Is it just me or Serum/Vital sounds without life?

[JO512]

It might be interesting to have a discussion in which we explore just what it is that makes a sound seem cold versus warm. Just to throw a few ideas out there, I think brighter sounds tend to seem cold and cutting while darker sounds tend to seem warmer. Fuzzier, dirtier, more saturated sounds can also tend toward warmth. If each note sounds almost perfectly like the last, as if you are replaying a sample, that can tend to sound much less alive and more monotonous.

For a warmer sound, I'd advise avoiding sounds that are excessively bright. Also make each note sound different from the last with some kind of variability or movement. One simple thing that can add some movement while also adding a bit of fuzziness is to combine slightly detuned oscillators so that you get some phasing that is out of time with the beat. You can also add a bit of subtle pitch drift. You might use LFOs to add a little subtle wobble to various things with modulations at both low and high frequencies, maybe preferably with some randomness. Adding a touch of noise can help give a sound that warm, fuzzy quality. Make sure the noise is filtered along with the oscillators. By default in Serum, the filter switch for the noise OSC is turned off. PWM changes can add something too.

You can make consecutive notes less perfectly alike also with velocity based modulations, note-on randomness, free-running LFOs that don't restart with each note-on that are not in sync with the beat, and so on. It can help to vary envelope parameters a bit. One of the things that makes a lot of electronic music sound so cold and lifeless to me is just that the envelope is so much the same from note to note, even with an old analog synth, which is very unlike traditional acoustic instruments.

I like to sometimes use very subtle FM or AM or noise-based modulations of things like oscillator pitch, filter cutoff, resonance level, the degree of effect of an envelope, parameters like the attack or decay of an envelope, and so on.

Harmor has this feature where you can make random alterations to all the partials with each note. I noticed that this can make it sound very much more similar to an acoustic string sound than any pluck that has the same distribution of partial levels with each note every single time. You can achieve something somewhat similar by making a wavetable in Serum in which you have something like a saw in every frame, but in each of the frames, you vary the partials slightly and also randomize the phases of the partials. Then you can do a note-on random selection of a wavetable frame with each note, so each note has slightly different partial levels and phases for those partials than the last.

And of course, various kinds of distortion/saturation can help. Reverb, IMO, is essential too, and has a much bigger influence on the perceived goodness of the sound than many realize, I think. A completely dry saw-wave is pretty hard to listen to. But butter it up with even a small amount of good reverb and it is night and day. It really helps to make it sound like it is in a room, especially if you are listening with headphones. Sounds sound more "real" when they sound like they are in an environment, since all natural sounds we hear are in an environment of some sort, unless we are in an anechoic chamber, which is notoriously uncomfortable to be in.

If you just use the init patch in Serum with a single saw oscillator and a basic filter and that's all, with no distortion or anything, and you don't add any movement anywhere, yes, it will sound rather lifeless and too clean. Each note will sound like the last. And it will have very little character. But that's even true of an actual analog synth. But with analog, it won't be absolutely perfectly the same every time, and you have a bit of noise and whatnot.

Movement. A bit of darkness. Noise. Randomness. Distortion. These are helpful in creating warmth.

[JO512]

Also, realize that distortion does something quite different when it is placed after a mix of notes or oscillators as opposed to before. I really, really like that in Phase Plant, there is a button at the top of each effects lane that says "poly". If you turn it on, and you have distortion in that lane, then that distortion is applied per-voice, and not to the mixed combination of voices. Try comparing playing multiple notes together with that button on versus off. Distortion placed after notes are already mixed will strongly emphasize the "beating" patterns in the sound created by the interference between the waves. It "sharpens" them up. Putting distortion after a reverb is also very different than placing it before, for the same reason. The distortion in Repro-5 is per-voice, which can change the waveshape. But you can't get that dirty, distorted interaction between the notes with per-voice distortion. For that, you need to add a distortion after. It is nice to be able to do either/both. Serum 2 now has per-voice distortion, which version 1 did not.

[djanthonyw]

No. I’m not just going by the presets. I’m using the demo from the initialized state and tweaking as well as listening to countless soundset demos from different sound designers, and watching countless youtube videos where it’s the synth people choose to make all their sounds.

As another example, David Guetta generally has great sounding mixes but he has a remix of Shouse “Love Tonight” that I find the lead sounding harsh and piercing. I found that he has a video on youtube of him making the remix and sure enough he uses Serum on that lead.



And it’s not just because I love analog emulations. A Virus, Nord, JP8K, Sylenth, etc do not sound analog but I also think they sound amazing.

Sure, Serum has the ability to make some acceptable sounds that sound ok or even good in some cases, but if you made those same sounds in other synths they will sound a lot better with some life and character, minus the harshness.

Sorry, Serum sounds bland and lifeless.

