24bit vs 16bit, the myth exploded!

rrod said:

   
[redacted] Oh nm, it's you analog 

Click to expand...

Even if I nm (redacted), RRod please try to go trough the entire post and, if required, read in between the lines .

analogsurviver said:

  Even if I nm (redacted), RRod please try to go trough the entire post and, if required, read in between the lines

.

Click to expand...

 
More I didn't want to rehash things when I know well your position. I agree with you on the DG sound, btw, though there's too much good material to avoid.

rrod said:

   
More I didn't want to rehash things when I know well your position. I agree with you on the DG sound, btw, though there's too much good material to avoid.

Click to expand...

No problem - I share the same opinion.
 
I never meant to imply to avoid DG altogether - music comes first. I only wanted to stress the point it simply is not a benchmark in SQ. Under the current umbrella of Universal, on average it is the worst of the three. No contest.
 
Perhaps the best, in terms of SQ, from the current Universal catalog, are the last Philips recordings done in analog. It was a memorable swan song !

soundsgoodtome said:

  So I found out dbpoweramp needs to be purchased to convert flac to mp3 after trial expires so I needed more time to get the CDs out of storage and ripped to mp3 via EAC.

Then this came in today, Steven Wilson's latest album HAND. CANNOT. ERASE. which includes a download in 16/44 both FLAC and MP3!


And so after I get a pattern down for a proper abx technique i was able to get an accurate hearing of wether a/b=x or if a/b=y. I will admit it wasn't easy and took several loops of the same 5-15sec segments but here's my result with my secondary chain of: Foobar ABX <cheap usb cable> Vlink 24/96 <spdif rca cable> Audio-gd NFB11 >> Fidue A83 triple hybrid iem. My He560 at home is a better rig and I'd expect the same results.

A bit blurry so 9/10 with 1.1% chance of guessing. ABX is not half as fun as actually listening to albums. Totally a snooze fest that requires concentration but yes I'll take my FLAC lossless anytime over lossy, even on my less than great Samsung Galaxy S3. YMMV



Tags: @kraken2109 @bigshot @sonitus mirus  @Music Alchemist @RRod @Stillhart  @AxelCloris @Ivabign

Click to expand...

 
So you weren't able to convert the file and ensure that it was encoded properly, you simply downloaded 2 separate files?  You don't even have the necessary tools to convert a file or rip a CD to a properly encoded lossy format?  Were the 2 files you tested even volume matched?  Were you allowing the spectrogram visualizer to run during the testing?  
 
If it was me, I would certainly want to investigate a bit more thoroughly. I'm able to identify a difference between the Tidal music's FLAC and AAC 320 test every time 15/15 in under 3 minutes, but it was discovered that there was a subtle EQ being applied to one of the formats that seemed out of place and did not occur when I encoded the same music to the same formats with EAC or dBPoweramp.  Maybe the files are different due to an inferior encoding process?  
 
Unless your only goal is to show others that you can hear a difference, there are a lot of issues to address with the methodology of your test.  If you really want to know for yourself if you can hear a difference, there is some more research required to be more certain.

Lol more hoops huh? I'll re-rip a CD in mp3 and recheck. I don't know how to run the spectogram during an abx so i wouldn't know. However a flac v mp3 abx from the same cd should make this irrelevant.

But I'm guessing the mp3 and flac rip from Kscope is the same master, I can email them to verify.

There's a sound difference and to me that's all I need to know I'm sticking with Flac.

sonitus mirus said:

So you weren't able to convert the file and ensure that it was encoded properly, you simply downloaded 2 separate files?  You don't even have the necessary tools to convert a file or rip a CD to a properly encoded lossy format?  Were the 2 files you tested even volume matched?  Were you allowing the spectrogram visualizer to run during the testing?  

If it was me, I would certainly want to investigate a bit more thoroughly. I'm able to identify a difference between the Tidal music's FLAC and AAC 320 test every time 15/15 in under 3 minutes, but it was discovered that there was a subtle EQ being applied to one of the formats that seemed out of place and did not occur when I encoded the same music to the same formats with EAC or dBPoweramp.  Maybe the files are different due to an inferior encoding process?  

Unless your only goal is to show others that you can hear a difference, there are a lot of issues to address with the methodology of your test.  If you really want to know for yourself if you can hear a difference, there is some more research required to be more certain.

Click to expand...

rrod said:

   
There's more to these tests than the final sample ratio, but getting that all worked out on a forum is a messy business that usually pisses someone off ^_^

Click to expand...

 

soundsgoodtome said:

Lol more hoops huh? I'll re-rip a CD in mp3 and recheck. I don't know how to run the spectogram during an abx so i wouldn't know. However a flac v mp3 abx from the same cd should make this irrelevant.

But I'm guessing the mp3 and flac rip from Kscope is the same master, I can email them to verify.

Click to expand...

 
See what I mean with my quote above

Yes. I'll say I'm just verifying for my own experience and everyone's mileage will vary along with tolerance of difference. If I were to simply enjoy music I think mp3 would be more than enough, however I want all the data audible or not in my files. Also, if we weren't spending so much on audio gear I'd go MP3 but since people pay hundreds for an extra detail in sound, why short yourself in the data?

rrod said:

See what I mean with my quote above

Click to expand...

I'm not mad at anything or anyone.  It was just that your comments suggested that you had very little experience converting a file or CD to a lossy format or with the ABX tools.  There were simply too many unanswered questions from what I read to indicate that the test was properly conducted.   I included a recent example where I had no control over the test files created, and in the end it was discovered that the files were actually different where there was no rational reason for there to have been any difference.
 
I'd be more curious about my results and would want to investigate further, but that is my personality.  I am a natural skeptic.
 
