Analog, Digital and Pedalboards

This article tries to explain what “analog” and “digital” means, why you may want an analog dry path, and why a pedal board full of digital pedals doesn’t make sense to me.

tl; dr

It’s hard to exactly define “analog” and “digital”; in music equipment, “digital” mostly means that audio is converted to numerical data. Typical features of digital audio include universal processing (in theory, one device can do all kinds of different sound manipulations) and latency (digital pedals typically take 1 to 2 milliseconds to process your signal, analog pedals are much faster).

When the very idea of digital equipment is “one device can do it all”, I don’t get why people would put a large number of digital pedals on their pedal boards. Especially not, when those pedals don’t come with MIDI controllable presets and just emulate analog pedals that you can get for the same, or lower, prices. My current pedal board choice is “mainly analog, with one or two digital pedals, in case of delay/reverb preferrably with analog dry path”, but I also have sympathies for the “one digital main device, like a Helix or a FX8/AX8; adding only a few extra analog pedals” concept.

Concerning marketing strategy, I’m not a fan of the Strymon / Source Audio approach of having several products with very similar hardware, but different algorithms. I prefer the Line 6 “everything is included” approach, or the Eventide one of “buy a core H9 and buy additional algorithms later”. For the future, I hope for a common standard and platform for digital guitar effects, similar to what VST is for DAWs: A system allowing me to have one digital device that loads algorithms from different manufacturers like EHX, Source Audio, Boss, Strymon, Line6, Neunaber and others. 

Analog and Digital ... what’s the difference?

It’s not easy find a accepted definition of what “analog” and “digital” actually means. Currently, there are no articles on “analog” or “digital” in the English Wikipedia; there are however entries for “Analog Signal” and “Digital Signal”. The latter explains: “A digital signal is a signal that is being used to represent data as a sequence of discrete values; at any given time it can only take on one of a finite number of values. This contrasts with an analog signal, which represents continuous values; at any given time it represents a real number within a continuous range of values”. An okay definition; still, it can hardly explain the difference between analog and digital photography.

There’s a consensus that old-school photos taken on a light-sensitive celluloid film, vinyl records or all-tube guitar amps are/work “analog”, while a MP3 file, a Netflix video stream or a Kemper modelling amp would be called “digital”. So, what are the characteristics?

Let’s start from: There’s something in the world that you want to reproduce technically. Like, a beautiful view that you want to conserve for the future. Or an animal making funny noises that you want to send to your best friend so you can both have a laugh.

The analog approach would be: Take some chemical or physical device that reacts to the incoming rays of light, or sound waves, in a proportional way. Like, the more light falls through a lense on a certain spot on a celluloid film, the darker that spot gets. Or, a membrane that vibrates from incoming sound gets connected to a stylus that cuts the vibrations into some material. The grooves in the material (e.g. a vinyl disc) are proportional, or analogous, to the air waves from the original sound. Now, you can connect another (less sharp) stylus to a big cone and play back the sound. This works even completely without electricity (old Grammophones were hand-operated). Analog electronics work similar: Like, in an electric guitar, the movement of the vibrating steel strings in relation to the magnetic pickups creates an electric signal pulsating proportional (analogous) to the string vibration. 

The digital approach would be: You measure the image or the sound by a defined grid and convert the measurements to numbers. Like, you measure how much light falls on each of 2 million defined points along a grid, a represent each of those points with a value from 0 (very dark) to 65535 (very bright). Or, you measure 44100 times per second “at which position is the vibrating membrane of my microphone currently”, from 0 via 32767 (neutral position in the middle) to 65535.

So, “analog” is defined by proportional signals, and by continuity. “Digital” is defined by chopping up the physical reality by making measurements at certain time and/or space intervals (grids), and storing each of these measurements as “one out of a range of possible values”.

Pros and Cons of Analog vs. Digital Technology

Historically, analog technology came first; it’s easier to build. This also means that many art forms that we love – photography, movies, pop music – have grown in the analog medium. Classic cinema look is defined by the specifities and limitations of celluloid film. The ever popular distorted electric guitar sounds are based on early analog tube amplifiers being pushed beyond their limits. So, we are part of a culture that has evolved around analog technology, and that might be the reason why in some areas, analog represents beauty for most of us. 

