Very cool KJ, thanks for the results and description. Nice that you did it at LSAF. I had a demo room there once.
The test looks very fair, well thought out and the results are indeed null. What were the drivers? I'm curious because when I did similar tests the drivers sounded different with different slopes. I did try to get the response thru the crossover area as close as possible.
The test looks very fair, well thought out and the results are indeed null. What were the drivers? I'm curious because when I did similar tests the drivers sounded different with different slopes. I did try to get the response thru the crossover area as close as possible.
Yes, but a null result means that statistically listeners didn't prefer one XO over the other. That is a pretty interesting for a null result.
I wonder if the preference Jon Bocani and I show for very steep slope crossovers is down to the way the DEQX agorithms in our processors work? Supposedly they address concurrent phase/timing alignment across 'all' frequencies. I'm not sure if this level of processing applies to all other DSP solutions.
When an experienced user set up miniDSP in my system a few years ago, it seemed to give quite a 'digital' glare to my (predominantly) analogue listening. DEQX certainly does not.
Last night I switched in real time between 2nd, 4th, 8th & 16th order from mid-treble as a sense check from following this thread. I was listening predominantly to vinyl replay but this was not a blind test & I tried hard not to have 'expectation bias' - difficult, I accept.
Whilst all retained the same sense of transparency, music became subtly more dynamic as the slope increased. Comparing 12dB with 96dB, the difference is most apparent - it's as if eq has been applied to roll off and gently soften the treble, at the same time taking some of the life out of the performance. Maybe my preference is to have music sounding this way. Probably the fact that amps & drivers are working with narrower ' cleaner' frequency bands will also be a factor.
A lot of so called 'audiophile' members of various forums have listened to my system and pretty well unanimously declared it to be equal to or better than anything they have heard so I remain convinced that DEQX is probably achieving something fairly unique.
When an experienced user set up miniDSP in my system a few years ago, it seemed to give quite a 'digital' glare to my (predominantly) analogue listening. DEQX certainly does not.
Last night I switched in real time between 2nd, 4th, 8th & 16th order from mid-treble as a sense check from following this thread. I was listening predominantly to vinyl replay but this was not a blind test & I tried hard not to have 'expectation bias' - difficult, I accept.
Whilst all retained the same sense of transparency, music became subtly more dynamic as the slope increased. Comparing 12dB with 96dB, the difference is most apparent - it's as if eq has been applied to roll off and gently soften the treble, at the same time taking some of the life out of the performance. Maybe my preference is to have music sounding this way. Probably the fact that amps & drivers are working with narrower ' cleaner' frequency bands will also be a factor.
A lot of so called 'audiophile' members of various forums have listened to my system and pretty well unanimously declared it to be equal to or better than anything they have heard so I remain convinced that DEQX is probably achieving something fairly unique.
from the opening post
by this are you in fact referring to bi-wiring?(i personally believe that bi-wiring is an insidious plot to sell speaker wire)
i've used PLLXO and two amps for two way systems to good effect but regard it as an "active" system (amps have gain)
by this are you in fact referring to bi-wiring?(i personally believe that bi-wiring is an insidious plot to sell speaker wire)
i've used PLLXO and two amps for two way systems to good effect but regard it as an "active" system (amps have gain)
i've always had a problem with extremely steep x-over filters, i find that they add a vail of distortion to what's passing through them.
to me it's a compromise low order filters allow drivers to interact and increasing the rate of roll off adds distortion. the only 48 db/oct. x-over i've ever liked was a BSS.
some may interpret the low level of distortion as a positive attribute?
to me it's a compromise low order filters allow drivers to interact and increasing the rate of roll off adds distortion. the only 48 db/oct. x-over i've ever liked was a BSS.
some may interpret the low level of distortion as a positive attribute?
Yes that is what it means and what is so interesting about it. Not sure about whether the listeners could tell the difference, but if there was not a preference, it doesn't matter much.
The drivers were composed of what I had on hand at the time that was suitable for the purpose, excluding the mids. Per cabinet:
4 X Adire Extremis 6.4 (2 each side of cabinet), 7" LF, 20Hz-350Hz
1 X ScanSpeak 10F/8424G 4" Mid 350Hz-2500Hz
1 X Aurum Cantus APR2.0 Ribbon HF 2500Hz and up
The cabinets were prefinished Parts Express 1 cubic foot models that I also had on hand.
System Response:
Measured from 50Hz up due to the measurement chamber LF limit. 2 meter mic distance. Because I wanted to limit the quantity of PEQ filters, it is not especially flat (although it looks pretty good at 10dB/division
).
Boundary LF measurement (performed at a large parking lot), 2 meter mic distance:
Thank you KJ, very interesting
To me, it looks a lot like most people (if not all) were not able to spot the difference. We don't know that, but it looks like it.
That's why i always proceed with Identification Blind Test format before making any Appreciation Blind Test. When you have participants that can spot differences, then you can ask them their preferences.
Now, you have 47/50 results (not to mention the 21 without preferences), and it could be the usual split because of random OR it could really be a split appreciation. The problem is: we don't know.
Identification, on the other hand, is very objective: either you can identify or you can't.
Anyway, i would not be surprised that people in general don't spot the difference between a 24db/oct and a brickwall slope. If i were to make an identification blind test i'd start with 6db or 12db against a brickwall, in order to maximise the chances of spotting them.
FYI,
I just redid my HT with active speakers after using passive for decades. In a theater with multi-channels etc. going active is much more complex than passive. In fact, in my case it was simply a PITA doing so. Conclusion, active may be the easier to design a crossover for many and it can tweak out a little better response, but in my system the audible differences were negligible or none. I now basically regret doing it as I am stuck with racks of amps and external MiniDSP modules, etc. and a maze of wiring that you cannot imagine.
