Software vs. Analog in 2025 – Has the Balance Shifted?

[zerocrossing]

This might be fun. Which is using analog which is using VSTs?





Breakdown reveals:

The first one made me want to teach a bully a lesson more than the second one.

[IvyBirds]

Hallelujah!

What's interesting about that video is that Anthony Marinelli is most famous for his work with the Synclavier II. He calls it "The Instrument that Changed his life"



The Synclavier was a Software Synth running on a computer, it had a dedicated controller called the ORK that wasn't MIDI as it predates MIDI, but may as well have been as all it sent digital code to the CPU where software turned it into sound. The controller also had controls for the DAW that was built into the Synclavier mainframe CPU

Everything we use in a modern DAW from plugins for sound, along with sampling, multitrack sequencing, and digital multitrack recording existed in the Synclavier

I am glad he is featuring Hainbach. Hainbach has made some pretty cool plugins and likes to do crazy ambient experiential stuff using bizarre cold war era electronics and tape, he also likes to use digital instruments the OP1 from Teenage Engineering for such things, along with plugins



I love doing crazy experiential stuff with sound and digital technology and software has made that possible

[pekbro]

Maybe you should share some of it, so we can see wtf you’re talking about.

[vurt]

never gonna happen!

[Funky40]

Hallelujah!

yeah, from 5:44 on one can see why patchcables in modulars do make sense.
(it´s not a modular here, well, to some parts,....but cabled anyway)

the idea of modulars without cables would work for something small. But not for something bigger. A "off live" idea. just sayin

[IvyBirds]

Maybe you should share some of it, so we can see wtf you’re talking about.

What would be the point? It's experimental and not really intended to be for public consumption plus the Internet is already full of synth fart sound

In software form


Or hardware if you prefer the presence and life that only hardware fart sounds can bring



But seriously when I am in my home studio I like to mess around and do all kinds of weird things just to see what they might sound like, it's not really musical it's experimental. Spent some time today in my yard making field recodings of the Cicadas in my maple tree. Going to try to see if I can turn that into a playable instrument and if I do it will sound nothing like a cicada in the end, that's the experimental stuff I like to do

Maybe I am successful, maybe I fail. In the end it doesn't matter as it's just me messing around and having fun enjoying Synths

If I want to seriously play some music I will just sit at my piano and play, I did that today also but I am sure most people don't care about classical pieces on the piano very much either

[SoftSynthLover99]

But do you make art only for the audience? Is every piece of sound / music you make only created for the entertainment of others?

The 3rd Wave was a snooze fest, so bye bye..

Curious what you didn’t like about the 3rd wave?! Everything I’ve heard from it sounds pretty amazing to me and I have that on my radar as a potential next synth. 24 voices of polyphony is very attractive as even most software that sounds worth anything doesn’t reach that high. And the overall sound to me seems very pleasing for a Hybrid!

[pekbro]

Fart sounds are always welcome around here

I would argue that analog farts are better than digital farts.

[ghettosynth]

The Synclavier was a Software Synth running on a computer.

The Synclavier II was a sampler. That's what changed his life, not the rudimentary FM synthesis. In context it was a ground breaking instrument, but, that it "changed his life" is not a win for software as a substitute for analog synths.

More substantial FM synthesis, e.g, Yamaha's approach would required hardware ASICS for a number of years. Most competing samplers in the early years had analogue filters. The real win for the Synclavier II was that it incorporated sampling and direct to tape. It was, in essence, a precursor to today's DAWs run with a very crude U/I. Of course that kind of technology changed his life.

The Nord Modular changed my life in a much less dramatic way, but it really helped me to refine my live shows.

In short, it's not really interesting that samplers were seen as ground breaking in the early 80s, of course they were.

[IvyBirds]

The Synclavier was a Software Synth running on a computer.

The Synclavier II was a sampler. That's what changed his life, not the rudimentary FM synthesis. In context it was a ground breaking instrument, but, that it "changed his life" is not a win for software as a substitute for analog synths.

