Why do Sequential/DSI synths sound "harsh" to me?

[LoboLives]

That’s just the way the Curtis Filter is designed in regards to its mixer. Filter envelope is also key. Listening to this video...I don’t really hear the harshness people are taking about
https://youtu.be/STaHYVmZBi8

Really? You don't hear that annoying nasality/buzziness/fizziness/scratchiness/metallicness/tinnyness/etc?

No, it’s bright when the filter is open but that’s about it. It really comes down to programming I find.

[jg666]

I don't hear any harshness with the Rev2. I'm not knowledgeable enough to get into the discussions about why some hear harshness and others don't   

It is good though that we all like different synth/filter sounds in my opinion as it would be boring if we all liked the same thing! There are other synths which people rave about that I don't like the sound of, whereas as soon as I heard the Rev2 in person in a music store I was blown away by it.

Edit - I think it's probably the bi timbrality of the Rev2 that makes the sound something more enjoyable for me. I had a play about with the Peak at the same time as the Rev2 in the shop and the Peak didn't hit the spot like the Rev2 did.

[Razmo]

I'm not sure if this is relevant, but in the early days of DSI synths, the cutoff parameter had a parameter range from 0->128 ... since that it was raised to be from 0->164 ... I wonder if from 128->164 is the extra "harsh range" some people seem to hear.

[LoboLives]

I'm not sure if this is relevant, but in the early days of DSI synths, the cutoff parameter had a parameter range from 0->128 ... since that it was raised to be from 0->164 ... I wonder if from 128->164 is the extra "harsh range" some people seem to hear.

Interestingly on the P5 presets in the P6, the oscillators are almost never fully open and the filter never goes past 1 or 2 o clock (which happens to me around 120 on the editor). Even on sounds you think the filter is wide open, it's not.

[Razmo]

I'm not sure if this is relevant, but in the early days of DSI synths, the cutoff parameter had a parameter range from 0->128 ... since that it was raised to be from 0->164 ... I wonder if from 128->164 is the extra "harsh range" some people seem to hear.

Interestingly on the P5 presets in the P6, the oscillators are almost never fully open and the filter never goes past 1 or 2 o clock (which happens to me around 120 on the editor). Even on sounds you think the filter is wide open, it's not.

Exactly... and I know that back then, DSI changed this range... and I bet it was because of the opposite... people back then wanted it to be brighter.... it's just a giveaway, that you simply cannot satisfy everyone ever... there will always be users complaining about something... increasing the value solved the brightness problem back then... but introduced it as "harshness" for other users not knowing that all they have to do is lower the range a bit on the filter cutoff... it's rather simple really... just lower the darn Envelope Amount knob, or control your modulation matrix amounts etc. 

[maxter]

...yes, and after all this is subtractive synthesis, so oscillators being very rich in higher harmonics (harshness?) is eligible in this case. The filter serves its purpose, and I'm actually fond of the "harshness" as it makes for more sound design options and can be easily tamed anyhow. Like a toolbox containing more tools, even if you mostly just use a couple of tools, it doesn't hurt to have the others available.

I'd bet that if the oscillators were lamer and/or the filter thinner, the lack of "harshness" would have some people complaining about the sound being "dull" (I probably would). That scenario would be a lot worse and less convenient to overcome, adding something not there to begin with.

I personally prefer the 2-pole mode with cranked up resonance compared to the 4-pole (even without resonance), if using the 4-pole I'll probably stick to using just 1 oscillator, probably attenuated as well. I think the 4-pole can add to the "harshness", which is great if/when you'd want that.

[AlainHubert]

It's not only a question of the number of poles, but much more the design of a filter that make up its qualities.
An Oberheim SEM filter is very nice, a Moog ladder filter too. A Curtis one from the late productions (i.e. CEM3396) not so much.
The CEM3320 sounded very nice too, but not quite as good as the SSM2040.
But it's all a matter of personal taste, I guess...

[Razmo]

There are so many factors and opinions that matter that there really is no right or wrong... a good filter (if it's YOUR opinion), is not so useful, if you have crappy oscillators... or none of these matters, if the engine is not flexible enough... and if the envelopes are not to your liking in response, it's even more bad... there are so many factors to take into account on what "a good and well sounding synth" is... yes... the Curtis is perhaps not the best compared to a MOOG, but the REV2 engine will beat most MOOG engines, giving you more options etc. etc... I never understood theese debates about a synth being good or bad... in fact I don't give a s**t about what others opinion are... I use what I feel makes me wanna make sounds and music... fair and square.

[jg666]

Like others have said, I don't care what others think of any of the synths I own   I liken all this kind of debate to something similar to liking particular music or a particular band.... example....

To some people, Neil Young is a God but to my ears his voice sounds like a cat being strangled and his guitar solos sound like he's dropped his guitar down a long flight of stairs whilst still plugged into the amp.

I'll never agree with those people who say he's great but they have the right to their opinion I'll never agree with criticisms of the sound I can produce with my Rev2 but everyone hears things differently and wants different things out of their keyboards.

