Testing out your headphones / speakers

I’ve written several times in the past about the importance of a nice set of headphones. And it’s still true after all these years! I actually have two sets:

  • HD 580
  • HD 600

The Sennheiser HD600 hardly need to be introduced. They are probably the most famous headphones released in the past 20 years. Following the Sennheiser HD560 Ovation, the German brand released the HD580 in 1993, which has formed the basis for all open-back models coming out after, including HD580 Jubilee (1995), HD600 (1997), HD650 (2003), HD6XX (2016) and HD660S (2017).

I recently replaced the headband foam and ear foam on both of these, which had degraded through time and use. Luckily these are common headphone models so it’s easy to find replacements from Amazon.

As of late 2017, I upgraded my office speakers to one of the the wirecutter recommended models, the Q 3020

Back in April I also upgraded my positively ancient 2003 era Logitech Z680 to the Z906. I did not use any of the speakers other than the subwoofer (which is tied to the “brain” panel), so I gave those away.

The Z906 is powering the array of four Q 3020 speakers plus a Q 3090c center speaker, so a classic 5.1 speaker arrangement.

But… how to test these speakers and headphones?

2 Likes

Having fun with some of the demo songs linked in the Wirecutter article:

So What – Miles Davis, that’s a given, right? I think it’s required by law to demo this track when reviewing speakers.

Sinnerman – which has always been an amazing track and pure joy to hear at any stage in your life unless you are, like completely dead inside. Go listen to it now!

Carnival – I didn’t know this as well, have heard maybe a few times, but I am a reasonable fan of Natalie Merchant. I support this choice.

Get Lucky – feels a bit too … recent and big-hit-y to be an ideal demo track, but whatevs.

Lost Cause – I absolutely ain’t gonna argue about. One of my favorite Beck tracks of all time and really quite underrated.

Giorgio by Moroder – Yet another daft punk track? Come on.

I went through a ton of “reference tracks reviewers use to test speakers” links and a lot of them boil down to personal preference. There’s probably zero track overlap in any of the links I found! But I did like this list…

… becaust it covers the philosophy of testing, rather than getting into opinion:

  1. Pristine, well recorded tracks. Simply put - if this doesn’t sound good, nothing will. These tracks have been recorded to such a high standard that, excuse the blasphemy, playing them through some Apple Ear pods would verge on acceptable, so pass it through some decent Hi-Fi and they’ll knock your socks off.

  2. Tracks with wide dynamic range and excellent sound stages. Listen out for spatial separation between individual instruments and vocals and feeling like your living room just turned into the Royal Albert Hall and you’re on the right track.

  3. Tracks with The Big One – Bass. Irritating if there’s too much, crushingly disappointing if there’s too little.

  4. Dense tracks. Songs that pack a lot in to each track and without capable audio equipment can sound, quite frankly, a mess. So when you get them right, that’s a whole heap of audio satisfaction (and a few smug engineers).

TL;DR crank some Bohemian Rhapsody. Otherwise try to select some tracks you prefer that fit in those above buckets?

My votes would look like

  1. Anything Steely Dan (like Hey 19), or similar. Studio wonk bands.
  2. Sinnerman - Nina Simone
  3. Edgar Winter - Frankenstein
  4. Pretty Vacant - Sex Pistols / Motley Crue - Kickstart My Heart
2 Likes

A couple audio reviewers recommended Axel F, and one I think recommended the Miami Vice theme.

An an example of how eclectic and personal and opinion-based these “tracks to review speakers by professional audiophiles” can be:

Oh yeah, Lincoln Mayorga, that guy is a musical TITAN! Everyone listens to his stuff to validate speaker quality. :roll_eyes:

Direct-to-disc recording was an interesting bit of audiophilia, though:

Direct-to-disc (or as some refer to them, “direct disc”) records were very popular with audiophiles in the early ‘80s, and they were considered to be the finest quality vine-yule. After the musicians were ready, they would cut an entire side of an album, without stopping. The signal from the control room wouldn’t go to tape (other than a tape backup), but it would go directly to the cutting lathe, and the master lacquer was cut in real time. The record stampers were then made from the master. Because of this there could be no do-overs and no editing or overdubs, and on occasion you can hear a clinker, a bumped microphone, rustling sheet music, or shuffling feet. But the fidelity of a properly executed direct-to-disc record is the apex of vinyl reproduction.

