In this episode, I talk about software plug-ins vs hardware. It is a question that I get frequently from listeners. It is relatively short.
Can a plug-in perfectly emulate a hardware device? What are the tradeoffs a plug-in designer needs to consider? What about a plug-in that has no hardware antecedent, but it something out of the mind of the designer? These are some of the topics considered.
Thanks for listening, subscribing, and commenting. You can reach me with your questions for a future podcast, or suggest a topic you would like to know more about. Send me email at firstname.lastname@example.org
85 Plug-ins vs Hardware November 22, 2023
I’m Doug Fearn and this is My Take on Music recording
Before I start this episode on Plug-ins vs Hardware, I’d like to take a minute to cover a couple of odds and ends.
First, this topic was suggested by a listener. I have a list of about a dozen questions and topics from listeners and I would like to include all of them in future episodes. However, some are fairly simple and can be answered in a couple of minutes. I plan to do a Q&A episode soon, covering multiple questions.
If you have questions you think I could answer, or topics you would like to hear My Take on, please send them to me. Email to email@example.com and I will add yours to the list. I probably won’t include some that are too specific, such as information on a single piece of gear, but otherwise, anything goes.
Second, as you probably know, this podcast is carried by almost all podcast providers, including Apple Podcasts and Spotify. If you already subscribe through one of those providers, thanks. But if not, it helps build my audience if you subscribe. Making this podcast is a lot of work, and I want to maximize the number of people who might benefit. So please subscribe. That way, each episode can automatically download to your listening device if you want, or you can get notifications and download the ones you want to listen to.
From the beginning, I have had a relatively small email list that I send notifications of each new episode at the time it is posted. I only send one email for each episode. If you would like to receive that, send me an email saying so and I will add you to the list.
I always enjoy hearing from my listeners and try to respond to every email I get.
OK. Today I want to talk about plug-ins and the hardware they emulate. I should tell you that I personally do not use many plug-ins myself, mainly because I have lots of hardware equalizers, and compressors that my company makes, plus a couple of hardware digital reverbs. So, I’m definitely not a plug-in power user.
But I do use some. I use a Flux Elixir limiter plug-in on the mix buss, after my usual VT-7 Compressor and VT-5 Eq. It doesn’t do much in my mixes, since I don’t like the sound of highly-compressed, very loud mixes. But the Elixir does provide protection against a few peaks that would otherwise reduce the overall level of the mix. A software limiter has some advantages over hardware, mainly because it can “look ahead” at the audio coming and capture those peaks in a transparent and music-friendly way.
I use Melodyne very selectively to correct off-pitch vocals and instruments – but only if I feel the note or notes detract from the overall impact of the song. And even then, I usually only correct the note to the point where it no longer bothers me.
Other than those two applications, I don’t feel the need for any additional plug-ins. I do not use eq or compression on individual tracks very much. I try to get the tracks I record to be as close to the final sound I want during the recording. That always seems to be a better way to do it, at least for me.
I should also say that there is a plug-in version of the D.W. Fearn VT-5 equalizer, which was created by Acustica Audio. And Acustica is working on a plug-in for my VT-7 Compressor. More on those in a minute.
I know that my workflow is a bit different from what most people do these days, so you can take my comments with a grain of salt. What I do works well for me, but it might not be best for you.
The basic question I get asked a lot is whether I think a plug-in can accurately represent the sound of a hardware device.
I think that at our current state of the art, plug-ins sound remarkably good. Do they sound exactly like the hardware? I don’t think so, but they are close and getting better all the time.
After designing the VT-5 Equalizer over a period of a couple of years, listening to it on a wide range of sounds, and tweaking it so that it became as good as I could make it, I have a very good idea of what it sounds like. After all, it’s the only eq I use.
So when plug-in designers started talking to me about the possibility of a VT-5 plug-in, I knew it was going to be challenging to get it perfect. Acustica came about as close as is possible, at least without having a dedicated computer just to run the emulation. That is always a trade-off for plug-in designers: how much CPU power can you devote to making the plug-in more accurate? It is a compromise. The VT-5 plug-in uses a lot of processing power to get as close as possible to the sound of the hardware, and that can become a problem for some people who might use several iterations of it in a mix, especially if they have an older, slower computer.
After a couple of years of refining the VT-5 plug-in, I was finally satisfied that it captured most of the essence of the hardware. Is it exactly the same? Not quite, in certain subtle details, but it is remarkably close. We went through many versions to get it to that point, and I am happy with the results.
Equalizers are one of the easier things to emulate, at least in their eq curves and phase shifts. For the Acustica VT-5 plug-in, called Ruby, the equalization matches the hardware in every way. Where there is a slight difference is in duplicating the sound of the vacuum tube electronics.
The difference becomes one of “feel” rather than measurement. Despite our ability to analyze and measure how a device performs, I believe there are still things we cannot measure, at least not today. I don’t even know how to describe those things. But when I listen, I can detect a difference in how the sound makes me feel.
Not very scientific, I know, but music is an emotional experience, so we have to let ourselves be open to how the music makes us feel. And the same applies to the process of recording that music.
The VT-7 Compressor emulation, on the other hand, is much more challenging. There is so much going on in a compression circuit that it takes a lot of computer horsepower to begin to duplicate it to any degree of accuracy. Will Acustica get there? I have confidence that they will. But it won’t be made available until I am totally happy with it. And the talented people at Acustica are doing their best to make that happen. The people involved can really hear! They are all great musicians as well as talented software engineers, and I know they can do it.
