howto pick-up 6 individual string sigs from guitar?
Hi all!
i'm looking for some sort of pick-up/mic or element system
that will pickup the signals of the 6 individual strings on
my guitar so i can use it as a controller for digital instruments,
anybody have some good ideas on this?
thanx a load
I think a google search for midi guitar pickup would be a good place to start :) This type of thing is very tricky to do digitally. It you don't want to separate the actual audio of each string into its own channel (not a realistic project for maxmsp imho) then midi would do the trick.
edit--- If you could manage to find any info on how the EHX POG circuit tracks polyphonically that would be very interesting to learn about as well!
Hi
Checkout any Roland compatible system ala GK 2, RMC etc . . .
Then get (or make) a breakout for the system Roland RMC etc all do something along this line
Then a six ( preferably 8 or more) input soundcard and your away
Doing something similar myself with VG99/GK2 equipped guitar etc . .
Max is definitely a place to be experimenting with this stuff, have fun . . .
Let us all know how you go
Cheers
Macciza
Great guys! thanx alot for the help
do you mean breaking the signal after it has passed through one of them
synths they associate with the elements, or you just use the element
and immediately breakout the signal from there?
dear Kazouki
The system is quite simple. You have and hex pickup (I prefer the RMC, more expensive but much higher quality). You can check the Roland one as it is an easy add on and there is also the ghost system that I have heard good from it.
The only thing that does is to have 6 pickups in one, i.e. one per string.
Then you plug that in a 'brain'. The options are:
1) the classic: a guitar synth machine à la VG99
2) the more open: a simple, quick HEX to MIDI converter: the fastest is the Axon AX-50 USB, or its big brother the ax-100 (same speed more guizmos)
3) the more roots: you take a fan-out and go to 6 pre-amps straight into MSP and start fiddling. There is a good paper in ICMC 08 (ADAPTING POLYPHONIC PICKUP TECHNOLOGY FOR SPATIAL MUSIC PERFORMANCE
Enda Bates Dermot Furlong Donnacha Dennehy) even giving the pinout for a more DIY approach.
My setup is going towards 2+3
pa
perfect! thats extremely clear, thanx a million
greetz thom
Here's an interesting product:
http://www.keithmcmillen.com/products/stringport/
Full disclosure, I do work for this guy.
mz
dear mz
thanks for that pointer. The only question is: how quik it is compare to the axon?
pa
Not sure. I expect the full hardware solution could be faster, but I know the KMI guys are working hard to keep latency down.
mz
Hi
Not sure that Stringport actually 'sends'midi even internally but it does provide interesting analysis data from the input that can be used to control effect parameters etc
Probably closer to a Roland style Virtual Guitar than an Axon midi setup . . .
Definitely having a bit more of a look at this setup . . .
Cheers
Macciza
A PhD colleague of mine at the University of Ulster is doing exactly this kind of thing, and has built himself a breakout box for multichannel audio and midi from a 'standard' hex pickup, his name is Ricky Graham:
http://www.rickygraham.net/
Brendan
About the StringPort:
"I expect the full hardware solution could be faster, but I know the KMI guys are working hard to keep latency down."
The latency is really ultra-low. I'm playing with in Max5 / AU Lab / Logic Pro: perfect.
(I do NOT work for this guy :-)
Cheers,
PG
'The latency is really ultra-low'
I like this testimony, but I am more likely to believe you with ms measurements... I might be less tolerant than you ;-)
pa
(Not cheap, the StringPort: I'm not tolerant at all about my money ;-)
Well... a stereo recording of the direct output of a guitar on one channel (say, left) and, in parallel, the output of the StringPort on the other channel... ?
Or something else I could do?
PhiG
Don't you think the latency depends on the pitch too ?
Because if the lower sound of a guitar is a E1 = 82 Hz, you'll need 13ms
to have a complete sine and detect it with 100% cert.
And guess… they invented vibrato arm and bass guitar :)
Your test will be very interresting PhiDjee
"Don't you think the latency depends on the pitch too ?"
Yes, but: say that I'm testing an audio interface (MotU, RME, Apogee, etc.), would I test the latency for the lowest notes? No :-)
"Because if the lower sound of a guitar is a E1 = 82 Hz, you'll need 13ms to have a complete sine and detect it with 100% cert."
Hmm, this thread was about a breakout box, not about pitch detection.
"Your test will be very interesting PhiDjee"
I'm testing again and again and more I'm testing, more I'm starting to play :-)))
Tests are done with:
- fretted and fretless guitars,
- 5 and 6 strings fretted and fretless basses,
- 5 and 6 strings violins (fretless only :-).