Mostly, I think audiophiles and wine reviewers are similarly full of :poop:. And their fields are … fertile. :rofl:

That being said, I was citing the CD mastering loudness wars recently and I ran across this data point:

One of the biggest albums of 2013 was Daft Punk’s Random Access Memories, with many reviews commenting on the album’s great sound. Mixing engineer Mick Guzauski deliberately chose to use less compression on the project, commenting “We never tried to make it loud and I think it sounds better for it.” In January 2014 the album won five Grammy Awards, including Best Engineered Album (Non-Classical).

1 Like

I checked out the link to the Q 3020 speakers and it seems everyone is sold out (a couple still listed on eBay. A visit to Q Acoutistics and nearly all their bookshelf speakers are Not in Stock… unless you want either their $600 Concept 20 pair or their Concept 300 pair for $3,500! I was just looking to replace an ancient pair of speakers that had been mounted on a wall in a spare bedroom turned “computer/SW radio room”. I’ll look again over the next few days.

I’ll be in the market for a new set of headphones as well. A certain bride thought my good set was her late mother’s and gave them away on me. :scream: I’m sure whomever she gave them to are enjoying them immensely! :unamused:

Maybe I’ll just splurge and get the Logitech Z906 surround sound speaker system. But that would be a little overkill for my laptop. :slightly_smiling_face:

2 Likes

One thing I’ve gotten into is external headphone amplifiers. I do think if you’re going to invest $200+ in headphones, it’s worth investing a bit in the amplifier, too. There are two types:

  1. Standard headphone amps for listening only, traditional audiophile style.
  2. “Gamer” headphone amps, because they have a connection for the microphone attached to the headset, and thus two-way communication.

After extensive research, I have some recommendations. For the gamer amp, I’d say start with this great, inexpensive all-rounder at $43, the Syba Sonic:

If you want something a bit fancier, the Schiit Fulla 3 at $120

As with anything audiophile, the sky is the limit beyond this… but I feel these are reasonable starting points. You can make a credible case that fancier headphones do require additional power to drive, and keeping the driver outside the PC, which is full of all kinds of ambient electrical noise, is a smart move. I can’t prove it, but here’s one little experiment:

As a counterpoint,

  • the difference between 1999 / 2010 era bargain bin, generic audio chips and today’s chips is … large. The built in motherboard audio has improved dramatically over the last 20 years. You could argue this matters less now than it used to.
  • Some nicer, more expensive motherboards do pay more attention to the sound circuitry, and can offer near-amp performance. So if you opt for a fancy motherboard, you might be able to skip the amp.

Per Wikipedia, the variables here are:

  • number of channels (mono, 2, 2.1, 4, 4.1, 5.1, 7.1, etc)
  • sample rate (44.1 kHz, 48.0 kHz, 88.2 kHz, 96.0 kHz)
  • bit depth (16, 24, 32)

So it’s a question of how often you are sampling the sound, and at what bit depth.

image

I did a lot of reading and

  • there’s a lot of skepticism about sample rate because most audio you listen to probably wasn’t originally captured at a super high sample rate, therefore listening to it at a different sample rate involves conversion and that can lead to potential conversion artifacts creeping in.

  • there is no skepticism whatsoever about bit depth, the more the better. Zero represents absolute silence, and the loudest sound, is the highest number possible allowed by the bit depth:

    • 16-bit → 65,536 values
    • 24-bit → 16,777,216 values
    • 32-bit → 4,294,967,296 values

Let’s take a look at our modest Syba:

For playback we get “studio quality” with up to 24 bit depth and 96 kHz sample rate. But to avoid potential conversion artifacts, most of the sites I read recommended sticking to the standard 44100 Hz or 48000 Hz sample rates, at the highest available bit rate, and 24-bit is … a lot.

For recording, this is clearly not the right gear for laying down a studio album… but since we’re only recording our voices on gaming headsets, I’d think 16-bit depth at 48000 Hz sample rate should be more than enough?

(It is kinda fun to set the recording levels really low, record yourself speaking, and then listen to it.)

Anyway, TL;DR – you want the greatest bit depth you can get away with, sample rate probably doesn’t matter too much beyond typical 44100 Hz CD quality, and, above all,

:mega: you can’t magically improve audio beyond the bit depth and sample rate it was originally recorded at!

If you want to get deep into audiophilia, look for those extra fancy recordings first :wink: … Super Audio CD is basically dead, but it was “comparable to a PCM format that has a bit depth of 24 bits and a sampling frequency of 88200 Hz”.

