I got introduced to music-making at age 9, way before I started with Design. However, as of today, Design has been paying my bills for more than 16 years, and I’m still a bedroom musician. That’s why it has been inevitable for me to look at the different music-making software that I regularly use with a critical eye, and I have spotted some curious tendencies and practices that deserve to be analyzed in and out of its context.
I have found a distinct trend in apps and software which pretend to simulate gear that exists in the “real world.” The vast majority of this software is centred around simulating effects pedals, amps and cabinets, signal processors, and modular synthesizers. Is evident that there’s an obsession to follow the terms established by hardware gear to great heights. They are betting big on skeuomorphic design. I was curious to see how beneficial or detrimental this trend was for their experience.
As a note, most of these samples are presented on a laptop, because conventional sound processing needs more processing power (and a proper Audio Interface), which also means there is a more significant offer of products on this platform. Yes, nowadays mobile devices are powerful enough to process sound with low latency, but any decent musician will leave it for rehearsal and recording basic ideas.
Analog vs Digital
Before we start, I feel like I should make a definition of terms we will use.
When speaking about real-world devices and their software counterparts, it would be easy to refer to them as analog vs. digital. An analog notepad on your desk vs. a digital notepad on your tablet, for example. In the context of music production, these terms mean other things.
You can have two physical pedals that aim for the same effect and look similar on the outside, but one can be analog and the other digital. Is all about how the signal is processed inside of the pedal. Analog effects can be achieved using ‘simpler’ electronic components maintaining the original signal (albeit transformed with the effect) that comes from the guitar. Digital pedals take that same signal, transform it into digital, process it using more complex electronics (microchips running software) and they turn it again into analog. Then, in both cases, the amp can take the signal and push the air that hits our ears and is interpreted by our brains as sweet music.
A similar thing happens with synthesizers but its case is more about how the waveforms are generated and processed. If they're created by electronic components that transform a voltage signal into a wave (analog), or if a CPU generates it (digital).
There’s an eternal debate between which one is better. How analog technology is ‘warmer’ in sound and how digital go close but no cigar. If you’re like me, probably cannot differentiate between the two most of the time.
If you start digging into this topic, you will find the term VST or Virtual Studio Technology. VST is the umbrella term for all software that can replicate studio hardware; from an instrument to a compressor. Steinberg created this technology in 1996 and now is an industry-standard. Most of the software described in this article runs as VST plugins on a Digital Audio Workstation or DAW (which is the software where you record, edit and mix).
More about DAWs here
With this in mind, I'll use “physical” or “real-world” for one realm, and “simulated” or “virtual” for the other.
Replicating the world
I came in contact with the word of simulated gear for the first time when I knew about Amplitube, an app that lets you connect your guitar to your iPhone or iPad using a connector called iRig. This first contact happened in the middle of the boom of skeuomorphic design . Everything in our mobile devices looked like their real-world counterparts, and this piece of software wasn’t the exception. Probably wasn’t a decision based only on a design trend, because not only Amplitube also existed on desktop devices at that time, but also because as of today, the app still looks very similar.
One of the reasons behind the skeuomorphic design was to use the touch characteristics of these new mobile devices and establish a direct relationship to the physical object which represented. Users can turn the page on their book app in the same way they do it in the real world, bringing a well-known behaviour to this new world of touch interfaces. These inherited patterns make a lot of sense in the context of touch devices, but this audio software is used way beyond the mobile realm, mostly as DAW’s plug-ins on machines that are usually more powerful than a smartphone.
Somebody may consider Amplitube as an entry-level software, and that the real pros would never use something like this. After all, this kind of skeuomorphic trend is very evident in one of the most popular entry-level DAW: Apple’s GarageBand. So, what about the professional software used on the big studios, like Avid ProTools (which costs almost $1k a year)?
A ProTools's compressor plugin looks like this:
It costs $300 along with other plugins that simulate famous compressors and is made by Avid (not a third-party). This one was explicitly based on this $3,800 beauty:
Also, have in mind that ProTools looks like this on a regular day of work:
Is not like they have space to spare on the user interface to justify the use of that needle indicator instead of a simple number, or is not like is going with the overall design of the software.
