Hey everyone! I just wanted to address the inaccuracies in this video that people have made me aware of! This was a video presentation for a college course I took and I was only expecting my class of 10 people and my professor to see it instead of the 93K (at the time of writing this) that have viewed it. Yes, I could take the whole video down and re-upload an edited one but that will take up more time than I have right now. Thank you to everyone who pointed out these inaccuracies! Inaccuracies: 0:48 - 英語のアニメ (Eigo no anime) should be changed to アメリカのアニメ (Amerika no anime) meaning American Anime / American Animation 2:28 - Ninendo should be spelled Nintendo 4:56 - The oscilloscope that was used as an example used a 2A03 Mod and was not the original sound produced by the NES 5:46 - The Commodore 64 (C64) is not a dedicated video game console but rather an 8-bit home computer that had the ability to run video games 10:16 - I pronounced flanging wrong 12:08 - Church modes should really just say modes (even though modern modes derived from church modes) 14:03 - 6.0 GHz/second should just be 6.0 GHz 19:52 - Marylou should be spelled Marilou Additional notes: I put the subtitles in before knowing I could add the correct ones via. TH-cam. I'm still learning how to edit my voice, I used a DeEsser and it sounded fine through my speakers but never test-ran it with headphones. I didn't come up with the term "chipsters", it's a term mentioned in Marilou Polymeropoulou's article titled "Chipmusic, Fakebit, and the Discourse of Authenticity in the Chipscene". My apologies for not including a warning for the flashing lights in the video! Flat out disregard the bit-depth part.
Good presentation :) To add one more nitpick: Core i9 is a (sub-)brand of processor, not a single product. Some i9 processors are specced for less than 5 GHz. Processors don't really have clock speeds themselves, but they are designed to be operated at a certain speed or range of speeds. Modern computers even slow down the clock on the fly if there isn't much work to be done. As for the speed at which it can process information, it's no longer as simple as comparing clock rates; modern CPUs can get much more done per clock-cycle as well, depending on the nature of the processing to be done.
On voice editing - I recommend using a program to remove background noise (I know audacity has this as a feature). This will make it sound cleaner and make the transitions in, out of, and between voice clips more seamless
From my experience, when I hear people talking about "8-bit music" they usually means squares, triangles, saws and noise while "16-bit" seem to be commonly associated with fm synth stuff and some basic effects like echo (not reverb) and panning
@@esmooth919when I hear chiptune I always think of the C64... That soundchip was something else man Just think of Rob Hubbard... Especially "Masters of magic"
@@SmplySilverThe SEGA Game Gear used stereo sound as well believe it or not despite the sound chip being shared with the SEGA Master System. I believe the TurboGrafx-16 also utilized stereo sound.
I've always just reserved "8-bit" for abiding by limitations, and "chiptune" for the general sound of a song produced with simple chiplike waveforms. I recognize it's arbitrary to use those terms that way, but it's still how I use them
Personally a specific bit (8, 16, 32, etc) is to me abiding by limitations, whereas chiptune is just. Sequenced music almost in general to me, MIDI can be chiptune, but for actual consoles the cutoff is DSi for handhelds and N64 or so for home consoles if that makes any sense..
That’s how we talked about the music in the early 1990s Amiga scene. Chiptunes were tracker songs that took a very small amount of space and sounded like the old computers.
@@makipri And with MIDI files, well, it could sound like a whole orchestra with a good synth (cough cough roland ST-32) or like an old computer depending on what your setup was. I guess chiptune can at times be more of a general _feeling_ rather than a concrete set of limits, yeah?
@@jeffystreet General MIDI added a lot of uncertainty to the mix for sure. Before that it was somewhat reliable what it would sound like when playing back a song. Of course different SID chips could play the song back all wrong but still the soundscape would be pretty similar. Anyway, in the early 1990s there was no word for fakebit, not really 8-bit either. We just called fakebit chiptunes and the real 8-bit tunes with their original sound chip or platform names.
@@spastorok I also found the focus on hertz to be unusual. I am given to understand that this way of measuring a CPU's processing speed is not very useful; at least these days. Perhaps someone with more knowledge can fill in the blanks. I did get the feeling that you were writing beyond your knowledge when it comes to the computery stuff in general; is that a fair assessment? Overall nice video. I particularly liked you drawing attention to how a word got imported from English to Japanese and then was imported back into English, undertaking a new meaning on its journey. Kinda nutty!
@@OctainServers until around the year 2000 (give or take several years), the Hz of a cpu was pretty good at telling the speed. since then the Hz has mostly stagnated, as the amount of power needed to push it faster isn't linear, so you're better off spending that extra power on parallel processing and other methods of speedup. also over time software had also improved, at least in the way that newer compilers work harder to make the same code run faster on the same hardware. that said, it has always been where no matter how fast the cpu runs, some operations are simply slower than others. for example, you could have one computer that adds in 1 cycle and multiplies in 10, and another that takes 1 and 5 respectively. same Hz, but the second is faster, but only when multiplying. there's a lot more nuance but yeah it's really hard to compare speeds by looking at the hardware itself. but in the end, if two computers have sufficiently different Hz, you can almost always expect that all the other variables favor the faster one anyways, so if anything it's an understatement.
Hey there! I was the director of a video game that utilized a chiptune soundtrack for six years and wanted to share my experiences with this subject because it's something that I've struggled with for a very long time. My game had a small team, but it was a relatively large teamm of composers. There were four of us for the majority of development, including myself as one of them, and the composers took up roughly a third of the total number of people involved with the project. From the veyr beginning, I envisioned the game to have a chiptune soundtrack (specifically emulating the limitations of the Game Boy), and throughout development I spent a lot of time cracking the whip and shooting the team down when they got a little too ambitious. I had two schools of thought when trying to justify this, one philosophical and one practical. The practical explanation ironically made more sense in theory than practice: it's so easy to find "8-Bit remixes" online that basically just throw a song into midi and replace every channel with a square wave and call it a day. Playing 8 square wave notes at once can sound somewhat retro, but it has this same inauthentic vibe as a lousy TV commercial trying to appeal to gamers with cheap pixel art and 90s buzzwords like "Level up!" I felt that by more faithfully following the limitations of a specific hardware, this sort of vibe could be avoided. In practice however, I often couldn't immediately tell the difference between a song with 3 voices and one with 5: the truth is that a composer with a genuine appreciation for the style will know how to use every channel with intent, and those channels serve a purpose. The philosophical reason is something I still subscribe closer to: if you're not interested in emulating a specific style, why bother with any limitations at all? Usually chiptune sounds have objectively fewer options than modern composition hardware can provide, so why hold yourself back to only Squares and Saws when you can utilize any instrument sample in conjunction with the sounds you enjoy from these waveforms? The answer I've come to there is that chiptune music does have a specific style. The line of thinking that it comes from limitation is like asking a pianist why they don't add percussion to their music, or asking an electronic band why they limit themselves only to synthesized instruments. The truth is that there is no "rule" to what you can use in any genre of music, but nonetheless certain choices in instrument and composition will affect what kind of song you create, and choosing to use the voices of a particular hardware isn't an agreement to adhere to limitations, but merely a desire to utilize the sounds of that hardware, regardless of what it's actually capable of producing. I still consider myself a purist in my own compositions, and I don't regret being so strict with my team, because we ultimately did produce an award winning soundtrack with a very distinct sound and I'd have it no other way. However, I acknowledge now that there are very genuine and valid reasons for fakebit to exist, and I think that other purists like me would benefit greatly from opening up to what those reasons might be. On the other side of the spectrum, I also think that fakebit "radicals" (that is to say, people who feel there should be no rules in chiptune at all) might benefit from understanding the perspectives around realbit better as well. This was probably a super long read, but as a longtime game developer who's often struggled conveying this argument to his team members, as well as a chronic video essay enjoyer, I found this video to be especially relevant to me, and I thought others who are in a similar spot might also enjoy hearing my thoughts on the matter. Thank you for reading all of that!
This was an interesting read, thank you! I can't call myself a deep fan of chiptune (as I'm not usually actively seeking the music in the genre), but I do enjoy both fake- and realbit with it being more than a half of my (admittedly patchwork-y) music taste. I'm kinda interested in the game soundtrack and the game itself. What's its name?...if it's not a secret ( "^^)
For me, chiptune that closely follows the hardware's original limitations and chiptune that surpasses the limitations to create a more unique sound both have their own places. I love listening to music that is so unique, you can't believe they're working under the limitations, but i also love the ones that just invokes the feeling "real" chiptune had, while not being held back by hardware limitations. It's not a "which is correct?" question, it's just a stylistic choice. I gotta say though, i hate when people lazily put 8-bit sounds and noises on their music because it's "nostalgic", it only really sounds cringe.
Not to be reductionist but it's basically the difference between a purist and innovation, and somewhere inbetween thats lost to nostalgia, real or borrowed. We see it all the time with demakes and "ps1" games these days. If we truly have ps1/n64 games again everyone would be miserable with the poor draw distance, terrible framerate, and somewhat endearingly artificial loading screens. Poor controls would be the cherry on top, because no one remembers that if you didn't have a dualshock controller back then you were playing ace combat and ape escape with a D-PAD. Meanwhile a lot people seem to forget, or more likely don't realise, how much pre-rendered textures/scenes were used to overcome limitations. Donkeykong, Final Fantasy, Fear Effect come to mind. Very few developers truly try to develop for the OG hardware but when they do it's an amazing technical feat. Personally I've stopped caring. The internet is one giant mixing pot and it's down to individuals to learn about history and share those discoveries. And I super appreciate people who post videos breaking down the sound channels of somegames, going so far as to compare soundfonts from the various consoles.
This whole arguement is like saying someone is not a real movie director for using digital editing and not 8mm film. Its up to creator to use whatever tools they like to acheive their creation.
11:04 I'm no expert but Tim Follin is kind of legendary with how he used early console sound chips, it's kind of an extreme departure from his peers to use him as an example.
Slightly irrelevant but Tim Follin also created imo the best video game soundtrack of all time with Ecco the Dolphin: Defender of the Future. Not chiptune, tho I think it was done entirely with MIDI instruments
A departure from console music of the time perhaps, but in terms of technical "tricks" most of what he did in e.g. Solstice had been pioneered years before on home computers. I would say that his style was somewhat distinctly European and was only a departure in the sense that most video games around the late 80s had distinctly a Japanese style. Follin's strength IMO is that although he makes good use of European style tricks like fast arpeggios and intricate parameter changes to embellish the sound with great sound design, he is also just an exceptional composer who used conventional composition techniques and ornamentation really well.
@@RyumaXtheXKing I'd love to hear other European chiptune composers with a sound like Tim's (besides Geoff Follin of course). Can you give some examples?
@@DustyMusician look up the soundtracks of european developed amiga games. there is tons of them! the amiga was primarily a european "console" so most games and music for it will originate there
At the end of the day, "Did I write good music?" is more important than "Did I write authentic music." Squares are square, and I like the way they sound. Creative limitation is great and important, but if I want to write a piece with tons of channels and lots of sound generation techniques, I will. But I'll still call it chiptune, because I've got to call it something.
It's great how you managed to provide viewers with precise definitions of technical terms while not leaving out important details and still keeping the definitions easy to understand. It also made the video dense with information, yet not boring at all. Hats off to you.
i clicked on this from recommended, watched the entire thing and didn’t realise how little views it had (800) until someone mentioned it in the comments omg. this is FIRE
This was awesome. All the things I always wanted to know about chiptune. That music imprinted on my soul along with pixelated graphics from my childhood era. Making the magic creatively despite the limitations made it feel like magic that you could deeply love. It wasn’t an easy thing to just churn out. We reveled in when it all came together and we felt it like magic. You rock! ❤✌🏾😎🔥
Chiptune composer here. I think a lot of the inaccuracies people have already pointed out would have been caught had you reached out to chiptune communities. In my opinion, one of the most important aspects of research projects like these is communication and collaboration with experienced people.