2 Likes

And as far as streaming services goes, for really high definition audio (beyond the expected CD quality), looks like your options are

Amazon Music HD

Tidal HiFi

HiFi audio is a superior sound but is still limited in its resolution—44.1 kHz /16 bit. TIDAL has partnered with MQA to deliver something substantially better: an authenticated and unbroken version (typically 96 kHz / 24 bit) with the highest possible resolution—as flawless as it sounded in the mastering suite. And exactly as the artist intended it to sound.

These both require native apps for playback, either on phone or desktop. And notice the precipitous dropoff in number of tracks: 70 million → 5 million!

1 Like

My motherboard (Gigabyte Z390 Aorus Pro WiFi) has some pretty hardcore audio settings as it turns out…

ALC1220-VB Immersive Gaming ViBes – All New Realtek High-End HD Audio Codec

ALC1220 120dB(A) SNR HD Audio with Smart Headphone Amp automatically detects impedance of your head-worn audio device, preventing issues such as low volume and distortion. With the new VB series audio controller, stream your voice to the world vibrantly with both front/rear microphone SNR up to 110/114dB(A).

Specs on the chip are hard to find, but here’s what I could dig up

Playback

up to 32bit, 192000 Hz :warning:

Recording

up to 24-bit, 192000 Hz :warning:

1 Like

OK, so I signed up for the Amazon Music HD just to give it a try

What audio quality does Amazon Music HD support?

Amazon Music HD offers lossless audio in two quality ranges: HD and Ultra HD.

HD tracks are 16-bit audio, with a minimum sample rate of 44.1 kHz (16/44.1 is also referred to as CD-quality), and an average bitrate of 850 kbps. Ultra HD tracks have a bit depth of 24 bits, with sample rates ranging from 44.1 kHz up to 192 kHz, and an average bitrate of 3730 kbps.

In comparison, most standard streaming services currently offer Standard Definition (SD) with a bitrate up to 320 kbps. These audio files use lossy compression, where details of the original audio are removed in order to reduce the file size. By contrast, Amazon Music HD preserves the original recording information to deliver the highest quality sound available, more than 2x the bitrate in HD and more than 10x the bitrate at the highest Ultra HD bitrate. Amazon Music HD will always play the highest quality content available, based on network, device capability and your selected settings.

Confirmed the recording quality matches the playback capabilities, by clicking that little yellow ULTRA HD indicator:

24-bit, 48 kHz seems to be the basic HD Audio bar, I’ll click around and see if anything else was recorded at crazypants quality levels.

:warning: All of this DOES require a native app to work, browser only is no-go; the above screenshots are from Amazon Music.exe

1 Like

Still investigating, but it seems like Lossless format is table stakes, so FLAC. The various badges have this meaning:

  • HD → 16-bit / 44.1 kHz

  • Ultra HD :yellow_square:24-bit / 44.1 kHz++

The main distinction between HD and Ultra seems to be bit depth. Not seeing a lot of sample rate variance, but I will keep looking! As you’d expect, the older recordings are at best just FLAC tracks ripped from the original CDs. So I’m concentrating on newer albums / newer artists recorded on the latest equipment to see if anything is really out there, quality wise…

Aha! Here’s a comparison of various HD streaming services

As far as mobile devices

Almost all iPhones and iPads released since 2014 running iOS 11 or later support HD and Ultra HD, limited to 24bit/48kHz playback. Anyone who wants to listen to higher sample rates needs to invest in a DAC that can support those rates.

I can find plenty of 24-bit content on Amazon HD but I can’t find much at a higher sample rate. I’ll keep digging. I did find this spreadsheet of high sample rate content on Tidal, which I put in this shared google spreadsheet:

Aha, duh, if you check the section in the Amazon Music executable titled “Ultra HD Exclusives” …

… your odds go up, better than any other section I’ve tried:

  • Bon Jovi / Bon Jovi has 24-bit, 96 kHz
  • Diana & Marvin has 24-bit, 192 kHz
  • Ariana Grande / Yours Truly has 24-bit, 96 kHz
  • Janet Jackson / Control has 24-bit, 96 kHz

Some others:

  • 24/96: Arcade Fire “Everything Now”, Attaca Quartet "Attaca Quartet, Jack Johnson “Brushfire Fairytales”, HAIM Women in Music Pt. III, Aerosmith “Back in the Saddle”, Al Greene “Greatest Hits”, Ben Harper “Welcome to the Cruel World”.

