Ladies and gentlemen, i think this is my last try for a proof of concept of the original demand to change the drivers and the passive crossover, again hope it helps the thread opener to come to some kind of intelligent conclusion to have a benefit and some fun to do it on his own.
Changed target function from Bessel 2nd order 2500 Hz to Linkwitz-Riley 4th order 2200 Hz with nearly the same number of components and mid-woofer and tweeter with waveguide with the same polarity
Changed target function from Bessel 2nd order 2500 Hz to Linkwitz-Riley 4th order 2200 Hz with nearly the same number of components and mid-woofer and tweeter with waveguide with the same polarity
@wafflesomd this isn't a black and white situation and WGs aren't the end all, be all solution to create an optimal 2 way design. I've mixed on many speakers and ultimately it comes down to how well you can interpret what you hear. Cheap speakers that rely on DSP to hammer the FR into shape aren't going to make up for inherent driver limitations which can't be fixed.
I think too many people are hung up on the whole WG thing these days. In order for the WG to work properly, you have to adhere to a specific crossover frequency, slope and woofer dimensions to get the full benefits. That means mostly lower than optimal transition points for the HF driver, often closer to Fs, which demands more driver linearity as well as operating in a higher distortion area, despite benefits of increased HF driver efficiency down low.
The other issue is crossing in the sensitive midrange area, where ears are more sensitive to phase shifts and jumps in FR. I always try to avoid crossing in the critical 1 - 4k area, which is why i favor 3 way systems. If you however have to do a 2 way (due to size limitations), it requires a bass-mid which can play clean up well past 3k with minimal breakup and controlled dispersion - no small feat. No amount of DSP will fix that if you want stable imaging that doesn't jump around depending on which way you turn your head.
Then you have the complexity of filtering, which often necessitates very extensive. complex passive networks or DSP, thr latter which I'm not a fan of and can produce quite audible pre-ringing. I realize the trend is going to perfect power response, but at the expense of low level resolution, THD and dynamics.
Just because a 2 way doesn't use a WG, doesn't mean its not of professional caliber or can't produce good results in the practical world. I've heard more bad 2 way monitors with WGs than decent ones without. You get what you pay for, with or without a WG.
I think too many people are hung up on the whole WG thing these days. In order for the WG to work properly, you have to adhere to a specific crossover frequency, slope and woofer dimensions to get the full benefits. That means mostly lower than optimal transition points for the HF driver, often closer to Fs, which demands more driver linearity as well as operating in a higher distortion area, despite benefits of increased HF driver efficiency down low.
The other issue is crossing in the sensitive midrange area, where ears are more sensitive to phase shifts and jumps in FR. I always try to avoid crossing in the critical 1 - 4k area, which is why i favor 3 way systems. If you however have to do a 2 way (due to size limitations), it requires a bass-mid which can play clean up well past 3k with minimal breakup and controlled dispersion - no small feat. No amount of DSP will fix that if you want stable imaging that doesn't jump around depending on which way you turn your head.
Then you have the complexity of filtering, which often necessitates very extensive. complex passive networks or DSP, thr latter which I'm not a fan of and can produce quite audible pre-ringing. I realize the trend is going to perfect power response, but at the expense of low level resolution, THD and dynamics.
Just because a 2 way doesn't use a WG, doesn't mean its not of professional caliber or can't produce good results in the practical world. I've heard more bad 2 way monitors with WGs than decent ones without. You get what you pay for, with or without a WG.
It's not really about waveguides, it's about getting good dispersion at the transition between drivers, waveguides are just a really good way to do that and probably the best way to do it with a typical two way speaker where you still need the woofer to be large enough to make any usable bass. I have not heard a typical dispersion mismatched two way that competes with one that has optimized it's dispersion. WG present what many would consider a solution to a problem and recommending one build some old school mismatched speakers doesn't make any sense to me.
Eh, I'd argue this is a bit of a dated take and probably more backed up by marketing from speaker companies who can't make very good speakers. If you're just listening, sure get whatever presents music in a way that sounds right to you, with monitors I totally disagree. The more neutral and better dispersing speaker IME always gives the user better translation and I have not run into a case that tells me otherwise. Not sure I agree with the second take either, look at the JBL M2. The horn/waveguide was the primary focus then the final integration with the CD is achieved with DSP. This is a pricey speaker, but the point is one needs context to determine whether DSP correction was the appropriate choice in the design. As far DSP not fixing problems, that's pretty much one of the biggest limitations of the mismatched two ways I mention.
You can get $25 tweeters with FS far below the xover points most practically sized waveguides will let you use. Don't really see this as an issue.
Seems most three way speakers, including commercial monitors are crossing the mid to the tweeter in this range. PMC specs 3.8khz for one speaker, ATC I believe is around there at 3.5k, just some examples.
I find active filtering far superior in just about every way for various reasons, and it appears commercial monitor makers agree with this.
If the goal is achieve a speaker that presents music in a neutral manner that offers the best potential room integration and mix translation, old school dispersion mismatched two ways are not the way to get there.
