Tape vs Digital Part 2

This is part two of a 2-part series contrasting recording to tape versus digital recording. In this episode, I look at the strengths and weaknesses in digital recording. And then I summarize both episodes with a comparison of the two approaches, in a number of categories like sound, ease of use, maintenance, and cost. For each comparison, I share my conclusion about which is better.

email: dwfearn@dwfearn.com
www.youtube.com/c/DWFearn
https://dwfearn.com/

Transcript

119                        Tape vs Digital – Part 2                                                   July 3, 2026

I’m Doug Fearn and this is My Take on Music Recording

 This is part 2 of a two-part series of episodes contrasting recording to tape verses recording digitally.

Digital audio recording theory goes back to the 1920s at Bell Labs. They had it figured out back then, although there was no practical way to use it. The technology to make It work did not yet exist.

Bell Labs invented the transistor in 1947, which made digital electronics practical. And one early use of that was for long-distance telephone calls, which used digital audio starting in the early 1950s.

That worked fine for the limited requirements of voices over the telephone, but it was far from the quality that could be recorded on magnetic tape. It wasn’t until computer technology became powerful enough that practical high-resolution digital audio recording became feasible.

The earliest digital audio devices were delay units. Later, equipment like the Eventide Harmonizer were developed, and digital reverb. The early digital devices were 16-bit and did not sound all that great. But they offered features that were appealing, some unique and some that emulated existing hardware at a lower cost.

In the 1980s, tape machines were developed that could record audio as digital data. They still used a tape transport, magnetic tape, and multiple tracks. Those machines were a breakthrough, but to me they didn’t sound as good as analog tape.

The digital tape machines did eliminate some of the problems in analog tape machines, like the distortion and noise. Some early digital recording systems still used preemphasis, but with a less severe curve. Digital machines had zero wow and flutter in the usual sense, but if there was a tape defect, or speed variation, they would disturb the audio for an instant. A tape problem could result in an audio dropout.

The machines were expensive and did not get much use except in studios that had massive budgets. It was a start, but digital audio had a long way to go. The era of digital recording to reel-to-reel tape was short-lived.

Various magnetic storage media was tried for storing digital audio, often using semi-pro  video recorders to store audio. Later, some, like the ADAT and the Tascam DA88 used tape cassettes that were made for consumer video recorders.

The early digital equipment was 16-bit or less. And the sample rates were in the 40kHz range. I was not impressed with their sound.

The Digital Audio Tape, DAT, recorders were a step up in some regards, but they were only 2-channel stereo. Still, they could record continuously for over an hour on one tape, which I liked because when I did classical recording on location, it was challenging to change reels during short breaks in the music. I started using a DAT machine as a backup, capturing everything in the concert. If a break was not long enough to change the reel, the DAT backup might save the day. I never had to use the backup, however.

I was interested in hearing the difference between my Studer 2-track machine and the DAT machine. It was easy to A/B the two, and, superficially, they both sounded good. But there was something about the analog tape that sounded better to me, and it took a long time for me to understand why.

The Studer, with Dolby SR noise reduction, was very quiet. It was the same or better than the 16-bit DAT. All the basic performance measurements like frequency response and distortion, were comparable. The DAT had no wow and flutter, and the Studer had almost none.

Still, there was a difference. Years later I concluded it was a lack of low-level resolution with the DAT recording. At 16-bit, there are around 65,000 distinct levels that can be recorded. That seems like plenty, but it’s not enough to capture things like reverb tails or low-level sounds of instruments and voices. What I missed most in the DAT recordings was that sense of realism in those low-level sounds.

If you want to hear the loss of low-level details, here is an easy experiment. Take a decent recording, in any format and record it at 16-bit, 48kHz sample rate. Then make another copy the same way, but drop the record level by 10dB. Do that a few more times, reducing the level by 10dB for each new version.

Now listen to it. You will have to increase the monitor level by 10dB each time a new version is played, so be very careful. You can damage your hearing and your speakers if you have the level cranked way up and then play a normal-level clip.

At -10 you probably won’t hear much difference. Even -20 might not sound too bad. But listen as the details start to disappear. By -40 it should be quite obvious. At -50 it stops sounding like music. At -60, it is mostly just noise modulated by the beat.

We can hear details well below that -60, so just imagine what it is doing to the reverb tails, snare rattles, or articulation sounds in a voice.

That’s why I stayed with tape until 24-bit recording became practical. My first system was a Radar 24, which could record 24 tracks of 24-bit audio at a 96kHz sample rate. For comparison, the DAT and other early digital formats were all 16-bit and 48kHz sample rate.

