Are we close to getting plugin compressors that sound like real ones?

[tetsuneko]

I kinda like the UAD Neve comp and the new 76's collection.. Sounds good enuff for me, but then again I'm not really a compressor nut.. So IMHO, we're already "close", whatever that means..

[poonna]

I've tried everything, have you tried them? I think it's essential that people compare things themselves and not care much about what anyone says (including me), I don't like sharing negative opinions on specific plugins unless there's a really good reason.

Yes, I've tried both, and was really impressed with them, even more so than a lot other stuff I have tried before. I haven't had the fortune to own a hardware compressor though, so I have no base to compare to except between plugins.

[jonahs]

One could argue that 10-15 years ago plugins were "lacking" on a number of fronts. Developers were trying and we were happy to support their efforts. Fast forward to now and we can appreciate the improvements. Where might we be 10-15 years from now? I imagine that plugins will be doing really cool things and not necessarily copying hardware equivalents, but going off into new directions. I can hardly wait. Still, developers will also get closer to capturing the essence of hardware units for those that want those qualities.

I'm not sure. One thing to keep in mind is that colored hardware is a "failure" and the engineers making do with the best they had. Software doesn't have that "advantage" so it has to know in exact details the underlying process of the errors and this is a much more complex area than just math.

The irony is that the failures have the potential to give people what they didn't even know they wanted. It has to go out of the engineer's hands and to the musicians ears (could be the same person) to be evaluated in a sonic context to hear if it works.

The other thing is how much expensive research is needed into the underlying physical properties and when you combine this with the shrinking music market and low cost "good enough" emulations of existing tools, I'm not entirely optimistic.

What I think what we will see is mix and match genetic re-combinations and remixes of what already exists and in practice this will provide a variety of new sounds.

[jariya]

Hi Jonahs, Nice comments. Thanks.

I'm not sure whether I should introduce a new question/focus on this thread, but it's related so I'll just lay it out there.

I think the consensus is: No, where not quite there but getting close, and the tools we currently have are "good enough," especially when in the hands of high-caliber mix engineers.

But, if we're not there, then WHY NOT?

Perhaps it's one of the following and/or a combination thereof:
• We don't fully understand what's really going on, so we're still learning, putting all the pieces together (from an understanding of human hearing to the physics(?) of sound)
• We've pretty much understood the processes for some time now, but the platform and technology are the constraining factors: cpu, memory, throughput … I hear the developers saying: if only the computers were faster we could build amazing things.
• We can do it, but don't want to because it's just not cost-effective. The resources (time, money, opportunity cost) it would require to do it right are prohibitive and the return on the investment won't cover the cost.

I also find it somewhat curious that no "name" developers have chimed in (at least openly) and offered a good explanation about the state of plugin development.

[Mutant]

We've pretty much understood the processes for some time now, but the platform and technology are the constraining factors: cpu, memory, throughput … I hear the developers saying: if only the computers were faster we could build amazing things.

No.

Sure we know what happens inside.

A hardware compressor is a very simple system if we compare it to an analog synthesizer.

And we have VSTi plugins so good that only bats and elephants would be able to tell the difference in a full mix.

There is no such black magic in compressors that would make it impossible to emulate them with our current technology.

[FabienTDR]

We've pretty much understood the processes for some time now, but the platform and technology are the constraining factors: cpu, memory, throughput … I hear the developers saying: if only the computers were faster we could build amazing things.

No.

Sure we know what happens inside.

A hardware compressor is a very simple system if we compare it to an analog synthesizer.

And we have VSTi plugins so good that only bats and elephants would be able to tell the difference in a full mix.

There is no such black magic in compressors that would make it impossible to emulate them with our current technology.

I think jariya is right on point with his comment.

Just a simple example, let's just have a look at a simplified compressor: The clipper. Even with today's extreme computer power, it's not possible to create a digital clipper that actually behaves like its analogue counterpart and computes in reasonable time - and it will never be (sure, it will improve). The main problem relates to the limited bandwidth of digital system, i.e. the creation of aliasing. Digital compressors (and many other processors too) still need an insane amount of (lossy) approximation and tricks to work in real-time at a reasonable quality. Even with today's possibilities.

The problems are usually well understood by the developers (it would be ignorant to say the opposite), IMHO it's really just about the technical constraints.

[jariya]

Thanks Fabien! Much appreciated.

[Mutant]

Compare Minimoog and Fairchild 670 schematics to see how simple a fairly complex famous vintage compressor is when compared with a famous simple 3 oscillator monophonic synth.

I think we are talking about different things here - i mean emulating the circuit to the point of being unable to hear the difference between software and hardware in a full mix and with human ears because thats a reasonable border beyond which we can say that the emulation is perfect.

An average fully usable and used by thousands of pro recording studio enginers around the world every day compressor plugin takes just about 5% CPU maximum and you are saying that for your purposes the perfect compressor plugin would have to be so complex that it would use more than 100% of a modern CPU to sound good to your ears ??????????????????

Come on ! You can't be serious here.