[skrohmer]

There are so many factors and opinions that matter that there really is no right or wrong... a good filter (if it's YOUR opinion), is not so useful, if you have crappy oscillators... or none of these matters, if the engine is not flexible enough... and if the envelopes are not to your liking in response, it's even more bad... there are so many factors to take into account on what "a good and well sounding synth" is... yes... the Curtis is perhaps not the best compared to a MOOG, but the REV2 engine will beat most MOOG engines, giving you more options etc. etc... I never understood theese debates about a synth being good or bad... in fact I don't give a s**t about what others opinion are... I use what I feel makes me wanna make sounds and music... fair and square.

agreed totally

Like others have said, I don't care what others think of any of the synths I own   I liken all this kind of debate to something similar to liking particular music or a particular band.... example....

To some people, Neil Young is a God but to my ears his voice sounds like a cat being strangled and his guitar solos sound like he's dropped his guitar down a long flight of stairs whilst still plugged into the amp.

I'll never agree with those people who say he's great but they have the right to their opinion I'll never agree with criticisms of the sound I can produce with my Rev2 but everyone hears things differently and wants different things out of their keyboards.

agreed totally

Nevertheless thousands of people are fighting for their opinions every day. I'm really glad to read such reasonable statements from time to time. Thanks guys! 

[Djinn]

Neil Young is great! 

[maxter]

Of course I, too, agree with the above. My previous point was not so much about individual taste concerning filters and "good" sound in general though, but more a kind of praise of the Rev2 oscillators and filters "going to 11" (as Razmo mentioned earlier), which opens up for more sound possibilities. Considering it's a subtractive synth, the more starting material to subtract from, the better (generally speaking, of course it depends on the actual material character/quality as well, and the ways and designs to subtract from and modulate it etc etc etc).

Only reason I mentioned my general preference of filter settings on the Rev2 (not every other synth, I may add), was to relate to the main topic of "harshness" in this thread in what way I think the filter could be adding to this effect, and an idea on how to avoid it (since it doesn't seem to be the character that the original poster likes). So, my "opinion" on this was not in any way to suppose what is "good" or "bad", in any way. Did anybody not miss my point?   

My opinion is this is an interesting topic.

[Sacred Synthesis]

It seems to be the claim here - at least among some - that the only way to avoid harshness in the Prophet '08/Rev2 is to somewhat close the filter.  So, in 4-pole mode, a fully open filter creates an inescapable harshness?  I wouldn't agree with this.  There are ways of keeping a wide-open filter from being overbearing, the two most obvious being to restrain the VCA level and to be wary of resonance.  With some patches, it even makes sense to slightly lower the VCA envelope sustain. 

Yes, at higher amplitudes and with excessive resonance the instrument can sound harsh, but it isn't necessary; nor would I call such harshness a characteristic of the instrument.  It's only a potential.

[Razmo]

Of course I, too, agree with the above. My previous point was not so much about individual taste concerning filters and "good" sound in general though, but more a kind of praise of the Rev2 oscillators and filters "going to 11" (as Razmo mentioned earlier), which opens up for more sound possibilities. Considering it's a subtractive synth, the more starting material to subtract from, the better (generally speaking, of course it depends on the actual material character/quality as well, and the ways and designs to subtract from and modulate it etc etc etc).

Only reason I mentioned my general preference of filter settings on the Rev2 (not every other synth, I may add), was to relate to the main topic of "harshness" in this thread in what way I think the filter could be adding to this effect, and an idea on how to avoid it (since it doesn't seem to be the character that the original poster likes). So, my "opinion" on this was not in any way to suppose what is "good" or "bad", in any way. Did anybody not miss my point?   

My opinion is this is an interesting topic.

Not at all ... I did not even have your comment in my head when I wrote the above... it was just a general statement about the general debate I often see people have, on why this synth is better than that synth... you can certainly debate WHY someone have a bad feeling about a particular synth... I'll even call this thread an ok question... i like these debates if they are constructive, trying to analyze the "problem" etc... and as long as the debating parties have respect and a decent way of going about the debate ... on other groups on the net, these kind of debates can quickly turn ugly... for no good reason really...

I actually tried to "hear the harshness" and feel like I "hear it"... with the filter wide open (or rather the range from 128-164)... I heard it in my small demo i made earlier... i too hear this "harshness" from time to time, but i always attributed it to the way the sound was designed, not as a general all-over trait of the Curtis chip... i often lower the filter envelope amount in my sounds, when they seem too bright/Harsh, i just never thought about it as a character of the entire synth... it goes away when the filter is not totally opened...

[Razmo]

It seems to be the claim here - at least among some - that the only way to avoid harshness in the Prophet '08/Rev2 is to somewhat close the filter.  So, in 4-pole mode, a fully open filter creates an inescapable harshness?  I wouldn't agree with this.  There are ways of keeping a wide-open filter from being overbearing, the two most obvious being to restrain the VCA level and to be wary of resonance.  With some patches, it even makes sense to slightly lower the VCA envelope sustain. 

