic, and would ensure that your audio would arrive in a particular time-frame, and asynchronous transmission that would allow control messages to be transmitted. So they took what was available in FireWire at that stage and then built the audio and MIDI transmission capability that they needed on top of that. They presented this to the 1394 Trade Association back in 1996. There were two parts to it: there was the transmission capability, and also a recommendation for connecting devices. The Trade Association said, well let's go for the transmission capability first of all, and this then developed into the international specification that is the IEC61883-6. Most manufacturers who transmit audio or MIDI data over FireWire utilize this standard.
mLAN has now appeared on a small number of products. Are there many others like yourself who are developing mLAN tools? Or is it a niche “No, there aren't many others developing mLAN tools, so it is somewhat of a niche at present. However, the intention is that the technology be accessible. The current position is best explained by going a bit into the history of mLAN. There was an mLAN version 1 that came out originally, and it was based on the 1394 Trade Association Audio/Video Control (AV/C) specifications enhanced with vendor specific features that mLAN required for professional audio connection management. A number of mLAN version 1 devices were created. However, these version 1 devices incurred large start up times when the network contained many devices. Furthermore, there was complex plug abstraction technology on the devices themselves – difficult for other manufacturers to construct on their devices. At the beginning of 2002, the mLAN project moved to a version 2. The main feature of this technology is a thin code layer on each device, and plug abstraction technology on a workstation, where it could easily be updated. This change mid-stream on the part of Yamaha was a bold move, but in the long run necessary for a lasting FireWire connection management solution. It is the effects of this mid-stream change that mLAN users are feeling, and that has caused many frustrations. My hope is that we can ride out this change as the version 2 technology matures. “As indicated previously, the nature of the technology used in version 2 has been released to the 1394 Trade Association as a Plural Node Architecture document. Also, the technology on the device side has been released in the form of the ‘Open Generic Transporter' specification that was released for standardization to the AES. For the workstation technology to be released there has to be certainty of its stability, and also there have to be resources to support the code and developer questions. There simply aren't these resources available inside Yamaha at this stage. Hopefully this will happen as the technology is used more pervasively. “I'm very positive about the future but I can also understand the wariness that people have because we have gone mLAN version 1 and there's been a complete change of course. That gives people the jitters – what's going to happen next? But my feeling is that we can really move forward with this mLAN version 2 technology and be very successful with it. “However I don't think this is niche market i.e. only geared to the small studio market. FireWire can now be transmitted over CAT-5 and optical cable, allowing for large transmission distances. This makes it feasible to move the mLAN technology into the commercial audio and broadcast markets.”
How about Ethernet? “When the mLAN project had to decide on a networking technology, they did of course consider Ethernet. The reason they chose FireWire was that it was designed for multimedia transmission, and in its standard form provided the necessary capabilities for high-speed deterministic transmission of audio and MIDI, as well as asynchronous transmission capabilities. It also allowed for the deterministic transmission of video. Ethernet, in its standard form, did not provide these capabilities. Ethernet is used successfully in the broadcast and commercial audio environments by companies with proprietary technology, and there are constraints on the configuration of the networks particularly if low latency is required.“
How did the Matrix Mixer come about? “To a large extent it came about with my frustration with using the 01x. I'd felt this to be a really powerful workstation and yet I wasn't in a position to easily utilize its features. The sort of thing I would do would be to page to the back of the manual, to the block diagrams, view what features were available, and then try and find from the manual how I would get at those features. So the thought occurred to me - why not lay out, in a simple way, what the block diagram is, and provide the user with the capability of going to the different areas of the block diagram and changing the parameters at those points. That led to the concept of a matrix, where there are a number of inputs as the rows, and a number of outputs or destinations as the columns. And that's all you see when you first bring up the Matrix Mixer. “In other words you're only exposed to the particular information that you need at a particular point. The first thing that I usually want to do with the 01x is to route a bunch of mLAN sources to some destinations – maybe I want to route them to the auxiliary bus or something like that. I maybe want to route them across to effects units. And I can do that very easily by just going to the cross-point of an mLAN source and clicking on the cross point where it meets with the effects unit column. Then if I want to do any processing, for example of the mLAN source, then I just click that and modify the parameters. “When I go to a source that I've just routed, and right click on it, up comes a parameter window and I can modify those particular parameters at that source point. Similarly if I want to modify the destination–for example the internal effects–then I click on the internal effect and just see the parameters that I can modify there. “So I only utilize and see information when it's relevant to me. I'm not confronted by a large amount of information. This is something that had been problematic for me on the current mixing desk control software - there's just too much information there. There are a lot of faders, a lot of parameters, and I tend to get overwhelmed by that.”
Where does StudioManager fit into this? “The Studio Manager is a great piece of software with wonderful graphics. It also provides many of the capabilities of the Matrix Mixer, and others besides. However, when faced with the StudioManager screen, I find that I am overwhelmed by the amount of information, particularly if I don't regularly use StudioManager. Also, the screen presents as a mixing desk panel. I wanted to present a screen that matched essential concepts–route from source to destination, process at source, process at destination.”
What are people going to get from the Matrix Mixer experience? “They're going to get simplicity and power. Without a manual, they can navigate through the features of the 01x and fully utilize those features. It makes it a very pleasant experience. “Behind the 01x Matrix Mixer is XML. We have a framework for both the data and for the display that can very easily be transposed to other mixing desks as well. My hope for the future is that there will be the same simplicity - that once you've worked with the 01x Matrix Mixer you can work with other Matrix Mixers for other mixing desks with the same level of simplicity.
Can we expect these in 2007? “Yes, and quite early on! To start with, the 03D and 01v96.”
Do you recommend using Matrix Mixer with StudioManager. or is there a danger of having too many controllers? “The way I would do it is to use the 01x Matrix Mixer and get familiar with all the features that the 01x has. I think that in the majority of cases it will probably be all you need. But there are features in StudioManager, for example graphic displays of equalization, that when you've got familiarity with the functionality of the 01x you can then go across to StudioManager and use such features; see what other features are there. Of course, if you like the mixing desk layout, and feel more comfortable with this, then StudioManager will suit you better at this stage.
How common is University involvement with commercial companies? “In general it is becoming more and more of a phenomenon, one of the reasons being the low salaries that University professors generally receive! However, from a more academic point of view, when you're working with companies you're very often at the edge of research. In our case we simply could not have got where we were without working with Yamaha. It was wonderful the amount of knowledge we got from them, and we've been able to write a number of leading edge papers on the subject of FireWire in professional audio. “We have started work with another South African University on a hardware project–creating bridge hardware that will allow for the increase in the number of FireWire audio devices that can be in a network.
Finally, where do you see mLAN in 5 years time? “I see standardization of all aspects of the technology. I also see a spread into other areas besides the studio environment such as live sound, installations, and broadcast, possibly even the home, and then an integration with video transmission.”
Thank you Richard.
|