An Interview with Vojtech Meluzin of MeldaProduction
Vojtech Meluzin is the prolific software wizard who founded MeldaProduction and is its driving force. We find out more about him and his company and in this in-depth interview.
by David Baer, Nov. 2015
MeldaProduction is a well-known and highly-respected vendor of all manner of music production plug-ins. The number of offerings in its catalog and the rapidity at which new offerings are introduced are mind-boggling. But what’s all the more astonishing is that a single individual is responsible for all the design and coding of this extensive catalog of software. That person is Vojtech Meluzin (pictured in the selfie to the right, presumably with fiancé, while on a rock climbing outing). We get to know him in the interview that follows.
SoundBytes: In just a few sentences, tell us who is Vojtech Meluzin?
Vojtech Meluzin: My full name is Vojtech Melda Meluzin, thus the company name – most people don’t get Meluzin and I don’t blame them. I’m from Czech Republic, and most of the people I work with are as well. The development is all on me though, so I’m to blame for bugs and the geniality too. I’m pretty much in my mid 30ths, engaged, but to a “foreign” girl, so no wedding yet … you know, classic Czech bureaucracy. What else? Oh man, I have done almost everything. Right now I’m rock climbing. Just simple stuff though, I’m too old for climbing to K2, and there’s the cold as well. Anyway, in my past I have been hooked on creating electronic devices, paragliding, and I even had a chemical laboratory (and yeah … there were a few “complicated” moments 😀 ).
SB: So please tell us about your early musical influences and education. What led to your involvement with music production?
VM: Well, I’m kind of a multi-instrumentalist, but mostly a drummer – quite good, I hope. And interestingly enough when I was in college (before you ask – Charles University in Prague studying informatics) with literally no money at all, I wanted to create some music. I was doing it using a computer mouse, but despite being a drummer, I was really too lazy to create drum tracks like that. So I thought “well, maybe I should create virtual drummer”. And I did, a super-simple one playing just one loop. But I used it on three different school projects.
Then I quite forgot about it, but a few years a not-so-good friend came to me and started challenging my pride about not finishing anything and suggesting that we should sell it. So I said “yeah, let’s do that!”. I improved that thing quite a bit, as you can guess. No business came out of that guy, but since I already spent so much time with it, I decided to give it a go. I worked for quite a long time, it was years actually, day work and the evening split between MDrummer and girlfriend – they had to share the time. And eventually I finished it, put a website up and, well, nothing happened. I had absolutely no understanding of the audio community, how it works, nothing. I just had MDrummer. I started some really stupid ads at KVR, where they quickly banned me 😀 , so I thought like “whoops – and the whole thing is over? Just like that?”. Hard beginnings … but then I came up with a great idea. Since MDrummer has so many effects in it, I could release them, for free. And I did and it didn’t take much time and people came, unbanned me at KVR, and after some months, the first purchase happened, and I went for quite some beers after that. But it took another, say, two years before I was able to quite the day job and get all my income from my own company. Lots of work in it … but thankfully things have kept improving since then, and now my main problem is to find time. Too many ideas, no time … damn!
SB: So I take it you first developed your skill at writing software code at the university? At the time, what did you perceive your goals and future to be?
VM: Not at all! I actually started what people usually call “coding” some time in elementary school – I could have been something like 10 years old. At that time it was all about qbasic, pascal and, well, assembler, which is actually a very usable skill in audio. It’s not so much about using it, but more about understanding the architecture though.
Anyway, at university, like most of us, I didn’t know what my future was to be. This was at the time of the Internet boom, so most people wanted to design websites and that sort of thing. I personally hated that. I always liked the high-performance, low-level stuff, like audio. I just didn’t know that at that time. I studied operating system design, so I knew I was going to work on some high performance stuff. I always thought it would be something with search engines, artificial intelligence, something like that. But I ended up with audio and wouldn’t change it for anything. It’s just great when you can push the technology forward in something so close to you, like music is for me.
SB: OK, so that takes care of the software technology education. But the other big aspect of what you now do involves some very heavy-duty DSP knowledge. When and how did you acquire that?
VM: On the road, so-to-speak. Basically I didn’t have the knowledge in the beginning at all! It took lots of books, experimentation, searching … lots of work and experience is all it takes. But it’s fun to watch lots of developers out there that seem to have no know-how at all 😀 . That’s the market … hard to say what is good these days.
SB: At this point I’d like to ask a highly technical question, with apologies to any readers who don’t speak Geek. You have produced an impressive number of deeply functional plug-ins in a relatively short amount of time and we see across-the-board improvements in your entire catalog being released with some frequency. Either you are the world’s fastest software coder or you are a master of object-oriented software engineering … and I’m betting on the latter. Can you tell us something about that?
VM: Well, I think it’s both, plus the fact that I’m a workaholic 😀 . Essentially this is all about planning ahead. While from my experience about 99% people just implement a solution to problem “A”, I’m always thinking in terms of “A” being generalized to something usable in the future for something else. I started with this already in high-school. I was basically putting a huge library together and every time I needed some algorithm or anything, I just added it to the library instead of “just using it”. So now I have vast amounts of code, my own cross-platform build method, plus the plugin kernel system, so building an update for all the 80 or so plug-ins takes about one to two hours, and to create a simple plugin including everything except for website stuff, it takes just a few hours as well. Of course, when the plugin is more complex … The downside is that any change in most subsystems can have many, many consequences, so keeping track of everything is very complicated.
