Thank you so very much for your appreciated reply!
I just logged in because I wanted to post the following:
"So very interesting news (for me) that I've ever read so far.
I would like to know more.
It is so interesting that I immediately wondered: why don't everyone do it?
The subsequent question was: how much does it cost to do it?
And then, what are the musical instruments you have established with the timbric sound identity compared to the recorded track in listening?
How does the bass and high response of your speakers result in your room?
And the 3D scene?
About your system I would like to know everything else too."
But practically you already answered me before I asked you my questions... 🙂
Just for the record, please note that I was referring to this your great comment (IMO)
https://www.diyaudio.com/community/...-of-measurements-in-audio.392799/post-7206577 👍
https://www.diyaudio.com/community/...-of-measurements-in-audio.392799/post-7206577 👍
I build these one:
https://www.diyaudio.com/community/threads/best-fullrangedriver.372476/page-13#post-7184394
https://www.diyaudio.com/community/threads/how-much-bass-eq-do-you-use.377768/page-4#post-7084495
https://www.diyaudio.com/community/...r-phase-bass-reflex-subs.372878/#post-6690217
https://www.diyaudio.com/community/...ion-with-a-2-way.334757/page-364#post-6711184
https://www.diyaudio.com/community/threads/best-fullrangedriver.372476/page-13#post-7184394
https://www.diyaudio.com/community/threads/how-much-bass-eq-do-you-use.377768/page-4#post-7084495
https://www.diyaudio.com/community/...r-phase-bass-reflex-subs.372878/#post-6690217
https://www.diyaudio.com/community/...ion-with-a-2-way.334757/page-364#post-6711184
Last edited:
I optimize only the loudspeaker with help of ARTA measurements and ATB PC 501/ Pro from Kirchner Electronics measured in the near field of 20cm. I never optimize the room as this is unpredictable and not common sense in audio engineering.
I made a near field recording of:
Big and small triangle
Plastic flute
Small wooden xylophone
My voice
Beer bottle plopped with finger
Mouth Harmonika type c
These instruments are resistant to humidity. A drum with leather is not so it sounds different every day.
The room dominates the bass sound in the room. As I had a good room I never had problems with good sound in the bass. It was sufficient to make small adjustments in the bass just controlling two dsp / eq filters: one in the region around 150 hertz and the other one at 45 hertz for the deep bass. This works well if the room behaves well acoustically.
3d Soundstage is good like with all decent fullrange drivers or good coax speakers.
Not everyone is a fan of loudspeaker point sources. So you always have a competition. If you listen to very processed recordings time coherence is hardly audible. For long time phase linearity (200hertz up to 5khz is audible) was neglected in the audio engeneering.
However today explains the big company RCF the difference since they use FIR for their products on a large scale.
Main difference is correct impulse reproduction, together with that correct color of sound and imaging.
Direct listening to instruments used the BBC in the sixties. As recordings were not that good they played in two identical rooms simultaneously: loudspeakers in one room, live band in the other with microphones wired to the speaker. The listeners fastly swapped the room and could compare the sound.
However I seldomly hear people use their own recording for optimizing loudspeakers. Today a zoom recorder (or others) of better quality will do. Untrained listeners are already happy with a zoom H1 for 100euro.
A loudspeaker builder needs in the end some control with real instruments as you never know if you are right due to daily moods and tricking yourself. So you would end up changing an already finished loudspeaker you shoudn't touch any more.
It is not good to use steep filters if they can be omitted or correct every deviation on axis if off axis it's correct or compensating the on axis response. It counts the completely emanated sound power into the room and that highs, mids and bass are in a correct level to each other.
However I seldomly hear people use their own recording for optimizing loudspeakers. Today a zoom recorder (or others) of better quality will do. Untrained listeners are already happy with a zoom H1 for 100euro.
A loudspeaker builder needs in the end some control with real instruments as you never know if you are right due to daily moods and tricking yourself. So you would end up changing an already finished loudspeaker you shoudn't touch any more.
It is not good to use steep filters if they can be omitted or correct every deviation on axis if off axis it's correct or compensating the on axis response. It counts the completely emanated sound power into the room and that highs, mids and bass are in a correct level to each other.
Thanks a lot, it is a huge amount of new news for me as so very interesting and also fascinating.
I had already guessed that almost everything is played on the frequencies to which the human ear is more sensitive and that therefore some inevitable compromises had to be there somewhere, but this applies to any system.
However, there is an enormous interest for me regarding your experiences excellently described.
The huge amount of information you provided requires an adequate study.