The Radar system was designed to emulate a tape machine, which made people like me more comfortable with the transition to digital. Radar recorded to a hard disc drive. But it didn’t use a separate computer. The computer was built into the Radar rack unit. And it didn’t run Windows or Mac OS. The operating system was a variation of Unix, and it was stripped down to only do one thing, and that was to record audio. It never crashed, not once during the many years I had it.

Radar was capable of doing variable speed, over a small range. That was a feature I used a lot in the tape days.

Radar was just a recorder and playback system. You could do some minor editing things within it, but basically it was just like a tape machine. It had a finite number of tracks available, 24. To mix, you had to use either an analog console or a DAW. The first DAW I had was a very early version of Pro Tools. I was impressed with what Pro Tools could do and adapted to it quickly as a way to mix projects recorded on the Radar. I could import the Radar wav files easily through networking the two systems.

Over the years, I updated Pro Tools and the Mac computer to run it. I upgraded to converters that sounded best to me.

But there was still something about digital audio that bothered me. I couldn’t quite explain it. It just didn’t feel comfortable to listen to. I wasn’t going back to tape, so I just made digital sound as good as I could.

And then I was introduced to DSD digital recording, which uses an entirely different approach to recording audio. Listen to the episodes, “DSD Recording with George and Geoff Hazelrigg,” from April 2021, and “DSD Digital Recording” from May 2025, for more details on how it works and why it sounds better to many people.

Those episodes also speculate on why conventional PCM recording does not sound as good to me.

DSD sounds great, but it is a pain to work with. Briefly, in their native state, you cannot mix DSD tracks, nor can you do any processing to them. It’s just like a tape machine, or the Radar 24 system in that regard. To mix, you either have to convert the tracks to PCM, that is, wav, files and mix in a DAW. Or you can mix it in analog with a console. I have done the mixing both ways, even using a passive mixer we built for smaller projects, with up to 8 tracks.

PCM recording has improved over the years, thanks to faster computers and improved converters. But there is something about DSD that sounds better, even when the tracks are converted to PCM for mixing.

Not everyone agrees that DSD sounds any different from PCM. You have to try it yourself to see if it is worth the hassle to you. There are only a couple of converters that will work in DSD, and only one DAW that I am aware of. The DAW that does DSD only records and plays back, of course.

I made the transition to digital in the early 1990s and I have rarely recorded to tape since then. For me, digital recording sounds very good, particularly in DSD. And the amount of easy manipulation you can do in digital would seem miraculous to an engineer from the 1980s and earlier.

I still have my Studer machine, although it needs some work. Like all tape machines, it is over 40 years old and it’s been decades since it was discontinued. Parts and repairs are difficult to obtain, as is the expertise and tools needed for repairs. I would like to have it working again because I think there is room in the world of recording for both digital and tape recording. But it is not a high priority for me.

You have many resolution options when recording digitally. You can choose a sample rate from 44.1kHz to 192kHz, or even higher with some systems. I don’t want to go into detail of the impact of sample rate on how a recording sounds, but as a general rule, I think you have to set it at 96kHz at a minimum for professional results.

The other variable is bit-depth. That ranges from 8 bits to 32 bits. Since those numbers are exponents of base 2, the number of distinct levels that can be resolved goes up a lot faster than the numbers would suggest at first glance. I would say that 24-bit recording sounds very good. There is an improvement by going to 32-bit recording, but I would probably struggle to tell the difference between 24- and 32-bit in a short A/B listening test.

My impression is that most engineers record at 24-bit, 96kHz sample rate. That is a totally acceptable sound. This podcast is recorded that way, although it gets reduced to a mp3 by the podcast providers.

For music recording, I use DSD256. If I mix in the box, I convert the DSD files to 32-bit, 192kHz. That sound pleases me, although it isn’t quite as pristine as the DSD original.

Every conversion results in more noise and more distortion. In fact, every aspect of the sound is degraded to some degree. For example, a hardware insert will require a conversion from digital to analog, and then back to digital. The loss is acceptable in all but the most extreme examples. But it is nothing like the loss in tape, where even one transfer from one tape machine to another results in a loss that is apparent to even the untrained ear.

Computers are now pretty fast, and storage is so cheap that even the most inexpensive home computer has a 1TB drive. The rate of improvement today has slowed from its rapid increase during the previous 20 years.

As fast as computers and storage are today, nothing is instantaneous. I am always amazed at how long it takes for a computer to process any data. I know there is a lot of math going on, and reading and writing to memory is not instantaneous, but still, the process is not very fast.

The result is latency, something that was never a problem with tape and other analog systems. In analog, there isn’t even the tiniest phase difference when monitoring the music being recorded. In even the fastest digital systems, there is always at least a few milliseconds of delay between when the music is performed and when you hear it. That can increase dramatically as sessions get more complex, and it can be a major problem for musicians performing while wearing headphones.