SM: MSpectraDynamics is certainly one of MeldaProduction’s crown jewels [Note – SoundBytes writer Dave Townsend covered this in great detail in SoundBytes last year – read that article here: http://soundbytesmag.net/spectraldynamicsdynamiceq/ ]. It definitely has some very enthusiastic fans out there. Tell us a bit about what inspired you to create this relatively unique processor.
VM: Actually MSpectralDynamics was more like an experiment. MDynamics was released right before it and the idea just came to my mind – why not to try to do that for each frequency in the signal. I thought it would be like a multiband compressor with lots of bands. Everyone was saying it’s a stupid idea though 😀 . But I wanted to try. The thing with spectral processing is that it appears to be really simple, but it needs to be done in a very, very smart way if it is to sound good. Otherwise you get the “alien” sounds you all know from badly compressed mp3 material. Anyway when I was nearly done with that, I was like “man, this is really cool”. And it indeed, it has become one of the crown jewels, which despite it having existed for many years, it still doesn’t have any competition in the market. And at that time I had no idea it could do things like de-noising …
SB: Let me ask you the same question about another product – your latest blockbuster, MXXX. While there are other products a little bit like it, there’s nothing remotely so flexible, deep and powerful. Can you tell us about the evolution of MXXX?
VM: Well, a few years ago I found out that we have almost everything in the portfolio. I accidentally checked uHe’s Zebra that time and, you guessed it, a lightning strike of inspiration came to me. I was never a fan of modular DAW’s/plugin’s. All the routing and stuff is just waaaaay too clumsy. But this uHe’s idea was just brilliant, so I just pushed it further and made it work so that the actual plugins can be used within the modular system. Then I finally realized that this is, in a way, the ultimate super-effect. That scared me, so I conducted a preset making action, which was a huge failure – a huge, huge failure. At that time I still believed in human race, so I just sent a temporary license to everyone, who registered to be participating in that action, which was nearly 100 people. And guess what, less than 10 of them actually delivered any presets. After that I just knew this would never happen again and I kinda realized on which planet we are. Anyway I put it on ice for some time, because there were so many improvements all the time, and now it is actually far, far better than before. It can really do pretty much anything now. And after the intervening years, there was now a new preset making action, which was a big success. Well, except for the psychopath from Chicago, who tried to blackmail me to get a license, bothered people at KVR, etc. Another point down for human race. 😀
SB: On the Melda forum at KVR, you have occasionally alluded something you’ve got in the pipeline that (forgive me if I don’t quote you exactly) “will revolutionize mixing”. Now, we realize that you can’t give up any trade secrets that will let the competition get an edge, but can you tell us anything at all about this intriguing development?
VM: Hehe well, I cannot say. Despite the fact that I’m pretty sure it would take years to clone it for competition. This one is so big that I want to keep it for myself. It will take lots of time, so you will have to be looking forward to it for quite some time. All I can say is that in my opinion the current way people are mixing is just medieval, because progress in audio world is unbelievably slow – most likely because of the big companies in the analog and digital worlds who took that piece of the pie in the beginning of the era and now are holding it using the money they made. But technologically they are not bringing anything new, they are even slowing down the process intentionally, just so they can hold the market longer. And if you think about what we do now from that perspective, it’s just ancient and there are so many ways we can make this better. With the “mixing revolution” I’d actually like to take more than one step to the future, we’ll see how that will go. So many words and I revealed nothing – good job, Vojtech! 😀
SB: OK, fair enough … but when the time comes that you need a beta tester, I am so in! 😀 But seriously, if MSecretWeapon is a long way off, is there anything near-term that you can let us in on? Can you share any hints about what’s in the pipeline?
VM: Hehe … well, I’ll keep you posted. I’d actually like to keep the list to myself, you know, so that someone doesn’t steal the ideas. But a few things are being worked on – one is a morphing plugin. On a larger scale there’s a spectral variant of MXXX, MYYY as the “ultimate synth”, a spectral synth. I often start with something and then get overwhelmed with the maintenance of existing stuff so much that, sadly, I need to put things on ice. That said, I need to say: “People, stop coming up with additional feature requests!” 😀 . No really, some feature ideas people come up with are great, but many times them are ridiculous and completely useless. And these folks often think “they understand it all”. They start being aggressive and stuff … and that complicates the development a lot. After all, the development time is basically the most expensive resource.
SB: With so many aspects of sound engineering covered with your current catalog of products, do you ever worry about running out of new things to build?
VM: No! I wish I’d get into a situation where the plate wouldn’t be full with a dozen new projects in the queue. I have way too many ideas, some may be silly, some impossible, who knows, but there are many. What I don’t have is development time – it seems that MeldaProduction will need a few more developers. The trouble is, I’m a very detail-oriented, critical and demanding … it would be very hard to find people that would satisfy my expectations.
SB: Vojtech, this has been a great pleasure. On behalf of our readers, I want to thank you for your time and I wish you the best of luck.
VM: Thanks David, my pleasure! And, mainly, good luck with your music! That’s why we do all this after all. 😀