Thank you again for your clarity and availability.
I had already guessed that almost everything is played on the frequencies to which the human ear is more sensitive and that therefore some inevitable compromises had to be there somewhere, but this applies to any system.
However, there is an enormous interest for me regarding your experiences excellently described.
The huge amount of information you provided requires an adequate study.
Thank you again for your clarity and availability.
Hi thank you very much for your very valuable advice ... you say "I derived from this webpage that if you build a loudspeaker the most important are the mid frequencies like 1 khz where the ear is most sensitive."
Ok ... but sensitive to what ? that is the problem 🤔🤔🤔
However this is extremely important i.e. to understand which is the critical range for listening Very important
I had the feeling that the ear is not equally sensitive across the audio range
As a consequence to place a xover cut close to that critical 1kHz could lead to issues imho
Anyway when i hear about "compression" of sound at higher SPL peaks the 1st thing that comes to my mind is distortion that rises above a certain level
When i see people raving about the bass of a mini speaker i think that they are false or unexperienced about what a good bass is Honestly
Years ago the reference were speakers with big woofers usually ... then the flats became smaller and the minimonitors appeared
I am pretty sure that many minimonitor owners have never experienced what a real bass gives to the music It cannot be explained in words Amazing
Minimoniotrs need a sub to provide a really satisfactory musical experience
A friend of mine had a pair of the LS3/5a ... a day he built two subs and when i heard the ensemble i immediately say ... now we reason
Before it was like listening to a good radio
I think of speakers like AR10 ... JBL L166 Speakers with big and powerful woofers
Now the focus is on look often and not the sound
Last edited:
I recently finished my own designed loudspeakers (3 ways 300W). So here are my 5 cents.
In my case to meter was a must, also to have a digital model of the speakers and adjust the model to the reality and then try digitally the changes and then in the real world. It was an iterative process.
It is super easy to see peaks and valleys in a curve, and compare them with previous curves, even after a long time. On the other hand, IMO, remembering sounds is difficult and tricky.
The final step is to listen, and for me the ultimate test is when your brain can not distinguish between what is coming from the speakers and what is "real", of course it happens only a few times with very specific sounds.
So final adjustments, the very small ones, are done by ear, because metering can't show everything and also because at the end we work for our ears.
In my case to meter was a must, also to have a digital model of the speakers and adjust the model to the reality and then try digitally the changes and then in the real world. It was an iterative process.
It is super easy to see peaks and valleys in a curve, and compare them with previous curves, even after a long time. On the other hand, IMO, remembering sounds is difficult and tricky.
The final step is to listen, and for me the ultimate test is when your brain can not distinguish between what is coming from the speakers and what is "real", of course it happens only a few times with very specific sounds.
So final adjustments, the very small ones, are done by ear, because metering can't show everything and also because at the end we work for our ears.
Attachments
Last edited:
love those and almost all Tannoy i listened to
maybe we can conclude that crossing at 1 kHz it's doable but a little tricky?
there is a nice thing in the DC driver
that the woofer cone acts as a lens for the tweeter
i have come to the conclusion that limiting the tweeter dispersion is a very good thing and supporting this i see more and more tweeters with lenses or felts absorbers like in the Dunlavy or the Lipinski loudspeakers
And also the test of the H and V dispersion vs frequency is very popular these days
Only distortion measurements are almost completely neglected Strange I like them
first of all congratulations for your great speakers
then do you mean that during the final adjustments you change parts in the xover by ears?
Yes, in my case final adjustments were done by ear, e.g. after a while I lowered the tweeters 1db to have a more balanced sound.
At this point it is a matter of taste, and also depends on the amplifier and source. They are ok with my tube amp and a charme when playing LPs, but a bit bright with the Marantz and digital audio.
At this point it is a matter of taste, and also depends on the amplifier and source. They are ok with my tube amp and a charme when playing LPs, but a bit bright with the Marantz and digital audio.
Big woofer monitors still exist, and are still used, but are useless for close listening like with nearfields. They are only fit for midfield to farfield monitoring.
But a JBL M2 uses a 15" and is a monitor in the sense that it's flat neutral response with controlled dispertion. And those are made by engineers who measure all the time.
A Neumann KH420G uses a 10", but if the space is big they recommend to combine it with a KH870 who has two other 10" for the lowest part, and that works as well from a distance, not close range. Then you're better of with the new KH150 (to stay in the brand) that uses a 6.5" driver and goes as low. So it all depend on the setting, the room and how far you're sitting from the speaker.
