How Deep Must the Bass Be?

[Bugal1998]

Certainly. I am more interested in averages than peaks, as it offers me a more general outlook as to a given track's makeup.

View attachment 325315

Same scale. Average included for comparison. Getting rid of the smoothing would require more substantial modifications.

Thanks for posting the comparison. And, yes, can definitely see the value of the averaged view. No need to remove smoothing.

With a +5db in-room bass curve that 50hz content is 30db above the 1khz level; even the 24hz content is 20db above.

Now time to go listen to the tracks you posted!

 

[Bugal1998]

You can do that in REW by using the RTA tool and import a WAV file.

I've done that in the past with Audacity, but now I pretty much only stream. Not sure if there's a way to access downloaded Tidal files for analysis in either program...

 

[Bugal1998]

Big panels usually have fast bass (but also sometimes not much bass).

I wonder if speed is how some listeners perceive and articulate frequency response. Anytime a salesman has talked about fast bass (Grado headphones, small bookshelf speakers, etc.) usually what I heard was not much bass.

I have no experience with panel speakers so I can't relate to the specific integration challenges or their perceptual impacts.

Edit: And thinking back, the one time I heard panel speakers the talk was all about their speed. Whatever it was, I enjoyed hearing them.

 

[olieb]

I've done that in the past with Audacity, but now I pretty much only stream. Not sure if there's a way to access downloaded Tidal files for analysis in either program...

The script of @Weeb Labs takes a WAV too. So in any case you need to record a stream first.
In Audacity I did not find an option for a peak level spectrogram.

 

[Bugal1998]

The script of @Weeb Labs takes a WAV too. So in any case you need to record a stream first.
In Audacity I did not find an option for a peak level spectrogram.

Ah. I do remember selecting specific time segments to analyze in Audacity, perhaps because it was always an averaged view. 'Twas years ago...

 

[dualazmak]

Ah. I do remember selecting specific time segments to analyze in Audacity, perhaps because it was always an averaged view. 'Twas years ago...

I still occasionally use the latest version of Audacity, but I mainly use Adobe Audition 3.0.1 (essentially free) as described many times in my past posts. Adobe Audition 3.0.1 has nice "Frequency spectrum analysis window" where we can get FFT spectrum of the pin-point cursor position or averaged FFT spectrum (Scan selection) of any selected region with various/flexible FFT parameter/window-type settings; I always prefer the "Blackmann-Harris" FFT window-type.

 

[rdenney]

Great post with good citations and low-freq information from orchestral repertoire to recorded electronics - Thank you...

Just curious - as the tuba world is rather small - did you ever know or converse with Roger Bobo (principal tuba - L.A. Philharmonic Orch.) who passed earlier this year?...

Post in thread 'Celebrity RIP Thread'
https://audiosciencereview.com/forum/index.php?threads/celebrity-rip-thread.30898/post-1522536

I didn’t know him beyond the occasional small talk at conventions, and he certainly didn’t know me. But I am friends with many of his students, chief among them (in my ears) Gene Pokorny of the Chicago Symphony (though Gene gives more credit for his training to noted A-list studio tuba player Tommy Johnson, who I also did not know).

I’ve always lived in Texas and Virginia and was never part of the tuba crowd in California.

But I certainly know his work.

Rick “whose first tuba solo LP was all Bobo” Denney

 

[computer-audiophile]

I also used to have to work in windowless lab rooms for a short hours, but luckily not often. What I see here is an absolutely professional studio environment, the room seems claustrophobic and repulsive to me. Not necessarily an attractive place to work, imo.

I like to listen to music in a comfortable living environment and only in stereo anyway. Of course, I also know interesting multi-channel performances of contemporary music e.g., but I prefer to listen to them in appropriately equipped concert halls, such as the Kubus at the ZKM Karlsruhe. There is no feeling of confinement there either. I don't need that at home.

Infrastructure | ZKM

The Hertz-Lab offers its guest artists a wide range of technical features for their artistic works.

zkm.de

 

[thecheapseats]

Post in thread 'Celebrity RIP Thread'
https://audiosciencereview.com/forum/index.php?threads/celebrity-rip-thread.30898/post-1522536

I didn’t know him beyond the occasional small talk at conventions, and he certainly didn’t know me. But I am friends with many of his students, chief among them (in my ears) Gene Pokorny of the Chicago Symphony (though Gene gives more credit for his training to noted A-list studio tuba player Tommy Johnson, who I also did not know).

