[sdiy] A different kind of polyphonic aftertouch

rsdio at audiobanshee.com rsdio at audiobanshee.com
Fri Jan 5 13:07:54 CET 2018

If all you want is Velocity, then the advanced sensors are not better than a switch. You do have a point there, but only in that specific case.

However, if you want to sense Position continuously, such as for Polyphonic Aftertouch, then you no longer need a switch at all. The ability to measure Position continuously allows you to know Velocity by the change in Position. I think the only reason anyone says anything about this paradigm change is that the Ensoniq inductive position sensor allowed them to save money on the switches, which also means that you never have to clean or replace those switches (since they aren’t there at all). Anyone familiar with vintage synth maintenance knows how common it is to have to pull apart the keybed and fix one or more switches that are dirty or broken.

The SQ-80 actually has an entire processor dedicated to the Poly AT keybed. It’s the same part on the EPS and other Ensoniq models with Poly AT. This connects over a parallel bus to the main computer that handles all of the synthesis, MIDI, and display (at least I don’t think Ensoniq used a separate processor for the display - they had a 68000 and an Ensoniq Oscillator Chip by the time they were designing with Poly AT). I’ve never found out whether the Poly AT circuit board uses MIDI (probably not) or some enhanced protocol to tell the synth what’s happening with the keys. So, yes, it has the numerical integration capabilities - which is not much more than addition and subtraction anyway.

Not every Ensoniq keyboard was clacky. I waited (for budgetary reasons) until at least the third hardware revision before buying my EPS, and I think it has the best keyboard action of any keyboard I’ve ever played. As Syntaur has shown, it’s possible to upgrade the clacky Ensoniq keyboards to be a lot quieter, so it should be possible to rescue the earliest models so they don’t suck.

The worst action in my opinion is on those keybeds where the return springs are way too strong and the keys have far too little mass. It’s hard to hold a chord without feeling like you’re doing too much work. Granted, I can’t deal with fully-weighted piano style action because I’m a guitarist, but there has to be at least enough mass to feel like you’re playing something real. The reason I like the (non-clacky) Ensoniq action best is that they weighted the keys just enough - not full piano, but not cheap plastic - to allow someone who wasn’t formally trained on piano to get a great response.

I’m really surprised that there aren’t better keybeds. Even back in the day, nobody made their own keybeds. Seems like they were all outsourced to a few dedicated keybed manufacturers. Why there isn’t at least one standard, high quality keybed today just doesn’t make sense. I’m also surprised that Poly AT isn’t more prevalent.

I agree with you that the physical feel and response are way more important than the underlying technology. As long as the technology works, it doesn’t really matter what the details are. Ensoniq got rid of the switches because they became redundant with Poly AT, but this didn’t stop keyboardists from using them just like switch-based keyboards.

My associate has done a lot of research (as part of a Masters degree) on human interaction with musical controllers, regarding sensitivity, timing resolution, and latency. It would be great if more manufacturers paid attention to these sorts of factors when designing their instruments, rather than just buying the cheapest parts and plugging them in.


On Jan 5, 2018, at 2:57 AM, Richard Wentk <richard at wentk.com> wrote:
> Why would these methods be better than a switch? 
> You’re only ever going to be measuring average travel time between two reference points. 
> Do you think an SQ-80 had the processor power to do some kind of continuous numerical integration, and even if it did, would that generate a “more accurate” velocity measurement that somehow felt hugely more responsive to a player?
> IME the biggest factor affecting playability is the quality of the mechanical action. 
> E.g the DX7 had possibly the best semi-weighted action ever, with a good-enough mono AT mechanism. 
> I don’t recall ever thinking it would have been improved with linear sensing. I do recall thinking “This is an exceptionally good action.” (Often followed by “I wish this had Poly AT...”)
> The Ensoniqs had a clacky action that was distracting to play. Poly AT was very nice to have, but - unlike the DX7 - the keybed was always in the way. 
> Nowadays most keybeds are crap. NI’s Kontrol series is bearable, Arturia’s keybeds have joke fake AT that ignores the sensor and simply fades in a simulated AT output with an envelope. Cheaper controllers (M-Audio, etc) are just horrible. The high end weighted Fatars are okay-I-guess, but a long way short of a good piano action.
> The most interesting thing to appear for a long time is the Roli Seaboard, which has true poly velocity, poly pressure, and poly position sensing, and is also very playable (although it takes some getting used to.)
> Bottom line is no one is bothered about linear sensing. They care how the keys feel under their fingers, and what they can do with them.
> R
> On 4 Jan 2018, at 22:29, Mike HEQX <mike at heqx.com> wrote:
>> As we see in the Ensoniq SQ-80 there is no physical switch. It was a change in inductance that was being measured and they did have a rubber pad under the key and they did give you polyphonic aftertouch ( not related I know ). So they used inductance, and someone else uses light, and yet another uses capacitance, and these are all linear sort-of systems that in my opinion are better than the two switch method of generating a psuedo-velocity sensitivity that measures time.
>> So really the majority of synth keybeds that claim velocity sensitivity are really just sensing attack velocity. Some offer a choice of attack curve, which is artificially generated for you. Very few are actual hammer actions ( in the piano paradigm ), and very few are doing anything unique. Long live linear sensing!
>> On 1/4/2018 2:36 PM, cheater00 cheater00 wrote:
>>> You put a spring or piece of rubber under the key and suddenly the
>>> linear sensor is a pressure sensor.
>>>> On Thu, Dec 14, 2017 at 6:46 PM, ASSI <Stromeko at nexgo.de> wrote:
>>>>> On Thursday, December 14, 2017 5:09:40 PM CET MTG wrote:
>>>>> https://deskthority.net/wiki/Adomax_Flaretech
>>>>> "There are no electronics in the switch modules themselves. Instead,
>>>>> surface-mounted infra-red optoelectronic components on the circuit board
>>>>> provide the sensing in conjunction with a prism in the module's
>>>>> transparent slider."
>>>> So a vertical mouse, sort-of.  Which means it's very likely measuring key
>>>> travel rather than pressure, which is a lot less useful for a musical
>>>> keyboard.

More information about the Synth-diy mailing list