<div dir="ltr">Several years ago we reached the point of diminishing returns for improvements to fundamental digital audio technology. I think this has been acknowledged in this thread. Technical opportunities for improved audio quality are elsewhere. Marketing opportunities are everywhere. Let's get back to loudness issues.</div><div class="gmail_extra"><br><div class="gmail_quote">On Wed, May 11, 2016 at 11:50 AM, Bob Katz <span dir="ltr"><<a href="mailto:bobkatz@digido.com" target="_blank">bobkatz@digido.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div bgcolor="#FFFFFF" text="#000000">
<p>Guys, isolating variables is really difficult. If you leaving to
conclusions was an Olympic event, audio engineers would get the
medal. I have my own conclusions about sample rate after many many
months of testing, but it's been impossible to engineer a blind
test that's reliable. So we go on, and we hope for the best. <br>
</p>
<p>I like the "idea" of MQA at least as an interim format until the
Internet gets faster, or I can't hear anymore, whichever comes
first :-)</p><span class="HOEnZb"><font color="#888888">
<p><br>
</p>
<p><br>
</p>
<p>Bob</p></font></span><div><div class="h5">
<p><br>
</p>
<p><br>
</p>
<div>On 5/11/16 11:06 AM, Grimm | Eelco
wrote:<br>
</div>
<blockquote type="cite">
<pre>Hi Thomas,
</pre>
<blockquote type="cite">
<pre>What is “difference”? When testing resampling filters, I’ve taken part in tests using a remote controlled Yamaha grand picked up by mics going only through the analog domain, and the same mics taking an AD-DA detour.
</pre>
</blockquote>
<pre>Mark that in an AD-DA path jitter is largely compensated. In a DA-AD chain (re-recording a track via a converter for instance) relevant jitter is even compensated almost completely. So in such a setup you're only testing the distortion of the ADDA chain, the quality of the decimation filters and the sigma delta modulators, plus the analogue path quality. Unfortunately jitter is one of the main issues with digital audio sound quality and it is not taken into the equation in a test like this.
In any case I agree that 48k can sound extremely well and that we should first solve the most significant problems...
Cheers,
Eelco
</pre>
<blockquote type="cite">
<pre>Sometimes subjects would prefer the latter, but results were never really significant. Should that happen, and if the latter was preferred, what would it signify? At the end, we need proof, not more claims.
There haven’t even been credible listening experiments showing that we favor 96 kHz sampled recordings over 48 kHz. It’s almost impossible to make experiements in this regards that eliminate confounders - and that’s what the dubious part of the hi-fi industry is living from.
However, with many of these exotic issues, predictability tends to go down as new patches are invented. That in turn drives FUD and takes away attention from what matters. High sample rates can result in unexpected interference products, sometimes even acoustically, and the same will be true with a “de-blurring” filter and most other processes that alter time/frequency domain parameters. Different, and so what?
It should be in the mastering engineer’s best interest to keep predictability of the downstream path high. Not if and if and if and if...
kind regards,
Thomas
</pre>
<blockquote type="cite">
<pre>On 11/05/2016, at 09.57, Grimm | Eelco <a href="mailto:eelco@grimmaudio.com" target="_blank"><eelco@grimmaudio.com></a> wrote:
Hi Bob,
</pre>
<blockquote type="cite">
<pre>Eelco, Yeah. To my experience I can hear a slight degradation after dithering a 2496 Down to 2096. It's small but enough to bother me. I'm on your side about the sample rate, that the bits are far more important. But I'll keep an open mind and try to listen. Right now it's very difficult... I wonder if I can get the Tidal decoder.
</pre>
</blockquote>
<pre>On second thought I must say I am not sure if Tidal offers the full decoding in their own API. The DAC's decimation filter compensation is probably not present. There's still some mist around MQA for me. I should have a chat with Bob Stuart in Paris at the AES to clarify things a bit. But maybe someone here knows the answers to these questions:
- Is in the current implementations the decoding locked to the DAC? Or just the compensation? Can you compensate without decoding an MQA stream?
- Will Tidal not decode MQA itself when the DAC has an MQA decoder built in?
- Is the AD decimation filter 'compensated'? If so, how?
</pre>
<blockquote type="cite">
<pre>You did some excellent comparison tests. Did you have an MQA digital decoder or just an MQA DAC?
