INTERACT FORUM

Please login or register.

Login with username, password and session length
Advanced search  
Pages: [1]   Go Down

Author Topic: Fidelity and other sound quality questions  (Read 5546 times)

MuseChaser

  • Recent member
  • *
  • Posts: 23
Fidelity and other sound quality questions
« on: March 28, 2016, 08:26:03 am »

JRiver MC 21.0.50, Win 7

I purchased JRiver a few months ago, and am still very much impressed with everything it can do.  The motivation behind the purchase was to be able to play various high definition audio files (24/96, 24/192, DSD) as I'd recently picked up two DACs that had those capabilities.  Since then, I've discovered the HUGE amount of other things JRiver does, and it's a pretty impressive program.  So, thank you!

A couple questions.  I've switched from past practice of ripping CDs to 320kb for car and mobile playback using winamp or WMC, and am now ripping to FLAC files using JRiver.  From those files, I convert them to mp3s as needed for various mobile usage.  So far, so good; the mp3s sound more than good enough for on-the-go sound.  I just ripped a Debussy/Ravel (Cleveland Quartet on Telarc) disc to flac files, then burned an audio CD copy from those files, all w/in JRiver.  The copy is obviously degraded from the original.  I've got good equipment, and I work on both sides of the microphone as a full-time musician and part-time recording engineer, so I may be more fussy than most.. who knows.  The copy isn't terrible; no obvious glitches, digital hash or distortion, or anything, but it's duller and flatter.  Almost anyone could pick out the difference instantly in an ABX test.. just did one w/ my wife and she called it immediately.  Of course, she's a classical musician, too.  Any ideas as the cause?  Unusual, or is it to be expected?  Web searches all return "FLAC files are perfect, and CDs burned from them are awesome, dude!" type of responses.

Another question/observation - there's an ongoing debate as to the worth of HD audio recording in general, pretty centered on the increased amount of TIM caused by the additional ultrasonic data present in higher sampling rate files.  I was thrilled by the first few 24/96 and 24/192 recordings I heard, but am now afraid that reaction may have been sort of an "emperor's new clothes" observation.  I've purchased and downloaded a few 24/96 recordings and have been playing them back using a Cambridge Audio DacMagic or a Tascam US366 DAC, and been very happy with the sound.  However, a few days ago I downloaded the 2015 Sampler from HDtracks.com.  I happened to own the CD (Sonny Rollins' "Saxophone Colossus")that one of those 24/96 FLAC sample tracks was taken from.  GREAT!  I could finally ABX the two formats and put my mind at ease.  The CD was played back using either a Parts Connection Assemblage or an Audio Alechemy Dac-in-the-Box w/ an upgraded power supply I designed and built.  The 24/96 FLAC file was played using the two aforementioned USB DACs.  I tried the ABX on two different systems (one was Adcom GFP-565/B&K ST-140/Audio Concepts LV/Sat, the other was Hafler DH-110/Hafler XL-280/Magnepan SMGc), and in every instance, the CD was superior in sound to the 24/96 FLAC in every way; high end extention, front-to-back depth, soundstage focus and stability, timbre, you name it.  The 24/96 FLAC, in comparison, sounded duller and deader.  NOT terrible, by any means, but not as good as the CD.  

I've verified the settings w/in the "Audio/Device" and "Audio/DSP & Output" fields were set to the correct and best device drivers (there's several for both the Cambridge and Tascam DACs; the USB/ASIO drivers give access to the 24/192 capabilities of both units) and that the DSP options were unchecked and no file conversion was taking place.  The FIRST checkbox, "Output Format," in the DSP menu is checked, however all formats from 192khz and down are specified as "no change."  The "channels" option has "2 channels/Stereo" selected, and the next field I've left as "JRSS Mixing (recommended)" because.. well.. who am I to argue w/ recommended settings, right?  ;)

Anyone have any thoughts?  Similar experiences?  What am I missing?

Thanks so much,

Barry
Logged

blgentry

  • Regular Member
  • Citizen of the Universe
  • *****
  • Posts: 8009
Re: Loss of fidelity and other sound quality questions/observations
« Reply #1 on: March 28, 2016, 08:35:55 am »

However, a few days ago I downloaded the 2015 Sampler from HDtracks.com.  I happened to own the CD (Sonny Rollins' "Saxophone Colossus")that one of those 24/96 FLAC sample tracks was taken from.  GREAT!  I could finally ABX the two formats and put my mind at ease.  The CD was played back using either a Parts Connection Assemblage or an Audio Alechemy Dac-in-the-Box w/ an upgraded power supply I designed and built.  The 24/96 FLAC file was played using the two aforementioned USB DACs.  I tried the ABX on two different systems (one was Adcom GFP-565/B&K ST-140/Audio Concepts LV/Sat, the other was Hafler DH-110/Hafler XL-280/Magnepan SMGc), and in every instance, the CD was superior in sound to the 24/96 FLAC in every way; high end extention, front-to-back depth, soundstage focus and stability, timbre, you name it.  The 24/96 FLAC, in comparison, sounded duller and deader.  NOT terrible, by any means, but not as good as the CD. 

Knowing the source of the audio in each recording gets to be a difficult job.  ONE resource for this is the Dynamic Range database.  I see that there are several CD versions of this album, each with a different dynamic range.  Surprisingly, the HD Tracks version has the highest dynamic range of all of them, including the CD with the most DR:

http://dr.loudness-war.info/album/list?artist=sonny&album=sax

Higher dynamic range usually correlates to lower average level.  Meaning that the HD Tracks version probably sounds quieter than the CD version.  As you probably know, small differences in absolute level usually make people pick the one that is louder as sounding "better" with many, many different descriptive terms used to describe how the louder one sounds better.  I've fallen victim to this myself.  It's wired into our ear/brain.

Maybe I'm on the wrong track, but that explanation makes sense to me.

Brian.
Logged

DJLegba

  • Citizen of the Universe
  • *****
  • Posts: 993
Re: Loss of fidelity and other sound quality questions/observations
« Reply #2 on: March 28, 2016, 08:53:08 am »

You could be hearing differences in the DACs, as you're playing the CD and the 24-bit downloads through different equipment. Rip the CDs and play them through your USB DACs if you want to compare.
Logged

MuseChaser

  • Recent member
  • *
  • Posts: 23
Re: Loss of fidelity and other sound quality questions/observations
« Reply #3 on: March 28, 2016, 09:10:03 am »

Knowing the source of the audio in each recording gets to be a difficult job.  ONE resource for this is the Dynamic Range database.  I see that there are several CD versions of this album, each with a different dynamic range.  Surprisingly, the HD Tracks version has the highest dynamic range of all of them, including the CD with the most DR:

http://dr.loudness-war.info/album/list?artist=sonny&album=sax

Higher dynamic range usually correlates to lower average level.  Meaning that the HD Tracks version probably sounds quieter than the CD version.  As you probably know, small differences in absolute level usually make people pick the one that is louder as sounding "better" with many, many different descriptive terms used to describe how the louder one sounds better.  I've fallen victim to this myself.  It's wired into our ear/brain.

Maybe I'm on the wrong track, but that explanation makes sense to me.

Brian.

Thanks for the thoughts, Brian, but that's not it.  I'm accutely aware of the "Louder is Better" phenomena, and made sure to level-match the samples during our listening tests.   You are correct in that I had to reduce the volume of the CD to match the lower average recorded level of the HDtracks cut.  Any other ideas?  If the source(s) is(are) two-channel stereo, does leaving the "JRSS Mixing" option enabled affect anything if the output is two-channel stereo?

Thanks again..

Barry
Logged

MuseChaser

  • Recent member
  • *
  • Posts: 23
Re: Loss of fidelity and other sound quality questions/observations
« Reply #4 on: March 28, 2016, 09:16:30 am »

You could be hearing differences in the DACs, as you're playing the CD and the 24-bit downloads through different equipment. Rip the CDs and play them through your USB DACs if you want to compare.

