mansr Posted August 8, 2019 Share Posted August 8, 2019 4 minutes ago, church_mouse said: For the non-experts, like me, is this suggesting that if we want to start screwing up our music then get the MQA version. However, if you want to complete the job make sure you get an MQA Dac too? Kind of a ‘‘double whammy‘‘. That's one way of putting it. Not a bad one either. Link to comment
John_Atkinson Posted August 8, 2019 Share Posted August 8, 2019 6 minutes ago, psjug said: JA seems to agree that the results are valid. But I assume he still maintains that leaky filters are OK because the impulse response is so visually appealing. Actually, it is the fact that listeners seem to tend to prefer the leaky DAC reconstruction filters with most kinds of music. Certainly on my commercially released recordings, which people seem to like the sound of, I use the Ayre QA-9 set to "Listen," which has a slow-rolloff antialiasing filter, as the master A/D converter. The possibility for there being some low-level aliasing energy seems to be offset against the better time-domain performance. See the discussion of this trade-off at https://www.stereophile.com/reference/104law/index.html Regarding the behavior of the upsampling performed by the renderer, I assume this will depend on which MQA filter is being used for the upsampling. Perhaps mansr could tell use which one is specified in his file's embedded MQA data? If a filter isn't specified in his data, then the spectral contamination will depend on which upsampling filter the renderer selects as its default. John Atkinson Technical Editor, Stereophile Link to comment
Popular Post mansr Posted August 8, 2019 Popular Post Share Posted August 8, 2019 4 minutes ago, John_Atkinson said: Actually, it is the fact that listeners seem to tend to prefer the leaky DAC reconstruction filters with most kinds of music. Got a citation for that? 5 minutes ago, John_Atkinson said: Regarding the behavior of the upsampling performed by the renderer, I assume this will depend on which MQA filter is being used for the upsampling. Perhaps mansr could tell use which one is specified in his file's embedded MQA data? It's filter 4. When I checked over 100 tracks on Tidal, this was used by 70% of them. 5 minutes ago, John_Atkinson said: If a filter isn't specified in his data, then the spectral contamination will depend on which upsampling filter the renderer selects as its default. It's not possible to not specify a filter. The Computer Audiophile, MikeyFresh and esldude 3 Link to comment
Popular Post botrytis Posted August 8, 2019 Popular Post Share Posted August 8, 2019 33 minutes ago, John_Atkinson said: Actually, it is the fact that listeners seem to tend to prefer the leaky DAC reconstruction filters with most kinds of music. Certainly on my commercially released recordings, which people seem to like the sound of, I use the Ayre QA-9 set to "Listen," which has a slow-rolloff antialiasing filter, as the master A/D converter. The possibility for there being some low-level aliasing energy seems to be offset against the better time-domain performance. See the discussion of this trade-off at https://www.stereophile.com/reference/104law/index.html Regarding the behavior of the upsampling performed by the renderer, I assume this will depend on which MQA filter is being used for the upsampling. Perhaps mansr could tell use which one is specified in his file's embedded MQA data? If a filter isn't specified in his data, then the spectral contamination will depend on which upsampling filter the renderer selects as its default. John Atkinson Technical Editor, Stereophile Problem being, MQA files are NOT what the Engineers and Music Artists wanted when they recorded the music in the first place. So, MQA saying they are more accurate is a pile of excrement. As far as time domain, they have NEVER showed any data that this is a problem so they have a fix to a problem that doesn't exist. MikeyFresh and Ran 1 1 Current: Daphile on an AMD A10-9500 with 16 GB RAM DAC - TEAC UD-501 DAC Pre-amp - Rotel RC-1590 Amplification - Benchmark AHB2 amplifier Speakers - Revel M126Be with 2 REL 7/ti subwoofers Cables - Tara Labs RSC Reference and Blue Jean Cable Balanced Interconnects Link to comment
John_Atkinson Posted August 8, 2019 Share Posted August 8, 2019 6 minutes ago, mansr said: 14 minutes ago, John_Atkinson said: Regarding the behavior of the upsampling performed by the renderer, I assume this will depend on which MQA filter is being used for the upsampling. Perhaps mansr could tell use which one is specified in his file's embedded MQA data? It's filter 4. When I checked over 100 tracks on Tidal, this was used by 70% of them. Thanks. John Atkinson Technical Editor, Stereophile Link to comment
The Computer Audiophile Posted August 8, 2019 Share Posted August 8, 2019 23 minutes ago, John_Atkinson said: Actually, it is the fact that listeners seem to tend to prefer the leaky DAC reconstruction filters with most kinds of music. At least those of us peasants can still select a different filter with Ayre and other non-MQA music / components 😁 Also, by saying "listeners" you mean more than two or just two? 😁 MikeyFresh 1 Founder of Audiophile Style | My Audio Systems Link to comment
John_Atkinson Posted August 8, 2019 Share Posted August 8, 2019 2 minutes ago, The Computer Audiophile said: Also, by saying "listeners" you mean more than two or just two? 😁 My own experience leads me to prefer "leaky" reconstruction filters for most kinds of music; as does that of some of Stereophile team of reviewers; as does that of readers with whom I have discussed the subject. I don't see why anyone would have a problem with people citing their own experience as a reason for having a preference. John Atkinson Technical Editor, Stereophile beetlemania 1 Link to comment
The Computer Audiophile Posted August 8, 2019 Share Posted August 8, 2019 4 minutes ago, John_Atkinson said: My own experience leads me to prefer "leaky" reconstruction filters for most kinds of music; as does that of some of Stereophile team of reviewers; as does that of readers with whom I have discussed the subject. I don't see why anyone would have a problem with people citing their own experience as a reason for having a preference. John Atkinson Technical Editor, Stereophile Agree. I have no problem with it. Sometimes it's all we have to go on, but it only carries the weight of what it is. Founder of Audiophile Style | My Audio Systems Link to comment
Popular Post botrytis Posted August 8, 2019 Popular Post Share Posted August 8, 2019 7 minutes ago, John_Atkinson said: My own experience leads me to prefer "leaky" reconstruction filters for most kinds of music; as does that of some of Stereophile team of reviewers; as does that of readers with whom I have discussed the subject. I don't see why anyone would have a problem with people citing their own experience as a reason for having a preference. John Atkinson Technical Editor, Stereophile Maybe old ears prefer it? It seems many reviewers in Stereophile are pushing 65+. I will say, there is nothing wrong with it UNTIL one waxes poetic and claims it is the best thing since sliced bread. Then you are pushing the envelope. Aren't these the same people who appreciate speakers and systems that are so accurate as to be annoying, like fingernails on a chalkboard? Then they add distortion to the system, artificially through listneing to MQA files? My mind is reeling...... MikeyFresh, Samuel T Cogley and crenca 2 1 Current: Daphile on an AMD A10-9500 with 16 GB RAM DAC - TEAC UD-501 DAC Pre-amp - Rotel RC-1590 Amplification - Benchmark AHB2 amplifier Speakers - Revel M126Be with 2 REL 7/ti subwoofers Cables - Tara Labs RSC Reference and Blue Jean Cable Balanced Interconnects Link to comment
manisandher Posted August 8, 2019 Share Posted August 8, 2019 1 hour ago, mansr said: The spikes at 66 kHz, 76 kHz, and 96 kHz are the result of intermodulation between the 10 kHz and 86 kHz frequencies. Surely the spikes at 66 kHz and 76kHz are simply the images of the 30 kHz and 20 kHz harmonics, respectively? 1 hour ago, mansr said: No idea what's going on at 94 kHz. Anyone care to speculate? Nothing of interest, as far as I can see. Did you mean ~70 kHz and ~90 kHz? Edit: I was looking at @vortecjr's FFT. Main: SOtM sMS-200 -> Okto dac8PRO -> 6x Neurochrome 286 mono amps -> Tune Audio Anima horns + 2x Rotel RB-1590 amps -> 4 subs Home Office: SOtM sMS-200 -> MOTU UltraLite-mk5 -> 6x Neurochrome 286 mono amps -> Impulse H2 speakers Vinyl: Technics SP10 / London (Decca) Reference -> Trafomatic Luna -> RME ADI-2 Pro Link to comment
crenca Posted August 8, 2019 Share Posted August 8, 2019 48 minutes ago, John_Atkinson said: Actually, it is the fact that listeners seem to tend to prefer the leaky DAC reconstruction filters with most kinds of music. Certainly on my commercially released recordings, which people seem to like the sound of, I use the Ayre QA-9 set to "Listen," which has a slow-rolloff antialiasing filter, as the master A/D converter. The possibility for there being some low-level aliasing energy seems to be offset against the better time-domain performance. See the discussion of this trade-off at https://www.stereophile.com/reference/104law/index.html Regarding the behavior of the upsampling performed by the renderer, I assume this will depend on which MQA filter is being used for the upsampling. Perhaps mansr could tell use which one is specified in his file's embedded MQA data? If a filter isn't specified in his data, then the spectral contamination will depend on which upsampling filter the renderer selects as its default. John Atkinson Technical Editor, Stereophile As one president said to another, "There you go again" 1) What "listeners"? Certain audiophiles with tipped up systems who like their HF blurred/distorted because it sounds "real" to them? I also occasionally listen to music with min phase (i.e. out of phase) filters, but then I don't falsely tell myself that I am listening to a "better time-domain" waveform because the opposite is true 2) Are you confusing the performance of a single DAC (a sample of one) with a trend in listeners preferences? Is the linear phase filtering on the Ayre QA-9 that bad? 3) Explain to us, exactly, how min phase filtering is in fact "better time-domain performance". You will of course have to start at the beginning (i.e. defining terms, etc.), and not these peculiar Stereophile/audiophile assumptions about "time domain" vis-a-vis min phase filtering you assume. esldude 1 Hey MQA, if it is not all $voodoo$, show us the math! Link to comment
Popular Post crenca Posted August 8, 2019 Popular Post Share Posted August 8, 2019 13 minutes ago, John_Atkinson said: My own experience leads me to prefer "leaky" reconstruction filters for most kinds of music; as does that of some of Stereophile team of reviewers; as does that of readers with whom I have discussed the subject. I don't see why anyone would have a problem with people citing their own experience as a reason for having a preference. John Atkinson Technical Editor, Stereophile Ok, you like the distortion of min phase filtering. Why would you want music in an "end to end" IP/DRM container that would force this particular preference on everybody? esldude, maxijazz and MikeyFresh 2 1 Hey MQA, if it is not all $voodoo$, show us the math! Link to comment
mansr Posted August 8, 2019 Share Posted August 8, 2019 7 minutes ago, manisandher said: Surely the spikes at 66 kHz and 76kHz are simply the images of the 30 kHz and 20 kHz harmonics, respectively? No. Non-linear distortion can only happen after the interpolation filter where the imaging occurs. Link to comment
John_Atkinson Posted August 8, 2019 Share Posted August 8, 2019 3 minutes ago, crenca said: Ok, you like the distortion of min phase filtering. If you say so.I always find it amusing when someone trues to "prove" that some else's preference is wrong. 🙂 John Atkinson Technical Editor, Stereophile Teresa 1 Link to comment
Popular Post mansr Posted August 8, 2019 Popular Post Share Posted August 8, 2019 Just now, John_Atkinson said: If you say so.I always find it amusing when someone trues to "prove" that some else's preference is wrong. 🙂 Nobody is saying your preference is wrong for you. They are saying it is wrong for them. MikeyFresh, botrytis, esldude and 2 others 1 2 2 Link to comment
Popular Post JoeWhip Posted August 8, 2019 Popular Post Share Posted August 8, 2019 I am no fan of MQA and try not to chim in on this thread, but after the latest analysis, why would anyone be in favor of MQA? Any sound manipulation can be accomplished either with selectable filters in a DAC or via playback software. There is NO need for the MQA format unless the whole thing really is for DRM. Manipulate your music to sound like whatever you like, but don’t force those choices on consumers who may actually prefer fidelity to the source. MikeyFresh, esldude, The Computer Audiophile and 2 others 2 3 Link to comment
manisandher Posted August 8, 2019 Share Posted August 8, 2019 5 minutes ago, mansr said: No. Non-linear distortion can only happen after the interpolation filter where the imaging occurs. It looks obvious to me that the spikes at 56 kHz, 66 kHz and 76 kHz are images of the harmonics at 40 kHz, 30 kHz and 20 kHz, respectively. (Much easier to see on @vortecjr's plot.) Main: SOtM sMS-200 -> Okto dac8PRO -> 6x Neurochrome 286 mono amps -> Tune Audio Anima horns + 2x Rotel RB-1590 amps -> 4 subs Home Office: SOtM sMS-200 -> MOTU UltraLite-mk5 -> 6x Neurochrome 286 mono amps -> Impulse H2 speakers Vinyl: Technics SP10 / London (Decca) Reference -> Trafomatic Luna -> RME ADI-2 Pro Link to comment
Popular Post John_Atkinson Posted August 8, 2019 Popular Post Share Posted August 8, 2019 5 minutes ago, mansr said: Nobody is saying your preference is wrong for you. My mistake. I was taking Crenca's words literally. Quote They are saying it is wrong for them. And I certainly don't have a problem with that. I haven't said otherwise. John Atkinson Technical Editor, Stereophile maxijazz, Teresa and MikeyFresh 1 1 1 Link to comment
Popular Post crenca Posted August 8, 2019 Popular Post Share Posted August 8, 2019 7 minutes ago, John_Atkinson said: If you say so.I always find it amusing when someone trues to "prove" that some else's preference is wrong. 🙂 John Atkinson Technical Editor, Stereophile 6 minutes ago, mansr said: Nobody is saying your preference is wrong for you. They are saying it is wrong for them. Really, your going to assert that min phase is on the one hand "better time domain performance" and on the other when asked to say exactly how it is better, more true to the waveform, your going to retreat into a subjective "ah shucks, its just my preference and your being an objectivist meany weany by asking me to prove it"?!? You can't have it both ways. Which is it? Is it objective "better time domain performance" or is it just a preference and distortion? Teresa, botrytis and esldude 1 1 1 Hey MQA, if it is not all $voodoo$, show us the math! Link to comment
John_Atkinson Posted August 8, 2019 Share Posted August 8, 2019 Just now, crenca said: Which is it? Is it objective "better time domain performance" Yes, a minimum-phase reconstruction filter offers better time-domain performance, for the reasons I outlined in my 2011 Richard Heyser Memorial lecture to the AES. Google will be your friend here. Just now, crenca said: or is it just a preference It is also my preference, as it is for the others I mentioned a few message back. Just now, crenca said: and distortion? I don't see why a filter having minimum-phase behavior is in itself a "distortion." John Atkinson Technical Editor, Stereophile Teresa 1 Link to comment
mansr Posted August 8, 2019 Share Posted August 8, 2019 14 minutes ago, manisandher said: It looks obvious to me that the spikes at 56 kHz, 66 kHz and 76 kHz are images of the harmonics at 40 kHz, 30 kHz and 20 kHz, respectively. (Much easier to see on @vortecjr's plot.) Although images of those frequencies would land there, that isn't what is happening. At the time of the digital interpolation, the harmonic distortion hasn't happened yet, so there's nothing at 20/30/40 kHz to create such images. The non-linear distortion (harmonic and intermodulation) only occurs in the analogue circuitry. Link to comment
Popular Post mansr Posted August 8, 2019 Popular Post Share Posted August 8, 2019 13 minutes ago, John_Atkinson said: And I certainly don't have a problem with that. I'm sure you don't. However, MQA does. 9 minutes ago, John_Atkinson said: I don't see why a filter having minimum-phase behavior is in itself a "distortion." It is a form of (linear) distortion, just as a non-flat frequency response is. See https://troll-audio.com/articles/linear-and-minimum-phase/ for some examples of what this distortion looks like. esldude and MikeyFresh 1 1 Link to comment
Popular Post firedog Posted August 8, 2019 Popular Post Share Posted August 8, 2019 14 minutes ago, John_Atkinson said: Yes, a minimum-phase reconstruction filter offers better time-domain performance, for the reasons I outlined in my 2011 Richard Heyser Memorial lecture to the AES. Google will be your friend here. But why do we need a proprietary file format to apply an MP filter? We can apply a minimum phase filter to standard flac files without MQA and it’s infrastructure. So again, please tell me why it is needed. On the other hand, at least you never referred to it as a scientific revolution on the level of Copernicus’ work....😉 JSeymour, esldude, MikeyFresh and 1 other 2 1 1 Main listening (small home office): Main setup: Surge protector +>Isol-8 Mini sub Axis Power Strip/Isolation>QuietPC Low Noise Server>Roon (Audiolense DRC)>Stack Audio Link II>Kii Control>Kii Three (on their own electric circuit) >GIK Room Treatments. Secondary Path: Server with Audiolense RC>RPi4 or analog>Cayin iDAC6 MKII (tube mode) (XLR)>Kii Three . Bedroom: SBTouch to Cambridge Soundworks Desktop Setup. Living Room/Kitchen: Ropieee (RPi3b+ with touchscreen) + Schiit Modi3E to a pair of Morel Hogtalare. All absolute statements about audio are false Link to comment
Popular Post mansr Posted August 8, 2019 Popular Post Share Posted August 8, 2019 1 minute ago, firedog said: On the other hand, at least you never referred to it as a scientific revolution on the level of Copernicus’ work....😉 No, he likened it to the creation by the lord almighty. MikeyFresh and esldude 1 1 Link to comment
manisandher Posted August 8, 2019 Share Posted August 8, 2019 4 minutes ago, mansr said: Although images of those frequencies would land there, that isn't what is happening. At the time of the digital interpolation, the harmonic distortion hasn't happened yet, so there's nothing at 20/30/40 kHz to create such images. The non-linear distortion (harmonic and intermodulation) only occurs in the analogue circuitry. Yep, I see what you mean. Still, it seems strange that the intermodulation distortion in the analogue circuitry sits exactly at 56 kHz, 66 kHz and 76 kHz. Main: SOtM sMS-200 -> Okto dac8PRO -> 6x Neurochrome 286 mono amps -> Tune Audio Anima horns + 2x Rotel RB-1590 amps -> 4 subs Home Office: SOtM sMS-200 -> MOTU UltraLite-mk5 -> 6x Neurochrome 286 mono amps -> Impulse H2 speakers Vinyl: Technics SP10 / London (Decca) Reference -> Trafomatic Luna -> RME ADI-2 Pro Link to comment
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now