24/192 music downloads are useless??

[3db]

Here's an article argueing against the futility of high rez downloads

24/192 Music Downloads are Very Silly Indeed

 

[adwilk]

Interesting. That was a very educational and enlightening article.

I'm thinking it makes sense... I don't know. My admittedly "not completely" scientific way of thinking is convinced that increased resolution will always matter in the right circumstance. To use the video/visual argument- The farther away from the image that you get or the smaller it is, the less the resolution matters. Get really close to a billboard, or sit in the front row of a theater, resolution becomes pretty important. Listening to low vs hi rez audio at low levels doesn't seem to make a difference. Crank it up and it SEEMS to matter. I'm glad I read the article and I'm also really confused.

 

[Hocky]

Interesting article.

 

[Josuah]

Also, higher resolution data does have an impact on measurements, which may or may not be audible and perceived either positively or negatively. Typically THD+N increases at the higher data rates.

But 88.2kHz over 44.1kHz also means the reconstruction filter can be implemented at a different frequency point and its impact on phase will be outside the audible range.

 

[DS-21]

Inferior, no.

An inefficient allocation of bandwidth, yes. For two reasons.

First, there are no audible differences compared to 16/44.1 (or 48, if that's easier to do - considering everything else in computerland is based on 2^x, the 44.1kHz sampling frequency of Red Book audio seems odd to me). Meyer & Moran, "Audibility of a CD-Standard A/DA/A Loop Inserted into High-Resolution Audio Playback," 55 J. Audio Eng. Soc. 9 (2007), at 775.

Second, that extra BW could be used for something that does make a marked audio improvement (when used well): additional discrete channels.

If Apple made well-mastered 5.1-channel 16/48 tracks available through iTunes, I'd never shop anywhere else for new music.

 

[Josuah]

Also, higher resolution data does have an impact on measurements, which may or may not be audible and perceived either positively or negatively. Typically THD+N increases at the higher data rates.

I should clarify the difference in measurements is not because of the higher resolution data itself, but from the processing of the higher resolution data. In other words, if you are sending the exact same input data (from the desired musical waveform) the output signal will be different depending on if you are converting it to analog at 44.1kHz versus 88.2kHz.

 

[mtrycrafts]

Here's an article argueing against the futility of high rez downloads

24/192 Music Downloads are Very Silly Indeed

What does he mean you cannot see X-rays?
There was a Russian girl who challenged the James Randi $1mil prize for just that ability.





She lost.

 

[Irvrobinson]

I've been sayin' it...

 

[avnetguy]

Meh, the article is ok but to me it just seems put together to dismiss 24/192. While I don't believe there is anything bad about listening to 16 bit 44.1/48kHz digital, hey it sounds great to me, I don't see how properly encoded 24/192 would be any worse other than potentially wasting space. It also seems to me, though I really don't look at digital acquisition at that level anymore and haven't for many years, that the statement of the signal being captured perfectly and completely for everything under the nyquist frequency is just not correct.

Steve

 

[TLS Guy]

I think the article is largely correct. However, the CD 16/44.1 as I have said before only just gets by.

For classical music there are often problems. I have a number of discs where in the mastering they ran out of bits, and that is audible.

For low level signal you must use enough white noise, dither, to avoid low level signal distortion, which is severe if not done properly.

When I master I generally use a higher sampling rate and then convert down. I always check the peak of a file, which WaveLab allows you to do easily, to make sure there is just a little head room.

I think it actually would be better to have a slightly higher sampling rate. It does provide a safety net.

HiDef downloads are a pain, as they are data files, and I usually end up with a long remastering job. There is no cue file with those downloads. So as far as I'm concerned they are not practical.

SACD almost always sound better than the CD layer, but this may well be mastering issues.

However the best sounding audio, and by a significant margin is on some of my recent BD discs. I'm not sure why this is, but I have no CDs that have the visceral impact space and detail of some of those discs.

Whether is is higher rates, or wider use of modern digital mics etc. I know not. But the improvement is not subtle.

At a recent AES chapter board meeting, the suggestion was made for some education sessions devoted to analog recording.

One prof. from McNally Smith and another from the Minnesota Media institute vowed that their 32 track analog Studer tape decks were superior to their digital hard drive recordings. I find this hard to believe, but may be in the pop world all this digital processing comes with penalties. Bouncing tracks in analog has definite generational loss.

They say for anything really important it goes on the Studers. However use is restricted as they don't want to wear them out!