Even when recording this podcast there is some latency in my headphones. It is only around 20mS, but it is still noticeable. If I were to add a plugin, that could jump to 100mS or more with some plugins. That is a tenth of a second and it is nearly impossible to speak or play with that much delay in your headphones.

I used to think that the latency problem would eventually disappear as computer technology advanced. But not much has changed in recent years. I suspect we will have to live with this problem for quite a while. Maybe we will never have computers and software that entirely eliminates latency.

Let’s summarize with a short comparison between analog recording to tape and current digital recording. This is not a complete list, but I think it covers the main areas that are important for use in the studio. Here are my thoughts on that, in no particular order.

Maintenance: Tape machines require daily adjustment and cleaning, and major expense for wear items. Digital has no daily maintenance other than keeping the files organized. Longer term, hard drives fill up and either need to be archived somewhere, or a new drive added.

Advantage: Digital

Performance and Sound: Both tape and digital have very good specs. Digital measures better in almost every regard, including noise, distortion, frequency response, phase error.

Tape has a unique sound, which may be appropriate for some music.

Advantage: Digital, but tape has unique properties that cannot be duplicated in digital.

Ease of Use: Tape machines are cumbersome beasts. They take time to rewind, they have limited recording time, and they require alignment tones on every reel. Tape has never been very reliable, deteriorates with age, and the sound can change from reel to reel. All tape manufacturers have periodic bad batches, which might not become evident until well into the project.

Digital is simple and easy to use. Software may be buggy, and computers are never totally dependable. But usually those are minor problems. Storage formats evolve over time, so for archiving, digital has to periodically converted to the current format. Tape formats are locked in time, and the machines to play the tape will eventually become just museum pieces.

Advantage: Digital

Availability of Repair and Maintenance parts: Digital has virtually nothing that wears out, except maybe hard drives or computer interface devices. Those are commodity items and very inexpensive. There is an issue of computer operating system maintenance, since most DAWs will only operate with very specific Windows or Mac OS operating systems.

Tape machines have been out of production since the 1990s. Repairs can be expensive or impossible. Parts, if you can get them, are very pricey. Lots of things on a tape machine must be routinely replaced. On the other hand, they do not depend on a computer.

Advantage: Digital

Cost of Media: Tape is very expensive -- hundreds of dollars for a reel of 2-inch tape, and that’s only good for 15 or 30 minutes, depending on the tape speed. There are only a couple of tape manufacturers. Tape is a low-demand item, so it is more expensive than it used to be when it was mass-produced.

The only media for digital recording are solid-state drives or hard drives. They are very inexpensive since they are used in billions of computers. Since all digital data is ephemeral, important audio files should be backed up in multiple places, ideally including one off-site or cloud based. A backup copy of tape is feasible if you have two machines, but the copy will always be of lower quality than the original. A backup will double the media cost.

Advantage: Digital

Length of Recording Time: Tape is limited to 30 minutes at most before you have to start a new reel.

Digital can continue recording as long as you have disc space.

Advantage: Digital

Basic Electronic Performance: Digital is better in all measurements. But that does not tell the entire story, since there are always artifacts in both tape and digital that can degrade the sound of the recording.

Advantage: A tie. You pick which artifacts that bother you the most.

Generational Loss: The sound of tape deteriorates quite noticeably when it is copied to another tape machine. That’s going to happen during mixing, and also if tracks have to be bounced on a multitrack machine. Noise, distortion, wow and flutter, frequency response anomalies all are doubled in a first-generation copy, and doubled again with every subsequent generation.

Digital audio is bit-for-bit perfect for a digital-to-digital copy. However, if the digital audio is converted to analog and back to digital, there will be a minor loss each time. Normally, this is not much of a problem, but it could be if there are a lot of hardware inserts in the mix.

Advantage: Digital

Punching: Punching-in a track on a tape machine is a skill that takes time to develop. A mistake can destroy a recording. There is no “un-do” button on a tape machine.

Digital punching is generally seamless. If you do it improperly, you can always go back to what you had before.

Advantage: Digital

Editing: Editing with tape is a cumbersome and prone to fatal errors. The tape must be cut with precision at exactly the right spot. The right spot has to be determined by rocking the tape back and forth by hand, listening to a vastly slower and lower pitch than the normal playback. The two ends before and after the edit are joined with a special splicing tape, which must be applied with great care so that it does not extend beyond the edges of the tape by even the slightest amount. It is a tedious and stressful process when cutting a tape from an expensive or one-of-a-kind session.

It can be difficult to repair a splice if it is not exactly where it needs to be. Even off by a tiny sliver of tape can ruin the feel of many types of music. Trying to put the edit back together and making a new splice just fractions of an inch away is likely to result in a dropout at the splice.

Splicing between different takes means keeping track of which direction the tape is supposed to go, which requires full concentration. A section of tape falling on the floor may never sound the same again if it gets dirty or encounters a magnetic field.

