Tape Emulation roundup

[mister-rz]

One I'm playing around with at the moment is roundtone 4 by sknote, can get a lot of different sounds and textures out of it, plus I like using the pitch shift for drums and pads, sounds dodgy so it fits in perfectly with my sound. My favourite one is tape by softube, like the front to back depth, gives a nice sheen to the sound, replaced vtm by slate for me, which I hardly use now.

Like using tapedesk for enhancing high frequencies on the n80 setting, has a nice sheen on 30ips, rarely use the other two options, mcdsp ac202 is very, very nice, keep meaning to use it more, low cpu and warm sounding, good for giving harsher sounds a soft coat, especially on the prickly end of the spectrum. Outside of that I'll use airwindows and the hornet tape, very rarely use my waves tapes anymore, liked the j37 for clarity and krammer for a more earthy sound. One on my wishlist is the tcs-68 by fuse audio labs, liked a lot of the black rooster stuff apart from the amp and tape, vcl-373 comp is amazing, so will definitely pick up tcs-68, interested in cassette emu's more now.

[v1o]

Does tape actually sound like anything? I seriously doubt it does, this whole tape emulation industry is just snake oil.

It has certain coloration, saturation and harmonics that most of the people like. Also there are many different tape machines out there, so there are many different tape emulations too.

So you achieve the same by using but crushers, EQs and compression.

[sascha]

Anyone with a basic understanding of physics and electronics can make up the fundamentals needed for magnetic tape recording. It's sometimes sad to read the same pre- and misconceptions of 'tape sound' in public forums over and over again, but it's also sad how certain software products have an impact on this because some don't behave at all like how a tape machine would act during recording and playback. I'm not mentioning names, but here's a small check list to find out if a plugin gets in the backpark of magnetic-tape at all:

- take Budde's VST plugin analyser and plot the frequency response at, say, -20dBFS, using both of its test signals (sine sweep and impulse)
- do the same at 0dB or higher (trying to crank the plugin somewhat), perhaps make more test, like -10, -3 etc.
- check the frequency response when the input-gain increases: the upper-mid & treble region should become lower in amplitude. This slew-rate limiting is one of the major aspects of 'real tape', it's a dynamic thing. It's a function of hysteresis, flux, gap width, induction and self-erasure. (we've shown that in the Satin manual, pp. 40-43)
And there is no inertia, so there's no point trying to mimic that through frequency-dependent dynamics compression. But it's dependent on the bias signal, which is another important factor:
- check the plugin's bias control: lower setting correspond to lower bias amplitude. In the real world, this is a HF sinewave (70..200kHz typically), and it pushes the signal into the most linear region of the hysteresis curve during recording. With no bias you'd end up with severe crossover distortion. Less bias = more treble but more crossover distortion. More bias = less treble but less distortion. Why more bias = less treble? Because of saturation and self-erasure, and partly the tape thickness plays a role here. Mind you, there are software products (again, no names...) that have it the other way round: if you increase bias and the sound becomes brighter, it's a major flaw in the modeling and just plain wrong. Do the test with VST Analyzer by measuring THD at various settings; the plot should reveal crossover distortion change properly, the upper harmonics should decrease when increasing bias (given reasonably low input, like -20dBFS).
- check if a head-bump effect exists at all. The magnetic head is not only core & wires, it's also its metal housing, and there's sympathetic resonance and mirror effects happening at lower frequencies, since the head geometry messes with this induced signal, e.g. this magnetic field interferes with the origin. On a real machine, this looks like a resonant filter tuned to very low frequencies, with some ripple bleeding in towards the lower mids. The 'good' manufacturers managed to dampen this effect, either mechanically or through filters on the replay board, but many people like that extra 'oomph', but sometimes it's very subtle. If a simulation shows no head-bump effect, then the entire tape transport and repro modeling is overly simplified and incomplete.

Sure, a 'good' machine and tape should sound transparent and not generate an own sound in the first place. But it's the tiny subtleties that make up the sound, not the distortion that comes up when overused. Try using a simulation like they did in former times: slap instances across multiple tracks and busses, it's the accumulated effect that counts, and often best used when it's not noticeable right away but chances are there's something missing when you bypass all instances.

[AdamWysokinski]

airwindows and Nebula are the best.

I agree re airwindows (FromTape, IronOxide5 and ToTape5) - best I've found so far. And free!

[4tune]

..here's a small check list to find out if a plugin gets in the backpark of magnetic-tape at all: ..

Thanks for that sascha, really gives an insight into the work involved for an in-depth model of some hardware!

[AdamWysokinski]

So you achieve the same by using but crushers, EQs and compression.

I agree, but using a dedicated plugin is simply quicker.

[sascha]

So you achieve the same by using but crushers, EQs and compression.

I agree, but using a dedicated plugin is simply quicker.

It might not only be quicker, it might also be more faithful. You can get tape-style HF smear by using a combo like high shelf (boost)->saturation->high shelf (inverse cut), but this is just a corase approximation (and misses bias linearisation, modulation/asperity and aspects like head bump completely). Tape has just so much interaction going on, one needs to actually mimic the process itself (like transport, bias generator, heads and the circuit boards) rather then the sonic outcome in order to capture the most of it. And it's interaction that makes things live and breathe. It's the oversimplification of complex phenomena that makes digital tools sound inferiour to analog, not the domain per se.

[bbtr]

It's the oversimplification of complex phenomena that makes digital tools sound inferiour to analog, not the domain per se.

You mean, objectively inferiour? In all possible cases?

And they're not oversimplifying, they are just NOT introducing unwanted artifacts/processes/whatever which in analog come from the materials, construction, etc.

[Giova942]

Is the DC offset introduced by ToTape5 not a big deal??

[bbtr]

He did it by mistake, so no, it isn't. Not more than a head bump. Which is not that great either.

[heavymetalmixer]

He did it by mistake, so no, it isn't. Not more than a head bump. Which is not that great either.

Is it true that using TDR NOVA or SlickEQ after ToTape5 eliminates that DC offset? Or do I need to set a HPF?

[robbmonn]

It's the oversimplification of complex phenomena that makes digital tools sound inferiour to analog, not the domain per se.

You mean, objectively inferiour? In all possible cases?

And they're not oversimplifying, they are just NOT introducing unwanted artifacts/processes/whatever which in analog come from the materials, construction, etc.

I agree with sascha. For example:

The softube Tape plugin. I love this plugin and use it all the time. But does it really emulate the signal chain to roundtrip a recording to tape? I don't think so. That signal chain would be:

sound in -> preamp in tape machine -> record head -> tape (wow/flutter/motor/tape physics) THEN, PLAYBACK: tape (wow/flutter/motor/tape physics) -> playback head -> output amp -> sound out.

Just take for instance the wow and flutter or "stability" emulation in the Softube product. After extensive listening it is a slightly chaotic LFO assigned to pitch. Same is true for Black Rooster, Waves and many other tape emu plugins. But, if you record and play back on the same tape machine you're going to have two modulations: one created when recording and one for playback.

If I listen to Twine by Taylor Deupree & Marcus Fischer I can hear this. There is even more than two different sources of pitch modulation, but definitely two main ones. And, to my ear they are different from each other. It seems likely that the artists used a different deck to record than to play back the loops.

To get the feel of 30ips studio tape you need to do a little bit of work and then maybe you won't get it. When I recorded in a studio I always felt the studio muti and 2-track recorders on the fancy tape decks sounded more perfect than anything else, CD quality at least. So, aside from hard to hear artifacts there isn't much to do. Get sweet preamps and a great mastering chain on your mix bus instead maybe.

BUT, for driven cassettes or old ass reel to reels playing loops slowed down or ???? you need to take a look at what you're trying to emulate and dig in. There is generally a lot going on. Roland Space Echos are easy to overdrive and (at least the one I used to have, sigh) have a lot of warble, and are set up for instrument, not line level. If you want that kind of a sound, set up your preamp emulation, your tape emu (remember to emulate the write and the read), the eq that would be in line to minimize noise, the output amps, etc. And be very careful about gain staging... when I ran complex guitar and vocal rigs the art really was in manning all those input and output level knobs across a signal chain. It took a long time to find the sweet spots... same is true with plugins. Push a preamp too hard or not hard enough and you're not going to sound the way you want. I don't think that there are shortcuts. This the assertion that the complexity is where the good stuff is and a lot of these plugins are just a shallow or a very specific implementation of the "real world."

For me, once i get into that space things get very interesting. Mixing plugins and being very careful about gain staging (sometimes multiple compressors or limiters in a chain) ... this can get you to a place where there are a range of good sounds to be had, not just "this sounds like tape."

At least that's my thinking.

[bbtr]

Why would you use ToTape5 in the first place? It's got NOTHING to do with tape.

As are most of his other plug-ins - not what they pretend to be. Some distortion, some frequency curve - et voila, his 3535432547874854-th plugin is ready to be unleashed upon the gullible fools who think that a two slider affair is the real thing and sounds 'great'. Well yeah, because distortion is a drug. That's why so many devs are pushing it.

[bbtr]

@ robbmonn

Sascha E. is a great and very knowledgeable developer. I've admired his work since Blockfish and Endorphin days.

And I agree with everything he said about tape. I don't agree with what he said about digital instruments being 'inferiour'. Which he meant (but didn't say) always. Which is patently untrue.

[sascha]

And they're not oversimplifying, they are just NOT introducing unwanted artifacts/processes/whatever which in analog come from the materials, construction, etc.

Seriously... come on.
I'm open for a discussion on a thorough & substantially technical level, and you're free to check the Satin manual, where we've given an in-depth magnetic-recording primer, along with thorough measurements and specs to also show what does what. If you look more closely at the topic (and actually use *real* wide-format tape) and dive into the physics involved, it should become more clear that taking everything into account is mandatory with serious tape-sound reproduction, and once you omit a seemingly unwanted aspect, it all falls apart and becomes a caricature of a thing. Fine if you like that but please don't call it tape then.

I don't agree with what he said about digital instruments being 'inferiour'.

You've misunderstood. Digital is not inferiour by design. It's only inferiour to analog if you leave out important aspects.