Five Myths About Tape

Since the earliest days of digital recording, there’s been a rather vocal subset of engineers who believe in the superiority of tape over digital recording. There are certainly differences between the two formats, but a surprising amount of the information disseminated about tape is, err, less than accurate. Let’s look at a few of the most egregious examples.

"Tape is ‘warmer’ than digital"

The truth is that the playback from analog tape can have an EQ boost in the lower frequencies due to something called head bump. This is a function of the playback head on the tape machine and is related to the physical characteristics of the playback head, when the wavelength of a recorded frequency gets close to the physical length of the gap of the pole pieces on the head. If you happen to be using companding noise reduction on your project, the head bump could be quite nasty. Analog tape also has another “feature” called self-erasure. Essentially, after many plays, the high frequencies will begin to disappear. The amount of high-frequency loss could vary depending on the tape used, the machine used, and the number of times the tape was played.

Analog tape recorders don’t have a flat frequency response — even a well-maintained and well-aligned tape machine has peaks and valleys in its response. But there are other issues: the changing sound of mixes (many engineers are creating brighter mixes these days), the sound of the consoles used (the transition from a transformer-based console to the more common ITB mixes these days), and even the sound of the instruments and the signal paths that are being recorded (compare John Lennon’s Gibson through a U47 to a brand-new Taylor with a small-diaphragm condenser) can make a mix sound brighter. In the end, it’s most likely that the “tape is warmer” crowd are hearing a number of different issues and lumping them all together to suit their preferences.

"It’s easy to get tape to sound good"

It’s pretty hard to compare apples to cantaloupes, so we should try to level the playing field. If you’re working with a high-quality analog tape machine and decent tape, as long as you don’t overload either the electronics or the tape, it’s easy to get a good sound. Of course, if you have decent converters, as long as you don’t overload the converters or any internal busses, it’s also easy to get a digital recording to sound good. But as many of us know, if you’re working with gear that doesn’t have a lot of headroom, it’s easy to get a bad sound, too. Besides, comparing an analog machine such as the legendary Studer A800 (easy to get a good sound) to the converters on your laptop isn’t fair, any more than comparing a digital system with top-of-the-line Burl converters to an analog machine such as the TEAC 3340 is fair.

These days, unless you’re pressing only vinyl, both analog and digital recordings spend time in both domains. Even if you’re a die-hard DAW user who mixes in the box, your microphones, preamps, and any other processing gear used on the way to the converters are analog pieces. And if you’re recording to 2-inch tape and mixing to a half-inch mastering deck, you’re still going to end up in the digital domain when you press your CDs. So unless you’re only pressing vinyl (and your mastering engineer eschews any digital storage in the mastering process), all recordings are really hybrid analog/digital recordings.

"Tape distortion sounds ‘vintage’"

When analog tape is hit harder than it should be, distortion occurs, just like when your DAW is overloaded. The difference is that when recording to tape, the onset of the distortion is softer, and the result (tape saturation) is more euphonic to most folks than the harsh distortion that comes from overloading a digital medium. Here’s an interesting thing, though — while tape saturation is now apparently one of the features of analog recording, professional engineers of the 1960s and 1970s worked hard to avoid it, considering it to be bad engineering. The electronics in a tape machine — whether it’s the tubes in an old Ampex or the transformers in an MCI 2-inch machine — can, if overdriven, add an audible (and measurable) amount of distortion as well. You can assume that if you’re saturating the tape, you’re also overdriving the electronics that will get that signal to the tape. So it’s kind of interesting that the quality that some engineers look for in analog tape is the result of a lack of quality in the recording process.

To many, especially those of us raised on overdriven guitar amps, “distortion” is “warmth.” There’s nothing wrong with that, of course, but if tape saturation is your goal, then you’re using tape as an effect rather than as a recording medium.

"Tape has infinite resolution"

This myth comes from a fundamental misunderstanding of the Shannon-Nyquist theorem, which states that when the sampling frequency is greater than twice the maximum frequency of the signal being sampled, the original signal can be perfectly reconstructed. As long as the Nyquist limit (the frequency that’s half the sample rate) exceeds the highest frequency of the signal being sampled, the original analog signal can be reconstructed without loss. The folks on the infinite resolution side of the controversy think that because the original signal is sampled 44,100 times per second, it must have some sort of stair-step effect and that information between those samples is somehow lost. But it doesn’t work that way. As nonintuitive as Shannon-Nyquist is, it has proven to be accurate for decades.

"At least tape has good archival qualities"

Not necessarily. While some engineers have spoken about tapes from the 1960s that still play back perfectly, not all tape was created equal. Through the decades that tape has been made, manufacturers tried to improve their wares in a number of ways. The original tape that US servicemen brought back from Germany after 1945 was basically made up of long reels of paper with red paint on one side. The color in the paint contained iron oxide, which would retain magnetic information. Improvements were not long in coming — a plastic polymer replaced the paper, the iron-oxide formulation was tweaked to allow for greater bandwidth and greater levels before distortion and even better adhesion (if the iron oxide didn’t stick to the plastic, the tape would be rendered useless). Anyway, one popular tape formulation turned out to absorb moisture from the air around it while in storage. This particular tape could be temporarily restored by baking it at 140 degrees for some time. (It turned out that the small ovens used to dehydrate fruit worked perfectly for these tapes). But not all old tapes responded well to baking; some brands would simply shed the iron oxide, while others would stick together. A surprising number of multitrack master tapes from the 1980s are unplayable today — with any technology. So while most of us have lost data due to hard drive failures (have we mentioned how important it is to back up your hard drives early and often?), we can’t assume that the archival quality of analog tape backups is that much better than online cloud storage. So back up your projects. And every few years, back up that backup.

Through the golden age of analog recording, engineers would strive for the highest fidelity, which meant that they went to great lengths to minimize these artifacts. It’s been long enough now that many young engineers don’t remember why digital recording was embraced by so many of the older generation. While the shortcomings of early digital were recognized (especially the brittleness that was an artifact of the first-generation converters), the long-standing issues that digital recording eliminated (wow, flutter, head bump, self-erasure, and print-through among them) made that new technology a better alternative for the vast majority of engineers.

The takeaway

If you choose to work with analog tape, embrace the sounds that you get — but don’t make the mistake of believing that analog recording is intrinsically better than digital; it’s simply different.