So if someone claims they had only two minutes of sleep in the past two weeks has to prove it?😀
I had my battle with absolute phase too but honestly how can i know what phase is right ?
I have to correct my post 270. Mr.Stoll was not talking about phase correction in the bass but about correcting the group delay ( making it flat ) to very low frequencies. That needs a lot of computing power and in general he has made the experience that to make the bass perform correct soaks up a lot of efford, cost and time wise not to mention the know how.
I have to correct my post 270. Mr.Stoll was not talking about phase correction in the bass but about correcting the group delay ( making it flat ) to very low frequencies. That needs a lot of computing power and in general he has made the experience that to make the bass perform correct soaks up a lot of efford, cost and time wise not to mention the know how.
Yes, Mr.Stoll uses DSP. I will first design the ZDL in the analog domain and maybe i will try DSP too later. To linearise the group delay in the low frequency range makes sense to me.
When we brought the first Audio Physic subwoofer to market in the early 90th it went down into subsonic frequencies too. We did that in the analog domain with a bridge circuit and partly current drive. It was designed to work in a corner and had quite a big cone area and a powerfull amp on board. The sonic result was a bigger and better defined soundstage and it improved the midrange too to our surprise. In those day a lot of competitors where laughing at us unlil they heard the product. It also had a quite sofisticated room equalizer on board. I myself have not done DSP experiments in the deep bass but Mr.Stoll tells me that he gets a result where the bass is absolutely deep, clean and transparent. In the 90th there was the EUREKA experiment that came to the same conclusion. We seem to be sensitive to time distortion in the bass.
When we brought the first Audio Physic subwoofer to market in the early 90th it went down into subsonic frequencies too. We did that in the analog domain with a bridge circuit and partly current drive. It was designed to work in a corner and had quite a big cone area and a powerfull amp on board. The sonic result was a bigger and better defined soundstage and it improved the midrange too to our surprise. In those day a lot of competitors where laughing at us unlil they heard the product. It also had a quite sofisticated room equalizer on board. I myself have not done DSP experiments in the deep bass but Mr.Stoll tells me that he gets a result where the bass is absolutely deep, clean and transparent. In the 90th there was the EUREKA experiment that came to the same conclusion. We seem to be sensitive to time distortion in the bass.
Still, after 10 pages, I'm still not clear of how you intend to align the drivers in the speaker. I would like to know how you go about the subject of time coherence for speakers like Herr Stolls and the Linkwitz speaker with the plastic pipes, where the woofer is positioned +/-90 degrees off the other drivers axes? Are you intending something similar?
This will be aproxomately how it will look. The tweeter is free standing on a nylon ball.
The midrange will be in this ball. Angling the midrange i can do time alignement. The woofer is an a conventional box. The reflex will be a big size port going from the woofer to the back. Between the midrange and treble i plan a low order crossover and between the midrange and bass it will be steeper.
The midrange will be in this ball. Angling the midrange i can do time alignement. The woofer is an a conventional box. The reflex will be a big size port going from the woofer to the back. Between the midrange and treble i plan a low order crossover and between the midrange and bass it will be steeper.
Attachments
Joachim,
You may wish to review the work of Blauert et. al. on spatial perception and frequency response as well on sensitivity in the human hearing to groupdelay "distortion" (I also found good correlation with the inaccuratly called Fletcher/Munson Curves).
I found doing that rather illuminating, for example regarding why a 10th order crossover at some specific frequencies is essentially inaudible while a simple 4th order LR at other frequencies (like 2KHZ) is brutally audible and really unpleasant, despite correct summation, flat frequency response and all...
Ciao T
You may wish to review the work of Blauert et. al. on spatial perception and frequency response as well on sensitivity in the human hearing to groupdelay "distortion" (I also found good correlation with the inaccuratly called Fletcher/Munson Curves).
I found doing that rather illuminating, for example regarding why a 10th order crossover at some specific frequencies is essentially inaudible while a simple 4th order LR at other frequencies (like 2KHZ) is brutally audible and really unpleasant, despite correct summation, flat frequency response and all...
Ciao T
I'm pretty familiar with Blauert's work, that's why I'm wondering what correlation between equal loudness curves and "spatial perception and frequency response as well on sensitivity in the human hearing to groupdelay" Thorsten is talking about.
Equal loudness contours of ISO226:2003 (which are derived from pure tones) don't correlate well to our loudness perception of music. See the curves applied by Audyssey and Dolby.
Last edited:
Assume the pure tones were tested with headphones. Normally the tones are tested for quite short durations too. To know more, among many other considerations, it would be valuable to have sensitivity with duration of the tone.
The ISO226:2003 data is based on several studies. Not only headphones were used: http://www.mp3-tech.org/programmer/docs/IS-01Y-E.pdf
Last edited:
Hi,
Suit yourself.
I said what I wish to on the topic, let the wise ones get use from it.
Ciao T
Suit yourself.
I said what I wish to on the topic, let the wise ones get use from it.
Ciao T
Maybe someone else can help out because I still don't know what your conclusion from looking at that data (which data in particular?) really was 😕
Hi,
My conclusion (also experimentally proven to my personal satisfaction) is that some specific frequencies are markedly better choices as crossover points than others.
Ciao T
My conclusion (also experimentally proven to my personal satisfaction) is that some specific frequencies are markedly better choices as crossover points than others.
Ciao T
- Status
- Not open for further replies.