Testing for Bit Depth and Sample Rate...aka Resolution

AsSiMiLaTeD

Do you guys know of a way to verify if a audio resolution in a file is what they say it is? I've got an album that I bought in CD format (nothing fancy just regular 16/44) that sounds more like an upsampled MP3 file to me. I'm looking for a way to test the file and see if it's native CD resolution or upsampled.

Either way it doesn't sound good to me, but I'd like to know who I should be irritated with the studio or the company selling me the music.

FYI I'd rather not say anything further about what the album is or where I bought it until I see if I can determine what the issue is.

Thanks

msg

Barring a piece of analysis software, I wonder whether you could try to obtain a media rip from someone in the same format and resolution you purchased, and then open and compare waveforms, zoomed in in something like Audacity or CoolEditPro?

K_M

Burn the CD you have to a hard drive,(lossless of course) look at resulting file size.

It might help if we can know what album it is. But I understand your privacy.
Knowing the album may help us determine the sound quality issue also.

One normal length song might be around 40 Mb in CD quality file size.
About 10 Mb or less in 320 Kbps MP3...

rooftop59

Perhaps I am confused, but if it is "upsampled," wouldn't it be the same size? LIke when your TV upscales a DVD to 1080p, it adds lines that aren't there? Hence the problem of trying to determine the original size...

The waveform idea might work, but what do you mean by media rip? I guess you could download the mps from Amazon and compare that way and if there is really no difference in the dynamics then I would be suspicious...

heiney9

Also if an exact copy wasn't made/obtained the size is going to be different. Not by a lot, but different.

H9

recoveryone

I have experience the same issues with CD's and I conclude that a bad master( poorly recorded) is always at the source of the problem.

AsSiMiLaTeD

Yeah I'm looking for a more definitive way of testing than comparing waveforms against another copy of the file because there are too many variables involved.

If I broaden my question a little, how do we know that anything we pay for online in FLAC format is what they say it is? I'm not just talking hi-rez stuff, I've bought plenty of regular 16/44 content from Pono and HDTracks...how do we know we're getting what we paid for?

I'd normally be in the "your ears will let you know the difference" camp, but with the wildly variable sound quality from album to album there's no way that approach is even remotely close to scientific...as in the sound quality from album to album varies by a much wider margin than the difference between even decent MP3 compression and a 16/44 file.

I should be able to look at or analyze a file somehow and know it's true resolution. is that not a thing? Am I the only one who cares about this? No way I could be the only one who cares about this...

heiney9

Other than comparing it audibly to a known copy, I don't think there is a way. I mean there probably is, but one would have to have the right software and knowledge to do it, as well as a known "good" copy of said recording.

That's one of the reasons I've shied away from paying for on-line sourced rips of music, especially when they're asking a premium price for said rip/recording. Unless the seller is the actual producer or source of the music.

Most recordings today are crap and just lesser crap. Even a remaster isn't guaranteed to sound better than the original or however many reissues have been released.

H9

mhardy6647

rooftop59 wrote: »

Perhaps I am confused, but if it is "upsampled," wouldn't it be the same size?
...

yes, I think it would have to. My recollection of the CD audio/WAV scheme is about 10 MB per minute. An 80 minute audio CD = 800 MB (roughly)
OK I looked it up: more like 700 MB (I will gloss lightly over the issue that 1k = 1024 bytes, and that there are 8 bits in a byte; work with me here )

ca. 700 MB/80 min = (ca.) 8.75 MB/min.

The music/data file on an audio CD would be 'padded out' to 8.75 MB per minute with... whatever. Interpolated data. Ads for Netflix. Subcode (remember "Subcode" from the dawn of the audio CD?).
Whatever.

mhardy6647

Good ol' subcode. The future was so bright, we hadda wear shades.
From the OM of the Sony CDP-102. There are, somewhat inexplicably, two of those beefy behemoths sittin' down in the basement someplace. They work, too (although the drawers are balky on both of them).

msg

Wellt, first I misread your original post and responded thinking you'd purchased a download.

Second, I understand that visual inspections of waveforms probably wouldn't be definitive, especially if it was just a comparison of one cd to another. I was thinking download vs physical media at first, which you touch on later in the thread. Indeed, I've wondered about this myself wrt actual quality/resolution in streaming services like Tidal.

