Ich hoffe, dass noch der Vintage Mode des P5/P6/OB6 für den REV2 kommt.
Wenn jemand hier im DSI Forum aktiv sein sollte, wäre es lieb, wenn derjenige mal nachfragen könnte
ich meine dort gelesen zu haben, dass für ein grösseres Firmware Update kein Platz mehr ist
Es steht
da aber auch das:
Seq devs have been mum on naming the specifics of the vintage knob implementation (trade secret and all) - but yeah, it's using the same general concept... per voice variance to tuning, envelopes, filters and amp sections... that's what Voice Component Modeling does as well... and that's what I measured when I recorded all the vintage synths. The way I go about it, and describe the implementation of VCM is through curated tables of stable offsets that target osc frequency, envelope ADR stage timing, filter and amp parameters on a per voice basis. With VCM on the Rev2 you can get extremely granular in the voice modeling design - setting the amount of virtual voices (4,5,6,8,12,16), and specifically defining the offsets per voice, as well as targeting any parameter in the mod matrix, or even macro type behaviors, like per voice frequency glitches, transient pitch settle effects, or other quirks. Once you build the VCM lookup table (via gated sequencer), the offsets are the same per voice, per component, every time you return to that voice, in round robin. The VCM method gives you repeatable voice behavior from session to session.
Based on discussions on GearSlutz, my interpretation is that the end result of Vintage Knob is similar (per voice variance to the same parameters), however the Vintage Knob implementation is probably not doing it with a curated static lookup table. From the info that was revealed, it sounds like it may be more "randomized" in the generation of offsets, either through a random seed type of algorithm, or a more refined slop motion type of implementation, which targets more than just frequency... adding envelope timing, filter and amp parameters.
From the Seq P5 marketing:
"We found that much of this desirable character was due to slight fluctuations and differences in the response times and frequencies of the individual VCOs, filters, envelopes, and amplifiers from voice to voice. To give the same mojo to the new Prophet-5, we created the Vintage knob. Use it to dial in as much old-school randomness as you’d like."
I would love to get more clarification on the exact method they're using
I'm guessing that with some test equipment / oscilloscope it can probably be mostly figured out, but I don't currently have a P5/P6/OB6.
A couple years ago I documented findings of the slop algorithm in the "appendix thread"... it was a bit tedious, but was actually a catalyst for me starting to do more research on vintage poly synths, and led to the articles/videos on VCO phase jitter and voice modeling.
There's a bunch more info about voice modeling on the
http://www.VoiceComponentModeling.com website, if you wanna check it out. Also, I've uploaded several more articles about my research here:
und kostet die hälfe mit 60% mehr voices. filter klingt identisch, oszis sehr ähnlich. dsi eben.
