Nyquist doesn't stop aliasing, it prevents any tone higher than half the sample rate from recording at all, what comes out is a very audible screech...
You got it! Exactly! That "audible screech" is the alias frequency created by having a too low sampling rate. (and too high a freq setting on your high-cut filter.)
...The benefit of digital for archives is that every copy can be perfect, and all you have to do is back up often enough that you are within the window of the formats still being readable. The other advantage is that storage gets cheaper and cheaper with time-- as the archive gets bigger, it doesn't cost you any more to back up the whole thing (probably less).
Agreed! And, getting back to the original article, it's not that hard to pull up your lossless audio file in old format 'A' and click [SAVE AS] lossless audio file in new format 'B' -- on your super cheap storage of course!
As noted tho, doubt most of the end users care a jot about the hardware or OS, end of the day, it (and they) have but one job to do.
Totally agree on this.^^^
re. Smoke: It is a real "Swiss Army Knife" of post, it has a full NLE (a real nice one, designed by editors, not coders,) compositing (the interface is more like the Flame's now; it used to be layers, like AE) titling, color correct, audio mix (that can even drive flying faders) and all your media is 'everywhere,' no import/export. Sadly, I think they've let Adobe get ahead of them on features and plug-ins.
I mis-spoke on the 'porting,' Your right, IRIX was native original.
Sad about 'Combustion' too. True 3D space with ray-trace reflections! Awesome particle system! Morphs! (and not those crappy 'luma' based morphs, either!) But it's not dead, I still use it.
If you are looking for a live music experience, a "musicians sitting in the room" experience, uncompressed digital is the closest we currently have.
Funny you should say that - there have been times when I've been listening to a mix where people were speaking and it seems just like that - kinda spooky, but also cool.
If you want 'spooky,' record digital, 2 ch binaural (dummy head, if possible.) Listen to it back on cans with your eyes closed. Craaaaazy!!
Yes, the 'shape' of the waveform is based on the sum of all the sinusoidal harmonics. Nyquist set the sample rate just above 2x the highest frequency to avoid aliasing. Please read up on it...
No, that's not how it works. You still get aliasing – unless you lowpass filter the signal before sampling it.
So that is how it works. Sorry I left out the LP circuit diagrams and chip pinouts.
Sorry art folks, you shouldn't ask for a new OS for every different use case. What you want can and should be achieved with UI tweaks.
I agree. I actually RTFA and it feels like he's asking (and answering) the wrong question.
Like if you were to ask: "What is the best kind of paper to print targets on to improve your firearm accuracy?" instead of asking "What is the best ammo/gun/sight to use to improve your firearm accuracy?"
I don't think the OS is the thing to streamline, the actual creative software UI is more important. I work in both Adobe Creative Suite, and Autodesk 'Smoke,' and they both have a fairly good interface for previewing and moving between applications and/or file/media types.
Smoke runs as a sort of single application, and you just click buttons or swipe between pages of apps. However you can easily move from paint to video editing to sound mixing, without leaving the main program. The UI is consistent from module to module. Smoke has (in the past) been ported for IRIX, LINUX and Mac OS.
In the Adobe CS you still have to launch different apps, but a good deal of file manipulation "common ground" is had in their 'Bridge' app. Moving media between apps is also streamlined, you don't need to render and export/import to move a project or media between apps.
That score of "5" needs to be stripped away because you really don't have a clue as to what you're talking about when it comes to analog. Everything you cited against magnetic tape can be debunked with a simple internet search. And yes, I've been using analog, open reel tape... since the 1970s.
Whoops! It appears that you've forgotten to log in before posting your comment. Log in and repost so we can continue this conversation.
But a 30 IPS properly aligned Studer multitrack will have frequency response up through 80Khz and dynamic range greater than what can realistically be achieved on most DAWs
$250,000 is a LOT of money to spend in the audio production world. Personally I would much rather spend that on a huge mic locker, top of the line preamps, acoustic treatment etc.
@ daftna: Dude, you are awesome. You made every point I wanted to make - and then some.
The 'money' factor aside - when I think of all the hours I've wasted with a non-magnetic tweaker in my hand, adjusting bias on a 24tr deck, track by track... hours that could have been used creatively...
...and I don't recall a single producer, or artist, or sideman ever saying, "Wow, I really like the tape hiss on this track!" or "Uh oh! The singer failed to clip (overload) the high note here. We'll have to take it again."
I want to hear what the people sitting in the studio heard. Digital get me closer to that than analog tape.
Amen brother! I love producing music with computers. I spend hours looking for sweet spots inside the resonant wavelength of the instrument to find the right sound.
Agreed! It's understandable that people who grow up with a certain technology become attached to the qualities (and even artifact) that it engenders. I understand why people shoot film, or record on tape, if they like the texture and artifact it imparts to the work.
However, it's not intrinsically 'better' than digital. And it's certainly worse if you don't want all the artifact and limitations tape imposes.
If you are looking for a live music experience, a "musicians sitting in the room" experience, uncompressed digital is the closest we currently have.
Unfortunately, analog (magnetic) tape starts shedding oxide after about 15 years
Successful efforts to improve that life were made in the 1960s so it's more than double that now. That doesn't help with old tapes though which is likely to be very delicate. My workplace still has plenty of reels of tape from the 80s and 90s and every now and again it turns out the current owners of the data do not have another copy and quite a few of those old tapes have been transcribed without incident. The data is digital but the medium is identical to the reels of audio tape from that time.
Anyway I think the current fuss is over lossy data formats and digital data in poorly documented formats that keep chaging over time - it's a no-brainer that analog is going to be better than that if you can keep the medium it is recorded on intact. An answer to both is some form of future proofing the digital data format so people can easily deal with files that come with zero documentation. Some other industries worked that out in the 1970s, professional audio may be a different story.
Agreed - and agreed.
Re. shedding, a lot depends on storage conditions (i.e. moisture and heat.) The problem is that it's progressive (otherwise our heads would never need cleaning, even on new stock) which means constantly losing the S/N ratio. But it is much better on the newer stock.
Of course, in the pro world you would always want a non-lossy format, and, much like people who 'pickle' old tape machines for playback, you have to: 1. Copy your work forward into new, non-lossy formats, as they develop. 2. Keep archived software that will play your older format files.
I understand Nyquist, do you understand harmonics and aliasing? The shape of the waveform matters.
Yes, the 'shape' of the waveform is based on the sum of all the sinusoidal harmonics. Nyquist set the sample rate just above 2x the highest frequency to avoid aliasing. Please read up on it...
Sample rate: a higher sample rate allows for higher frequency representation. As in, if you have a sample rate of 48,000Hz, you can play back a frequency of 24,000Hz (already above the range of human perception). Higher sample rate = more high frequencies you can't hear.
First, I know *nothing* about audio.
So a question: Sure, there are frequencies humans can not hear. However, in to context of "playback", do those frequecies effect how we hear what frequencies we do hear, eaither due to the audio equiptment (speakers) or interaction with other frequencies?
Just askin...
First: the sample frequency needs to be slightly higher than the frequency you wish to reproduce. (Otherwise you will wind up with alias frequencies.) So 44.1k would easily get you up to 20k. 48k would very handily get you up to 22k.
Second: Humans; perfect, young, humans, can hear up to about 20k Hz. Humans can't hear anything above 20k (although dogs can!) so there is no need to reproduce it. Women and girls can hear better, longer than most men. As you age, you naturally lose high frequency hearing. If you are exposed to *loud noise you lose high frequency hearing. (That "ringing in your ears" after that loud party/concert. Ouch.) The higher range is not used for the 'notes' of the song, it is just the high harmonics of the instrument that you hear up there. Like the 'sizzle' of a cymbal crash or triangle. The core 'note' (or 'fundamental') of the triangle is much lower.
To most of you, music reproduced up to about 15k would sound pretty good. The extra range up to 20k would just give more 'air' and sizzle to music to a younger and/or female listener.
*Sadly, Sir George Martin is almost totally deaf now.
Digital has no problem with low frequencies. High frequencies are what it has trouble with. Digital is better than analog at low frequencies. Your booming pipe organ will have no problem with digital, but the violins and flutes may.
48k samples would not have any trouble reproducing anything a violin or flute could produce.
Analog tape playback is still available, after almost a century...
Unfortunately, analog (magnetic) tape starts shedding oxide after about 15 years. Magnetization starts to print through and creates pre and post 'echos.' As the magnetic signal weakens, the signal-to-noise ratio degrades. To be archived, you have to bake the tape (in an oven) and then you get one playback on your analog deck, so it can be digitized for archive.
