One Synth Challenge #109: Fathom Synth by Seaweed Audio (mmGhost wins!)

[løkken]

Thanks for the update Everett! Can you please clarify one thing? I found out wavetable oscillators ignore the "Raw Samples/Sin Partials" toggle. Is this intentional? I would like to use the "Raw Samples" mode for my experiments with granular synthesis

[bftucker]

BlitBit, Thanks, Brilliant Work!

Fathom Version 2.9.4 is now released!

Amazing. Your hard work and many updates, your responsiveness to user problems, and often surprising additions are a delight.

"Thanks. Brilliant work!"

[FathomSynth]

bftucker, wow you are still here! I had not seen you in the thread for a while, thanks.

lokken, Hi, yes that is for sure a bug. It should work for both the wave table and wave draw waves equally.
I'm make sure that gets fixed.

[574X]

This is a seemingly extremely well done mono synth.

If anybody wants to play chords you can use Tale's midi router for ReaJS.

https://www.taletn.com/reaper/mono_synth/

Just route the midi from the midi router's track through one midi channel to each instance of any mono synth.

Here's a demonstration:

You can also make chords using the detune. For example use 3 voices and set values to three different semitones.

Agreed, I love the versatility of the detuner module in Fathom. To clarify, in a mono synth with similar functionality, that method provides static intervals based upon input of a single midi note (true, you can automate the intervals), whereas with midi routing to multiple instances, you can literally "play chords" in the traditional sense, as if you were playing a polyphonic synthesizer or keyboard.

However, my pad was sounding a little thin because I was using one instance of free fathom and the detuner module method to save my vintage processor. I'm interested in how quickly the polyphonic version could thicken the atmosphere, and how much more optimized it is compared to the multi-instance approach.

I've loved making some great drums with Fathom, and I use one instance for each element for routing purposes, but it would be really nice to hear that mind-opening, single-instance stock Fathom drumkit in all its polyphonic glory, and with some additional percussion in the patch.

A couple massive boosts to workflow efficiency and potentially cpu usage optimization that I'd love to figure out are whether or not Fathom has multi-out routing for keyboard splitting, or the ability to duplicate modules. Does anyone have any info about that?

[Carl_saved]

A couple massive boosts to workflow efficiency and potentially cpu usage optimization that I'd love to figure out are whether or not Fathom has multi-out routing for keyboard splitting, or the ability to duplicate modules. Does anyone have any info about that?

I don't think Fathom has more than a stereo audio output, if that is what you mean. You could pan one sound to the left and one to the right and then split the two channels by means of your DAW, resulting in two independent mono sounds from one Fathom instance and split them by a Keyboard tracker inside Fathom
Duplicating modules can be done by opening the section "copy" under "Audio Components" on the left hand side. All the existing modules are listed there and can be copied by dragging them into the module matrix. However, this feature could break in version 2.7.x of Fathom, but should work reliably in versions 2.8.x and later.

[functionform]

This puppy is hungry. One patch I worked on with one WT osc and two filters through a delay was up to 80% cpu during patch design! This is on 2.9.4 too. When it was simply playing back it comes back to 4-6.

One really nice thing about this OSC is that having such an easy ability to create almost anything is forcing me to learn stronger approaches to sound design and mixing. The really bad thing about this OSC is that having such an easy ability to create almost anything is forcing me to learn stronger approaches to sound design and mixing.

[574X]

Duplicating modules can be done by opening the section "copy" under "Audio Components" on the left hand side.

Thanks Carl, it's amazing how I can miss things in plain sight sometimes... At the OSC I find myself reading more manuals than anywhere else, but this was a rare exception where I just jumped into making tracks. It just serves as a reminder that a little study along the way can work wonders for productivity.

This puppy is hungry. One patch I worked on with one WT osc and two filters through a delay was up to 80% cpu during patch design! This is on 2.9.4 too. When it was simply playing back it comes back to 4-6.

You might find some sections at the end of the quick start guide to be of help. There are a couple sections about optimization in DAWs and about what's using lots of processing. I did notice the detuning being surprisingly hungry.

[doctorbob]

I believe that closing the GUI will save cpu cycles as well when playing back. Even with my beast of a machine, sometimes there was a significant lag between moving a knob and hearing the result. Patience with Fathom is a virtue! Sounds like a good name for a track!

