Interview with the demoscenter - Smash ^ Fairlight

This time we have an interview with Smash from the Fairlight group, who has been involved in the life of the demoscene for a long time, and is well known as one of the best coders. He makes extremely versatile demo - sentimentally pop with great graphics, but did you know that Smash started as a musician? Also in an interview, he will tell you how his best works were created, share the secrets of his dem's popularity, and how he writes code for his masterpieces.


First, tell us briefly about yourself?


')
My name is Matt Swoboda, I am also known in the demoscene as Smash. I am currently in the Fairlight group as a coder (programmer).

Let's start from the beginning ... When did you first sit down at the computer?

I got to know computers quite late. In my opinion, we didn’t have a computer in my house until I was 10 years old, and even then it was a gloomy computer for running a business that my father used. At school we studied at BBC Micro, but the classes were too rare to interest anyone with programming. I had a real computer at the age of 12 - it was a PC.

How was your baby? Have you always been creative, or were you obsessed with video games ... or dinosaurs?

I think I was a creative child. I liked to collect LEGO and other designers, but I could never boast of success in drawing or playing musical instruments. Although I studied well at school, I was generally rather lazy.

Ok, when and how did you run into the demo scene? What prompted you to become part of it?

I saw a couple of demos and crectros at my friend on the Amiga, completely unaware of what it was. My real discovery of the demoscene happened when in 1993-94 I bought the magazine PC Format, which was packaged with a compact disc with a bunch of rubbish. There were several demos, among which was " Second Reality ", and a lot of tracker music in the .mod format.

At first I was interested in programming, I had a crazy idea to make games and I was constantly messing with BASIC at Acorn Archimedes and at home on the PC, trying to figure out how it all worked. Then I realized that I would need some kind of music for the game, and I began to study all these trackers and tried to write my own tunes.

I listened to a lot of mods, and in each I read all the lyrics that the musicians left in the samples, and that interested me, I think it started this way. In 94-95, the Internet appeared in our home and, some time later, I discovered IRC for myself and quickly found the #trax channel where the tracker musicians were playing.

In the end, you did the game you wanted?

No, writing music, at that moment, seemed to me more interesting, and I completely abandoned programming for a couple of years. Later, I participated in the development of several games. Maybe someday I will make my own game.

It would be interesting ... so, it means that at first you were a musician, I even saw your name in some works from the 90s, but now you are definitely a coder, what made you go back to programming?

My discovery of the demoscene began with music, and I, in fact, did not know very well what a demo is and how they are made. I just wrote music, put it on the Internet on the Hornet Archive and chatted with the people on the #trax channel, even participated in several contests. I was in the NOISE and Theralite groups. On the channel #trax mainly there were people from the USA, Canada and Australia, countries where the demoscene was not as developed as in Europe. The guys just wrote good music, not much bothering about the demo. But gradually I began to look more and more demos and more and more ignited this kind of creativity.

When I started to achieve good results in writing music, my friends started taking my music for their dems, several tracks went to discmags, then they started asking me to write music for a certain demo to participate in demopati. Then, a friend of mine introduced me to guys who wanted to raise the banner of Razor 1911 on the PC scene, and I joined them.

At some point, I even became a popular musician at the demoscene and several of my tracks were released on a Miasmah type CD, I won several competitions for small demopatis, I went preselect for large parties like Assembly, but did not win prizes. At that time, I spent several hours creating tracks daily in Fast Tracker 2 for 3-4 years. But I never created anything really cool. I was not a musician by nature, of course, after so many years of studying music, I learned the basic laws and some features, but it was not some gift or vocation.

Everything changed when I entered the university (in 1998, when I was 18). First, I had freedom, some pocket money, and I started going to demopati. The first was "Wired 98" in Belgium with a large number of participants from the UK. It was then that I got it! In 2000, I left Razor and joined Fairlight, and we have become a close-knit team since day one. Around the same time, tracker music began to give way to mp3 tracks and soon almost disappeared from the demo. Unfortunately, the transition from Fast Tracker 2 to professional audio editors was very difficult for me, and I could not achieve more or less decent quality compared to other musicians, and my interest in music began to fade.