[enCiphered]

Serum has always been "transparent" by design.
It gives you a blank canvas. It doesn’t impose vintage coloration or analog imperfections unless you add them. That’s flexibility. You want drift, noise, nonlinearities? Use warp modes, LFO modulations, filter drive, FX section, or external processing. It´s everything there, use it.

Calling it lifeless probably says more about how you're approaching it than about what the synth is actually capable of.
If you prefer baked-in analog coloration, that’s valid but it's not an objective flaw in Serum. It’s a preference. Many professionals use Serum for its surgical precision and then shape the tone to taste.

[djanthonyw]

Yea, I get that point. However, even if you program sounds on serum to add movement and saturation as an attempt to add life, it’s still not going to sound as good as other synths by comparison.

[writheuk]

How many hours has it taken you to come to the conclusion that Serum is "lifeless"?

[JO512]

As another example, David Guetta generally has great sounding mixes but he has a remix of Shouse “Love Tonight” that I find the lead sounding harsh and piercing. I found that he has a video on youtube of him making the remix and sure enough he uses Serum on that lead.

The sound that you are pointing out here is piercing by design, it seems to me. He apparently wants it that way because he likes it. There is nothing about Serum that forces you to make sounds like that. Such sounds are a possibility that Serum offers, not a necessity.

[JO512]

... in other synths they will sound a lot better with some life and character, minus the harshness.

I made a patch in Serum to see what you think. I applied some of the suggestions I gave. I am curious if you think this has the qualities like harshness you object to. There is a lot going on in this sound that I wouldn't normally do, but I wanted to make it vary quite a bit so you can hear some variety in it, to show some range. There are two files with different arps. The MIDI sequence in the first arp is something I stole from a Starsky Carr video demoing a synth which itself was taken from a pack of things by Deadmau5 that was on Splice, if I am not mistaken. The second is mostly to show the sound in both low and higher registers.

The patch is all Serum 1, including the effects, which consist of a bit of single-band compression and some reverb and delay. I didn't add distortion here or a phaser or flanger or anything.

https://soundcloud.com/user-260326741/s ... al_sharing

https://soundcloud.com/user-260326741/s ... al_sharing

[JO512]

As another example, David Guetta generally has great sounding mixes but he has a remix of Shouse “Love Tonight” that I find the lead sounding harsh and piercing. I found that he has a video on youtube of him making the remix and sure enough he uses Serum on that lead.

Also, in the video you linked to, it sounds like it might be the case that what we are hearing is the audio coming out of his laptop speakers! There is no low-end and it sounds thin and tinny. I looked for the actual audio of the track he made and sure enough, it sounds quite different.



I think that lead sound is pretty nice, honestly. And judging by the comments, most other people seem to like it too. Seems like a pretty popular remix. 24M views. Raving comments. This isn't normally my kind of music, but I think it sounds pretty good, actually. Not sure what is bothering you about it.

[djanthonyw]

As another example, David Guetta generally has great sounding mixes but he has a remix of Shouse “Love Tonight” that I find the lead sounding harsh and piercing. I found that he has a video on youtube of him making the remix and sure enough he uses Serum on that lead.

The sound that you are pointing out here is piercing by design, it seems to me. He apparently wants it that way because he likes it. There is nothing about Serum that forces you to make sounds like that. Such sounds are a possibility that Serum offers, not a necessity.

No, it's just the way Serum sounds. Generally you can have a sound that cuts through the mix without sounding sterile and harsh. As mentioned above, I'm not just solely talking about that one example. I said that I've used Serum, not just demoing presets, but starting from an initialized state. Pretty much any soundset demo from various designers sound lifeless too.

Here's examples from the same sound designer:

Serum


Pigments


Serum sounds like a basic soft synth from 1998 while Pigments sounds like hardware.


Here is an example of the Love Tonight lead I made that does not sound lifeless and brittle:

https://www.dropbox.com/scl/fi/1g2ktrry ... j4dsn&dl=0

I listened to your examples, and of course they don't sound harsh because the filter is cut so much, but I don't think they have any life in them.

And of course I'm not referencing just the sound from the video of him demonstrating the remix. I have the full quality remix he made, not just what is coming from his speakers.

[Iores Pyadzee]

I don’t like Serum as well, however the creativity is fine.

[djanthonyw]

Yes, the GUI and features are very good but none of that matters if the sound isn’t there especially when the CPU usage is unjustifiable considering how bland the sound is.

[writheuk]

Yes, the GUI and features are very good but none of that matters if the sound isn’t there especially when the CPU usage is unjustifiable considering how bland the sound is.

I hope you didn't pay for it...

[djanthonyw]

I demoed it.

[mjolnir]

To my ears, a lot of the new additive synths could benefit from an additional resonant multimode filter section with ADSR.
I know that the additive algorithms can emulate filters, but I still think they need something extra.
The only additive that I could really groove on was Image-Line's Harmor, but the interface was too intimidating.