You're definitely not wrong! I'll keep it in mind if I decide to do a video essay again - this was just a final project for a college course I'm taking, so apologies for the unprofessionalism regarding the research.
Nice video some feedback, for whatever it may be worth the central question of the video and its title isnt really even asked until the end. A lot of the conclusion could have been the intro. I dont think the intro fully worked because you talked about the history and evolution of anime, but not framed around limitations. This makes the thematic pivot to video game music limitation feels disconnected. This makes the video feel a bit unmotivated because (like i said) the central question isnt really asked until the end It also felt like most of the video is spent on the history of video game sound tech, then you go through chiptune cultural history. But because Faketune and the subcultural history isnt mentioned until the second half, its hard to understand whats important to know about the tech history. I think the history lessons could have been combined Overall, i think you need to better motivate in the beginning of the video why its being made, whats the central question/thesis as well as better sync it all into a single narrative. The editing and pacing was good, i did watch until the end! Great first video i hope you make more.
Hey! Thanks for taking the time to provide feedback! This video is actually my final project for a music history course that I'm taking and I needed to choose something in the realm of video games (hence the chiptune focus). I definitely agree though with everything you said regarding the disconnection from limitations - I'm not much of a youtuber and this is only my fourth video using premiere pro. I'll keep these in mind if I decide to do any more videos in the future!
The limitations of NES hardware have brought out the more creative side of me and many other composers I've met. A lot of music nowadays relies very heavily on good sound design or tembre but with the NES everyone has to use the same sounds so the question now is how do you make your song stand out from the rest? I've discovered so much about music that i don't think I would have using a traditional DAW (not to discredit DAW composers). I'm so glad you covered this in this amazing video!
Of course YOU'RE here lmao Honestly? I'm still jamming out to your original works, especially for a certain orange executable file and his compressed canine
Another chiptunist here. I really hate when people say they made a "Genesis cover" of something only for it to be a MIDI slap with Sonic 3's soundfont. No channel limitations whatsoever.
I am a chiptune composer, and I also use a variety of DAW arrangers. I think the most important thing is how you view music. After all, producing good-sounding music is the most important, but if you are exposed to a little bit of chiptune, you also have a little music. Knowledge, you will find some of the mysteries and fun brought by chiptune is good. Of course, many retro sounds are now being widely integrated and used, but I think why everyone is arguing about this issue is mainly because of the so-called "fakebit" Among them, there are many bad works. They probably don’t understand things like “square” and “SAW”. They will use DAW to the extreme to perform large-scale EQ compression processing and stereo expansion. It is best to get a distorted but similar is about retro video game music, so I think the most fundamental question is how to produce good music that all people can appreciate
I feel like the meme of people importing entire album tabs to fl studio and replacing all the instruments with video game soundfonts with just a few clicks did a lot to mislead people online into thinking that it was done manually and/or that the import actually sounds good when they don't. Ngl that meme inspired me to get into using DAWs and learn music composition, but I do see better how low quality and lazy many of those uploads actually were.
For me chiptune will always mean "tracker chiptunes", which are basically tracker modules with very simple and small samples. That's what was called chiptunes in the 90s, at least in the demoscene. This modern usage referring to music made with PSG:s came much later.
You're absolutely right, and the author of this video is pretty badly wrong. His terminology is extremely revisionist and doesn't match the true roots of the term chiptune. C64 SID tunes were SID tunes. Etc. Only Soundtracker / Protracker etc. small sample tracker tunes were called chiptunes.
Same here. I wonder if there are any examples of usage of a 'chiptune' or 'chipmusic' terms before trackers came in. I mean - did people in the '80s even know these terms?
came here to post this as well. there’s even a link between this and the anime/animation intro: the term chiptune started as a sample-based “borrowing” of PSG music (like japan borrowing western animation and naming it anime) without much regard for the authenticity. the idea of debating what is or isn’t chiptune (or anime) came much later, in both cases, as did “chiptune-influenced” (or anime-influenced) media, sometime in the late 00s.
The original meaning of chiptunes predates trackers by about half a decade (specifically, by about as much as Commodore VIC-20 predates Amiga). It is exactly what the word says: Tunes written as code (yes, they had to write assembly code, all of those musicians, that was the skill game companies were paying for) for sound chips in 8-bit computers. Edit for clarity: I grew up in the 80s and belong to the generation that migrated from 8-bits (C64) to 16-bits (A500). The original meaning was literally "code which generates music". You are, however, right that using the term "chiptune" for a specific musical style did emerge in the tracker scene, specifically to denote tracker mods which aped the sound and style of the music written for PSGs using samples.
@@gearwatcher You're saying the term "chiptune" was used already on the VIC-20? Until someone manages to refer to a usage that predates the tracker scene, I'll remain skeptic.
6:10 There were these absolute madlads I recall, demoscene programmers who managed to stufff an Amiga MOD into the c64 and got it play the mod back! Crazy stuff, very similar to the Genesis Toy Story title On that note, the SID chip has amazing filters that allow musicians to create super unique sounds. Take a look at Eliminator on the C64. Jeroen Tel managed to use filters to get a super fat bass out
I make chiptunes (or fakebit) since the early 2000s. I was never really a purist but thought that at least some of what I did is considered chiptune. Where do you stand on "classic Dos Modules" (xm, it, s3m and mod)? That's pretty much universally agreed "chiptune". It's somewhat sample-based and can be played on an old computer. Sounds like according to your definition that wouldn't be chiptune either. But I'm really happy to see that a video about 'chiptune vs. fakebit' get so many views. Chiptune is so niche, so seeing this is really amazing.
My thoughts: Chiptune music doesn't have to be made on old machines using archaic software or even manual assembly programming. However, the end result should either be able to be _played back_ on an original chip(s), or at the absolute very least conform to the limitations of an existing chip. Once you cross the barrier of just throwing waveforms together, that is simply synthesized or electronic music, which admittedly chiptunes are a subset of. The limitation is what inherently separates chiptunes from other electronic music in my mind. A song that's simply made up of various waveforms you threw together in FL Studio is not a chiptune, unless those waveforms conform to the limitations of an existing chip, like a 2A03 or 6581/SID chip, or a combination of chips like was so commonly seen on the Famicom back in the day (2A03 + FDS, VRC6, Namco, etc).
@@mal2ksc Using multiple chips kinda blurs the line. It was a thing done back in the day by Famicom games and hardware hackers, but it still uses actual hardware synthesis chips and not just slapping waveforms together in FL Studio so I say it counts.
@@dorukdogauysal8299 I'd say a 24-bit address space would be my limit for music "in a game" or 16-bit address space for "just music". These are arbitrary limitations going by my gut feeling.
If I use three square (or sawtooth) wave channels (no FM, glissando, or "unlimited" volume levels) working from a "library" of note frequencies spanning three octaves, should that still be considered somewhat authentic, even though created on a modern computer? Also would have a 64-bar sequence in (at maximum resolution) 16th notes in 4/4 time.
@@TheRedCap Well that sounds promising, i would like to hear some of your work if you are a musician yourself! Do you have a TH-cam channel that you post music stuff or a SoundCloud etc.?
I had my finger resting on the skip forward button the entire video, but never once felt compelled to press it. You kept holding my attention evenly througout the video. Bravo
I didn't see it mentioned in here, but I feel like a more important memory limitation to bring up was how much the soundchip could hold at a time. I definitely don't understand it fully, but the example that always made me realize this was a thing was the island theme in Startropics. I think most people know by now, the harmony track of that song was messed up, and someone had fied it like 20+ years later. He explained that the reason the track would die out and then start back up out of time, was because the code for the track was too long. As in, it contained more instructions than could be loaded into a single track for a song. He fixed it by adding in repeater instructions for parts that repeated, and removed just doubled up instructions for those parts. Which he said was a common trick at the time. I found that particularly fascinating because that had to play a HUGE part in how the songs were written. Either requiring massive amounts of repeated measures, or maybe even some form of swapping of the loaded instruction set (sort of how the later MMC chips would swap out loaded graphics rapidly mid screen draw to do cool stuff). But never really knew it because he was the only one I ever saw even mention this fact. Would love to know more about that specific limitation, because there is no way it didn't massively shape how these songs were written. Well, outside of the ones they just made super duper short as a whole to not worry about it.
The limit is not a hardware limit but just a limit of that particular (Startropics) sound player software implementation. Presumably the composer software did not have that limit since they put in more note data than what the player could then address. If i take a guess, the limit was likely due to the simplicity of using an 8-bit index in the player rather than memory constraints per se.
@@qwertypolis Interesting. The way I thought he made it sound was that you could only load a certain number of commands (I think it was 256 yeah) for a single song . I didn't think they had unique uhh, code for playing music though. I thought it was all the same on NES kind of like how graphics are always loaded the same. Just they started being able to load different sections of ROM into the console rapidly.
@@nickfarace9339 Old consumer architectures rarely have any fancy hardware controlled logic; usually the CPU is responsible for banging a bunch of hardware registers to make things happen. Not that I know anything about the NES in particular, but quick check gives that there are about four hw registers per voice to control things like frequency, volume etc. To play music, there needs to be a player routine written for the CPU that sets these registers to proper values at the correct time to play the song. Such information is then usually stored in some form of song data that the CPU routine reads. From the description of the Startropics player routine, it seems its song format was quite simple in that each hardware channel had allocated 256 bytes of song data. Usually such data includes commands such as effects (glissando, vibrato etc) beyond the bare note data, and apparently one of these commands for Startropics was a command to repeat notes or sections of notes. Sometimes a song format is standardized, for example, for the original Playstation Sony specified a SEQ format that was basically MIDI + sound samples, which was included in the SDKs together with tools that game developer could use, but for the early systems, it was usually up to the developers themselves to come up with their own thing.
is there any video of the song on youtube? Found some guy with gameplay and songs but the songs arent right edit: Found it th-cam.com/video/-GBvq609sW0/w-d-xo.html
I used to make tunes for crack intros in the 80s, 90s, 2000s, and for demos in 2010s and 2020s. Most are made with Amiga, but I've also used synths with modern daws to create tunes. I didn't know there is such clear cut division or generations, realbit and fakebit. To me it is just the aesthetics as well as, to some extent, the restrictions (but less so). Edit: nevertheless, I enjoyed your take on chip tunes!
As a chiptune composer myself, I think that chiptune music is a style rather than a way of composing. Once you start working with trackers you start to learn techniques and ways to overcome the limitations and the music start to sound close to the so call chiptune style. That could be easily translated using modern hardware, but there's certainly a specific style, specially in modern LSDJ compositions that is different from old chiptunes and there are many modern chiptune composers using the same style with moderns hardware. Is it chiptune just because you make a song using a tracker that at the end of the day IS NOT real hardware and is emulating the real chip? I don't know, but I'm certain that chiptune music is more a style and a technique rather than the tools used to make it.
spot-on. The genres were always needed to highlight the aesthetics, using them to create divisions based on technicalities is a really dick move. It's like saying there's "real" metal music played with electric guitars with metal strings, and "fake" metal music played with everything else. Absurd IMVHO.
you got some technical info messed up in the "what is 8-bit" section. The NES does have an 8-bit CPU but that doesn't have any bearing on the bit depth of the audio hardware, its two separate things. I'm an audio engineer not a computer scientist, so my explanation here might be a little weak, but when someone refers to an 8-bit CPU that means that it can only address 8 bits in a single cycle, where the bit depth of digital audio refers to the maximum number of amplitude levels a sample can exist at. The NES sound chip works with 4-bit resolution for the Pulse, Triangle, and Noise channels, and the Sample channel can play either 7-bit PCM samples or 1-bit samples using Delta Modulation (tbh i am unfamiliar with DM, but based on what i skimmed out of this tech sheet i found for the 2A03 chip it sounds like a really advanced form of PWM used to generate some MEGA crunchy samples that don't take up as much room on a cartridge as standard PCM) Digital sounds is kinda hard to explain without visual and audio examples so I'll link a video that explains it better than I can here: th-cam.com/video/Gd_mhBf_FJA/w-d-xo.html
When using audio encoded as DPCM, the equivalent bit depth of the 2A03's DMC channel drops to 6 bits. That is, DPCM " steps " never touch the DACs least significant bit. You'd think that, with four channels with 4-bit DACs and a 7-bit DAC for PCM, 8 bits would be enough to represent the audio coming out of the NES. But, in reality each channel has its own separate DAC. The mixing between channels is done in the analog realm and has a lot of non-linearities, so you end up needing way more resolution than that to record NES music.
