Saturday 18 December 2021

Upsampling: Native DAC Playback, and SoX PCM/DSD upsampling of 1kHz signal. (And the "Beatles: Get Back" documentary, "As the artist intended"?)

Notice one of the waveforms NOS.

A few weeks ago on this blog in a comment discussion, Bennet / Dtmer Hk talked about showing what it looks like when upsampling a 1kHz 16/44.1 signal in PCM side-by-side with DSD upsampling.

Sure, no problem! We can have a quick peek at the 1kHz sine tone from a couple of DACs for comparisons between direct playback with built-in filtering, upsampled in high quality with PCM SoX, and then using high quality DSD conversion with SoX-DSD.

Plus, let's also have a look at the recent Beatles: Get Back documentary series and some thoughts which I think relate not just to the "music lover" which I hope is in all of us, but also to the "hardware audiophile" side of this hobby.

First, let's show the 1kHz sine upsampled in various ways as noted above. We start by defining the parameters I'll use in SoX 14.4.2. Starting with a clean 16/44.1, -0.2dBFS (a little overhead so as not to clip), 1.00227kHz (reduce rounding off at 44.1kHz) sine wave:

Converted To PCM 24/384 with:
     sox input.wav -b 24 output.wav rate -v 384000 dither -S

These parameters were as recommended by Bennet and look like excellent choices if you want to do high quality upsampling. This will create a 24/384 (we could aim for 352.8kHz as well but honestly it makes no practical difference) output file with "very good" filtering. Also with high-pass dithering.

Converted also to DSD64/128/256 - using Måns Rullgård's SoX-DSD mods:
      sox-dsd input.wav output.dsf rate -v 2822400 sdm -f sdm-8
       
<DSD rate = 2822400 for DSD64, 5644800 for DSD128, etc.>

These are the same parameters as discussed a few weeks back when testing DSD performance with the various DACs.

Using the TEAC UD-501 and Topping D10 Balanced DACs, captured with E1DA Cosmos ADC (mono mode), here are a few side-by-side high resolution FFTs - captured using FlexASIO at 32/384 (192 bandwidth), 512k-points:


=========================

As you can see, the TEAC graph includes the NOS mode where we can see all the ultrasonic imaging artifacts when we run the DAC filterless, top right is the standard direct playback with the typical "sharp" filter I use the vast majority of the time. The more modern Topping D10 Balanced doesn't have a NOS mode but can handle DSD256, pushing the ultrasonic noise shaping further out. As shown back in August, the D10 Balanced has by default a steep minimum phase filter for direct playback.

If we do some pixel peeping of the SoX PCM upsampled FFTs, we can make out the steeper and deeper post-Nyquist attenuation compared to the DAC's direct playback thanks to SoX's "very" high quality sharp linear filter. Of course, DACs do have noise floor limits so there's only so much attenuation possible. Even if the low-pass filter applied is capable of mathematically achieving 32-bits with better than -190dB attenuation, that's just a numbers game (that sometimes gets tossed around as if meaningful) and in real life we're not going to see that magnitude in actual playback from the analogue outputs.

Note that distortion measurements like the THD / THD+N do not and really should not change much** with various PCM and DSD processing/upsampling. Feeding a DAC with 24/384 data compared to 16/44.1 over USB could account for some of the change and likewise DSD vs. PCM will also lead to variation. Do not assume that big numbers like 384kHz upsampling or DSD256 will automatically lead to higher fidelity - as you can see, this is not necessarily the case. This is also why having test data for your specific DAC is essential.

[** Interestingly, the biggest difference I see is with the Topping D10B going from 16/44.1 to 24/384 PCM. Notice the increase in THD. I wonder if we're seeing the same phenomenon as the Topping D90SE at and above 192kHz suggesting that there are potential optimizations that can be tweaked with the ESS THD compensation parameters at higher samplerates.]

Like with the measurements, I don't hear a difference in the sound between various filters for the most part although there are exceptions. For example, the slow roll-off filter with the Ayre/PonoPlayer a few years ago is an example of one that did change the sound noticeable. Sure, we could do fancy stuff to allow even more precise filtering - Chord's WTAHQPlayer upsample, even try out PGGB if you want. You can also play with minimum or intermediate phase settings (as discussed previously).