Then I started trying to write code. I programmed at the university all sorts of laboratory work and calculation tasks for Java and C. And I really liked it. I zababahal a three-dimensional shooter in space for the programming exam, learned a little OpenGL, and by the end of my studies at the university in 2001, I wrote my first demo (actually it was 64k intro). It was quite crookedly done, in some places I didn’t understand how it works, but it was a useful experience that taught me a lot, from that moment I really got infected with the demo stage, and, more importantly, I received my first work as a 3D graphics programmer.

After I started programming at work and run demos in my free time, I began to learn very quickly. My work colleagues raised my level even more, from whom I learned a lot and, eventually, began to understand what I was doing. I also found that I absolutely had neither the time nor the desire to play music ... I became a coder! :) At that time, I myself made music for my works (usually they were intro) and I very quickly realized that making music and writing code takes a lot of time at the same time. Much faster and better is obtained if the music is written by someone else more talented in this kind of art.

And, of course, you collaborated with different groups then?

I think it is necessary to clarify what kind of cooperation it was. I think you ask about my joint work with CNCD , Orange and others. In fact, this was not some kind of group collaboration. Just a few friends who were in different demo groups decide to make demos together, and then include the names of their groups in the credits.



Suddenly ... how then did the idea of ​​what you do appear? Did you start with a general concept, effect, mood?

I invent new effects and chips all the time. Usually I get some kind of “big idea” and I carry it for several months, usually it happens in the fall or spring. I experiment with all sorts of fashionable algorithms such as fluid dynamics, particle systems, ray tracing, volumetric lighting, and other fashionable ones. Then at some point we decide to make a demo for some major demopati such as Revision / Breakpoint or Assembly , we select the most successful effects that look impressive enough to become the basis of the demo, and flexible enough to allow you to use several just by changing the parameters. Thus, it turns out that we are doing "technological demo". For me it is important that the demo was built on some kind of technology, so that it was the main feature, the soul of work. If the demo doesn't look technologically cool, why bother doing it at all? Why strain and do something real-time, if it does not require complex calculations, and does not work at the limit of the possibilities of iron? Well, of course, there are people who do it just for fun, or for some other good reason, but for me the main technicality is the main point to release something.

"Techno-demo" ... first, algorithms and technologies? Have you ever made music first?

Ideally, yes, we have to pick up the idea, technology, music, and adjust the picture and sound so that they emphasize each other. But it rarely happens. It seems to me that when all this is taking shape, we get a really breakthrough work. For example, we did " Agenda Circling Forth " ( video ), " Numb Res " ( video ) and " Blunderbuss " in this way and, in my opinion, this is noticeable). The overall design is based on music and follows the music.



We don't even have a regular musician in the band, and the soundtrack is always a problem for us. :) We have several acquaintances who we can ask to write music, but more often we find a track on the side, hearing it on Soundcloud or Bandcamp, or someone accidentally sends us the perfect thing. Music is a critical part of the demo and it is important to pick it up correctly.

If we do not have a soundtrack along with the idea of ​​a demo, then we have to write it in the process of working on effects. This has its advantages, we can change the music (for example: “1 minute in this place is too long, cut out a couple of bars.”), But at the same time, not hearing the track is completely incomprehensible how to build effects. In my opinion, the best option is to make a demo, adjusting the effects to the music. The general mood and transitions between visual parts should be combined with music as much as possible.

What about titles? How do you choose a release name?

Oh, this is usually done at the very least when the general idea is realized and polished. Some of our works are called not very impressive. In general, the online anagram generator is a great thing, we used it a couple of times for the selection of names.

Come on! It is hard to believe. Then tell me how you do your part of the work? Where do you get inspiration?

First of all, we are with Desop or I alone decide to make a demo at all or not. This always happens with the goal of unleashing on some kind of demopati, about a month or two before the event. At Christmas, we usually think about a demo for Revision / Breakpoint / The Gathering and in the summer for Assembly. Sometimes it turns out only for one season, sometimes even for one season.

First we go through what we have been doing lately. This is usually some kind of technology or effect, but sometimes it can be some kind of visual modeling or animation technique. We look at what it looks like from a democenter. We also recall the video clips that we saw recently and discuss the ideas seen in them. We are always looking for ideas outside the demoscene - video clips, works of art, anything on vimeo or behance.net and similar sites.

Your work is very popular among demoscenters (and not only among them). Do you have a secret or do you use something special when doing a demo?

