Gibson to Embed Guitars with Ethernet 474
caseyuw writes "Gibson is planning to roll out their Magic this year with the delivery of guitars using Cat 5 instead of analog cables to connect instruments and amplifiers. The debate over the quality of digital vs analog signal processing is not new, but using a 'Magic' Les Paul would force you entirely into the digital domain." We mentioned this last year, but the above article has much more information.
Re:Ethernet, not Firewire? (Score:5, Informative)
Also, the max length of a FireWire cable is 4.5 meters [firewiredirect.com], while Ethernet can do 100 meters [homenethelp.com]before needing a repeater.
Not sure how much bandwidth a gee-tar takes up, but I'd bet that cable length was the deciding factor in this design.
Not the problem (Score:3, Informative)
Re:Wireless? (Score:5, Informative)
As a guitarist, that seems good enough.
Re:Wireless? (Score:5, Informative)
Might be that 54mbit wireless has good latency though.
One thing that annows me about the main post is the statement that Magic will 'force' people into digital. This is nonsense of course.
From the article: Those initial Magic guitars will also have traditional analog pickups. "It will essentially be two guitars in one: You don't have to go digital if you don't want to," said Arora.
Wrong... (Score:4, Informative)
Also, the max length of a FireWire cable is 4.5 meters, while Ethernet can do 100 meters before needing a repeater.
Not sure how much bandwidth a gee-tar takes up, but I'd bet that cable length was the deciding factor in this design.
From Apple's Firewire 800 [apple.com] page:
"FireWire 400 delivers data over cables of up to 4.5 meters in length. Using professional-grade glass optical fiber, FireWire 800 can burst data across 100 meter cables."
-T
Line6 already won in this domain (Score:4, Informative)
Now, the laziest could also check out Steinberg's Virtual Guitarist [harmony-central.com]...
Re:Has to be said (Score:1, Informative)
Re:Wireless? (Score:4, Informative)
You'll never get that across today's wireless LAN technology.
Re:Please, lord. . . no! (Score:3, Informative)
Try to keep up, like Jimmy does.
[OT] MIDI over Ethernet - DMIDI (Score:3, Informative)
The original UDP version, from a performance timing perspective, was tight and the network was transparent to musicians. The Ethernet version seems to be even tighter!
http://www.dmidi.org [dmidi.org]
Re:Hey there's a video presentation (Score:5, Informative)
Read the specs, it's all open. Biggest advantage will be user created software synths, better compression/normalization, it's adaptation of highly availiable technology and it's open sourceness (which I know the mods will love me for mentioning)
The one major weakness with midi is it's ring topology. This is just straight up ethernet, any topology that ethernet supports this new standard will.
Here's a quote from the PDF specification.
1.
Physical Layer
: consists of the mechanical and electrical specifications required
to form the physical network. This layer is compatible with the IEEE 802.3
Ethernet physical layer.
2.
Data Link Layer
: as defined by the IEEE 802.3 Ethernet protocol. It views bits
transported by the Physical Layer as defined sequences called frames that can be
transported across any standard Ethernet-compatible network.
3.
MaGIC Application Layer
: uses the frames transported by the Data Link Layer to
encapsulate MaGIC-specific information into packets that allow MaGIC devices
to exchange real-time bi-directional audio and control data.
The MaGIC application layer is independent of the two layers under it thereby providing
the ability to easily change the mode of physical transport based on available technology.
As you can see, it uses just good old 802.3 Here is a list of what Mechanical interfaces it works on.
2.4 Mechanical Interface
The MaGIC protocol is suitable for a variety of physical interfaces. Examples include:
the IEEE 802.3 Ethernet physical layer, the high-speed multi-link Optical Interface,
wireless interfaces, the Ethernet Gigabit-based physical layer, etc.
This specification only describes the MaGIC Link based on the IEEE 802.3 100-Megabit
Ethernet physical layer, which uses standard Category 5 (Cat 5) cables, and RJ-45
I could go on and on about why it's so much better than midi. Check out the pdf, it's got more info than the videos.
tubes from an engineering standpoint (Score:5, Informative)
Now, here are some reasons why tubes might sound better:
first of all let's start with some tube basics: you heat a plate (cathod) and electrons jump off it. the electrons pass through a grid, and gets obsorbed at another plate (anode). You can vary the voltage on the grid and control how much eletrons pass - hence the amplifying.