I feel most 2 way designs cross the tweeter too low, with or without WG. The WG helps a little to offset the lower mid loss, but often doesn't make up for the lower xover induced rise in THD vs a higher xover without WG. Most of the time the tweeter can't keep up at higher output levels in either scenario. Hence the favoring of 3 way designs as a first option.
Driver spacing is also an issue and requires a lower xover to compensate for it, which forces the situation even more so. Id rather have less driver spacing and a higher xover if the LF driver allows without excessive beaming. I see both sides of the argument.
I avoid DSP at all costs. It boxes you into a corner with your signal chain and allows less flexibility with amplification. The increase in noise at low levels with high efficiency drivers is often a problem with DSP. Passive filters don't add noise. They may have losses, but the tradeoff is worth it to me in most cases.
I'm not sold on DSP, unless its of highest caliber. Mini DSP doesn't cut it for me, as most off the shelf consumer rack gear. A well designed passive filter usually sounds better to me, especially if your application calls for a fullly analog signal chain.
Call me old school if you wish. I don't use social media, drink Starbucks or watch TV. It kills the senses. I use a smart phone, but only because I have to. Otherwise I'd own a flip phone and use a land line if I could. LOL
Driver spacing is also an issue and requires a lower xover to compensate for it, which forces the situation even more so. Id rather have less driver spacing and a higher xover if the LF driver allows without excessive beaming. I see both sides of the argument.
I avoid DSP at all costs. It boxes you into a corner with your signal chain and allows less flexibility with amplification. The increase in noise at low levels with high efficiency drivers is often a problem with DSP. Passive filters don't add noise. They may have losses, but the tradeoff is worth it to me in most cases.
I'm not sold on DSP, unless its of highest caliber. Mini DSP doesn't cut it for me, as most off the shelf consumer rack gear. A well designed passive filter usually sounds better to me, especially if your application calls for a fullly analog signal chain.
Call me old school if you wish. I don't use social media, drink Starbucks or watch TV. It kills the senses. I use a smart phone, but only because I have to. Otherwise I'd own a flip phone and use a land line if I could. LOL
The reasons commercial monitors manufacturers use dsp and active solutions are more of a practical nature, to offer all in one system. It does not mean it is the only way to good sound.
Loudspeaker with wg does not automatically sounds better then non waveguided. People who build many different projects, different concepts, with different drivers, listen to commercial products....they have the right comparison and insight. When I talk to others diyers and hifi enthusiast, they often mention good sound is not about wg or dsp, these things are not what decides.
I did 3way with wg and octo dac and nanosharc. So we can say it got state of the art dsp and dacs. It sounded great, but I am 100% surre passive version will be cleaner, more relaxing, not synthetic sounding. These are mild difference that are not obvious immediatelly, it takes many hours of listening and comparison to with other speakers.
Loudspeaker with wg does not automatically sounds better then non waveguided. People who build many different projects, different concepts, with different drivers, listen to commercial products....they have the right comparison and insight. When I talk to others diyers and hifi enthusiast, they often mention good sound is not about wg or dsp, these things are not what decides.
I did 3way with wg and octo dac and nanosharc. So we can say it got state of the art dsp and dacs. It sounded great, but I am 100% surre passive version will be cleaner, more relaxing, not synthetic sounding. These are mild difference that are not obvious immediatelly, it takes many hours of listening and comparison to with other speakers.
Last edited:
Hmm, perhaps this forum is not so much science or research oriented as I once thought. OP wants a monitor and most of the suggestions are pushing him away from that.
Good luck OP, you'll absolutely need it.
Good luck OP, you'll absolutely need it.
DSP has its benefits. No denying that. The flexibility and user friendliness is a massive plus. I'm rather old school. I dont march to the beat of one drum. I use my own ears. Some people just don't agree and thats ok.
I think the OP can make up his own mind based on the info given. Its like going to see a Dr. There are multiple ways to treat the same illness. Each way has side effects. It depends on which ones you can deal with. I'd rather have all the information and come to my own conclusion. Thats why the good Lord gave us a brain. Unfortunately we Americans like to be told what to do and not think for ourselves.
I think the OP can make up his own mind based on the info given. Its like going to see a Dr. There are multiple ways to treat the same illness. Each way has side effects. It depends on which ones you can deal with. I'd rather have all the information and come to my own conclusion. Thats why the good Lord gave us a brain. Unfortunately we Americans like to be told what to do and not think for ourselves.
That's simply not true and a THD measurement shows quickly if a tweter is the limmiting factor. Have a look: https://www.neumann.com/de-de/produkte/monitors/kh-120-ii/
THD level over 2kHz stays very low and K3 is incredibly low. MaxSPL is by no means restricted from the tweeter.
With active electronics you CAN do a crossover at 1,5-2kHz without problems but it needs some time and knowledge to get it right. You still have all teh restrictions and problems to deal with as with a passive crossover (it's silly to think you can correct problems of the speaker with DSP) but you have WAY more possibilities to do it correct (like eq/delay the driver to get perfect crossover shapes).
Most MiniDSPs also don't cut it for me, but Hypex or the Neumann DSPs or Genelec Ones (don't know the cheaper ones) don't add noise - they are so low in noise and THD that most signal chains absolutely benefit from it when useing the digital input.
