[lantian]

   
Judging by some of the comments you read, I think a lot of the audiophiles are already experimenting with this 

 

In my opinion that is how hifi kicked of, but hey that's just me

 

[watchnerd]

  We should include alcohol consumption in the next study

 
why limit it to alcohol?

 

[StanD]

You guys have had a reality forum show going on, I missed the entertainment. Must be the audiophile grade of LSD that's been going around, makes one hear things.
So what kind of precision does one need when multiplying 2 different 16 bit numbers for sound effects like ring modulation? Certainly 16 bit precision in not adequate, unless one is willing to compromise by scaling down the two inputs prior to multiplication.

 

[castleofargh]

please, no more drug stuff for most obvious reasons on a public forum where people are welcome at 13.
 
Quote:

Yeah, for me, music always sounded way better when I'm happy.

When I try to remove all the biases I can though, I can't hear any difference. The more I think I hear something, the more I hear it. Then I swear I hear the opposite once I confirm that it actually wasn't what I was playing.

D:

Key to enjoying music: be happy!

I guess most will agree about this,  but what if
"I'm only happy when it rains
I'm only happy when it's complicated"?

 

[U-3C]

  I guess most will agree about this,  but what if
"I'm only happy when it rains
I'm only happy when it's complicated"?

Hey, my best music listening experiences are from when it is raining and I listen/dance/swing my hands wildly like a madman to music on the balcony, with my Q701 plugged into my computer indoors~
 
I'm still perfectly fine and alive.
 
*On a side note*
 
Th...thanks AKG for your 20 feet cable...
 

 

[pinnahertz]

  You guys have had a reality forum show going on, I missed the entertainment. Must be the audiophile grade of LSD that's been going around, makes one hear things.
So what kind of precision does one need when multiplying 2 different 16 bit numbers for sound effects like ring modulation? Certainly 16 bit precision in not adequate, unless one is willing to compromise by scaling down the two inputs prior to multiplication.

Ring modulation would be a kind of analog synthesis effect, and as such, I doubt high precision is even desired.  Remember, early FM  synths were 12 bit, then 15 bit.  With today's interest in vintage instruments you probably want that 12 bit grit. 

 

[gzubeck]

so what makes some recordings sound insanely smooth vs. recordings that make you want leave the room? Whats the higher sampling rates for? Doesn't a higher sampling rate improve anything? Is it strictly poor equipment and increased loudness?

 

[pinnahertz]

  so what makes some recordings sound insanely smooth vs. recordings that make you want leave the room?

Since your observation relates to a difference in the recordings themselves, that would be where to look. There have been good, bad, and inbetween recordings since the beginning. Generally a bad sounding recording, these days, is evidence of lack of skill or misplaced artistic priorities.

Whats the higher sampling rates for?

There is a strong psychological coupling between the concepts of "high" and "better", regardless of the actual result. Proponents would say they are trying to capture all of the audio, regardless if its audible or not. Opponents would say there's no need to capture it if it's not audible, since no difference can be heard. A realist would recognize that simply running a higher sampling rate doesn't ensure the capture and reproduction of ultrasonic frequencies and their delivery to the hear. And it turns out, delivery to the ear is one of the harder things to do.

Doesn't a higher sampling rate improve anything?

It increases the maximum frequency that can be captured. That's not necessarily an improvement.

Is it strictly poor equipment and increased loudness?  

Back to those bad sounding recordings are we? It's actually kind of hard to buy really poor equipment these days. The loudness war has resulted in some really horrible sounding recordings. Mastering a recording to be loud is highly destructive, and is one of those misplaced artistic priorities. Ostensibly it's supposed to sell more music. It doesn't. It's much more of a ego thing.

 

[gregorio]

Originally Posted by pinnahertz /img/forum/go_quote.gif
 
[1] It's the "law of uncorrelated summing", you build level 3dB every time you double the channel count (assuming uncorrelated signals at the same level). It's why even analog mixers provide for lots of extra headroom at the mix bus, and a bus master.
  [2] In all the forum battles I've read and been involved with here and elsewhere I've never actually seen a conversion take place in either direction.

