Mammoth Fuzz inspired patch

#1

Hi guys,

Here's a patch inspired by the awesome Zvex Wooly Mammoth fuzz pedal for our beloved Axolotii. I've experimented a lot with the 40106 IC which is a schmitt trigger, to make oscillators/synths, wild distortions... So i hoped it could work that way with the axoloti. Thanks for your schmitttrigger @johannes, it was a very nice surprise!

Today i soldered a few potentiometers (some oldschool B5k and B50k both work like a charm, i tried B100k as well, but it didn't reach the maximum range) to the axoloti to see how it worked, i wanna say thanks to all of you who posted questions, it helped me greatly to get the hang of it, so i left the gpios, just replace them with the ones you use.

4 controls:

(PAO and PA1)
Drive 1 // Drive 2: both are reading a delay line, with these two controls you can create tons of different kind of fuzz / high gain distortions. Very authentic/usable sounds, not noisy, but i wish i could add a little noise gate at the end, but i didn't know how to do it.

(PA2)
EQ: the pedal has the same kind of 1st order LP filter.
(Edit: i know i didn't use a 1st order LP here lol, it's only because it was the LP which came first)

(PA3)
Volume: output volume, adjust to match your ideal gain staging.

I tried the patch with an 8 string guitar, it should work for both bass and guitar, and i'm sure all kind of sources as well.

I'll try to make other patches with a 'guitar/bass pedal' approach in the future!

Have fun!

Flo

flo x mammoth fuzz.png1146x860 67 KB

flo x mammoth fuzz.axp (6.7 KB)

Edit 2: here's an audio demo of the patch with an 8 string guitar:

#2

Thanks for sharing!

It's not hard to follow the signal path, but I'm not sure what the purpose of the delay line and double delay/read is?

Also, have you taken into account execution order? I can see a few places where additional an unneeded delay is being introduced (because objects earlier in the signal chain are being executed after objects that are later).

What makes this particularly a Mammoth emulation? Is it something to do with the curious usage of the delay line?

#3

I was more experimenting with these delay lines to try to find a 'new sound' or a 'new way' to make an usual sound! I'm always looking for ways to freshen up my fun/inspiration while making sonic tools like that, so when i use them in a session i can stumble on 'happy accidents' more than i could with a regular fuzz pedal/sound! For me that's my sole goal when diying a hardware effect, or coding a vst, is to make things that i can't get from a 'regular effect'. I try to flee 'know how' with music, because it's a fun killer, and you need that fresh enthusiasm to make quality tracks. I compose from 300 to 500 tracks per year, believe me over the years, the fun can be eroded haha!!

I haven't taken in account anything but the pleasure brought to me by playing the patch haha :slight_smile: The delays are unnoticeable when you play directly thru the patch, so it's all good to me! But as i said in my first post (shoegaze chorus), i'm really not a master of it at all, and i'm really taking all possible advices you guys can give me. I'm an experienced producer who knows which final sound i want, so i chase that, and make it work with the programming, and if something can go 'wrong' i'll be pretty happy, if it's usable and expressing an emotion / energy, i'll take it over anything 'regular'!

What makes this an emu of the wooly mammoth is the sound! What makes this a different thing is the way you can tweak the distortion sound and give it another character!

#4

Does it sound identifiably more like the Zvex Mammoth than it does any other fuzz pedal? Because it doesn't even have the same set of controls as the original, I find it a bit of a stretch to call it an emulation.

Execution order is worth paying attention to, even if this patch doesn't suffer any noticeable issues. One day a lack of attention to execution order will bite you in the arse and there'll be some problem that you can't work out the cause of.
Objects get processed in 'western' reading order - so in rows from left to right. The bit that matters is the level the top of the object is on, if one object in a chain is lower than the rest then it delays the entire thing. This has greater implications than latency as it can result in parallel chains going out of sync and all other kinds of problems.

In your fuzz patch, you have the two delay reads - one above the mixer and one below. This kind of patching is problematic as they reach the mixer at different times because the one below the mixer misses its chance and has to wait for next time round, by which point the upper delay read is 16 samples later.
The mixer then goes to a filter which is slightly higher than it, causing the signal to 'miss the boat' again as the filter has already been processed in that pass. And then the output is higher again causing another delay.

I'm of the opinion that if it sounds good, it is good, so if this sounds good to you then it's a job well done. I am however hoping to make you aware of the potential issues with such a layout so that it doesn't cause you problems later.

So for general purposes, the signal flow should go across the page like lines of text with the output at the bottom right.

#5

Does it sound identifiably more like the Zvex Mammoth than it does any
other fuzz pedal? Because it doesn't even have the same set of controls
as the original, I find it a bit of a stretch to call it an emulation.

-> it only misses the 'threshold' knob as i said in the description, which is a way to apply the distortion to the signal depending on the set level of this threshold! It may be far fetched to call it a 'proper' emulation since i'm not using the same components, but believe me in the sound, it's an instant reminder of it! It's kinda like the diy community, they won't necessarly use the same components, but they'll say it's a clone of something, because it sounds like it.

This has greater implications than latency as it can result in parallel chains
going out of sync and all other kinds of problems.

-> if you think about phase issues, then don't worry because i'm pretty aware/sensitive to it, i will only do uncommon things with it, if it benefits the final sound!

In your fuzz patch, you have the two delay reads - one above the
mixer and one below. This kind of patching is problematic as they reach
the mixer at different times because the one below the mixer misses its
chance and has to wait for next time round, by which point the upper
delay read is 16 samples later.