And the monitor that most mastering engineers now use, the Kii Three BXT uses 12x 6.5" woofer for that. No M2 can do what that speaker does. And again, developped by engineers who measure all the time and rely on measurement, not on subjective sighted hearing. This is the studio of one of Europe's most asked mastering engineers, Jerboa (aka Frederik Dejongh) in Ghent, Belgium with those Kii's.
But a JBL M2 uses a 15" and is a monitor in the sense that it's flat neutral response with controlled dispertion. And those are made by engineers who measure all the time.
A Neumann KH420G uses a 10", but if the space is big they recommend to combine it with a KH870 who has two other 10" for the lowest part, and that works as well from a distance, not close range. Then you're better of with the new KH150 (to stay in the brand) that uses a 6.5" driver and goes as low. So it all depend on the setting, the room and how far you're sitting from the speaker.
And the monitor that most mastering engineers now use, the Kii Three BXT uses 12x 6.5" woofer for that. No M2 can do what that speaker does. And again, developped by engineers who measure all the time and rely on measurement, not on subjective sighted hearing. This is the studio of one of Europe's most asked mastering engineers, Jerboa (aka Frederik Dejongh) in Ghent, Belgium with those Kii's.
IME that's hardly good enough for a SOA or near SOA system. The zoom recorder is junk in terms of accuracy compared to some reproduction systems.
Good point. We use live instruments and optical phono as references.
Also, just as a point of reference relative to reproduction speaker technology, good electrostatic speakers can far outperform electromagnetic box speakers. Some technical info available at: https://www.soundlabspeakers.com/wp-content/uploads/2019/07/The-Complete-White-Paper.pdf
Most of the system here is custom discrete electronics, except at this point for the electrostatic speakers, dual mono block power amps, optical phono cartridge, and mechanical turntable components. If interested, DS W3 optical phono cartridge tech is shown near the bottom of the page at: https://ds-audio-w.biz/info_optical_cartridge/
Additinal info: https://ds-audio-w.biz/cms/wp-content/uploads/2021/11/Technical-information20221212_c.pdf
...Google Chrome can translate to English if needed. The whole phono setup is roughly as complicated to get right as the custom dac, although the dac is mostly a set of electronics problems whereas phono is more in the way of mechanical complications.
Last edited:
Go check the Room acoustics section of this forum. There you find treads like these:
https://www.diyaudio.com/community/threads/diy-diffusors.327763/
https://www.diyaudio.com/community/threads/diy-rpg-skyline-diffuser-panel.24519/
...
Hi thanks and very impressive
However I would look at other types of drivers if EQ and DSP are used and the listening room is big
like this one ... https://www.bcspeakers.com/system/dragonfly/production/2021/01/17/16_03_16_852_file_b.jpeg
plus sub
I am a believer in high efficiency
I don't think that is fit for hifi with a very narrow dispertion of the tweeter. If you want hi sensitive coaxials nothing beats the Altec (or GPA clone) 604E, certainly if you use EQ to flatten out the response. And a bit cheaper, the Radian 5212 is also much better for hifi. Both have a much better dispertion.
The B&C you mention (and all their coaxials) are aimed at pa line arrays where dispertion needs to be small and the sound need to reach far, not for home use. The difference is the horn used on the tweeter.
But for very neutral speakers with controlled dispertion smaller coaxials are way better, like used in the Genelec 83xx series. They use also subwoofers in those cabinets, but in a bandpass config below 80Hz or so. Kef coaxials are similar (but less good integrated). That is because the dispertion on smaller drivers is way better in the higher frequencies of the woofer in general (depending on the design off course).
The B&C you mention (and all their coaxials) are aimed at pa line arrays where dispertion needs to be small and the sound need to reach far, not for home use. The difference is the horn used on the tweeter.
But for very neutral speakers with controlled dispertion smaller coaxials are way better, like used in the Genelec 83xx series. They use also subwoofers in those cabinets, but in a bandpass config below 80Hz or so. Kef coaxials are similar (but less good integrated). That is because the dispertion on smaller drivers is way better in the higher frequencies of the woofer in general (depending on the design off course).
hi thank you for the very valuable advice
but you posted a picture of a big studio room i was referring to that
Moreover why narrow dispersion must be a bad thing in general expecially considering a non treated room
actually i see many designers limiting tweeter dispersion with felts like Lipinski or Dunlavy
constant narrow dispersion could work bass aside of course
by the way I see dispersion graphs usually starting from 1kHz up
i listen alone right in the sweetest spot
i could save on acoustics treatment
no waf to cope with
Last edited: