John Bowen Synth Design

All processing inside Solaris is digital (and very high quality, running at 32 bit, 96kHz). The software code implements highly accurate models of the components of some (highly desireable) vintage analog synthesizers.

Yes, the firmware (i.e., software stored inside the synthesizer's memory that enables its operation) can be updated via the CF card. This would potentially give your Solaris synthesizer expanded, or even entirely new, capabilities.

The Solaris sequencer is a step sequencer that outputs 4 rows of control values. The Overall rhythmic timing is established by Sequence A; the other 3 rows can have different lengths, but they will still use the rhythmic pattern set by Seq A. Even though you can set up 4 independent 'channels' per preset in Solaris for articulation, the overall Amplitude envelope controls all final output (i.e., you can set each signal path's VCA to be controlled by separate envelopes, but those ADSR time values have to 'fit' within an overall Amp EG setting). So, my gut feeling is that it is not polyphonic in the sense you may be wondering, but the real point of the sequencer is that it can produce some fairly complex sounds. You can play polyphonically, and all voices played will be controlled by the sequencer, with the root pitch determined by the notes you are playing.

Sample files are loaded from Compact Flash cards in this method for the 1.0 release:
1) Sample files need to be converted to the .raw format. There are quite a number of programs that can do this.
2) The samples need to be described in a text file. The name, number, sample start and end, loop start and end, root key, fine tune values, low and high keys need to be listed in this text file (see this forum post for example). This text file is what makes up a 'sample pool', and you can have several on the CF card.
3) The samples and text files are loaded via computer onto your CF card.
4) The Sample Pool is selected from the System page and loaded into the Solaris for use in any of the four oscillators.

The CF card does not need to be a certain brand nor high speed. It uses the FAT file format, which is how most CF cards are formatted.

Expect 10 voices with the system going full - all 4 filters, oscs, etc. (note that for the first OS release there won't be DSP offloading so most likely no matter the configuration there will still be 10 voices.)

The Solaris v1.0 Ss will not have a Multitimbral Mode. For Keysplits, you can use one of the four Key Tables to define any ranges across the keyboard as you wish - even multiple "split points". You can then apply this key table output to the VCA of any channel to create a "split".

All knobs are rotary encoders except the Master Volume.

The final decision is to not use any detents for the Solaris encoders.

Currently the algorithm for the knobs takes into account the range for the current parameter. So for filter cutoff, where the range is 12 octaves, it takes 1 and 1/2 turns to reach the whole range, sweeping by semitone, whereas with Resonance, you have 0-100%, which happens in 1 turn, moving by 1% increments. Holding down the Shift key inverts the current control algorithm per knob - changing the cutoff sweep moves by 1/10 of a semitone, and Resonance moves by 5% increments. We will continue to tweak this until I think it's at a an optimum point for most parameters.

Yes, definitely! This feature is based on a voice assignment level and the detune is set upon the key-on event. This means that if you change the Detune value while holding down a key, you will not hear any change until you select a new note. The Unison stack can be 2 to 8 voices, All, or Chord. If play is set to Polyphonic, the stack of voices will 'divide' to allow polyphonic stacking as much as is possible, even if All voices are assigned. Chord Mode allows you to use the Unison stack to play a chord in parallel, based on the notes you hold down before pressing the Unison button. In this case, Detune is not activated.