In the spirit of bfv, my favorite Eddie Van Halen moment, Greensboro 2007, where the Tascam Portadat was accidentally switched from 44.1kHz to 48kHz, thus forcing an encore of "Jump" to be played impromptu at the experimental temperament of A=478.911Hz, or approximately "A sharp and then some." Eat your heart out, Merzbow.
Omg if I'm Zooming someone and I route audio through the Zoom audio driver so I can pipe in my computer audio straight to the Zoom attendees, this is exactly what happens. The first time it happened, I didn't know I had to switch it back from 48 kHz manually, and then when I loaded a digital tuner into Ableton, it was like "every single one of your strings is flat (or sharp, can't remember which) by like 1.5 semi-tones", and I chalked it up to thermal shenanigans. Had recently changed strings. Then I started trying to write a digital piano part in the same key that I thought it was in based on the guitar part's fretboard positions... Omg it took me like an hour to realize that my sample rate was the culprit. A sample rate set by Zoom automatically. It's almost like audio engineering is a legit job, "But since I can hear audio, I think I totally understand it." I'm sure you totally can't relate. ;) ;) ;) Gotta respect Eddie's talent. I could try to guitar like he did, but, as reality would have it, I am me and not he.
Yeah I shoulda said DA-30 Mk.II because that's undoubtedly what it was. It almost always is. Either that or SV-3700 but the DA-30s were black and therefore cooler. Can you... find the troublemaker? General audio practice is "more samples are better" and also "if this ever touches video holy god it better be 48kHz" so it would make sense that the tape is going to be 48, not 44.1 but, you know. CDs. What's funny is that computer audio has absolutely no idea what to do with external clock other than go "well I guess the world is 11% faster for some reason." When I feed the laptop, which usually plays Tidal, without messing with the word clock it'll play fine but fast for about a song and a half and then start skipping as its buffer goes "WTF are you doing". Amusingly enough I also run watch/clock timing software on that one and if it gets a different word clock on its pro audio interface than, for example, Zoom thinks it should be at, the software gets so upset that it dumps its log files, erases its authorization and resets its preferences. It literally has a stroke. Which is extra dumb because it's got provisions for external wordclock in, but if it's not on a BNC (because the author has it on BNC) it loses its fucking mind. To be fair to Tascam, most of their stuff won't slave to any old word clock. It will go "hey buddy - you told me we were at 44.1? And we're at 48? So... I'm just going to flash my display at you angrily and refuse to pass audio until you remedy the situation." And to punish Avid/Digidesign, they decided that a 44.1/48kHz word clock was wholly inadequate to their purposes so they have "superclock" which is 4x the rate, but only really works at 192kHz, and also passes cleanly through some systems and totally breaks things in others. If I switch from a 48kHz movie project to a 44.1 music project I have to switch two physical switches, change a menu item on another piece of hardware, then go into the computer, close pro tools, change the slave speed in the audio hardware driver, open pro tools, and open an existing project at the proper sample rate, close it, then start a new project at the sample rate I want. Failing to do this in the proper order can result in crashes and 0dBFS blasts of pink noise through five very expensive Genelec speakers. So yeah. Let's just put that innocuous little switch right up on the front where bored guitar techs start looking for things to go wrong. Right, Eddie?
'cuz if you're recording for video you're at 48kHz and if you're recording for music you're at 44.1kHz. And if you bought the gear to do this shit, you get to switch your shit a lot. And it's always this inconvenient. Sampling rate is related to Nyquist Frequency which is kind of like the "half-wave optics" of audio - if you want to reproduce a signal with content at 10kHz, you need to sample it at 20kHz or better in order to accurately reproduce the transients of the waveform. And back when everyone was figuring out digital audio the dorks were "how 'bout 50kHz" or "how 'bout 100kHz" (I had a Yamaha TX-16W whose native sample rate was 50kHz; a Fairlight CMI would go to 100kHz) but the guys who actually had to spec the memory were at "duh think in eights" so that 100kHz Fairlight CMI was 16-bit. So the labs like Philips and Sony were pretty much at "12-24-48kHz, that gives us 4kHz headroom over 20kHz, which everyone thinks is the limits of human hearing, even though it's actually like 16kHz in adults and 18-19kHz in kids but maybe there's some transients over that but good god if we double the sample rate we halve our play time and nobody wants to do Betamax again so... 48." And then Sony said "fuck yeah let's do Betamax again" so when Norio Ohga looked at Philips' recommendation that a CD be about the same size as a cassette tape he said "naah that won't work, it has to hold enough audio for all of Beethoven's Ninth Symphony" and Japan being Japan, and the late '70s being the late '70s, you sure as shit don't tell your CEO he's an idiot you irrevocably shittify digital audio for everyone forever. This is the same Sony who, when given a chance to change audio for the better by standardizing on 24-bit, 96kHz 8-channel audio like literally everyone else said "naaaah, 1Ghz, 1-bit, 'Super' audio CD!" thereby killing high resolution audio in its crib with a whack-ass format war that never even got to the point where Sam Goody carried the shit. And then they put rootkits on all their CDs.
this whole situation reminds me very strongly of how A=440 Hz is a totally arbitrary standard that we don't even follow half the time. It's total fucking madness.
