• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Multi-Channel, Multi-Amplifier Audio System Using Software Crossover and Multichannel-DAC

Thank you for the info on Fq dependent smoothing in Acourate; it is interesting for me knowing the difference from psychoacoustic smoothing* in REW which I cited in my post #708; I wrote there, "nowadays, I always do not like psychoacoustic smoothing* ...."

*In REW user manual: "Psychoacoustic smoothing uses 1/3 octave below 100Hz, 1/6 octave above 1 kHz and varies from 1/3 octave to 1/6 octave between 100 Hz and 1 kHz. It also applies more weighting to peaks by using a cubic mean (cube root of the average of the cubed values) to produce a plot that more closely corresponds to the perceived frequency response."
 
Last edited:
Another comparative listening, on-the-fly vinyl LP vs. Remastered CD: analog piano solo recorded in 1967

Hello friends,

In my posts #688 and #697, I shared my successful revival installation of DENON DP-57L + DL-301II (MC) in my DSP multichannel multi-driver multi-amplifier stereo audio setup for on-the-fly real-time listening/playback of LPs.

And in my post #688, I touched on comparative listening of on-the-fly LP vs. remastered CD for HiFi music tracks, pipe organ music album, originally digitally recorded and first released in LP format then later-on released in CD format. I found that the sound quality of original LP is really wonderful; almost no difference in sound quality between on-the-fly LP sound and the ripped remastered CD in my present setup, thanks to the still amazingly excellent performance of DP-57L + DL-301II together with AUDIO-TECHNICA AT-PEQ30 and TASCAM US-1x2HR.

Yesterday, I carefully and completely washed and further cleaned-up (see post #695) another LP (I played it last time about 25 years ago!); this LP was analog recorded in London in July 1967, and released in 1970 in Japan (and also abroad, I believe), "Mozart Recital" by (young!) Vladimir Ashkenazy.
WS00005189.JPG


The vinyl "material" of this LP looks very nice and durable; after my intensive wash and real-trace clean-up (83 rpm for three times!), no pop nor scratch noises can be heard which is really impressive.

Ashkenazy's "Mozart Recital" was analog recorded in 1964, and I can easily hear the rather warm-taste sound characteristics which was the typical one of LONDON label in that era. Listening to the remastered CD, on the other hand, I can easily notice the intention of the remastering sound engineer making the sound of each track slightly more clear/transparent for CD re-release.

It is very interesting for me that if I gained-up the midrange driver channels (L&R Yamaha 8 cm Be-dome JA-0801) in about 2.6 dB from my present standard gain, i.e. when I dial-up from -17.0 dB to -14.4 dB by Accuphase E-460 directly dedicatedly driving the L&R midrange drivers, the on-the-fly LP sound became almost identical to the remastered CD sound I heard under standard -17.0 dB gain position of E-460.


BTW, I have almost no further intention of washing/cleaning any other my old LPs bringing them to my on-the-fly LP listening by the revived DP-57L + DL-301II in the DSP multichannel multi-amplifier setup.

I would like purchasing, on the other hand, a few of recently released HiFi 45 rpm 30 cm LPs in jazz and classical genre, if available. I would highly appreciate your possible recommendation and/or suggestion on such HiFi 45 rpm 30 cm LPs which would well fit as my standard real-time-listening reference LPs in my DSP multichannel active audio setup.
 
Last edited:
For educational purposes I like to share some basics about IIR and linear phase crossovers and the associated step responses.

For the image below I calculated a 3-way LR crossover (IIR). The phase of the green signal has to be inverted as this is the only way the addition of the tree parts will give a linear amplitude response (blue).
Let's now take a look at the step responses below the amplitude chart. The step response of the woofer is positive and rises slowly. The green signal is inverted (starts going down) and the brown signal shows the positive step response for the tweeter.
Adding these tree responses you get the blue step response of the overall LR 3-way crossover. As you saw the amplitude response is flat after the addition but the step response is NO step at all! The reason for that is that IIR LR crossovers shift the phase so it adds up perfectly in the frequency domain but NOT in time domain.

1671800243877.png


For comparison I calculated linear phase crossovers. This also 2. order LR but those cannot be realized with IIR filters. You need FIR for that.
In the amplitude section you see the same as before, a perfect addition (blue). Here this can be achieved without inverting the green channel!

