Hacker News new | past | comments | ask | show | jobs | submit login
They don't make them like that any more: the Yamaha DX7 keyboard (kevinboone.me)
336 points by ingve 21 days ago | hide | past | favorite | 217 comments



>With hindsight, the DX7 and its ilk look like a weird diversion in the path of musical progress. Operator synthesis makes no logical sense. It became popular because we had the capability to do digital sound generation, but lacked the technology to do it in a comprehensible, systematic way. So the DX7 enjoyed huge popularity for a few years, before fading into obscurity.

That's a weird take. I mean "Operator synthesis makes no logical sense"? He writes like FM went away.

While DX7 itself was eventually retired (and synths like M1 became popular), Yamaha made several further flagship FM synths, and FM synthesis is still widely popular, used in both plugin form (including several DX7 emulations plus lots of fresh FM implementations), modern dedicated FM synths (e.g. Korg Opsix, Elektron Digitone), and of course as FM sound engines in multi-engine synths (FM-X in Yamaha's keyboards like Montage and MODX, Hydrasynth's FM mode, Nord's FM engine, to name but a few). There are even boutique options (ReFace DX, Volca FM).

And of course even vintage DX7s go for a quite high price in second+ hand market...


FM/PM synthesis makes no "logical sense"... until you spend the twenty minutes required to understand it.

Really people gave up on the DX7 because of the limited interface, somehow FM got the blame. A subtractive synth with a knob for each function is a fantastically intuitive thing, but when it's three buttons and a line of abbrv txt it's not very nice either.


FM is inherently more difficult to understand than subtractive, even with a very good interface. It's the double-edged sword of FM - you can create very complex behaviours from a small number of simple operators, which makes FM both uniquely powerful and uniquely difficult to think about intuitively.

I can model subtractive and additive patches quite accurately in my head. That's not a brag, it's just not very difficult if you understand the basic principles and you've put in a bit of practice. FM still takes me by surprise after decades of trying to master it.


Subtractive is intelligent design. FM is discovery :)


FM is more accessible now


I heartily recommend the Korg Opsix to anyone who wants to experiment with FM synthesis.

As the article mentions, the UI of the DX7 made programming it, modifying existing sounds or coming up with new ones entirely, difficult to say the least.

With the Opsix, much of the ability to tweak aspects of the operators, envelopes, LFOs and so on, are given dedicated knobs and sliders, or easily accessible menus whose parameters are tweakable with physical knobs. The hands-on UI of the Opsix turns what once was an ordeal of menu diving and incremental buttons into a much more rewarding, interactive experience.

I play one live in a band and use an old Yamaha KX76, the dedicated controller bigger brother to the DX7 as a controller for the Opsix. Best of both worlds. The feel on those old keyboards is superb.


The Opsix looks very straightforward. I bought the Digitone Keys about three years ago and it's a lot of fun, along with having that sequencer and Elektron Interface. It does not appear to have been a successful product, and I do not think it's really possible to use a full instrument, but I have enjoyed using it as a plaything to learn and understand FM synthesis. No doubt there is some creative wizard like Blue Lab Beats who probably uses this and a Digitakt to make whole albums.


For people interested in this product but not sure buying a hardware synth: opsix native is the name for the software only version of it available at the Korg website…


I have one and agree wholeheartedly. The opsix makes FM fun and relatively easy. Excellent for sound design.


They also didn't get why FM synthesis was so appealing, in spite of being difficult to understand for the common musician.

It's in the tone.

Acoustic instruments respond in a complex way to the variation in strength of input: when you strike the key in the piano faster, pluck a string harder, or blow air info the saxophone stronger, you don't merely get a louder sound: the harmonic content, the timbre of the sound changes as well.

Analogue synthesis struggled accomplishing this. The classic analog synth would have an envelope generator ("ADSR") controlling the loudness of the tone, and another, most commonly, controlling the filter (the thing that makes the synth do a wowowow sound on the same note), but responsive fading and evolution of the harmonics wasn't readily available.

On the Yamaha DX7, it was built into the core idea of FM synthesis.

You don't know it when you hear it, you know it when you play it: the way the keyboard responded to the touch was alive, magical.

You didn't need to rely on the modulation wheels and joysticks and knobs to vary the timbre as you play. You could simply play the keyboard.

On my Yamaha Reface DX (which overcomes the drawbacks of FM user interface), I can easily make a tone whose character (not loudness! - or not just loudness) changes when I simply play harder. It's like having several instruments at once at your disposal, blending between them on the fly.

It's that playability that makes FM make sense — and it was what other digital synthesis technologies went for, too. Roland's "linear arithmetic", vector synthesis, and M1's multisampling all explored that area — but they came after DX7.

What makes FM synthesis unique is the heavily non-linear response of the tone to the dynamics. At worst, it's unpredictable, but once you figure out where the sweet spots are in the parameter space, you get a tone like nothing else. A bell that's also a string orchestra. A guitar with a soul of the saxophone, but not mistaken for either; an identity all of its own.

Yamaha DX7 heavily leaned into this aspect in instrument's design, via providing additional parameters that controlled the sensitivity of operators to velocity depending on where on the keyboard you are, so that the lower tones would have a different character from higher ones.

The "diminished brilliance" the author writes about was likely that — i.e., the author not figuring out how FM sound design works, which they openly admitted. It was matter of taste of whoever made the presets; without programming those curves in, the higher notes can easily sound screeching.

The point, again, was that the instrument wasn't merely responsive in a way that analogue synths couldn't dream of, but that the way in which it was responsive, tone-wise, was programmable, and varied not just from patch to patch, but across the scale and velocity range.

Again, think about how plucking different strings on a guitar harder produces a different variation in tonal response. Each string has its own character.

This is the soul of the mathematical idea of FM synthesis: that the tone evolution should not merely be controlled by time passing (as it is on most analogue synths, via envelope generators and LFO's), and not by knob twiddling (modulation wheels, knobs, sliders, joysticks,...) — but by playing the instrument itself.

And on a keyboard, what you really play with is where on the keyboard you strike a key, and how fast.

Yamaha DX7 allowed the player to vary the timbre by playing the instrument, with both hands, by having all tone generators depend on these two variables in a programmable, non-linear, interesting way.

FM synthesis of Yamaha DX7 therefore can't be separated from the physical keyboard it shipped with. The way the tones felt as you played them were determined by the response curves which simply don't map in the same way to a different keyboard.

The fact that the DX7 was a digital synth obscured the fact that it was a very analog instrument in that way; that to get a truly good FM preset, you need to tune it to the keyboard response (i.e. velocity curves), and that involves the analog components.

It's also for this reason that DX7 only has membrane buttons, and no knobs or sliders. It didn't need them. The 60 keys were your knobs and sliders, the means to control the tone.

That's why the ePiano on the DX-7 was on 60% of the new releases. It didn't merely emulate the Rhodes (which, by all means, wasn't a rare instrument).

What it did was it gave keyboard players a way to play with the tone of their instrument while playing the instrument, something the Rhodes would have a more limited range for, as the variation in tone response was constrained by how similar the actual metallic forks that made the sound were to each other, and how similar the hammers are across the octaves — and the digital DX7 didn't have that limitation.

It also gave the people used to playing the synth with one hand (to be able to tweak the sound with the other) the freedom to play truly polyphonically, and use the keyboard itself to control the tone dynamics.

Playing it was a liberating experience, and it still is, because while intricate multi-sampling can also give you that effect (at no less difficulty, mind you, even if you have the samples!), FM does it differently.

The musicians didn't need to be mindful of all that; the absolute majority (Brian Eno expected) were outright oblivious to why and what made DX7 the instrument that you had to have.

You just felt it.

And yes, new FM synthesizers keep coming. Because emulating acoustic instruments is not just easy with sampling these days, it also isn't enough. You can just hire someone to play the real instrument, after all.

You need a bit more than that to craft a distinctive sound — especially a new one.

Liven XFM, Korg Opsix, Arturia Minifreak all go boldly where manmade sound didn't go before, and these are just three novel FM synthesizers from this decade.

Reface DX came out less than 10 years ago; and its FM engine is different from DX7 (as is the UX — you can finally change the tone while playing it with live controls).

And for all the talk of how FM is old, I've yet to see someone not be captivated by the ePiano patch that comes stock on the Reface DX when I let them play it when I bring the instrument around with me on trips (which I often do).

Current developments in the controllers (like what ROLI is doing) will allow all the existing sound generation techniques to shine in new ways, including FM.