I had considered that the DACs might be part of it, which is why I auditioned the CD not only through the Parts Connection Assemblage, which was considered an excellent DAC when it came out years ago (20 years, maybe?), but also through the Alchemy Dac-in-the-Box, which was about the cheapest DAC maybe 25 years ago that could be called high fidelity equipment.  The DITB, especially, shouldn't be as good as the Cambridge I used on the 24/96 files, much less any better.  The other issue is, of course, if I rip the CD, then I'm no longer listening to the CD, I'm listening to a FLAC rip.  Yeah, I know.. it's lossless.  It's perfect.  But it's not the CD.  I want to know if the 24/96 files sound better than the CD, not better than a CD rip.

Your point about the DACs is, of course, 100% valid, though; there shouldn't be any variables other than the source being evaluated.  I could use MC to play the CD, then go back and forth between that and the file played by MC.  I'll give that a shot, although you can't switch back and forth instantly between the two so a true ABX is hard to pull off.  I'll give it a shot and report back; thanks for the idea.

Barry
Logged

blgentry

  • Regular Member
  • Citizen of the Universe
  • *****
  • Posts: 8009
Re: Loss of fidelity and other sound quality questions/observations
« Reply #5 on: March 28, 2016, 09:35:27 am »

Any other ideas?  If the source(s) is(are) two-channel stereo, does leaving the "JRSS Mixing" option enabled affect anything if the output is two-channel stereo?

JRSS mixing shouldn't affect anything.  Ideally, you'll want to have all other DSP turned off.  Just leave Output Format checked so it can do sample rate conversion if necessary.  I would disable volume, or use Internal Volume to set 100%.

Maybe... maybe the USB connection to your DACs is poor.  Maybe it's introducing jitter, or some other phenomenon that affects playback.

As for ripping your CD... there's no reason to think that the data from the rip will be any different from the data on the CD.  The method of sending the bits to the DACs is different between a spinning CD with an internal DAC versus a ripped file with an external USB DAC.  But the data itself should be the same.

I too suggest that you rip the CD (with a good ripper in secure mode, or with Accurate Rip), and then A/B the ripped CD version, versus the HD Tracks version.

If they both sound bad (ripped CD and HD Tracks) then it points to something in your computer playback chain.  If the ripped CD version sounds better, then there's something going on with the sources of these two versions of the song.

Brian.
Logged

DJLegba

  • Citizen of the Universe
  • *****
  • Posts: 993
Re: Fidelity and other sound quality questions
« Reply #6 on: March 28, 2016, 09:49:48 am »

When I bought my DAC the first thing I did was connect the digital out from my Meridian 508 CD player to the optical input on the DAC. That way I was able to compare the DAC in the CD player vs the external DAC (exaSound e22). Flipping back and forth they were close. Sometimes I preferred the Meridian, sometimes the e22. But after extended listening through the e22 (without the flipping back and forth) I realized I was enjoying my CD collection a lot more. I don't think it was the "new equipment bias", because that was over a year ago and I still find the e22 really compelling.

I'm not sure how a FLAC rip could sound different from the original CD when played through the same hardware with no additional digital processing, unless you've accepted a non-secure rip (permits read errors during the rip). My Meridian started to get fussy (which was my original incentive for buying the DAC) and there were several discs in my collection that I had to copy with my computer because there were tracks on the original the Meridian wouldn't play. The copies always sounded identical to me.
Logged

Arindelle

  • Citizen of the Universe
  • *****
  • Posts: 2772
Re: Fidelity and other sound quality questions
« Reply #7 on: March 28, 2016, 10:20:41 am »

Personally, all things the same FLAC and WAV and CD almost always sound identical to me (never ever heard any difference bewteen wav and FLAC) Although I have messed up cds that sound worse than their ripped tracks.

have to agree on 2 points already mentioned.

To compare ...

the DAC has to be the same

the volume (gain) has to be identical.

the rip is on a "secure" type mode and is an identical copy to FLAC

Assuming also you have set up JRiver correctly and for this type of test you are not adding any DSPs and are using either Internal volume or disabled volume options. (avoiding the windows mixer). Just a bit more output gain will be noticeable and again as mentioned, most people subjectively respond positively at first listen to just a touch more volume. The power supply could have something to do with it I suppose ... but unless your PC is running on the same plug as your refrig I sort of doubt it.