If instead you record it onto another magnetic tape, you've just added more tape hiss and distortion that wasn't there in the original.
If you want true 'analog,' go to a live concert with no amplification. If you want fidelity: record, mix and deliver digitally.
I don't want to sit at home and listen to analog tape hiss, or wow, or flutter, or dynamic or frequency limitations. (or for so-called vinyl "purists": vinyl record noise.) I don't want to hear all the limitations and artifact of the recording media. I want to hear what the people sitting in the studio heard. Digital get me closer to that than analog tape.
The ICE, duh! Didn't your RTFA? Oh, wait, this is Slashdot.
The gas-liquid-solid gets thicker and thicker as you go down. The pressure gets so great that the hydrogen molecules get their electrons squeezed out of them, which means that the hydrogen becomes a sea of freezing liquid metal.
"The "mean" is the "average" you're used to, where you add up all the numbers and then divide by the number of numbers." Used without qualification, the word average tends to imply this usage.
"The "median" is the "middle" value in the list of numbers."
...that average I.Q. is 100......What you need to remember is this : half the population has an I.Q. of 100 or lower.
I'm guessing your company isn't selling Wolfram Research 'Mathematica.'
You're confusing 'average' and 'median.' The 'median' is the number that half the numbers in the set are above, and half the numbers in the set are below.
From Wiki: "The current scoring method for all IQ tests is the "deviation IQ." In this method, an IQ score of 100 means that the test-taker's performance on the test is at the median level of performance..."
Slashdot Mirror
Nyquist doesn't stop aliasing, it prevents any tone higher than half the sample rate from recording at all, what comes out is a very audible screech...
You got it! Exactly!
That "audible screech" is the alias frequency created by having a too low sampling rate. (and too high a freq setting on your high-cut filter.)
...The benefit of digital for archives is that every copy can be perfect, and all you have to do is back up often enough that you are within the window of the formats still being readable. The other advantage is that storage gets cheaper and cheaper with time-- as the archive gets bigger, it doesn't cost you any more to back up the whole thing (probably less).
Agreed! And, getting back to the original article, it's not that hard to pull up your lossless audio file in old format 'A' and click [SAVE AS] lossless audio file in new format 'B' -- on your super cheap storage of course!
As noted tho, doubt most of the end users care a jot about the hardware or OS, end of the day, it (and they) have but one job to do.
Totally agree on this.^^^
re. Smoke: It is a real "Swiss Army Knife" of post, it has a full NLE (a real nice one, designed by editors, not coders,) compositing (the interface is more like the Flame's now; it used to be layers, like AE) titling, color correct, audio mix (that can even drive flying faders) and all your media is 'everywhere,' no import/export. Sadly, I think they've let Adobe get ahead of them on features and plug-ins.
I mis-spoke on the 'porting,' Your right, IRIX was native original.
Sad about 'Combustion' too. True 3D space with ray-trace reflections! Awesome particle system! Morphs! (and not those crappy 'luma' based morphs, either!) But it's not dead, I still use it.
Whoops! It appears that you've forgotten to log in before posting your comment.
Log in and repost so we can continue this conversation.
Thanks!
If you are looking for a live music experience, a "musicians sitting in the room" experience, uncompressed digital is the closest we currently have.
Funny you should say that - there have been times when I've been listening to a mix where people were speaking and it seems just like that - kinda spooky, but also cool.
If you want 'spooky,' record digital, 2 ch binaural (dummy head, if possible.) Listen to it back on cans with your eyes closed. Craaaaazy!!
Yes, the 'shape' of the waveform is based on the sum of all the sinusoidal harmonics. Nyquist set the sample rate just above 2x the highest frequency to avoid aliasing. Please read up on it...
No, that's not how it works. You still get aliasing – unless you lowpass filter the signal before sampling it.
So that is how it works. Sorry I left out the LP circuit diagrams and chip pinouts.
Happy to see you read up on it, though. :)
Sorry art folks, you shouldn't ask for a new OS for every different use case. What you want can and should be achieved with UI tweaks.
I agree. I actually RTFA and it feels like he's asking (and answering) the wrong question.
Like if you were to ask: "What is the best kind of paper to print targets on to improve your firearm accuracy?" instead of asking "What is the best ammo/gun/sight to use to improve your firearm accuracy?"
