EssenceFM Filter
pandar1 | Hi, I’ve been experimenting with the filter, and I’ve noticed it seems to be one of the areas that could benefit from some improvement. I’ve come across discussions mentioning possible hardware limitations. Could you share more details about the filter? Was it based on an existing design or developed from scratch? It would also be great to get some technical specifics, such as the slope, minimum/maximum cutoff frequencies, and how the low-pass and high-pass filters interact. At 50%, does the dry signal split, with part of it going to the high-pass and the other to the low-pass, and then recombine in parallel? Additionally, does the filter cutoff frequency track the played note, or would I need to set up a modulation slot for that? It is correct that the first point in the envelope sets the initial cutoff frequency, while the rest of the envelope shapes how it evolves over time? Is the filter fully digital and implemented via DSP? Are the limitations due to computational demands? Would it be feasible to introduce a more advanced 24 dB ladder filter, even if it meant a reduction in the number of voices? For example, could there be an option to choose between 300 voices with the standard filter or 200 voices with a higher-quality filter? I understand the filter may not be the main focus of an FM synth, but having a high-quality filter would certainly be a great addition. Lastly, I think it would be helpful if the manual provided more technical details about the filter, similar to the ones I’ve asked about. This could give users a better understanding of its capabilities. Thanks in advance for your time and support. |
KODAMO | Hi, The filter in the EssenceFM may appear to behave weirdly because it's a 6dB/oct filter, which is rather unusual for a synth. It's primarily meant for sound shaping purposes (eg. if you notice your sound needs a bit less highs, more mediums or a bass boost, it's easier to use the filter than to mess with operators' volumes). It can still provide high resonance and create interesting sounds on its own but it will not behave like most synthesizers which use the filter as the main voicing tool. Changing it was discussed before but would break a lot of existing sounds, so we advised against. It may still be possible to add a switch somewhere, although it may distract users from the FM synthesis itself which is the whole point of the EssenceFM. <blockquote>It would also be great to get some technical specifics, such as the slope, minimum/maximum cutoff frequencies, and how the low-pass and high-pass filters interact. At 50%, does the dry signal split, with part of it going to the high-pass and the other to the low-pass, and then recombine in parallel?</blockquote> Dry/Wet is like having a mixer before the filter, one end sends to the output, the other to the filter (which then sends to the output) Inside the filter, the LP/HP setting allows to mix different levels of low-pass or high-pass (like a continuous variable filter). So at 50%, you have 50% of the LP, 50% of the HP sound. |
| pandar1 | May I ask why the choice of a 6 db/oct instead of a more conventional 12/24 db? I think you could add a 6/12/24 db switch in the filter page defaulting to 6 db for existing patches, so no disrupting of old patches would occur... And what about the keyboard tracking? |
| KODAMO | Because for its initial purpose (sound shaping) most of the time it does the job. A better filter like the one in the Mask1 is more DSP-intensive yes. Maybe the EssenceFM has enough resources to emulate a filter with a steeper slope, we have to try. Keyboard tracking can be done using the mod matrix : note number => cutoff (or cutoff env if you want to preserve the envelope) |
Jazz Paladin | Interesting. I have always wondered why the filter seemed to be less extreme than what I am used to hearing. This explains ! |
| pandar1 | Yes, I think the implementation of a more aggressive filter would open up a world of possibilities... |
Reply