So I did a calibration of my Studer/Revox A 700 and critically listened to some of my best old analog master tapes.

I have to say the sound quality gave no quarter to CD, and may be was more pleasing. However I know that all Studer machines have a very slight boost in output at around 30 Hz. You can not adjust it out on any Studer machine I have worked on. I have long been of the opinion, that this was deliberate on Willi Studer's part. May be that little extra warmth is all it takes to keep analog recorders alive.

 

[Irvrobinson]

For classical music there are often problems. I have a number of discs where in the mastering they ran out of bits, and that is audible.

For low level signal you must use enough white noise, dither, to avoid low level signal distortion, which is severe if not done properly.

When I master I generally use a higher sampling rate and then convert down. I always check the peak of a file, which WaveLab allows you to do easily, to make sure there is just a little head room.

Recording sessions are, IMO, the real justification for long word lengths. It's better to have the headroom and down-convert later.

I also have a couple of classical CDs where they have the tell-tale distortion of "running out of bits", which is really word length. In my case those recordings were made directly to 16 bit format, not the more typical 18, 20 or 24 bit format more modern recordings were made with.

 

[Irvrobinson]

I think it actually would be better to have a slightly higher sampling rate. It does provide a safety net.

How does a higher sampling rate provide a safety net?

 

[TLS Guy]

How does a higher sampling rate provide a safety net?

The issue is that CD 16/44.1 at least in the classical arena is sailing very close to the wind. However it was the best achievable back then.

So to have a bit more leeway, bigger files I think would be a benefit.

CDs can be downloaded with cue file now. I think CD probably could move to download.

For hard copy, I think BD audio only and BD with video is the way to go.

I really am very impressed with the audio of BD with multichannel and having the possibility of a picture with it a huge bonus.

Basically I'm just tickled pink with most of my BD purchases.

The whole problem now with CD, it is just good enough, and there are enough mastering engineers who get it wrong, because of the fine tolerance, I think there is advantage in moving on.

By the way it is not a frequency response issue, may power amps to not reproduce anything over 20 kHz.

 

[cpp]

The whole problem now with CD, it is just good enough,

CD to my ears is a lot better than crappy iTunes computer generated music. Good quality music from a CD well it all depends on the equipment used to make the disk, the mic's, the recording equipment and the room. A crappy made CD is just that, crappy music. Same for computer music, a sorry sound card, cheap speakers or those cheap ipod headphones well, you get what you get.

"Most recent remasters of popular music only sound like higher-resolution crap in their 24/96 24/192 glory."

 

[avnetguy]

I would have to agree that the biggest limitation is not the CD format but the all to common now, sub-par mastering. While 16 bit PCM sampled at 44.1kHz may loose a little bit of detail in the highest of heard frequencies, I doubt many (if not all) could tell the difference.

Steve

 

[djreef]

To read this article, much of I do agree with, you'd think this guy believes we've gone as far as we can with this from a software perspective. Sorry, I ain't buyin it. We're still a long way off from 'you are there live', and it can't all be hardware related.

DJ

 

[TLS Guy]

To read this article, much of I do agree with, you'd think this guy believes we've gone as far as we can with this from a software perspective. Sorry, I ain't buyin it. We're still a long way off from 'you are there live', and it can't all be hardware related.

DJ

I don't think it is the software which is the biggest barrier to the original sound right now.

I have some multi channel BD discs that get you very close indeed to the I'm there experience. In fact I think what I hear here is almost certainly better than all but the best seats in the house.

The spectral balance of all instruments is correct, with space around all the instruments and there is a good sense of the space and size of the hall. The dynamics and spl. are the same is the concert hall experience. There is no localization to the speakers, in fact no real cues to the fact it is reproduced music.

Even the audience applause puts you right in a seat in the concert hall with the illusion of the room being much, much bigger than it is.

So you, you can now get very close indeed to the original sound.

So recordings can now get you very close indeed to the original sound. In fact as good as makes no difference.

The bigger problem is that at this time it is not easily achieved.

 

[djreef]

I don't think it is the software which is the biggest barrier to the original sound right now.

I have some multi channel BD discs that get you very close indeed to the I'm there experience. In fact I think what I hear here is almost certainly better than all but the best seats in the house.

The spectral balance of all instruments is correct, with space around all the instruments and there is a good sense of the space and size of the hall. The dynamics and spl. are the same is the concert hall experience. There is no localization to the speakers, in fact no real cues to the fact it is reproduced music.