</pre>
</blockquote>
<pre>As far as my memory serves me (I'm not good at remembering things in high detail that happened in the past) Alan had access to an MQA encoder so he could select his own track. In any case we had the original side by side with the MQA encoded and decoded versions. They were all in wav files so we could use Alan's DAC. We also listened to an MQA DAC (since part of the decoding trick is to 'compensate' for the DAC's decimation filter anomalies and that's obviously only available in the DAC itsefl). The DAC sounded worse than Alan's reference DAC, so I cannot really say anything about that.
</pre>
<blockquote type="cite">
<pre>Did Alan agree with your conclusions?
</pre>
</blockquote>
<pre>No. It's also a matter of taste probably. But in my ears MQA was 'different' from the original and the decimated version just slightly reduced in quality. In the past months I have learned that the quality of the decimation filters has to be _extremely_ high to not be audible. But I repeat: I cannot draw a conclusion based upon just one listening test.
</pre>
<blockquote type="cite">
<pre>Comparing MQA to HDCD is not a good analogy.... MQA is far more sophisticated.
</pre>
</blockquote>
<pre>Sure! But it is fun that the same idea of packing > Nyquist energy in the lower bits got rejuvenated and I'm just curious to hear what the performance improvement level really is.
Cheers,
Eelco
</pre>
<blockquote type="cite">
<blockquote type="cite">
<pre>Hi Bob K,
</pre>
<blockquote type="cite">
<pre>Dear Bob L. I have written to Bob Stuart but not gotten any response yet. Of course I want to audition MQA, but I'm using a digital crossover, and for stereo I have 4 channels of DAC so I need a digital MQA "decoder". Any thoughts on that?
</pre>
</blockquote>
<pre>The Tidal MQA decoder should take care of that.
I listened to MQA at Alan Silverman's studio last AES. Alan was very enthusiastic, but in my ears MQA was not identical to the original. Compared to a downsampled version you loose some and you gain some. MQA is based on the presumption that high sampling is more important than more bits: the HF info is coded into the bottom 4 bits, which effectively means you have to choose between 48 / 24 versus 384 / 20.
We compared:
1. Original 96/24 file
2. Saracon downsampled 48/24
3. MQA non-decoded ("48/20")
4. MQA decoded ("96/20")
For me personally #2 came closest to #1. I have to admit I was a bit disappointed, but it also confirmed my experience that with real high quality decimation filters 48 kHz tracks are hard to distinguish from the high sampling original (as a delivery format that is - during production high sampling is a must). But bits do count. Of course my experience could also be effected by MQA's "de-blur" algorithm, which claims to undo the ripple of the original AD converter (or subsequent decimation filters), but I have too little information about it to know what it does exactly and how it 'sounds'.
I'd certainly love to do furter testing on MQA. And just out of curiosity I'd love to compare it to HDCD - which was a similar process, though just with 44/16 packing space, which is obviously worse than 48/24.
All the best,
Eelco
</pre>
<blockquote type="cite">
<pre>On 5/9/16 11:24 AM, Robert Ludwig wrote:
</pre>
<blockquote type="cite">
<pre>Hi Ian,
Just to reiterate, Thomas is 100% right in my opinion, needlessly crushing music is by FAR the #1 item that needs to be fixed for better sound, nothing else matters when you start with unsatisfying music. However, assuming you’ve got great sounding music, then all the rest comes in to play!
The TIDAL MQA material isn’t officially launched yet, but Bob Stuart said there are people who know the URL that would get one to the beta site.
I don’t know how much you know about MQA, but it is several “things” under one MQA name. It is an audio de-blur algorithm, it is a “folding” or data compressing algorithm folding 384 kHz or less into either a 44.1kHz or 48kHz/24 bit file (which can be streamed at 1 - 1.3MB/sec) and it is a digital copy protection scheme where theoretically you could make a one-off DAC that is coded to your converter and only people with the same converter could hear the music!
As far as MQA examples go, without an MQA DAC you can’t hear the whole thing, but one could hear some of the results of the de-blur process using a normal converter. I’m trying to see if I can find you a before and after example. I’ll get back to you.
In the mean time, you can check out Morten Lindberg's 2L label “test bench”. Even though he is recording at 352.8kHz, Morten still finds the MQA to create a superior sound.
It’s a mind blower to see a 44.1kHz/24 bit file feed the MQA converter and the resulting FFT goes right out to 176.4kHz!!
