Potentiometer verse Encoder Smackdown

[Markyboard]

Carlo (aka Marino) and I revived our longstanding dispute over synths with pots vs endless encoders in his most excellent GearLab Sequential Circuits Pro-3 Impressions. I know this has been discussed more than once but thought it worthy of its own thread. Plus I need some backup so...gather round pot heads .

 

From Carlos Pro 3 thread....

 

 

 

POTS vs. ENCODERS.

This is a subject which I have discussed many times. Simply, having potentiometers â as opposed to endless encoders - on a synth with patch memory makes no sense to me. I hear several fellow musicians saying, 'I prefer to have visual feedback of where the parameters are'. Well, I can understand it in the case of a Minimoog, where by looking at the panel, you can almost have an idea of what it will sound like. But in case of a synth with patch memory, in *no case* you have visual feedback. Whatever editing system you use, 'Jump', 'Pass Thru' or 'Relative' (SCI offers all three), when you recall a patch, every parameter is set at a certain value, and its relative knob is set somewhere else. Then when you start editing, you'll have some knob which position matches the actual value of its associated parameter, and some other (the one which you haven't touched yet) which are placed differently. And when you first use a knob, you are *forced* to look at the screen, to check the actual starting value that was stored in memory.

 

So the logic solution is to use endless encoders, which always start from the stored value, then go in increment or decrement from there. I have learned this in the old times on the Matrix-12. Of course, you have to have the screen showing both the parameter and its value as it changes. And perhaps, also a visual reference associated with the knob, LED ring or whatever. What's sure to me is, with pots you can *not* have immediate visual feedback.

SCI uses a combination of both pots and encoders, which to me is even more confusing. I mean, it's manageable, but irrational and counter-intuitive. Once again, the Xenophone only uses encoders: Both dedicated ones, and four 'soft' encoders over the main screen. Hats off.

Endless encoders, like OLED displays, are more expensive and have a shorter life span - but in this case, I feel that they should be adopted universally. In more than 30 years, I have replaced the encoders on my Matrix-12 exactly once. I think that's more than acceptable!

.

 

As usual I think some of this boils down to your application and how you work. But I"ll start with the easiest part, my subjective opinion;

 

I like how pots feel. I like that they tend to have more physical resistance and they have a consistent feel and performance. They always start at about 7:00 and stop at 5:00. You control the rate of change, i.e. there"s no fast mode vs slow mode which often annoys me. In general pots feel more substantial to me. This is even true for those tiny Eurorack pots spaced an ant"s pecker width apart (thanks for that one EscapeRocks ) although admittedly quality plays into it big time.

 

For real time control, live performance etc. I think pots in relative mode is the way to go. It avoids the sudden jump in the parameter value while still always being active. Unfortunately not all pot laden synths have this feature and I totally get the detrimental affect of sudden jumps during the music. Of course just like preset jumps with keyboards that don"t sustain the current sound through the switch-over we learn to plan around these things.

 

Now for my application which is mostly sound exploring, experimenting, programming...

 

I don"t care about jumps in sound when modifying the called up preset. It"s true there is no useful visual information for the pot positions when first starting to modify the preset. But I often work in sections, for example anything affecting the filter. I may be modifying 6 or so parameters in short succession, back and forth etc. until I"m at a point of fine tuning them. Those 6 pots quickly do show their contribution to the overall affect with their physical position, something I value. Not so with encoders unless they have the illuminated rings which granted are way better than those without. I don"t care about the other pots that haven"t been moved yet. If I do I"ll turn the pot(s) of interest, let the jump happen and use my ears to get it back to where it was or more likely where I want it to be. Now there are some parameters that can quickly destroy the sound in which case I use the display and or compare function to tell me where it was set in the preset.

 

 

I didn"t mention sliders but just not a big fan mostly due to reliability. My Polyevolver started with encoders, they were very poor quality and eventually I upgraded to pots. This was even after replacing some of the bad encoders with better quality ones. Once I transitioned to pots I never looked back. They feel wonderful many years later and never a problem.

 

Ok that"s enough from me for now.

[The Real MC]

The biggest problem is when a control becomes defective and you have to hunt down a replacement that fits the EXACT PC BOARD FOOTPRINT.

 

Slidepots are my least favorite because of their limited life (unless they're Penny & Giles). Their footprints vary all over the map, and resellers do not stock very many varieties of new ones. NOS is not a solution because they already lost some life, they do degrade from inaction. Scavenged slidepots are even worse as they have less life left out of the gate. Some of my devices do have slidepots; Lexicon model 200, Korg SDD-3300. These devices are the exception because they get sounds that very very few others can. I just acquired a basket case ARP Avatar which will be getting new Bourns slidepots along with gmusynth.com adapter boards.

 

Not a fan of encoders for many of the same reasons but they are better on devices with shallow feature sets that won't get edited often, like my Eventide 2016. The opto encoders are way more reliable due to fewer moving parts.

 

Give me rotary pots any day. While there are types out there that are short life, you can get better ones for not much more money. If a rotary control goes bad they are a lot easier to find a direct replacement, or if possible just use one with solder eyelets and wire them to the PC board.

 

While I prefer relative mode for editing, it does have its thorns. When I starting playing with the band I am in now our soundman's obsession with subwoofers produced shock waves that were enough to knock my Andromeda into self-edit mode. Changing to pass-thru cured that. Other gear offers a panel lock mode for exactly that reason.

