best way to back up alot of tunes?

[delcecchi]

Cool, I thought I was the only one listened to Sisters of Mercy.

[Kyhl]

I don't remember which order I placed the pressings of this corrosion but the brighter one is the mobile fidelity reissue.

I will have to watch for a merciful release pressing of the album since I don't have it.

[delcecchi]

Which album, floodland? I have the CD but no vinyl. About the only vinyl I saved of any interest was the first edition of the Velvet Underground with the banana on the front, from mid 60's some time.

[Kyhl]

The MoFi was from a reissue of Floodland LP.

The Merciful Release was a 12" single with Torch and Colours on the B side.

[delcecchi]

Interesting. I gave up vinyl years ago. Just too inconvenient and I personally can't hear the difference.

Isn't pretty much all music recorded and mastered and mixed and all that digitally these days? What is the sample rate and word size they use?

I am sort of curious is all. I could probably eventually look it up, but I really don't know where to start. And the audio field, being involved with human beings and their ears and brains, has a lot of folklore mixed in with reality and I can't tell the difference very well.

[Kyhl]

I think, don't know for sure because there is no standard, most is recorded at 24/96 these days.

Except Sony, they are using DSD instead of PCM. Sony is also archiving to DSD.

There are some bands recording to analog, White Stripes comes to mind.

But CD can't do more than 16/44 so LP'S are the only way to get the higher resolution in a physical format. You can download DSD from Sony or higher rates from HD Tracks but I like having a physical backup.

I've been looking into the reasoning behind the higher rates and it seems to be related to anti - alias filtering working better when pushed away from the audio bands.

[fishwater]

I've been looking into the reasoning behind the higher rates and it seems to be related to anti - alias filtering working better when pushed away from the audio bands.

Higher rates (and sample depths) help create a better stereo soundstage. The human hearing is limited to 40khz, but stereo human hearing can pick differences regarding time and space that is perceptible between ears down to 80khz.

[Kyhl]

I just lost a long explanation of this phenomenon. Thanks HSO.

I'll try retyping it later. I hate rework!

[Kyhl]

Let's try this again.

Higher rates (and sample depths) help create a better stereo soundstage. The human hearing is limited to 40khz, but stereo human hearing can pick differences regarding time and space that is perceptible between ears down to 80khz.

I notice the same thing, better defined imaging from higher rates.

The higher sampling rates help with imaging, in my experience.

I have seen references to an ABX study that showed the participants noticed a differences between music played being filtered at 20khz versus music played without the filter. This suggests that we hear tones up to 20khz but can senses frequencies higher than that.

The question in my mind is, were they hearing the filter or the reduction in frequency response? I don't know for certain and won't bother to look it up because my experience is that the higher frequency response in the recording allows for better playback.

The higher bit rates reduce quantization errors creating less work for the filters. Plus moving the filters farther away from the audio band with the higher sampling rate seems to help even more. It's a win win.

Oddly, I recently ran across this information while researching digital playback in regards to some people's claims that a certain $500 USB cable created better sound than a $20 USB cable. It didn't make sense to me before researching and still doesn't make sense after digging into it.

I did get a better understanding of the way audio is transmitted via USB. Is is not transmitted as buld data like computer data would be tranmitted. Therefore packets lost are lost forever. But packet loss doesn't really happen on a quality cable anyway so I'm still at a loss for why it should sound better.

Jitter was one explanation. But all USB DAC's use a buffer to read the data and reassemble it into audio. The buffer has to be reclocked so there should not be a jitter affect from the cable or the PC clock.

I did find one interesting nugget. Some people have found that insulating pins 1 and 4 of the USB cable will reduce jitter in the DAC. Pins 1 and 4 are the power lines for the USB cable. If the USB socket on the DAC is self powered it does not need the power lines from the USB cable. Disconnecting them removes the dirty power supplied by the PC which lowers timing errors (jitter) withing the DAC.

This is not possible on a Toslink or SPDIF coax. The industry seems to be going to data over USB because it has lower timing errors. Isolating the PC power seems to lower the timing errors even farther.

[delcecchi]

Depending on the filter style, a give cutoff filter can induce "ripple" in the response in the pass band (well below the cutoff frequency). That could account for some of the results. Also note that much of the "science" put out by cable companies and others is bovine excrement.

There are a number of different ways to make a filter based on the placement of the poles and zeros. Some also can affect the phase response as well as the amplitude. I do confess it has been a pretty good while since I studied complex variables, Bessel Functions, Chebychev polynomials and all that.

Also, much of the "science" and assertions in high end audio publications are also bovine excrement.

I put most of it in the same category as those who assert that speaker wires need "break in" or are directional.

[Kyhl]

I'm with you Del. That is why I was studying the implementation.

Early on people were saying how much better their picture was after spending a wad on an HDMI cable. Until they became educated on how HDMI works. A better cable makes no difference there. I expect it is the same snake oil for USB.