But I think it's the physical package of the keyboard, the algorithm, and the presets tuned to the combination of the two is what made the DX7 such a success.

A new FM instrument could easily be a hit with these factors, particularly if they don't skimp on including built-in speakers and making the presets sound great on them. FM truly shines when all the pieces are aligned in a performer's instrument.

Reface DX comes close to that point, but the presets it ships with are more of an engine demo than sounds to make music with, the speakers are not loud, and the mini-keys (which I love!) were a turn-off for many people — because in the Internet age, people would judge a machine without actually playing it, and that's the only way to understand what's so damn special about FM synthesis.

It's the way you feel when you play it.

A sound demo simply won't get that across.


>the speakers are not loud

Well, let's be clear: the speakers are just for noodling.

Nobody who performs would use the built-in speakers of such a keyboard, not even street performers.

- the only exception is the big-ass speakers in digital home pianos, and still those are only for home use.


>Well, let's be clear: the speakers are just for noodling.

Just for noodling. And trying it in a store. And playing with friends (you don't need much to play along an acoustic guitar). Or playing and singing in a room, or next to a campfire.

All of these applications don't require big cabinets; beefing up the speakers to 5W (and adding a bit more bass response) would do a lot for the reface DX.

>Nobody who performs would use the built-in speakers of such a keyboard, not even street performers.

Well, that's exactly my point (though I've yet to see a street performer with a Reface DX, other than myself, that is).

The portable amps I do use with the DX can be easily built into the body (I have the keytar strap, and even duck taping a micro amp to the keyboard gives enough firepower for street performances).

The built-in speakers on 1980s keyboards were much louder.

>- the only exception is the big-ass speakers in digital home pianos, and still those are only for home use.

This is not quite true.

Check out the small amps coming out these days, that make Roland Cube Street look huge in comparison.

Blackstar Fly is a respectable example; there are many others.

There is no good reason why something like that can't come built into the keyboard.


>Just for noodling. And trying it in a store. And playing with friends (you don't need much to play along an acoustic guitar). Or playing and singing in a room, or next to a campfire.

Who plays an FM synth on a campfire? Not even Kraftwerk! And I don't think it's for jamming with friends with acoustic guitars either. It's more for electronic musicians and keyboardists wanting something portable to noodle at home, or indie musicians for when playing live.


Me? I downloaded a wurly patch “wurly logic” that I can pretty much play like a piano on the reface dx.

I bring it camping all the time.


>Who plays an FM synth on a campfire?

Reface DX owners, mostly, because that's the only modern FM synth with speakers.

I usually use the ePiano patch, which I modify to sound louder; it blends perfectly with the guitar.

>And I don't think it's for jamming with friends with acoustic guitars either

Then you think incorrectly. It's a very versatile instrument, which is for anything that people use it for.

Pigeonholing it as either a home studio or stage instrument is precisely what makes Reface DX such an underrated instrument.


It’s a very portable synth, the small factor, the included speakers and 6xAA battery operation give it playability on the go. I for one use it at home due to limited available space, I use the power adapter and ignore the built in speaker.


>It's also for this reason that DX7 only has membrane buttons, and no knobs or sliders. It didn't need them.

I wouldn't romanticize it as anything other than cost-cutting, later flagship models got sliders.


Sure, but they could get away with this cost cutting, because the synth had enough to offer without them.

A classic analog synth wouldn't sell with such controls.


Beautiful summary of a rich and poorly commonly understood subject


Thanks so much!

I'm in love with my Reface DX, and have a soft spot for all FM synth. I'm glad it came through!


I second the GP's thanks. I know nothing about synths and really appreciated your explanation, and the passion with which you expressed it.


The DX-7 is one of the era-defining music instruments like the CS-80 and the TR808 among others. Here you you can hear some tracks that feature it (complete with weird hair too)

https://m.youtube.com/watch?v=w18E9C8ZG2Q


Very nice.

The DX7 is a musical instrument in its own right. Samplers are emulators of something else.


Well, samplers are also musical instruments in their own right.

Nobody said samplers are confined to playing emulations of acoustic instruments. There are tons of creative users to make patches, and after the initial sample is added (which could be anything), most samplers have a full blown set of filtering, envelope, modulation, fx, audio manipulation etc options.


Could you recommend a patch set for the reface? I bought mine off someone who had some great presets and I loved it, but one day I accidentally factory reset it and there's only a few factory defaults that are really musical as you say.


There's a guy who is a genius at creating Reface DX patches, and he uploads them all for free on SoundMondo. I recommend everyone checking this video to see how wonderful Reface DX can sound. After discovering those patches I've sold my Reface CP because I've liked the DX simulated Rhodes just as much.

https://www.youtube.com/watch?v=zWcEruvc9eA&t=72s https://soundmondo.yamahasynth.com/user/7744


Thanks! Those sound amazing.


You could make your own set, mix and matching from various patches available (including the built-in).

There are some patch editors and librarians to help you: https://ctrlr.martintarenskeen.nl/


I gave away my Reface DX after I figured out that the non-tactile capacitative touch buttons would trigger randomly when exposed to sunlight while I played it in the park.


>I gave away my Reface DX after I figured out that the non-tactile capacitative touch buttons would trigger randomly when exposed to sunlight while I played it in the park.

You might have had a defective one.

I played mine in every possible setting (walking the streets, in the forest, on a mountain top, in the desert, indoors, outdoors, at night, during the day, etc), and never had that issue. (FWIW I applied all firmware updates when they came out).

I did have other issues though:

— once in a blue moon, it'd factory reset itself on startup while running on batteries (a gentle reminder to back up your patches). I didn't have that issue after switching to powering off a power bank with a USB-to-12v adapter (with USB-C supporting 12V natively, that's just a cable).

— sometimes, F and Bflat would stop responding in all octaves. A gentle whack would fix that. Taking it apart and making sure all the connections are tight seems to have fixed it for good.

What I'm getting at is that there absolutely were some QC issues with an otherwise nearly indestructible instrument (between all the drops and two Burning Man trips, boy did that thing take a beating).

Might be worth giving it another go if that's the only problem.


I remember that some keys would randomly stop responding on mine as well. That was frustrating. I do not view having to take apart a modern Yamaha digital synthesizer to make it work as acceptable.


Interesting how that could work. The keys might have had an IR element for force sensing:

https://www.planetanalog.com/force-sensing-eliminates-false-.... (See the bit on the VCNL3030X01 sensor in the middle. I tried to text fragment link, but HN seems to strip that off)


Attempt to summarize one part (the “playability”): You are suggesting the velocity curves and their interaction with the algorithms gave the DX7 a lot of its nuance?


Fine tuning with curves and offsets is what made DX programming an art.

Think of a patch as a vector in the parameter space. For analogue (subtractive/linear model) synthesisers almost all of the space makes some kind of sense. You get a usable sound even if it's weird. With FM (non-linear) synthesis a lot of the parameter space is completely unusable and the great sounds are clustered in little islands around which tiny changes in any parameters has wild effects - especially for patches that use a feedback operator with high settings.

A good DX patch is a finely balanced creation. IIRC the sample rate is 60KHz and the oscillator control resolution is 14 bits, That doesn't give you as much control as with digital virtual synths today. Setting up fine control of key-tracking and velocity is absolutely essential to making the DX preformative.

Now one of the lovely things is how you can download literally tens of thousands of DX7 patches, all the Yamaha cartridges and compilations of peoples personal patch collections from the past 40 years. But because of the extreme sensitivity of the programming not all of them work perfectly with the various emulator plugins and so they need manual tweaking.


Interesting! I share a keen interest in synthesizers.

If you play these patches on the DX7 itself, how good is the reproducibility? I would assume the digital settings to match perfectly. Is there anything else going on that might make patches feel different on different DX7s?

For example, I could imagine some oscillators having subtle differences, perhaps with variation with temperature? If the signal path is all downstream of the same clock, at least until the final analog conversion, I’d expect negligible variation.


As far as I know the engineering standards within Yamaha are excellent and consistent. From what I've heard DX9 patches sound perfect from DX7 but without velocity, and TX7, TX816 and other modules sound identical. Also DX7 patches import perfectly into later Yamaha FM products.

About variance; I'm talking about the design of emulators. There's quite a lot of VSTs and other plugins (that all sound amazing) but the same patch doesn't necessarily sound identical on each. Two classic voices to test from the original ROM presets are Tuberise and GrandPNO2.

BTW these are digital oscillators so temperatures and component tolerances are not a factor.


Thank you for defending the DX7's honor so vociferously so that I didn't have to!


