Minimoog Softsynth Shootout: Diva MiniV3 Monark Legend Minimonsta vs Model D

[Ingonator]

I'm just a curious man, but from my reading, CEM3320 doesn't seems to be a chip implementation of the SEM circuit - seems to be two different things.

correct. AFAIR, the SEM was a 2-pole 12db/oct filter, with no self-oscillation, whilst the CEM is capable of 24db/oct, and will self-oscillate. As a filter core chip, rather than a full filter design, the CEM was somewhat reconfigurable toplogy-wise...

http://electricdruid.net/cem3320-filter-designs/

OB-Xa used two CEM3320 chips per voice (overall 16 chips) to do both a 24dB LPF and a 12dB LPF. OB-8 seemed to be able to do both the 24dB and 12dB LPF from a single CEM3320 chip per voice.
Prophet 5 Rev. 3 seemed to be based on the CEM3320 too but had a single 24dB LPF.

All 3 synths used CEM3340 chips for the VCOs and CEM3310 for the EG (envelopes).

Overall technically the OB-Xa and OB-8 seem to be closer to a Prophet 5 Rev. 3 than to the OB-X that was built on discrete SEM type filter and VCO circuits.

The new OB-6 seems to use the SEM circuits too and also includes a morphable multimode filter like the original SEM (or the Oberheim 4 voice and 8 voice that were based on mutiple SEM units).

[egbert101]

<delete>

[whyterabbyt]

I don't think people are getting the point. A 12db or 2-pole filter is a 12db 2-pole filter. Or is it? That's all I'm trying to understand between the difference with the SEM 12db 2-pole filter and the Curtis integrated version. If they are functionally the same, even if the circuits are slightly different, then the result is the same tone or sound

No. Just because two filters of a given pattern have the same number of poles (and thus the same slope) that does not mean the result is 'the same tone or sound'. There's a lot more to the tone of a filter, and its actual effect on the frequency content of a signal, than just the slope.

[whyterabbyt]

correct. AFAIR, the SEM was a 2-pole 12db/oct filter, with no self-oscillation, whilst the CEM is capable of 24db/oct, and will self-oscillate. As a filter core chip, rather than a full filter design, the CEM was somewhat reconfigurable toplogy-wise...

http://electricdruid.net/cem3320-filter-designs/

OB-Xa used two CEM3320 chips per voice (overall 16 chips) to do both a 24dB LPF and a 12dB LPF. OB-8 seemed to be able to do both the 24dB and 12dB LPF from a single CEM3320 chip per voice.
Prophet 5 Rev. 3 seemed to be based on the CEM3320 too but had a single 24dB LPF.

what's your point?

[Ingonator]

I don't think people are getting the point. A 12db or 2-pole filter is a 12db 2-pole filter. Or is it? That's all I'm trying to understand between the difference with the SEM 12db 2-pole filter and the Curtis integrated version. If they are functionally the same, even if the circuits are slightly different, then the result is the same tone or sound, but what might be making the difference between the OB-X and OB-Xa is oscillator drift, saturation, hiss, or something else going on.

It just seems logical to conclude the difference in sound is because the Curtis chips are by nature more stable, resulting in a more tune perfect sound, but less analog pleasing sound (i.e., less drift, less hiss, less saturation and so on).

Anyway, I won't waste any more of people's time about it.

AFAIk the CEM3320 is not based on the SEM design. The original SEM filter was a state variable filter that could morph between different modes like e.g. LPF, BPF(or Notch) and HPF. It also did not self-oscillate and seemed to keep the low end at high resonance and/or even get louder than more queit at high resonance.
Not sure how far the SEM on OB-X is different to that in the original SEM but OB-X did not seem to self-oscillate too.

CEM3320 seemed to be a ladder filter that was able to self-oscillate but also looses low end at higher resoanance.

[egbert101]

<delete>

[Ingonator]

what's your point?

You mentioend 24dB while OB-Xa and OB-8 could do 12dB too while OB-Xa used a second chip for that per voice.

[egbert101]

<delete>

[whyterabbyt]

what's your point?

You mentioend 24dB while OB-Xa and OB-8 could do 12dB too while OB-Xa used a second chip for that per voice.

Yes; I said 'capable of 24dB', pointed out its topology was configurable, and provided a link to a list of topology implementations based on it that included the OBXa 12db and 24dB filters.