I think emotional union with the audience is the key to whether the demo will be just good / impressive / strong or it will become a classic that is remembered and watched for many years after its release. It is very difficult to predict what exactly will hook the viewer, but when this happens, the demo turns out to be something more than just a sequence of effects. We have such works turned out "Numb Res" and "Agenda Circling Forth" I think in these works we have found something that touches the beholder for the deepest strings of the soul.



For comparison, we have releases that did not achieve such success, despite the fact that they were made of excellent effects. In the demo, the magical emotional unity with the viewer is achieved by the connectedness of the visual and musical components. And it seems to me that the above two works did not accidentally become so popular, in both cases we first found the music, and then lined up the effects around it.

This confirms my feelings, I always felt that the music in your demos is not just for the background or on minor roles ...
Since we started talking about music, this is an interesting question: “If there were no copyrights and licensing, what tune would you like to make a demo for? Any, from Bach to Madonna. ”

I think it depends on the demo we are working on. For each demo I did (when we didn’t have a soundtrack initially), there was always a tune in my head that would fit perfectly. For example, for the last work “Apocalypse When” it was the composition “Phat Planet” by the group Leftfield.



Very dramatic ... and the effects and music seem like the beginning of a movie :)

The demo should have a development and a final, there should be certain places for the main action and quiet transitions. I heard such a thought that a simple formula lies at the heart of demodisance - you need 3 key points, the first one almost immediately after the beginning, after a calm introduction, the second in the middle and the third at the end, after which the main action is minimized. Selecting the right moments for these key moments, people seek the presentiment of an event from the viewer and then finish it off with this event. And the music plays one of the main roles here.

If you select the most destructive effect and insert it at the most emotional moment of the soundtrack, it will shoot. If you make a mistake, if the music is calm, when everything is spinning and sparkling in the demo, you will get just cool effects to the music, not the demo. But music should also have elements of development and climax, and they should be memorable and engaging, as well as visual effects. That is why I prefer to do a demo on an already existing track, but at the same time it should be bright and dynamic enough for the effects to fall well on it.

For me, the perfect soundtrack (for the Big Demo, and not for the usual, where templates and stamps stick out of the monitor :)) should be memorable, have a catchy melody, people should remember it even when they walk down the street, they will hum it and certain parts of it will be remembered and how they were in sync with the visual part. I really like it when they use vocals in demos. This is one of the chips that is remembered, especially if the words are reinforced by what is happening on the screen. We can say that the vocals on the soundtrack are half of its success. Even if people do not remember the melody, they will remember the words.

I like pop and rock music, and I would like to make a demo with a similar soundtrack, although it will be very difficult ... maybe one day a good visual idea will come to my mind to put it on rock music. I always wanted to stir up something under 65daysofstatic, that would be awesome!

Okay, if any readers know who 65daysofstatic are, let me know.
Unfortunately, I would not understand a word ... but perhaps among readers, there will be someone who also wants to make a demo under 65daysofstatic.
Let's talk about what software do you use when writing a demo? Do you use any of your work?

We have a demo on which everything we do is based. This is what we have been creating and modifying since the 2000s, but the current version has not changed since 2007, and the interface has not changed either. Externally, work in our environment is similar to AfterEffects or eyeOn Fusion, only in real time, and allows artists to work separately from coders, although this does not always work out. That is, this means that the work on the demo can be divided: the coder can make new effects and logic, and the artist then arranges everything beautifully, paints, generally does what is called “design”, and it is very important that they work in one environment, it allows them to be more productive. Even if I do some kind of release alone, I use the demo tool, because there it is more convenient to create a general composition and then work on the details.

We used our demo tool for all released projects: all demos made since the mid-2000s, all 64k intro, two issues of Zine magazine, various interactive projects, several particle simulators used for pre-winders and even for a couple of video clips. We have a very cool and flexible environment.

Where do you usually make demos? Are you doing something special in the process, are you listening to music or drinking beer, or just kodish in the dark?

Usually I don’t have enough free time to purposefully devote it to creating a demo. Sometimes it happens when I go on the train, at lunchtime at work, in the evenings, nights, weekends at home, sometimes lingering after work, on the plane ( translation, by the way, also made on board flight SU1730 ... comment. Translator , in cafes, for a cup of coffee, etc. etc. More and more I work on a laptop, it certainly slows down, but then I can code almost anywhere. And this is awesome!