A web search for 'audio resolution analysis' turned this up
Haven't used it so can't say much more about it.
http://www.deephouseamsterdam.com/this-handy-tool-checks-the-real-quality-of-your-music-files/

'audio fingerprinting' may be another handy search term to try.

CoolJazz

Traders Little Helper I think is the name of the program that will help you. It's freeware and been around. Does the job of identifying pretty well the content that's been upsampled from MP3 to linear.

I've used it and it's only rarely been fooled for me. Not perfect but pretty good.

It is just a tool for the up to redbook stuff, unless their is a newer version that does more I haven't seen.

CJ

halen

Danny,

A couple years ago I was developing my own music interface, like JRiver or Kodi. The most important aspect of the development was analyzing the actual data. The meta data is already known, such as album and track name and file size. I guess you could say the objective is to write out the bits before and after it has been massaged by the DAC. Take the before and after and perform a bit compare, like code compare. I would make use of a CD player installed on the PC for cd or just extrapolate the data if it was a music file. I don't have the algorithm used to up or down sample.

For a long time I attempted but could find no solutions or API's that gave the functionality. All I could come up with his compiled data, not the source. Languages used were C Sharp and Java. I know an API must exist but proprietary and not open source. Lastly, I am not skilled enough to just write an API to interact with the CD laser and its reading.

Good luck on your search!

AsSiMiLaTeD

msg wrote: »

A web search for 'audio resolution analysis' turned this up
Haven't used it so can't say much more about it.
http://www.deephouseamsterdam.com/this-handy-tool-checks-the-real-quality-of-your-music-files/

That one did the trick, thanks. I tested it out with some files of varying known resolutions and it identified them all correctly. I then took a known 320 mp3 file, upsampled it using dbpoweramp, and it also identified that one correctly with the original bitrate.

It looks like there's still a way to trick these programs but those methods look to leave fairly obvious fingerprints in the file that seem to be easy to spot if you're looking for them. There are some more advanced ways to fool spectrum analysis but those actually appear to introduce enough noise that anyone actually listening to the file would be able to tell.

So between that program and my ears I think I may be set, although now that I know that spectrum analysis is a thing I'll look into it some more.

I appreciate the help

AsSiMiLaTeD

Also...

I did some quick scans of my "high resolution library" and they confirm alot of what I've been complaining about for a while now. No wonder alot of these don't sound better in hi-rez...they're not IN hi-rez.

msg

Cool.
Now I'm curious to have a look at some of my own.

K_M

AsSiMiLaTeD wrote: »

Also...

I did some quick scans of my "high resolution library" and they confirm alot of what I've been complaining about for a while now. No wonder alot of these don't sound better in hi-rez...they're not IN hi-rez.

I have found the master tape and recording determine if it really sounds good or not so good, more than any other variable.

voltz

Danny, when I play a file from Tidal using my Auralic mini the apple app I use tells me its 16bit/44 or 24bit/48 when playing some master files.

Also when I play a file from the internal SSD I have inside my mini it tells me which one it is as I have a few songs that I have 3 copies of on my mini : like 16 bit and 24 bit and DSD file and it lets me no which is which even before I play the file.

AsSiMiLaTeD

voltz wrote: »

Danny, when I play a file from Tidal using my Auralic mini the apple app I use tells me its 16bit/44 or 24bit/48 when playing some master files.

Also when I play a file from the internal SSD I have inside my mini it tells me which one it is as I have a few songs that I have 3 copies of on my mini : like 16 bit and 24 bit and DSD file and it lets me no which is which even before I play the file.

That's just the issue though, just because a file SAYS it's 16/44 or 24/96 doesn't mean it actually is. The info readout you're getting on any player or software like that is going to read the bit rate and sampling frequency but doesn't actually tell you what's packed inside the file.

Think of it like FedEx dropping off a new DAC you've just ordered. The box may SAY it's an Auralic Mini but you actually have to open and inspect the box to make sure someone didn't throw a piece of coal in there and mail it off to you.

voltz

okay that makes sense.