It takes a while to find the spot in the song, rock the tape to find the exact place to cut, mark that spot with a grease pencil, transfer the tape to a splicing block, make the cut, apply the splicing tape, re-thread the tape, back it up, and listen.

All those challenges are multiplied on a multitrack tape, where a cut through the tape makes the transition from previous to next section differ slightly across the tracks.

With tape, you can only edit the entire tape, with all the tracks. It is possible to do a “window splice,” where a section of tape is cut out and replaced just for one track. But that is a last-ditch process to save a performance. It could easily take all day to do a window splice.

Editing a digital file is easy, with a waveform display to guide you to the exact spot. If you make a mistake, it’s easy to un-do it and try again. This encourages experimentation. Edits that would be unthinkable on tape are possible in digital.

Advantage: Clearly digital

Cost: Tape is expensive. A used tape machines in good condition is expensive. Maintenance costs are high, and require a lot of specialized knowledge and experience.

Digital systems are cheap. You can entirely equip a simple but high-quality digital studio for the cost of a single multi-track tape machine. There is not much on-going cost with digital, except maybe upgrading the computer and converters periodically.

Advantage: Overwhelmingly digital

Effects: Digital offers access to thousands of manipulations in the form of plugins. The speed of the tape, on the other hand, allows you to create unique, artifact-free effects that digital cannot do. Listen to the episode, “Variable Speed Recording,” from December 2025 to learn more.

Advantage: Digital is better at most, but tape has some unique capabilities.

Latency:  Tape has no latency at all, under any circumstances. The musicians recording an overdub have no problem since they hear what they are playing exactly when they hit a note.

Digital always has latency. There can be ways around this, by monitoring the input to the converter instead of the output, but that adds a lot of complexity to a session. And not all converters have a “direct out.”

Advantage: Tape

Session Flow: Digital has the advantage of nearly instantaneous return to a point in the song. That makes it easy to keep trying an overdub, or a punch-in, over and over until the performer gets it right. That process can be automated. That may make it possible to assemble a performance that could not be done all at once.

Starting a new take is also very rapid. And there is no break to change reels.

Tape is a slow and cumbersome medium. It takes time to rewind, or find a punch-in spot, or go back to a previous take. But that workflow also helps make a slower-paced session that might leave more time for contemplation and creative ideas.

Advantage: Depends on your goals, but mostly digital

Medium dependability: Digital files could last forever unless something goes wrong. But formats evolve. DAWs come and go. Computers are always advancing. A digital recording from 30 years ago may not be playable, unless it has been periodically converted to the current format.

Files can be accidently erased, misnamed, or lost in the file system. Multiple backups are required for serious recording projects.

Tape has never been very dependable. Tape manufacturers always have bad batches. Tape formulations are discontinued. Current tape type selection is limited. Proper tape storage is vital, and that means over decades. Old tapes may not be playable.

Tape is subject to damage if not properly wound on the reel. Magnets near the tape can erase it or cause a major increase in noise, loss of high frequencies, and decrease in level. Noise reduction systems, like Dolby and dBX, have long been out of production and a working unit is required to decode old tapes.

The tape is worn out a bit with every pass. A 24-track project with lots of overdubs and punching may start to lose high frequencies. Worst case, the oxide may deteriorate and flake off. The recording can disappear.

Advantage: Digital

Mixing: Tape requires analog mixing unless converted to a digital format. Digital mixing offers tremendous creative opportunities. However, tape can do some things in a mix that require innovation by the engineer that can be very satisfying compared to just using a digital plugin.

Advantage:  Digital

===

Clearly, in this comparison, digital recording wins in almost every category. Tape does have some unique tricks, and after all, tape was used for some of the most creative albums ever made. Tape can be a pain to work with, but that pain can inspire innovative solutions.

I have made my decision to work entirely in digital. I may use tape on occasion for its unique sound and capabilities. I wish that DAWs all had artifact-free variable speed, which is trivial to achieve with tape but challenging in the digital domain. I hope the latency problem improves with technological advances.

There is much more to this comparison, but I think I have covered the main points that show the contrast between digital and tape.

There are many people who have spent their entire recording career in digital, and a tape machine may have an allure that is difficult to resist. If you can find a tape machine in decent condition, keep it running properly, learn how to align that machine, and if you can find high-quality and consistent tape to use, then I encourage you to try that experience. You will learn a lot about recording, and a new appreciation of those albums made before the digital age.

But be prepared to deal with a lot of frustration. You will need to spend a lot of money. It’s not a medium for the faint of heart.

Thanks for listening, subscribing, and commenting. You can reach me with your thoughts and suggestions at dwfearn@dwfearn.com

This is My Take on Music Recording. I’m Doug Fearn. See you next time.