I just redid my HT with active speakers after using passive for decades. In a theater with multi-channels etc. going active is much more complex than passive. In fact, in my case it was simply a PITA doing so. Conclusion, active may be the easier to design a crossover for many and it can tweak out a little better response, but in my system the audible differences were negligible or none. I now basically regret doing it as I am stuck with racks of amps and external MiniDSP modules, etc. and a maze of wiring that you cannot imagine.
If you consider that a pair good quality passive 3rd or 4th order xover for <1 kHz frequencies can cost several hundred euros, active becomes very attractive, even more so, if you consider the cost of tweaking the xover (zero euros for active, several tens for passive).
My only gripe is home theater setup. If you use digital xover and want to avoid unnecessary DA-AD conversion, only pre-pro option is Trinnov at ~18000 euros for the base model. Or give up the convenience of a standalone blu-ray player and deal with the fustercluck state of playing blu-rays on a PC. Or do away with optical disks altogether and deal with the uncertainties about audio-video quality and availability on streaming services. Or get really resourceful and retrofit I2S or S/PDIF outputs on your everyday average normal AV receiver.
My only gripe is home theater setup. If you use digital xover and want to avoid unnecessary DA-AD conversion, only pre-pro option is Trinnov at ~18000 euros for the base model. Or give up the convenience of a standalone blu-ray player and deal with the fustercluck state of playing blu-rays on a PC. Or do away with optical disks altogether and deal with the uncertainties about audio-video quality and availability on streaming services. Or get really resourceful and retrofit I2S or S/PDIF outputs on your everyday average normal AV receiver.
I didn't vote because I didn't feel any of the options really fit.
I have passive crossovers on my MTM's I don't like the idea of DSP based crossovers (ludite) and to do an analog active crossover for them (with corrections I have in the passive crossovers) would have been too difficult (for me).
I have active crossovers between my woofers and my MTM's. Depending on your definition of EQ I also have some corrections to deepen the bass response. They are analog, and use simulated inductors, which allowed me to design them just like a passive crossover so I could get the acoustic slope I wanted.
So for me, use what best suits the speakers you have. If that's passive then use passive, if it is active go active. All slopes are acoustic 4th order bessel. That seems to work best for me
Tony.
I have passive crossovers on my MTM's I don't like the idea of DSP based crossovers (ludite) and to do an analog active crossover for them (with corrections I have in the passive crossovers) would have been too difficult (for me).
I have active crossovers between my woofers and my MTM's. Depending on your definition of EQ I also have some corrections to deepen the bass response. They are analog, and use simulated inductors, which allowed me to design them just like a passive crossover so I could get the acoustic slope I wanted.
So for me, use what best suits the speakers you have. If that's passive then use passive, if it is active go active. All slopes are acoustic 4th order bessel. That seems to work best for me
Tony.
or you could go with one of these...
XM126 Vacuum Tube Electronic Crossover Network 2-way, 3-way or 4-way
passive can not add anything it is merely substractive no corrolation to before or after an amp.
active has gain/amplification and has the ability to add or multiply the signal.
I cheerfully paid the cost of an 8 channel amp for the convenience and instant gratificiation of being able to implement and tune my XOs in DSP.
I don't think a Trinov is necessary. So far, I've avoided multiple A/D D/A conversions by using my TV as an HDMI audio extractor. All my digital sources (computer, DVD, cable TV) have HDMI outputs. My AVR aggregates them to the TV. The TV has a 2 channel TOSLINK output to my DSP. My DSP can switch between TOSLINK and SPDIF. My analog sources are AtoDed and then come in via SPDIF.
Interestingly, in order to get the AVR to enable the audio on its HDMI output, I had to in effect tell it that a downstream device was providing the audio. When I hookup my surround speakers, I may find that they are silent because of this. The workaround would be a 2nd AVR or an HDMI switch to use in place of the original AVR, freeing it to handle surround and center channel duties.
Big deal. Not even remotely audible if you set it up correctly.
I own over 2500 DVD/BR. All are ripped to my server (full rez .mkv), and I have zero difficulty in playing any of them back.
I couldn't disagree more Earl.
My HT has been active (6 ch 4 subs) for 8 years, and once the rack was built, it has hardly been touched. I's in an understairs cupboard so not visible/audible in normal use. It is a tad more complicated than a passive 2ch set up but needs no more maintenance and as it's less complex than my last PA system and considerably less complex than an OB van or TV studio to build, it was a doddle to build.
It's my experience that active allows advantages that passive can't begin to bring and sounds better when well implemented, always. Plus it's a hell of a lot easier to adjust.
I first used active about 40 years ago. It was "cool" - all Crown stuff.
About 10-20 years ago I tried active several times and never found it to be a significant audible improvement in direct back to back tests.
My recent experience is the most audible difference to date mostly because of better testing capability, but still not that big a difference.
So I suppose "I couldn't disagree with you more".
When I think of the number of times I set an active crossover, adjusted its slope or frequency, measured, listened & then changed again until I was completely satisfied, I dread to think how much it would have cost me to try even a few of these variants with passive components.
That, for me is one of the greatest advantages & even if active becomes somehow inferior to a purist, can the user be absolutely sure that what they hear couldn't be bettered?
That, for me is one of the greatest advantages & even if active becomes somehow inferior to a purist, can the user be absolutely sure that what they hear couldn't be bettered?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.