There’s however also some “hard fact” advantages of analog technology, including:

• It distorts softer (tube-based analog circuits even softer than transistor-based ones). When digital devices reach the limits of their measurement range, there’s a hard cut. For analog devices, there’s mostly a gradual transition. (Most digital overdrive/distortion guitar pedals emulate analog distortion; you don’t actually overdrive their circuit).
• It’s faster. At least for audio. Analog pedals and amps will process incoming audio almost immediately. Digital pedals and amps always have a slight lag (latency).
• It can be more durable and easier to service/repair (mostly true for oldschool analog stuff like 1960s tube amps; less true for modern analog devices with ICs and SMD components).

Digital technology has other advantages; the main ones being:

• Once recorded, storing and transmitting digital data will not degrade the data or introduce noise (unless you use lossy compression formats like JPG and MP3).
• Since the digital representation of sound/video is available as data/numbers, you can do all kinds of calculations/modifications to them; including stuff that would not be possible with analog electronics (such as polyphonic pitch shifting, or simulating actual room reverb).
• All of those modifications can be done with the same device: A computer. While in the analog domain, you need different circuits/devices e.g. for dynamic compression, equalisation, adding echo, making a multi-track recording of various audio signals and mixing them – in the digital world, one device can do all of this, with the same universal computing technology, just using different algorithms.  

One thing that can be both positive and negative is that analog equipment “lives”: It will produce slightly different results each time you use it, it may be sensitive to temparature ... while digital equipment always will sound exactly the same, unless you deliberately implement randomisation.

Here’s a very interesting Brian Eno quote: “Whatever you now find weird, ugly, uncomfortable and nasty about a new medium will surely become its signature. CD distortion, the jitteriness of digital video, the crap sound of 8-bit - all of these will be cherished and emulated as soon as they can be avoided. It’s the sound of failure: so much modern art is the sound of things going out of control, of a medium pushing to its limits and breaking apart. The distorted guitar sound is the sound of something too loud for the medium supposed to carry it. The blues singer with the cracked voice is the sound of an emotional cry too powerful for the throat that releases it. The excitement of grainy film, of bleached-out black and white, is the excitement of witnessing events too momentous for the medium assigned to record them.” 

While analog music technology always operates in the range of (some) failure, digital music technology rarely does; it rather emulates failure. Making it more reliable and top-40-band-proof, but maybe less interesting artistically. If you want to here real digital failure, listen to the self-oscillating digital delay at the end of Radiohead’s Karma Police, or get some obscure pedal like the Iron Ether FrantaBit or the WMD Geiger Counter. 

The Grey Area in Between: Bucket Brigade Delays

Is there such a thing as “half digital”? If we define “digital” by chopping up all dimensions of a continuous signal into discrete samples, then the answer would be “yes”. Movie recorded black and white on celluloid would be an example: Space and brightness are continuos here, there is no fixed grid (well, maybe there is discontinuity on molecular level, but not with the exact same grid for each frame). Time however is not continuos, but chopped with a “sampling rate” of typically 24 images per second. 

Another example would be classical analog delay pedals. Other than real tape delays, that are without question fully analog, those pedals work with bucket brigade (BBD) technology: An analogue “sample” of the guitar signal is stored in a capacitor, and is then moved through a line of thousands of capacitors, one step at each clock cycle. So, BBD delay pedals have a sampling rate; which is typically changed when altering the delay time. This is why at very long delay times/low clock rates, BBD delays create artifacts similar to low-bitrate digital signals.

Still, neither celluloid movies nor BBDs work with numerical data that can be arbitrarily processed/calculated; that’s why it’s probably more accurate to call them “discrete-time analog” rather than “semi-digital”.  

Latency, Conversion Loss and Analog Dry Paths 

As said above, digital audio devices always introduce a small measurable delay (latency), while analog devices don’t. This is because the process of converting the signal from analog to digital, running it through a microprocessor at a certain clock rate and converting it back to analog consumes more time than electricity running through analog circuitry (which happens almost at the speed of light). 

There’s little information online about the actual latency of different digital pedals; in a thread on The Gear Page, these figures are listed:

• Eventide Timefactor: 2.3 ms
• Torpedo CAB: 1.7 ms
• Strymon Deco: 0.8 ms

Recently, I did some own measurements of my Line6 M5 and HX Stomp, and found:

• Line 6 M5: 1.2 ms
• Line 6 HX Stomp: 1.55 ms (activating the FX loop doubles this to 3.1 ms). 

This is not much latency. At room temperature, sound takes about 2,92 milliseconds to travel one meter, so already standing only one meter away from your all-tube guitar amp speaker will give you more latency than any of the above listed pedals. Some older pedals may however have significantly more latency. Also, if you use the Eventide H9 in pre/post mode, its latency will double (documented here). 