I don't think the OS is the thing to streamline, the actual creative software UI is more important.
I work in both Adobe Creative Suite, and Autodesk 'Smoke,' and they both have a fairly good interface for previewing and moving between applications and/or file/media types.
Smoke runs as a sort of single application, and you just click buttons or swipe between pages of apps. However you can easily move from paint to video editing to sound mixing, without leaving the main program. The UI is consistent from module to module. Smoke has (in the past) been ported for IRIX, LINUX and Mac OS.
In the Adobe CS you still have to launch different apps, but a good deal of file manipulation "common ground" is had in their 'Bridge' app. Moving media between apps is also streamlined, you don't need to render and export/import to move a project or media between apps.
That score of "5" needs to be stripped away because you really don't have a clue as to what you're talking about when it comes to analog. Everything you cited against magnetic tape can be debunked with a simple internet search. And yes, I've been using analog, open reel tape... since the 1970s.
Whoops! It appears that you've forgotten to log in before posting your comment.
Log in and repost so we can continue this conversation.
Thanks!
But a 30 IPS properly aligned Studer multitrack will have frequency response up through 80Khz and dynamic range greater than what can realistically be achieved on most DAWs
$250,000 is a LOT of money to spend in the audio production world. Personally I would much rather spend that on a huge mic locker, top of the line preamps, acoustic treatment etc.
@ daftna: Dude, you are awesome. You made every point I wanted to make - and then some.
The 'money' factor aside - when I think of all the hours I've wasted with a non-magnetic tweaker in my hand, adjusting bias on a 24tr deck, track by track... hours that could have been used creatively...
I want to hear what the people sitting in the studio heard. Digital get me closer to that than analog tape.
Amen brother! I love producing music with computers. I spend hours looking for sweet spots inside the resonant wavelength of the instrument to find the right sound.
Agreed! It's understandable that people who grow up with a certain technology become attached to the qualities (and even artifact) that it engenders. I understand why people shoot film, or record on tape, if they like the texture and artifact it imparts to the work.
However, it's not intrinsically 'better' than digital. And it's certainly worse if you don't want all the artifact and limitations tape imposes.
If you are looking for a live music experience, a "musicians sitting in the room" experience, uncompressed digital is the closest we currently have.
Successful efforts to improve that life were made in the 1960s so it's more than double that now. That doesn't help with old tapes though which is likely to be very delicate.
My workplace still has plenty of reels of tape from the 80s and 90s and every now and again it turns out the current owners of the data do not have another copy and quite a few of those old tapes have been transcribed without incident. The data is digital but the medium is identical to the reels of audio tape from that time.
Anyway I think the current fuss is over lossy data formats and digital data in poorly documented formats that keep chaging over time - it's a no-brainer that analog is going to be better than that if you can keep the medium it is recorded on intact. An answer to both is some form of future proofing the digital data format so people can easily deal with files that come with zero documentation. Some other industries worked that out in the 1970s, professional audio may be a different story.
Agreed - and agreed.
Re. shedding, a lot depends on storage conditions (i.e. moisture and heat.) The problem is that it's progressive (otherwise our heads would never need cleaning, even on new stock) which means constantly losing the S/N ratio. But it is much better on the newer stock.
Of course, in the pro world you would always want a non-lossy format, and, much like people who 'pickle' old tape machines for playback, you have to:
1. Copy your work forward into new, non-lossy formats, as they develop.
2. Keep archived software that will play your older format files.
I understand Nyquist, do you understand harmonics and aliasing? The shape of the waveform matters.
Yes, the 'shape' of the waveform is based on the sum of all the sinusoidal harmonics. Nyquist set the sample rate just above 2x the highest frequency to avoid aliasing. Please read up on it...
Sample rate: a higher sample rate allows for higher frequency representation. As in, if you have a sample rate of 48,000Hz, you can play back a frequency of 24,000Hz (already above the range of human perception). Higher sample rate = more high frequencies you can't hear.
First, I know *nothing* about audio.
So a question: Sure, there are frequencies humans can not hear. However, in to context of "playback", do those frequecies effect how we hear what frequencies we do hear, eaither due to the audio equiptment (speakers) or interaction with other frequencies?
Just askin...