I have a now older but certainly a very good CD player/dac from Linn. Comparing flacs from JRiver and CDs can/will sound different. Not better necessarily, but different. To AB test from JRiver you can set up 2 separate zones and adjust the gain in one so that it is identical ... as a musician sound engineer you probably have something to measure this, even an iphone ap would get you pretty close. you can also set preamp gain under dsp studio, then hit the overview under Playing Now in the tree and shoot back and forth between zones. That way you can set 30 seconds or whatever short duration your brain memory can handle pretty easily.

Remember expectation bias can play a big role even for the most discerning and objective listener. Even blind I can pick out acoustic guitars, and some "brands" I prefer more than others .. I often will pick the guitar I'm used to. Same goes with DACs.

To really do an ABX test there are some plugins that work, but I only know ones that would work comparing the same files through JRiver like for comparing DACs .. maybe your wife could help you on that.

I suppose you should say "arguably" to be politically correct, but using the same DAC and same output volume a FLAC file would/should sound the same or even better as it doesn't have the additional "transport" involved in a spinning piece of plastic, nor a potentially gunked up laser.
Logged

MuseChaser

  • Recent member
  • *
  • Posts: 23
Re: Fidelity and other sound quality questions
« Reply #8 on: March 28, 2016, 10:48:34 am »

Thanks for all the thoughts and replies, everyone.  I've done a little bit more listening after having ripped my CD copy securely to FLAC files.

I've played back the original CD via JRiver on my laptop so it's going through the same signal chain as the FLAC and the HDTracks 24/96 file.  Here's the observations...

1.  The ripped FLAC file sounds better (more open, more sparkle on the cymbals, a small vocal grunt by the percussionist early in the track is more clearly heard, more depth to the soundstage) than the CD when played back using the laptop's drive.  The laptop is on battery power and not plugged in to AC.  This was the first suprise.  Again, the difference wasn't subtle.

2.  The ripped FLAC sounds better than the 24/96 from HDTracks, for the same reasons above, although not as obviously; the 24/96 sounded better than the CD played via the laptop, too.

These were all played from w/in MC going into a Tascam US366 using the ASIO drivers, and ALL DSP & Output boxes unchecked, and no JRSS mixing.  Playing the physical CD was a clear last in quality.  The 24/96 was a little better, but the securely ripped FLAC was a lot better.  Very weird.

After doing that, I put the CD back in my regular transport and played it through the Assemblage DAC.  No comparison; hands down better than anything else above.

GRANTED.. this wasn't near as easy to do objectively as instantaneously switching back and forth between level matched sources as I did earlier, but I'm pretty confident in what I heard.  

Is it possible that a $449 DAC kit from 1995 is still a much, MUCH better piece of equipment than two different $150-$200 modern-day USB DACs that get decent reviews?   I CAN use the TASCAM US366 as a USB-to-S/PDIF converter and feed the Assemblage, but the Assemblage can't handle anything above 48khz sampling frequency, sadly.

Thanks again for the thoughts.  Any others?  :)
Barry
Logged

blgentry

  • Regular Member
  • Citizen of the Universe
  • *****
  • Posts: 8009
Re: Fidelity and other sound quality questions
« Reply #9 on: March 28, 2016, 11:22:31 am »

The 24/96 was a little better, but the securely ripped FLAC was a lot better.  Very weird.

They are from different sources, so probably different mixes, and probably different Dynamic Range.  Check the Dynamic Range library field for each one and see!

As for the CD spinning in your computer player sounding worse... I have no idea what might be going on there.

Quote
Is it possible that a $449 DAC kit from 1995 is still a much, MUCH better piece of equipment than two different $150-$200 modern-day USB DACs that get decent reviews?

Maybe.  Are you aware that almost all modern day DACs are Delta Sigma?  While the majority of older DACs are R2R "ladder DACs".  The over-sampling and slicing and dicing that go on in a D/S DAC might be doing something to the sound that you don't like.  This is one reason that Schiit's Multi-bit DACs are getting so much attention:  They are NOT D/S DACs and therefore do not slice and dice the digital samples.

