How to recreate an EFM-BD sound?

Elektron Gear Machinedrum
#1

A couple of months ago, I created a kit on my MD for a specific track and recorded it in a Cubase project. Recently, I changed the arrangement of the track and wanted to re-record the MD, only to find out that I had tweaked the kit before the original recording and forgotten to save the final changes to the main EDM-BD kick drum.

The previously saved kit gives me a starting point with the settings, and with the help of an oscilloscope and playing the original recoding phase reversed against the current sound, I was able to match the PITCH/DEC/RAMP/RDEC parameters to fit the original recording. But I can’t find the proper settings for the FM parameters MOD/MFREQ/MDEC/MFB.

I already spend way too many hours trying to match the sound and sometimes I got close, but there was always something different in the frequency spectrum. The crazy thing with those EFM parameters is that they can give wildly varying results from one value to the next, e.g. MFB 66 sounds completely different than 65 or 64. So “zooming in” on the target values isn’t really possible.

Yesterday, I even created tracks in Cubase to automate stepping through the parameters while constantly playing the sound against the original recording. Which is rather tiresome.
It’s driving me nuts!

What’s the best approach to recreate an EFM sound? Is there any scientific method that can be applied?

Any hints are welcome :hugs:

Sample attached, if anyone would like to give it a try… :elmd: :crazy_face:

PS:

Here are the values I’ve come up with so far:

PITCH DEC RAMP RDEC
24 87 72 30
MOD MFRQ MDEC MFB
111 (?) ?? 28 ??

PPS: Caveat: I’m triggering the MD via MIDI, which seems to produce a slight shift of the harmonic spectrum for EFM sounds… (which is crazy and worth its own topic)

2 Likes
#2

That’s a beefy bass drum! I’ll try to get there if I can find the time tonight.

1 Like
#3

Apparently there’s a bug that causes notes triggered by midi to have an inverse phase or something. There’s been talk about it for a while AFAIK, I think there was a big thread on the old forum.

#4

Thank you for giving this a thought.

I know and verified that bug. If triggered via MIDI, the sounds are all phase inverted. However, phase inversion does normally not change the spectrum of a sound. It is therefore not detectable by the ear except in conjunction with other sounds, which leads to different phase cancellation patterns.

But with (at least some) EFM machines, in addition to the phase inversion, the spectrum of the sound changes when triggered via MIDI - they sound slightly different, seem to have shifted harmonics.

Now, IF within the EFM machines, the phase inversion introduced by MIDI triggering is not applied to the final waveform, but e.g. only to some parts of the algorithm, THEN that could actually explain the change in timbre.

#5

Thank you, looking forward to your findings!

I forgot to mention, that the Filter Width setting for the sound is around 68, possibly with a bit of resonance, but all other parameters are most probably at their 0 (default) settings.

In the mean time I experimented some more, and found that I also can get pretty close with these base settings:

PITCH DEC RAMP RDEC
24 87 55 32
MOD MFRQ MDEC MFB
?? ?? 28 ??

But I also found out, that all the parameters interact with each other, which means a different value for RAMP or RDEC will give a whole new set of harmonics for all the combinations of the FM settings. So there are 128^5 (34,3 Mrd) possible combinations for the full range of five parameters or if we drastically reduce the parameter values to only a quarter of the range, there are still 32^5 (33, 6 Mio) combinations - which are way too many to explore in a lifetime.

Thus, it is pure luck if anybody ever finds the same combination of values I had while recording that kick.

The fascinating aspect of this quite frustrating discovery is that due to the characteristics of FM sound generation, the MD EFM machines can produce way more different sounds than one might think by looking at the number of parameters.

1 Like
#6

Unfortunately the next day I shipped my MD to Sweden so it will be a couple of weeks before I can look into it. The EFM BD is probably my fav machine and I don’t think I’ve ever made a sound like that.

1 Like