I recently strapped-on the trusty 6-string and went-out to make some noise. I have a modest pedalboard with all true-bypass pedals (that means that when they are switched-off there is nothing but a switch and a bit of wire in the signal path - no electronics at all). Now TB pedals are held by many tone-facists to be the best way to preserve the thoroughbred sound of our beloved guitars, and for my sins I'm one of them. If it isn't true bypass it gets modified or it doesn't go on the board, and that is my position. Absolutely and without doubt! Except that...
...and if you don't know who Pete Cornish is - well, he's a (really) big hitter in the effects and electronics world and he favours buffered systems. And what's a buffer? In this context it's unity gain (or close) amplifier with a high input impedance and a low output impedance which helps to drive a guitar signal to minimise the tone-robbing effects of capacitive cables which act as a high pass filter to earth.
Cables dull your tone, long and/or poor cables dull it more, switches and sockets make it worse. To a significant degree low impedance sources are immune, high impedance sources are not (note that contrary to the opinions of many experts on the web, you don't get high or low impedance signals, the electrons really don't have a clue about this stuff, they just go wherever seems easiest to them, the output and input impedance of your kit helps them decide)*
Fancy to pass a few hours and end-up none the wiser, Google "do I need a pedal board buffer" and read a few of the 305,000 hits. OK? Want a proper answer? Read on.
Ready? - here it is - "I don't know". I can't tell you, but a listen to this might help you to decide for yourself.
I figured that I could record a short piece of audio with and without a buffer and listen to the difference to see what it did to my signal. So far so easy, but here's the small print (I've typed it the same size to make it easier to read).
· Firstly you need to decide what is the "standard" that you will reference. In the simplest case, plug your guitar into your amp with a single lead. Let's work with this.
· Then you need to find a way to play the test piece exactly the same for each take. I used a Line 6 recording at 44.1kHz 16 bit - it's not perfect and it introduces a couple of levels of AD/DA conversion but it is consistent. I plugged this in with a short 6" cable.
· Then you need to reduce the variables as far as possible, so I took the output from the pedalboard directly to an instrument input on a DI and recorded that. All of the pedals are bypassed, and the buffer, in this case from a Boss NS-2 pedal (because it was sitting looking at me) was in a switchable hard-wire bypass-loop. This simplifies the signal path but does result in a slightly "quacky" recording. Boss pedals have a buffer which is active when the pedal is bypassed, ie not true-bypass, they are one of the better considered mass-produced pedals and this unit is found on the pedal boards of some big-big name players. Opinion is very split - some say it's tonally transparent, others say it could suck-start a Harley; it's all in amongst those 305K Google hits.
The recording is of some very simple open chords played on a Bill Puplett strat with Kinman pickups with slightly tired strings. This is a very quiet guitar and good leads were used throughout.
I recorded 4 passes into Cubase. The files are normalised because the Boss buffer isn't EXACTLY unity gain but are otherwise completely raw.
- Buffer bypassed - the Back Track runs through wire and bypassed pedals into the DI.
- Buffered - as above but with the Boss buffer at the start of the circuit.
- Noise Suppressor - as above with the pedal active.
- Direct - just a short lead between the Back Track and the DI.
- The above edited into a single take.
All the passes are sample aligned and unprocessed.
Audio sample 1 - Buffer Bypassed
Audio sample 2 - Buffered
Audio sample 3 - Noise Suppressor in
Audio sample 4 - Direct
Audio sample 5 - Edited together
Just to put a perspective on the difference, here's Audio sample6 - the edited together version with an amp sim
Nulling; this is where you flip the polarity of a signal and add it to another, if the source signals are identical then the flip completely cancels the signal and you get silence. If they are not identical then the result is the difference, so the less the result the more similar the sources? Make sense? OK
Here's the really telling bit
Audio sample 8 - Buffer Bypassed nulled with Noise Suppressor in
Audio sample 8- Buffer bypassed nulled with Direct
In this case none of it makes very much difference, and I don't need a buffer. YMMV.
Oh, but there's a but, and it's a big but! You may prefer the sound with one, or without, or with high capacitance leads, or whatever. This is a test - it doesn't actually give answers! 42! Many of my favourite guitarists use junk signal paths and don't care at all about maintaining signal integrity at all - only what comes-out of the speakers. Take a look at some of the Rig Rundowns in Premier Guitar magazine (it's available free online). Eric Johnson, the man who can apparently hear a difference between different brands of battery in a pedal has some Boss, and the connects his pedals with huge loops of cable because he prefers the sound; and Paul Reed Smith whom I had the pleasure of meeting some years ago built a "sweet switch" into his early guitars to mimic the sweet HF roll-off of a long cable. Did that help?
So that's me sorted, add a comment and tell me what you think.
*This isn't actually LITERALLY true.