I’ve always lived in Texas and Virginia and was never part of the tuba crowd in California.

But I certainly know his work.

Rick “whose first tuba solo LP was all Bobo” Denney

ok _ I don't watch the Celebrity RIP Thread up here... I met him on a panel we were on at USC's music school in the early eighties - what a great guy and we had a common background as we were both born in L.A.... I was much younger and he gave me some valuable 'working' advice although his focus was concertizing while mine was studio work... he also encouraged me to get conducting skills with a teaching referral... He had monster chops (as we say)... great guy...

 

[gnarly]

That example is a bit of a thought experiment. Once the motor stops the cone itself has non linearities…

I do not think anyone knows DEFINITIVELY what accurately captures the difference between fast and slow, but one interesting discussion is the phase interaction at the crossover:

Hi, been meaning to get back...a few 2C...

My problem with using fast and slow as descriptions for bass is simply how utterly nebulus they are......
Are we talking subjective sound, are we talking something physical like motor/cone velocity, air velocity,.... or what?
Most often I see folks talking subjective sound....

Personally, I can't help but equate the idea of fast with anything other than transient response.
And the definition of transient response is frequency response (complex, mag and phase).
The wider the frequency response bandwidth, the better the transient response.
The flatter the complex freq response, the better the transient response.

The fastest sound there is? ..... full-range flat mag and phase.
So what slows it down?

For subs regardless of room...
Limited bandwidth...for example, the sub with the earliest acoustic high frequency roll-off will sound "slower".
(Ironically, subs with higher THD are often described as sounding faster, when played at higher SPL. A good front loaded horn sub, with nearly zero THD even at higher SPL, can be so clean it sounds fat and slow in comparison.)
Phase lag....the sub with the steepest acoustical high-pass, from either porting or EQ's boosting low end on sealed, can sound "slower"..( via the misnomer known as group delay)

For speakers...
non flat freq response
wonky crossovers
multiple drivers and their lobing
disjointed directivity
........anything that diminishes flat complex freq response (or rather the particular house curve in play)
........anything that negatively impacts polars

And then there's the friggin room.....!
Room modes..
Multiple sources / reflections....comb filters galore.
Yikes...

So for me, back to Audio 101 and simplicity.
Perfect transient response = perfect freq response = mo' faster



Ok, to help pay for that loadwinded sting on the soapbox, here's something maybe folks will enjoy.
I've had the idea for a while, that tone bursts can be used to help with unconventional measurements.
I've used them for fine tuning sub xovers, due to their high degree of repeatability in measuring low frequency time-of-flights.

@GXAlan.... for the motor stop idea you mentioned....I've tried a single cycle sine burst, to see how well the acoustic waveform matches the electrical.
Here's the REW Generator and the scoped electrical sound card output.

Here's a comparison of acoustic captures of that burst, with a mic about 4" from sub's dust cap
.
The left is with a few small minimum phase EQ's to flatten mag and phase.
The EQ's made no difference whatsoever to the single freq burst.
So, the left could be said to be the raw response also.

The right has the same set of EQ's, but implemented in a linear phase FIR filter.

The idea was to see what kind of oscillations occur around the one cycle burst. To see how damped the sub is.
I was surprised to see the overall differences, min phase (or raw) vs the lin phase.
If it weren't for an earlier onset of negative oscillation on the lin phase, I'd call it hands down superior to the raw / min phase

In both screens the center of the bursts are on the center vertical crosshair. The two outer vertical cursors are one 50 cycle, 20ms period apart.
Voltage cursors the same...

 

[dualazmak]

Hello again @gnarly,

Thank you for sharing the details of your intensive sine tone burst experiments.

You maybe aware of that I did the same experiments and observations for my sub-woofers and woofers on my project thread.

Spoiler

- Measurement of transient characteristics of Yamaha 30 cm woofer JA-3058 in sealed cabinet and Yamaha active sub-woofer YST-SW1000: #497, #503, #507

I would highly appreciate if you could share the 3D (Gain-Fq-Time) color spectrum of the recorded air sound (hopefully by using Adobe Audition) so that you/we can semi-objectively visually observe 3D energy distribution of those low Fq transient air sound.