I agree that this synth makes you think more about sound design, and also the mixing of those sounds.

e.g. I have a crash patch which (to my ears and for my purposes) sounds not bad at all. BUT, once deployed into the mix is awful, and a complete flopperoo. I have 15 days to fix this - but it's really really really annoying, coz I know it's not wonderful (and therefore all you lot out there will think the same - bah!). I will probably have to do some EQ ducking -at least I can do this with reaEQ with parameter modulation.

So there you go, being forced to use yet another mixing technique ...

dB

[FathomSynth]

Everyone, I broke the ADSR glide in the latest release 294.

I will be fixing it and releasing a new version immediately.

[574X]

e.g. I have a crash patch which (to my ears and for my purposes) sounds not bad at all. BUT, once deployed into the mix is awful, and a complete flopperoo.

I've been experiencing this phenomenon surprisingly often, lately. Some things sound shockingly different in isolation than when in a mix. Luckily it hasn't been a concern, especially with my attitude of just going with the flow when it comes to synthesis. I'd imagine that if I really wanted to preserve the signal, the ways to make room would be the standard methods which you're probably already trying. - Lowering the levels of conflicting elements, filtering them out if needed, panning similar frequency to opposing sides, compressing the dynamics of the foremost element, and depending on the mono comparability requirements of it, adding a stereo widening effect to make it stand out.

[doctorbob]

I'd imagine that if I really wanted to preserve the signal, the ways to make room would be the standard methods which you're probably already trying. - Lowering the levels of conflicting elements, filtering them out if needed, panning similar frequency to opposing sides, compressing the dynamics of the foremost element, and depending on the mono comparability requirements of it, adding a stereo widening effect to make it stand out.

Yup been there and done most of that!!! So, it's careful EQ ducking of the conflicting elements - but, with maybe 24 instruments banging on with the crash, it's gonna take some time to sort out. The odd thing is that it REALLY does not sound the same at all - just a pathetic "floosh", solo it, and bingo nice "crash" ... so, if I can sort this out I will have learnt something about mixing - thanks to the OSC, again.

Starting point is to bring in the instruments one a t a time and note which ones start to affect the crash sound, and these can be EQ ducked, maybe with some volume automation to force the ducking as well.

Cheers,

dB

[Scrubbing Monkeys]

Well,....I think I I am done.......except........I cant seem to get my final level right.
By right I mean peaks tickling cloee to zero and rms some where around -6db. max. preferably hanging around -12db.

This is my first mix in Mulab. I now understand the need 3rd party meter plugins. I was completely lost trying to set up my normal -18 db per channel.
I finally went back and inserted TB audio free meter on every channel and ended up adjusting everything quite a bit. ( pain in the #$@!!). Probably my ingnorance but there has to be a way to make gain automation relative.

Anyway, I am using Limiter no 6.

All is good, all sounds ok. I put TBaudio meter last on the msster bus.
Peaks hold below zero. needle still twitching........on the peak selection. All of the rms selections are pegged.

I exported the file and brought it into Reaper where I know the meters.
I have never had RMS levels that high. I didnt know you could clip rms without clipping peak.

If I get the RMS down using the limiter I am afraid it will be weak.

Sound wise it is not bad but watching the red on the Rms is freaky

Any advise?

[] Peter:H []

The Kybernetic Gods Preview (NOT A SUBMISSIOM YET)
https://soundcloud.com/p-e-t-e-r-h/the- ... ds/s-hjeZN
All wavetables handcrafted. Many rhythmic elements are just clever wavetables.
This song has not been mixed at all by now, apart from 2 Eq plugins. There's no extra reverb, only 1 fader touched by now...
Kybernetic machine music

---

An earlier Fathom Draft
https://soundcloud.com/p-e-t-e-r-h/fath ... 01/s-Fc4Jt

[functionform]

Two questions:

1. Does anyone know where you can change the pb range?
2. Is there a way to save an OSC configuration as a new wave? Specifically I've configured an additive source, and then added FM to it and I want that output as a new wave option.

@Scrubbing, have you looked at the Goniometer? Might want to check to see if you've produced some dc offset.

[Scrubbing Monkeys]

Two questions:

1. Does anyone know where you can change the pb range?
2. Is there a way to save an OSC configuration as a new wave? Specifically I've configured an additive source, and then added FM to it and I want that output as a new wave option.

@Scrubbing, have you looked at the Goniometer? Might want to check to see if you've produced some dc offset.

You cant save those waves yet, but the dev said it was in the plans.

Thanks, Ill check out Goniometer. The wave looks semetric and the body is a goob bit below the peaks. Actually looks quite normal.