- ...I asked some infos to an ex-QM engineer; he proved to be very kind and helpful, so I thought it could be interesting to share with you part of his words...
"QM used to work with DSPs from a french company called "Dream". These DSPs had different fixed algorithms for sound synthesis already in there. We had around 7 different types of algorithms. Sample playback with 12 dB filtering was one of the algorithms. But we also had FM algorithms. Some of the FM algorithms had also a feedback control, others not. For most of the organs we used a simple FM algorithm with parallel operators and sine signals. This algorithm had more operators than the algorithms with feedback. The filter
parameter (...) is mapped to the feedback parameter in FM sounds.
So - at some sounds you hear nothing when you control the parameter, at some you only hear what happens when one rises up the feedback of an algorithm and on sample based sounds the control is assigned to the cutoff frequency. The engine was really terrible. (...) most of the time you just tried hard to get sounds out of it, that are not distorting or aliasing too much. For that reason we made the decision to allow users only access to some offset values instead of hundreds of parameters. Trial an error with all this parameters can easily create results, that are not predictably and mostly noisy and frustrating. Especially the FM algorithms had a limited dynamik range. Just a little bit higher on one value can cause absolute random noise instead of something useful. (...)
There are more FM sounds inside than you maybe would expect. Most basses for example are FM sounds too."
I also asked if it was true that RavenMax had 112 mb of samples inside, because you can read in many websites that the Raven Max has 14 mb of samples (while the Max manual claims 122 mb); he said that the information was meant as bits and not Bytes, so it was just a marketing trick.
"It was a hard job to get so many different samples into this limited memory. (...) Some special FX samples (...) are 8 Bit, 8 KHz sample rate..."
--- In Quasimidi-The-Raven@yahoogroups.com, "somethingkillingyou" <somethingkillingyou@...> wrote:
> so, the M.A.S.S. (Multi Algorithm Sound Synthesis) engine does Subtractive (PCM samples involved) but also Additive and FM synthesis...
> but is there a way to know which sounds of the Raven are using FM and additive synthesis?
> we know that the filter does not work in the additive and fm sounds, and that in the FM sounds in particular the 'tone control' change the operator feedbacks...
> now I imagine that the only way to tell is to try each sound?!
> but there are 1000 sounds in the Raven Max!!!
> "FM Percussion" group sounds are fm sounds for sure, because using wheel 2 you can hear the sound change but it's not cutoff, in fact it is the operator feedback that changes...
> I think the sounds with FM in their name are FM too indeed (like AcbassFM, Jazzy_FM etc.)...
> but what about additive sounds?
> and are there other FM sounds that doesn't have the FM letters inside their names?!
> anyone knows more than that!?
- Very interesting!
So Raven has FM algorithms inside it. Does it have virtual analog algorithms (sine, saw, square oscilators etc.)?
I am trying to reproduce Klaus preset on other synth. One part of it is "SuperSol". I am wondering how is it made. Is it sample from ROM or VA algorithm? And how to reproduce it on other VA synth? May be somebody tried...
--- In Quasimidi-The-Raven@yahoogroups.com, "nikitamrzlyakov" <nikita@...> wrote:
> Very interesting!
> So Raven has FM algorithms inside it. Does it have virtual analog algorithms (sine, saw, square oscilators etc.)?
> I am trying to reproduce Klaus preset on other synth. One part of it is "SuperSol". I am wondering how is it made. Is it sample from ROM or VA algorithm? And how to reproduce it on other VA synth? May be somebody tried...