 

[gnarly]

Hi dualazmak, many thx.

Sorry, I rarely peer down into measurements that deeply. (I'm still a keep with the big picture type speaker tuner.)
Besides, the bursts I put up were taken at 4 inches from the dust cap. I can't see how they could relate to energy distribution out in the room.

 

[MickeyBoy]

20Hz to 20kHz is more than you need. That’s what we can hear not what is found in music.

While you can find organ music that goes down below 20 Hz, I cannot say that I routinely listen to music containing pipe organs. I do listen to piano.

Then, thinking about pianos, how many pieces use the very lowest or very highest note of the piano? It’s actually quite rare to use A0 but two mainstream pieces I know of are

• The Bells of Notre Dame from the Disney movie The Hunchback of Notre Dame.
• The Heart Asks Pleasure First from the movie The Piano.

But these are movie soundtracks, so you get to take advantage of LFE channel delivering that lowest note

Chopin’s Fantasie Impromptu goes down to 35 Hz. So in my mind, getting down to 35-40 Hz is full range for my music and compromises can be made because the question that then arises is if giving up some bass extension results in a better speaker elsewhere?

On the other hand, bass for movies is unlimited in appetite. This doesn’t mean that you need high SPLs just low extension.

Gran Turismo, the movie, is mediocre story telling but a sonic extravaganza. I saw it in “4DX” 6 axis motion which was great as well. I imagine tons of bass helps with the visceral element.

In Top Gun: Maverick, when Darkstar fires its afterburners, you need a lot of extension to replicate the feeling you get at an airshow.

THE REAL QUESTION
Once Dirac ART and Trinnov Waveforming are in play, don’t we want our main speakers to go as deep as the lowest room node so we have the ability to actively manipulate the room sound?

In that case, since positioning a subwoofer in every region of the room is harder than going with slightly larger full range speakers, the advice may be very different.

I imagine that bass freqs would sound very different with Dirac ART. Would be more than curious to hear from someone who has used it.

 

[gnarly]

I well understand your points.

I would be still very much interested in seeing 3D transient energy distribution of your low Fq tone bursts both given by minimum phase and linear phase, measured/recorded at very near (4 inches) to the SP unit as you described, only if possible!

Judging from your air sound spectrum data, I assume very little difference in 3D transient energy distribution between the two though; it would be interesting also to confirm by 3D (Gain-Fq-Time) color representations.

BTW, if you would be seriously interested in installation of Adobe Audition 3.0.1 (essentially free) on your Windows 11 (or Windows 10) PC (if you have one), I would be more than happy sharing the installation files and instructions, as I recently wrote here #452.

You are very kind with your willingness to offer folks the audio files and assistance with software you use. (I still owe you some feedback on some previous generosity as I remember.)

Must admit I can't see how any valid 3D energy distribution can be gleaned from a single 1/4" mic capsule over a one cycle burst, so close to the sub.
Am I missing something?

 

[pablolie]

Part 1 is even better, classical music bassheads (shake your heads) :-D!

PS: A double bass can go down to 31Hz, when tuned so - but some say it can get lower. Part 1 is in the same album, and I think it hits more prolonged lower notes with the double bass anchor on the right.

 

[Chrispy]

I wonder if speed is how some listeners perceive and articulate frequency response. Anytime a salesman has talked about fast bass (Grado headphones, small bookshelf speakers, etc.) usually what I heard was not much bass.

LOL best description of "speed" I've seen yet. I just figure it's more like "fleeting".

 

[dualazmak]

PS: A double bass can go down to 31Hz, when tuned so - but some say it can get lower.

Thank you for your sharing the YouTube clip. I quickly analyzed the audio track of "Part 2" by Adobe Audition 3.0.1 after extracting audio into non-compressed AIFF format using JRiver MC 31. Yes, the meaningful music tones go down to around 15 Hz, and we can even see hall reverberation(?) around 5 Hz. It is really interesting hearing/feeling only the R&L subwoofer channels, heavy and large Yamaha YST-SW1000 (ref. here) covering below 50 Hz, by pushing the two solo buttons in DSP EKIO's output subwoofer channels.

I also analyzed it by MusicScope 2.1.0:

We can find that the YouTube clip was uploaded with sharp low-pass (high-cut) filter at 15.7 kHz.