  • 24/192: Jackson Browne “Running on Empty”, Art Blakey & the Jazz Messengers “Just Coolin’”, Cat Stevens “Greatest Hits”, Creedence Clear Water Revival “Chronicle: 20 Greatest Hits”, Eagles “The Very Best of the Eagles (2013 Remix)”, Fleetwood Mac “Fleetwood Mac”, James Taylor “Greatest Hits”.

It is kinda fun to listen to Salt & Pepa sing Shoop at 24 bit, 192 kHz :loud_sound::rofl:

1 Like

Since a few people have asked, this is just me experimenting, not an actual argument for this kind of precision in audio. A classic essay on this, well worth reading, in the Internet Archive by the Ogg Vorbis author at

24/192 Music Downloads …and why they make no sense

TL;DR

  • There is no point to distributing music in 24-bit/192kHz format. Its playback fidelity is slightly inferior to 16/44.1 or 16/48, and it takes up 6 times the space.

  • for increased sample rate you may end up reproducing ultrasonics and other weird artifacts; it’s basically excess precision that’s more likely to harm than help.

  • for 24+ bit depth, while there are reasons to use it in recording and production, the effective in-practice dynamic range for 16-bit is 120dB which is plenty, more than 15 times the 96dB claim. So while it may not be harming the sound, it’s basically a waste of bandwidth and space.

And of course the science:

This paper presented listeners with a choice between high-rate DVD-A/SACD content, chosen by high-definition audio advocates to show off high-def’s superiority, and that same content resampled on the spot down to 16-bit / 44.1kHz Compact Disc rate. The listeners were challenged to identify any difference whatsoever between the two using an ABX methodology. BAS conducted the test using high-end professional equipment in noise-isolated studio listening environments with both amateur and trained professional listeners.

In 554 trials, listeners chose correctly 49.8% of the time. In other words, they were guessing. Not one listener throughout the entire test was able to identify which was 16/44.1 and which was high rate [15], and the 16-bit signal wasn’t even dithered!

I did run my own experiment with MP3 bitrate and found similar results to the above. See if you can hear the difference between these five samples:

I got 4/6, I only consider picking “128kbps mp3” a failure :x: … and I blame the whispers on Tom’s Diner for throwing me off on one of those!!

What does actually work to improve sound quality, per the article?

  • Better headphones :raised_hands: :headphones:
  • Lossless formats (mainly to combat potentially badly encoded files, and to avoid generational loss in quality)
  • Better base masters for future generations, so I believe recording masters today at 24-bit and 48 kHz makes sense.
1 Like

The above is wired audio. Wireless audio quality is its own whole subject as well, since you have the new variable of transmitting the wireless audio data from the computer / device to the wireless headphones…

Bluetooth versions by year:

5.2 Dec 2019
5.1 Jan 2019
5 2016
4.2 2014
4.1 2013
4 2010
3 + HS 2009
2.1 2007
2 2004
1.2 2003

So it’s incredibly likely you have at least a Bluetooth 4 device, and probably closer to Bluetooth 5.

5.2 does have one nice and highly relevant feature, LE Audio

  • LE Audio: Announced in January 2020 at CES by the Bluetooth SIG, LE Audio will run on the Bluetooth Low Energy radio lowering battery consumption, and allow the protocol to carry sound and add features such as one set of headphones connecting to multiple audio sources or multiple headphones connecting to one source It uses a new LC3 codec. BLE Audio will also add support for hearing aids.
  • Enhanced Attribute Protocol (EATT), an improved version of the Attribute Protocol (ATT)
  • LE Power Control
  • LE Isochronous Channels

Here’s kind of a fun test, if you happen to be on a bluetooth audio device right now:

http://www.brentbutterworth.com/bluetooth-blind-test.html

I checked some recent bluetooth purchases and they can be surprisingly coy about what version of bluetooth they offer beyond “5”… Interestingly, on a windows machine here’s how you tell what version of Bluetooth you have

LMP 6 Bluetooth Core Specification 4.0
LMP 7 Bluetooth Core Specification 4.1
LMP 8 Bluetooth Core Specification 4.2
LMP 9 Bluetooth Core Specification 5.0
LMP 10 Bluetooth Core Specification 5.1
LMP 11 Bluetooth Core Specification 5.2

So my mobo has Bluetooth 5.0.

Here’s how to check on Android, and how to check on iOS. Both my iPhone Pro 12 and the Xiaomi Mi 9 support Bluetooth 5.0.

1 Like