[DuX]

thats not really true.. take some 10 years old compressors like waves c1 or rcomp / rvox or a eq like the waves Q 10 ..gui are old but it would be re released today with a nice gui and a new name i bet everyone would get fooled.

what evolved the most is how good looking the ui are.
sound get better very slowly ,compare a plugin like the old sonalksis eq and equality for example , i can't really hear any difference that will really improve a mix

Exactly. What I'm worried about is that in the next 10-15 years 95% of the code will be GUI, since DSP code in itself shouldn't be big. People are fooled by simple GUI tricks into thinking that this new great looking compressor is the end of them all... but what's underneath is still not more sophisticated than the code from 10 years ago.

See, I still like MJCompressor, for instance, very much. It's great. I listen with my ears, not eyes, and I use VST-analyser to check my findings. However, if I can point out some great plugins that were released lately, I'd definitely point my finger into the Klanghelm DC8C direction. Very nice behaviour on that one. When I close my eyes and tweak it, I can really pinpoint the settings nicely. I would still prefer the GUIless plugin without the excessive 2MB or more of unneeded BMP graphics, though, but OK, I can force myself to close my eyes and listen. This way of mixing has made a world of difference for my mixes. A world of difference. You don't even know how stupid our brains are and easy to trick.

You know, a 100 plugins with nice 2MB consuming GUI use 200MB of memory. 100 GUIless plugins use much, much less. So, I don't need to go 64bit any time soon, since the progress is mostly in looks, and less in quality, but people... are people.

The problem is that huge majority of people doesn't understand computers, and majority of people still think in terms of hardware. I don't. I've been with the computers since times when we didn't have any fancy GUIs. Plugins were completely bare. Waves changed that with their GUIs [and no better quality] in the late 90s, and started this whole GUI makes a plugin sound good shi*. It's just a picture ffs! Hardware devices tended to sound different because they WERE DIFFERENT, but a different GUI doesn't make a plugin sound any different!

Open your eyes, kids. Or rather CLOSE THEM.

Cheers!

[@midnight]

I realize this is a difficult question to ask and to answer. Let me try to explain. In terms of EQs, it seems (to me) that many musicians / engineers who contribute to various forums support the general view that developers have gotten really close to matching their outboard equivalents. But, not so with compressors …at least not yet.

One reason that I bring this up is because of the good work that is being done regarding console and tape emulations (UAD, Toneboosters, and most recently SlateDigital's VTM). Given the more thorough analysis behind our understanding of how such real world equipment works, why our ears appreciate the qualities they impart to sound, and improvements in algorithm design, it should be no surprise that "better" plugins are the result.

Also, I realize that some of the quality differences can be appreciated only once the entire signal chain is upgraded as well - better converters, speakers, etc.

I've also noticed new developers appearing on the public scene with their take on compressors, and meeting with positive reactions. I've tried several of the latest offerings, but I'm not yet convinced. I have too many already, and I'm a bit tired of being disappointed. Let's say I'm in the market for a quality plugin compressor that "feels, behaves and sounds" like it could exist in the real world. Now, I know that I'm treading into very subjective territory, but the analogy with EQs is there and one day should work in this case.

So, are we "almost there" with compressors? One plugin that received some good feedback about a year or so ago was the Vertigo compressor. Any others? Perhaps the new version of the UAD 1176 (although I can't test it as I no longer have access to a UAD card since I'm only working with a Windows laptop)?

whats so funny

Well, if I may be so frank.

I think it's hilarious that the old "do my plugins sound as good as hardware compressors" is still being raised.

It doesn't even matter. Seriously.

[Trakstar]

It matters to some people thus the point of this thread, and all you have done is rub it in thier faces so what are you some kind of know it all or something? And these are Not the old plug ins we are disgcussing,Softtube CL2B and FET compressor really hits the marks.As spoken by thousands of professionalsmake people look like you an absolute joke you pompous prick

[Uncle E]

Waves API 2500 compressor is the bee's knees.

It is and it also sounds different from the hardware it's emulating.

[Uncle E]

I wasn't necessarily looking for an exact plugin replicate of a specific hardware unit, only wondering whether current advancements in the development area have allowed us to get really close to the behavior, feel and sound of real world units.

We did a shootout of the Alan Smart C1 and Rupert Neve Portico 5043 hardware compressors against the UAD Neve 33609 software compressor. In that test, each unit sounded different from one another and none of them sounded clearly superior to the others, they all sounded like amazing compressors!

Similarly, I was using the $6,000 Shadow Hills Mastering Compressor for a project and found that I was able to cop its sound using a UAD LA2A into a UAD 33609. The LA2A couldn't be hit quite as hard as the Shadow Hills' optical section but the 33609 absolutely nailed its VCA section. I ended up using the plug-ins for that project because they were cleaner.

[Uncle E]

thats not really true.. take some 10 years old compressors like waves c1 or rcomp / rvox or a eq like the waves Q 10 ..gui are old but it would be re released today with a nice gui and a new name i bet everyone would get fooled.

I think most compressor plug-ins suck, no matter how nice their GUI's are. Furthermore, some of the compressor plug-ins that I think are excellent do not have photo realistic GUI's.

[Uncle E]

I tried the normal demo and it blew the analog eq out of the water so Im guessing if you really want to sample a"Proper" analog EQ or compressor your going to have to be ready to spend at least a good few hundred quid, more so with the Equalizers anyway.

There are some exceptions, such as the Golden Age stuff, but you're right for the most part. When us geezers are reminiscing over hardware, we're usually not thinking about ART or Alesis.