Bits in a CPU don't matter at all. Say you want to add two 32-bit numbers: it only takes 1 instruction on a modern x86-64 CPU. But how do you do that on an 8-bit CPU? You do it like you'd calculate that with pen and paper, take the first 8 bits, add them, see if there's overflow, take the next pair of 8 bits, add them, add overflow, etc. until you're done with all 32 bits. It will take a shitload of instructions instead of just 1, but you're not limited by the CPU architecture in any way. CPU clock rate will be an issue though, but a 3000 MHz 8-bit CPU will calculate things faster than a 1 MHz 32-bit CPU (unless our hypothetical 32-bit CPU can do 3000 operations in parallel in one cycle, which makes clock rate a very unreliable measure of performance too, but that's another story). Instead of bits, I just use chip names because 8 bits in the NES chip sound very different from 8 bits of C64. (honourable mention: extension chips like Konami VRC6)
imagine explaining some 19th century composers the idea of making music with a programmed sound chip from a portable electronic gaming device you can fit in your pocket
It's a sign of how little traction the Spectrum got outside the UK and Russia that an entire video about chiptunes doesn't mention it once. I never had any kind of games console growing up (or even since, actually), so 8-bit to me is synonymous with either the AY sound or the beeper manipulation of the Speccy 48k.
Wonderful first video essay! The flow and pacing felt appropriate. I'm not a chiptune musician, so the only feedback I can give is the intro could've leaned a bit more with the idea of limitations, as other comments have pointed out. Lovely work!
As someone who's made a lot of mostly fakebit music, I've enjoyed dabbling in the various limitations and sounds of other things. Heck, my most popular vid's a cover using Mega Man X samples, which...honestly, kinda surprising considering what's usually there. I've been working on a video game for 7 years at this point, but I think it's because of trying to break through the limitations provided to me in the fakebit music constructor BeepBox that I'm now using UltraBox to the best of my ability, having evolved to JummBox, Paandora'sBox, and now Ultra. I think honestly I've kinda seen this kind of thing myself, and now I'm making stupid music, sure, still, but now I'm doing so in nearly PlayStation 1 quality, and I think honestly that's kind of impressive. I feel like the growth there and the growth in the authentic chiptune scene are rather similar, but nonetheless extremely impressive what people can and will do
Inaccuracy: At 10:34 you said that video games tended to sacrifice some channels for sound effects, leaving less channels available for music. You gave Super Mario Bros. as an example of this, and showed that the Dpcm/Pcm channel were used for the sound effects in that game. If you actually take a listen, 1. the sound effects don't sound like samples at all and 2. the "breaking a brick" sound effect is made out of *noise* in the *noise channel* playing different frequencies in quick succesion. What's actually happening is that the 2 pulse wave channels and the noise channel all are responsible for playing different sound effects, and whenever a sound effect needs to be played in any one of those channels the music being played in that channel is interrupted, and the sound effect is played.
I've always wanted to see a video like this! I was so sick and tired of all these channels making "8 bit" covers and I was pissed that people thought it was actually legit! Great video!
Hello from France ! I don't know how you appeared in my recommandations with so few views and like but I am glad you did. This is a really great video !
As an outsider that knew nothing about how this stuff worked, I just called this sound "primitives" for triangle, saw, square, and "impulse" for samples and PWM. So, 8-bit purists would be devoted to "primitive waveform" music... All of this came from just messing around with various audio software. When I learned that FM just stands for "frequency modulation", I realized that all FM synths would still be using "primitives" and "impulses"... If you'll allow me to take a tangent; Then I learned about Fourier Transforms and my mind was blown to learn that these days almost all synthesizers are just sine-waves of different volume and phase... That was like 8 years ago, and music has just become much more fascinating since. I keep rediscovering things organically because I just experiment and fiddle with tools. I think my new favorite thing is "convolution". I have no idea how it works from a programming perspective, but it seems like it uses a sample of a loud click played in a specific location to capture how each frequency decays, acting as a filter, reverb, delay, etc based on a physical location, and it sounds amazing. Sometimes I try following tutorials about this stuff, but I always remember that the reason I like making music isn't the music itself, but rather the experimentation and self-learning.
Just wanted to say that I love the sound of the Gameboy via LSDJ, I've been listening to Nonokuso for years at this point even if it looks like he has disappeared...
As an 8-bit music enthusiast I found this analysis pretty interesting. Honestly I've never heard about this Realbit vs. Fakebit dilemma, mostly because I'm more into the Famitracker scene (which would be fakebit even if the tool is made to make music as close to the hardware as possible). For me the realbit line should be drawn for any tool, being real hardware or software emulation, which purposely follows all the core aspects of the original hardware and doesn't give the composer a way to "escape" from the original limitations. In any case, I thank the algorithm to letting me watch this amazing video!
@@jc_dogenYeah I was going to say i thought famitracker was almost the definition of realbit, although I'm definitely on the periphery of the scene. I haven't composed much lately, but I've been known to make original pieces for SPC hardware for the Super Nintendo. I really like the sound of even more retro games than that though, hopefully I get back into composition soon and I can try out some new styles 😊
Here's the thing - in this very video, Sean calls 'software emulation' fakebit, which is what Famitracker is doing. So there's this weird grey area produced by the definition where if you record the music from Famitracker, it would be 'fakebit', but if you export the NSF and stick it on a NES, it wouldn't be? I don't really like the fakebit definition used in the video, because it explicitly screws up stuff like Famitracker. Presumably, the big problem is calling it 'fake' - it's a highly volatile term. Edit: He does raise the question of 'playback through original hardware' during the conclusion, but I reckon it's an overly restrictive distinction to make...
@@doopuprime no, because famitracker is emulating the hardware, so a recording from famitracker is a very accurate simulation. fakebit refers to chiptune made without adherence to limitations and/or without attention paid to details. the gray area would be chiptune made in software that can't actually export for playback on real hardware, but adheres to limitations in a philosophical sense, not a practical one.
Sure, it's a great emulation, but it's still a software emulation of a hardware component - if you had something like an FPGA implementing the soundchip so it's hardware emulation of a piece of hardware, I think Sean would still argue that it's 'fakebit' under his definition. For what it's worth, we're largely on the same page - I think this 'fakebit' distinction is pretty unhealthy when it's applied to software that aims to perfectly capture the soundchips of the NES etc. And again, the bit at the end of the video tries to handwave it - there's this whole digression in LSDJ and 'making stuff using the tools of demoscene' which doesn't really sit great with me either.
It's always a tough question but at the end of the day I don't think the answer really matters all that much. Anamanaguchi is chiptune in my mind but considering they add so many other instruments it wouldn't fit under many other folks description. I make chiptune all the time, sometimes on hardware, sometimes with Plogue synths, sometimes with Raw waves in a wavetable synth. I think what Chiptune actually is has expanded greatly since the 'chiptune' era and terms like Fakebit only really serve to gatekeep and discourage others. We also can't discount the fact that not everyone can get a hold of a DMG, GBC, C64. Older consoles are getting harder and harder to find and parts are disappearing. It can be pretty costly and not everyone can afford that. Really appreciate everything you touched on in the video.
"I think what Chiptune actually is has expanded greatly since the 'chiptune' era and terms like Fakebit only really serve to gatekeep and discourage others." I suppose I consider my self a chiptune purist largely from nostalgia. For me "fakebit" that contains anything other than an accurate approximation of an actual period sounchip (note I didn't say real hardware, I don't count using the likes of Famitracker or composing music in an emulator as "fakebit") is more chiptune inspired electronic music. "We also can't discount the fact that not everyone can get a hold of a DMG, GBC, C64. Older consoles are getting harder and harder to find and parts are disappearing. It can be pretty costly and not everyone can afford that." VICE, DOSBOX and WinUAE (although that can be a bugger to set up) are there although you're still limiting yourself to period software, but we also have Famitracker and OpenMPT! What I am disappointed though is that people seem to think that chiptune means "sounding like the NES" or just using square, triangle, sawtooth and noise. This probably has a lot to do with Famitracker existing, but not a "SIDtracker" for a modern OS, especially when the C64 had a pretty sweet sound chip which could produce music that still holds up today! Heck, I'd like to see Famitracker (or a fork of Famitracker) allow you to add instuments based on your retro soundchip of choice! A 2A03 here, a POKEY there, add a little SID... call it Electric Orchestra (codename Speedwagon) and have the hardware geeks develop ways to network real hardware, use them as instruments and have it called Real Electric Orchestra!
Didnt know your channel before and by curiosity scrolling on yt i clicked on this video, turns out this was a very rewarding experience. I love chiptune music regardless of the time because i like any electronic music in general because of how much limitations you break trough to create different sounds, play at any speed, and pretty much like the ultimate music sandbox. However claim a certain ammount rules for a genre to be part of it, its as valid as dont care an keep snaking in and out of the rules to create something very special to someone.
"Back in the day" (circa 2002-2006) I used to frequent the "microdisco" monthly night club in Stockholm, Sweden. Johan Kotlinski, nitro2k1, Goto80, nullsleep, eat rabbit, firestARTer, Psilodump, and all then active and famous musicians on Game Boy, NES and C64 used to play. Cities with similar events were New York, Paris, Berlin and Tokyo; and artists from these cities used to fly in to play. We used to call music made by real 8-bit hardware in the style of rapid arpeggios "chip music" or "chiptunes". This also included simplet video game sounding music. The music that was made with the real 8-bit hardware but that sounded completely different was/is called "Bitpop". The appeal (for many and esp for me) was that a DMG Game Boy had a unique sound because of the very low and powerful bass. This bass was cut in future models to save battery. Also, you can't hear the entire bass if you activate the 75hz filter common on many audio mixers. So the idea was to visit a microdisco event and hear a DMG Game Boy monster bass connected to concert speakers. And it was always a surprise: would this evening be chip music with lots of arpeggios and music sounding like an old video game, or would it be a bitpop artist, making something completely different, using the 8bit equipment? A good example of this is "Huoratron - $$ Troopers" th-cam.com/video/PPwCmhvmHeM/w-d-xo.html music made with 8bit stuff sounding completely different than video game music. When people asked us why we would go to and dance to game boy music we would say that the game boy may be limited in what style of music it can play with only blips and blops, but... those blips and blops are made with infinite sound quality which is why it always sounded different live or on LP/vinyl than when you heard it as an MP3 file. This is also a distinction between musicians making "chip style" music with a computer, they're still making digital music, it can't compare with a DMG Game Boy, with an analogue chip and a PA speaker / subwoofer.
Haven't watched the video yet, but from what I recall, the term "chiptune" originated from the Amiga cracktro scene as just lightweight blippy music that resembled the computers prior such as the c64. Squares tris and saws were very easy to sample and fit on the floppy disk with compositions. It wasn't really adopted to be defined under a real thing til trackers became accessible for use with.. well, chips!
To describe the speed of a chip, you would use just Hz, not Hz per second. Hz per second would actually describe how fast a pitch changes if sliding between two pitches in your chiptune, or playing a glissando on a violin or trombone.
