Fabrice Gabriel is well known and respected in the audio/music industry for the plugins that he designs with Slate Digital, which he co-founded and is the CTO off, as well as his own company, Eiosis, that he founded in 2012.
Fabrice and Steven Slate made a name for themselves when they introduced the FG-X to the market back in 2010 and with their highly regarded tape machine, VTM, and their many other high-quality, and affordable, plugins since.
In this interview, Fabrice talks about the painful experience with their first release, the FG-X, the unusual things he discovered about tape machines when working on the Virtual Tape Machine, Slate vs Acustica Audio, how being creative is to be organised and so much more.
Could you tell us why you went to AES back in the day and how you met Steven Slate there?
I went to AES with my company, Eiosis, to show my product and to meet new people. Steven came up to talk to me because he knew my plugins and he was struggling to find someone who could work for him. We shared some ideas between us and decided to start working together.
The first product you made together was the mastering plugin, the FG-X, how was the experience of developing that first plugin and what role did you specifically play back then?
The experience was painful and difficult because this was in the early days and I was designing all the algorithms myself, now I can’t because we have too many products. Also, right after the release, we had to go back to modify it.
How was the first initial reception when it came out?
It was really good but it was also quite controversial because we tried to bring an easy process to a market that didn’t necessarily wanted it, so, we got a lot of criticism because of that. Then there was also some issues at the beginning because it was a very innovative plugin and there were some situations we didn’t anticipate.
We took some heat and some people were a little pissed that we released that plugin. Even some mastering engineers didn’t want to use it because they thought it made it a bit too easy.
We didn’t know that at the time, but when a plugin has a lot of success, there will be a lot of people criticising it and that’s what we learned happens after a successful release.
Both you and Steven have a drive for perfection and to only release something when you are 100% happy with it. Could you tell us your process and how you achieve this “perfection”?
Nowadays there’s a need to have a process because there are more people involved and I never thought about our process until recently when I had to analyse it because I had to teach it.
The process is dependent on what you do but if you want to make the best products, for example, hardware emulation plugins, you have to keep asking yourself, “Does this sound the same” all the time. The more you listen the more your ear gets refined so you can hear more, subtle, things. And if you have the ability to focus that will come pretty quickly.
Also, you can’t be lazy and assume that people will not use a plugin in a certain way, or that it’s good enough, people will find out. Our main job is actually to figure out and sell products that can save people time because they will not have to figure out how to use some things that we spend a tremendous amount of time figuring out for them.
When you worked on the Virtual Tape Machine, you said that you had to understand what the magic was with tape machines, did you ever find out, scientifically, what it was?
Of course, it’s only math so there is no magic, however, the tape machine does some stuff you would not expect it to do and if I didn’t look precisely at what it does and how it does it, I would never have the idea of doing that kind of process to audio. That’s the magic, it’s many instances of unusual audio processing. Very strange and unexpected. So, when you know what it does you don’t want to process your audio with that, but it was very interesting to study it because now I understand why it sounds good.
Any specifics you could tell us of what that weird stuff was?
There’s a kind of saturation that is produced by the combination of the tape, the bias and the way the signal is reading of the tape and the recording head. Everything is participating to that sound. The saturation is very strange, for example, the high frequencies start to saturate at very specific moments in very specific conditions with very specific frequency conditions. There’s a lot of phase and phase issues too and these things would not be interpreted as very positive in any situation.
For example, you wouldn’t take the phase and rotates it very violently and very strangely and expect that to sound good.
But this is the sound of a tape machine and I think people back in the day tried to reduce these effects as much as possible and I think the difference between tape machines that sounded good versus machines that didn’t, was in their separate ways of handling those inherent issues and make them sound musical.
The interesting paradox here is that, back in the day, the engineers tried to reduce those issues but we are trying to repeat them.
Why do you think your emulations stand out from many other manufacturers out there?
I think we stand out because we spend enough time on it and we try to go all the way, but, of course, there are very good manufactures and plugin developers out there. Also, back in the early 2010 people didn’t care too much about accurate replicas of analogue equipment so I think we set a trend by reproducing it more precisely than before.
For example, with the tape machine, the difference was to make sure that everything was properly done, for example, we didn’t ask ourselves the question, “Is this something we want to see in an emulation?” instead, we asked ourselves, “Is this feature present in the original hardware nor not?”. This was the only thing we cared about.
The same approach went into making the VCC where I wanted to make sure that I couldn’t tell the difference, both scientifically and sonically, so I pushed the envelope and made sure that all the tiny details of what I saw were faithfully reproduced.
If you compare that to other products that came out then, for example, emulations of analogue EQ’s, they were usually just approximate curves but there’s so much more to an analogue EQ’s sound than the curve. However, at that time, other developers didn’t care about the frequency response nor the harmonic or dynamic aspects of the circuits, but it was fine for everyone because they still had their analogue equipment.