Let's now take a look at the step responses.
As you can see they are symmetrical in linear phase crossovers and if you add them you get a PERFECT step.
THIS IS HOW IT SHOULDS BE! Linear amplitude and perfect step response! This also results in a constant group delay over the whole frequency range.

A perfect step response can never be achieved with standard IIR filters (with one exception see below). This means shifting speakers or adding time delay will never result in a perfect step response if you use standard IIR crossovers. With IIR this by theory is impossible.
1671800793120.png


Now we come to the only exception that can create a perfect step response with IIR and this is the subtractive delayed approach.
So what did I do? I started with the low-pass part of a IIR LR crossover at 150Hz (red signal). Next I delayed the input signal
by 80 samples and from this I substracted the low-pass. As result I got the green signal. As we can see there is also some kind of symmetry in time domain.

Caused by the time delayed subtraction you get a perfect linear frequency response and also a perfect step response (blue) if you add the red and the green signal.
This means with one IIR low-pass, a delay and a subtraction you can create a crossover that is perfect in frequency and time domain.
More regarding the subtractive delayed approach can be read in the paper I attached here for a second time. With subtractive delayed approaches it is important to have a look at the polar response which was done by John Kreskovsky in the attached paper.
1671801629173.png


In any case getting a perfect step response is a design goal you might have and that is absolutely correct from a theoretical point of view. At the end it will come to the question if crossovers with a perfect step response sound better or not. In other words: Is a distorted group delay, produced by IIR LR crossovers audible?

You can find a lot of research papers regarding this topic with results that often say "it depends...". I'll try to sum it up: If you are an experienced listener you hear the difference especially with wideband signals having steep slopes. Percussion is for example one thing, where I can clearly distinguish between a perfect step response or a distorted one.

Best DrCWO
 

Attachments

  • 1671800767593.png
    1671800767593.png
    50.3 KB · Views: 93
  • A New High Slope, Linear Phase Crossover Using the Subtractive Delayed Approach.pdf
    247.5 KB · Views: 1,364
Hello again @DrCWO

I thank you indeed again for your above in-depth educational post. Reading the above, I feel I could almost fully understand your theoretical approach and accomplishment. Your above kind post would be very nice reference for many ASR friends periodically visiting here.

I still believe that any of the theoretical achievements/implementations should be validated/proved by "independent well validated objective measurement(s)", hopefully on room air sound at our listening position.

I also believe that in each step of our audio setup-changes (aiming towards improvement), we should also subjectively check very carefully the difference (worse, no change, better) trusting our own ears and brain, as Keith of @Puite Audio kindly wrote here;
"You must hear equipment in your own room in your own system, compare unsighted (close your eyes), if there isn’t an immediately apparent difference/improvement. To go further if there isn’t a significant improvement then don’t change anything, the largest gains are speakers and room. Keith".

Since I have been carefully objectively measuring-observing-tuning the Fq-domain and time-domain wave-shape matching of actual air sound recorded by measurement microphone (as summarized in my post #520 and #687), I feel and assume I have been accomplishing almost unconsciously what you are talking about and what you are targeting to.

I feel, at least for my ears and brain (not a perfect sound recognition system!), and at least in my room acoustic environment (not the "perfect audio room") with my "limited" audio gears (not the "perfect" audio gears), I have already done almost all the possible improvements objectively measured and subjectively determined/confirmed.

Consequently, I am now gradually going back to the enjoyment of listening to my favorite "music" rather than listening to the "air sound". (You would please visit my remote thread here, only if you would be interested in early classical music...)


You wrote; "Percussion is for example one thing, where I can clearly distinguish between a perfect step response or a distorted one."

Yes, I fully agree, I too can.

I always include many of such "sharp transient sound" music tracks in my "audio sampler music playlist" (#669, #670), and also I use "Sony Super Audio Check CD" having super-nice transient sound check tracks, throughout this project thread.

And, on each step of my progress in this project, I (we) very carefully listened to all of the "sampler playlist" (and some tracks of the "Super audio Check CD"), for improvement or not. Especially some jazz tracks in my "sampler playlist" has been always the critical choice for assessment of improved "transient" behavior of the "total sound" at my listening position in my listening room environment (please refer to #687).

In case if you would be seriously interested in having all the raw music tracks of my "audio sampler playlist", as well as "Sony Super Audio Check CD", please simply PM me.
 