Thank you for this comment! And also, would love to hear your DX7 story :)


Agreed. It's nice that the TFA is celebrating the DX7 because it was absolutely a seminal synth but it's a pretty uninformed take on FM synthesis. Depending on how you count there's really only a handful of fundamental synthesis approaches used for musical tone generation (subtractive, additive, granular, wavetable and frequency modulation). Most synthesizers being made today just use evolved forms and combinations of these basic techniques. As you said, FM synthesis is one of those fundamental building blocks and it didn't go away. Instead, it went pretty much everywhere - including into the ringtone chips in the vast majority of mobile phones of the early 2000s.

Just last month I got a cheap RF doorbell chime for $3 from AliExpress with a dozen different selectable tones and they all have that distinctive FM sound. I was a little surprised that even today FM synthesis remains a cheaper way to generate complex tones than sampled waveforms. The article didn't even explore the fascinating development of FM synthesis which made an unlikely leap from esoteric academic research at Stanford to an incredibly successful product that had global pop culture impact. https://usa.yamaha.com/products/contents/music_production/sy...


DX7s don't go for a high price. The market is flooded with them. If you pay more than $600, you paid too much. You can easily find one in the $400 range.


>DX7s don't go for a high price

It's a 40 years old digital synth (so trivial to replicate one to one), with no special hands on controls as a keyboard, and one with 150,000+ units sold, with about a dozen modern replications.

$400 would already be an impressive sum for such constraints.

But it regularly goes for 800+ in Reverb.com for example.


> (so trivial to replicate one to one)

Good luck finding a 60-key MIDI controller with velocity-sensitivity and aftertouch for less than $200, and it's without getting into replication (of either the controller or the synth).


I rebuild these things. I know the market.


>And of course even vintage DX7s go for a quite high price in second+ hand market...

Everytime I read about some cool old synth, I check how much they cost and they are always hundreds or thousands of dollars, I think DX7s are like $500+.


The DX7 might be the best selling synth of all time. There's a lot of them out there.


Trent Reznor famously destroyed dozens, if not hundreds of them while on tour.

They were so commonplace people were donating theirs to NIN so he could smash them during the show. It was a whole thing.

I don’t think it really moved the needle on the number out there, but it does show how ubiquitous they were.


They are still excellent midi keyboards, in addition to an iconic synth engine. So I suspect DX7s are more widely applicable, and thus are much sought after, which keeps the prices not very low.


The keyboard may have one of the best synth actions ever made - smooth but fast, mid-weighted, not at all spongy.


I read sometwhere that the DX7IID had corrected the original DX7's whimpy full velocity numbers delivery but found that not to tbe the case. I felt I had to use an Anatek POCKET CURVE with its dedicated DX7 compensation curve and beyond when using my IID to program the TX802 and 81Z.


> The DX7 might be the best selling synth of all time

They were, for a long time, but the figure I've seen is something like 100k for the original model. The market is so much bigger now I don't really believe it's true anymore.


The number I'm seeing is closer to 200k in the first 3 years. It was on the market for 6 years so maybe a little more than that.


>There's a lot of them out there.

True, but everyone seems to want $500 for barely working ones and significantly more for decent ones.


I think the cheapest you could reasonably expect to get a useable synth with a decent keyboard is $300, independent of age and coolness. That's where you'll find a Microkorg or a Mininova or something.

Of course you can get romplers and keyboards with preprogrammed sounds much cheaper. And you can probably do something quite nice with a Behringer JT-4000M (once it hits the shops) and a cheap MIDI keyboard.


Depends on your definition of 'useable' but you'd certainly get an early VA board 2nd hand for $300-400 like the Yamaha CS1X.

For monophonic new you have options like the

Korg Minilogue Behringer MS-1 IK Multimedia UNO Synth Pro Novation Bass Station II Yamaha Reface series


Yamaha reface dx goes for 250 on the used market. It’s totally worth the money if you understand its limitations.


oh cool, i've looked at the reface line but didn't realize they go down to that cheap.


Plenty of people hoard classic synths like trophy pieces and never use them. Which sucks for people interested in making music or starting a new hobby with some old gear.

It’s probably one of the most popular things to do in the music scene.


God bless our lord and saviour Uli Behringer for delivering us from the dark ages of Virtual Analog c.1995-2012.


It's not so bad as long as the gears are taken good care of, if this isn't a thing we'll find them defunct in a dumpster yard. There are still positives about having a vintage collector scene, like USB floppy emulator kits, LCD kits, new mods, hacks, sharing old drivers etc.


Hundreds of dollars for an antique long out of production seems extremely reasonable to me. I'm not sure what you were expecting, $10?

It's also a pretty substantial object. They have a metal case! It's quite a large and heavy thing to ship and store.


>Hundreds of dollars for an antique long out of production seems extremely reasonable to me. I'm not sure what you were expecting, $10?

No, it's just that the supply of old synths in general appears to exceed the demand, but because there is this idea that old synths are valuable, people are willing to sit on them with high listing prices instead of adjusting their prices to move them. It's not really an efficient market and it's frustrating.


“DX7 … before fading into obscurity.”

Uh, yeah, that is quite a take. Like most synths, the DX7 went through an untrendy era, but it’s certainly bounced back and you’ll hear its patches used all over the place today. That Top Gun tubular bell patch, for example, shows up quite a bit - most recently heard it on a track called 4AM (the Fauns Remix) by Power Glove.

Not to mention there are a ton of software emulations, including the free Dexed. There are literally thousands of DX7 patches available online, and what’s great about Dexed is it makes it easy to transfer them to a real DX7 using MIDI sysex.

What’s also great about Dexed is that it makes programming the DX7 a lot easier, especially if you have the first revision, like me, with the membrane buttons that look cool in all their multicoloured glory but are complete ass to use.

I get the sense there’s a really thriving community around this excellent and highly influential synth.

And that’s not even going in to all the excellent new FM synths that you’ve talked about, which trace their heritage back to the DX7.


>I get the sense there’s a really thriving community around this excellent and highly influential synth.

Not specific to DX7, but its OPL cousins are also highly regarded and extensively researched in game emulation community.


The DX7 is literally the only synth I've heard of. Well, that and the DX9 and the mini moog. Don't count the Fairlight since it's basically a computer.


> Don't count the Fairlight since it's basically a computer.

Yeah, but what a computer it was!

Sampling, waveform editing, sequencing. Incredible in the early 80s when it came onto the scene. But, of course, it was wildly expensive, and even by the late 80s its capabilities were being eroded by equipment that mere mortals could actually hope to afford, including home computers like the Atari (MIDI + sequencing) and Amiga (sampling + sequencing), along with the nascent audio scene on PCs and Macs.

Surprisingly, Fairlight still (sort of) exist although I have to squint pretty hard to see them as exactly the same company: https://en.wikipedia.org/wiki/Fairlight_(company).


Some great synths are taxing to operate and tweak in real time even if you know what is happening. So I appreciate your point about variations on the interface.


From what follows after that I think they were referring to the configuration interface being effectively unusable by most people when they said “makes no logical sense“.


The tiny LCD UI was a typical cost-cutting measure for synths from that era and isn't unique to FM or even digital synths.

I bought a vintage Yamaha TX81z (they're cheap on ebay) and while it features a tiny LCD UI it also comes with a very logical diagram printed on the top of the case. Yamaha's user manuals are also well written.

I've enjoyed making dozens of FM synth patches on all sorts of instruments (perhaps because I am attracted to the FM sounds that you still hear all the time in pop and dance music.) At the end of the day most synthesis methods boil down to two knobs: signal amplitude and spectral complexity, and FM isn't that different, especially with a single oscillator/operator pair.


I'd argue the opposite, the DX7 worked great because it came with good presets and you could expand it with those cartriges, although programming your own was annoying as heck. But most people cared about getting the newest sounds and at max tweaking them slightly, which was a pain. But the FM engine was/is immensly powerful and the sounds are to that day extremely iconic.

Just as a visualization for the parameters that DX7 hides from its users behind that interface, Matrixsynth made thebDT7 which puts all parameters on knobs and it looks like this: https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEj...


I have a MODX and although it’s so damn powerful when it comes to sound design - each of the 16 parts can have 8 layers each, it’s a lot of work to do anything with it.

I’ve seen piano patches that sound REALISTIC because in 8 layers, you can model the hammer, the initial string hit, the cabinet echo, string resonance, and a whole lot of other timbres… but it takes SO MUCH EFFORT. So unless you’re a /real/ sound designer or you buy patches online, you’re better off getting a Digitone for “simpler” FM or just stick to the MODX’s excellent sampled acoustic instruments.