By the way, distance does matter. We once tried to record our drummer and bass player standing about 10 meters apart from each other in a studio; we needed to give each of them both drum and bass on their headphones, else the natural 29 ms latency between them would have killed the groove. 

Latency of digital pedals may become a problem when you chain several digital pedals: Their latency will add up, and so will signal degradation introduced by analog/digital/analog conversion in each pedal. Will this make an audible diffference? Well, it depends – on the actual pedals that you chain and on your sensitivity to this kind of stuff.  

That’s were analog dry paths come in: Effects like echo and reverb – which leave the original signal intact, but add repeats/room on top – can be built to have an analog dry path: Only the effected (wet) part of the signal will run through A/D/A conversion. In delay and reverb pedals, the wet signal is delayed anyway per definition, so there, latency doesn’t matter. Best of both worlds, if you ask me. There’s lots of pedals with analog dry path nowadays, including:

• All Strymon delay and reverb pedals
• Boss DD-500 and RV-500 
• Source Audio Ventris 
• Empress Echo System, Reverb and Tape Delay
• Neunaber Wet

There’s still even more delay, reverb and multieffect pedals without analog dry path, including:

• Everything by Line6
• Source Audio Nemesis
• Empress Superdelay
• Eventide Timefactor, Space, H9

A well made modern digital pedal may introduce so little latency and conversion loss that an analog dry path is not really necessary (check this guys vid). However, I can definitely hear what the digital dry path of my Line6 M5 does to the signal; providing the M5 with an external analog dry path (like, in a parallel loop of my ES-5) does make an audible difference, which you can hear in this demo I made. 

Why a Board Full of Digital Pedals Doesn’t Make Sense to me

We learned above that a core feature and strength of digital technology is its universality: Provided with the right algorithms, one device can do it all. That’s why it feels very strange to me when people put 6+ digital pedals on their boards. Especially when they all come from the same company (typically Strymon, sometimes Source Audio).

To me, it would make much more sense to have ONE Strymon device, and to be able to load their various algorithms in there. Of course, it’s a clever marketing strategy to have people buy 7 different pedals of which the 3 big ones, and the 4 small ones, share very similar hardware, but have different algorithms (and colours). But I prefer Eventide’s strategy of selling a core device (the H9) and allowing you to buy additional algorithms later. I would love to replace a few algos on my Strymon Timeline (like Digital, Dual, Swell, Trem) with the ping-pong algo from their DIG and some of their finer reverbs; and I would gladly pay some money for that. But I have zero motivation to spend 832 € on a BigSky and a DIG that wouldn’t fit on my pedal board anyhow.

Also, cumulating conversion loss bothers me: Why don’t we have digital audio jacks on our guitar pedals, when my HiFi equipment had them already 27 years ago? (According to Source Audio, most digital audio ICs are prepared for SP/DIF interfaces anyhow, so they may incorporate that interface in future pedals). 

So, when going mostly digital, for me, it makes much more sense to have all digital stuff going on in ONE device. Like, a Line6 Helix or a Fractal Audio AX-8 (of course, there’s also Kemper, but they are not that well-developed in the effects section). Then again, I don’t like the idea of all algorithms coming from one company. The fun (and artistic value) of building a pedal board for me comes from trying different stuff and finding my own individual combination. Of course, this is also possible with a Helix or AX-8, with all the different algos they offer. Still: In 2010, I saw Broken Social Scene in concert. One of their guitar players had a Line6 M13 as his pedal board. Having played a M9 for quite some time, I was constantly thinking “yep, I know that sound ... and again, something familiar Line6-ish”. 

So, what I would really want is: A digital standard for algos from different manufacturers. Like the VST standard for digital effects in DAWs. There could be several companies manufacturing the hardware, with different user interfaces and horsepower; but they should all be able to load any algo that follows the standard. I want ONE box with a flexible signal chain that can give me EHX POG2 sounds, WET and Flint reverb algorithms, Nemesis Tape Delay, DL4 Auto Volume Echo, Kemper amp profiles ... and I’m ready to pay for those algos, if the price is reasonable.

As long as manufacturers won’t come up with this standard, I will gladly stick with my board consisting mainly of analog pedals (some of them digitally controlled), with the Strymon Timeline as digital pedal with analog dry path, and the M5 as the only fully digital pedal (which I normally only use for special FX sounds, or in a ES-5 parallel loop for Tape Delay sounds).