As a rule, I listen to music at this time, I drink little coffee or diet coke, but sometimes I tear myself off and write code for beer on demopati. I concentrate easily, so I do not need a secluded quiet place to work fruitfully. I work best on the train, because there is no Internet (the main irritant, and because I know in advance the period of time that I can devote to work. I choose some kind of problem for an hour and do it. It is easier for me to concentrate on a short period of time. If I work long and continuously, I get bored and I get distracted or work less efficiently, and for one hour I can get rid of the whole world quite easily.

Train coding is obtained. By the way, your effects in the demo are more often realistic than abstract. Is this what you specifically do or just goes by itself?

With reality, this is what happens - people perceive reality subconsciously and this is a very powerful tool of influence. For example, I made some kind of abstract “garbage” and nobody understands how big or small it is. It is very difficult to understand on abstract objects. But if I put the building next to this "garbage", then everyone will know what size it is. Now it’s not just “bullshit”, now it’s “bullshit on the street” :) Playing with the size and perception of familiar objects is a key feature in my demos. I think this is very cool. Why make some kind of smoke that turns into text when you can build a building and blow it up.

It seems to me that realism in the demo is very useful in order to establish a connection with the viewer, especially if the viewer is far from the demostsena. Even if I do something synthetic and abstract, I always add realistic shadows to it. I can not stand the visually flat ugly effects and always try to get rid of them one way or another.

How do you create such images? Do you see something in life first and then program it in graphics?

When I come up with a new effect, I often look at a lot of different visual material, both real (as a rule, it is more useful) and synthetic, trying to render what I want to see using different software for 3D modeling. I can take some effect that everyone uses, model it in software, dramatically improve it so that it cannot be reproduced in real time, and then gradually simplifying, eliminating visually unimportant parts, I try to code it so that it works already in demo. I read all sorts of scientific research if this refers to what I am trying to reproduce, of course, I furiously google everything that has already been created on this topic, etc.

I heard that they speak about you as one of the best coders on the scene (I even heard that they took autographs from you on one Japanese demopati!). In this case, allow me to ask a few questions about programming (they may seem ridiculous to you, because I myself have never written programs.
What is the most difficult in programming?

The most difficult thing in programming ... perhaps it is deadlines and deadlines, you never have enough time to do something the way you intended. You can run into a bug that takes you several hours that you did not plan to spend on debugging. But the most difficult thing about programming demo is the visual reality of what you are trying to reproduce mathematically, and programming is not difficult. Code is just code, you write it if you know how to write it. The hardest thing is to understand how to do what you want to do, and not the process of programming this idea, when you already understand it.

I am not very cool in mathematics, I understand it well, but no more. I often get blunt when I try to understand some multi-storey equation in those scientific papers that I have to read, it is easier for me to understand something if I see it in the form of source codes, sometimes it is easier for me to reproduce something in my own way than to thoroughly understand all these mathematical calculations. Mathematics, yes, perhaps mathematics is the most difficult thing about programming demos.

So the main thing is design and mathematics ...

And Inspiration. Sometimes it's hard enough to come up with something new. Also, a certain complexity is that you need to come up with such a thing, so that it is complex, new, and realizable on the current hardware. It's easy enough to program something cool, but it works very slowly. It doesn’t matter how cool it looks and that no one else has done this on stage if it works at a speed of 1 frame per second. There is always a line for which you can not intercede. On the other hand, if you did something working at 200 frames per second, then maybe it's just crap, why do another crap? :) Or if you are doing some kind of extreme visual art, incomprehensible to most people, you will not be understood either. You need to be able to find a middle ground, use different tricks, which is usually difficult to do, because simplifications and tricks kill an idea. It is very difficult to find a balance. You need to choose goals wisely. There are several effects that I came up with many years ago, but it became possible to realize them only now.

And what do you do when you reach a dead end?

The wonderful side of programming is that there are always 100,500 things to do right now and most of them require 0% inspiration, tasks that you just need to sit down and do. If you do not want to engage in any effect, you can fix a couple of bugs in the demo-tule, or put in order the menu and buttons. Often you just need to get away from the problem, and it will be solved by itself when you think about it again.

