Messages

1

preenfm2 and preenfm3 / Re: Sawtooth waveform

« on: December 23, 2018, 06:34:28 PM »

I think the built in saw wave is heavily bandlimited, IIRC Xavier said it was additively generated with something like the first 20 partials.

To my ear, the sudden stop after the top partial gives the impression of resonance.

I believe it's due to Gibbs Phenomenon. I've been meaning to try synthesising my own bandlimited saw with a bit more of a dropoff in the harmonics to lose the ringing but I haven't got around to it.

2

preenfm2 and preenfm3 / Re: pwm

« on: February 21, 2018, 02:23:42 AM »

Just off the top of my head, I imagine this approach would cause a lot of aliasing but don't let that stop you trying! I'd love to hear how it turns out.

3

preenfm2 and preenfm3 / Re: Subtle distortion on one channel

« on: November 15, 2017, 03:47:55 PM »

Good idea, will try and do that this evening.

4

preenfm2 and preenfm3 / Subtle distortion on one channel

« on: November 15, 2017, 01:50:12 PM »

My PreenFM2 seems to have a subtle distortion on one side of the stereo image. It doesn't sound like clipping, but rather that kind of digital distortion you get when the steps in the DAC aren't quite the same size. I've ignored it for a long time as it's very subtle and I rarely use it in stereo but I'd like to diagnose it now if I can.

Can anyone suggest any steps? Swapping the DAC ICs would be an obvious one... see if the distortion follows the DAC or not.

5

preenfm2 and preenfm3 / Re: November 5th 2017 -> 2.08

« on: November 14, 2017, 03:15:50 PM »

hmm ok. so long delay is out of sight :-)

free verb seems doable, and it sounds very nice especially for synth sounds:

https://ccrma.stanford.edu/~jos/pasp/Freeverb.html it uses comb filters and allpassfilters.

What's the sum of all the delays used for the combs and allpasses?

6

preenfm2 and preenfm3 / Re: November 5th 2017 -> 2.08

« on: November 14, 2017, 03:14:20 PM »

I think i remember there's 60Ko of RAM free on the preenfm2.
That's 30Ko per channel (stereo).
Each float takes 4 bytes to store so that's 7500 samples, at 40.000Hz your delay can be 0.18 second long... Could be a chorus maybe.

I don't know to much how a reverb algo works, but i assume the CPU usage will be big.

You could double the length by using 16bit fixed point for the delay, double again if you made it mono. Maybe not worth it since it would still be super short. Chorus/flange would be fun though.

7

preenfm2 and preenfm3 / Re: PreenFM2 SMD version

« on: November 07, 2017, 05:47:33 PM »

While I have no need of another PreenFM2, I definitely want to see how the build comes out.

8

preenfm2 and preenfm3 / Re: preenfm2 editor 2.08.1 (windows and macOS standalone version)

« on: October 29, 2017, 05:21:06 PM »

Copying and pasting envelopes is super useful!

9

preenfm2 and preenfm3 / Re: Thoughts about operator tuning

« on: September 07, 2017, 11:46:12 PM »

Hooray, thanks lokki for implementing it and thanks Xavier for the firmware being open source!

10

preenfm2 and preenfm3 / Re: [Feature Request] Adjustable Seq/Arp Length

« on: January 09, 2017, 01:15:50 PM »

Just wanted to add support for this idea. Always in favour of being able to reduce the number of steps in a sequencer for other time signatures and cross phrasing.

11

preenfm2 and preenfm3 / Re: Thoughts about operator tuning

« on: May 03, 2016, 11:57:40 AM »

It probably wouldn't be too hard, but I wouldn't know where to begin in terms of setting up the build environment because I'm not really a programmer. But you can get a linear tuning offset by adding directly to the phase increment.

12

preenfm2 and preenfm3 / Re: Thoughts about operator tuning

« on: March 29, 2016, 12:04:47 PM »

That would be fun!

Or Hz based detune per timbre, for flanger-like effect when using same patch on different instrument slots.

I might actually try this, when I make me some more free time...

Computation-wise I suppose it wouldn't be to hard on cpu, just adding simple increment value (related to detune value) to phase increment
I haven't yet touched anything in preenfm2's UI code, so that might get a bit tricky there, I don't know...

Yeah, I can't imagine it's difficult computationally - it's just adding an offset to the phase increment at a later stage (cent detune is obviously before the 12th root conversion).

13

preenfm2 and preenfm3 / Thoughts about operator tuning

« on: March 22, 2016, 05:15:07 PM »

I noticed that some FM synths allow detuning in cents and Hz. I originally assumed that it was just to allow finer degrees of detuning, but now I understand more about synthesis and sound in general I realise the purpose

 Hz is of course a linear measurement whereas cents are a (very small) musical interval. So a detune in Hz is actually a different interval depending on what note you're playing. "How could that be useful!?" I hear you think. Well anyone who is familiar with detune by cents on a digital (therefore stable) synth may have noticed that the beat frequency is proportional to the note you're playing. A note an octave higher will beat twice as fast.

The beat frequency is the difference in Hz between the two oscillators, so the beating is changing frequency because when you play an octave higher you've doubled the frequency of both oscillators, so the difference between their frequencies has also doubled - doubling the rate of beating.

If you detune by a fixed number in Hz then you get the same beating rate on every note. This is particularly useful in FM where detune between modulators and carriers causes timbral changes and note just a 'chorusing' sound.

This isn't a feature request, just some thoughts.

I would propose it as a third "FTyp" parameter, maybe called KeHz or something. Freq would still be a ratio setting on this mode, but Ftun would be in Hz instead of cents. This would allow for sounds that have a cycling quality that is at a consistent rate across the keyboard. I think 2 decimal places (allowing for 0.01) is good as it allows incredibly slow phasing.

14

preenfm2 and preenfm3 / Re: Newbie questions

« on: March 22, 2016, 04:55:09 PM »

Yes sin^2 seems to be useless.
All the sin variant are not great and take some memory. I only keep them for preset compatibility since the first versions of the preenfm1.
But now you have user waveforms....   

Yeah, it definitely needs to stay for compatibility. I just remembered that someone told me that sin^2 had a different spectrum and I believed it until more recently when I'd learned enough about these things to realise it wasn't true.
It does have a use though, it allows a wider ratio range so if you want a wide space between modulator and carrier, this allows for an extra octave.

15

preenfm2 and preenfm3 / Re: Newbie questions

« on: March 21, 2016, 04:00:03 PM »

@SirPrimalform But sin and sin squared do not have exactly the same spectrum, even apart from the doubled octave. In the following link I have plotted sin(x) vs (sin(x))^2: https://www.wolframalpha.com/input/?i=plot+sin+x+and+(sin+x)%5E2

As you can see their shapes differ subtly, meaning their spectra should differ (I guess you get a bunch of subtle overtones when you square the sin).

The difference is that Sin^2 is an octave higher, half the amplitude and with an offset of +0.5.
But there's no fundamental difference in shape. Look: https://www.wolframalpha.com/input/?i=plot+sin+x+and+((sin+(x%2F2%2Bpi%2F4))%5E2)*2-1


plot sin x and ((sin (x/2+pi/4))^2)*2-1

x is divided by 2 to drop it an octave, +pi/4 is to correct the phase, *2 is to correct the amplitude and -1 is to centre it around 0.
I did test this before posting, the maths and the theory work out.

EDIT: Just saw your edit.

p.s. This is a really nice online graphing calculator: https://www.desmos.com/calculator/fnynlawfti