This is such a fantastic video! I was thinking of covering the topic myself, but you seem to have way more knowledge and insight than I do. There is one interesting idea I've had though. I think there's a difference between "genre" and "style", and I think it's interesting to look at genre as something more tied to culture, while style is more tied to the sound. Chiptune as a genre can be seen as the more purist scene, while the perception of chiptune nowadays is a style that can be applied to other genres (Drum & Bass, Brostep, etc). I think it helps to explain why there is so much confusion, not only around Chiptune, but music genres in general. Once again, this was such an amazing video that I've learnt a lot from. Thank you.
also another thing to consider, the term "chiptune" was originally coined very specifically to describe music on the commodore amiga that emulated the sound of the older 8 bit machines like the commodore 64, atari 8-bit series and to a lesser extent the NES and sega master system. chiptunes were mostly used in little intros for cracked software and used in the demo scene for their very small file size edit: i hadn't gotten that far in the video yet
I'll spare pointing out inaccuracies as others seem to have done that already. I definitely have to put myself in that "first gen" group you established here. One mark of someone like me is that we regurgitate our last meal when we are subjected to a MIDI file converted with GXSCC which not only produces fakebit out of MIDI but also desecrates the Konami SCC wavetable expansion used in many of the MSX computer games. After a bit of time you develop an ear for it, every little millisecond mistake you'll catch if it was inaccurate because it can come down to the exact implementation of a circuit, pseudo-random generator and how these behave and all these fine details can be ignored by people when trying to replicate the sound. As an example, consider that the NES Mini by Nintendo itself has sound inaccuracies despite it just being emulation of their perfectly researched and reverse-engineered NES. They got the periodic mode of the noise channel wrong, because the real NES plays this type of sound in a way that whatever was playing a normal static noise before influences the timbre you get in the periodic mode. The NES Mini likely uses a sampled version, which sounds the same any time it's played. I caught this when I was playing Mega Man 2 on it at a retro game exhibit and threw the Metal Blade weapon... which uses the aforementioned sound. It's enough to be caught and labelled as "eh, not so real" by a hardcore purist.
GXSCC "8-bit" is one of the most unlistenable sounds to me. When like... every instrument sounds like a sine wave. It's probably synesthesia but it sounds like how soggy, cold carrots feels to the mouth.
I’ve always just called the 8 bit era inspired music I made chiptune despite knowing it wasn’t true chiptune (like a program from LSDJ). I remember making 3xOscs in FL Studio that came close to having that synth sound, and would always limit myself to 3 instruments (bass, drums, melody/lead), as I was most familiar with music from 3 piece rock bands. I usually limited the lead to play 2 chords at most at a time, which matched the NES 4 channel limitation. To me chiptune was always about the heart and soul of the old limitations. I always attributed “8bit” and “16bit” music genres as the “true chiptune”. Also I used to compose TONS of music on Google’s Chrome Music Lab, as I loved how limiting it was as you could only use one instrument sample with a kick and snare drum. I especially loved the 8 bit sounding synth on it.
FYI "hertz" and exponents thereof are already "cycles per second". CPU frequency is measured in hertz, not hertz-per-second. It's much better to use "FLOPS" (FLoating point Operations Per Second) when comparing processors, because that takes into account not just the rate at which data moves through the system, but how much work it does with each step. That i9 doesn't just cycle 3350x as fast, it does exponentially more work per cycle (operating on 64 bit blocks of data versus 8 bit) and has all kinds of modern tools that make it able to do even more things per step! It really puts in perspective how miniscule the GameBoy's power is - it did less than 1 FLOP per cycle as it had no floating point instructions, so they had to be done manually by combining other instructions. It's also crazy to think that it was consistently cheaper to design custom chips containing analog synthesizers for every new console than to include enough memory and power to do the work in software for the Commodore and SNES. Those things are prohibitively expensive now, while just adding more memory for samples is practically free!
There are some wild takes in these comments jeez xD I thought this video was a reasonable attempt and I get that this was for a school project with limited time to do any research. I echo those who say that this video would have been better if the framing wasn't around the 'real chip vs 'fake' chip'' debate. As someone who both makes chiptunes in Famitracker that can be exported to a format that can be played on a real Famicom, and chiptunes in OpenMPT that utilise like 50 channels and incorporate completely non-chiptune instruments, I've always felt weird about this debate and I don't care for it very much. I think nowadays (at least primarily amongst people who make chiptunes) there's an implicit consensus that both the purist definition (chiptune as a medium/instrument) and the more modern/"chipster??" definition (chiptune as a genre/style) are valid. Also I think neo-chiptune is a neat term.
Thanks for being forgiving. I didn't mean to intentionally upset anyone through making this personal project but I can also understand their frustrations (especially if this is a community they've been a part of for a while). I've definitely learned a lot from these comments including yours! Thanks!
Fantastic video! I am someone who occasionally dabbles in fakebit, according to your definition, and I have tremendous respect for authentic chiptune artists who go through the effort of doing things the "right" way. I was always unsure how to label my 8-bit-style tracks, but now I'm a lot more confident about where I fit into the scene, thanks to your video!
Great video! It's good to have a video to point to whenever someone thinks they can make chiptune music just by throwing a bitcrusher on something. I used to write music in FamiTracker and it was interesting to work within the limitations of these simple waveforms and limited channels. Kudos~
soundtracker was not used to make chiptunes. it was the first tracker software, used on the amiga and is pure sample based. the amiga had 4 8 bit PCM channels, but no waveform generation capability. i like how you pointed out the memory and storage limitation of the time but playing even 8bit sound file would have taken does CPUs to its limit. my personal definition of a chiptune, if it uses any kind of realtime sound generation for instruments is a chiptune. everything else are sound tracker/screem tracker / octa tracker or noise tracker mod files. also the file size often gives it away as they are usually in the 1kb to 10kb range as they do not contain any sampling information.
As someone who loves a good 8-bit cover but has never really looked into it, this was a fascinating video! I loved that you took the time to talk about how composers dealt with the limitations of their hardware, the different modes, and the different chiptune subcultures. I also really appreciated the subtitles (though I think you could also upload the text file to the video itself, and let youtube sync it - that way it'll actually show up as having subtitles in searches!)
Two things about hardware: 1) The C64's SID chip also had a freely assignable analog filter, which made it more in line with analog subtractive synthesizers. 2) The audio capability of the Japanese NES (Famicom) could be expanded with on-cartridge sound chips, which immediately broke the 4 channel rule, with one game even adding FM sound synthesis capabilities. 3) Arcade hardware is very special, since they didn't had to conform to cost restrictions of a home console, which made certain things appear there way before the home market, and even then often a way more upmarket fashion.
Also, don't forget that the Commodore SID chip has an envelope generator, amplitude modulator and low/band/high pass filter per voice. As well as ring modulation and oscillator synchronization. With those combined as well as artists changing the parameters on fly made the ultimate sound chip.
Okay but here me out, if Cory in the Housr can be recognized as an anime by many, doesn't that mean you could get away with saying Weezer is authentic chiptune?
I'm really glad that you (someone) decided to make a video on this topic, but unfortunately this video has a lot of problems. I'm willing to write up a full review if you're interested.
It's a little bit more complicated than that 14:18 [a modern i9 is probably closer to six orders of magnitude faster than the [modified] 65010A that is found in the NES due to the fact that it has multiple cores, each of which can do multiple instructions for a cycle.
This is what humans do. We try to put everything on a shelf. Labbled and categorized. There are Multimillions of sounds, tastes, smells, and what have you to enjoy in this world. It doesn't need labels it doesn't need shelves and it definitely doesn't need gatekeepers. Chiptune? Fakebit? There is nobody with a trophy for gatekeeping music at the end of it all. Enjoy and bathe in the sound. In the end whether on original hardware or using a daw somebody worked hard on it and if you like it like it. Everything else is a flawed exercise in superiority. Also, Spongebob is an anime. I mean really have you seen some of the plots?
Correction: Even though the SNES is a "16-bit" system, its sound samples are actually 4 bits per sample(with a header byte every 16 samples, so technically 9 bytes per 16 samples), and its actual sound output is 32khz *8*-bit stereo.
The thing that puzzles me is that a person could use a DAW and emulated instruments, but also follow all of the limitations of the original hardware. What would that be considered? By most defintions it would be fakebit, but that's like comparing a duck to an ostrich just because they're both birds. Some people just don't have an N64 lying around, or coding ability. And what about the trackers?
In the early 90s Amiga scene, a chiptune was generally considered music made with samples so short that you could practically fit them onto an old sound chip. No reason that can't be applied to modern sample based trackers doing just the same. For hardware purists I'm certain this wouldn't be seen as real chiptune (historical example or not), but many also argue that this is where the term 'chiptune' originated from in the first place. I have no clue where they would stand on Amiga trackers like AHX that aimed to replicate a C64's SID synthesis and did not support any samples.
Following " all of the limitations of the original hardware " is really difficult in a DAW, *unless* you specifically use something like the Chipsynth plugins. Even having each channel in a different track in your DAW and adjusting their volume already violates one of the main limitations many of these chips had which is the very small dynamic range/volume control precision. A lot of " chiptune " plugins out there just generate pulse waves in the laziest ways possible and call it a day. *Actually* mimicking the sound of a NES, for instance, requires generating the audio using period counters at ~1.8 MHz, accurately emulating the weird way in which the NES mixes the audio channels and then downsampling everything. Anything else will generate tons of aliasing and not emulate how channels influence each other in the NES.
If you are able to nail every detail, how many volume levels are possible, the exact waveforms, the exact frequencies possible (oh boy, those are also limited based on how wide the related register of any sound chip is) and many many more fine details, then I'd say yeah, that's pretty real... except you made it in something that cannot export it in any way for an actual console to play it.
Hey everyone! I just wanted to address the inaccuracies in this video that people have made me aware of! This was a video presentation for a college course I took and I was only expecting my class of 10 people and my professor to see it instead of the 93K (at the time of writing this) that have viewed it. Yes, I could take the whole video down and re-upload an edited one but that will take up more time than I have right now.
Thank you to everyone who pointed out these inaccuracies!
Inaccuracies:
0:48 - 英語のアニメ (Eigo no anime) should be changed to アメリカのアニメ (Amerika no anime) meaning American Anime / American Animation
2:28 - Ninendo should be spelled Nintendo
4:56 - The oscilloscope that was used as an example used a 2A03 Mod and was not the original sound produced by the NES
5:46 - The Commodore 64 (C64) is not a dedicated video game console but rather an 8-bit home computer that had the ability to run video games
10:16 - I pronounced flanging wrong
12:08 - Church modes should really just say modes (even though modern modes derived from church modes)
14:03 - 6.0 GHz/second should just be 6.0 GHz
19:52 - Marylou should be spelled Marilou
Additional notes:
I put the subtitles in before knowing I could add the correct ones via. TH-cam.
I'm still learning how to edit my voice, I used a DeEsser and it sounded fine through my speakers but never test-ran it with headphones.
I didn't come up with the term "chipsters", it's a term mentioned in Marilou Polymeropoulou's article titled "Chipmusic, Fakebit, and the Discourse of Authenticity in the Chipscene".
My apologies for not including a warning for the flashing lights in the video!
Flat out disregard the bit-depth part.
Good presentation :)
To add one more nitpick: Core i9 is a (sub-)brand of processor, not a single product. Some i9 processors are specced for less than 5 GHz. Processors don't really have clock speeds themselves, but they are designed to be operated at a certain speed or range of speeds. Modern computers even slow down the clock on the fly if there isn't much work to be done. As for the speed at which it can process information, it's no longer as simple as comparing clock rates; modern CPUs can get much more done per clock-cycle as well, depending on the nature of the processing to be done.
I feel like eigo is fine in that context, because its not like only americans ever made animations. and english is synonymous with the west.
On voice editing - I recommend using a program to remove background noise (I know audacity has this as a feature). This will make it sound cleaner and make the transitions in, out of, and between voice clips more seamless
That's enough, no need to re-upload!
Thanks.
