Thanks, your explanation is consistent with my experience/experiments and my everyday "coherent minimalist" M.O. (msg #51).
Yes the sub amp must have continuous-knob phase adjustment for aligning with woofer.
Last edited:
Ouch, even at subwoofer XO frequency? Original's 290hz presumably.
My two different subs, Jamo 8" pair and unknown-make big plate-amp triad, go up to ~200hz probably LP 2nd-order. Years ago when I replaced the satisfactory phase-matched Jamo subs with Monitor Audio AlMg RST drivers over closed tubes with passive 1st-order aircore litz inductors (to support Mo's Studio 2 rear port plugged), I did think it was a significant improvement -- poor man's Studio 60. But I didn't analyze why and didn't know how to tweak phase, just played around with parts until the sound clicked.
Last edited:
I'm not sure a litz inductor is going to do much at low frequencies, but a slow first order crossover should be easy, and helpful since the wider range of multiple sources can have room benefits.
They usually have a tunable allpass circuit. And as such they can only help you getting the phase at the crossover frequency right. So it is handy for achieving a flat-amplitude response but not for transient accuracy.
If transient-perfect speakers really were that easy in the analog domain we would only see transient-perfect speakers for ages. 😉
Regards
Charles
If transient-perfect speakers really were that easy in the analog domain we would only see transient-perfect speakers for ages. 😉
Regards
Charles
Yep. I figured he meant all-pass. How many amps have you seen with them? Only ones I've ever seen are prosound DSP amps..
Processors with all-pass? Sure, plenty.
I guess we have to define what we mean by transient accuracy.
If we mean flat mag and phase across the spectrum, which also equals correct time alignment, then I guess all-pass can't help transient accuracy since all-pass can only add phase rotation, not reduce it.
If we mean as smooth as possible mag and phase, with as minimal as possible phase rotation, then i'd say alll-pass can help achieve transient accuracy.
Just depends on how 'transient perfect' we're talking, huh? If we really mean perfect, I say lin phase FIR to the rescue ! (for any speaker 3-4ways and up)
(Aside: I've spent all week setting up a 4-way syn fully IIR. What a pain in the butt compared to lin phase FIR....ridiculously so!!!)
Processors with all-pass? Sure, plenty.
I guess we have to define what we mean by transient accuracy.
If we mean flat mag and phase across the spectrum, which also equals correct time alignment, then I guess all-pass can't help transient accuracy since all-pass can only add phase rotation, not reduce it.
If we mean as smooth as possible mag and phase, with as minimal as possible phase rotation, then i'd say alll-pass can help achieve transient accuracy.
Just depends on how 'transient perfect' we're talking, huh? If we really mean perfect, I say lin phase FIR to the rescue ! (for any speaker 3-4ways and up)
(Aside: I've spent all week setting up a 4-way syn fully IIR. What a pain in the butt compared to lin phase FIR....ridiculously so!!!)
Hi Mark
Interesting! I expect its for your Syn 10´s? Are you going to share about that, in your SYN 10-thread maybe? What did you do (Chris´s zero´s order approach?)? and how did it go? Is it worth it some how?
Although my project is on hold it would interest me what difference there are IIR vs FIR, as my first option is IIR. Fir needs further investments.
Regards
Steffen
Interesting! I expect its for your Syn 10´s? Are you going to share about that, in your SYN 10-thread maybe? What did you do (Chris´s zero´s order approach?)? and how did it go? Is it worth it some how?
Although my project is on hold it would interest me what difference there are IIR vs FIR, as my first option is IIR. Fir needs further investments.
Regards
Steffen
Hi there,
looking at the cost no object equipment the thread opener is using like the Accuphase DF 45 with an original price point of 14.000 Euro's my advice would be to try out what an automated measurement and room correction system would do with the sophisticated speaker system like just as an example the miniDSP Flex 8 DL (the model doesn't matter - but you need the Dirac Live full range signal correction for each of the 4 ways actively driven - the miniDSP is the cheapest way to get it done, but with some degradation of the sound quality due the 48 kHz / 24 bit limitation)
i was trying to program a DSP on my own when i was young in the mid / end 80's and i can imagine that the amount of man hours invested in the Dirac measurement and correction algorhythm was huge
to achieve similar results by hand you can spend the rest of your life measuring and measuring again
just my humble opinion - Stefano
looking at the cost no object equipment the thread opener is using like the Accuphase DF 45 with an original price point of 14.000 Euro's my advice would be to try out what an automated measurement and room correction system would do with the sophisticated speaker system like just as an example the miniDSP Flex 8 DL (the model doesn't matter - but you need the Dirac Live full range signal correction for each of the 4 ways actively driven - the miniDSP is the cheapest way to get it done, but with some degradation of the sound quality due the 48 kHz / 24 bit limitation)
i was trying to program a DSP on my own when i was young in the mid / end 80's and i can imagine that the amount of man hours invested in the Dirac measurement and correction algorhythm was huge
to achieve similar results by hand you can spend the rest of your life measuring and measuring again
just my humble opinion - Stefano
Last edited:
How about gated?
Nothing wrong with what I see, but gated is the best I think. It will be smoother, but possibly may look a little different. It will be good to about 300hz most likely, so no need to show anything below that.
Nothing wrong with what I see, but gated is the best I think. It will be smoother, but possibly may look a little different. It will be good to about 300hz most likely, so no need to show anything below that.
