According to some comments on the KVR forum, the 808 I painstakingly
modelled straight from Roland's schematics, with some help from
e-licktronic's Yocto 808 clone on github, is now one of the closest
TR-808 plugins on the market.
I hope those comments are right, because I never got to test a real
TR-808. I used every sample pack I could get my hands on as reference.
Not sure I'm doing things differently than the other companies that
made 808 plugins, but it works. Whenever a sound came out wrong I went
back to the circuit, and every single time my math was wrong, not the
drawing. It paid off. Nothing was fitted. No matching EQ on anything,
ever. What comes out of the plugin is a consequence of getting the
circuit right. Accents are a higher voltage in the model, the way the
real trigger bus works, so they should behave like the real unit.
But don't take my word for it. Install it, make a beat, play it loud.
Listen to how it comes alive — the little differences from hit to hit.
That is not a humanize knob. I actually took a random generator OUT of
this thing. What's left varies because the noise circuit is six
oscillators that never stop running and don't get reset when you hit
the pad, exactly like the hardware. About a third of a dB, hit to hit.
Samples are stale and always the same. This isn't.
Every step can also be parameter locked to its own pitch. And the pitch
is not a pitch shift — it swaps the capacitor in the circuit, the same
as retuning the real machine. You can hear the difference: pitch a
sample up and it gets shorter and thinner. Tune this kick up and it
gets LOUDER, because the click's ring scales with the cap and the
retrigger doesn't. Nobody would program that on purpose. It's there
because the circuit does it.
That's the reason I went down this rabbit hole in the first place.
If I'm wrong about any of it, I'd genuinely like to hear about it.
Thanks for your Time,
Paul , BeatForge
modelled straight from Roland's schematics, with some help from
e-licktronic's Yocto 808 clone on github, is now one of the closest
TR-808 plugins on the market.
I hope those comments are right, because I never got to test a real
TR-808. I used every sample pack I could get my hands on as reference.
Not sure I'm doing things differently than the other companies that
made 808 plugins, but it works. Whenever a sound came out wrong I went
back to the circuit, and every single time my math was wrong, not the
drawing. It paid off. Nothing was fitted. No matching EQ on anything,
ever. What comes out of the plugin is a consequence of getting the
circuit right. Accents are a higher voltage in the model, the way the
real trigger bus works, so they should behave like the real unit.
But don't take my word for it. Install it, make a beat, play it loud.
Listen to how it comes alive — the little differences from hit to hit.
That is not a humanize knob. I actually took a random generator OUT of
this thing. What's left varies because the noise circuit is six
oscillators that never stop running and don't get reset when you hit
the pad, exactly like the hardware. About a third of a dB, hit to hit.
Samples are stale and always the same. This isn't.
Every step can also be parameter locked to its own pitch. And the pitch
is not a pitch shift — it swaps the capacitor in the circuit, the same
as retuning the real machine. You can hear the difference: pitch a
sample up and it gets shorter and thinner. Tune this kick up and it
gets LOUDER, because the click's ring scales with the cap and the
retrigger doesn't. Nobody would program that on purpose. It's there
because the circuit does it.
That's the reason I went down this rabbit hole in the first place.
If I'm wrong about any of it, I'd genuinely like to hear about it.
Thanks for your Time,
Paul , BeatForge