I dunno, man, the minute someone invented the harpsichord it was gonna be One Temperament To Rule Them All for the simple reason that retuning a harpsichord/clavier/piano is a nightmarish ordeal. Frankly if it was possible to drag one pipe organ over to play next to another we would have settled on something in like 300BC.
Maybe, but we're still arguing about pitch now. Yeah, if you're playing with a keyboard, the you tune to the keyboard, and yeah 440 is the standard, but orchestras routinely use 442, Vienna is at 446, some people advocate for 432 (they're mostly nuts). Baroque musicians use a different arbitrary standard, 415, even though A has historically been as low as like, 392, and as high as 460something. That doesn't even go into intonation systems. We use one equal temperament, but there are others - 1/4 comma is popular in some baroque circles, and groups without a keyboard instrument often aren't even using equal temperament, they're using a form of just intonation! Shit's wild, fam. Frankly if it was possible to drag one pipe organ over to play next to another we would have settled on something in like 300BC.
Personally? I like Indian music with the movable frets so "all over the fuckin' place" is A-OK with me. But I also know that most people experience recorded, not performed music these days and most of that is sample libraries. And if I had to worry about my Vienna samples being at A=446 while my Synthogy concert grand is at A=440? Interoperability is this thing normies like and tweakers eschew just to feel special. Nobody gives a crap how many key signatures Dream Theater goes through except the people who won't shut up about Dream Theater. And we know better than to ask them.
Are you saying I shouldn't invite you to my 24 hour restream of Change of Seasons? Seriously, though, I get that these are not as wildly different as the Van Halen example. It just reminded me of a similar phenomenon. Alternatively, it's a bit like calling out a classic tune at an Irish folk jam and having 4 people start in 4 different keys because "that's the one I was taught it in". Nobody gives a crap how many key signatures Dream Theater goes through except the people who won't shut up about Dream Theater. And we know better than to ask them.
LOL there are things that are impressive that are not enjoyable. And I mean, humans both hear and see relatively, not absolutely so if Eddie and... whoever the bassist was at that show had ratcheting whammy bars on their axes no one would have noticed a thing. The only reason it's a problem is that Sony needed to fit 74 minutes of stereo program audio on 120mm of lexan to be read by a 780 nm laser and physics says you can't do that at 48kHz. - random googling occurs to see how far back DASH dumbness goes; author is reminded of "44.056kHz" which you never see but which exists; the author is again reminded of PCM adaptors which are well before his time but goddamn it a pet theory dies: - Okay kids ignore that Norio Ohga Beethoven's Ninth thing. We're stuck with 44.1 because backintheday if you wanted to record digital audio, you were doing it on a 3/4" U-Matic VCR through a PCM transcoder that turned sound into video and so that it would work with NTSC and PAL VCRs, it HAD to be 44,100 Hz because it's the least common multiple that works with 29.97 frames @ 60Hz, and also 25 frames @ 50Hz. And if you were PAL you were actually recording at 44,056 Hz because of dropped frames.Are you saying I shouldn't invite you to my 24 hour restream of Change of Seasons?
It is simplest if the same number of lines are used in each field, and, crucially, it was decided to adopt a sample rate that could be used on both NTSC (monochrome) and PAL equipment. Since NTSC has a field rate of 60 Hz, and PAL has a field rate of 50 Hz, their least common multiple is 300 Hz, and with 3 samples per line, this yields a sample rate that is a multiple of 900 Hz. For NTSC the sample rate is 5m × 60 × 3, where 5m is the number of active lines per field, which must be a multiple of 5 (the rest used for synchronization), and for PAL the sample rate is 6n × 50 × 3, where 6n is the number of active lines per field, which must be a multiple of 6. The sampling rates that satisfy these requirements – at least 40 kHz (so can encode 20 kHz sounds), no more than 46.875 kHz (so require no more than 3 samples per line in PAL), and a multiple of 900 Hz (so can be encoded in NTSC and PAL) are thus 40.5, 41.4, 42.3, 43.2, 44.1, 45, 45.9, and 46.8 kHz. The lower ones are eliminated due to low-pass filters requiring a transition band, while the higher ones are eliminated due to some lines being required for vertical blanking interval; 44.1 kHz was the higher usable rate, and was eventually chosen.