This kind of decision seems to be directed to hit an aspirational nerve. There are quite a lot of plugins that try to emulate the work of the original LA-2A, why no go with the one that looks like it? It also seems there can be beneficial for people that learned to work with the original gear, making the transition to software gear way easier (one of the main objectives of skeuomorphic UI design).
There are other scenarios for this aspirational case. We have the bedroom musician that cannot afford physical effect pedals, and that wants to experiment and make music with them before even thinking of investing in one. What better way to say “it sounds like it” than operating like it and looking very close to it. If someday you get the real deal, it will be like you already use it before.
Create a visual link to the real gear is the rule. Most effects pedals have the same physical features (a box with a few knobs), but they can be easily related to the (famous) real-world version just using colours. An orange Phaser is easy to link to the popular MXR line, and a green Overdrive can be immediately identified as an Ibanez Tube Screamer.
Brands know about his. That’s why in some cases this trend goes beyond aspiration, and it works more like a promise. Amplitube serves as a good example, having components from different brands which aim to give the closest experience to the physical gear. Both in visual and in sound.
Reverb.com, the online music store made a comparison between a physical Fulltone OCD pedal and its plug-in.
You can judge here if they are fulfilling their promise of true-to-life emulation.
There’s a specific part of a guitar emulation setup that is hard to imagine without a certain degree of skeuomorphic design. Guitars, even when they can be directly connected to a recording console throughout a Direct Interface, are usually recorded just locating a mic in front of an amp, and capturing the distinct sound of it. Everything on this step is a variable, but how to manage the location of the microphone and the distance to the speaker or speakers? A visual approach might be the obvious way to go since there's no other easy way to translate all these variables to more minimalistic terms.
There’s a final argument for the use of a realistic approach. Users can double-down on the use of simulated pedals and buy a controller to manipulate them as physical ones, undoubtedly useful for live performances. In this case, the software becomes a middle-step, and the real interaction keeps happening on a physical object that works as other pedals. If the user is not going to be manipulating things on a screen, it seems that the apparent limitations of the skeuomorphic design are minimized.
Not Plug-n-Play
The mobile world that once was plagued with realistic textures and photo-realistic emulations have since moved beyond that. Not only as a design trend, but people now have been adopting mobile patterns that used to be new and strange, and they don’t need real-world cues to guide them through an experience. We already saw a few reasons why product designers might be still attached to this type of hyper-realistic approach. Now the obvious downsides of this method.
There’s a logistical downside to this pursuit for graphics realism. I love how it looks, but most of these true-to-life pieces of gear are fixed to a maximum resolution that it seems off for today’s screens, especially in laptops. The windows are less larger than desired, at least for me. Less size means the controls are less accessible.
On iPads, most of these apps look amazing because iPad users find more value on this. iPad apps must look sharp on that beautiful screen. However, on the side of the development, these interfaces make an app bigger on disk space, require much work, and they start looking old faster than other design trends (not that the user needs to care about any of these things, these are development issues).
Industrial Design must work its way out to accommodate to impositions and limitations of the material world. Most importantly, they have been designed to be used in entirely different contexts. For example, pedals need space to guard a circuit board, free space for the footswitch (you don’t want to step on the knobs) and probably an internal 9v battery. The same with tube amps, which are heavy because of the magnets inside and they need room and airflow for the delicate tubes. Why are we replicating this enormous space and wasting precious real-estate on digital interfaces? The most prominent example of this is on Amplitube’s stomp section:
On mobile, the problem changes. The screen is smaller, which makes the real estate more valuable. On these small screens, the switches have a tolerable size, but you have fewer pedals on the screen at the same time. This scenario invites us to check what’s the primary use case for each device. Experimenting and tweaking is a nightmare in smartphones because the user needs to navigate many screens to modify the settings. If tweaking is the primary use case of the app in this specific device, shouldn't they adapt the interface for it? Alternatively, did they found that the users find less value on a visually depurated experience?
At the end of the day, we are abandoning all the principles of UX design. Instead of constant improvement to find the best experience and an always-present necessity to adapt to changes, we go blindly with what there’s already there, even when this approach has been proved wrong in every other context or industry.