 
1. We can't assume uncorrelated signals though, particularly when mixing music and, they're not at the same level. Nevertheless, even though the "law of uncorrelated summing" is extremely imprecise in practice (the real world of mixing real acoustic sound signal), as we increase the channel count we would build level. However, none of this is directly relevant to the point I was trying to make because 0dBFS (max peak) has exactly the same reference level at both 16 and 24bit. 24bit has no additional headroom relative to 16bit, what it has is more "foot-room". So when recording, 24bit doesn't provide more headroom in terms of allowing us to record a louder signal, it provides more headroom in terms of allowing us to record the signal quieter (by reducing the mic-preamp), thereby increasing the gap between the peak signal level and the max system level (0dBFS). The reason in film we often submix down to a 24bit file rather than 16bit is because there's likely to be numerous further processing steps applied to that recorded file and that result will form part of another submix, which may in terrn form part of anther submix. This isn't the case with music because even though we often employ a (relatively, very simple) submix topology, those submixes don't need to be recorded (to a file) and therefore truncated (to 16 or 24bit), they never have to leave the original (48bit/64bit) virtual mix environment. I'm sure you're probably aware of all this but I thought I'd (attempt to) clarify for the wider benefit.
 
2. I can't remember seeing such a conversion here (on head-fi) but I've certainly seen it elsewhere. In fact it's very common on pro music/sound engineering forums, where newbies are usually infected by at least some audiophile myths (if not most of them). The difference with pro sound engineering forums is that there is a deeper knowledge, both at the technical and personal level. For example if a newbie quotes some famous producer/engineer/studio/musician extolling the virtues of some dubious product, there will be those present who know first hand why that famous producer/... made that quote in the first place. Although unrelated, I find it interesting that there appears to be a relationship between the pro audio world and the audiophile world, a relationship commonly separated by a decade or more. Some issue will arise in the pro audio community and become a bit of a widely discussed "thing"; the how, why and what of particular observations and solutions sought. Occasionally the issue might be based on a fallacy (which is ultimately dispelled) but more commonly it's a real issue related to the technology at that point in time. 10 or 20 years later that exact same issue with the exact same initial arguments/observations appear in the audiophile world. Although typically it's completely fallacious because either: 1. It was fallacious to start with in the pro audio world. 2. The technology has moved on and the issue no longer exists or 3. The issue was never applicable to consumers anyway. Anti-alias and reconstruction filters (types, ringing, audibility, etc.), quantising error, resolution, noise/dither, bit depth, sampling rate, audio compression, data compression, jitter (types, audibility, etc.) and jitter clocking/re-clocking/rejection, transmission of data down cables (signal types, loss/distortion), to name but a few. It's almost as if audiophile manufacturers are trawling pro audio forum archives looking for some 20 or so year old issue (which never existed or no longer exists), so they can create a product which solves that (non-existent) issue or use it to support an existing product.
 
Quote:

  [1] So what kind of precision does one need when multiplying 2 different 16 bit numbers for sound effects like ring modulation? Certainly 16 bit precision in not adequate, [2] unless one is willing to compromise by scaling down the two inputs prior to multiplication.

 
1. I'm not sure I understand what you're asking. In general, 16bit is enough when multiplying two different 16bit numbers (for whatever sound effect). The problem starts to arise when taking that result and multiplying it with something else and gets worse if you then take that result and multiply it with something else again, and then keep repeating that procedure.  Certainly, 48bit or 64bit float is more than enough for any conceivable, realistic number of such processes/procedures and btw, a 48bit fixed virtual mixing environment is not a new thing, it was introduced about 20 years ago and was pretty much ubiquitous by about 15 years ago.
 
2. In practice, the 48bit systems did provide considerable headroom, in fact the mix busses were actually 56bit accumulators, with the additional 8 bits used as an additional 48dB of headroom (above 0dBFS). And of course with floating point (32 or 64), headroom is effectively astronomical, as all you're doing in effect is changing the position of the decimal point. Either way though, you do ultimately have to scale it down, either on the input to the various processing (multiplication for example) or the output of all the processing (from the mix buss) because this output has to ultimately go to a file and/or DAC with an integer output which cannot exceed 0dBFS. For example, with a 32bit float mixer you can in theory go up to about +500dBFS (if memory serves) without distortion but you obviously can't actually output that signal, you'd have to reduce the output (fader) level so the peak is less than 0dBFS, so as not to clip the DAC.
 
G

 

[TheoS53]

  so what makes some recordings sound insanely smooth vs. recordings that make you want leave the room? Whats the higher sampling rates for? Doesn't a higher sampling rate improve anything? Is it strictly poor equipment and increased loudness?