I seriously considered purchasing this music so that I could analyze the downloads that you tested.  So, I certainly have issues.  

It's a really good album. Buy the BD-A and use the one-time code provided (free) to get both mp3 and flac in a single zip file.

The CD might have the same codes too but I'd verify before buying.

analogsurviver said:

  I like DSD particularly for the fact that it can not be mangled with PCM>DSP every time some computer geek would like to sell yet another DSP software of one kind or another.
 
IF you personally can fathom the thought of making a recording and playing it back without any computer (except the recorder)  within miles, then you might, just perhaps might start hearing the advantages of DSD over PCM. With live mike feed as a reference. And not remasters of masters done by third and fourth party, at and during which you most definitely were never present.

Click to expand...

 
Sorry, but this seems like willful ignorance. I don't know of a single headphone or loudspeaker (even high end) which would not benefit from digital EQ. And yes, that EQ will bring you closer to the live feed you mentioned. That is just one of several reasons why I am only interested in a computer-based digital audio system.
 
As for DSD releases, they are mostly captured and mixed using PCM anyway, and the final result is transcoded to DSD, a process which just introduces distortion and noise (not that it will be audible in the real world though). Still, it is always better to get the PCM release if there is a choice.

soundsgoodtome said:

Lol more hoops huh? I'll re-rip a CD in mp3 and recheck.

Click to expand...

 
AAC 320, not MP3

soundsgoodtome said:

If I were to simply enjoy music I think mp3 would be more than enough, however I want all the data audible or not in my files.

Click to expand...

 
That's fine, but you are saying you can hear what has always been inaudible for everyone else. If you just want nice big chunky file, save everything as WAV and fill up hard drives like crazy. If you don't care if lossy sounds exactly the same or not, there's no reason to take the test.

lamode said:

   
Sorry, but this seems like willful ignorance. I don't know of a single headphone or loudspeaker (even high end) which would not benefit from digital EQ. And yes, that EQ will bring you closer to the live feed you mentioned. That is just one of several reasons why I am only interested in a computer-based digital audio system.
 
As for DSD releases, they are mostly captured and mixed using PCM anyway, and the final result is transcoded to DSD, a process which just introduces distortion and noise (not that it will be audible in the real world though). Still, it is always better to get the PCM release if there is a choice.

Click to expand...

Sorry, if the headphone or speaker/room requires more than a decent analog parametric EQ, it should be optimized differently. There is a reason why manufacturers strive for inherently as flat response as possible. 
 
I agree that recording PCM and then bounce it to DSD does not present the proper way to do it. Recording DSD and editing with the least invasive PCM tools only at and around the edit point yields approx 90% pure DSD that can be provided to the end user. That means those 90% are pure DSD that never was in any form of PCM. Provided musicians are really up to the task and allow minor errors in playing to be issued, that percentage can reach 100%.
 
The inability to edit in pure DSD is to me an advantage; it is essentially direct to XYZ process, giving an honest account of the musicians' capabilities. It is the PCM fix/correct/master to 100 % note by note perfection we grew so accustomed with and spoiled by that is creating this problem in the first place. One could rely on the recording to be a true sound of musician(s) far better before PCM started to change things - sometimes to unrecognizability. 
 
With today's tools (PCM, DSP ), my howling or piano banging may well win a Grammy - if I find a competently crazy enough recording engineer and/or computer geek and can afford his/hers fee. 

analogsurviver said:

  Sorry, if the headphone or speaker/room requires more than a decent analog parametric EQ, it should be optimized differently.
 

Click to expand...

 
Why should I use analogue EQ when digital is so much better?
 

Originally Posted by analogsurviver /img/forum/go_quote.gif
 
Recording DSD and editing with the least invasive PCM tools only at and around the edit point yields approx 90% pure DSD that can be provided to the end user. That means those 90% are pure DSD that never was in any form of PCM. Provided musicians are really up to the task and allow minor errors in playing to be issued, that percentage can reach 100%.

Click to expand...

 
It doesn't work that way - if you want to use more than a stereo pair of microphones you'll need to mix the channels and that requires PCM too. Same goes for anyone recording in more than one take.
 

Click to expand...

Originally Posted by analogsurviver /img/forum/go_quote.gif
  The inability to edit in pure DSD is to me an advantage;

Click to expand...

 

Click to expand...

But impractical for almost all modern recordings, so basically irrelevant. Post tools can be abused but they are also very useful and advantageous in the right hands. Somewhat similar to Photoshop and photography.
 
At the end of the day, artists want to mix, and use multiple takes, so PCM will be necessary and DSD becomes irrelevant. DSD is difficult to use and offers no advantage, only disadvantages.

lamode said:

   
Why should I use analogue EQ when digital is so much better?
 

 
It doesn't work that way - if you want to use more than a stereo pair of microphones you'll need to mix the channels and that requires PCM too. Same goes for anyone recording in more than one take.
 

Click to expand...

 

Click to expand...

But impractical for almost all modern recordings, so basically irrelevant. Post tools can be abused but they are also very useful and advantageous in the right hands. Somewhat similar to Photoshop and photography.
 
At the end of the day, artists want to mix, and use multiple takes, so PCM will be necessary and DSD becomes irrelevant. DSD is difficult to use and offers no advantage, only disadvantages.

Click to expand...

Why do you think analog EQs offer equalizing WAY above 20 kHz - with center frequency as high as 50 kHz ?
 
Why do you think I avoid more than two microphones and mixing - like a pleague ?
 
It can be recorded in several takes and DSD changed to PCM for editing only around/at edit mark.
 
Post tools never equal a well captured natural sound. "Photoshoped" music is precisely as unatural as photoshoped pictures - in real life, it does not exist.