[Nathanael_I]

I prefer encoders for things with patch memory at a conceptual level- but with one hugely important observation... Encoders only work when a huge amount of design thought and expertise is applied to parameter scaling. A Moog filter cutoff knob is a thing of artistic usefulness. The endless encoder on the filter parameter of my Solaris does not bring the same joy or sensitivity - it sets parameters, but I map the mod wheel if I want real time control. Visual feedback is helpful (rings like Nord, Yamaha Montage, etc). So there's the rub. Architecturally, an encoder is a better choice. But practically - they often aren't as musically useful as a pot. Dave Smith gets this really right whatever he is specifying on the OB-6. Moog gets this.

 

Haptics - the final frontier for getting electronic instruments to be as expressive (without multiple lanes of MIDI controller data) as acoustic ones....

[mate stubb]

Pots.

 

In theory, I could love encoders with LED rings for current value, but in practice I have not laid hands on any that feel good.

[SamuelBLupowitz]

Yeah, I think this is a workflow thing more than anything. I don't know that I'd argue that I prefer the "feel" of any encoder of a nice smooth pot, but I don't know that that's important enough to me to overcome various practical concerns if encoders improve my workflow. When I switched over to a software-based synth setup, I got the Behringer X-Touch Mini as a knob/slider controller over the various Korg and Alesis options because I knew I'd be blazing through a lot of presets, but I would want to be able to tweak various paramaters from patch to patch, so I wanted something with knobs that would "move" with the patch changes and show me where each paramater was set with the LED ring. There are enough little things for me to keep track of, I like to not have to wonder "how much headroom do I actually have here? Where is the resonance going to jump when I move this?" Much the way I like to see the drawbar setting that's about to come out of the organ, I like to see the knob positions of my synth parameters.

 

But yeah, if we're dealing with all (or mostly) real-time control rather than a situation where I'm scrolling through preset patches every song (or within songs), I want the nicest, smoothest knobs I can get.

 

I don't think we've had a fight about patch memory in the last few days, have we? :wink:

[Al Coda]

Functionality wise, I don´t have a problem w/ both.

 

But I have a problem w/ todays pots quality.

The pots in my Minimoog D are great since about 40 yrs ... (which says it all)

 

A.C.

[AnotherScott]

So the logic solution is to use endless encoders, which always start from the stored value, then go in increment or decrement from there. I have learned this in the old times on the Matrix-12. Of course, you have to have the screen showing both the parameter and its value as it changes. And perhaps, also a visual reference associated with the knob, LED ring or whatever.

If you've got the LED rings then "Of course, you have to have the screen showing both the parameter and its value as it changes" is no longer necessarily true. My favorite Nord Lead 3 has dedicated encoders-with-rings for every relevant parameter. You don't have to look at the screen at all when building or editing a sound. It's the only way to have a board with patch memory AND the advantage of all knobs instantly and simultaneously displaying their correct positions.

 

For real time control, live performance etc. I think pots in relative mode is the way to go.

But what if you need to increase (or decrease) a parameter and its associated knob is alread all the way up (or down)?

 

When it comes to synths with patch memory, my feeling is that if you are the kind of player who tweaks parameters during performance, endless encoders are best (at a minimum, encoders for those parameters you are most likely to tweak in live performance), these controls are the only ones that really work, i.e. will always do what you want when you move them. Whether Jump, Catch, or Relative, any fixed pot will at some point not behave the way you want. But if you do not tweak during performance, meaning that the knobs are only there for you to do your "offline" pre-gig patch generation, then I see the virtue of pots, especially over encoders that do NOT have LED rings.

 

Related, I would not want to use a knob-per-function style synth with patch memory that lacked endless encoders UNLESS there was a mode that made the entire front panel "live" so that, with one button, the sound and the controls would all be instantly in sync. (I'm not sure off-hand whether any such boards lack this feature?)

[Markyboard]

 

For real time control, live performance etc. I think pots in relative mode is the way to go.

But what if you need to increase (or decrease) a parameter and its associated knob is already all the way up (or down)?

 

Good point Scott. "Pass Through" mode solves that but that "dead space" is not something I ever cared for. You have the same issues with pedals and other real time controllers as well. How many times have we read someone forgetting to reset their transpose button?

 

 

I suspect when controlling synths in real time at a live show it always require planning work-arounds for any of these type inconveniences.

 

I believe most of my synths have the Live Mode feature. I never use it. I'm almost always returning to the same place after power cycling my equipment so the knobs are often where they should be. Some of my synths power up from where I left off editing them last which is really nice. And if not or if I call up a new patch I just don't care about the jumps for what amounts to a very short period of time in my geek out sessions.

 

Life is rough for us basement studio rats.

[Doerfler]

Life is rough for us basement studio rats.

 

at least the commute isn't bad.

[Mike Warren]

There are nice feeling encoders available. One of my oscilloscopes has nice ones. They feel just like pots without end stops. No clicks.

 

My preference is encoders with LED rings, but I'd like to see twice as many LEDs as the ones I've seen.

 

Or motorised pots would do too.

[EricBarker]

Motorized pots are the holy grail. Then it would be a toss up between traditional pots and encoders with a really good LED collar, though as many have said, most companies don't get it right. Absolute dead last would be encoders without a collar. Unfortunately, that's what my Arturia board has, and I barely map anything to do them. Software synths just haven't developed a really good strategy of interfacing with controllers, IMO.