Only it's not just "a sampler" it's an Additive Synth, with FM. The sampler was added later. Thanks to the Synclavier being a computer able to load new software

When it first got introduced it was Additive/FM only and that's what changed his life

If you watched the video and understand his body of work you would see that it's the additive/FM parts and you can see many demo videos of him programming additive and FM sounds that he knows like the back of his hand

And he owns a TON of analog synths, that he has had since the 1980s, but it was his work with software inside of the Synclavier mainframe that changed his life

You are right the Synclavier wasn't a substitute for his analog synths, they VASTLY outperformed them personally and professionally

[ghettosynth]

Related, it was software that allowed the Prophet 5 to exist. This was Dave Smith's real contribution at the time. The P5, all revisions, are largely datasheet designs with minor tweaks with respect to the analog side. Dave, however, got his start building sequencers and was good with digital and so jumped into microprocessors. The real win with the P5 was the voice assignment code that Dave wrote. Once that nut was cracked. All polysynths of the day used the ideas that Dave pioneered in the P5. That is, all analog polysynths required software to function. Even the controls are sampled and stored, again, all digital electronics that is necessary to provide cost effective patch storage. You cannot make a reasonable analog polysynth without software, you can't make one at all without digital electronics unless you make it fully polyphonic.

It's never as simple as "software" vs "hardware." Software provides a vital role for most synthesizers since the early 80s. Even many monophonic ones. The revered TB-303 cannot function without its microprocessor. All of the CEM3387 series synths (even the modern ones) required the processor to apply a wave shape correction voltage that's computed from pitch. The dominant technology that has caused Eurorack to blow up is software. Many of the really popular modules are either partially or fully digital.

None of that has any bearing on the question in this topic. It's even subtlety wrong to frame it that way. The issues arise from the limitations of models. So, even though we can think of FPGAs as digital designs, it's still not the same as DSP running on an embedded CPU or even a desktop CPU that also has to run a consumer operating system. Sample rate and latency have real impacts on the quality of the model. One of the things that made the Nord Modular sound better than alterantives, e.g., Synthedit at the time was that it ran at 96Khz.

I'm not trying to diminish the P5. Most manufacturers went a similar route largely for costs savings. Some chose more subtle designs, e.g., the MemoryMoog did not compromise on the Moog filter even though it uses CEM VCOs, (at least some) VCAs, and EGs. However, the CEM based oscillator is more fiddly to tune and uses a less sophisticated tuning algorithm than others, e.g., the P5, again, Dave with the software. Some designs were truly, for lack of a better word, lazy, e.g., the Akai AX73. Oberheim started with discrete designs, but dropped that for cost reasons. It's why I really dislike the OBXa, it's just a CEM synth with a lot of limitations.

When we discuss analog vs models of analog, it always has to be contextual. Sometimes the variance is so small it doesn't matter, sometimes it's more significant. The context is detailed and technical. You can't just say that software wins because someone's album that people like used plugins. You also can't realistically do an experiment. The macro context is too complex. So we're stuck in this place where the Starskys of the world will play back raw oscillators outside of a dynamic context thinking that they're actually showing something that matters and the difficulty of evaluating full tracks in a context where it's nearly impossible to evaluate the question experimentally.

Which is why I am of the opinion that you will never convince everyone that the model is right. George box nailed it: "All models are wrong, some models are useful."

[IvyBirds]

But do you make art only for the audience? Is every piece of sound / music you make only created for the entertainment of others?

The 3rd Wave was a snooze fest, so bye bye..

Curious what you didn’t like about the 3rd wave?! Everything I’ve heard from it sounds pretty amazing to me and I have that on my radar as a potential next synth. 24 voices of polyphony is very attractive as even most software that sounds worth anything doesn’t reach that high. And the overall sound to me seems very pleasing for a Hybrid!

The 3rd Wave is the only hardware synth to come out in the last 10 years that interested me at all besides the Montage M and I got that to gig with thanks to it's keybed. I have been wanting a PPG Wave forever and this seemed amazing on paper

I spent a lot of time with one at Sweetwater in Indiana and was ready to drop serious coin for it and bring it home with me from Indiana

In the end I didn't because it's literally a plugin in a box with some analog filters bolted on in the end, and the analog filters are just kind of meh.

There is a digital filter modeled off of an SEM/Oberheim design that they don't like to advertise is digital, that sounds better but then you just have a Digital Wavetable or Digital Virtual Analog Oscillators running entirely in the digital domain that would be going through a digital filter model

It has a lot more modulation and envelope options than the original PPG Wave but again that's all software and still very limited compared to software options

It's literally software which is fine but it's not nearly as flexible as the vast majority of software Wavetable synts or VA Synths that cost close to $5 Grand less

I think it's great for the people who think it's hardware so it has to be be awesome, or for people who want to post pictures of it on Instagram

If you want a hardware PPG Wave get the Behringer one for $700 and save four grand, if you want something that sounds like PPG Wave in a modern plugin get the new Waldorf Microwave Plugin, if you want a deep Wavetable synth with tons of filter and modulation options there are so many phenomenal ones in software it's impossible to count

[ghettosynth]

Also, if it's not clear, I'm not saying that SSM/CEM chips were not excellent analog design, they were. It just wasn't Oberheim's or Sequential Circuit's design.