[delcecchi]

My working assumption is that anyone trying to sell you a "high grade" cable at a high price for any purpose in Audio using any kind of technical rationale is lying through their teeth. That includes both digital and analog.

I also have an issue with saying that vinyl has "more resolution" than a CD but that is another topic for another time.

[LMITOUT]

This thread has wandered away from the original subject but I saw this the other day and thought it would be worth sharing:

Quote:

Beyond MP3: New push for high-resolution music so clear you can hear a pin drop

The iPod marked a huge leap forward for the music industry -- and a step backward, too.

To make music conveniently portable, digital music players rely on compressed songs, which strip out parts of what otherwise would be very large music files.

It’s the only way to fit 10,000 files on that iPod, and in theory your ears don’t notice what’s missing.

But theory isn’t reality, experts say, and a growing number of musicians, composers and sound engineers aim to fix that by putting the fidelity back into your digital hi-fi, and getting the output -- the song you listen to -- to match the input that went into it at a mixing board. Your ears won’t believe the detail and clarity, they say.

“It’s like cleaning off a dirty windshield,” said Mike Mettler, former editor in chief of Sound and Vision magazine. “You can’t quite see what’s there, and then all of a sudden…”

Mettler spoke Tuesday as part of a panel at the 2014 Consumer Electronics Show that delved into high-resolution digital music, a term the panel of top names in the music business struggled to define. At its most basic, it refers to a file’s bit rate (the number of bits stuffed into each second of music) and its sampling rate (the number of slices of sound taken per second). The higher, the better.

But to these men and women, the term “high resolution” speaks to the poetry of music and engineering: That intake of breath before a trumpet solo, a mic stuffed in a drum and wrapped in blankets to perfectly capture each sharp thump, an echo bouncing around a room until it can reverberate no more.

“When you’re learning to taste wine, you don’t learn how to taste it. You learn how to describe it. It’s the same thing with music,” said C. Jared Sacks, head of Channel Classics Records, a classical music label. “At the end of the day, it’s emotion.”

Put simply, the higher the resolution -- the more bits of data stuffed in a second and the more samples of audio taken -- the more sharpness, the more detail, the more breaths, the more echoes. It makes the music more realistic.

CD-quality music is 16 bit / 44.1 kHz -- a standard that was defined in 1980, when Jimmy Carter was president and “The Police” had just started recording. But even that doesn’t cut it. A rate of 24/48 is generally regarded as the baseline for high res, experts say.

“Past 24/48 the door opens to the discotheque of the high-res world,” said Cookie Marenco, founder of Blue Coast Records. Tracks run up from there. A popular recording of the Pretenders’ 1979 debut album can be had at 24/192, for example.

At 24/48, the recordings should meet or exceed the capability of the human ear, which is generally believed to hear sounds at frequencies between 20 and 20,000 Hz, said Mark Waldrep, head of AIX Records and founder and president of iTrax.com. Music ought to fill that range and go further. But few digital files do.

“We have the technology for the first time to not only record that way, but deliver it that way -- if we so choose,” Waldrep said.

Few consumers are even aware of the esoteric details that these engineers and enthusiasts pore over. Indeed, most people are unaware that digital files can be had outside of iTunes and Amazon and Spotify. To get high-res files, you need to visit specialized websites such as HDTracks.com, iTrax.com, and Super HiRez.

To spread the word, the Consumer Electronics Association has jumped headfirst into high-def; on Tuesday it launched the HSOforum hiresaudiocentral.com, which will serve as a clearinghouse of information, expert content, gear and more for those looking to boost sound quality. Mettler currently runs a site called soundbard.com; in his spare time he will be chief content officer for the new CEA site.

Newbies to high-res audio will be struck immediately by the cost. High-res is expensive, sometimes three or four times the cost of an album from iTunes. Files are also much bigger: A 3.5 minute song can be 120MB or more, rather than 3 to 4 in a typical MP3.

Newbies will also confront unfamiliar terms: Files are distributed not as MP3s but as AIFFs or FLACs -- or DSD or PCM files, two uncompressed digital file formats. Engineers have been using both to mix songs for decades. The panel differed as to which format is better, but it doesn’t seem to matter in the end.

“The thing is, with DSD or PCM or hot dogs or bananas, the thing is how it’s implemented,” said David Chesky, founder and creative director of Chesky Records and a legend in the field.

You must choose, then. But choose wisely.

High resolution digital music is also tied into surround sound, or at least the consumer electronics industry hoped it would be. A decade and a half ago, Philips and Sony came out Super Audio CD (SACD), which was far higher resolution audio that hinged on having lots of speakers. At the same time Panasonic and a few other companies came out with DVD-Audio, a very similar type of disc.

Both were utter failures on the commercial market.

Among other factors, many consumers found the sheer number of speakers to be daunting.