[whyterabbyt]

But that does not answer my question. Maybe someone can understand my frustration.

Ive answered it, so no I dont understand your frustration. Filters of the same pattern (LPF etc) with the same slope can and do sound different from each other, because there's more to the sound of a filter than the pattern or slope.

[Ingonator]

I don't think people are getting the point. A 12db or 2-pole filter is a 12db 2-pole filter. Or is it? That's all I'm trying to understand between the difference with the SEM 12db 2-pole filter and the Curtis integrated version. If they are functionally the same, even if the circuits are slightly different, then the result is the same tone or sound, but what might be making the difference between the OB-X and OB-Xa is oscillator drift, saturation, hiss, or something else going on.

It just seems logical to conclude the difference in sound is because the Curtis chips are by nature more stable, resulting in a more tune perfect sound, but less analog pleasing sound (i.e., less drift, less hiss, less saturation and so on).

Anyway, I won't waste any more of people's time about it.

AFAIk the CEM3320 is not based on the SEM design. The original SEM filter was a state variable filter that could morph between different modes like e.g. LPF, BPF(or Notch) and HPF. It also did not self-oscillate and seemed to keep the low end at high resonance and/or even get louder than more queit at high resonance.
Not sure how far the SEM on OB-X is different to that in the original SEM but OB-X did not seem to self-oscillate too.

CEM3320 seemed to be a ladder filter that was able to self-oscillate but also looses low end at higher resoanance.

But that does not answer my question. Maybe someone can understand my frustration.

As mentioned above the SEM and OB-X semed to use discrete VCO circuits while OB-Xa, OB-8 and Prophet 5 Rev. 3 used CEM3340 chips for the VCOs which should not be similar to the SEM circuits.
So both the VCF and VCO circuits are different which of course should result in differences between OB-X + SEM and the CEM based ones.

The SEM and OB-X filter was only 12dB LPF while the CEM could also do a 24dB LPF besides a 12dB LPF which could make a difference in sound too.

Last but not least the resonance bahavior of SEM based and CEM based filters semed to be different too (as i tried to mention in my post above).

But that does not answer my question. Maybe someone can understand my frustration.

Ive answered it, so no I dont understand your frustration. Filters of the same pattern (LPF etc) with the same slope can and do sound different from each other, because there's more to the sound of a filter than the pattern or slope.

Exactly that!!!

The possible sound differences of filters with a similar slope were quite obvious to me so i might not recognized that this might be the problem of egbert101 here...

For example U-He Diva seems to include 5 different Lowpass filters with a 12dB slope that could sound quite diffeernt.
With many filters the biggest differences are found at higher Resoanance amount while without any Resoannce tehy might sound closer to each other.

In my Bass Station 2 at higher Resonance i coudl also find such difference with teh "Acid" filter which is a diode ladder 24dB Lowpass filter and the "Classic" 24dB Lowpass filter that is a transistor ladder filter.

The 24dB Lowpass of CEM3320 will sound different to a Moog 24dB Lowpass and so on...

The Legend uses the 24dB Minimoog LPF and Repro-1 uses a CEM3320 so you could spot the differences there, especially at higher resonance. Besides that the CEM3320 in the Pro One and Repro-1 has a different Cutoff range as it has e.g. in a Prophet 5 Rev.3 or OB-8.

You could include Arturia SEM V2 in the comparison and especially at higher resonance you should find a difference between the filters in The Legend, Repro-1 and SEM V2.
Opposing to a real Minimoog The Legend also has a 12dB LPF (and Diva has it too) which could be quite useful for certain sounds.

[chk071]

So, will Synapse emulate one of the great Oberheim poly's then? Seriously, i'd buy that for sure. Pretty please.

[AnX]

So, will Synapse emulate one of the great Oberheim poly's then? Seriously, i'd buy that for sure. Pretty please.

Hopefully not. Obies sound dull as f**k to me. I've never understood the fascination. Personally, i'd like something else. Luckily, my opinion counts for nothing, so dont get ya panties in a knot

[chk071]

Fair enough.

[spunkmuffin]

Comparison video between Oberheim OB-8 and Oberheim OB-Xa. Strictly for nerds.

It's very clear to me. The OB-Xa sounds like a biscuit, while the OB-8 sounds like a cookie.