Go to eleven
For now, let’s stop giving Amplitube a hard time and talk about my biggest UX pet-peeve, found everywhere in the realm of music production: Knobs
Knobs are an excellent solution to allow the manipulation of levels on some elements on a device. If you have a bunch of things to control in a semi-precise way, knobs are a great solution. Sometimes you can even put two in one! That’s why you see them everywhere on sound gear. When you change the size of the knob, you don’t sacrifice much precision, as it might happen with other devices like sliders or “volume controls” (which are widely used on graphic equalizers because it allows you to see the “shape” of the band's manipulation).
You manipulate knobs with your fingers, which means it should translate well to touch devices. Sadly, replicating the same gesture on touch screens would be a pain. Imagine turning a small knob on your phone with two fingers, like in the real world. The wise decision was to manipulate them with just one finger, and a new pattern was born: Use the knob like a slider, dragging up to increase the level. This behaviour might be now a standard but is just off from my perspective. Most knobs work (if you look at it as a clock) going from around 8 o’clock (for minimum) to 4 o’clock (for maximum), this means going from left to right drawing a big curve, not down upwards (maybe the first half of a knob is manipulated this way).
You might say that this dragging upwards interaction is akin to the real world because a person can put their finger on the side of the knob in a tangential way and move it up, this way the knob does its circular movement. However, this is also true for any clockwise movement. A virtual knob could have a “drag from left to right” behaviour, following the same logic.
It can be worse. What if you transfer the same action to a mouse? The relationship with the original real-world pattern is almost completely lost, now it behaves like a volume control than melted and bent. Look at the example below from Ableton Live (Image 11), even the skeuomorphic visual have disappeared, but inherited the behaviour from other places.
Is not common to see many designs using knobs outside of the musical context. You can see the use of big knob controllers or dials to illustrate levels that (hopefully) required a high level of precision or a long scale that needed to be wrapped into a circle. This became a trend a few years ago (I'm blaming this on the popularization of the Nest thermostat), but it seems it is now used less and less in favour of interactions that fit better to touch devices.
Beyond pedals and effects
I left out many examples of VST instruments because of one reason. The simulation of pedals, amps, compressors, etc., aims to have not only an identical result but also an identical process. Is like if you transfer your gear into the virtual world, and they are the same, but they no longer take space in your studio. Other virtual instruments aim to have identical results, but the rest of the process changes. A virtual guitar, for example, needs input from a MIDI keyboard, or “drawing” the notes on an interface. Besides, the plug-in also might depend on a library of sounds produced previously with a real guitar. There's no way to compare the experiences because playing the guitar and organizing notes on software that sounds like somebody else's guitar when you hit "play" are entirely different things.
With drums, there's a different situation. They rely entirely on a library of sounds, but modern electronic kits can fell to a certain degree as a real kit. Toontrack’s Superior Drummer offers a virtual kit to configure, while the rest of the app goes with a more straight-to-the-point. Having the simulated representation on the screen can be nice when building your kit, but as soon as you want to change something, you face the reality that all of it is just a facade, and you have to choose the drum or cymbal from a plain list. I’m afraid that with just a few buttons (and way fewer pixels) the experience is not much more effective than the one on the kit’s sound module.
Going back to gear simulation, Guitar Rig 5 wants to evolve. Instead of pedals, they had every effect or component included in something similar to a studio rack and got rid of useless things like footswitches, simplifying everything while they still try to create a relationship with the gear they are emulating (with colours and textures). The guys at Native Instruments are going in the right direction, even if they still have knobs everywhere.
DAWs on the other side, being the intricate pieces of software that they are, can’t go dealing with this kind of skeuomorphic approach because they have so much to deal with it. These big players leave all the bombastic designs for the plugins, but as soon as you have yours set and ready, you’ll need to manage a sophisticated interface to deal with the complex pipeline of music production. Even so, as we saw earlier, they still have their version of knobs. They love that stuff.
What about synths?
