Sujet Choix d'un synthétiseur.

Citation : Now the bad part. The really bad part. The unbelievably bogus bad part.

At home in the studio where the noise floor is _significantly_ lower than the NAMM show floor, I could hear all kinds of harmonics when the oscillators were set to sawtooth. At first I thought maybe the waveshaper circuit just wasn't up to snuff, but the more I played with the unit the more I knew something was amiss. What really opened my eyes were some horrible anomalies with the oscillator sync and ring mod settings. When set to oscillator sync there is a horrible clicking/beating sound at about 1/4 the beat frequency of the two oscillators. The ring mod also has some "steppiness" right around the null point that I couldn't put my finger on.

So how do I determine the source of my woe? Time to crack the unit open and take a look. Immediately upon extracting the PCB from the case I notice something is amiss. There is no analogue circuitry in the oscillator sections. Over by the filter I only find enough trannies and OTAs to support the filter and VCA. So where are the oscillators? At the bottom left of the board is Fujitsu microcontroller of some sort so lets go look that up on the net. Turns out it's a fairly powerful Fujitsu F2MC 16 bit MB90F497G microcontroller with 128kB of ROM and 4kB of RAM, as well as a nice complement of built in A/D. (datasheet available at http://www.fma.fujitsu.com/pdf/e713713.pdf) It only took a minute to trace out the circuit path of the tuning controls in the oscillator section to see they're feeding the microcontroller. After studying things a little more and putting an oscope on the micro-p it's obvious what's going on. Despite what MFB's own website, marketing literature, user manual, and unit itself says, this unit has

*** 100% DIGITAL OSCILLATORS ***

What a crock of #@$*in' @#$@!!!! No wonder the sawtooth oscillators have harmonics - the unit uses a 3 bit psuedo-DAC to make the sawtooth. Even worse is that the "ring mod" is digital. Don't even get me started on the pulse wave shape being positive DC. Or the bug in the last note priority code. Or that the LFO sawtooth waveform modulates the oscillators inversely.

So what's this all mean? If you want a small synth with the Oberheim detune sawtooth sound you best look elsewhere. In fact, unless you filter the sawtooth waveforms heavily with the filter plan on only using square waves and forget oscillator sync or ring mod unless you can live with rhythmic audible glitching. I'm really disappointed because up until I noticed the sawtooth waveform and ring mod/sync glitches I really thought this was on par with a non-patched Oberheim SEM. It would be so awesome to have a sub-$500 MIDI analogue synth with the features of the OB SEM. And the MFB almost delivered. *** ALMOST ***

I'll probably hang on to my MFB Synth Lite. It's a nice little box and will be a fun addition to rig at work. I will be following up with MFB and Geoff Farr (US distributor) so that their website and all documentation is corrected and that they do not misrepresent this unit as having analog oscillators. For those interested I have a high resolution picture with key components labeled on my site at http://www.retrosynth.com/gear/synthlite/synthlite_pcb_labeled.jpg.

PS: I'm hoping someone at EM and/or Keyboard mag plagerizes this article when they do a review of this synth.

Citation : Oui j'ai dejà lu cet article, mais MFB parle clairement de DCO sur son site http://www.mfberlin.de/Produkte/Musikelektronik/MFB-SYNTH_LITE/MFB-SYNTH_LITE_e/mfb-synth_lite_e.html ... feraient-ils de la publicité mensongère?

Citation : Despite what MFB's own website, marketing literature, user manual, and unit itself says, this unit has

*** 100% DIGITAL OSCILLATORS ***

Citation : Sinon je sais que ce n'est pas le même prix mais une bass sur un synth 100% discret c'est quand même plus chaleureux et organique.