[ghettosynth]

Yepp, take that back. Then take this!

Whatever one thinks of this guy, or, thinks about marketing as an explanation, it still remains that there is a claim on the table that a model improved over time which required an update. The assumption is that the model is now, perhaps asymptotically, closer to the thing that it is modeling. Note that it doesn't matter whether that claim is true or not in this case. This is a common story and it has been true for many analog models. That said, while marketing may have had a hand in this, and they're definitely triggering your FOMO, it's very likely that there is some truth to the fact that the model has been improved. Here's a dangerous statement, the model can be further improved. Whether it's audible to most people in a mix is a different question, but it's very unlikely that my statement there is false. It is, in fact, a story that has repeated itself over and over. Artuira claimed back in 2004 that:

But what about the core of the Minimoog, the sound? Thanks to TAE®, Arturia’s proprietary technology for emulating analog circuits, the minimoog V offers an unchallenged quality of sound. Sharp filters, aliasing-free oscillators, soft-clipping guarantee you have the best analog emulation on the market.

https://web.archive.org/web/20040611002 ... oogv.lasso

Well, that's some slippery language there guvnor, no? It's certainly not claiming to be perfect, just the best. Yet, we know that it has been improved at least twice since then, and by improved, I mean remodeled. This necessarily implies that the old model wasn't close enough and there was room to get closer. Of course there was.

And yet, some of you were still claiming, at the time, that software had solved all of your woes over twenty years ago (emphasis mine).

viewtopic.php?p=555489&hilit=Minimoog#p555489

Now, its just simple, un-cluttered, and easy to get on with making music. Software synths sound better to my ears (yes that includes over-blown supposed classics like the minimoog!) and whole orchestras can be loaded into the computer. Best of all - I've been able to convert 30 square floor space into another recording room so I can record the best thing of all - real acoutic instruments.

Its never been better and my output has quadrupled.

Some of you couldn't hear it then because you were playing stuff where it didn't matter, even though at that time it was painfully obvious. It's definitely better than it was in 2004, but you might consider that you are that guy that couldn't hear it then updated to the 2025 edition.

[zerocrossing]

But do you make art only for the audience? Is every piece of sound / music you make only created for the entertainment of others?

The 3rd Wave was a snooze fest, so bye bye..

Curious what you didn’t like about the 3rd wave?!

...In the end I didn't because it's literally a plugin in a box with some analog filters bolted on in the end, and the analog filters are just kind of meh.

That's it, in a nutshell. My hope was that it would give me a better representation of the PPG type of wavetable reproduction than Waldorf's pretty good PPG 3 plugin. It just didn't. I posted some a/b comparisons of the raw wavetables and some presets on GearSpace, but no one seemed interested in making a guess which was which. They just wanted to hold on to their dream.

That's not to say that the 3rd Wave is a bad synth. Obviously it's not, but when I took PPG wavetables and put them in Dune 3, I was able to get results that were as good and often better. Aside from oscillator FM, the 3rd Wave's mod matrix doesn't update fast enough to do good audio rate modulation. Dune does a much better job. It's also got a lot more going on, more filter models, oscillator effects, great oscillator unison options, etc. You can even model voice card offsets using the mod matrix. Dune's a kick ass synth, and I'd put it up against most any wavetable in software or hardware, though Serum 2 has been really kicking ass lately, though it's also kicking my processor's ass as well. Multi/Poly's also a newcomer that's been very good.

So, while the M did really capture the sound of the Microwave, at least in that respect. The 3rd Wave is a failure at sounding like a PPG to the degree I was hoping for, IMO. The Behringer might do a better job, but around the time I was being disappointed by the 3rd Wave, Waldorf came out with their Microwave 1 plugin, and I decided that it satisfied by "vintage wavetable" slot, and I gave up on the PPG. I think the Behringer would bug me too much, with its limitations and that stupid interface.

Don't worry, I'm still a hardware club member in good standing. My Nina came with the PPG wavetables loaded up in it (though it doesn't really do that PPG thing.), and it's a kick ass sounding synthesizer. I don't know how to describe it, but there's something about it that I just love. The VCOs have a very distinct tone. I've never heard anything else like it. The filter is pretty Moogish. The wavetable oscillator is ridiculously hot, so you can really push the VCF and VCA.