Even the audience applause puts you right in a seat in the concert hall with the illusion of the room being much, much bigger than it is.

So you, you can now get very close indeed to the original sound.

So recordings can now get you very close indeed to the original sound. In fact as good as makes no difference.

The bigger problem is that at this time it is not easily achieved.

No, I agree completely. Look, I defer to you when it comes to hardware issues, as you know infinitely more than I about such things. My perspective comes from more of a philosophical, theoretical background in that we may be selling the software perspective short given our acute limitations where the hardware is concerned. Until we figure out what the limiting component/s is/are in the hardware chain how are we to know whether the bitrate is adequate for true 'you are there' realism - dynamics, and all? Seriously, and I've proposed this in other threads - what is the next big thing that takes us there?

BTW Your personal experiences with 5.1 leads me to believe that your thinking is this is the future of audio. I dunno, I'm kinda hating the sound of that given my current setup.

DJ

 

[TLS Guy]

No, I agree completely. Look, I defer to you when it comes to hardware issues, as you know infinitely more than I about such things. My perspective comes from more of a philosophical, theoretical background in that we may be selling the software perspective short given our acute limitations where the hardware is concerned. Until we figure out what the limiting component/s is/are in the hardware chain how are we to know whether the bitrate is adequate for true 'you are there' realism - dynamics, and all? Seriously, and I've proposed this in other threads - what is the next big thing that takes us there?

BTW Your personal experiences with 5.1 leads me to believe that your thinking is this is the future of audio. I dunno, I'm kinda hating the sound of that given my current setup.

DJ

A number of observations.

In surround systems finding or building a suitable room is by far the biggest barrier. Honestly to bring off what I have done requires a purpose built room. Odd sized rooms, rooms with openings, asymmetrical furniture layout and poor dimensions ratios are not going to get you there. In these cases you probably are better off with 2.1 or 3.1.

Main and center speakers too close together is another big and frequent problem. Of the pictures posted by forum members I would guess in excess of 90% have their mains too close together. 8 to 12 feet is the range to aim for, with 9 to 11 ft being the real sweet spot in my view. It is absolutely impossible to throw a believable sound stage with main speakers closer than 8 ft.

So architecture is the toughest hurdle as far as I'm concerned. I understand from data dedicated rooms are becoming less rather than more popular. However if you poll my grandchildren, this is their absolute favorite room! My wife really likes it also and my children and spouses.

The next issue is that I don't think the any old puny speaker is adequate for realistic multi channel audio. All speakers have to be very good and neutral and able to achieve high spl when required. After all you would have to expect the reflected sound to match the forward sound, just like off axis and axis response of a speaker, or the result will be a downgrade no an enhancement.

I do think BD is better than SACD, and I think it is all to do with the ability to set the delays correctly in PCM systems. Correct delay is absolutely essential and it is never the measured distance.

As far as further improvements, it has to do with getting rid of passive crossovers and the wider introduction of zero phase shift digital crossovers and well as the development of affordable wide band drivers.

 

[jinjuku]

A number of observations.

In surround systems finding or building a suitable room is by far the biggest barrier. Honestly to bring off what I have done requires a purpose built room. Odd sized rooms, rooms with openings, asymmetrical furniture layout and poor dimensions ratios are not going to get you there. In these cases you probably are better off with 2.1 or 3.1.


Main and center speakers too close together is another big and frequent problem. Of the pictures posted by forum members I would guess in excess of 90% have their mains too close together. 8 to 12 feet is the range to aim for, with 9 to 11 ft being the real sweet spot in my view. It is absolutely impossible to throw a believable sound stage with main speakers closer than 8 ft.

Agreed. Having room is the only reason I did the Statements. If I didn't have the proper spacing I would have never attempted. Plus rooms that force one or more speakers into tight corner boundaries is all to common.

As far as further improvements, it has to do with getting rid of passive crossovers and the wider introduction of zero phase shift digital crossovers and well as the development of affordable wide band drivers.

Let me know if you are up to helping me design my first custom speaker. Using the Infinity Kappa 100.9w's sealed in a WMTMW where the MTM is open baffle.

The entire driver set is:

Kappa 100.9w (LR2 at ~300hz)
Zaph ZA14W08 for the mids
RS28F for the tweets (figure LR3 or 4 at ~1800hz)

The X-over points are just estimations.

I plan on a sandwiched frame that the woofer bins will be placed in that I can tighten / un tighten to toe in on the X axis.

Be using my DCX 2496 for X-Over duties.