Yeah… the MODX/Montage UI is really bad and feels like something out of 2004.


But what's the point of making piano patches that sound realistic, except as a tour de force? It's sort of a "Pierre Menard, author of Quixote" sort of thing.

What I'd be interested in is how viable it is for producing intetesting, distinctive sounds, and tweaking them to your liking in a comprehensible way, not poking blindly.


>what's the point of making piano patches that sound realistic, except as a tour de force?

pianos are enjoyable, but expensive and very heavy


And noisy.

VS plug a headset in and stop being interrupted by irate neighbours annoyed by your enthusiastic musical expression, while still enjoying "real piano" dynamics.

(or waking the kid that you just put to bed in the adjacent room and finally it's that me-time of the day)


Patches / parameter sets / whatever else you call the way to describe it are way more compact and should be much less expensive, shouldn't they? I honestly would expect a competent (if not exquisite) grand piano to be among presets in a good synthesizer.


>But what's the point of making piano patches that sound realistic, except as a tour de force? It's sort of a "Pierre Menard, author of Quixote" sort of thing.

Because via modelling (not necessarily FM-based) they can be made more realistic or more expressive than pre-recorded samples (see Pianotte or SWAM for examples).

Also you can then make the piano tonal characteristics as you want them (instead of confined to a fixed real piano's sound).


This is true, but this likely has to be done once, by the manufacturer? Or maybe several times, in search of perfection. But likely it's not what every musician is expected to do. Also, doing this should still be easier than building your own physical grand piano :)


yeah I don't know how about making one personally, but physical modeling (particularly Pianoteq) is near best-in-class if you want a highly realistic piano sound that you can then tweak to your personal preference over something more traditional like a Sampler lib such as Kontakt's The Gentleman.


I would be extremely interested in a video or article describing that kind of patch (one with a physical analogue/interpretation). Are you referring to something specific?


You might be interested in physical modeling synthesis. It's got a long history. Here's a good run down: https://www.perfectcircuit.com/signal/what-is-physical-model...


Yes, I have a hobby of implementing physical modeling synths. I am off the deep end in digital waveguide synthesis, and I was curious about the FM equivalent.


I’m pretty sure my original comment was referring to a YouTube video I saw of physically modelling a piano on the Montage. Can’t find the video now though :(

… but if you’re into modelling… man, I’m hankering for an Erica Synth Steampipe! Just found out about them yesterday. Dear Santa…


Digitone II is available since a few weeks and is an excellent choice for exploring FM!


> He writes like FM went away.

Absolutely. The whole genre of dubstep is based on FM synthesis :)


Its legacy is far from a "weird diversion"


There was mention of the sound becoming 'dull' at the higher octaves. This is not a limitation of the DAC as was suggested, but caused by the phase modulation algorithm aliasing badly at the top of the range. Basically the modulation causes higher harmonics to appear in the output, and if these are higher than nyquist (half the sampling rate) they reflect back into the audible range, and produce unpleasant very digital sounding artifacts.

So, the DX designers avoid this by adding keyboard scaling for the operators - as you play higher up the keyboard, sounds could reduce the amount of modulation applied, and hence reduce the level of the aliases till they were inaudible. Unfortunately if done too aggressively, this will lead to a noticeable reduction in the harmonic complexity of the tone, which is perceived as a dull tone.


Great post! I always thought that the keyboard scaling was entirely for musical purposes. Such as emulating the difference in timbre in a plucked instrument as you go up in pitch.

One of the main DX7 patents talks about adding an averaging filter in the path of the feedback mechanism. Averaging the current sample with the last one helps prevent the high-frequency aliasing. There's also a 16Khz low-pass filter on the main output.

Here's a very in-depth look at the behaviour of feedback FM: https://ristoid.net/modular/fm_variants.html


The one at the output cant really fix aliasing. At this point bad things have already folded in the normal spectrum.


Yes, you're right. I could have made that clearer. I think the filter is just there to remove harsh high frequency artefacts from the final output for the sake of the overall sound.


I know there is a bunch of new DX7 implementations on more modern hardware, i.e., the Raspberry Pi.

When implemented on modern hardware, wouldn't it be possible to run the algorithm at a higher CPU processing speed, to reduce the aliasing at the higher notes, and avoid the need for keyboard scaling and thus preserve the timbre of the higher pitched notes?


You mean, would it be possible to oversample the algorithm, run it effectively at a much higher sample rate and reduce the audibility of the aliases? Yes, it'll definitely help, but no, it's not possible to eliminate this effect, even with monstrous levels of oversampling.

The problem is that, say, a piano keyboard runs from 30Hz to 4Khz, so to get an even tone, at the top of the keyboard, you are only hearing 4 harmonics of the fundamental, whilst at the bottom, you potentially have roughly 1,000 harmonics in the audible spectrum.

Without fiddling with the voice with keyboard scaling, that same tone with 1,000 harmonics played at the top of the keyboard will generate a 4Mhz frequency, so you'd need a 2Mhz sample rate to avoid the aliases reflecting into the audible spectrum.

So that's 45x oversampling, which is a scary amount of effort to throw at the problem, to resolve the issue for this one contrived sound.

Now if you modify the above sound, and double the modulating frequency (the operator pitch) then you'll double the oversampling required, and you have basically limitless control to create higher and higher levels of harmonics to tame, so you can always fiddle with the setting to produce aliasing if you try hard enough :)


If people are interested in the DX7's technical details, I've done some work reverse-engineering it, and some of Yamaha's other FM synths: https://ajxs.me/blog/tag/DX7.html

Ken Shirriff has also done some amazing work analysing the DX7's sound chips: https://www.righto.com/2021/11/reverse-engineering-yamaha-dx...

A really awesome documentary on the DX7 by MadFame: https://www.youtube.com/watch?v=sXt_NXjc7oY

There are a lot of software emulations of the DX7 that are quite realistic, like the amazing Dexed. Here's one that deserves a lot of credit for how accurate it is: https://github.com/chiaccona/VDX7


The design of the DX7 ASIC is really quite astonishing. They used every trick they could think of to make it work given the technology limitations, like a 96-step shift register and using a lookup table to produce the sine wave. The lookup table is clever, too, storing only half the table, and storing log2() of the values instead of the original value so they can use addition instead of multiplication to compute the operators.


and all these tricks to generate oscillators are still used to this day in digital hardware and software synthesizers:

https://aaltodoc.aalto.fi/server/api/core/bitstreams/803386d...


>Here's one that deserves a lot of credit for how accurate it is: https://github.com/chiaccona/VDX7

I read "bit-accurate" and immediately went Ctrl+F "rate": "It then resamples from the native 49khz sample rate", yeah, this one is serious!


There’s a decent supercollider implementation as well

https://github.com/everythingwillbetakenaway/DX7-Supercollid...


Dexed is great, and Arturia has another that is just as accurate.

https://www.arturia.com/products/software-instruments/dx7-v/...


Somehow related, another excellent youtube channel on old retro keyboards: https://www.youtube.com/c/KeenOnKeys


Anthony Marinelli (film composer and synth programmer for many artists including Micheal Jackson) recently made a series of documentaries about FM Synthesis and its inventor John Chowning, the DX7 is mentioned multiple times alongside the Synclavier which used the same synthesis method but predated the DX7.

About Chowning and the DX7 https://youtu.be/Mu8lHX-xuSg

Other videos about FM and contemporary applications https://youtu.be/Pl7BY4C9Cg4 and https://youtu.be/Uoiqy5mkUKE

Aboout Synclavier https://youtu.be/G5T7NBLSY0c


The synclavier went even further with resynthesis capabilities:

You would feed the synth some audio and the synclavier would attempt at reproducing the audio snippet through FM synthesis.


In the future you can get a synclavier on your iphone.

We live in that future.

https://www.synclavier.com/synclavier-go/


Oh, and apparently there's a new hardware desktop Synclavier!

Synclavier Regen weighs in at $2.5K, but then again, it was never a budget brand in the first place:

https://www.synclavier.com/synclavier-regen/


And if you want to hear it:

1. Download and install dexed: https://asb2m10.github.io/dexed/

2. Download and unpack https://hsjp.eu/downloads/Dexed/Dexed_cart_1.0.zip

3. In dexed, load the cartridge "Original Yamaha/DX7 ROM/ROM1A"

4. Load the preset E PIANO 1 and play some Whitney Houston

5. Load the present BASS 1 and play the intro to Danger Zone. Listen to how note velocity affects the attack!


Dexed is great!


