Xhun Audio RawPrime : an original synthesis concept, established on... Aliasing (on purpose) !

[Jopmanajop]

I've loved LittleOne since it was released

Same here, love the sound the most, no other synths sound that huge and strong, but browser bug and sync bug are such a dealbreakers. Feels like everyone recording it live or with the smallest buffer, since no one else complaining about syncing

[Jopmanajop]

but didn't upgrade last time because it sounded like the preset browsing/saving bug still wasn't fixed

BTW, just upgraded several days ago in hope for the future fixes, also looking at the Whole preset pack bundle, but with that browser bug...

[wasi]

I'm a huge fan of Mr. Alias (and a pro owner) so I'm naturally excited about this.

[xhunaudio]

Ok, let's try to move the first steps in order to define what this Device actually is...

As well as for the Yamaha OP- / YM- chips, in this Device all starts with a "wavetable" stored in memory.

Note : I said "wavetable", not "multi-dimensional wavetables". It means a single waveform stored in memory. Remember, we have to stay cheap as in 1976 !

Or better, YM- had a 8-bit "quarter of sine wave period" stored, and it reconstruct a sine wave period by recombining it 4 times, resulting in a 9-bit, full period wave.

But the similarities with YM- stop here. Yamaha uses 2 of that sine wavetables in phasemod (or additive) in order to generate partials.

Instead the Device uses two (7-bit, full period) sine wavetables (separately) in a different, original fashion...

Let's start from the... top. What's possible to achieve with this new synthesis approach ?

Here are two preliminary examples of what this Device is able to produce.

The most important thing to consider here is :

- the first audio demo uses only 2 sine waves to produce the entire bass line.

- the second audio demo instead uses only 1 sine wave to produce each voice.

Both audio demos use a delay processor (included). Audio demo A makes use of the analog filter (included).

Audio demo A : Using only 2 sines per voice (#1 voice in total, monophonic)
https://soundcloud.com/xhunaudiorawprim ... -mega-bass

Audio demo B : Using only 1 sine per voice (#2 voices in total, polyphonic)
https://soundcloud.com/xhunaudiorawprim ... -it-simple

[xhunaudio]

FM/Phasemod synthesis is *NOT* involved in any way with this unreleased Device/synthesis method.

Here an example of a clavinet-like patch. Four voices polyphonic, each voice produced using only two sine waves.

https://soundcloud.com/xhunaudiorawprim ... -clavprime

[stoopicus]

Your approach sounds a lot like the Ensoniq ESQ-1. As do your audio samples

[xhunaudio]

Ahah, it's the same thing I noticed yesterday, while making some presets !

I'm just talking about the *flavor* of certain presets, reminding early samplers... not its architecture / synthesis method.

Indeed the Device does *NOT* use samples or wavetables at all. Just one single-cycle wavetable of a 7-bit sine wave. Plus... some magic stuff I'll explain later.

Not time yet for the full details - usually I make a post here every time I write a section for the manual, so I can keep track of everything and I don't forget anything.

[DrGonzo]

Example B grabbed my attention. Very nice!

[El°HYM]

So will this have some good Oversampling?

[xhunaudio]

So will this have some good Oversampling?

Well, this is a hybrid D/A design Device.

Yes, oversampling is applied to some specific (mostly analogue) components but... "oversampling" is not the key point here, definitely. This could lead to a misconception about what the Device is.

This Device is *NOT* a simple synthesizer producing aliased tones/sounds.

This Device 100% produces sound *exclusively* by using aliasing tones/partials.

What you hear in the Example A, Example B, Example C audio demos is not a wavetable with its aliasing, or a multisample with its aliasing, or a generated waveform with aliasing, etc.

What you hear is produced *entirely* by "pure" aliasing only.

[El°HYM]

This was actually meant as a joke. But will we have control over Jitter?

[xhunaudio]

Ah, ok - sorry for not understanding

Some jitter is there as a consequence of how it's made, but there will be no End-User control over it (= no jitter amount knob).

It is built as a late 1970s device - so in that case "jitter" is a bad thing you probably don't want to mention either. It's there but you don't want End-Users have any control on it.

[El°HYM]

I am not so sure about this but can also understand it, in this case.

Looking forward to try it out and in hearing some more examples.

[xhunaudio]

I am not so sure about this but can also understand it, in this case.

Yes, things like "changing the jitter amount" go beyond the context of this synthesizer and do not represent a vantage in this case (also in terms of the sound produced).

I want to stay focused on the device context. The User Interface (and its parameters) will mimic a hardware synthesizer : a clear arrangement, all in plain seight.

It has been a little bit difficoult to accomplish that, since there will be no less than 80 knobs on the panel ! But now all has been placed correctly and the Device has the very compact look and feel I was looking for.

Its look reminds an Apple II or Amiga 500 or Atari ST in terms of "compactness".

[El°HYM]