-> this difference in the reading of the delay lines is intentional, it's the way to modify the distortion character, it's like a phase manipulation!

The mixer then goes to a filter which
is slightly higher than it, causing the signal to 'miss the boat' again
as the filter has already been processed in that pass. And then the
output is higher again causing another delay.

-> That's exactly why i posted this patch, to get some feedback from more knowledgeable persons to learn from :slight_smile: So when you say that the filter is slightly higher than it, you mean that 'it should be' a value below or equal to the mixer? Like the top object should never be higher than what's before it?

How different is it from gain staging when you program patches for audio signals like that?

So for general purposes, the signal flow should go across the page like lines of text with the output at the bottom right.

-> you mean that visually? I'm sorry if i don't understand you well, i'm trying to haha :slight_smile:

#6

i just added an audio demo to it!

#7

Well not quite. Your "Drive" controls set two separate delay times, which has nothing to do with drive amount but rather would control a sort of feedforward comb filter effect. Perhaps we're looking at different pedals, because the Zvex Woolly Mammoth I'm aware of has one drive knob (Wool) and a bias control (Pinch). I don't know what you mean about there being a threshold control.

There's a difference between making a clone that uses different but equivalent parts but still works in the same way and making something that works completely differently and calling it an emulation. See here: http://mutable-instruments.net/forum/discussion/7675/cool-tutorial-for-diy-solina-like-synth/p1

Cool, I was just making sure you knew that it worked that way so you knew what to avoid when there are times you want to avoid it.

Well, this isn't achieving anything you couldn't just achieve with your 'Drive' controls since they're just altering the delay time. The backwards execution order is just adding a 16 sample offset to one of the controls.

I mean the actual physical positioning! Axoloti's execution order is defined by the positioning of the objects on the page. If you look at the screenshot you posted, the filter object is one space higher up the page than the mixer before it.
As I say, the execution order works like writing on a page, so it starts at the top left and does everything on that line, then it goes back to the left and does the next line.

Pasted image1092x637 62 KB

So the red line I drew shows the order your patch is being executed in - very different to the signal flow! As you can see, the filter is being executed before the mixer. What this means is that on the first pass, there is no signal for the filter to process. Then the mixer sends its signal to the filter, but the filter has already been executed for this pass so the signal has to wait. On the second pass the filter processes the signal from the first pass and then the mixer sends the actual signal from the second pass - but again the filter has already run so that signal has to wait for the third pass and so on!

Gain staging is important but nothing to do with the stuff I was referring to. It's definitely worth consideration but the rules are different with digital audio and I won't go into it here.

That's exactly what I meant! Unlike with analogue stuff where everything happens simultaneously, in a computer things have to be done one at a time. If it was up to the patcher to decide what order to do things, it might sometimes make decisions that stop your stuff from working so Johannes made it up to the user to decide the order and the way you do that is through positioning of the objects. It took me a while to get used to after a couple of weeks of patching without any regard for execution order and getting lots of weird bugs that seemed to appear and disappear by themselves because I'd inadvertently moved something. I think it was worst when I was trying to make a drum machine. :slight_smile:

#8

thank you so much, i never thought of that, i thought the cables made the connections like patching audio hardware, and that it was what was important lol! I will try that for my next patch for sure, it can be another form of inspiration, just what i like :)!

Well not quite. Your "Drive" controls set two separate delay times,
which has nothing to do with drive amount but rather would control a
sort of feedforward comb filter effect. Perhaps we're looking at
different pedals, because the Zvex Woolly Mammoth I'm aware of has one
drive knob (Wool) and a bias control (Pinch). I don't know what you mean
about there being a threshold control.

Yes this sort of feedforward comb filter effect is exactly why i did that, there would be no interest to me in just replicating the wooly mammoth as it is, i already used it, i made a few circuits out of it so i've ran out of fun for that box, finding that way to make a different sound with a similar character was what thrilled me to do it!

Pinch is just a glorified threshold believe me, i'm not a pro at these details, but i got good ears haha ! A threshold as in a compressor threshold meaning of course!

#9

i edited the first post so it's more clear that it's inspired by and not a clone of :wink: thank you!!

#10

Well the cables do make the connections, the positioning just decides the order that the processing is done since it can't happen all at once like with analogue stuff.

I'm fairly certain it's a bias control, you can get a gating effect with some bias settings. You might be able to emulate that by adding an offset to the audio before the Schmitt trigger. The gating happens when the "zero line" of the input rests on one side of the tipping point so that audio below a certain level won't cause it to switch.

#11

I'm fairly certain it's a bias control, you can get a gating effect with
some bias settings. You might be able to emulate that by adding an
offset to the audio before the Schmitt trigger. The gating happens when
the "zero line" of the input rests on one side of the tipping point so
that audio below a certain level won't cause it to switch.

I'm really not saying that it's not a bias control, all i'm saying is that i interpret it/use it/understand it as if it was a threshold, because to me it acts like a threshold does. Enabling more or less of the sound to excite the distortion, as a threshold would on a compressor. Believe me i really understood since your first reply that you're more advanced than me at linguo and programming, i'm just expressing it how i perceive it as a musician, with my musician cultural background.

I tried to do what you're saying but i didn't succeed, so i left it out, and to make you laugh, i tried it with a threshold from a compressor :wink:. What would be the name of the object that i can use to add an offset to the audio?