It is!

Also, something I have only just noticed... if using the on-screen keyboard, the lower down the key you click, the harder the note is struck.


Yamaha’s engineers did a whole lot with very little. I can only speak to the OPL2 as used in the original AdLib PC card (and to a much smaller extent the OPL3 in Sound Blasters) but IIRC all the sounds came from a lookup table of 10-bit integers representing one quarter of a sine period. Through bit shifts and addition/subtraction, out came a stream of 13-bit floating-point-ish data that the DAC converted into countless game soundtracks and sound effects.

Part of the character of the sound, aside from the whole FM-with-feedback deal, was how abrupt and coarse the envelope steps were. There was no averaging or fine interpolation; just the brutality of fast integer math.


Here's some more technical information about the OPL2, including some images of the decapped chip: https://yehar.com/blog/?p=665

The 4-operator FM synths that Yamaha produced from the DX100 (OPP/YM2164) onwards get that characteristic 'lo-fi' sound from the length of the log-sin table. The 4-operator chips used 256 samples for the quarter log-sin wave, whereas the DX7 had 1024 samples. As you increase the length of the sine table, the timbre gets much 'smoother'.

See also: https://www.righto.com/2021/11/reverse-engineering-yamaha-dx...


Interesting. Thanks!

For the longest time (over 30 years) I've thought that things like the OPL3 were analog. I had no idea that those synth things were happening in numberland.


>These instruments were mostly of academic interest – I can’t think of any band that used one on stage (but I could be wrong).

Very wrong. The Prophet 5, Oberheim 4/8 voice, Roland Jupiter 8, MemoryMoog, and especially the Yamaha CS80 were all available before the DX7, and very widely used on stage and in the studio.

But they really were stupendously expensive - far too expensive for mere mortals.

(There were also fully polyphonic string synthesizers, which had a very, very simple synthesizer under each note, but they were based more on vintage organ technology.)

So by the time the DX7 appeared, Roland and Korg had already produced semi-analog) instruments with digitally controlled oscillators and analog everything else.

The DX7 stood out because the keyboard responded to velocity, and it had an incredible 16 notes of polyphony. So you could play your fancy jazz chords on the epiano sound with the sustain pedal down, and the notes would respond to your fingers and wouldn't cut off too obviously.

The maximum for a pure analog polysynth was 8 notes, and cheaper models had between four and six notes. No fancy jazz chords. Especially not with sustain.

And the DX7 sound cut through - it was much thinner and more acoustic-sounding. The big analog polysynths sounded much bigger - lush, but less versatile, because they were best suited to string and brass fill-in pads, occasional special FX, and perhaps the occasional lead.

Yamaha did try to take synthesis up another level with physical modelling, which is based on more precise models of strings, reeds, pipes, and other resonators, and they produced a couple of very interesting, very very expensive instruments, before abandoning the tech. Sampling is better for imitating real sounds, and the synthetic sounds you get from physical modelling mostly don't sound all that much more interesting than FM.

We're about due another revolution, because there's so much processor power now a lot of synthesis techniques that used to be impossible are now practical. But culturally there just isn't the interest in experimentation. Electronic sounds live in a few small niches - even with modular - and there's not much of a push to explore outside of them.


I'd also like to quibble with this assertion the author makes at the end of the article:

> but almost all digital music production these days is based on sampling

This is very much not true. Subtractive and FM synthesis are still very widely used.


Digitone 2 ($1k USD) just released this quarter, has a strong FM synth in a package with a few other rad synthesizers + killer sequencer.


> The maximum for a pure analog polysynth was 8 notes, and cheaper models had between four and six notes. No fancy jazz chords. Especially not with sustain.

Some had total polyphony though, by using divide down oscillators.

> The DX7 stood out because the keyboard responded to velocity, and it had an incredible 16 notes of polyphony. So you could play your fancy jazz chords on the epiano sound with the sustain pedal down, and the notes would respond to your fingers and wouldn't cut off too obviously.

The dx7 was capable of making sounds that were simply impossible to create with regular analogue subtractive synthesis, because FM (phase modulation in practice) allowed drastic tonal changes in a timbre "through time". So the attack of a patch could have a completely different "waveform" than its release, thanks to operator interaction. That's what the true novelty, a realism no analogue synth could achieve.

https://www.youtube.com/watch?v=1ApmgnKkqaI


Came to quibble with the “no bands gigged polysynths” slander as well. They did, but only for about a 5 year window where it was economically feasible.

There’s Sequential Prophet 5 all over Talking Heads - Stop Making Sense. Van Halen famously used an Oberheim. Basically any record in the first half of the 80s if it had synths.


And again today. There are lots of analog polysynths available now for reasonable prices.


Gary Numan and the PolyMoog. It was his core sound.


Physical modeling is gaining steam these days. A good example is the Osmose, which combines a physical modeling sound engine with a keyboard that not only has polyphonic aftertouch, but lets you wiggle the keys sideways to individually shift their pitches. I have one, and it's easy to sound like you're playing an expressive acoustic instrument; perhaps one that exists in the real world, perhaps not. It's worlds away from any FM I've heard, and more expressive than samples.

They've gotten a lot of professional musicians to rave about it so I'm not convinced experimentation is all that dead.

Osmose isn't the only PM around either. Pianoteq, one of the most popular piano plugins, is pure physical modeling and sounds fantastic (though I wouldn't call this "experimental"). Some of the workstation keyboards use PM too.

https://www.expressivee.com/2-osmose

https://www.soundonsound.com/reviews/modartt-pianoteq-8


we had one in my band nadir bliss. sean loved that thing, he probably still plays it everyday. it's a popular sound texture in the bedroom pop world, aka indie internet artists from like 2014-2020. we played a sort of loud chaotic post punk but some of the songs had a bedroom pop thing. https://nadirbliss.bandcamp.com/album/sharp-distance-2


wait actually this might be the wrong keyboard. anyway listen to my friend's band: https://nadirbliss.bandcamp.com/track/open-a-window-2


The key difference, though, as you pointed out, was the price. These analog synths were incredibly expensive, which limited their accessibility to major artists


Nothing brings home quite how much money there was to be made in the music industry in the '80s as seeing the top bands' keyboard rigs. In the video for "The Reflex", Duran Duran's Nick Rhodes is surrounded by a Fairlight CMI, a Jupiter-8 and a Prophet-5. Adjusting for inflation, that's about $250,000 worth of equipment.


How different the industry was back then...


> The use of mathematics to produce sounds resembling real instruments quickly became obsolete once we could just go out and sample the real thing. Sampling is a boring, brute-force approach to sound synthesis, but it’s a relatively straightforward one, now we have the computing power and memory.

I guess this is true, but the next thing to come along that made me leave the DX7 at home was an Ensoniq EPS-16+.

https://www.vintagesynth.com/ensoniq/eps-16


On the subject of Ensoniq, here's a great article written about them and the huge impact they had on the music industry: https://reverb.com/news/history-of-ensoniq-samplers-mirage-e...


Fun fact, one of the founders of Ensoniq, Bob Yannes, came up with the legendary SID chip found in the Commodore 64.


And to be fair, there's physical modeling, which in general comes much closer to simulating, well, physical instruments.

I think FM is inferior at this if only because harmonics/overtones have less variability.


There is an obvious sounds to pianoteq which is basic the best place ano modeler out there. We are closed to the real thing but so far at the same time.


The article mentions PC sound cards, but this family of synthesizer chips was pretty common in video games overall ca. 1985-1995. They were used in Master System, Genesis, Neo Geo, Capcom CPS1/CPS2, Sega System 16, Midway Y-Unit/T-Unit, Namco System 1/2, and too many one-off Data East, Konami, and Taito boards to count.


I worked on the sound effects and music for the game Wilderplace and used WebDX7 for all of the sound (everything but some sculpted white noise for rain as DX7 doesn't offer pure noise or filters). All the sequencing/synthesis happens real-time in the game. DX7 was crazy fun to work with, like working with clay or something...

DX7 patches have this way of layering with each other that feels both clear and warm at the same time. They don't accumulate mud in the same way that sample-based instruments do. They do have this challenge of accumulating digital-sounding cruft, but I think this is part of their gritty, organic beauty.

Okay but that said I'd love to experience a stereo GPU-based FM synth with high dynamic range and a fully configurable operator graph.


My DX7 take is this: I don't think time has made any other synth less desirable to own than the original DX7's.

- Keyboard velocity maxes out at 100 and the action sucks.