There’s a YouTube video where Slate Digital’s algorithmic approach is compared to Acustica’s Volterra Kernel technology, what is your thought about these two technologies, do you prefer one over the other?
It’s just different, the Volterra approach is limited in a way and our approach is limited in another way. The limitation of Volterra is that the resolution is limited, for example, it can only emulate a certain number of harmonics and that’s it.
They did a great job in developing their technology but they can only go as far as their model can go, which sometimes can be efficient but sometimes not. What’s great about their method is that they can be very quick with measuring and reproducing anything. However, the level of detail is difficult to know because the relative detail depends on what the hardware does.
On our side we take a different approach, we make everything by hand, we compare every extreme case and have a much more details approach, therefore, it’s a much longer, more difficult and a much more expensive process. We also make sure every single plugin sounds the same as the hardware and that the waveforms will be the same, which is not the case with the Volterra Kernel. For example, you can analyse the same waveform and in most cases, it should look the same but in extreme cases when there’s heavy saturation or clipping, it might not look the same. But again, they don’t spend that much time analysing those things so they can also release plugins much quicker.
Check the mentioned video below.
Is the discussion of digital versus analogue still relevant?
It’s been 10 years since we started this discussion and I hope that by 2020 it will be over.
Having said that, the difference that I try to make comes from the fact that if we promise an analogue emulation, it will be an analogue emulation. This was not the case a few years ago, for example, back in 2005 when I was analysing some plugins and compared it to the analogue hardware, I went, “What the fuck, this is not an emulation”
Also, I was always struggling with EQ’ing vocals in-the-box, but once I tried a piece of analogue hardware, it just worked. This also made me realise that, back then, the plugin manufactures were lying to us and really didn’t respect us. If they would have told us that this was a close replica of the hardware, it would have been fine.
I think it’s the marketing team that wants to say, “This is the exact same thing, we spent five years of research and development, etc.” which is obviously a lie. I don’t like lying to anyone and specifically not to customers because if people are giving you money for a replica, let’s do a replication, let’s do the job.
My motivation was always to be faithful and to deliver something you promised rather than to satisfy people.
What does the relationship between art and science mean to you?
If you are only a technician you cannot thrive, you have to be imaginative and creative to find solutions - that’s the artist in you. The relationship between artistry and to be a good technician/engineer is very important and needs to happen even more because we do tools for artists.
For instance, there are some companies where everything is dived, for example, the algorithms, the controls and the GUI all get developed separately. What we try to do is to integrate those processes across the team because it’s very important that everyone is included in the artistic process.
That’s why I’m trying to teach our team to participate in all the processes and to understand the needs of our clients, and be able to listen to the algorithms. I’m also trying to make sure that ideas flow within the team rather than just being a transactional thing. This, I believe, will make better products.
Also, what I learned is that if you compartmentalise too much then the products become tasteless, therefore, the relationship between the user interface, the algorithms, the look and feel, the graphics, the way that you move a button, are very important. I’ve been pushing the quality of the graphics a lot and I think I have made several graphic designer cry because of this. Everyone has to understand the job of the other for them to be able to create something better, technical speaking and artistically speaking.
Once again, when I tried some other plugins back in the day, they would tell me that it sounds great, but they seem to spend two days on the graphics. I was opposed by that so I always wanted to reflect the work we have been doing on the algorithms for months on the graphical user interface.
Could you expand on what sort of things you are teaching them when it comes to creativity?
A few years ago some people in the company, who are gone now, believed that creativity was magic and that you are either creative or not, which to some extent is true. However, I tried to remove that aspect of creativity and magic and that there are, in fact, some processes that lead to creativity. Some people are more or less sensitive to those processes but you can teach them and that's what I’m trying to do.
So, being creative is being organised, perseverant and having a set of methods to think about when you start to work on a project. Sometimes you start working on a project and think you will add this or that because some people have asked for them, however, you always have to go back to some core questions, and that’s the core of creativity, such as “why do you do this or that, who are you developing this project for”
It’s the basic idea for all artist, like a singer who sings without intention or with the wrong intention, it will not work. Same with products, you have to define the intention and the why. Once this is defined everything becomes easier and you can start to develop a creative process on top of that. I believe that most people are creative but most of the time they need to be structured, which might sound counter-intuitive, but creativity is better when there are more constraints, for example, if you tell a producer or musician to do whatever they want, usually it will be a big pile of garbage, instead, if you give the producer some constraints such as, "I want a song with this tempo, this length, this sound and this style” the producers will now have to express his or her creativity in another, more restricted way.
That’s when people become really creative, rather than allowing them to do whatever they want. In order to be creative people needs structure and the intention behind it.