From my experience, when I hear people talking about "8-bit music" they usually means squares, triangles, saws and noise while "16-bit" seem to be commonly associated with fm synth stuff and some basic effects like echo (not reverb) and panning
to be fair, some 8 bit machines like the game boy had stereo support
When most people talk about 8-bit, they seem to be referring to the NES, not realizing that the NES wasn't the only 8-bit system that existed.
@@esmooth919when I hear chiptune I always think of the C64... That soundchip was something else man
Just think of Rob Hubbard... Especially "Masters of magic"
yeah, for most people 8-bit means NES chip, and 16-bit is usually Yamaha YM2612
@@SmplySilverThe SEGA Game Gear used stereo sound as well believe it or not despite the sound chip being shared with the SEGA Master System. I believe the TurboGrafx-16 also utilized stereo sound.
I've always just reserved "8-bit" for abiding by limitations, and "chiptune" for the general sound of a song produced with simple chiplike waveforms. I recognize it's arbitrary to use those terms that way, but it's still how I use them
that is such a based and reasonable take honestly
Personally a specific bit (8, 16, 32, etc) is to me abiding by limitations, whereas chiptune is just. Sequenced music almost in general to me, MIDI can be chiptune, but for actual consoles the cutoff is DSi for handhelds and N64 or so for home consoles if that makes any sense..
That’s how we talked about the music in the early 1990s Amiga scene. Chiptunes were tracker songs that took a very small amount of space and sounded like the old computers.
@@makipri And with MIDI files, well, it could sound like a whole orchestra with a good synth (cough cough roland ST-32) or like an old computer depending on what your setup was. I guess chiptune can at times be more of a general _feeling_ rather than a concrete set of limits, yeah?
@@jeffystreet General MIDI added a lot of uncertainty to the mix for sure. Before that it was somewhat reliable what it would sound like when playing back a song. Of course different SID chips could play the song back all wrong but still the soundscape would be pretty similar. Anyway, in the early 1990s there was no word for fakebit, not really 8-bit either. We just called fakebit chiptunes and the real 8-bit tunes with their original sound chip or platform names.
Small nitpick: Hertz means "cycles per second", so "hertz per second" are "cycles per second per second".
Ugh, you’re right!
@@spastorok I also found the focus on hertz to be unusual. I am given to understand that this way of measuring a CPU's processing speed is not very useful; at least these days. Perhaps someone with more knowledge can fill in the blanks.
I did get the feeling that you were writing beyond your knowledge when it comes to the computery stuff in general; is that a fair assessment?
Overall nice video. I particularly liked you drawing attention to how a word got imported from English to Japanese and then was imported back into English, undertaking a new meaning on its journey. Kinda nutty!
@@OctainServers until around the year 2000 (give or take several years), the Hz of a cpu was pretty good at telling the speed. since then the Hz has mostly stagnated, as the amount of power needed to push it faster isn't linear, so you're better off spending that extra power on parallel processing and other methods of speedup. also over time software had also improved, at least in the way that newer compilers work harder to make the same code run faster on the same hardware.
that said, it has always been where no matter how fast the cpu runs, some operations are simply slower than others. for example, you could have one computer that adds in 1 cycle and multiplies in 10, and another that takes 1 and 5 respectively. same Hz, but the second is faster, but only when multiplying. there's a lot more nuance but yeah it's really hard to compare speeds by looking at the hardware itself.
but in the end, if two computers have sufficiently different Hz, you can almost always expect that all the other variables favor the faster one anyways, so if anything it's an understatement.
@@rubixtheslime And the game changed with processors using multiple cores.
smy my head
Hey there!
I was the director of a video game that utilized a chiptune soundtrack for six years and wanted to share my experiences with this subject because it's something that I've struggled with for a very long time.
My game had a small team, but it was a relatively large teamm of composers. There were four of us for the majority of development, including myself as one of them, and the composers took up roughly a third of the total number of people involved with the project.
From the veyr beginning, I envisioned the game to have a chiptune soundtrack (specifically emulating the limitations of the Game Boy), and throughout development I spent a lot of time cracking the whip and shooting the team down when they got a little too ambitious. I had two schools of thought when trying to justify this, one philosophical and one practical.
The practical explanation ironically made more sense in theory than practice: it's so easy to find "8-Bit remixes" online that basically just throw a song into midi and replace every channel with a square wave and call it a day. Playing 8 square wave notes at once can sound somewhat retro, but it has this same inauthentic vibe as a lousy TV commercial trying to appeal to gamers with cheap pixel art and 90s buzzwords like "Level up!" I felt that by more faithfully following the limitations of a specific hardware, this sort of vibe could be avoided. In practice however, I often couldn't immediately tell the difference between a song with 3 voices and one with 5: the truth is that a composer with a genuine appreciation for the style will know how to use every channel with intent, and those channels serve a purpose.
The philosophical reason is something I still subscribe closer to: if you're not interested in emulating a specific style, why bother with any limitations at all? Usually chiptune sounds have objectively fewer options than modern composition hardware can provide, so why hold yourself back to only Squares and Saws when you can utilize any instrument sample in conjunction with the sounds you enjoy from these waveforms? The answer I've come to there is that chiptune music does have a specific style. The line of thinking that it comes from limitation is like asking a pianist why they don't add percussion to their music, or asking an electronic band why they limit themselves only to synthesized instruments. The truth is that there is no "rule" to what you can use in any genre of music, but nonetheless certain choices in instrument and composition will affect what kind of song you create, and choosing to use the voices of a particular hardware isn't an agreement to adhere to limitations, but merely a desire to utilize the sounds of that hardware, regardless of what it's actually capable of producing.
I still consider myself a purist in my own compositions, and I don't regret being so strict with my team, because we ultimately did produce an award winning soundtrack with a very distinct sound and I'd have it no other way. However, I acknowledge now that there are very genuine and valid reasons for fakebit to exist, and I think that other purists like me would benefit greatly from opening up to what those reasons might be. On the other side of the spectrum, I also think that fakebit "radicals" (that is to say, people who feel there should be no rules in chiptune at all) might benefit from understanding the perspectives around realbit better as well.
This was probably a super long read, but as a longtime game developer who's often struggled conveying this argument to his team members, as well as a chronic video essay enjoyer, I found this video to be especially relevant to me, and I thought others who are in a similar spot might also enjoy hearing my thoughts on the matter. Thank you for reading all of that!
This was an interesting read, thank you!
I can't call myself a deep fan of chiptune (as I'm not usually actively seeking the music in the genre), but I do enjoy both fake- and realbit with it being more than a half of my (admittedly patchwork-y) music taste.
I'm kinda interested in the game soundtrack and the game itself. What's its name?...if it's not a secret ( "^^)
@@V972 The game is called Axial Disc 1 (and its sequel Axial Disc 2!) The soundtrack is all up on TH-cam if you're interested!
For me, chiptune that closely follows the hardware's original limitations and chiptune that surpasses the limitations to create a more unique sound both have their own places.
I love listening to music that is so unique, you can't believe they're working under the limitations, but i also love the ones that just invokes the feeling "real" chiptune had, while not being held back by hardware limitations. It's not a "which is correct?" question, it's just a stylistic choice.
I gotta say though, i hate when people lazily put 8-bit sounds and noises on their music because it's "nostalgic", it only really sounds cringe.
Not to be reductionist but it's basically the difference between a purist and innovation, and somewhere inbetween thats lost to nostalgia, real or borrowed.
We see it all the time with demakes and "ps1" games these days. If we truly have ps1/n64 games again everyone would be miserable with the poor draw distance, terrible framerate, and somewhat endearingly artificial loading screens. Poor controls would be the cherry on top, because no one remembers that if you didn't have a dualshock controller back then you were playing ace combat and ape escape with a D-PAD.
Meanwhile a lot people seem to forget, or more likely don't realise, how much pre-rendered textures/scenes were used to overcome limitations. Donkeykong, Final Fantasy, Fear Effect come to mind.
Very few developers truly try to develop for the OG hardware but when they do it's an amazing technical feat.
Personally I've stopped caring. The internet is one giant mixing pot and it's down to individuals to learn about history and share those discoveries. And I super appreciate people who post videos breaking down the sound channels of somegames, going so far as to compare soundfonts from the various consoles.
This whole arguement is like saying someone is not a real movie director for using digital editing and not 8mm film. Its up to creator to use whatever tools they like to acheive their creation.
11:04 I'm no expert but Tim Follin is kind of legendary with how he used early console sound chips, it's kind of an extreme departure from his peers to use him as an example.
Slightly irrelevant but Tim Follin also created imo the best video game soundtrack of all time with Ecco the Dolphin: Defender of the Future. Not chiptune, tho I think it was done entirely with MIDI instruments
A departure from console music of the time perhaps, but in terms of technical "tricks" most of what he did in e.g. Solstice had been pioneered years before on home computers. I would say that his style was somewhat distinctly European and was only a departure in the sense that most video games around the late 80s had distinctly a Japanese style.
Follin's strength IMO is that although he makes good use of European style tricks like fast arpeggios and intricate parameter changes to embellish the sound with great sound design, he is also just an exceptional composer who used conventional composition techniques and ornamentation really well.
He is not an „extreme departure“. He comes from the European school of chiptune while most people are used to the Japanese style.
@@RyumaXtheXKing I'd love to hear other European chiptune composers with a sound like Tim's (besides Geoff Follin of course). Can you give some examples?
@@DustyMusician look up the soundtracks of european developed amiga games. there is tons of them! the amiga was primarily a european "console" so most games and music for it will originate there
At the end of the day, "Did I write good music?" is more important than "Did I write authentic music." Squares are square, and I like the way they sound. Creative limitation is great and important, but if I want to write a piece with tons of channels and lots of sound generation techniques, I will. But I'll still call it chiptune, because I've got to call it something.
It's great how you managed to provide viewers with precise definitions of technical terms while not leaving out important details and still keeping the definitions easy to understand. It also made the video dense with information, yet not boring at all. Hats off to you.
i clicked on this from recommended, watched the entire thing and didn’t realise how little views it had (800) until someone mentioned it in the comments omg. this is FIRE
OH SHIT YOU'RE RIGHT
This was awesome. All the things I always wanted to know about chiptune. That music imprinted on my soul along with pixelated graphics from my childhood era. Making the magic creatively despite the limitations made it feel like magic that you could deeply love. It wasn’t an easy thing to just churn out. We reveled in when it all came together and we felt it like magic. You rock! ❤✌🏾😎🔥
We are witnessing a legend being born.
8 bit universe is the “so retro” of chiptune music
"spongebob is not an anime" is the best conclusion here, I think
Chiptune composer here. I think a lot of the inaccuracies people have already pointed out would have been caught had you reached out to chiptune communities. In my opinion, one of the most important aspects of research projects like these is communication and collaboration with experienced people.
You're definitely not wrong! I'll keep it in mind if I decide to do a video essay again - this was just a final project for a college course I'm taking, so apologies for the unprofessionalism regarding the research.
Furnace server guy
Furnace server guy
I recognize you as a chiptune person but I have no recollection as to what you do
Furnace server guy
Nice video some feedback, for whatever it may be worth
the central question of the video and its title isnt really even asked until the end. A lot of the conclusion could have been the intro.
I dont think the intro fully worked because you talked about the history and evolution of anime, but not framed around limitations. This makes the thematic pivot to video game music limitation feels disconnected. This makes the video feel a bit unmotivated because (like i said) the central question isnt really asked until the end
It also felt like most of the video is spent on the history of video game sound tech, then you go through chiptune cultural history. But because Faketune and the subcultural history isnt mentioned until the second half, its hard to understand whats important to know about the tech history. I think the history lessons could have been combined
Overall, i think you need to better motivate in the beginning of the video why its being made, whats the central question/thesis as well as better sync it all into a single narrative. The editing and pacing was good, i did watch until the end!
Great first video i hope you make more.