- Membrane buttons.

- Weighs way too much.

- Dinky 2-line screen for controlling dozens and dozens of parameters using one slider.

You can run the entire thing completely correct off of a chip the size of a fingernail these days, so there's no practical reason to own one. No wonder the prices have dunked lately. Even the FB-01 is a much better proposition, at least it only takes up half a rack.


Nah, I’d recommend the Yamaha DX Reface for something that’s a direct DX linage but compact (or you can use a MIDI controller if you want more than 37 keys).


Seconding the Reface DX recommendation.

It makes FM programming enjoyable.

I never figured Dexed or FM-7 out, even though the soft synths and the screen real estate would seem like an ideal match for FM synthesis.

But no, I feel like I only started to grok FM with Reface DX.


Yep, I really like my little reface DX, it's reasonably easy to tweak sounds, and with the help of the free reface DX editor panel ( https://refacedx.martintarenskeen.nl/ ) you can load any patch for 4-op FM synth of yore (DX 9, DX 11, TX81Z) into it.


The key action sucks? The FS keybed is considered one of the best synth actions by many, it’s interesting you say that. It was used in the Korg M1 and other Korg synths for example and lives on as the lead-free FSX in the Montage flagship synths of Yamaha.


But they still make great stuff like Korg Volca FM or Dirtywave M8 and many, many other beautiful synths!

Apart from that, with music equipment, it's completely different than with computers. In our case, great hardware becomes outdated and goes out of use. You could say it disappears forever because its time has passed. With music equipment, it's entirely different. There, everything remains compatible, almost nothing "goes out of use" and it cannot become obsolete.


Any recommendations for a starter synth? Something polyphonic, and has the ability to mess around to create new sounds (so analogue controls, but it could still be digital behind the scenes). I'd like some presets, but not be limited to them. What is on the market at a reasonable price point (under or around $500) that fits this profile?


There's a lot of affordable choices depending on what you want, but everything has tradeoffs. I love my Microfreak, but it only has 4 voices of polyphony, a weird keybed and no onboard effects. That said it makes amazing sounds and has a very discoverable set of controls, and since I play guitar I have external effect pedals at home anyway.

If you're not committed to hardware, a nice midi controller and a software synth will let you learn without spending big bucks. And even then the midi controller is optional.


Hydrasynth Explorer

Minifreak

buying used off of ebay / reverb is fine

Or get a Arturia MiniLab 3, it comes with a synth package and can control anything on your computer.

Good free synths:

- Dexed

- Vital

- Odin

- Pianobook / Decent Sampler


Synthstrom Deluge is also going to have DX7 emulation soon.I can’t wait to have a DX7 in a powerful groove box.

https://m.youtube.com/watch?v=84EENKfeV8M


The longevity of music gear is one of the coolest aspects of the industry


> And once we have sampling technology, there’s little point using computational techniques to make the once-novel sounds of earlier synthesizers: we can just sample the instruments themselves

You can't tweak the parameters of a filter envelope if it's "printed" to a sample.

You might sample a variety of different envelopes, but there's still fundamentally no way you can get the sound of the filter attack gradually increasing using samples.


Assuming the details of the DX7 are similar to some of the PC Yamaha FM synth hardware, you could do things like set the attack/decay of a modulating operator to be very slow, the attack of a carrier to be fairly fast, and subsequent key presses wouldn't necessarily sound the same. And with a big chained algorithm like the DX7 can do, I agree, you couldn't capture the full behavior through a sample.

You could replicate the most common behavior, though.


Which is part of what makes the Mellotron such an interesting instrument.

https://en.wikipedia.org/wiki/Mellotron


"Tuning a Mellotron doesn't." —Robert Fripp


You can totally tweak the filter and amp of samples. Don’t believe be - run a sample with the LP turned all the way down :)


If the sample is the master output from the synth then the filter won't be the same.


Of course it won’t be the same, but the comment said won’t work not won’t be the same


Sure you can. Check out what Ensoniq did with digital-analog hybrids.


Hey this is the synth I started with way back! I used this[0] 3 hour video to teach me the ins and outs of sound synthesis from the NY School of Synthesis by Dean Friedman.

It still holds up today as it did back then for anyone trying to get into synths.

[0] https://www.youtube.com/playlist?list=PLFa4_po__8wMNscyFfDy6...


On the subject of the DAC used, and why it was 'only' 12 bit. I believe the instrument actually runs all of the voices through a single DAC, so it is basically 16x oversampled, and the voices are held by a sample and hold circuit for all 16 voices before they are summed to the analog output.

If the later DX mark II is anything like the TX81z (which is also said to have a 16 bit DAC) it actually uses a 16 bit dynamic range DAC, but with a 10 bit resolution (a YM3012 for the TX - https://www.alldatasheet.com/datasheet-pdf/pdf/1179489/YAMAH...)

This is definitely the age of interesting DACs, with all sorts of interesting noise profiles. We're really spoilt these days with the quality from delta/sigma DACs, thinking that this has always been the case!


I had one. I wrote MIDI software. I wasn’t a particularly good keyboardist.

Pretty damn powerful, and a "pure" synthesizer. Too complicated for a lot of musicians.

Things really took off, when samplers hit the scene. People were much more comfortable modifying existing sounds, as opposed to creating them from whole cloth.


> Building an analog synthesizer that was polyphonic – where many keys could be played at the same time – was stupendously expensive. The same electronic circuitry had to be duplicated many times, sometimes for each key. These instruments were mostly of academic interest – I can’t think of any band that used one on stage (but I could be wrong).

Divide-down synths like the Arp Solina and Omni were widely used in live bands and weren’t that difficult to trek around for the period. Certainly lighter than the contemporary transistor organs of the era

True Analog polysynths were also used extensively live going back to the late 70s with the arrival of the Prophet 5, Roland Jupiter 4 - and by the early 80s (before the DX7 arrived) there were a tons of artists performing with them live.


Whats hard (and impressive in regards to those who came up with meaningful patches in the 80s) about starting with FM synthesis isn’t only the programmability (modern FM synths have that covered, see the other comments). It’s not an intuitive thing to work with algorithms and you need to learn what is happening.

I can recommend FM Theory and Applications by Musicians for Musicians for everyone who wants to create own patches:

https://www.burnkit2600.com/manuals/fm_theory_and_applicatio...


In addition to this wonderful resource, I can also recommend getting Dexed and just playing around. It's free, open source and, importantly, as unintuitive as an actual DX7. Great for digging deep into how and why a change in an operator creates a change in a sound.


As I noted in another comment, you can start out, and get pretty far, with just a single operator pair, which basically determines the waveform and spectral complexity. For me it has a similar feel (unsurprisingly enough) to analog FM, ring modulation, or oscillator sync.


Yes! Dexed is great and I would recommend trying the samples from the book with it. You need a bit of exploration, to get a grasp whats going on.


I lived that era, and don't miss the DX7 at all. Yes, when it came out it was revolutionary and if you wanted to produce very thin sounds, (some) credible organs, really nice electric pianos, punching basses, and metallic sounding instruments that one was the keyboard to use, no doubt, but the drawback was that almost everything else would sound terrible, I mean really awful. Additive synth is great only if done well, and the first DX synthesizer were a lot limited in some fields; if you tried to make a track only using a DX7 it would be a compromise wrapped around other compromises. For starters, operators could generate only sine waves, and every other waveform had to be emulated by frequency modulating an operator by another synced operator; that could be used to make some really interesting effects and sounds, but no way one could produce for example believable strings; they were just terrible. Also it had no filters, that is, no VCF/DCF/whatever, so no way to produce credible sweeps or warm pads that would have been trivial to produce with a 10 years older keyboard. DX7 has been revolutionary for sure, but had its share of limitations, and (plugins aside, of course) rather than spending money on a vintage DX7, I'd get a smaller and much cheaper FM expander to be driven together in unison with a traditional analog sounding one; the 1st one for the aggressive attack, the 2nd one for more warmth. As for additive synth, I deeply regret having sold my Kawai K5000 years ago; that was really a beast. Heck, That applies pretty much to all my old gear, but sometimes one is forced to, only to regret much later.


Yamaha basically still makes the DX7 in a modernized form in the Reface DX. It just has 4 operators instead of 6, but it can create various waveshapes whereas the DX7 can only create sine waves so it requires more operators to achieve the more complex shapes.

Also, Korg makes the Volca FM, which is actually a clone of the original Yamaha engine and you can actually directly load DX7 patches into it.