Or maybe there's something about the frequency "tilt" of the CD player's DAC that you like?  Hard to say.  It would be interesting if you could try another modern DAC with a good reputation, *particularly* a Mulit-bit DAC.

Brian.
Logged

Arindelle

  • Citizen of the Universe
  • *****
  • Posts: 2772
Re: Fidelity and other sound quality questions
« Reply #10 on: March 28, 2016, 11:52:34 am »

Is it possible that a $449 DAC kit from 1995 is still a much, MUCH better piece of equipment than two different $150-$200 modern-day USB DACs that get decent reviews?   I CAN use the TASCAM US366 as a USB-to-S/PDIF converter and feed the Assemblage, but the Assemblage can't handle anything above 48khz sampling frequency, sadly.

Thanks again for the thoughts.  Any others?  :)
Barry

Sure its possible! cheaper DACS maybe have some more advanced DAC chips, but the rest of their components can be shite.
I'd remove the word objectively from the vocab first off . My Linn Dac/Transport was bought in the late nineties too. it certainly still sounds pretty darn good ... if it had a digital input I'd still be using it. I like it because it sounds "warm" ... but is it better than my Cambridge yes, but not better than a Schitt or a PS Audio that I have. subjectively... But my Linn player was outrageously expensive, and I wouldn't spend anywhere near that much today.

Now I'll repeat myself .. personal opinion only. I think when most people talk "SQ", I'm not convinced they are actually hearing what they claim to.

Be that as it may, do you need to spend more than 200$ for a DAC today ... maybe unless you are good at kits and soldering ...or more than 700$ on a DAC -- I personally don't think so, but some insist on it. They can sound different. Sometimes (but not always). But better? thats something else again. What is a SPDIF converter worth its salt going to cost? 150$ to much more. Seems a waste of money IMO (opinion guys, lets not go ape on this) paying 150-200USD currently is not going to get you a great DAC. It not just the DAC chip, its the other stuff going on like its power supplies, clock, sampling choices etc. For 400-600$ there is a large choice of very good external DACS around (check out Schitt dacs ... they blow out DACs 3 times the price and you can keep it for a month to test it for a small fee). On the high end I find it silly to spend more than $700, but hey thats just me. Or my ears ...

Frankly I went down the high-end path and in the digital part of the system you don't have to spend a lot ... better to put the dinero in the analogue (speaker then amps) part. With JRiver you can even do active crossovers; do without a preamp and just go into monoblocks saving a great deal of money if you know what you are doing.  But there is still is a balancing act .. you have a DAC that cost the same amount as your speakers? or your speakers cost 5K and you have a 150 dollar DAC?  one might want to test some gear out to level it a bit. But otherwise its really subjective.

I'm a musician too and I try more and more to listen to the music and not my system. Is it transparent, do I like the soundstage "depth". If so, I'm a happy camper and probably should put more time in speaker placement and room conditioning than paying for expensive gear -- and go to more live concerts and supporting the artists music.

Just my 2 cents of course, but from what you have described its the DAC not the format.  I see Brian mentioned Schiit too while I'm writing this ... might be worth looking into emulating your DAC type and sampling method in a more modern bit of kit.
Logged

sdmarquart

  • World Citizen
  • ***
  • Posts: 122
Re: Fidelity and other sound quality questions
« Reply #11 on: March 28, 2016, 12:08:34 pm »

JRiver MC 21.0.50, Win 7

A couple questions.  I've switched from past practice of ripping CDs to 320kb for car and mobile playback using winamp or WMC, and am now ripping to FLAC files using JRiver.  From those files, I convert them to mp3s as needed for various mobile usage.  So far, so good; the mp3s sound more than good enough for on-the-go sound.  I just ripped a Debussy/Ravel (Cleveland Quartet on Telarc) disc to flac files, then burned an audio CD copy from those files, all w/in JRiver.  The copy is obviously degraded from the original.  I've got good equipment, and I work on both sides of the microphone as a full-time musician and part-time recording engineer, so I may be more fussy than most.. who knows.  The copy isn't terrible; no obvious glitches, digital hash or distortion, or anything, but it's duller and flatter.  Almost anyone could pick out the difference instantly in an ABX test.. just did one w/ my wife and she called it immediately.  Of course, she's a classical musician, too.  Any ideas as the cause?  Unusual, or is it to be expected?  Web searches all return "FLAC files are perfect, and CDs burned from them are awesome, dude!" type of responses.

