Analog 4 DCOs aliasing

White_Noise · 2023-05-26T07:20:09.627Z

LyingDalai:

You can also have aliasing with the encoders, the position of which are linked to a digital value. A filter sweep with high resonance for instance can show this on DSI synths, iirc.

That’s closer to quantization error than aliasing. Whereas aliasing is determined by sample rate, quantization error (or lack thereof) is determined by the bit depth of each sample. Steppy synth controls are caused by using too few bits to represent the control’s location, which translates as obvious stepping in the parameter when you adjust it. For instance, I pulled up the rev2 manual and that has 165 steps for filter cutoff, which is something like 8 bits of resolution.

If all the controls used something like 16 bits, you’d never notice the stepping, but it would take more processing power to make that happen (for a start all the presets would take up twice as much storage space) and for most controls would be overkill. MIDI 2.0 is going that way though, so synths may follow. I think 8 bit controls are the standard because MIDI uses 8 bits, but I could be wrong on that or otherwise oversimplifying.

pumpkin_head · 2023-05-26T08:09:06.076Z

Yes, but four means oven in french. Surely the analog four can get hot after a while.

mess5 · 2023-05-26T08:11:04.187Z

Have you heard A4 aliasing?

Purusha · 2023-05-26T09:16:25.868Z

Stepping on the filter perhaps, but surely not aliasing?

homealone · 2023-05-26T09:40:53.391Z

MIDI controls only use 7 bits, which is why the max value is 127. But they are still transmitted as 8 bit bytes, just with the highest bit always set to 0. Some controls use two bytes together to provide 14 bit values up to 16,383, or +/- 8192 in the case of pitch bend.

Analog 4 has digital controls which must get converted to CV internally at some point through a DAC, but from what I’ve seen the encoder resolution is much larger than MIDI, so ‘zipper noise’ as it used to be called, is minimised.

thermionic · 2023-05-26T18:37:40.443Z

Obviously not or we’d be talking about something beyond links to rants on Gearspace about how DCOs are “not analog” and arguments based on DCOs from 1972 or whatever.

@iono hopefully the rants aren’t taken in your direction, I don’t think it’s a problem to ask these questions or improve your understanding*, I just think some of the Synthbro-ier arguments on Gearspace can be force of will from people who want to feel superior to others over informed discussion of EE and synth design. Complex topics are fine, “controversial” topics are generally simpler and involve some idiot trying to project their egos into topics they don’t understand and are thus immune to learning and growing.

I highly suggest that when you note some topics as “controversial” you note that the experts’ opinions are more interesting and important than someone who argues against them for their own self-importance and need for internet validation.

Gearspace has gotten better since the name-change from Gearslutz lately if but for that decision seems to have passively rooted out some of the genuinely least intelligent and most reactive trolls (a bunch of them melted down and tried to incite some anti-mod revolution through sockpuppets, nobody followed!) but some people are just energy vampires and better ignored than assuming they have any reason to argue with the unearned self-confidence that may cause others to think they have any valid points beyond rhetoric.

Granted that insecure overconfidence “works” in a lot of aspects in life to bring persons influence (hi Twitter!) but I can only speak towards narcissistic contrarians in our communities

*Trolls can be an excellent passive excuse to learn and better understand things, if you’ve got the headspace and kind and generous engineers have the energy to patiently respond to the attacks they’ll inevitably receive from the stubborn and perma-ignorant. It wouldn’t be a “controversy” if the trolls ever stopped doubling down, adjusted the initial hot take, or acknowledged initial ignorance. Once one joker has their hot take, other contrarians (or sockpuppets) jump in to attack whatever “establishment” that exists counter to their reality.

It’s sort of a microcosm of how information is exchanged on this here internet.

Anyway! Have a nice weekend and enjoy your non-Aliasing A4

homealone · 2023-05-26T18:51:02.837Z

One note off topic: Gearspace can be an excellent resource for learning. Some of the very old threads are full of useful information and are noticeably absent of ego bullies. But as a community in which to actively participate it’s like willingly taking a job as an emergency callout sewer drainage engineer, for free…

Ok, back to the wholesome Analog 4 talk.

J3RK · 2023-05-26T19:04:44.435Z

The CEM3396 / PA397 (same IC basically) found in the Matrix 1000/6, Prophet Rev2, Prophet 08, etc. is simply clocked with a pulse. It’s still an analog integrator internally. This is why the Rev2 can have all of its oscillators basically free-running like VCOs. Each clock can be independent. In general, you’re not going to see a clock generator for these go anywhere near a frequency where there could be side effects. They’re probably not generated in a way where this could even be potentially an issue. They’re probably a standard on/off logic pulse.