First: the sample frequency needs to be slightly higher than the frequency you wish to reproduce. (Otherwise you will wind up with alias frequencies.) So 44.1k would easily get you up to 20k. 48k would very handily get you up to 22k.
Second: Humans; perfect, young, humans, can hear up to about 20k Hz. Humans can't hear anything above 20k (although dogs can!) so there is no need to reproduce it.
Women and girls can hear better, longer than most men.
As you age, you naturally lose high frequency hearing.
If you are exposed to *loud noise you lose high frequency hearing. (That "ringing in your ears" after that loud party/concert. Ouch.)
The higher range is not used for the 'notes' of the song, it is just the high harmonics of the instrument that you hear up there. Like the 'sizzle' of a cymbal crash or triangle. The core 'note' (or 'fundamental') of the triangle is much lower.
To most of you, music reproduced up to about 15k would sound pretty good. The extra range up to 20k would just give more 'air' and sizzle to music to a younger and/or female listener.
*Sadly, Sir George Martin is almost totally deaf now.
Digital has no problem with low frequencies. High frequencies are what it has trouble with. Digital is better than analog at low frequencies. Your booming pipe organ will have no problem with digital, but the violins and flutes may.
48k samples would not have any trouble reproducing anything a violin or flute could produce.
Please go look up "Nyquist."
wants to bomb a bunch of brown people, just like Hitler,
I believe it was Hitler who was poison gassing his own citizens. So.......
Analog tape playback is still available, after almost a century...
Unfortunately, analog (magnetic) tape starts shedding oxide after about 15 years.
Magnetization starts to print through and creates pre and post 'echos.'
As the magnetic signal weakens, the signal-to-noise ratio degrades.
To be archived, you have to bake the tape (in an oven) and then you get one playback on your analog deck, so it can be digitized for archive.
If instead you record it onto another magnetic tape, you've just added more tape hiss and distortion that wasn't there in the original.
If you want true 'analog,' go to a live concert with no amplification.
If you want fidelity: record, mix and deliver digitally.
I don't want to sit at home and listen to analog tape hiss, or wow, or flutter, or dynamic or frequency limitations. (or for so-called vinyl "purists": vinyl record noise.)
I don't want to hear all the limitations and artifact of the recording media.
I want to hear what the people sitting in the studio heard. Digital get me closer to that than analog tape.
A whale and a bat breed?... that is just sick beyond imagining.
Not according to rule 34.
The ICE, duh! Didn't your RTFA? Oh, wait, this is Slashdot.
The gas-liquid-solid gets thicker and thicker as you go down. The pressure gets so great that the hydrogen molecules get their electrons squeezed out of them, which means that the hydrogen becomes a sea of freezing liquid metal.
Many military jobs require TS special clearances
First it was Don't ask, Don't tell. Then they allowed the homos. Now you can't even join the army unless you're a tranny?
Yeah. Those gay Spartans were terrible soldiers.
Smaller is ALWAYS better. Period. I can think of so many uses for these cubes my mind is racing.
Agreed. The minute I saw the pic I was picturing a wearable with this... or maybe a gamer/VR or AR backpack with four inside.
Don't you mean short and to the left?
No. Ron Jeremy is 60 now...
Not only is the median a type of average, but for a normal distribution -- a Gaussian "bell curve" -- all three types of average -- median, mean, and mode -- are the same value.
From your own article above:
"The "mean" is the "average" you're used to, where you add up all the numbers and then divide by the number of numbers."
Used without qualification, the word average tends to imply this usage.
"The "median" is the "middle" value in the list of numbers."
And the US is in a position to be talking about "fundamental freedoms"?
Otto: "Shut up. We didn't lose Vietnam. It was a tie!"
Archie: "I'm tellin' ya baby, they kicked your little ass there. Boy, they whooped yer hide REAL good!"
If you're not paying for a product, you ARE the product. I'm quitting today.
"I'm not in the business. I am the business."
FTFY
...that average I.Q. is 100... ...What you need to remember is this : half the population has an I.Q. of 100 or lower.
I'm guessing your company isn't selling Wolfram Research 'Mathematica.'
You're confusing 'average' and 'median.'
The 'median' is the number that half the numbers in the set are above, and half the numbers in the set are below.
From Wiki:
"The current scoring method for all IQ tests is the "deviation IQ." In this method, an IQ score of 100 means that the test-taker's performance on the test is at the median level of performance..."