Hey! Thanks for taking the time to provide feedback! This video is actually my final project for a music history course that I'm taking and I needed to choose something in the realm of video games (hence the chiptune focus). I definitely agree though with everything you said regarding the disconnection from limitations - I'm not much of a youtuber and this is only my fourth video using premiere pro. I'll keep these in mind if I decide to do any more videos in the future!
The limitations of NES hardware have brought out the more creative side of me and many other composers I've met. A lot of music nowadays relies very heavily on good sound design or tembre but with the NES everyone has to use the same sounds so the question now is how do you make your song stand out from the rest? I've discovered so much about music that i don't think I would have using a traditional DAW (not to discredit DAW composers). I'm so glad you covered this in this amazing video!
Of course YOU'RE here lmao
Honestly? I'm still jamming out to your original works, especially for a certain orange executable file and his compressed canine
The way you make your soundtrack stand out is by using extra chips inside the cartridge that give you extra sound channels
Cringe. "I think you are the best EVER". Obviously written by a guy who have no clue about oldskool chip music (8-bit home computer/consoles).
@@jmp01a24.Tim folin comes out of the dark.
Another chiptunist here. I really hate when people say they made a "Genesis cover" of something only for it to be a MIDI slap with Sonic 3's soundfont. No channel limitations whatsoever.
I’m glad you included the Pictionary title theme as an example. It goes so hard
I am a chiptune composer, and I also use a variety of DAW arrangers. I think the most important thing is how you view music. After all, producing good-sounding music is the most important, but if you are exposed to a little bit of chiptune, you also have a little music. Knowledge, you will find some of the mysteries and fun brought by chiptune is good. Of course, many retro sounds are now being widely integrated and used, but I think why everyone is arguing about this issue is mainly because of the so-called "fakebit" Among them, there are many bad works. They probably don’t understand things like “square” and “SAW”. They will use DAW to the extreme to perform large-scale EQ compression processing and stereo expansion. It is best to get a distorted but similar is about retro video game music, so I think the most fundamental question is how to produce good music that all people can appreciate
I feel like the meme of people importing entire album tabs to fl studio and replacing all the instruments with video game soundfonts with just a few clicks did a lot to mislead people online into thinking that it was done manually and/or that the import actually sounds good when they don't. Ngl that meme inspired me to get into using DAWs and learn music composition, but I do see better how low quality and lazy many of those uploads actually were.
For me chiptune will always mean "tracker chiptunes", which are basically tracker modules with very simple and small samples. That's what was called chiptunes in the 90s, at least in the demoscene. This modern usage referring to music made with PSG:s came much later.
You're absolutely right, and the author of this video is pretty badly wrong. His terminology is extremely revisionist and doesn't match the true roots of the term chiptune. C64 SID tunes were SID tunes. Etc. Only Soundtracker / Protracker etc. small sample tracker tunes were called chiptunes.
Same here. I wonder if there are any examples of usage of a 'chiptune' or 'chipmusic' terms before trackers came in. I mean - did people in the '80s even know these terms?
came here to post this as well. there’s even a link between this and the anime/animation intro:
the term chiptune started as a sample-based “borrowing” of PSG music (like japan borrowing western animation and naming it anime) without much regard for the authenticity.
the idea of debating what is or isn’t chiptune (or anime) came much later, in both cases, as did “chiptune-influenced” (or anime-influenced) media, sometime in the late 00s.
The original meaning of chiptunes predates trackers by about half a decade (specifically, by about as much as Commodore VIC-20 predates Amiga). It is exactly what the word says: Tunes written as code (yes, they had to write assembly code, all of those musicians, that was the skill game companies were paying for) for sound chips in 8-bit computers.
Edit for clarity: I grew up in the 80s and belong to the generation that migrated from 8-bits (C64) to 16-bits (A500). The original meaning was literally "code which generates music". You are, however, right that using the term "chiptune" for a specific musical style did emerge in the tracker scene, specifically to denote tracker mods which aped the sound and style of the music written for PSGs using samples.
@@gearwatcher You're saying the term "chiptune" was used already on the VIC-20? Until someone manages to refer to a usage that predates the tracker scene, I'll remain skeptic.
6:10 There were these absolute madlads I recall, demoscene programmers who managed to stufff an Amiga MOD into the c64 and got it play the mod back! Crazy stuff, very similar to the Genesis Toy Story title
On that note, the SID chip has amazing filters that allow musicians to create super unique sounds. Take a look at Eliminator on the C64. Jeroen Tel managed to use filters to get a super fat bass out
I make chiptunes (or fakebit) since the early 2000s. I was never really a purist but thought that at least some of what I did is considered chiptune.
Where do you stand on "classic Dos Modules" (xm, it, s3m and mod)? That's pretty much universally agreed "chiptune". It's somewhat sample-based and can be played on an old computer. Sounds like according to your definition that wouldn't be chiptune either.
But I'm really happy to see that a video about 'chiptune vs. fakebit' get so many views. Chiptune is so niche, so seeing this is really amazing.
My thoughts:
Chiptune music doesn't have to be made on old machines using archaic software or even manual assembly programming. However, the end result should either be able to be _played back_ on an original chip(s), or at the absolute very least conform to the limitations of an existing chip. Once you cross the barrier of just throwing waveforms together, that is simply synthesized or electronic music, which admittedly chiptunes are a subset of.
The limitation is what inherently separates chiptunes from other electronic music in my mind. A song that's simply made up of various waveforms you threw together in FL Studio is not a chiptune, unless those waveforms conform to the limitations of an existing chip, like a 2A03 or 6581/SID chip, or a combination of chips like was so commonly seen on the Famicom back in the day (2A03 + FDS, VRC6, Namco, etc).
@@mal2ksc Using multiple chips kinda blurs the line. It was a thing done back in the day by Famicom games and hardware hackers, but it still uses actual hardware synthesis chips and not just slapping waveforms together in FL Studio so I say it counts.
@@TheRedCapYou can use multiple synthesizer chips, but what about the cpu and rom/ram limitations?
@@dorukdogauysal8299 I'd say a 24-bit address space would be my limit for music "in a game" or 16-bit address space for "just music". These are arbitrary limitations going by my gut feeling.
If I use three square (or sawtooth) wave channels (no FM, glissando, or "unlimited" volume levels) working from a "library" of note frequencies spanning three octaves, should that still be considered somewhat authentic, even though created on a modern computer? Also would have a 64-bar sequence in (at maximum resolution) 16th notes in 4/4 time.
@@TheRedCap Well that sounds promising, i would like to hear some of your work if you are a musician yourself! Do you have a TH-cam channel that you post music stuff or a SoundCloud etc.?
I just like it when beeps and boops sound good.
I had my finger resting on the skip forward button the entire video, but never once felt compelled to press it. You kept holding my attention evenly througout the video. Bravo
I see Chiptune as "Music created by simple sound waves, Electronically." Or "music that sounds like it came from a sound chip."
I didn't see it mentioned in here, but I feel like a more important memory limitation to bring up was how much the soundchip could hold at a time. I definitely don't understand it fully, but the example that always made me realize this was a thing was the island theme in Startropics. I think most people know by now, the harmony track of that song was messed up, and someone had fied it like 20+ years later. He explained that the reason the track would die out and then start back up out of time, was because the code for the track was too long. As in, it contained more instructions than could be loaded into a single track for a song. He fixed it by adding in repeater instructions for parts that repeated, and removed just doubled up instructions for those parts. Which he said was a common trick at the time. I found that particularly fascinating because that had to play a HUGE part in how the songs were written. Either requiring massive amounts of repeated measures, or maybe even some form of swapping of the loaded instruction set (sort of how the later MMC chips would swap out loaded graphics rapidly mid screen draw to do cool stuff). But never really knew it because he was the only one I ever saw even mention this fact. Would love to know more about that specific limitation, because there is no way it didn't massively shape how these songs were written. Well, outside of the ones they just made super duper short as a whole to not worry about it.
Hardsynch.
The limit is not a hardware limit but just a limit of that particular (Startropics) sound player software implementation. Presumably the composer software did not have that limit since they put in more note data than what the player could then address. If i take a guess, the limit was likely due to the simplicity of using an 8-bit index in the player rather than memory constraints per se.
@@qwertypolis Interesting. The way I thought he made it sound was that you could only load a certain number of commands (I think it was 256 yeah) for a single song . I didn't think they had unique uhh, code for playing music though. I thought it was all the same on NES kind of like how graphics are always loaded the same. Just they started being able to load different sections of ROM into the console rapidly.
@@nickfarace9339 Old consumer architectures rarely have any fancy hardware controlled logic; usually the CPU is responsible for banging a bunch of hardware registers to make things happen. Not that I know anything about the NES in particular, but quick check gives that there are about four hw registers per voice to control things like frequency, volume etc. To play music, there needs to be a player routine written for the CPU that sets these registers to proper values at the correct time to play the song. Such information is then usually stored in some form of song data that the CPU routine reads. From the description of the Startropics player routine, it seems its song format was quite simple in that each hardware channel had allocated 256 bytes of song data. Usually such data includes commands such as effects (glissando, vibrato etc) beyond the bare note data, and apparently one of these commands for Startropics was a command to repeat notes or sections of notes. Sometimes a song format is standardized, for example, for the original Playstation Sony specified a SEQ format that was basically MIDI + sound samples, which was included in the SDKs together with tools that game developer could use, but for the early systems, it was usually up to the developers themselves to come up with their own thing.
@@qwertypolis alright this makes a ton of sense thanks!
That track from Dominator 6:27 sounded immensely more advanced than it should've
is there any video of the song on youtube? Found some guy with gameplay and songs but the songs arent right
edit: Found it
th-cam.com/video/-GBvq609sW0/w-d-xo.html
I used to make tunes for crack intros in the 80s, 90s, 2000s, and for demos in 2010s and 2020s. Most are made with Amiga, but I've also used synths with modern daws to create tunes. I didn't know there is such clear cut division or generations, realbit and fakebit. To me it is just the aesthetics as well as, to some extent, the restrictions (but less so).
Edit: nevertheless, I enjoyed your take on chip tunes!
That's fantastic! I love tracker music, it's one of my inspirations as an amateur composer.
that’s sick i wanna hear
As a chiptune composer myself, I think that chiptune music is a style rather than a way of composing. Once you start working with trackers you start to learn techniques and ways to overcome the limitations and the music start to sound close to the so call chiptune style. That could be easily translated using modern hardware, but there's certainly a specific style, specially in modern LSDJ compositions that is different from old chiptunes and there are many modern chiptune composers using the same style with moderns hardware.
Is it chiptune just because you make a song using a tracker that at the end of the day IS NOT real hardware and is emulating the real chip? I don't know, but I'm certain that chiptune music is more a style and a technique rather than the tools used to make it.
spot-on. The genres were always needed to highlight the aesthetics, using them to create divisions based on technicalities is a really dick move. It's like saying there's "real" metal music played with electric guitars with metal strings, and "fake" metal music played with everything else. Absurd IMVHO.
you got some technical info messed up in the "what is 8-bit" section. The NES does have an 8-bit CPU but that doesn't have any bearing on the bit depth of the audio hardware, its two separate things. I'm an audio engineer not a computer scientist, so my explanation here might be a little weak, but when someone refers to an 8-bit CPU that means that it can only address 8 bits in a single cycle, where the bit depth of digital audio refers to the maximum number of amplitude levels a sample can exist at. The NES sound chip works with 4-bit resolution for the Pulse, Triangle, and Noise channels, and the Sample channel can play either 7-bit PCM samples or 1-bit samples using Delta Modulation (tbh i am unfamiliar with DM, but based on what i skimmed out of this tech sheet i found for the 2A03 chip it sounds like a really advanced form of PWM used to generate some MEGA crunchy samples that don't take up as much room on a cartridge as standard PCM)
Digital sounds is kinda hard to explain without visual and audio examples so I'll link a video that explains it better than I can here: th-cam.com/video/Gd_mhBf_FJA/w-d-xo.html
When using audio encoded as DPCM, the equivalent bit depth of the 2A03's DMC channel drops to 6 bits. That is, DPCM " steps " never touch the DACs least significant bit.