There are actually other types of DCOs, but I’m not aware of any synths that employ them that would have artifacts from higher frequencies.

I guess it’s technically possible, but when you factor in the pitch-range limitations, and unliklihood of high frequency modulation in the synths that had them, this would be a non-issue in real-world use.

As to the original question, no the oscillators in the Analog Four will not alias. Period.

I’m actually wondering if anyone bringing this sort of thing up (on the other sites (not here)) might have been recording digitally, and getting aliasing or other digital artifacts from a recording issue, or compression, or maybe confusing a synth with digital oscillators for a synth that has DCOs.

Also, there are digitally controlled VCOs too. The VCO pitch control voltage is generated by a DAC.

thermionic · 2023-05-26T19:46:19.105Z

J3RK:

Also, there are digitally controlled VCOs too. The VCO pitch control is generated by a DAC.

Yeah, I guess the delimiter of what gives an opportunity for interesting knowledge over “controversy” is the open-ended question-

“What is analog?”

over… some boring prescriptivist answer-peddling based only on that someone cares too much about their own opinions.

homealone · 2023-05-26T19:53:11.277Z

At the end of the day it all has to come out of speakers so we can hear it, and a speaker cone can’t move between two points without travelling in between (unless it’s a quantum speaker .)

J3RK · 2023-05-26T20:01:35.861Z

Quantum speakers are always collapsing waveforms though…

sicijk · 2023-05-26T20:17:48.330Z

Schnork · 2023-05-26T20:49:14.253Z

J3RK:

I’m actually wondering if anyone bringing this sort of thing up (on the other sites (not here)) might have been recording digitally, and getting aliasing or other digital artifacts from a recording issue, or compression, or maybe confusing a synth with digital oscillators for a synth that has DCOs.

I’ve seen several posts in which users thought DCOs were digital. They were probably thinking VCO=analog, DCO=digital. I mean, I can see how someone would think that way…

Also aliasing has been a favourite topic of the the plugin youtuber niche for quite some time, sometimes portraying it as the sneaky enemy of jolly bedroom producers, so I guess some people might be a bit confused at this point.

lowph · 2023-05-26T22:04:49.913Z

I think aliasing is fascinating and now I wish it did happen on the Analog Four. Sometimes you find really cool artifacts which can be used creatively.

I remember when they added anti-aliasing to the Absynth soft synth, and then people complained their old patches didn’t sound good anymore. I wasn’t an Absynth user, but several people I knew mentioned it.

What are some synths that are notorious for aliasing? Clearly not DCO/VCO synths

pselodux · 2023-05-26T22:24:25.748Z

Your Mom:

I think aliasing is fascinating and now I wish it did happen on the Analog Four. Sometimes you find really cool artifacts which can be used creatively.

This is why I keep my Nord 2X. I thought it wouldn’t be as gritty sounding as the original Nord Lead but the artifacts really start to emerge when it’s pushed.

J3RK · 2023-05-26T22:31:17.345Z

Agree completely!

I don’t want it in analog hardware, or digital where it’s not intended, but I absolutely love it in the right context.

Amiga MOD music
Vintage wavetable synths, (Waveform City, Miniwave, Wave, Microwave)
Old OPL hardware
Chip tune oriented synths
Vintage samplers like the Mirage

Filters like to chew on that grit too. Good stuff!

felixtv · 2023-05-26T22:31:40.032Z

My DSI Mopho X4 is having 12 or even 16 bit control on filter only and it’s a good workaround. This crucial control isn’t feels steepy.

J3RK · 2023-05-26T22:33:03.977Z

12 bit or above is generally great for controls, especially if they have smoothing algorithms or circuits.

warpigs330 · 2023-05-27T02:20:47.987Z

The model cycles aliases nicely.

obscurerobot · 2023-05-27T03:06:33.239Z

natehorn:

Disclaimer: I am not a certified synthologist

I’m a Discordian Pope and hereby ordain you a Certified Synthologist!

Your Mom:

Cuatro Analógico

Death Note in Japanese, probably.

thermionic:

“What is analog?”

Technically everything digital is also analog. I don’t even like music or synths, I’m only here for the amazing pedantry opportunities.