Sometimes I throw some effect for several months, until I accidentally find a solution, doing something else. This is the thrill of coding. Everything is intertwined in one tangle, especially in graphics programming. The knowledge that you get in the first steps in this area remains relevant for the rest of your life. Software rendering algorithms used 15 years ago, now used when programming the GPU. Knowledge tends to accumulate. Of course, it will take years to gain enough experience to make cool demos, but you can still do a lot even at the simplest level.

Ok, then what's the cool thing about programming?

As for graphics programming, this is a bunch of tricks, chips, algorithms. When you encounter a problem or when you need to create a new effect from scratch, you just take a few ready-made pieces created earlier, collect them together and everything works! Many "new" effects are actually combinations of technologies and tricks that were previously used.

Sometimes you come up with a new chip for the demo-tula and you are happy to discover that it solves a bunch of old problems that you could not solve "head on." For example, this year I managed to do a raytracer (which is already used in the 5 Faces demo) fast enough to draw complex polygonal models in real time. It was a big breakthrough, because there are a lot of problems that can only be solved with the help of raytracing and then BAM! it all worked! :) I now recall some abandoned ideas that I had previously and finally understand how I can implement them. Switching to DirectX11 was another big revelation and a big revolution in the head. , - - , , .



- « »?

It's a difficult question. , — , , , . , .

?! . , -, ?

, , , ( , ) , , . , , . , , , , , , .

And now I will express a non-standard opinion. I believe that a good programmer should be able to work with different people. To be able to explain to specialists in other areas the features of their work, to be able to understand the course of their thoughts and their limitations in solving your common task. A demo coder should do it all for free. The ability to conduct a project independently, to convince people that they also want to work with you - this requires leadership, the ability to understand others, maybe even in some way and charm. It turns out that the best demo coder is a smart, collaborative-oriented person.

— . - . , , , . 64 , Windows, , , / , , … .

. , . : « , , , :)»

… , «!» , Paper ( video ) from Psychic Link and Acme, an indescribable effect when a scene springs from a sketch. So I really like Square ( video ) from Pulse and 303 ( video ) from Acme. These works of Statix ​​from the 90s set a standard demo for me then. And they are still relevant, they are the embodiment of the perfect combination of technology and visual art, amazing work. I also like the work of Moppi Productions Assembly 2004 invite ( video ). Small and perfect.

, , «» . . , , ?

. , . - , , , , . ? : « » , pouet.net , . . , , . , : « » « , » ( , . :) . , — , «». , - , , , , - . , . , . — .

- ?

Of course! 64 , . . ( , ) — , , . , , , -- , , . , 14 , , , .

14 … :) , . , . :)

Yes, the world is changing ... The Demoscene is more and more behind the rest of the technology world. The true demoscene is a thing of the past - old platforms, a terrible limit to the resources and capabilities of iron, the old design style, nostalgia for the “old school” are now increasingly found in modern works. The rest of the world caught up and in many ways surpassed the demoscene in technical terms, all that we have left is the art of real-time graphics, and then, more and more often, these techniques are found outside of the demoscene. And many who adopted these technologies that we invented in the demoscene now make incredible things of them. Now we need to learn from them. The problem is that everyone who remained on the scene does not want to do this, all that they want to do is the same thing that was in the 90s, because they grew up on it and want to keep it in mind. Demosceneas it now exists, it inhibits itself, but nothing is permanent and eternal.

?

, , , . . , , - , .

Ok, ? ?

I have an idea to make some short demos. It usually takes a lot of time and effort to make a six-minute demo of 10 parts, and it seems to me that now you don’t need to do such things. Sceners will definitely like it, but people far from the demoscene will think: “Why all this is mixed in a bunch, I don’t understand, I don’t see any connection.” I'd rather do some cool two-minute dems united by one idea.

Mmmm ... I will wait with impatience. In conclusion, what do you want to wish to the demostsenters?

Go and do something! If you want the demoscene to exist, go and make your contribution to this amazing art.



- Despite the fact that some of my questions were strange, Smesh answered everything and generously shared his knowledge. Thank you Smash, for your patience and attention!

, , " direct to video " Fairlight pouet.net. Displayhack . AssemblyTV Fairlight, .

.