So those are my personal experience with plug-ins vs hardware, from a different perspective than the typical user might have.
But what about other plug-ins that attempt to duplicate real hardware devices? I have tried many, from a variety of plug-in manufacturers, and most are pretty good. I also encounter a lot of plug-ins that just sound bad to my ear.
Do they duplicate the hardware? Probably not, especially the ones that sound bad to me. The sound of most vintage gear has degraded over the years, even if they have been rebuilt. Modern electronic parts can sound different from those used 60 years ago. On top of that, those old products tend to all sound slightly different when you compare several different units, so the plug-in sound may not sound like another piece of the same hardware.
Does that make them unwise to use? For most people, no. If you are getting the sound you hear in your head out of the plug-in you use, that is an indication that it is doing the job.
I urge you to evaluate a plug-in not only by how it accomplishes what you want, but also by how it feels to you.
I’m sure every hit record you hear these days used a bunch of plug-ins. That did not stop the records from being very successful. So that might also be a criterion to consider.
Could the recording have been better using the hardware? Probably. But in the world of popular music, sometimes OK is good enough.
I am cursed with the need for perfection, and although I have never achieved that, I keep trying.
What would really help a lot of recordings is a simplification of the recording process. The less you torture the audio, the better it sounds. If you get the instruments and vocals to sound the way you want during the tracking process, a chain of plug-ins on every channel may not be necessary. Microphone choice, pattern choice, mic placement, the place in the room, the bleed from other instruments, etc. all shape the sound. You can get a lot of control over the sounds that way. I have covered this in the past, so I won’t add anything more about it today.
Classic hardware can be very expensive, perhaps 250 times more than a plug-in version. That is the deciding factor for many recordists. They cannot usually justify the tens of thousands of dollars for a classic hardware device, so the plug-in is the logical alternative. And there is not much to go wrong with a plug-in, unlike the constant need of maintenance to keep old hardware running.
And that’s true for new hardware, too. I know the D.W. Fearn products are very expensive. I wish that wasn’t the case, but to build no-compromise hardware, with 50+ years life expectancy, costs a lot.
Especially when starting out, many recordists have to use plug-ins to achieve the sound they want. Maybe someday they will be able to afford the real hardware, but in the meantime, plug-ins are a good solution. I can’t argue with that.
Plug-ins can be a cost-effective way to evaluate the sound of hardware you might buy later. I know our VT-5 hardware sales increased when the Ruby plug-in version became available.
Additionally, you can use multiple versions of a plug-in as needed, and not be stuck with just one piece of hardware. Although you can use the same hardware on multiple tracks if you want to spend the time to create a new version of it on another track, the convenience of a plug-in cannot be denied. If you have both hardware and software versions, you can use the better-sounding hardware on the most critical tracks, and the plug-in version on the others.
So far, I have only talked about typical outboard gear emulations – things like compressors and equalizers. Plug-ins can emulate other outboard gear, like reverbs, delays, pitch-changers, and other things, some of which are impossible to do in hardware.
Those types of devices, in my experience, are marginally good at their job, whether based on hardware or not.
I have heard some plug-in reverbs, for example, that sound remarkably good at first listen, but ultimately, they do not hold up when mixing. I prefer the old-fashioned approach of using individual sends from the tracks that need reverb, all sent to one hardware reverb device, with the return coming up on a separate fader. There is a cohesiveness of sound obtained that way the I find preferable to a bunch of different reverb sounds. Your experience may be different.
In my studio, I use two hardware digital reverbs: a Quantec and a Bricasti M7. In a sense, they are both essentially the same as a plug-in because they emulate the sound of real rooms or classic reverb devices like plates, or an acoustic echo chamber. The advantage, to me, is the versatility they provide.
However, most people use plug-in reverbs exclusively. I am sure your DAW has a way to route your echo sends and returns to achieve the same overall reverb sound as I use. You might be surprised at the difference.
I plan to do an episode on reverb next, so I will leave that topic for then.
Another interesting use of a software emulation of real hardware is when a hardware designer develops an interesting circuit as a one-off, and then it is analyzed and made into a plug-in. I have had several hardware product ideas that I think would be interesting to try, but some electronic components required are so rare that manufacturing is out of the question. There are many old vacuum tube types from the 1930s and 1940s that have very interesting properties. But they are rare. Classic audio transformers are another example of a way to create a unique-sounding hardware device.
At some point, I want to build a device using parts that are not readily available, and if it works well enough that I would use it in my own sessions, I would see if it could be made into a plug-in. Only one hardware version would be necessary, or perhaps even possible.
Another plug-in design approach is to create a device emulation from scratch, without any hardware antecedent. This is becoming more prevalent as designers’ imagination allows them to create totally new virtual devices, perhaps with unique properties. That could be pretty exciting to try, and maybe I will work on that at some point.
I don’t have the coding skills to do it myself, but I certainly know the people who could. The challenge is translating a sound in my head into a description that a software expert could interpret.
So those are my thoughts on the current state of plug-ins. They are an incredible tool in many cases. Their ability to emulate actual hardware devices is very good, but not perfect. There is a big opportunity to create software that could do things that no hardware ever could.
My only caution is to always use the minimal amount of sound manipulation as possible to achieve your goal. A chain of plug-ins sometimes makes the sound worse, not better. Capture it expertly at the tracking stage, and the need for further manipulation can be minimized.
This is My Take on Music Recording. I’m Doug Fearn. See you next time.