Tell me if can report something interesting for you.
Toast ;-)
Philippe
you guys are awesome.
can this be expanded for an 88-note piano? hehe
...or maybe a home-grown "harp" that uses super-cheap strings? I mean, once you sense the sound/amplitude, it doesn't really matter what the pitch is, you just want the action to trigger other stuff...
anyway, fun thread :)
"can this be expanded for an 88-note piano? hehe"
Oh yes, this is called a piano, by Yamaha.
(What?)
"...or maybe a home-grown "harp" that uses super-cheap strings?"
And this one too: http://www.apple.com/quicktime/guide/hd/animusic_rc.html
:-)))
Well, [expr no fun == no music]
PG
Just to make it clear once more, there are basically two methods to get the signals from individual guitar strings for later processing:
----------
Audio
You need a hexaphonic pickup, be it magnetic, piezo, laser, infrared, whatever. You pick up the string vibrations and get an audio signal from each string. You can send these six audio streams through six preamps and have your raw signals. To get healthy audio signals, you need a preamp for each string. The sound quality depends on the pickup, the cabling and the preamps.
A usual guitar pickup is the Roland GK, which is magnetic. Godin uses piezo pickups on the bridge. I think there are several piezo systems on the market.
----------
MIDI
This is a second stage. Once you got your audio signals from the pickup, you can convert each string signal to MIDI. There is a variety of converters available. Some have built-in synthesizers and some not. The Roland GR-20 has a synthesizer and sends MIDI, is good for live playing but its tracking is not that good. The two Terratec converters are pretty fast, the Axon AX 100 has built-in sounds which are not very good The AX 50 USB has no synthesizer but USB. And there is still the old Yamaha G50 around. I did not test it side by side against the Axon AX 100 (which I have) but from my feeling the two are equally fast.
I think for your purposes a Roland GK pickup and either an Axon AX 50 USB or a used Yamaha GR50 would fit and they are not too expensive.
----------
Custom built ...
... is no great deal but a lot of work. You can make your own small pickups and mount them anywhere you want. And you can buy or build preamps and Audio-to-MIDI converters.
----------
Another approach:
What about the Yamaha EZ-AG? I am not sure if and how it sends MIDI out but it may be an option. As a silent electronic guitar os as a fancy guitar-like MIDI controller:
http://music.yamaha.com/products/main.html?productId=189
I played one in a shop and it was quite funny. Unfortunately I didn't check the MIDI system but I am sure you can find out what this thing does. The string controller works surprisingly good, certainly good enough for software synths if there is MIDI out (what I think).
----------
> seejayjames wrote:
> can this be expanded for an 88-note piano? hehe
Actually there is not much room for invention left in this area. The Yamaha Disklavier was introduced 1986 and is still available in several flavours:
http://www.yamaha.com/Products/Disklavier.html
Because if the lower sound of a guitar is a E1 = 82 Hz, you'll need 13ms
That is the great thing about the Axon. They say (and it sounds like) that they need only a quarter of the period to trigger... slower in the lows, but faster than the Roland...
Anyway, what we need is to do A-B-C testing with a recorded note to be sure... or ask one of the manufacturer to do a public test ;-)
pa
You could not play seriously with 13 ms latency. The lower E string is no problem for a usual hardware converter, be it Axon or Roland. Their problems start with bass guitar. But they work on another level anyway, they don't just measure the frequency. Axon for example can recognize three different pick positions in a range of about 20 cm or less. Means, before a "home grown" Analyzer knows the frequency, they have already analyzed the wave characteristics and know where you picked the string. I think they don't even measure the root note but rather a harmonic frequency above. But this is my theory, I don't know for sure. It would explain the speed of the devices.
that said, the hardware are fast and usable for my 6 string basses (except the low B and E which are slow but you get used to it...)
I really would like to see an A-B-C test of these 3 hardware solutions...
pa
Hi PA
Met you in Sydney at NIME. Recently got a StringPort to add to my VG setup, Will see about doing some comparative tests if you are still interested. If you want MIDI only from your bass you might check out http://www.industrialradio.com.au - (I don't work for them but they are Aussies) It 's by Steve ChickI think who pioneered what beacame the Peavey? MidiBass back in the '80s. 5 to 10 ms delay across the board.
And you can play with 13ms delay seriously, it;s like being 13 feet from the speaker, if you had headphones on and stood 13ft back it would be the same as instant audio from 13 ft . . .
Cheers
MM