I suspect that it is this assumption that has lead to the belief that a higher sampling rate is better. Someone might listen to 2 recordings of different sampling rates, and it just so happens that the higher sampling rate one sounds better, so they assume that it is because of the sampling rate that the song sounds better. But, we cannot assume that, even though it's the same song, that they actually came from the exact same recording  

 

[lantian]

Higher sampling rates are simply producing more life like resolution of your music.  Human ear is perfectly capable of hearing the difference between higher sample rates. The so called auditory resolution of human hearing is about 5-7us, a 44.1khz audio has each slice of 20us, 96khz has something like 10us, and only 192khz can offer time resolution of 5us. That would seem perfectly smooth to our auditory systems/brain. Go ahead and correct me If I am wrong on this.

 

[TheoS53]

  Higher sampling rates are simply producing more life like resolution of your music.  Human ear is perfectly capable of hearing the difference between higher sample rates. The so called auditory resolution of human hearing is about 5-7us, a 44.1khz audio has each slice of 20us, 96khz has something like 10us, and only 192khz can offer time resolution of 5us. That would seem perfectly smooth to our auditory systems/brain. Go ahead and correct me If I am wrong on this.

have a look 2 pages back of the images I posted.... in 24 bit resolution it's impossible for any human to hear the difference between 192khz, 96khz, and 48khz, as the files are IDENTICAL below 23khz

 

[lantian]

  have a look 2 pages back of the images I posted.... in 24 bit resolution it's impossible for any human to hear the difference between 192khz, 96khz, and 48khz, as the files are IDENTICAL below 23khz

I am not so sure if that is correct after all audio works on a logarithmic scale, so while they seem identical they are not, I do not posses enough knowledge about this, but  fact is the time resolution changes and human auditory system can pick that up.
Never mind should not have said anything

 

[gregorio]

  [1] so what makes some recordings sound insanely smooth vs. recordings that make you want leave the room? [2] Whats the higher sampling rates for? [3] Doesn't a higher sampling rate improve anything? [4] Is it strictly poor equipment and increased loudness?

 
Essentially as pinnahertz said but put slightly differently:
 
1. There's the initial problem of how you define "insanely smooth". For example, a fan of thrash metal would probably have quite a different definition of "insanely smooth" compared to say a fan of Debussy, if a thrash metal fan would even think "insanely smooth" desirable in the first place! Let's say, for argument sake, we all want "insanely smooth" and all define it the same, still there is no simple answer to your question as there can be many causes of a recording not being smooth; most typically a musician not performing smoothly, the instrument they're playing or the acoustics not responding smoothly but just as possibly; not optimally chosen mics, mic positioning, mic-preamp and pre-amp setting or downstream, some processing which reduces smoothness and further downstream still, some consumer equipment or environment not responding as smoothly to the frequency content of that particular piece of music/recording as to another.
2. Essentially as far as the consumer is concerned, they're for marketing purposes.
3. For the consumer, no. In fact it's likely to actually make it worse rather than improve anything, but we're only talking about slightly worse.
4. No, as mentioned in #1 there are all sorts of potential causes and not all of them about poor equipment. I'm going to disagree with pinnahertz somewhat here; mastering a recording to be loud is not necessarily "highly destructive" it can just as easily be highly beneficial and likewise, not mastering a recording to be loud can be in effect even more "highly destructive", it all depends on our listening equipment, environment and what we're actually listening for at any particular instant in time.
 
G

 

[castleofargh]

  Higher sampling rates are simply producing more life like resolution of your music.  Human ear is perfectly capable of hearing the difference between higher sample rates. The so called auditory resolution of human hearing is about 5-7us, a 44.1khz audio has each slice of 20us, 96khz has something like 10us, and only 192khz can offer time resolution of 5us. That would seem perfectly smooth to our auditory systems/brain. Go ahead and correct me If I am wrong on this.

I'll let those who didn't stop using math the day school ended

, explain in detail if they feel like it .but basically it's the same mistake as the great classic "humans have 140db of dynamic range so of course 16bit isn't enough!". the values come from something real but specific, and are then abused out of context to try and make a point.
get me loud iterated ripple noises and square waves as the only sounds in your favorite song, and then we can discuss some 5 to 10µs delays audible to humans in "music".
 
for the "sample rate to period" thing, it doesn't mean what you seem to think it means. the actual signals in music are sine waves, the actual resolution will depend on the bit depth and loudness of the signal, and most certainly does better than 20µs. to degrade to any not even significant values, we'd need to go look at very low levels, which of course negates the 5µs of the paper anyway as nobody would ever pass the test at very low volume levels.
so it's pretty safe to say that you are wrong on this. the results of a specific test are valid for the specific conditions of the test.