Last edited:
Hello friends,

It is a little bit of my surprise (and great honor) knowing that as of yesterday views/visits on this thread reached 170,000 times since I started on April 9 2020.
WS00005228.JPG


Your continuing visits/participation and kind encouragements will be highly appreciated.
 
Hello again, Igor! Thank you indeed for your visit here again on this thread after some while.

Your latest(?) system diagram at your home (I believe it is not in your recording/mixing studio) is really interesting again for me, especially the utilization Q-SYS Core 110f. I well remember you kindly shared the photo of your entire wonderful listening room in your post here on this thread.

I can well understand the active DSP configuration for front L&R NS-1000x and the 5x KEF Q900 center array.

Let me ask you what would be the DSP configuration for the rear L&R NS-1000x in non-XO-ed full range drive (??) by 2x HYPEX amp? And, what would be the XO Fq and slope between SVS PC2000 subwoofer and rear pair of NS-1000x?

In any way, I myself still have been sticking to pure 2.0 L&R stereo (not audio visual) music-only enjoyments as you may find throughout this thread; even though I occasionally use my Panasonic TH-55HZ1800 55-inch OLED 4K TV as PC monitor (by HDMI connection to audio dedicated PC) for visual-listening as I wrote here.
 
Last edited:
I try for a new thread

Oh, you have edited you post on 7:08 PM Japan Time after my response on 3:09 PM Japan Time to it (in original simple one)!

I well know your recording/mixing (I believe so based on your old post here on July 12 2020) for ECM on that Arvo Pärt music "The Deer's Cry" is just stunning and wonderful.
Edit:
Yes, I could confirm; I found your name on discogs.com site for "Arvo Pärt: The Deer's Cry"
WS00005232.JPG


I understand you will soon start your new thread here in ASR Forum, right?
If this would be the case, I am (we are) very much looking forward to your new thread and my (our) possible participation on it.

Only if you would be interested, please visit my remote thread; "Lute Music and Other Early Music: Stunning Recordings We Love".
 
Last edited:
Oh, you have edited you post on 7:08 PM Japan Time after my response on 3:09 PM Japan Time to it (in original simple one)!

I well know your recording/mixing (I believe so based on your old post here on July 12 2020) for ECM on that Arvo Pärt music "The Deer's Cry" is just stunning and wonderful.
Edit:
Yes, I could confirm; I found your name on discogs.com site for "Arvo Pärt: The Deer's Cry"
View attachment 258740

I understand you will soon start your new thread here in ASR Forum, right?
If this would be the case, I am (we are) very much looking forward to your new thread and my (our) possible participation on it.

Only if you would be interested, please visit my remote thread; "Lute Music and Other Early Music: Stunning Recordings We Love".
 
Hello audio friends,
i wonder if there's a smart way to compare DAC's. I want to compare a Motu Ultralite-mk5 and an Okto DAC8PRO. A comparison by earphones left me without noticeable difference. A comparison via speakers is so hard to judge because it takes 1-2 minutes to change my setup (cabling etc). What's the best 'material' to listen to?, which details should i focus on?
 
A comparison by earphones left me without noticeable difference.
Thanks because in all likelihood there isn't.
 
Hello audio friends,
i wonder if there's a smart way to compare DAC's. I want to compare a Motu Ultralite-mk5 and an Okto DAC8PRO. A comparison by earphones left me without noticeable difference. A comparison via speakers is so hard to judge because it takes 1-2 minutes to change my setup (cabling etc). What's the best 'material' to listen to?, which details should i focus on?

Your inquiry is somewhat off topic and beyond the perspectives of this project thread, I assume.

Just for your reference, we have huge and intensive thread entitled "Serious Question: How can DAC's have a SOUND SIGNATURE if they measure as transparent? Are that many confused?".

I myself have several DACs; KORG DS-DAC-10 (4 of it!), OPPO Sonica DAC, ONKYO DS-DAC-1000, OKTO DAC8PRO and some others. In my DSP-based multichannel multi-amplifier project on this thread, I use 8-Ch OKTO DAC8PRO all the way, and I have been much satisfied its 8-Ch capabilities and superb sound quality as very well evaluated and discussed many places in ASR Forum including the dedicated thread on DAC8PRO.
 