Crossing over at 3-4kHz ... has profound downsides, there is a reason we went away from it.
Have you done a listening comparison of active/DSP vs. passive/analog filters, where both filters implemented the very same transfer function? I did exactly that during the development of the OSMC, and the passive filter sounded much better to my ears. Adding the costs for good DACs/ADCs and and the extra amps used for active/DSP systems, I don't see their advantage.
With all due respect Mbrennwa: i've done the same with my own monitors and came to the inverse conclusion.
There is too many points to take into consideration imo: some of the benefits i see might come from multiamping, some from the programation of software used, dac quality, whatever... and it's the same with passive: change the type of capacitor= different sound, type of coil: different sound, type of resistance ( maybe?) a different sound...
It seems we all agree there is differences between approach and outcome.
Only experience with both can tell which one you prefer and why.
That said, studio monitoring is historically well documented and there is trends and evolution ( past the hype of the moment).
One thing is sure since mid nineties waveguided dome ( even 3 ways, i mixed on Genelec 1037a circa 97) and active filtering/ multiamping is the norm and for good reason imho.
The issues about reliability are true mainly in entry/mid level gear. I've seen zero issue with Atc Scm110A or the Genelec's own electronics in multiamped/active filtered gear for the cumulated 10years i spent with both units.
There is too many points to take into consideration imo: some of the benefits i see might come from multiamping, some from the programation of software used, dac quality, whatever... and it's the same with passive: change the type of capacitor= different sound, type of coil: different sound, type of resistance ( maybe?) a different sound...
It seems we all agree there is differences between approach and outcome.
Only experience with both can tell which one you prefer and why.
That said, studio monitoring is historically well documented and there is trends and evolution ( past the hype of the moment).
One thing is sure since mid nineties waveguided dome ( even 3 ways, i mixed on Genelec 1037a circa 97) and active filtering/ multiamping is the norm and for good reason imho.
The issues about reliability are true mainly in entry/mid level gear. I've seen zero issue with Atc Scm110A or the Genelec's own electronics in multiamped/active filtered gear for the cumulated 10years i spent with both units.
Last edited:
What DSP did you use? These Hypex modules sound exellent, esp. with the digital input. It's a few hundred for really good DACs and world class power amp.
How good did you match the frequency response? You need at least +-0,5dB which is not that easy, esp when the speakers stand on different spots in the room.
But the main point - it doesn't make sense to use DSP and then restrict your filters to what analog stuff can do! Delay alone to really get perfect phase for the crossover is worth the upgrade.
Do a proper DSP setup and THEN compare to the passiv crossover ... my tests had clear results.
Oh. There is a reason why MiniDSP doesn't specify S/N and THD with these units ...
I would suggest you try a really good DSP/amp combination when you want to judge sound quality. The Hypex are hard to beat in this regard if you can accept the software ... 🤓
These also look pretty good. Not sure if they can compete with the Hypex in terms of S/N but analog input seems better and software is pretty interesting.
https://www.four-audio.com/en/products/ppa-series/
I would suggest you try a really good DSP/amp combination when you want to judge sound quality. The Hypex are hard to beat in this regard if you can accept the software ... 🤓
These also look pretty good. Not sure if they can compete with the Hypex in terms of S/N but analog input seems better and software is pretty interesting.
https://www.four-audio.com/en/products/ppa-series/
I didn't care for my minidsp HD, felt like a veil was put on the signal. Don't even know how to describe it really.
Running the filters in my pc sounds comparatively transparent.
Running the filters in my pc sounds comparatively transparent.
Yes software+good semi pro/pro soundcard are a good answer for some years now.
This solutions are now in the same league as good dedicated loudspeaker management units. In fact if you don't need to move your stuff too often, can accept to have a ( small) computer dedicated to filtering it's the way to go imho.
4audio products seems very nice. Dante enabled is great, now i've tried it on my unit i'm convinced it's a great tool. I'm sold on it.
This solutions are now in the same league as good dedicated loudspeaker management units. In fact if you don't need to move your stuff too often, can accept to have a ( small) computer dedicated to filtering it's the way to go imho.
4audio products seems very nice. Dante enabled is great, now i've tried it on my unit i'm convinced it's a great tool. I'm sold on it.
Last edited:
There has been done a thorough investigation, say a technical review, and the Minidsp HD isn't perfect but has very good s/n and thd figures and this has been documented and can be found on the forum. For the price it is very good value i.m.o.but obviously not high end (high spend). It does well enough for the majority of users, otherwise it would not be such a popular product.
I don't know why it is so popular. I've not heard the latest one but had the first gen to play with imo it wasn't on par with dcx2496 which wasn't on par with my own expectation either.
Except for the usual suspects ( Bss, Dbx, Dolby) there wasn't so many good loudspeaker management units 15 years ago...
Except for the usual suspects ( Bss, Dbx, Dolby) there wasn't so many good loudspeaker management units 15 years ago...
That's the answer: they got better with the HD models 😉 , but if I would have the money, Legend Audio has a nice dsp offering I lust for.