AsSiMiLaTeD

K_M wrote: »

AsSiMiLaTeD wrote: »

Also...

I did some quick scans of my "high resolution library" and they confirm alot of what I've been complaining about for a while now. No wonder alot of these don't sound better in hi-rez...they're not IN hi-rez.

I have found the master tape and recording determine if it really sounds good or not so good, more than any other variable.

This is very true, and is also why you can't rely on your ears alone for what I'm after. I'll repeat what I said above, the variability in recording and sound quality in general from album to album is more than enough to mask the small gains in going from a 320 MP3 to a 16/44 file.

So your next logical question is probably gonna read something like "If that's the case when why do you care". Fair question, I'll answer it and the general "Why does he care" questions that some may have by using a real world scenario.

Let's go with the album I just talked about in another thread, To the Bone by Steven Wilson. There are currently two ways you can buy that album, you can buy the CD from Amazon for $10 or you can head over to HDTracks and pay almost twice the price for the 24/96 version. For my money, I want to know that the 24/96 version I'm paying for is actually that and not just an upsampled CD rip.

If the mastering between both versions is the same (as it often is) then the only advantage of the 24/96 that warrants the extra $8 is the higher resolution, if that number is a lie I want to know that and save my money, simple as that. It's not a big deal on one album, but with a huge CD collection it adds up.

halen

I believe HDTracks was called out years ago for claiming higher bit rate when in fact it was not native. I come from the camp that most are upsampled.

CoolJazz

halen wrote: »

I believe HDTracks was called out years ago for claiming higher bit rate when in fact it was not native. I come from the camp that most are upsampled.

HD Tracks sells what the labels provide them. Because the labels did this, HD Tracks had to get deeper into paying attention to the province of the material just to do what they can to avoid this misguided repeat of wrong information. It's silliness to blame HD Tracks for what someone else does. To me it's an excuse to not try to better the playback experience.

And even sillier yet is to believe "most are upsampled". Nothing could be further from the truth. I've yet to ever find one.

I find the non-discerning crowd to repeatedly confuse the quality of the music production with the carrier of the info. More info certainly carries the potential of carrying better quality or exposing poorer quality. If you wish to mask it with the limitation of lower bit rates and depths or worse, bit rate reduction then you're missing out on a whole 'nother level of musical enjoyment.

Why spend all the money and time and effort for a system to stop just one jump short of much more enjoyment?

CJ

pitdogg2

To be honest I think this whole HD thing has been somewhat of a scam from the get go. It meant to be a money grab. Why should extra bits cost more? I'm firmly in the camp that they are just upsampling 16/44 to whatever they can sell for more. Are the masters recorded at 24 bit 32 bit and 96khz or 192 khz? If they are does that really cost more to do? These days in digital all it takes it seems to me is for someone to include the code to trip a light. They are selling you 16/44 with a code to trip the 24/192 light.
Color me sceptical

heiney9

AsSiMiLaTeD wrote: »

This is very true, and is also why you can't rely on your ears alone for what I'm after. I'll repeat what I said above, the variability in recording and sound quality in general from album to album is more than enough to mask the small gains in going from a 320 MP3 to a 16/44 file.

That is probably true for some recordings and maybe even some genre's of music, but isn't that small gain what most of us in this hobby are looking for?

I mean really, to just say "well it's recorded crappy so might as well get an MP3", it's still degraded vs. the original recording.

I'm not 100% sold on hi rez stuff because of the above, but in many cases I feel the difference is much smaller than going from 320 MP3 to 16/44 file because of the limitations of the source material in some cases. Meaning hi rez may not sound much different than the original because the limitation is the source material not the technology.

But you have to take it on a case by case situation.

MP3's are bad and one should strive to avoid them at all costs if you're serious about audio reproduction. But as always, we are the end consumer and we will always be at the mercy of the production/recording staff. If they produce a poor recording then no amount of tweaking is going to really change that.

I will also say that if something is recorded particularly poorly, I probably wouldn't listen to it all that often, unless it's got other values that help my enjoyment.