Spectacular thread @dualazmak

I don't saw what processor you're using in the PC audio server. I'm thinking if a fanless mini PC with a Celeron 5105 would work with stereo DSP and ADC interface ... or you need to go to i3 processors (as an example).

Maybe you tested something like that ... The Pi 4 isn't enough, lots of sound skips really.
 
I don't saw what processor you're using in the PC audio server.

Thank you for your visit!

- Completely silent audio (audio-visual) dedicated Windows PCs in this project: #225
Also, you would please find my latest system setup here including the two audio dedicated PCs and all ASIO signal path.

Also please find here (on this thread) and here (remote thread) the Hyperlink Index for this project thread.
 
Last edited:
Another comparative listening; on-the-fly remastered vinyl LP vs. remastered CD: Bill Evans jazz piano trio, analog recorded in 1977; remastered (2021) vinyl LP (45-RPM 188-gram) vs. its CD release

Hello friends,

I am happy sharing again my recent audio listening enjoyment here in this post.

This post is a follow-up of my post here (#688) and here (#722) regarding comparative listening of on-the-fly vinyl LP versus CD release using my DSP-based multichannel multi-driver multi-amplifier audio system having DENON DP-57L TT plus DENON DL301MarkII MC Cartridge connected to AUDIO-TECHNICA AT-PEQ30 phono preamplifier into TASCAM US-1x2HR audio interface (ADC); please refer to my post here (#688) for the details of the revival of vinyl LP TT in my audio setup.

Several weeks ago, my “that” serious jazz fanatic friend (please refer to post #438) kindly came again to my home, and we fully enjoyed our 6-hour audio listening session. This time, as he has promised before, he kindly brought a 2-LP album of 45-RPM on heavy and thick vinyl of 180-gram each (so indicated on the jacket). Using my present on-the-fly LP listening setup (#688 and #697), we listened to only the first two tracks of the LP and I was very much impressed by the sound and music performance. After he left my audio room, therefore, I quickly ordered the 2-LP album together with its exact CD release at Amazon Japan site and they arrived on my desk a week ago.

Even though I am not a serious jazz enthusiast, I know that the 2-LP album is very famous among the jazz lovers. Bill Evans (piano), together with Eddie Gomez (bass) and Eliot Zigmund (drums), recorded (fully analog tapes) the album "You Must Believe In Spring" in August 1977 at Capitol Studios, Hollywood, CA, U.S.A., and the original LP album was release in 1981 which was the first memorial/mourning release after Bill passed away in September 1980. I heard many people recognize this recording as one of the Bill's Schwanengesang performances.

Here I dare not say much about the self-destructive (long-year semi-suicidal) life of genius jazz musician Bill Evans.

In 2021, the 40th anniversary year of its first LP release, Mr. Paul Blakemore of Concord Mastering very carefully re-mastered the album from the original analog tapes, and Mr. Kevin Gray of Cohearent Audio took the remastering cut pressed onto two 45-RPM180-gram heavy-thick vinyl at RTI. The 2-LP 45-RPM album (made in U.S.A.) was accordingly released under the label of CRAFT RECORDINGS in 2021.

2-LP: CR00455 CRAFT RECORDINGS, Made in U.S.A.
WS00005314.JPG


WS00005315.JPG


Very fortunately, the 2-LP of heavy-thick vinyl arriving on my desk “looked” perfect with its nice jacket and inner envelopes, very clean with no bending at all. I actually measured the weight of the LPs, and found it is really 188-gram each!

Before putting the LPs onto my audio rig, I carefully and rigorously washed them with aqueous (distilled water) diluted neutral pH kitchen detergent and a hard plastic brash; after complete drying-up, I further cleaned the LPs by actual 86-RPM trace with 3.5-gram stylus pressure using NUMARK PT01 portable TT for five times (see #695). Of course, each time I cleaned-up the PT01's cartridge stylus with a small fine brush and pure isopropyl alcohol. These are my “routine” cleaning procedures for old LPs as well as newly purchased LPs.


Together with the LPs, the exact CD release was arrived on my desk containing three bonus tracks (the bonus tracks were separately remixed by Mr. Seth Presant).

CD: UCCO-5612 CRAFT RECORDINGS, Made in Japan:
WS00005316.JPG

Using dBpoweramp CD Ripper (release 17.3 64-bit), I carefully ripped the CD into non-compressed 16-bit 44.1 kHz AIFF format, and the CD album was incorporated in JRiver’s SSD music library.
WS00005317.JPG

(Please refer here for my SSD digital music library organization.)

