You've watched a dozen demo videos of the Behringer Pro-800 and the Arturia MiniFreak, and you still can't tell which one you'd actually keep. That's not you failing to listen carefully. It's because most demos run different patches, different gain levels, and different reverb tails, which makes any real comparison impossible.
Specs won't save you either. Voice count, filter type, and price tell you what a synth has, not what it does to your ears. A Waldorf Protein and a Dreadbox Nymphes might both list "eight-voice" or "six-voice" polyphony, but one uses emulated CEM filters on a wavetable engine and the other runs true analog VCOs with a Juno-style architecture (gearnews.com). Those are fundamentally different sound sources, and no spec sheet captures how they behave under a filter sweep or a stacked unison patch.
The fix is a level-matched A/B test you run yourself, in your own DAW, with your own ears. Here's how to build one.
Step 1: Pick Your Two Test Subjects
You don't need to own both synths to run this test. If you're shopping and only have access to plugin versions or demo units at a shop, that's fine, the protocol still works.
A useful real-world pairing right now: the Behringer Pro-800, an analog six-voice synth modeled on the Sequential Prophet-600 architecture with dual VCOs and a 24dB low-pass filter per voice, against the Arturia MiniFreak, a six-voice hybrid with digital multi-engine oscillators and an analog filter per voice (gearnews.com). Or go fully digital-vs-analog with the Dreadbox Nymphes, which uses single VCOs with continuously variable waveforms and a 24dB low-pass plus 6dB high-pass filter per voice, against the ASM Hydrasynth Explorer, a wavemorphing digital engine with poly-aftertouch (gearnews.com).
If you're using software stand-ins, pick one plugin modeled on true analog circuitry and one built on digital synthesis (FM, wavetable, or virtual analog). The goal is contrast in sound engine, not brand loyalty.
Step 2: Strip the Patch Down to the Sound Engine
This is the step most bedroom comparisons skip, and it's why they mislead. A factory preset with chorus, reverb, and a filter envelope sweep tells you about the preset designer's taste, not the synth's raw voice.
On both instruments, set up the following:
- One oscillator only. Disable OSC 2 or any sub-oscillator layering.
- Sawtooth waveform, since it's harmonically rich and exposes filter behavior clearly.
- Filter cutoff open enough to pass most harmonics, resonance at zero.
- A simple ADSR: instant attack, no decay stage, sustain at 100%, release short enough to avoid overlap.
- All effects off. No chorus, delay, reverb, or drive.
- No LFO modulation routed to pitch, filter, or amplitude.
Play a single sustained note, same MIDI velocity, same key. You're now hearing the oscillator and filter stage in isolation, which is exactly what you want.
Step 3: Level-Match Within 0.5dB
Your ears will lie to you if one synth is even 2dB louder. Louder almost always sounds "better" or "fuller" in a blind test, regardless of actual tonal quality. This pairs well with headless guitar tuning stability: a daw test method.
Route both synths into separate channels in your DAW. Record a 4-8 second sustained note from each at the settings from Step 2. Use your DAW's loudness or peak meter, ideally an RMS or LUFS-style meter if your DAW includes one, to measure the average level of each clip.
Adjust the output gain (not the synth's own volume knob, since that can change tone on some analog gear) on whichever track reads louder until both clips sit within 0.5dB of each other. Bounce both to audio so the level match is locked in and won't drift if you touch a fader later.
Step 4: Build Your Listening Checklist
Now you're ready to actually compare. Play both level-matched clips back to back, ideally through the same monitors or headphones, at the same time of day when your ears aren't fatigued. Listen for four specific traits:
Attack transient. Does the note snap in instantly, or is there a soft, almost rounded onset? Analog VCOs often have a slightly less uniform attack because the circuit needs a few milliseconds to stabilize, while digital oscillators start at a mathematically precise sample value every time.
Filter resonance behavior. Slowly raise resonance on both synths while holding the note. A true analog low-pass filter, like the 24dB design in the Pro-800, tends to self-oscillate and add a musical, slightly unstable whine as resonance climbs. Digital filter modeling can sound cleaner or thinner at high resonance because it's calculating an approximation rather than driving an actual feedback loop.
Voice-to-voice pitch drift. Play a five-note chord and hold it for 15-20 seconds. On an analog polysynth, listen for subtle pitch wandering between voices as the circuitry warms up, this is thermal drift, and it's part of what gives analog chords their organic, "breathing" quality. Digital oscillators lock to sample-accurate pitch and won't drift at all, which some listeners hear as sterile and others hear as reliably in tune.
Unison width. Switch both synths to unison mode if available (the Pro-800 offers eight-voice unison) and play a single note. Analog unison gets its thickness from genuine oscillator-to-oscillator detuning and drift. Digital unison often relies on phase or pitch offset algorithms that can sound wider but more mechanical, without the same slow, organic movement.
What to Listen For, and Why It's There
The differences you're hearing come down to how each synth generates sound. Discrete analog VCOs are voltage-controlled circuits, and voltage is inherently a little unstable, subject to temperature, component tolerances, and power supply fluctuation. That instability is thermal drift, and it's the reason analog chords rarely sound perfectly static. For more on this, see full coverage of sidechain noise gate settings for guitar: step-by-step.
Digital oscillators calculate a waveform mathematically, sample by sample, so they don't drift unless a developer deliberately programs in emulated drift or noise.
Filters split along similar lines. A synth like the Nymphes uses a genuine analog 24dB low-pass and 6dB high-pass filter, where resonance is a real feedback path that can self-oscillate. A digital filter, even a well-modeled one on something like the MiniFreak's analog-hybrid design, is either running true analog circuitry per voice or approximating that behavior in code. The MiniFreak actually splits the difference, pairing digital oscillators with an analog filter per voice, which is worth listening for specifically since it isolates the oscillator-vs-filter contribution to tone (gearnews.com).
Common Mistakes That Wreck the Test
The biggest error is comparing presets instead of raw patches. A factory patch with built-in chorus and a slow filter sweep will always sound "more interesting" regardless of the underlying synthesis, so you're rating the preset designer, not the synth.
The second is skipping level-matching entirely. Even a small volume gap biases your perception toward the louder signal, every time, regardless of your experience level.
The third is testing tired or in a noisy room. Ear fatigue flattens your ability to hear subtle transient and resonance differences within 20-30 minutes of critical listening, so keep test sessions short and take breaks.
Finally, watch out for confirmation bias if you already know which synth costs more or which brand you prefer. Consider recording both clips, renaming the files something neutral like "A" and "B," and listening blind before you check which is which.
Your Comparison Worksheet
Use this simple template for each pair you test:
Synth A: ________ Synth B: ________
Also read: see adobe audition won't open on windows 10? fix it fast
Level-matched (within 0.5dB)? Y / N
Attack transient: A feels ________ / B feels ________
Filter resonance at 75%: A does ________ / B does ________
Pitch drift over 15 seconds (chord): A ________ / B ________
Unison width and character: A ________ / B ________
Overall preference for my style of music: ________
Run this test on two or three synth pairs and you'll start noticing your own pattern, whether you consistently prefer the unstable warmth of analog drift or the precision of digital oscillators. That pattern is more useful than any spec sheet, and it's the thing that actually predicts whether a new polysynth earns a permanent spot on your desk.