Thanks so much,

Barry

Barry - no need to re-rip FLAC files to 320kbps. Use the JRemote on your mobile and transcode at "high quality" and it will stream all your FLAC files at 320kbps. It's an awesome feature!

Thanks,
Spence
Logged

imeric

  • MC Beta Team
  • Citizen of the Universe
  • *****
  • Posts: 1467
Re: Fidelity and other sound quality questions
« Reply #12 on: March 28, 2016, 01:24:12 pm »

This is fun to read thanks folks!!

IMHO a hard drive is a much better transport than a CD so I don't even bother listening to CDs I just rip them right away to ALAC (for compatibility but FLACs are as good/convenient).

Some will say AIFF or WAV should be used vs any of the lossless format such as FLAC or ALAC. Personally I can't tell the difference bwn the 2 and I don't see why there would one.  Especially if you play from memory...Should there be one? Maybe..I'm not one to tell...On a very slow PC/Software combo introducing some sort of Jitter I guess?

I often have different versions of the same albums...Vinyl, redbook, SACD/DSD and HD 24/88.2, 96 or 192 and they all sound quite different (not talking about remaster versions but the "same" version...Even then how do we know for sure we're getting the exact same source/original recording??  Most likely we're not getting it from the master tapes...So many CDs I own that you can easily tell with the snap, crackle and pop they are coming from a Vinyl...And I'm talking about legally purchased CDs here...

In general I do tend to prefer DSD or vinyl but Redbook is still amazing even though it MAY sound less opened/dynamic or fluid (silky high ends...)

And yes I do have Redbook/CD versions that I prefer over "HD" or vinyl...

Everything is SO subjective and yes volume level plays an important part and couldn't agree more on this alone...

On this topic I often have the impression that when I reduce volume in MC (or on the e22 which has an MC plugin which will reduce the digital gain in the DAC itself...This is nice!)

I "feel" (Or maybe I just want to believe) it will have some sort of impact on SQ...Placebo? Maybe...Yes it's all done at 64 bits in the digital domain but I usually leave volume levelling OFF and put my e22 volume at 100%.  I'm fairly sure I can tell when Volume Levelling is ON in MC..Then again is my brain playing tricks on me? Maybe...

And yes DACs will definitely sound different (well the analogue part of the DAC will for sure :)...

And softwares will sound different also...I did an A/B with the Wife of MC vs Foobar and I got 9/10 right...(And I really wanted Foobar to win but it lost against MC...:) )
 
Logged

MikeO

  • Citizen of the Universe
  • *****
  • Posts: 792
Re: Fidelity and other sound quality questions
« Reply #13 on: March 29, 2016, 04:25:22 am »

Hi

I assume you have loaded the Cambridge Audio USB 2 audio driver that you can download from CA. I believe it makes a difference.

Just my 2 penneth...

The DAC Magic wasn't the best of CA's DAC, the DAC Magic Plus was meant to be better. That said I had the DAC Magic for a long time about the time I was ripping my collection of CD's . Maybe my ears are getting old but I couldn't spot any difference using my Sennheiser HD800's

I currently listen "independent of JRiver, I use the Cambridge Audio CXN with an External USB HDD , this is the same DAC as the DAC Magic Plus, and the same effect,  little or no difference. I have taken to replacing some of my rips of older CD's with HD tracks (eg Layla & Other Assorted Love songs, Santana Abraxas, Tango in the Night.... Invariably I can spot minor differences but is that down to format or more care and attention in the mastering process, we'll never know. Many of these CD's date back to the 70's so what do we expect?

Mike
Logged
Pages: [1]   Go Up