Then, we (myself and my wife) went into intensive comparative listening sessions for the LPs vs. the CD with our audio rig in exactly the common/same configuration parameters (i.e. common Fq response settings).
WS00005588.JPG


WS00005125 (2).JPG


Having the first on-the-fly listening to the LP, we were deeply impressed by its “physical quality” and “sound quality”.

We hear no audible pop nor scratch noise all the way through from the silent lead portion of the LP, music tracks, the silent interval between the tracks, and end silent portion; we could only hear very subtle scratch/pop noises in final loop of the silent end portion in each of the two LPs which is quite common in many other LPs and no problem at all for our listening enjoyments.

Thanks to the very well manufactured (and well QC-ed) 188-gram thick-heavy vinyl with no bending, we hear no audible wow-flatter nor off-center (eccentric) sound fluctuation. Of course, these are also greatly attributed to the still-excellent performances of DP-57L TT and DL301MarkII MCC as well as to my “routine” intensive LP cleaning procedures.

Furthermore, the merit of the “45-RPM sound quality” pressed on high-quality vinyl material is just amazing; it would be rather difficult describing it properly by words since English is not my mother language. I would like to suggest you, therefore, just trying this 45-RPM LP if you have excellent HiFi LP listening audio setup. It is very much interesting and impressive looking at the track-7 of only-6-minute music actually occupying the whole surface of the B-side of the second LP; what an extravagant/luxury cutting and LP press!

The total sound quality of wonderful music performances is just perfect and stunning felt as one of my best on-the-fly LP listening experiences; I would dare not say too much about my subjective impression, but you would please imagine the heart-ease reproduction of the slightly nostalgic 1977 analog recording in really wonderful total sound balance, low distortion, high S/N, of course no clipping, suitable and comfortable 3D sound stage of the stunning jazz piano trio performance. We could hear and much appreciate the excellent and successful re-mastering and re-mix engineering.

You may easily understand my preference and fascination to this album when you would notice that I have all the CDs of Karel Boehlee Trio; please refer to my post here on this thread.


On the other hand, what was our subjective impression on the remastered CD listening?

Again, the total sound quality is really very nice; nice reproduction of slightly nostalgic 1977 analog recording. We (myself and my wife), however, could easily hear/distinguish the CD sound, in comparison with LPs, having a little bit of edge and gain enhancements in piano sound which maybe, I assume, intentionally remixed adjusting to contemporary low-end to middle-end HiFi audio gears and/or headset/earphone listening (using smartphones?)

Of course, our such hearing “observation” would greatly dependent on the difference of LP playback sequence and ripped SSD CD digital library sequence even all the audio configuration including the Fq-response remained unchanged for both of the LP and the CD listening settings.

It was interesting for us finding that the CD sound became “almost” identical to the on-the-fly LP sound by reducing the midrange Beryllium driver (500 Hz to 6 kHz) in 1.6 dB (i.e. change the standard -17.0 dB down to -18.6 dB with ACCUPAHSE E-460’s digital gain control) and reducing the tweeter + super-tweeter (6 kHz to 21 kHz) in 2.0 dB (i.e. change the standard -17.0 dB down to -19.0 dB with SONY TA-A1ES’s digital gain control).

Even we could/can apply such fine relative gain tuning for the CD sound, however, at least with our audio rig and room acoustic environments, we always do prefer the LP sound over the CD sound for very much impressive heart-ease listening to the wonderful music by Bill Evans Trio throughout our intensive comparative listening sessions.

By the way, we two really do not like the three bonus tracks in the CD since these three tracks have very different music style and recording quality in quite “heterogeneous” with the original release’s seven tracks. Even we listened to the three bonus tracks only one time, I have quickly deleted them from my JRiver digital music library.

In summary, I highly recommend the 45-RPM 188-gram heavy-thick 2-LP album, CR00455 CRAFT RECORDINGS, Bill Evans Trio: “You Must Believe In Spring” if you have proper HiFi LP on-the-fly listening audio setup.

At least in my audio rig, this gem 45-RPM 188-gram 2-LP album should serve as one of (or the best?) golden reference(s) for on-the-fly real-time LP listening enjoyments.
 
Last edited:
Back
Top Bottom