What's important IMO is that we keep the effect of upsampling in context and realize that this is just low-pass filtering, not more-complex DSP like FIR room correction. As per this post from Måns, you don't need "megatap/megaorder" filters to achieve more than adequate very low passband ripple and large stopband rejection. As he showed, ~400-taps is more than enough, providing 180dB stopband attenuation for any amplifier, and certainly for any human ears. Make sure to let that sink in so that we can be better informed consumers.

Again, big numbers might seem impressive at first and perhaps can spur sales based on specs, but in reality, this is just not needed and eventually people get wise to the marketing and become unhappy if pushed beyond the realm of believability. Looking at the SoX PCM upsampling debug log, it's "only" using a 573-taps FIR with 186dB stopband attenuation for the 44.1 upsampling to 32/384kHz internally. This can be increased to 645 taps with 210dB attenuation if we use -u instead of -v for "ultra" quality.

As usual, for those who advocate for really really complex upsampling algorithms and enormous tap lengths, it would be nice to see evidence that this makes a difference with actual DAC output rather than supposed "subjective" claims of improvement with no apparent basis in physical reality. Even if one insists that ears are the best tools to appreciate differences because one doesn't believe that measurements can capture supposedly audible changes, let's get some controlled listening (like blind testing) to demonstrate/confirm the claims!

Be mindful of these ideas when you're on audiophile forums and people are talking about using stuff like powerful Intel i9, Xeon, AMD Ryzen 9, Threadripper CPUs, and leveraging the power of GPUs to perform all their upsampling whether to PCM 768+kHz or DSD512+ as if this too is some kind of impressive feat. I think we've all seen examples of multi-thousand-dollar computers being used for this purpose with all kinds of claims that it improves sound (like this product). IMO, this is all highly unlikely unless one is also applying room-correction or other type of DSP that can definitively change the sound.

Audiophiles spend a lot of money (and time) on things which I think are unnecessary and suspicious, with claims of subjective improvements. Hey, if this is part of one's "pursuit of happiness", sure, go have fun! Just make sure in principle not to entertain too much snake-oil or be too giddy about fantasies, because at the very least, that would be bad for the reputation of the audiophile hobby as a reality-based technical endeavor.

--------------------

To end, recently I finished watching the 3-part Beatles: Get Back documentary I mentioned a few weeks back on Disney+. I think one needs to be a bit of a music/Beatles history fan to get through all 7+ hours; my wife managed just over an hour. It's more of a "meditative" work to show the process and hectic leadup to the performance of the January 30, 1969 Rooftop Concert at 3 Savile Row and much of the content that would come out later on the Let It Be album in 1970. It's still a glimpse based on what was filmed/recorded and I'm sure there was a ton of interesting stuff happening in the background off-camera!

Nice seeing George Martin and Glyn Johns at work. Great capture of the moment that the tune "Get Back" came to McCartney in Episode 1. It's also interesting watching the interplay of technology and the music production process throughout as they complained about the acoustic quality of Twickenham Film Studios, had issues with noise on their original Apple studio recording setup, needed to borrow the EMI REDD console and George Harrison's 8-track recorder, commented about hum over the PA system, etc. (much of this in Episode 2).

There were a cute couple of minutes on Episode 3 around 57 minutes where they were fooling around with the "Stylophone mini pocket organ" (you can still buy something like this, although I would recommend the Gen X-1 for more features). Especially fun watching Billy Preston playing with it! For the modern audience, it is amazing watching that intersection between artistic curiosity and technology even though the synthesizer "toy" sounded so remarkably primitive. Note that the Beatles would later incorporate the Moog synthesizer on Abbey Road recordings later that year in the summer of 1969 which would be the last time they recorded as a band even though the Rooftop Concert was their last live performance together.

Starr & Preston playing with the Stylophone. Lennon taking a sip.

Yoko Ono looked bored (stoned?) much of the time, but otherwise seemed harmless at this point which is a couple of months prior to getting married to John in March. While there were disagreements, the interpersonal conflicts all seemed quite pedestrian; at least what was shown on film and in the audio clips. Nothing like the words used among cantankerous audiophiles in forum wars ;-).