H9

CoolJazz

pitdogg2 wrote: »

To be honest I think this whole HD thing has been somewhat of a scam from the get go. It meant to be a money grab. Why should extra bits cost more? I'm firmly in the camp that they are just upsampling 16/44 to whatever they can sell for more. Are the masters recorded at 24 bit 32 bit and 96khz or 192 khz? If they are does that really cost more to do? These days in digital all it takes it seems to me is for someone to include the code to trip a light. They are selling you 16/44 with a code to trip the 24/192 light.
Color me sceptical

You should stop letting your skeptic side outweigh your inquisitive side. Try it, it's awesome hearing the greater refinement.

It's nothing to do with turning on a light! It's the moo-sic!

Get a high quality recording, not just high rez, but high quality and setdown with freeware program and downsample it. Then go listen to the difference and you'll be a skeptic no more!

Or download a professional one that's already done and wonder no more.

And by the way, some of the best are the analog sourced older stuff! The quality really, really shines!

Get more for your time, effort and money and go high-rez, you'll never look back!

CJ

pitdogg2

My hi-rez says SACD and DVD-Audio. Its all I need. If i can't have physical media i do not need it.

halen

CJ,

I can't help but to question the authenticity of all these high res files. Suddenly you have a huge selection of albums, which are all high res. I don't think it is native high rez but upsampled.

I have around six tb of albums ranging from 16/44 up to double or quad DSD. I do question these claimed high res files.

Anyhow. Cheers!

K_M

heiney9 wrote: »

AsSiMiLaTeD wrote: »

This is very true, and is also why you can't rely on your ears alone for what I'm after. I'll repeat what I said above, the variability in recording and sound quality in general from album to album is more than enough to mask the small gains in going from a 320 MP3 to a 16/44 file.

That is probably true for some recordings and maybe even some genre's of music, but isn't that small gain what most of us in this hobby are looking for?

I mean really, to just say "well it's recorded crappy so might as well get an MP3", it's still degraded vs. the original recording.

I'm not 100% sold on hi rez stuff because of the above, but in many cases I feel the difference is much smaller than going from 320 MP3 to 16/44 file because of the limitations of the source material in some cases. Meaning hi rez may not sound much different than the original because the limitation is the source material not the technology.

But you have to take it on a case by case situation.

MP3's are bad and one should strive to avoid them at all costs if you're serious about audio reproduction. But as always, we are the end consumer and we will always be at the mercy of the production/recording staff. If they produce a poor recording then no amount of tweaking is going to really change that.

I will also say that if something is recorded particularly poorly, I probably wouldn't listen to it all that often, unless it's got other values that help my enjoyment.

H9

Im not sure we are talking about just poorly recorded stuff, but the vast amount of normal and fairly good sounding stuff, either the difference is so small, or maybe it is just up rezzed, or maybe different masterings...or maybe.......

In the end, its hard to say anything for sure about Hi-res, that is an absolute and positive and completely true.
I think a lotta people are just not impressed, with all these uncertainties.

lightman1

pitdogg2 wrote: »

To be honest I think this whole HD thing has been somewhat of a scam from the get go. It meant to be a money grab. Why should extra bits cost more? I'm firmly in the camp that they are just upsampling 16/44 to whatever they can sell for more. Are the masters recorded at 24 bit 32 bit and 96khz or 192 khz? If they are does that really cost more to do? These days in digital all it takes it seems to me is for someone to include the code to trip a light. They are selling you 16/44 with a code to trip the 24/192 light.
Color me sceptical

Throw away all your silly fancypants cables, Ivan.

pitdogg2

lightman1 wrote: »

pitdogg2 wrote: »

To be honest I think this whole HD thing has been somewhat of a scam from the get go. It meant to be a money grab. Why should extra bits cost more? I'm firmly in the camp that they are just upsampling 16/44 to whatever they can sell for more. Are the masters recorded at 24 bit 32 bit and 96khz or 192 khz? If they are does that really cost more to do? These days in digital all it takes it seems to me is for someone to include the code to trip a light. They are selling you 16/44 with a code to trip the 24/192 light.
Color me sceptical

Throw away all your silly fancypants cables, Ivan.

done I made my own from Canare..... smarty pants. Not to mention Ken S. made some for me from a coax RG62 I believe. It's an air core RG6.