When synths were born, they had to create their language. Synths try to emulate the sounds of existing instruments, but its process is not based on the results, they are based on the means to achieve it. If you want to achieve something similar to a Violin, with those prolonged notes with soft edges, you don’t go to a Violin simulator. You modify the envelope to have musical notes with a slow attack and a lot of sustain (and then you modify another 200 components). Sadly, this process included tons and tons of knobs.
Synths also inherited the same simulation issues when they made the transfer to the software world. However, modular synthesis is different and is not as simple as moving knobs or connect pedals. The whole purpose of modular and semi-modular synths is the ability to transform the instrument modifying the connections. This approach is why even a small modular set looks like a mess.
Yes, they took all of this to the software world. Synthesis has a steep learning curve, so it seems the purpose of this is, as we discussed before, to make a smooth transition of knowledge between physical and software. Being as complex as it is, transferring modular synths to the software without the idea of “connect point A to point B” would require a new paradigm.
This software in the image above is called Softube, a softsynth (software synthesizer) that simulates the same look of real-world Eurorack modules. The advantages are that you have unlimited cables and they can fade out to easily manipulate the modules (and of course, you can mount a whole new synth with drag and drop, instead of screwing each module to a wooden rack).
Fortunately, there’s another approach. Again, the people at Native Instruments tried to separate the brains and the beauty. In Reaktor, the user has the chance to manipulate the modules (usually with knobs, sadly) and then manage the connections below the hood. This way the complexity of the connections and the modules won’t be in the middle of the main UI, and it allows new complex behaviours on each module.
Nevertheless, this interface is very unfriendly and very little intuitive (even if you know modular synthesis), but nothing about modular synthesis is intuitive, I’m afraid. One cannot just start playing with cables as you can do on a guitar.
Between two worlds
This entire post is about how little the world of music production has learned from the software world in terms of experience. However, there are some examples of real gear relaying of software or learning from digital interfaces to improve the design of physical products. After all, MIDI controllers stopped pretending to look like other instruments and now are mostly pads that look more like their software counterparts and try to adapt to different musical contexts. Even for a drummer, the electronic pads long ago renounced the idea to be a fake drum, and now they are made to be efficient.
In the world of pedals, there’s one that cannot be simulated in software because it’s already there. The Eventide H9 is a very popular pedal that has dozens of “algorithms” on the inside, that’s it if you can afford to buy all the add-ons (talking about adapting software market techniques). Without a laptop, tablet or smartphone, the user won’t be able to modify all these algorithms, only to navigate through presets and alter key features (limited by the physical knobs and buttons on the pedal). That’s because the user can only access the full pedal via software. Because most of these algorithms are based on other Eventide pedals, the software interface is more of the same. Ironically, while the real pedal looks minimalist and modern compared to others, the software UI is a very traditional simulation. To add insult to injury, the design of the original H9 app pre-dates the iOS7 look and feel, and it still looks like this, yes even on Android and PC.
With synths, there’s a lot to unpack, but I want to mention the work of Teenage Engineering from Sweden. Their OP-1 is a small, beautiful (and expensive) piece of gear that has a lot to offer (like with everything synth) and relies a lot on its screen interface to do that. Its keyboard keys don’t even look like keys anymore, because they are made for much more than that. You can say that the OP-1 is the Macbook of synths, but that would be a disservice for the fantastic work they have made.
A lot of the things from the OP-1 can be seen in other synths and is just a sign of an industry that has different players with multiples approaches, all trying in their way to make the best of physical limitations and also blessed with a user base that is very open to experimentation.
Coda
We live in a golden age of pedals and effects. Small boutique manufacturers who build weird pedals by hand are able to make themselves a name, while people can still find low-cost alternatives or even get the parts to build them themselves. This context is the reason why the software aspect of the business is only a follower and won’t be able to do anything else besides simulations and emulations for a while.
We all know the joke from Spinal Tap of “this go to eleven.” That scene was just an example of how the same thing can be repacked and sold has an improvement. With gear simulation is the opposite case, a new thing trying to be rebranded has the same, because we think that their sole value is on being the same as the original, even if this completely holds up the potential of a product and go beyond simulation.
I only hope that we can find an alternative to knobs.