You'd think that, with four channels with 4-bit DACs and a 7-bit DAC for PCM, 8 bits would be enough to represent the audio coming out of the NES. But, in reality each channel has its own separate DAC. The mixing between channels is done in the analog realm and has a lot of non-linearities, so you end up needing way more resolution than that to record NES music.
Bits in a CPU don't matter at all. Say you want to add two 32-bit numbers: it only takes 1 instruction on a modern x86-64 CPU. But how do you do that on an 8-bit CPU? You do it like you'd calculate that with pen and paper, take the first 8 bits, add them, see if there's overflow, take the next pair of 8 bits, add them, add overflow, etc. until you're done with all 32 bits.
It will take a shitload of instructions instead of just 1, but you're not limited by the CPU architecture in any way. CPU clock rate will be an issue though, but a 3000 MHz 8-bit CPU will calculate things faster than a 1 MHz 32-bit CPU (unless our hypothetical 32-bit CPU can do 3000 operations in parallel in one cycle, which makes clock rate a very unreliable measure of performance too, but that's another story).
Instead of bits, I just use chip names because 8 bits in the NES chip sound very different from 8 bits of C64. (honourable mention: extension chips like Konami VRC6)
imagine explaining some 19th century composers the idea of making music with a programmed sound chip from a portable electronic gaming device you can fit in your pocket
It's a sign of how little traction the Spectrum got outside the UK and Russia that an entire video about chiptunes doesn't mention it once. I never had any kind of games console growing up (or even since, actually), so 8-bit to me is synonymous with either the AY sound or the beeper manipulation of the Speccy 48k.
Wonderful first video essay! The flow and pacing felt appropriate. I'm not a chiptune musician, so the only feedback I can give is the intro could've leaned a bit more with the idea of limitations, as other comments have pointed out. Lovely work!
Tim Follin is one of the most insanely talented composers out there for sure!
I still go back to the Silver Surfer NES OST once in a while.
Goat plok
Dang, "Dominator" was a super heavy tune. That was some metal stuff back in the 80s!😂
As someone who's made a lot of mostly fakebit music, I've enjoyed dabbling in the various limitations and sounds of other things. Heck, my most popular vid's a cover using Mega Man X samples, which...honestly, kinda surprising considering what's usually there. I've been working on a video game for 7 years at this point, but I think it's because of trying to break through the limitations provided to me in the fakebit music constructor BeepBox that I'm now using UltraBox to the best of my ability, having evolved to JummBox, Paandora'sBox, and now Ultra. I think honestly I've kinda seen this kind of thing myself, and now I'm making stupid music, sure, still, but now I'm doing so in nearly PlayStation 1 quality, and I think honestly that's kind of impressive. I feel like the growth there and the growth in the authentic chiptune scene are rather similar, but nonetheless extremely impressive what people can and will do
Inaccuracy:
At 10:34 you said that video games tended to sacrifice some channels for sound effects, leaving less channels available for music.
You gave Super Mario Bros. as an example of this, and showed that the Dpcm/Pcm channel were used for the sound effects in that game.
If you actually take a listen, 1. the sound effects don't sound like samples at all and 2. the "breaking a brick" sound effect is made out of *noise* in the *noise channel* playing different frequencies in quick succesion.
What's actually happening is that the 2 pulse wave channels and the noise channel all are responsible for playing different sound effects, and whenever a sound effect needs to be played in any one of those channels the music being played in that channel is interrupted, and the sound effect is played.
I've always wanted to see a video like this! I was so sick and tired of all these channels making "8 bit" covers and I was pissed that people thought it was actually legit! Great video!
Hello from France ! I don't know how you appeared in my recommandations with so few views and like but I am glad you did. This is a really great video !
Great video! This was informative and entertaining.
As an outsider that knew nothing about how this stuff worked, I just called this sound "primitives" for triangle, saw, square, and "impulse" for samples and PWM. So, 8-bit purists would be devoted to "primitive waveform" music... All of this came from just messing around with various audio software. When I learned that FM just stands for "frequency modulation", I realized that all FM synths would still be using "primitives" and "impulses"...
If you'll allow me to take a tangent;
Then I learned about Fourier Transforms and my mind was blown to learn that these days almost all synthesizers are just sine-waves of different volume and phase... That was like 8 years ago, and music has just become much more fascinating since. I keep rediscovering things organically because I just experiment and fiddle with tools. I think my new favorite thing is "convolution". I have no idea how it works from a programming perspective, but it seems like it uses a sample of a loud click played in a specific location to capture how each frequency decays, acting as a filter, reverb, delay, etc based on a physical location, and it sounds amazing. Sometimes I try following tutorials about this stuff, but I always remember that the reason I like making music isn't the music itself, but rather the experimentation and self-learning.
Just wanted to say that I love the sound of the Gameboy via LSDJ, I've been listening to Nonokuso for years at this point even if it looks like he has disappeared...
4:58 genuinely never noticed chiptune used vibrato up until now…
As an 8-bit music enthusiast I found this analysis pretty interesting. Honestly I've never heard about this Realbit vs. Fakebit dilemma, mostly because I'm more into the Famitracker scene (which would be fakebit even if the tool is made to make music as close to the hardware as possible). For me the realbit line should be drawn for any tool, being real hardware or software emulation, which purposely follows all the core aspects of the original hardware and doesn't give the composer a way to "escape" from the original limitations.
In any case, I thank the algorithm to letting me watch this amazing video!
famitracker isn't fakebit, the entire point is that it emulates the hardware and can export to it. unless you're using multiple expansions at once.
@@jc_dogenYeah I was going to say i thought famitracker was almost the definition of realbit, although I'm definitely on the periphery of the scene.
I haven't composed much lately, but I've been known to make original pieces for SPC hardware for the Super Nintendo. I really like the sound of even more retro games than that though, hopefully I get back into composition soon and I can try out some new styles 😊
Here's the thing - in this very video, Sean calls 'software emulation' fakebit, which is what Famitracker is doing. So there's this weird grey area produced by the definition where if you record the music from Famitracker, it would be 'fakebit', but if you export the NSF and stick it on a NES, it wouldn't be? I don't really like the fakebit definition used in the video, because it explicitly screws up stuff like Famitracker.
Presumably, the big problem is calling it 'fake' - it's a highly volatile term.
Edit: He does raise the question of 'playback through original hardware' during the conclusion, but I reckon it's an overly restrictive distinction to make...
@@doopuprime no, because famitracker is emulating the hardware, so a recording from famitracker is a very accurate simulation. fakebit refers to chiptune made without adherence to limitations and/or without attention paid to details. the gray area would be chiptune made in software that can't actually export for playback on real hardware, but adheres to limitations in a philosophical sense, not a practical one.
Sure, it's a great emulation, but it's still a software emulation of a hardware component - if you had something like an FPGA implementing the soundchip so it's hardware emulation of a piece of hardware, I think Sean would still argue that it's 'fakebit' under his definition.
For what it's worth, we're largely on the same page - I think this 'fakebit' distinction is pretty unhealthy when it's applied to software that aims to perfectly capture the soundchips of the NES etc.
And again, the bit at the end of the video tries to handwave it - there's this whole digression in LSDJ and 'making stuff using the tools of demoscene' which doesn't really sit great with me either.
It's always a tough question but at the end of the day I don't think the answer really matters all that much. Anamanaguchi is chiptune in my mind but considering they add so many other instruments it wouldn't fit under many other folks description. I make chiptune all the time, sometimes on hardware, sometimes with Plogue synths, sometimes with Raw waves in a wavetable synth. I think what Chiptune actually is has expanded greatly since the 'chiptune' era and terms like Fakebit only really serve to gatekeep and discourage others.
We also can't discount the fact that not everyone can get a hold of a DMG, GBC, C64. Older consoles are getting harder and harder to find and parts are disappearing. It can be pretty costly and not everyone can afford that.
Really appreciate everything you touched on in the video.
"I think what Chiptune actually is has expanded greatly since the 'chiptune' era and terms like Fakebit only really serve to gatekeep and discourage others."
I suppose I consider my self a chiptune purist largely from nostalgia. For me "fakebit" that contains anything other than an accurate approximation of an actual period sounchip (note I didn't say real hardware, I don't count using the likes of Famitracker or composing music in an emulator as "fakebit") is more chiptune inspired electronic music.
"We also can't discount the fact that not everyone can get a hold of a DMG, GBC, C64. Older consoles are getting harder and harder to find and parts are disappearing. It can be pretty costly and not everyone can afford that."
VICE, DOSBOX and WinUAE (although that can be a bugger to set up) are there although you're still limiting yourself to period software, but we also have Famitracker and OpenMPT!
What I am disappointed though is that people seem to think that chiptune means "sounding like the NES" or just using square, triangle, sawtooth and noise. This probably has a lot to do with Famitracker existing, but not a "SIDtracker" for a modern OS, especially when the C64 had a pretty sweet sound chip which could produce music that still holds up today!
Heck, I'd like to see Famitracker (or a fork of Famitracker) allow you to add instuments based on your retro soundchip of choice! A 2A03 here, a POKEY there, add a little SID... call it Electric Orchestra (codename Speedwagon) and have the hardware geeks develop ways to network real hardware, use them as instruments and have it called Real Electric Orchestra!
Great points! @@GeoNeilUK
Very interesting video! My favorite part was the use of the Gameboy only as a music player, so cool!
Didnt know your channel before and by curiosity scrolling on yt i clicked on this video, turns out this was a very rewarding experience. I love chiptune music regardless of the time because i like any electronic music in general because of how much limitations you break trough to create different sounds, play at any speed, and pretty much like the ultimate music sandbox. However claim a certain ammount rules for a genre to be part of it, its as valid as dont care an keep snaking in and out of the rules to create something very special to someone.
"Back in the day" (circa 2002-2006) I used to frequent the "microdisco" monthly night club in Stockholm, Sweden. Johan Kotlinski, nitro2k1, Goto80, nullsleep, eat rabbit, firestARTer, Psilodump, and all then active and famous musicians on Game Boy, NES and C64 used to play. Cities with similar events were New York, Paris, Berlin and Tokyo; and artists from these cities used to fly in to play.
We used to call music made by real 8-bit hardware in the style of rapid arpeggios "chip music" or "chiptunes". This also included simplet video game sounding music.
The music that was made with the real 8-bit hardware but that sounded completely different was/is called "Bitpop".
The appeal (for many and esp for me) was that a DMG Game Boy had a unique sound because of the very low and powerful bass. This bass was cut in future models to save battery. Also, you can't hear the entire bass if you activate the 75hz filter common on many audio mixers. So the idea was to visit a microdisco event and hear a DMG Game Boy monster bass connected to concert speakers. And it was always a surprise: would this evening be chip music with lots of arpeggios and music sounding like an old video game, or would it be a bitpop artist, making something completely different, using the 8bit equipment?
A good example of this is "Huoratron - $$ Troopers" th-cam.com/video/PPwCmhvmHeM/w-d-xo.html music made with 8bit stuff sounding completely different than video game music.
When people asked us why we would go to and dance to game boy music we would say that the game boy may be limited in what style of music it can play with only blips and blops, but... those blips and blops are made with infinite sound quality which is why it always sounded different live or on LP/vinyl than when you heard it as an MP3 file. This is also a distinction between musicians making "chip style" music with a computer, they're still making digital music, it can't compare with a DMG Game Boy, with an analogue chip and a PA speaker / subwoofer.
Haven't watched the video yet, but from what I recall, the term "chiptune" originated from the Amiga cracktro scene as just lightweight blippy music that resembled the computers prior such as the c64. Squares tris and saws were very easy to sample and fit on the floppy disk with compositions. It wasn't really adopted to be defined under a real thing til trackers became accessible for use with.. well, chips!