>There remains a niche interest in generating instrumental sounds using mathematical modelling, but almost all digital music production these days is based on sampling.

Yeah...that's just completely untrue.


Korg Opsix can also load Dx7 patches, and has all six ops. And many waveforms beyond sine. And other weird shit you can use instead of an oscillator in an operator spot, like a comb filter and other things. Tons of fun.


Does anyone know a great mini (2 octaves or so) keyboard that is portable, fits in a backpack, has a built-in grand piano, touch sensitivity, and supports bluetooth?

EDIT: I found this [1], which comes close but does not have bluetooth.

[1] https://eu.loogguitars.com/products/loog-piano


I recently was obsessed with FM for a while! If you want a bit of fun, here’s a synth I made, the codes fairly simple if you want to poke around, you can fool with parameters easily. Doesn’t work great on mobile though. https://chrishulbert.github.io/you-synth/


> "My Kawai stage piano has about forty different instrumental sounds, all produced from samples of real instruments. Apart from the string sounds: I believe it creates those by tormenting cats on Xanax."

I don't quite know why but the author is really correct on synth strings. I've got a few friends who are fantastic semi pro strings players and the expressionism of their instruments are phenomenal -- I've sung with them a lot and played orchestrally as well. Synth strings inevitably sound like synth strings. I'm sure there's a realistic sampler somewhere on the planet but I've never come remotely close to discovering one that is any good ...


Synth strings sound like the original "string synthesizers" like the Solina String Ensemble (famously present in Jarre's Oxygene, albeit played on somewhat different instrument, and through a phaser). They tend to be scratchy, nasally, thin, etc.

Sampled strings cen be really convincing. Listen e.g. to https://vi-control.net/community/threads/phaeton-symphony-by...


Regarding synth strings, maybe Roland Integra-7 (and stuff like that) can do the trick.


Brian Eno, of course, insisted that the problem with the DX7 was entirely the fault of the UI. He had a custom controller made, although I can’t find a photo of it. These days, the DT-7 attempts to emulate it.

Edit: It wasn't a custom controller, image in reply.


The programmer was a commercial product, rather than a custom design made specifically for Eno. It was manufactured by a German company named Jellinghaus:

https://reverb.com/item/21724471-jellinghaus-dx-programmer-f...


Thank you for the correction! Managed to get it under the edit window as well so I can acknowledge it.


FM synth is tough to tame, but the possibilities are endless. More so than the standard subtractive VA that most synths use. I feel that we haven't yet fully explored what FM has to offer!


>I feel that we haven't yet fully explored what FM has to offer!

Processing power prevented a lot of fun possibilities, if one day consumer CPU become power enough to do 1Mhz upsampled DSP with ease, we can brute-force a lot weirder wave shaping algorithms without too much aliasing.


Try listening to early 2010s Skrillex!


That's totally wavetable synthesis though right?



I learned to play the keyboard on my father's DX11, in the late80s/early 90s, and remember the sounds with warmth. The DX11 doesn't seem to have been at all as well known or iconic though, and have only seen the DX7 being written extensively about.

It seems to have been quite closely related to the DX7 though, based on some info I found:https://www.vintagesynth.com/yamaha/dx11


>They don't make them like that any more: the Yamaha DX7 keyboard

Greg Kihn Band Breakup Song with the unforgettable lyric, "they don't write 'em like that any more" did not use the DX7 at least because it was not available in 1981, but the keyboardist Gary Philips was known to use Minimoog, Prophet-5, Wurlitzer, Rhodes and Yamaha pianos with the band. In 1983 Gary Philips was an early adopter of the DX7. They don't make them like that any more either.


I didn't know how to play the keyboard but the DX7 was seen in so many music videos in the 80s that it became an object of desire for me. If I were going to learn how to play the keyboard, it was going to be on the DX7! To me it was the Lamborghini of keyboards. I had zero understanding of synths or even basic music theory. I just knew all the cool bands had a DX7 and I wanted to be like the cool guys on saw on MTV. We eventually got some cheap consumer keyboard from K-Mart which was completely forgotten by New Years Day.

As an aside, I find it's a bit weird how overlooked the Greg Kihn Band is in current 80s culture. It might not have been as big as Michael Jackson or Chicago but they were quite well known. Weird Al even did a parody of one of their songs but their cultural impact seems to have not survived as well as other acts of that era, which is a pity.


As an aside to an aside, Greg Kihn became a DJ for a classic rock station in the Bay area that I used to listen to. Even then I only know one of his songs.


Same, but in my case it was the Roland D-50.


All these erudite comments were fun to read. I can't help comparing with guitar enthusiasts who are looking out for really old Fenders and Gibsons to get the authentic old fashioned sounds. To my mind the value of a DX7 is just like that. It has a sound relevant to its time in history. It is what it was and has a soul of its own. Has time and technology produced more sophisticated instruments? Sure, but I don't think that's the point.


I hope they DON'T make them like the DX7 anymore.

What the article says is true, the DX7 sound dominated pop music for a long time.

Many musicians were too uncreative (or perhaps it was too much of a hassle) to change it from the boring, over-used factory pre-sets. I seem to recall it was really, REALLY an unpleasant chore to make interesting sounds-- endless modes and everything set up with a single button.


If you want a professional Yamaha FM synth from the 1980’s, the V50 is a bargain. It was the last of the DX’s and is essentially two DX9’s with a sequencer and a drum machine.

For clarity, the later 4 op FM chips, have multiple wave forms and provide a sonic space as powerful as the earlier sine wave only six Op chips…they are not identical but equivalent.


Also of note, the SY77 which bridges the best of Yamaha FM with AWM2 rompler tech.


absolutely. i have an sy99 and they often go for less or equal to a dx7, and they are much more of a powerhouse. can't recommend enough


DX7 will always have a special place in my heart being the spiritual predecessor to my favorite soft synth, FM8. If you want a modern FM synthesis experience, FM8 has been that for over a decade. Low CPU usage, incredibly versatile, really easy to work with in the mix (FM synthesis tends to be easy to shape and layer spectrally).


> being the spiritual predecessor

There was the FM7 before. It already fixed the UX/UI problems of the hardware version.


Yes they don't make them like that anymore. Most synthesisers nowadays have much better user interface.


> An interesting development was the discovery that, if the operators started with square, rather than sine, waveforms, it didn’t take as many operators to create interesting sounds.

Is this true? How and why? I imagine the square basis also contributed to the brilliance, not being band-limited.


It has the most harmonics.


Just to add: harmonics are sine waves, in essence.

A pure sine wave has only one harmonic.

You need to add up a bunch of sine waves of different frequencies (multiples of the fundamental) to get something as, in a way, unnatural as a square (or sawtooth) wave.

That's why they're great building blocks for subtractive synthesis too. Pass them through a resonant filter, and you get something interesting.

They are also very easy to make digitally (a square wave is just on-off-on-off..).


DX7 is a great example of how new technologies can ruin a previously-great user experience. Going from cables and knobs and an "everything is visible at once" UX on analogs to a small LCD panel with data entry buttons was a huge regression.


The DX7 is an incredible synth, but one of the joys of playing is hearing the sound change to your adjustments and the DX7 just doesn't do that. But it is great fun to feed through filter, delay and reverb pedals. Thats kind of the best of both worlds


just for the audience, the DX7 used a newly invented type of synthesis (FM but that doesn't tell you or anybody else anything) and it was alone among synths at the time to produce metallic, brass, bell, etc. sounds. Performers flocked to it to get away from the same old analog synth sound everybody else had. FM synths are programmable, but not easily, and the menu diving was not practical while playing.

Analog (subtractive) synths are very much back in popularity right now and they offer lots of knobs to give bwahgrrrssshhhhwow type transformations


>The DX7 is an incredible synth, but one of the joys of playing is hearing the sound change to your adjustments and the DX7 just doesn't do that

...but Yamaha Reface DX does.

I've been changing the sound while playing the synth to go from the ePiano preset to an aggressive synth lead as a demo of that interface.

It's pure joy, and an underrated piece of hardware.


It's even better with the Dtronics expander, but still too much faff for me.


Thanks, I didn't know about it - and it seems like an awesome addition to the machine!


I can appreciate these instruments now, especially in their historical context, legacy in music, and from a technical perspective. But I still carry a 30 year old dislike of "that FM sound", for the dumbest of reasons:

I was in college in the 90s. My roommate had a PC with a Soundblaster card, and he played this Star Wars game incessantly. It has a terrible, insipid, repetitive, infuriatingly annoying soundtrack that I grew to despise with every molecule of my being, mostly because I was subjected to it as an eternal everlasting torment, but also because I had an Amiga and my computer was clearly superior in every way especially when it came to sound.