Yes, at higher amplitudes and with excessive resonance the instrument can sound harsh, but it isn't necessary; nor would I call such harshness a characteristic of the instrument.  It's only a potential.

I recently read a bit from the CEM3396 datasheet, and it actually mention that they created exactly TWO VCA's to circumvent a particular distortion situation if it only had ONE VCA... it was something about the VCA envelope that would get possible distortion, if all volume was to be controlled thru only a single VCA CV... now I'm not sure if this has anything to do with this, but if you experience that this harshness/brightness goes away when lowering the Volume VCA, then it may be that it originate from the VCA curcuit maybe... i do not see how anything should be capable of overdriving the VCA stages, as they are the last in the signal chain inside the chip... the raw oscillators are obviously fed directly into the filter, and from there into the two VCA's after that... maybe it's a combination of the oscillators and the filter with resonance... i read in many Sequential manuals, that you CAN clip the Curtis chip with high resonance... so maybe the VCA's inside the chip can get driven too hard from the filter... or maybe if the output from the filter, when it's too low gets amplified and distorted in the VCA's somewhere? ... in other words, I do not know... but it would be fun to experiment a bit with the signal flow of the Curtis chip some day.

[Sacred Synthesis]

I recently read a bit from the CEM3396 datasheet, and it actually mention that they created exactly TWO VCA's to circumvent a particular distortion situation if it only had ONE VCA... it was something about the VCA envelope that would get possible distortion, if all volume was to be controlled thru only a single VCA CV... now I'm not sure if this has anything to do with this, but if you experience that this harshness/brightness goes away when lowering the Volume VCA, then it may be that it originate from the VCA curcuit maybe... i do not see how anything should be capable of overdriving the VCA stages, as they are the last in the signal chain inside the chip... the raw oscillators are obviously fed directly into the filter, and from there into the two VCA's after that... maybe it's a combination of the oscillators and the filter with resonance... i read in many Sequential manuals, that you CAN clip the Curtis chip with high resonance... so maybe the VCA's inside the chip can get driven too hard from the filter... or maybe if the output from the filter, when it's too low gets amplified and distorted in the VCA's somewhere? ... in other words, I do not know... but it would be fun to experiment a bit with the signal flow of the Curtis chip some day.

We haven't quite defined the meaning or cause of the "harshness" in this thread, so things are a bit vague.  What I was referring to above was a harshness that was the result of slight distortion.  Because I use layered patches all the time, distortion is a serious and constant problem for me.  High audio levels, distortion, and a fully-open filter are a recipe for musical disaster.  But again, I do think a full open filter can still be made to sound pleasant, rather than harsh.

[guyaguy]

I recently read a bit from the CEM3396 datasheet, and it actually mention that they created exactly TWO VCA's to circumvent a particular distortion situation if it only had ONE VCA... it was something about the VCA envelope that would get possible distortion, if all volume was to be controlled thru only a single VCA CV... now I'm not sure if this has anything to do with this, but if you experience that this harshness/brightness goes away when lowering the Volume VCA, then it may be that it originate from the VCA curcuit maybe... i do not see how anything should be capable of overdriving the VCA stages, as they are the last in the signal chain inside the chip... the raw oscillators are obviously fed directly into the filter, and from there into the two VCA's after that... maybe it's a combination of the oscillators and the filter with resonance... i read in many Sequential manuals, that you CAN clip the Curtis chip with high resonance... so maybe the VCA's inside the chip can get driven too hard from the filter... or maybe if the output from the filter, when it's too low gets amplified and distorted in the VCA's somewhere? ... in other words, I do not know... but it would be fun to experiment a bit with the signal flow of the Curtis chip some day.

We haven't quite defined the meaning or cause of the "harshness" in this thread, so things are a bit vague.  What I was referring to above was a harshness that was the result of slight distortion.  Because I use layered patches all the time, distortion is a serious and constant problem for me.  High audio levels, distortion, and a fully-open filter are a recipe for musical disaster.  But again, I do think a full open filter can still be made to sound pleasant, rather than harsh.

That’s why I was asking for a definition of harsh earlier on. One man’s harsh is another man’s super saw. 

Speaking of saws...I wonder if the perception of harshness would change if the default wave were triangle or square...

[Sacred Synthesis]

One man’s harsh is another man’s super saw.

Masterful.  This is a proverb that deserves neon lights. 

[AlainHubert]

A super saw cuts both ways... Some like it a lot, some not.

[synthboy]

Thanks. Yes. I’ve played the same 5 notes chord first on the Rev2 then on the OB6.
What, you hear harshness on the OB6 now? I can clearly hear some on the Rev2, and the OB6 sounds much smoother to me...

Had to listen again (using Sennheiser Momentum Wireless headphones) and yes i still think the first sound is more smooth and muffled. The second more harsh and metallic. Seems that more than me have the same opinion.
I would prefer the first sound playing solo but the latter in a mix.