I couldn't help but feel sad that Lennon seemed a bit distanced and presumably not quite at his creative peak by this time (heroin use an issue by 1968).

Nice footage of the Bobbies responding to the noise complaints during the rooftop performance.

There were some cool instruments used (also see here). Check out this article with discussion of the studio gear. No fancy speakers or "high end" amplifiers as far as I can tell in the control room. Maybe I missed it, but I didn't see any audiophile cables used in the production and nobody seemed all that worried about it. I'm not sure if modern audiophiles are actually hearing the performance "as the artist intended" as a result. ;-)

Here's a still shot from Episode 3 around 2:08:40, near the end:



Anyone know what kind of speakers these are in the control room? Do they sound any good?

Speaking of "as the artist intended", I think Let It Be is a good example of why this phrase is often meaningless. First of all, I suspect that our modern day equipment would have significantly higher fidelity than the old studio gear. The actual released album in 1970 was the studio work of the infamous Phil Spector. Which speakers, amps, playback gear did Spector use for finalizing the stereo mix? Would he be the person we should ask to "authenticate" the sound (if still alive)? Furthermore, by the time Let It Be was released in May 1970, the Beatles had already broken up. One wonders if every member of the band even heard the final mix before it got into the record stores. And even if they did, I'm sure they didn't all sit in a room together with the same playback equipment to "authenticate" that this was the sound they intended (much less envisioned), right?!

The bottom line is that, IMO, the only meaningful target we can aim for as a landmark for high quality playback of music is that of achieving high fidelity which means low distortion and "transparency" to the source material whether it's what's on tape, LP, or the digital data - this can be objectively assessed. Everything else is of questionable reliability, based on subjective impressions (subject to the limitations of human hearing and cognition), or simply unverifiable.

There's no need to have any pretensions about whether it "sounds live" or captured the "actual event" or even what we think the artist(s) heard in the studio or afterwards. All that stuff is for the artists and audio engineers to decide anyhow way before it shows up on a CD, LP, or digital stream! All we can do as consumers and audiophiles is judge whether our equipment/room fairly reproduced the recordings and whether we like what the artist and production folks gave us. This again does not mean we can't have our own idiosyncratic preferences or consciously choose devices which can add "euphonic" embellishments (ie. distortions, "colorations").

Here's the Internet Archive's Complete Rooftop Concert 1969 if you want a FLAC copy - far from the best audio quality but an interesting piece of history.

Also, make sure to check out Let It Be... Naked (2003 release, DR10) if you have not heard the non-Spector mix which many (including myself) consider to be superior. I can appreciate the different track list, inclusion of "Don't Let Me Down", and the more live sounding arrangements after watching the documentary.

--------------------

I enjoyed watching the new Steven Spielberg West Side Story the other night at the theater. Nice retelling of the classic! The West Side Story soundtrack (2021, DR13) sounds excellent and I really appreciate the high dynamic range.

Happy to see Neil Young & Crazy Horse's Barn is DR12. Although the music isn't "hi-res" and the noise floor can be a little high, well done on maintaining the dynamic range. At least the distortions sound more intentional ("Human Race") than a result of nasty clipping due to the mastering. The quieter tracks like "Song Of The Seasons" and "Tumblin' Thru The Years" sound clear. Let's hope that more modern albums continue to keep up the double-digit DR at least and get us back to more natural sounding recordings again instead of the limp, distorted, low DR records that have been the norm for way too long.

As usual, dear audiophiles, I hope you're all enjoying the music over this holiday season!

17 comments:

  1. Hi Arch, the speakers should be Altec 612 with leak TL/25 Amp below.
    Found some pictures on the internet
    How do they sound? I don't know
    Thank you for the excellent work as usual

    ReplyDelete
    Replies
    1. Thanks chancedon!

      Awesome to know what kind of gear to keep an eye out for in the future to have a listen to... Maybe even opportunity to measure if able!

      Delete
  2. You could certainly argue "Let it Be" wasn't as the artist intended. McCartney didn't have approval of the mix/album and was engraged with what Spector did with some of his material - especially "The Long and Winding Road". That's the reason for "Let it Be...Naked".
    I guess it depends on how you define "the artist".

    ReplyDelete
    Replies
    1. Yup, good point Unknown.

      There are the stories of the music itself, and then the stories of the whole production process before the final mix and product.

      I do hope that claims around "as the artist intended" blows off over time as if this has any actual meaning or even that this somehow enhances the authenticity of what we buy.

      If an artist intends that there be some kind of "authenticity" process, then let's just have a description of the system, room, volume level, etc. to describe the day the artist finally listened to "Album X" and gave a thumbs up. Be done with it.

      Otherwise, I honestly don't think the consumer cares to any meaningful degree as to have companies make money off this kind of claim!

      Delete
  3. I'd say there is at least some merit to HQPlayer, since some of its filters are apodising (you can usually tell those by the fact that they cut off around 21.02khz rather than the usual 21.97), and work well on badly mastered files.

    ReplyDelete
    Replies
    1. Sure Techie. Maybe that's beneficial.

      I wonder if anyone has formally tested this? Could be a great little research project!

      Delete
  4. I wonder how many people aged 40+ would be able to discern a difference between a 21.97kHz and 21.02kHz filters?

    ReplyDelete
    Replies
    1. I'm guessing 100% of mainstream audio reviewers in non-blind tests. (Happy holidays everyone)

      Delete
    2. +1 to Phil.

      40+ is very generous. Let's talk 10+. :-)

      Happy holidays to you and yours Ralph and Phil!

      Delete
  5. Did you see the interview with Taiko's CEO https://youtu.be/ggby6Vhi_9E ? (The second part of the interview is released today).

    ReplyDelete
    Replies
    1. "Thanks" for the link Geert.

      Painful series of videos. Thankfully YouTube allows 1.5x playback speed ;-).

      I honestly think these guys either:

      A. Hallucinates what they believe they hear. Or are delusional. Either way they're psychotic and have lost touch with reality and insight.

      or...

      B. Shilling. Snake-oil salesmen. Are in it to make money or achieve some ego-based attention.

      I would bet it's Option B more likely than A, ya never know!

      Delete
  6. Thanks for the tests. So PCM1795's relatively short (compared to PCM1792A/PCM1794A) PCM filter does not seem to cause any penalty. Harmonic distortion and noise shaping dominate the ultrasonic frequency region with both on-chip filter and SoX resampling. On the other hand, DSD128 still has considerably more ultrasonic noise than PCM.

    ES9038Q2M on the Topping shows an obvious dip at around 22kHz followed by a sharp rise of noise when using DSD64. For DSD128 and 256, while the rise of noise is pushed to higher frequencies, the dip at around 22kHz is much less obvious, this phenomenon is similar to ES9038Pro: higher <20kHz noise floor when using higher DSD rates.

    ReplyDelete
    Replies
    1. Yup Bennet. A little bit of this noise or that distortion...

      But nothing I'd be too worried about with either the segmented TI/BB DAC or multibit-SDM ESS. I remain unconvinced that the underlying DAC technology sounds different if the objective results look similar with low level variation...

      Happy Holidays man!

      Delete
  7. I tried to watch that Beatles docu, but I guess I'm not a big enough Beatles fan so I couldn't bare longer than 40 minutes. Kinda annoying thou with the horrible picture quality, first way to much noise reduction and then some bad looking ai-upscaling on top of that. They tried to make it look modern, and I guess they succeeded since it looked like a bad smartphone camera..

    ReplyDelete
    Replies
    1. Good point Tell about the picture quality with clear post-processing done. I think also pumped up the saturation for the colors.

      Yeah, I definitely agree with the sentiment about how much you could take. As I mentioned, my wife managed about 1 hour. We at least had some popcorn to start and enjoyed some 18-year aged, single malt whiskey at the end of the hour ;-).

      A fun piece of history for the fans. And targeted to more of the "hardcore" demographic.

      Delete
    2. The original was intended for 1969 UK TV, and so was filmed in 16mm.
      For the 1969 theatrical release they turned the 16mm into 32, so it looked like S***, as it was just optically pumped up.
      This new digital version looks way better than the theatrical release 50 years ago.
      So it's all relative. Can't really compare video/movie tech from the 60's to today.

      Delete
  8. Could you test some Asus motherboard with headphone amp? Saw your reviews for Gigabyte (they suck from personal experience and what I saw from other people online) and MSI, but no Asus.

    ReplyDelete