Anyway, hearing this - https://www.youtube.com/watch?v=TexOZwatuLg - constantly caused me to associate all FM sounds, even in good music, with this cheap "wah wah wah ee ee wah wah wah" noise, which I've only partly shaken decades after the trauma. Also, when he wasn't playing that one, he was playing this Dune game (https://youtu.be/gOscXf0Fpmk?si=hUZ8g0GZ48CKfKzh&t=7394) with a more varied but equally cheap sound, and on top of that playing the game triggers the most repetitive samples ever devised by man. Reporting, acknowledged, reporting, acknowledged... oh, the torture.

Where was I? Ah, that's why I'm not a fan of FM synthesizers.


For those of you more interested in the synthesis than the keyboard, the Elektron Digitone is one of the easiest FM synth to use, while still capable of an incredible variety of sounds.

For a software recreation of the DX7, see the Dexed plugin.


Still use mine for bass sounds.

Nightmare to program by hand on the fly, but it has its classic sounds.


FM synthesis was developed by John Chowning on the Stanford AI lab PDP-6 and PDP-10.


FM synthesis is a bit like Lisp--everybody says it's cool, but getting acceptable results is harder than with another approach. When I need FM, I use Arturia's _DX7 V plugin and only play presets.


One of my favorite videos on the internet involves DX7: https://www.youtube.com/watch?v=eL_N3A5gVhQ


That was painful, but I couldn’t quite look away. At 4:00, I was the one with the distressed look.

But I can’t deny that the end result was startlingly realistic in general, though obviously still a bit overdone.


I am kind of still confused why the industry links „keyboard controllers“ with „synths“. there should be just good keyboard controllers that can plug into good synths (like Eurorack). What am I missing?


You’re missing MIDI controllers, of which there are plenty. :) Synth versions without the keys are often called “desktop versions”.


There are a huge number of MIDI controllers and keyboardless synths


> am kind of still confused why the industry links „keyboard controllers“ with „synths" there should be just good keyboard controllers that can plug into good synths (like Eurorack). What am I missing?

There are good keyboard controllers that you can plug into a Eurorack if you're a Eurorack player. Many synths haverack mount versions. You're missing nothing in that sense.

Your question isn't about the industry. It's about musicians. You're asking why musicians other than you still want complete instruments.

The answer is simple. People who play other instruments need other instruments. Of which the keyboard can be an integral part.

You might as well ask why people still buy hardware synths when perfectly good software synthesizers exist.

The answer is the same: they play differently (that applies both to the machines and the performers).

Here are some reasons why synths come with built-in keyboards.

The most important reason:

— Keyboards are mechanical instruments with particular response curves. The patches the synth comes with can be optimized to play and sound good on the keyboard the synth ships with.

This guarantees quality out-of-the-box, consistency, and makes it so that your presets would feel the same when your play them on another device.

Yamaha DX-7 played with an 88 weighted keys keyboard is a different, and a worse instrument. As is a Clavinova played when DX-7 as a controller.

— Ergonomics matter. Each synthesizer is unique in the way it builds its sounds, which is why many come with dedicated controls (buttons, knobs, sliders, screens).

Where these controls are in relation to the keyboard matters.

I can play drum pads while playing the keys on my Akai Mini Play 3, for example, but not on my larger Yamahas. I can tweak the patches on the Reface DX live, but not if I used the MX-88 as a controller.

— Physical layout matters. Performers to look the audience, not the knobs; the audience wants the performer to look at them.

Memorizing the layout by touch isn't possible when the layout isn't fixed.

For gigging musicians, the following are important:

— portability

— physical robustness

— time to set up on stage

— space on stage

— physical setup

— ability to move while performing

Having more cables to connect isn't something gigging musicians want when you have 5 minutes to set up for a 30 minute set at a dimly lit bar.

And then there's the setup. You put a stand on a stage, there goes your controller. Where does the synth go? Oh, you need another table or stand for the synth.

Now to tweak a patch. Your controller may not have all the knobs your synth engine uses. The sound module has them, but it's on the table over there, while you're playing facing the audience over here. Bummer.

Finally, I have straps for some of my keyboards, so I can move around the stage during the show.

1/4" cables are long, and made to withstand abuse; USB plugs agent, and MIDI cables, while better, aren't made for that either.

There are also marketing reasons:

— A synth that people can try in a store is more appealing than one they can't. Rackmount synths don't make for good displays.

If people have inconsistent experience trying the instrument, it doesn't bode well for the brand.

— When someone plays a Nord keyboard, you know it, because it has a distinctive look. Musicians take note of other musicians' gear.

— An instrument that can't make any sound on its own is not an instrument, it's a module.

It's a different product.

And then we get to psychological reasons:

— Performers are deeply attached to their instruments.

A sound module without a keyboard isn't an instrument. A controller isn't an instrument. Together, they form a system made up of components; it doesn't feel as an instrument the same way a guitar does.

— Looks matter. Performers want to look good on stage. And a generic controller simply doesn't look as good as a complete instrument, with a dedicated interface.

Feels matter. The feedback one gets from touching and playing the instrument affects the performance.

And that's before we get into things like:

— busking

— jamming with other people

— playing outdoors (I played my Reface DX in the mountains and at campfires, in the streets and on the road)

The TL;DR, though, is that if you want a controller for your Eurorack, you've got it.

Saying that's how everyone should make music, though, is quite a reach.


Yamaha keyboards were top notch. I had a SY-85 with the nicest keyboard I've ever used for my purposes. Good feel, velocity sensitive but non weighted, fully polyphonic aftertouch.


Fun fact: The Taco Bell bell gong sound is an E on the DX7

https://www.youtube.com/watch?v=4DBrC_l-VCs


It’s a perfect example of how the DX7’s FM synthesis could create unique, bright sounds


sounds like the bell in the opening of the top gun theme


If you have a techie friend from India ask if they have an old TVS Gold keyboard.

Pretty nice model m clones.


As a former owner and synth head, I feel the DX7 was good for poorly simulating a broad range of superior instruments. Also it wasn't merely difficult to program - it made sound design a necessarily separate activity from playing due to the notorious programming UX. Revolutionary and a nifty museum piece.

Relating to it's influence on 80s music, this oft repeated claim is laughable. Tom Oberheim and the late Dave Smith had demonstrably greater influence with the OB-X* and Prophet lines reigning supreme and offering lasting sonic contributions.


>poorly simulating a broad range of superior instruments

Nothing touches FM's velocity responsiveness until physical modeling, it sounds lame when you get used to it, but the same thing is true to the romplers that replaced it.

>Relating to it's influence on 80s music

80s is quite broad, claims like this basically just showed which artists the writer prefer. Personally, it's PPG and Moog.


> Nothing touches FM's velocity responsiveness

??? You can make any synthesis velocity responsive. Oberheim had it, there were Rolands in the DCO era. Vangelis mastered it on the CS80.

The DX7 simply made velocity sensitive keys widely available, and critically Yamaha baked velocity sensitivity into every patch.


The FM trick is that the available parameters that can be modulated through velocity result in a wider range of tonal variation, where traditional subtractive synthesis is usually simpler, and velocity changes affect parameters that don't result in wide changes to the tone, such as volume or filter envelopes.

With FM, modifying an operator envelope can result in a completely different sound. A well-crafted patch that takes advantage of this, played by a good keyboardist, will feel alive, in a way traditional synths of that time couldn't achieve.


I get it, you’re an FM fan, and I think that’s admirable.

To the analog fans it’s VCOs that feel alive, whereas the DX7 is cold and sterile. It’s all a matter of what timbres you enjoy.


Not forgetting aftertouch.


> The use of mathematics to produce sounds resembling real instruments quickly became obsolete once we could just go out and sample the real thing.

No. One of the most popular piano VSTs on the market is Pianoteq, which uses physical modelling of the instrument rather than being sample-based. It's often considered one of the most realistic-sounding piano VSTs all the same (they have models of many popular acoustic models, authorized by their builders), and offers a much higher degree of dynamism and reactivity than sample-based ones, and much more flexibility in simulated microphone placement, etc.

This approach is still leading-edge in places.


Now it’s the Nord Stage series that is the crown jewel of gigging keyboardists. I sold off my Fender Rhodes and Hohner Clavinet years ago when I bought my first Nord. Nothing can take the place of a real B3/Leslie though!


Now, how about drums; an instrument everybody loves?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: