I love the way you pluralized that. I don't think it's right, as it isn't hyphenated or a Norman root, but I really enjoyed it.
I'm honestly impressed anytime I see someone add a pluralizing 's' without adding an apostrophe. It's almost unheard of!
It seems pointless to be discussing everything but the elephant in the room. What RAM is going into the gaming machines?
This is a blatant attempt to refocus gamers on something they could actually afford to buy. RAM? No more RAM upgrades available. Because AI. I just hope that AI doesn't suddenly need keyboards too. 'Cause then even us writers are gonna be in big trouble.
The value is in not needing to upscale and downscale again. Every change to the original audio has the potential to induce issues, so the why from my perspective is simple enough.
The change being introduced here is fixed. You cannot convert analogue to digital without a world of signal processing in the way. But that signal processing is very very good. We live in an age where resampling, sample rate conversion, delta sigma conversion, bitdepth conversion etc, etc, are so good that we can't even measure the resulting difference with some of the best gear we have, we instead have to model it in the purely digital domain. Then you get some algorithms which introduce harmonics at -160dB from the primary signal, and those harmonics are not only not audible, they are many orders of magnitude fainter than our best gear can reproduce.
Fun fact, ever sound you hear through a computer that isn't played using some hardware exclusive mode (e.g. WASAPI exclusive mode for Windows Core Audio) already goes through multiple cases of conversion. All volume controls in applications and in windows themselves convert bit depth before and after. Your source material is rarely the same sample rate as your audio path, resampling is also done everywhere.
But all this is even ignoring the true theoretical WTF moment. If audiophilia is about reproducing as accurately as possible what is produced by the studio, and the studio output is by its nature a creative process, who are you to argue they are doing it wrong?
Just a dork that has spent a lot of time in studios. Including his own.
The idea that tracked audio should need to jump through more digital conversion processes just hits me wrong, and always has. I've been doing the recording game for almost thirty years now, and been playing instruments for over forty. It's not like I'm some, "Oh, I used to use a tape recorded to pull songs off the radio" uninformed idiot when it comes to the subject.
In fact, despite your seeming expertise on everything, I'm willing to place a wager this is one field where I have slightly more experience than you do.
To suppose? What do you mean by suppose? Assume? Yes there is a reason not to assume. Theorize? No, there is no reason not to theorize, so how do you test that theory?
Well, to start with, you need to create a new universe from scratch. From there, it's just a matter of waiting around for a few billion years to see the results.
So we've got Docker containers housing sandboxes, that run javascript code that does API calls to web based LLMs(nobody is running local cause you can't buy the fucking hardware cuz AI). It's just an incomprehensible madhouse of spaghetti at this point.
NVIDIA makes no real explanation of how this increases security or how to do the "guardrails". As if the existing MESS of endless layers of
And NVIDIA want to bring this to the masses? I thought I was pretty techno-savvy. But, this shit is starting to look like the NFT bubble. Nobody know what it is or what it does, but "You gotta have it!".
But just think of all the money changing hands! I mean, sure, it all goes in a big circle, with little added and removed from the circle at any point, and doesn't really trickle out into the rest of society, but the tech companies get to look flush by continually passing cash around and around and around and around and around.... MY GAWD, MAN! THINK OF THE POOR TECH CEOS!
Yeah, but right now it is extremely lucrative, which for many people shapes the morality.
Living with the concept of profit and greed being your only moral compass tends to really fuck up a society. And we've been living with that here in the US, and exporting it as fast as we can, for well over forty years. Forty years is convenient as a cap, since it's around the time I became politically aware, and also around the time Reagan's power began to fully cement the reality of Greed as God for us here in the US. Others played the game before him, but he managed to turn it into a fully fleshed-out, easily digestible message for the masses.
Your ears couldn't when they were at peak. 96kHz sample rate can faithfully replicate 48kHz waveforms, which is over double the range of human hearing. Even 48kHz replicated 24kHz waveforms, which are well above human hearing. Studios are not capable of it, because -why would they be-? That's like making a 4800K 36-bpc movie camera. 192kHz is just fucking clownshoes. For the mastering process, there is value in working using upscaled PCM audio, so that you don't introduce additional aliasing where you have very little room to absorb it, but there is no value whatsoever in recording it, or playing it.
The value is in not needing to upscale and downscale again. Every change to the original audio has the potential to induce issues, so the why from my perspective is simple enough. Why upscale when you can record at what you need for mixing/mastering and only need to downscale for the final result?
Seriously, if a person is betting six figures on the whether an interceptor worked, that person needs to re-evaluate their priorities.
I have this sneaking suspicion some top members of our (USA) government are involved in placing at least some of those bets.
I keep betting that your mom will suck dick for cash. No one is willing to bet against me.
Sorry, man. Already spent all our fun money on getting your mom to suck our dicks.
Encyclopedia Britannica still exists. Who knew?
What? Big tech WANTS age verification. That should be pretty obvious by now. Who do you think is lobbying for it? If big tech didn't want it, it would be crushed. Age verification benefits Palentir, Meta, Microsoft, Google, Oracle and many others. It's NOT about age verification. It's the first step towards full digital identification and removal of all anonymity.
That's honestly taking a lot longer to be pushed through legal channels than I thought it would. There was a time a decade or so ago where Zuckerbot was saying he wanted Facebook to become the defacto digital ID for the United States and to have it used for everything from banking to social media ID tied to a real human being. The fantasy of big tech has been to tie all online activity to a real human for almost as long as big tech has existed. While I'm not surprised they're getting close to pulling it off now, I am surprised how long it's taken to get from fantasy drawing board to legal discussion.
You need that on the studio side because you're manipulating the sound through all those shitty plugins. But if you think you can hear the sound difference, I have a $175,000 ethernet cable to sell you.
Yeah, I doubt I do. My ears are shit since I've been in thrash and death metal bands since the 80s. But the argument that studios aren't capable of more than 48k? That sounded like somebody talking about a subject they were completely unfamiliar with.
Basically all music you listen to starts out as 48kHz. Even the "192kHz" tracks are just upscaled.
When an artist records an album they generally record in 48kHz - and basically anything you listen to will have a 48kHz master.
Now, some artists do record in 96kHz usually in very specialized recording studios set up for it (most aren't). Generally you're not going to be listening to a 96khz originated recording because almost no studios have equipment to do so. AIX Studios and Skywalker Ranch are two examples that can do 96kHz and have tuned pipelines to work at 96kHz. But AIX generally only does their own concert recordings of public domain works (e.g., classical pieces) and people who use Skywalker Ranch at 96kHz are very, very rare.
99.9% of common music you can buy will be recorded at 48kHz, and the whole "high res" thing is to be able to sell you upscaled music at a higher price even though there's no added content.
In an age when a lot of home studios record at 96k and up, I find a lot of this post to seem out of touch. Commodity recording equipment has been capable of 192k and up for about twenty years now. I don't see how its possible that professional studios wouldn't have this capability, even if they don't typically use it. I've been recording at home at 96k for about ten years or more. I'm sure you can argue it's pointless to do so, because a lot of people do, but disk space has been cheap enough and system performance is impacted so little by it I don't much care about the arguments against it.
What does Meta need all this AI compute capacity for? Running Moltbook at scale?
You need a *LOT* of datacenters to refactor all data available on the internetwebz every couple months when they release a new model.
yeah i agree!!! just throw it out and dont even bother!! nothing is ever as good! we should all just give up!!! fuck living!!!
Hello, Sam Altman. Thank you for your contribution to the conversation.
One person's error is another person's data.