The difference between a tube amp and a FET amp is that tube amps have some insane amount of dynamic range that is very nice and linear. somethinge like 40V (or more, depending on the tube). It goes by the name "high voltage, low current."
Now, for the same power, FETS can't touch this range because most fets don't operate at that high voltage level - and if you push it then it will saturate / turn off and you won't be linear anymore.
So for the same power, FETS would go toward "low voltage, high current." This is cool and all, and theoretically if you stay within the linear region you are all good, right? wrong. All the EE books teaches you one thing that you never do in the real life - that is to assome a nice ground.
ground is never nice - especially when there is a lot of current, ground tend to float here and there - which would give you crap and distortions that we all know and love. Of course, throughout the years engineers (hey we don't have a life, after all) figured some ways around it - but AFAIK all of these are either 1) very expensive, and 2) not completely effective (usually it's both). (btw, one of these is to make as much of the system digital as possible.)
So... In the end, tube amps still reign. I heard that RCA made the best tubes, no confirmation on this, though.
Just for the few who thought "well when we get lots of superconductors then finally FET amps will be better!" That's not correct either. Unfortunately superconductors we know of are only good for no resistance at DC, and the ground does not play nice because of AC concerns.
So, there you have it. For the record I don't know any engineers who thought "oh yeah I can duplicate a tube response through other means," but they might have told their bosses shit like "I can make it damn close and you can't tell the difference" (which is usually a lie) so to keep their jobs.
And Tubes are considered solid-state. A tad fragile (there are stainless steel ones for the military, if anyone is interested), but still solid state last I checked...
Re:OT question about transmission lines (Score:1, Informative)
In fact, electrons move slowly - there's no reason for them to travel at the speed of light, although for some reason people are led to believe that this is the case. Indeed, if Einstein is worth his salt, it would not be possible for an electron to move at the speed of light (they do have a mass).
In wires, electrons actually move at speeds measured in CMs per second (once again, I forget the actual figures, but it's either CMs or MMs). The reason that it appears so much faster is that, at the risk of oversimplifying things, the whole chain of electrons is shunted along when the first electrons move. As an analogy, imagine holding a long stick and moving it backwards and forwards. The actual movement at the other end of the stick would appear almost instantaneously, despite the fact that the stick would never actually be traveling more than say 20MPH.
It's the same with free electrons moving through a conductor - they may not move fast, but as electrons are passed to atoms, other free electrons in the vicinity are passed along - the net result is very fast data transmission.
That answers the question of why it goes so fast, and hopefully you see why that is the question, rather than why it goes so slow...
Re:Wireless? (Score:3, Informative)
That's what you might get with modern cards and, say, ASIO drivers/Cubase VST.
Re:Wireless? (Score:5, Informative)
On the other hand, sound travels through air at about 1 foot per millisecond (roughly). So that 100 ft cable would put Jimi 100 ms away from the monitor speakers. At 4/4 time and 120 beats per minute that is almost a quarternote of latency. Clearly no riff-master would ever want to be off by that much.
Likewise, you can think of latency as "adding distance between you and the speaker". For example, 1 ms of added latency is like adding another foot between you and the speaker. For most applications 5 ms isn't going to be noticeable, but the signal chain for a concert or studio can be long. And all those 5 ms delays really add up.
Gibson Magic is really just a CobraNet wannabe (www.peakaudio.com). CobraNet has been around longer, is more of an established standard, and has more sophisticated network management and routing than Magic. In contrast to CobraNet, Magic is a latecommer that was developed by people who should stick with guitars rather than 100Base-T. More to the point, CobraNet is supported by more than 30 different companies while Magic has maybe one supporter if you don't count Gibson itself.
And isn't this just a repeat post? It seems that Gibson Magic pops up here every so often but that they don't have any real new news...
Illegal != unsponsored (Score:2, Informative)
The issue the RIAA has with is recorded music [the "R" in the acronym]. Performing someone else's music live is completely legal [if unsponsored].
The only way you'd be 'taxed' is if you released your recording of someone else's song...which is a royalty payment. And btw, you're already taxed playing cover songs if you perform - the clubs [big ones anyway] send you a 1099 for taxes, and [IIRC] either the company hiring you or the venue itself has a license [ascap/bmi?], lest the local music union folks visit.