I woke up on New Year’s Day and watch 22 minutes of me not understanding a single thing. Great video
Really good video. I struggle to stay interested in a video for over 15 minutes without saving it for later, but I just kept watching this.
To describe the speed of a chip, you would use just Hz, not Hz per second. Hz per second would actually describe how fast a pitch changes if sliding between two pitches in your chiptune, or playing a glissando on a violin or trombone.
This is such a fantastic video! I was thinking of covering the topic myself, but you seem to have way more knowledge and insight than I do.
There is one interesting idea I've had though. I think there's a difference between "genre" and "style", and I think it's interesting to look at genre as something more tied to culture, while style is more tied to the sound. Chiptune as a genre can be seen as the more purist scene, while the perception of chiptune nowadays is a style that can be applied to other genres (Drum & Bass, Brostep, etc). I think it helps to explain why there is so much confusion, not only around Chiptune, but music genres in general.
Once again, this was such an amazing video that I've learnt a lot from. Thank you.
Amazing essay!! Definitely subscribing, that ending destroyed me lmao
All praise the changes in the algorithm that brought this nifty video to my attention.
I love gatekeeping random things. "It's not authentic" just let people create what they want and call it what they want.
thank you immensely for embedding subtitles into your video. it makes the video so much easier and nicer to watch
If been waiting for a video like that for ages. It was worth it
also another thing to consider, the term "chiptune" was originally coined very specifically to describe music on the commodore amiga that emulated the sound of the older 8 bit machines like the commodore 64, atari 8-bit series and to a lesser extent the NES and sega master system. chiptunes were mostly used in little intros for cracked software and used in the demo scene for their very small file size
edit: i hadn't gotten that far in the video yet
Such an incredible video
I'll spare pointing out inaccuracies as others seem to have done that already. I definitely have to put myself in that "first gen" group you established here. One mark of someone like me is that we regurgitate our last meal when we are subjected to a MIDI file converted with GXSCC which not only produces fakebit out of MIDI but also desecrates the Konami SCC wavetable expansion used in many of the MSX computer games.
After a bit of time you develop an ear for it, every little millisecond mistake you'll catch if it was inaccurate because it can come down to the exact implementation of a circuit, pseudo-random generator and how these behave and all these fine details can be ignored by people when trying to replicate the sound. As an example, consider that the NES Mini by Nintendo itself has sound inaccuracies despite it just being emulation of their perfectly researched and reverse-engineered NES. They got the periodic mode of the noise channel wrong, because the real NES plays this type of sound in a way that whatever was playing a normal static noise before influences the timbre you get in the periodic mode. The NES Mini likely uses a sampled version, which sounds the same any time it's played. I caught this when I was playing Mega Man 2 on it at a retro game exhibit and threw the Metal Blade weapon... which uses the aforementioned sound.
It's enough to be caught and labelled as "eh, not so real" by a hardcore purist.
GXSCC "8-bit" is one of the most unlistenable sounds to me. When like... every instrument sounds like a sine wave. It's probably synesthesia but it sounds like how soggy, cold carrots feels to the mouth.
Great video dude! I'm so surprised this only has 7k views! I'm definitely gonna check out more content from you.
Masterpiece!! Thank you for creating this video ❤
I’ve always just called the 8 bit era inspired music I made chiptune despite knowing it wasn’t true chiptune (like a program from LSDJ). I remember making 3xOscs in FL Studio that came close to having that synth sound, and would always limit myself to 3 instruments (bass, drums, melody/lead), as I was most familiar with music from 3 piece rock bands. I usually limited the lead to play 2 chords at most at a time, which matched the NES 4 channel limitation.
To me chiptune was always about the heart and soul of the old limitations. I always attributed “8bit” and “16bit” music genres as the “true chiptune”.
Also I used to compose TONS of music on Google’s Chrome Music Lab, as I loved how limiting it was as you could only use one instrument sample with a kick and snare drum. I especially loved the 8 bit sounding synth on it.
Great video
Absolutely amazing video
FYI "hertz" and exponents thereof are already "cycles per second". CPU frequency is measured in hertz, not hertz-per-second. It's much better to use "FLOPS" (FLoating point Operations Per Second) when comparing processors, because that takes into account not just the rate at which data moves through the system, but how much work it does with each step. That i9 doesn't just cycle 3350x as fast, it does exponentially more work per cycle (operating on 64 bit blocks of data versus 8 bit) and has all kinds of modern tools that make it able to do even more things per step! It really puts in perspective how miniscule the GameBoy's power is - it did less than 1 FLOP per cycle as it had no floating point instructions, so they had to be done manually by combining other instructions.
It's also crazy to think that it was consistently cheaper to design custom chips containing analog synthesizers for every new console than to include enough memory and power to do the work in software for the Commodore and SNES. Those things are prohibitively expensive now, while just adding more memory for samples is practically free!
There are some wild takes in these comments jeez xD
I thought this video was a reasonable attempt and I get that this was for a school project with limited time to do any research. I echo those who say that this video would have been better if the framing wasn't around the 'real chip vs 'fake' chip'' debate.
As someone who both makes chiptunes in Famitracker that can be exported to a format that can be played on a real Famicom, and chiptunes in OpenMPT that utilise like 50 channels and incorporate completely non-chiptune instruments, I've always felt weird about this debate and I don't care for it very much. I think nowadays (at least primarily amongst people who make chiptunes) there's an implicit consensus that both the purist definition (chiptune as a medium/instrument) and the more modern/"chipster??" definition (chiptune as a genre/style) are valid.
Also I think neo-chiptune is a neat term.
Thanks for being forgiving. I didn't mean to intentionally upset anyone through making this personal project but I can also understand their frustrations (especially if this is a community they've been a part of for a while). I've definitely learned a lot from these comments including yours! Thanks!
crazy high quality video, genuinely fascinating
The mode ("church mode") point is true of any music & composition. The TH-cam channel "8-bit Music" just uses this genre of VGM to teach music theory.
This video slaps. Keep up the good work!
Shocked to see a nice video essay in my recommended, especially a lesser known one. Nice work, guy.
Thanks, guy.
6:14 This is based on the score for The Transformers: The Movie (1986)
14:05 "GHz per second" 💀💀
This is a fantastic documentary, thank you.
Don't trust this video, it has a lot of disinformation about chiptunes and poorly research
Fantastic video! I am someone who occasionally dabbles in fakebit, according to your definition, and I have tremendous respect for authentic chiptune artists who go through the effort of doing things the "right" way. I was always unsure how to label my 8-bit-style tracks, but now I'm a lot more confident about where I fit into the scene, thanks to your video!
Great video! It's good to have a video to point to whenever someone thinks they can make chiptune music just by throwing a bitcrusher on something. I used to write music in FamiTracker and it was interesting to work within the limitations of these simple waveforms and limited channels. Kudos~
This video was a pleasure to watch. Major throwback.
I had a fun time with this! Nice stat to a channel!
"Church modes" is crazy
Great vid bud! Algorithm brought me here and I'm glad
Amazing video!
This video is so well done!
soundtracker was not used to make chiptunes. it was the first tracker software, used on the amiga and is pure sample based. the amiga had 4 8 bit PCM channels, but no waveform generation capability.
i like how you pointed out the memory and storage limitation of the time but playing even 8bit sound file would have taken does CPUs to its limit.
my personal definition of a chiptune, if it uses any kind of realtime sound generation for instruments is a chiptune. everything else are sound tracker/screem tracker / octa tracker or noise tracker mod files. also the file size often gives it away as they are usually in the 1kb to 10kb range as they do not contain any sampling information.
As someone who loves a good 8-bit cover but has never really looked into it, this was a fascinating video! I loved that you took the time to talk about how composers dealt with the limitations of their hardware, the different modes, and the different chiptune subcultures. I also really appreciated the subtitles (though I think you could also upload the text file to the video itself, and let youtube sync it - that way it'll actually show up as having subtitles in searches!)
Fantastic video essay! This is informative, factual, and engaging, I love it.
This is really well done, nice job
Two things about hardware:
1) The C64's SID chip also had a freely assignable analog filter, which made it more in line with analog subtractive synthesizers.
2) The audio capability of the Japanese NES (Famicom) could be expanded with on-cartridge sound chips, which immediately broke the 4 channel rule, with one game even adding FM sound synthesis capabilities.
3) Arcade hardware is very special, since they didn't had to conform to cost restrictions of a home console, which made certain things appear there way before the home market, and even then often a way more upmarket fashion.
Also, don't forget that the Commodore SID chip has an envelope generator, amplitude modulator and low/band/high pass filter per voice. As well as ring modulation and oscillator synchronization. With those combined as well as artists changing the parameters on fly made the ultimate sound chip.
Okay but here me out, if Cory in the Housr can be recognized as an anime by many, doesn't that mean you could get away with saying Weezer is authentic chiptune?
This video is epic and underrated, also you are a hero for adding sources
Loved this video awesome work.
I'm really glad that you (someone) decided to make a video on this topic, but unfortunately this video has a lot of problems. I'm willing to write up a full review if you're interested.
It's a little bit more complicated than that 14:18 [a modern i9 is probably closer to six orders of magnitude faster than the [modified] 65010A that is found in the NES due to the fact that it has multiple cores, each of which can do multiple instructions for a cycle.
This is what humans do. We try to put everything on a shelf. Labbled and categorized. There are Multimillions of sounds, tastes, smells, and what have you to enjoy in this world. It doesn't need labels it doesn't need shelves and it definitely doesn't need gatekeepers. Chiptune? Fakebit? There is nobody with a trophy for gatekeeping music at the end of it all. Enjoy and bathe in the sound. In the end whether on original hardware or using a daw somebody worked hard on it and if you like it like it. Everything else is a flawed exercise in superiority. Also, Spongebob is an anime. I mean really have you seen some of the plots?
Chirpy Chips would like a word
came up in my recommended and hardly ever give lower view videos a try, but this is great it hits all the marks
Heyy that's a pretty good video! I'd only drop the subtitles pasted into the video. Great job, loved the ending :D
Correction: Even though the SNES is a "16-bit" system, its sound samples are actually 4 bits per sample(with a header byte every 16 samples, so technically 9 bytes per 16 samples), and its actual sound output is 32khz *8*-bit stereo.
The thing that puzzles me is that a person could use a DAW and emulated instruments, but also follow all of the limitations of the original hardware. What would that be considered? By most defintions it would be fakebit, but that's like comparing a duck to an ostrich just because they're both birds. Some people just don't have an N64 lying around, or coding ability.
And what about the trackers?
In the early 90s Amiga scene, a chiptune was generally considered music made with samples so short that you could practically fit them onto an old sound chip. No reason that can't be applied to modern sample based trackers doing just the same.
For hardware purists I'm certain this wouldn't be seen as real chiptune (historical example or not), but many also argue that this is where the term 'chiptune' originated from in the first place. I have no clue where they would stand on Amiga trackers like AHX that aimed to replicate a C64's SID synthesis and did not support any samples.
Following " all of the limitations of the original hardware " is really difficult in a DAW, *unless* you specifically use something like the Chipsynth plugins.
Even having each channel in a different track in your DAW and adjusting their volume already violates one of the main limitations many of these chips had which is the very small dynamic range/volume control precision.
A lot of " chiptune " plugins out there just generate pulse waves in the laziest ways possible and call it a day. *Actually* mimicking the sound of a NES, for instance, requires generating the audio using period counters at ~1.8 MHz, accurately emulating the weird way in which the NES mixes the audio channels and then downsampling everything.
Anything else will generate tons of aliasing and not emulate how channels influence each other in the NES.
If you are able to nail every detail, how many volume levels are possible, the exact waveforms, the exact frequencies possible (oh boy, those are also limited based on how wide the related register of any sound chip is) and many many more fine details, then I'd say yeah, that's pretty real... except you made it in something that cannot export it in any way for an actual console to play it.
The music at 14:31 is the Ice Path theme from Pokémon Gold/Silver/Crystal!