I only prefer sub (plate) amps which have a continuously variable phase adjustment. That weeds out the cheaper models that only have a 180 deg switch option. The SpeakerPower SP1-700 plate amp is an excellent sounding unit that doesn't draw any attention to itself when correctly set up. Thats not easy to achieve with class D amps that have some amount of correction built into them. The hypex amps are hit and miss for me. They don't have as subtle of an EQ section that tends to be a bit more "detached" from the main speakers at lower crossover points. I'm not really sure why that is, but I can say hypex has a little more inherent latency built into their processing. All of the SpeakerPower plate amps I've so far used can be run past 200 hz without having audible integration issues.
reasons for gate imo...
..we have crappy measurement conditions,
..we think we can ignore what are measurements say below a certain frequency
..we need to feel good about our measurements
Don't mean to be an a-hole...just reality, as i see it...
Measurement conditions are not ideal. The speaker is very large, and I believe it was already said in the thread that they were not going to be moved much.
We can ignore the response below a certain frequency! To look at the cross at 2k, and at 7k, we really don't need to look at anything below 400hz.
I'm not sure how gating makes someone feel good about measurements. Gating will remove the effect of room reflections, and will be in my opinion, more relevant.
We can ignore the response below a certain frequency! To look at the cross at 2k, and at 7k, we really don't need to look at anything below 400hz.
I'm not sure how gating makes someone feel good about measurements. Gating will remove the effect of room reflections, and will be in my opinion, more relevant.
Fascinating. High XO very much out-of-phase at 7.5khz (dip) but steep-order mooted it by low-8khz. Wonder how this affected IR and transcients. Would like to know how well (or poor) the time-delay-aligned HF (original mid driver) and VHF (new super-tweeter) actually phase-aligned away from XO overlap.
Thanks for the discussion. Even after semi-monopolizing the Fullrange Photo Gallery with multi-way coherent minimalist point-source-wannabe experiments, I still find it puzzling and mysterious -- the cause-effect relationships among time-alignment, phase-alignment, transcient-perfect response, venue-airyness, pinpoint-imaging, and holographic-depth. Perhaps SOTA DSP with rigorous protocol could test, disentangle, and demystify.
I'm convinced time-misalignment reduced perceived depth (and I'm not alone). I can have it all, and pretty quickly, by aligning acoustic centers first, then phase.
It's been truly enjoyable to whip up a coherent speaker or two, that placed musicians quite specifically and beyond the wall, and then to hear and discover new things in old music.
Thanks for the discussion. Even after semi-monopolizing the Fullrange Photo Gallery with multi-way coherent minimalist point-source-wannabe experiments, I still find it puzzling and mysterious -- the cause-effect relationships among time-alignment, phase-alignment, transcient-perfect response, venue-airyness, pinpoint-imaging, and holographic-depth. Perhaps SOTA DSP with rigorous protocol could test, disentangle, and demystify.
I'm convinced time-misalignment reduced perceived depth (and I'm not alone). I can have it all, and pretty quickly, by aligning acoustic centers first, then phase.
It's been truly enjoyable to whip up a coherent speaker or two, that placed musicians quite specifically and beyond the wall, and then to hear and discover new things in old music.
Last edited:
Not similar.. better than Dirac. Obviously when you fix acoustic problems, Dirac isn't going to be necessary but more to the point acoustic problems require acoustic fixes.
It doesn't only take time, but knowledge. However it doesn't take as long as that, and any improvements count so there's no reason anyone should feel it's beyond them.
Haha, I believe transcient near-perfect is (and was, >60 years ago) quite easy, just much less important for selling rock/disco/AV "HIFI" than the more-in-your-face aspects of sound "quality". A couple of unusual examples/counter-examples:
My once-SOTA all-AlMg 1st-order but not offset-time-aligned Mo Studio 2 with phase-adjusted Jamo subs (flipped to face front) properly upstream-optimized (including custom PC/OS, Chord Hugo DAC, all Au24 OCC wiring etc.) rendered -- as if -- pianist's hand-and-fingers positioned to strike each note (uniquely) in a true performance not routine play. Ultra-fi and pinpoint imaging. Yet, side-by-side easily bested in perceived distance, or soundstage depth, by my 1/10-cost-thirty-years-later Mark Audio 2"-AlMg "holographic" fullrange driver (Maeven TLonken). What does this imply?
My XO-less 2-way micor55 LX rendered Irish harp real-sized beyond the wall, harpist arms crisscrossing over its face (even though adjacent strings weren't spatially separated). Thirty years ago ARC D70II->ESL63 (without the benefit of Hugo) rendered this track with incredible palpability -- but no harp face, moving arms.
And I have some hope for 12" Zenith/Aura pseudo-isobaric, Aura driven by phase-matching plate amp.
Last edited:
Duntechs were IMO ultimate in their ideology of 1st order passive speakers. But still had some issues like any loudspeaker - compromise of properties.
https://www.stereophile.com/content/dunlavy-audio-labs-sc-iva-loudspeaker-measurements
https://www.stereophile.com/content/dunlavy-audio-labs-sc-i-loudspeaker-measurements
https://www.stereophile.com/content/dunlavy-audio-labs-sc-iva-loudspeaker-measurements
https://www.stereophile.com/content/dunlavy-audio-labs-sc-i-loudspeaker-measurements