“The challenge was convincing my wife to have five speakers in the room,” Sacks said, laughing. “And that’s not going to change.”

But, as with high-res music itself, the technology is addictive. “Once you’ve heard good multichannel, it’s hard to go back.”

Still, the new HSOforum alone is unlikely to convince people to buy, acknowledged John Newton, founder and president of Soundmirror, a recording studio with over 80 Grammy nominations.

“We’re fooling ourselves if we think that the person who has an iPod plugged into a desktop device while they’re cooking or cleaning is ever going to experience or care about high-res sound. They want that in the background. And that’s a different market than the people that listen more seriously to music.”

The weight of a consumer electronics giant might make a difference. Sony recently threw its weight (again) behind high resolution audio, with a new line of consumer electronics such as the HAP-Z1ES, a music player that can handle DSD files and won an innovation award at this year’s CES show.

“With a giant multinational company getting behind this and exposing it to people, we’re going to see if people take to it,” Chesky said. “The masses are now going to have an opportunity to be exposed to high res.”

And like a longtime McDonald’s eater who finally tastes foie gras, ears may be permanently changed, Chesky said.

“We’re in the fine food and fine wine business.”

[Kyhl]

I think they are over selling it a bit, and I'm for hi-res digital.

You can hear all of the things they mention with mp3 and CD quality.

In my experience, what you gain in hi-res is better spacial queues. If your system is set up properly you should get an image that floats in air between your speakers. This happens even with mp3. What hi-res brings is better definition of that space in the air. An example, in my old room I could close my eyes and tell when the singer was behind the mic or moved their head to one side or the other (while still singing at the mic). That definition is lost as you throw out samples.

Unfortunately, my current room loses that definition and it has been a struggle to regain some of it.

I have blind tested a 320k mp3 versus .wav file and the difference was minimal. Surprisingly, the .wav produced a flatter image in my system.

I've also done the mp3 invert test and played it back for kicks. This is where you convert a .wav to mp3. Invert the mp3 and convert it back to .wav. Then sum the two .wavs.

The result of that sum shows the data that is lost when an mp3 is created. At 320k, it isn't much when using a recent version of LAME to do the mp3 encoding. I'm told that older versions of LAME were worse. LAME 2.98 or better should be fine. I used LAME 2.99 for my test.

All of this talk of hi-res though goes back to my original post stating that it is important to keep a full definition version (lossless) of the CD ripped digitally. Lossy versions have their place but once made lossy they can never become lossless again. You can't replace bits later that were thrown out in the original rip. So it is best to keep at least two versions. A lossless version, and a lossy version. It is better to keep multiple lossless versions at multiple locations.

[delcecchi]

Unless the comparisons were done very very carefully and double blind, I can only think of the line from "the boxer"

Quote:

Still the man hears what he wants to hear

And disregards the rest

All those guys you quote have a monetary interest in convincing you that some fancy new thing is way better than what you have already.

[Kyhl]

I can tell you that my evaluation of mp3 vs .wav was done on a blind test.

It wasn't a true ABX. It was AB. So you can still poke holes.

Someone else created a file with a sample the same song from the same 16/44 digital source. One sample was the .wav at 16/44. The other sample was the .wav converted to mp3 then converted back to a .wav at 16/44.

They were not the same files. After the test I inverted and summed them and they were different.

I played both .wav's and one had a defined flat image extending between the speakers but was flat like a piece of paper. The other had a smaller image floating between the speakers. The smaller one did not extend to the speakers and was undifined. It was just there, a blob in the middle.

Sonically, I expected the mp3 to sound bad. I expected the reverb trails to trucate on bass and high frequencies. They didn't.

I expected the sound of the stick stricking just before the cymbol crash to be missing. It wasn't.

I expected the bass drum striker to be lost. It wasn't.

The mp3 sounded identical to the cd, except for the image they portrayed.

Science can tell me that I can't hear. I'll believe my ears.

The sales guy can tell me what I am missing. I'll believe my ears.

It's nice to see science catching up.

humans can discriminate a sound's frequency and timing more than 10 times better than the limit imposed by the Fourier uncertainty principle.

Del, there is also the theory that if you observe something but it does not show up in the measurements, then you measured incorrectly.

[delcecchi]

I agree with you about the observations being the ultimate test. But the tests have to be done very carefully. Audio seems to be particularly susceptible to minor things messing up results since we are trying to test human perception. It doesn't help that lots of people are trying to sell us something.

And ultimately there should be an explanation of where the difference is coming from, why it sounds different. Intramodulation distortion was an example from long ago. The use of brick wall filters when mastering CDs in the old days, with the filters producing pass band artifacts, is another.

There is so much baloney in the field of high end audio that it is hard to sift the wheat from the chaff.

The Fourier uncertainty principle? Not familiar with that. I know about Fourier analysis, Nyquist, Shannon, etc.