<a href="http://www.2l.no/hires/index.html" target="_blank">http://www.2l.no/hires/index.html</a>
?
Best,
Bob
</pre>
<blockquote type="cite">
<pre>On May 8, 2016, at 6:20 PM, Ian Shepherd <a href="mailto:ian@mastering-media.co.uk" target="_blank"><ian@mastering-media.co.uk></a>
wrote:
Bob L,
Can you recommend a release that really highlights the benefits of MQA ? Is there a way to tell which files on TIDAL are encoded with MQA ? And is there a converter you’d recommend ?
Although I’m afraid Thomas may well be right, I’m curious to give it a try.
Ian
</pre>
<blockquote type="cite">
<pre>On 8 May 2016, at 14:27, Bob Katz <a href="mailto:bobkatz@digido.com" target="_blank"><bobkatz@digido.com></a>
wrote:
Guys:
It's a good thing I buy my music either on CD or from HD tracks or similar vendors. I only buy the AAC from the iTunes store when one of my clients wants me to reference a CD that I don't have and I have to listen to it "now". If then I like that music, then I go out and buy it, usually on CD. Amazon has become my major vendor, I pity the brick and mortar stores, though Best Buy seems to be hanging on, they don't have the CD collection I can get online from Amazon. I can then transfer the CD into iTunes and use on my iPhone if I like.
For my highest-sonic-quality personal entertainment on my iPhone I use an incredible app called HFPlayer from Onkyo..... in conjunction with an Oppo DAC that connects via Lightning directly to the iPhone and plays any high res. flac files I have on the iPhone up to 384 kHz/24 bit (and also plays DSD directly into the ESS Sabre DAC in the Oppo DAC).... It beats the crap out of the Astel and Kearn stuff that is so unreliable and uber-expensive.
Bob L., I tried the Astel and Kearn player and got it to crash in 15 minutes, three times before I called it flakey. And their communication with their dealers and reps is exremely shakey. I do not like this company. If they wanted to make a dependable music player, why did they pick the Android operating system?
Also, sorry, Neil Young, this Oppo/iPhone combination beats the Pono because I can also take phone calls :-(. If I'm jogging or don't want to carry the DAC, the HF Player will downsample the high res material so it will work out of the iPhone's regular headphone jack.
Tip: the Oppo PM-3 headphones are killers! Maybe the most accurate and impacting headphones you can get and carry with you to play music.
If you want to salivate about the best portable playback on earth, in my opinion, read my review of the Oppo pieces and the HF plyaer: visit
<a href="http://www.innerfidelity.com/content/katzs-corner-episode-11-oppo-explosion#RiXt3ToxveQ4BMf5.97" target="_blank">http://www.innerfidelity.com/content/katzs-corner-episode-11-oppo-explosion#RiXt3ToxveQ4BMf5.97</a>
These PM-3 headphones have bass that goes down to the center of the earth, are extremely accurate and tonally neutral.... With a little of the 64-bit EQ in the HF-player you can get even more accurate playback.... and the Oppo DAC/headphone amp is clean and punchy and open....
I have tons of high res flacs that fit in a 64 GB iPhone. If Apple lost for me the purchased AAC music that I have from the iTunes store I'd be pissed, but not that pissed because I just don't have that much money invested in AAC material I bought from the iTunes store. It seems to me that music must be in the iPhone backup that's on my Mac, though, isn't it???? I'm only semi-attached to the cloud. I use icloud for Notes, Calendar, etc., but not for music (I think). I share my iTunes library with a few OSX computers.... maybe that's through iCloud? Yuk....
Bob K.
On 5/7/16 2:39 PM, Thomas Lund wrote:
</pre>
<blockquote type="cite">
<pre>Hi Bob,
This casual treatment of tracks fits with what I'm seeing in iTunes. One has to actively disable cloud replacement time after time so iTunes doesn’t steal your known-good tracks. From Tidal I’ve heard how record companies require them to carry only the newest (crappiest) version of a track, and make sure originals aren’t available.
Compared to such demolishment, I’m afraid MQA is a sheer marketing detail which is bound to disappoint consumers just like the myriad of previous attempts at promising “hidef". 24 bit, 16 bit, 14 bit, 12 bit resolution: Doesn’t matter. Dither or not: Doesn’t matter. Sample rate DSD, 192k, 96, 48, 44.1k: Doesn’t matter. Lossless, 320k, 256k, 128k: Doesn’t matter. The version of a track/squashing is what generally matters.
cheers,
Thomas
</pre>
<blockquote type="cite">
<pre>On 07/05/2016, at 00.20, Robert Ludwig <a href="mailto:gatewaybob@mac.com" target="_blank"><gatewaybob@mac.com></a>
wrote:
<a href="http://www.iphonehacks.com/2016/05/apple-music-isnt-deleting-music-libraries.html" target="_blank">http://www.iphonehacks.com/2016/05/apple-music-isnt-deleting-music-libraries.html</a>
I think this article explains it pretty well. It isn’t as horrible as the blog describes, but it is still taking way too much control for me to ever want to use their service.
BTW, Tidal has 300 albums with MQA so with an MQA converter you can hear streaming 96kHz or even 352.8kHz music!
MQA is really revolutionary.
Best,
Bob L.
On 5/6/16 5:01 PM, Grimm | Eelco wrote:
Dear friends,
If this story is true it is unimaginably horrific.
<a href="https://blog.vellumatlanta.com/2016/05/04/apple-stole-my-music-no-seriously/" target="_blank">https://blog.vellumatlanta.com/2016/05/04/apple-stole-my-music-no-seriously/</a>
Can anyone please tell me it is not true?
Cheers,
Eelco
_______________________________________________
MLA mailing list
<a href="mailto:MLA@grimmaudio.nl" target="_blank">MLA@grimmaudio.nl</a>
<a href="http://grimmaudio.nl/mailman/listinfo/mla_grimmaudio.nl" target="_blank">http://grimmaudio.nl/mailman/listinfo/mla_grimmaudio.nl</a>
--
_______________________________________________
MLA mailing list
<a href="mailto:MLA@grimmaudio.nl" target="_blank">MLA@grimmaudio.nl</a>
<a href="http://grimmaudio.nl/mailman/listinfo/mla_grimmaudio.nl" target="_blank">http://grimmaudio.nl/mailman/listinfo/mla_grimmaudio.nl</a>
</pre>
</blockquote>
</blockquote>
<pre>--
If you want good sound on your album, come to
Bob Katz <a href="tel:407-831-0233" value="+14078310233" target="_blank">407-831-0233</a>
DIGITAL DOMAIN MASTERING STUDIO
Author:
Mastering Audio
Digital Domain Website
No trees were killed in the sending of this message. However a large number
of electrons were terribly inconvenienced.
Please excuse the weird typing - this message dictated by microphone. Contact me by regular email.
_______________________________________________
MLA mailing list
<a href="mailto:MLA@grimmaudio.nl" target="_blank">MLA@grimmaudio.nl</a>
<a href="http://grimmaudio.nl/mailman/listinfo/mla_grimmaudio.nl" target="_blank">http://grimmaudio.nl/mailman/listinfo/mla_grimmaudio.nl</a>
</pre>
</blockquote>
<pre>***
Ian Shepherd
Mastering Media Ltd - Audio Mastering, Blu-ray and DVD Excellence
<a href="http://twitter.com/ianshepherd" target="_blank">http://twitter.com/ianshepherd</a>
Mastering Media Ltd,
Kemp House, 152 City Road, London, EC1V 2NX
Registered in England and Wales
Company No. 073500450
Tel. <a href="tel:%2B44%280%291487%20842498" value="+441487842498" target="_blank">+44(0)1487 842498</a>
_______________________________________________
MLA mailing list
<a href="mailto:MLA@grimmaudio.nl" target="_blank">MLA@grimmaudio.nl</a>
<a href="http://grimmaudio.nl/mailman/listinfo/mla_grimmaudio.nl" target="_blank">http://grimmaudio.nl/mailman/listinfo/mla_grimmaudio.nl</a>
</pre>
</blockquote>
<pre>
_______________________________________________
MLA mailing list
<a href="mailto:MLA@grimmaudio.nl" target="_blank">MLA@grimmaudio.nl</a>
<a href="http://grimmaudio.nl/mailman/listinfo/mla_grimmaudio.nl" target="_blank">http://grimmaudio.nl/mailman/listinfo/mla_grimmaudio.nl</a>
</pre>
</blockquote>
<pre>--
If you want good sound on your album, come to
Bob Katz <a href="tel:407-831-0233" value="+14078310233" target="_blank">407-831-0233</a>
DIGITAL DOMAIN MASTERING STUDIO
Author:
Mastering Audio
Digital Domain Website
No trees were killed in the sending of this message. However a large number
of electrons were terribly inconvenienced.
Please excuse the weird typing - this message dictated by microphone. Contact me by regular email.
_______________________________________________
MLA mailing list
<a href="mailto:MLA@grimmaudio.nl" target="_blank">MLA@grimmaudio.nl</a>
<a href="http://grimmaudio.nl/mailman/listinfo/mla_grimmaudio.nl" target="_blank">http://grimmaudio.nl/mailman/listinfo/mla_grimmaudio.nl</a>
</pre>
</blockquote>
<pre>_______________________________________________
MLA mailing list
<a href="mailto:MLA@grimmaudio.nl" target="_blank">MLA@grimmaudio.nl</a>
<a href="http://grimmaudio.nl/mailman/listinfo/mla_grimmaudio.nl" target="_blank">http://grimmaudio.nl/mailman/listinfo/mla_grimmaudio.nl</a>
</pre>
</blockquote>
<pre>--
If you want good sound on your album, come to
Bob Katz <a href="tel:407-831-0233" value="+14078310233" target="_blank">407-831-0233</a>
DIGITAL DOMAIN MASTERING STUDIO
Author:
Mastering Audio
Digital Domain Website
No trees were killed in the sending of this message. However a large number
of electrons were terribly inconvenienced.
Please excuse the weird typing - this message dictated by microphone. Contact me by regular email.
_______________________________________________
MLA mailing list
<a href="mailto:MLA@grimmaudio.nl" target="_blank">MLA@grimmaudio.nl</a>
<a href="http://grimmaudio.nl/mailman/listinfo/mla_grimmaudio.nl" target="_blank">http://grimmaudio.nl/mailman/listinfo/mla_grimmaudio.nl</a>
</pre>
</blockquote>
<pre>
_______________________________________________
MLA mailing list
<a href="mailto:MLA@grimmaudio.nl" target="_blank">MLA@grimmaudio.nl</a>
<a href="http://grimmaudio.nl/mailman/listinfo/mla_grimmaudio.nl" target="_blank">http://grimmaudio.nl/mailman/listinfo/mla_grimmaudio.nl</a>
</pre>
</blockquote>
<pre>_______________________________________________
MLA mailing list
<a href="mailto:MLA@grimmaudio.nl" target="_blank">MLA@grimmaudio.nl</a>
<a href="http://grimmaudio.nl/mailman/listinfo/mla_grimmaudio.nl" target="_blank">http://grimmaudio.nl/mailman/listinfo/mla_grimmaudio.nl</a>
</pre>
</blockquote>
<pre>
_______________________________________________
MLA mailing list
<a href="mailto:MLA@grimmaudio.nl" target="_blank">MLA@grimmaudio.nl</a>
<a href="http://grimmaudio.nl/mailman/listinfo/mla_grimmaudio.nl" target="_blank">http://grimmaudio.nl/mailman/listinfo/mla_grimmaudio.nl</a>
</pre>
</blockquote>
<br>
<div>-- <br>
<div>
<div>
<div>
<div>
<div>
<div>
<pre cols="80"><font face="Courier">
If you want good sound on your album, come to
Bob Katz <a href="tel:407-831-0233" value="+14078310233" target="_blank">407-831-0233</a> </font><font face="Courier">DIGITAL DOMAIN MASTERING STUDIO
Author: <b>Mastering Audio </b></font><font face="Courier">
<a href="http://www.digido.com/" target="_blank">Digital Domain Website</a>
No trees were killed in the sending of this message. However a large number
of electrons were terribly inconvenienced.<font color="#ff0000">
<b>Please excuse the weird typing - this message dictated by microphone. Contact me by regular email. </b></font></font></pre>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div></div></div>
<br>_______________________________________________<br>
MLA mailing list<br>
<a href="mailto:MLA@grimmaudio.nl">MLA@grimmaudio.nl</a><br>
<a href="http://grimmaudio.nl/mailman/listinfo/mla_grimmaudio.nl" rel="noreferrer" target="_blank">http://grimmaudio.nl/mailman/listinfo/mla_grimmaudio.nl</a><br>
<br></blockquote></div><br></div>