Yikes!
Next step is to tell you only believe in current and not voltage or resistance in ohm's law because only peak to peak matter and not rms...
🙄 😀
Please not you Camplo! Already had to deal with this kind of things here and it is too much weird for most people... ( it is much too weird! Don't fall into the dark side of the force - of your brain).
Next step is to tell you only believe in current and not voltage or resistance in ohm's law because only peak to peak matter and not rms...
🙄 😀
Please not you Camplo! Already had to deal with this kind of things here and it is too much weird for most people... ( it is much too weird! Don't fall into the dark side of the force - of your brain).
Last edited:
What do you mean, Im expanding variables, not the opposite
Spl matters and the start up transient matters
Spl matters and the start up transient matters
Both are the same thing expressed differently... as with the ohm law, same relation expressed in different ways ( 3 dimentional expression).
But tbh i'm not this good at explaining this kind of things in my own language so won't even try in English.
Sure others will do better than i could.
How i understand things you try to explain is you forget than gd is a derivative of other phenomenons, and it could explain things by itself. I rather see this as a symptom of other related phenomenon.
But i'm limited in this so won't expand too much on this ( as i'am frustrated to not explain myself as easily as i wish).
But tbh i'm not this good at explaining this kind of things in my own language so won't even try in English.
Sure others will do better than i could.
How i understand things you try to explain is you forget than gd is a derivative of other phenomenons, and it could explain things by itself. I rather see this as a symptom of other related phenomenon.
But i'm limited in this so won't expand too much on this ( as i'am frustrated to not explain myself as easily as i wish).
Last edited:
The change in the amplitude at the beginning of these burst, have an audible effect...when the group delay is high enough, it takes a longer time to go from 0 amplitude to max amplitude....this is what I believe we hear at the point and past the threshold of audibility
This involves a distortion of amplitude over time.
Attachments
Last edited:
How do you generate the data in the images? Sure it looks like it should be audible but is it due to group delay?
They are recordings of a loudspeaker playing tone burst from REW.
Wether I chose high/low group area to play tone burst, or if changed GD using xo filter....the trend was there.
I already knew about this effect along time ago though....I'm not sure if I read it or if I theorized it based on peoples experience with it.... what helped me to stay confident is the idea that I can hear it in the bass reproduction where group delay is really bad....Because of my experience in signal design when I hear something, I can do a good job on hypothesizing what it might look like, in this particular instance its really easy to visualize a small amplitude rising to a larger amplitude, overtime, based on what I heard in those bass reproductions with bad group delay...
Wether I chose high/low group area to play tone burst, or if changed GD using xo filter....the trend was there.
I already knew about this effect along time ago though....I'm not sure if I read it or if I theorized it based on peoples experience with it.... what helped me to stay confident is the idea that I can hear it in the bass reproduction where group delay is really bad....Because of my experience in signal design when I hear something, I can do a good job on hypothesizing what it might look like, in this particular instance its really easy to visualize a small amplitude rising to a larger amplitude, overtime, based on what I heard in those bass reproductions with bad group delay...
Last edited:
And at which frequency? I mean the issue track frequency so if you are two octave under 'cutoff' ( cutoff for sake of explanation not to trigger a fight) it'll be different than 2 ocyaves over...
The frequencies are mentioned on the graphs. He's doing it on a horn with the axi driver and slowly but surly comes to the realization that this horn won't be able to cross over at 200 Hz or another ridiculously low number for the horn used right now.
Last edited:
Also not forget that a "tone burst" is never a single frequency - can't be. The rest depends on the frequency response in the band around this frequency, if not on the whole FR. Actually these burst will have pretty wide spectra, as they all seem to start rather abruptly.
Last edited:
I specified which was which, in my original post
The above left is a 4" mid vented, tuned to 70hz....the right is the Axi2050 on a 350hz horn. Which was only intended to be used to 400hz but since I have ordered a 150hz horn to be used to 300hz or lower if possible.
The above left is a 4" mid vented, tuned to 70hz....the right is the Axi2050 on a 350hz horn. Which was only intended to be used to 400hz but since I have ordered a 150hz horn to be used to 300hz or lower if possible.
Last edited:
The frequencies and delays written in the pictures seem arbitrary, it is a bit hard to follow. The steady state sine in the image is at frequency you have marked? where the group delay number comes from? Trying to get handle what is it that we are looking at.
Can you repeat the test at higher frequency? The frequencies are quite low, difficult to get good measurements without reflections.
Can you repeat the test at higher frequency? The frequencies are quite low, difficult to get good measurements without reflections.
Filter effect on transient response is a known thing...in a bass driver with really high GD the filter causing this is a high Q acoustical filter on the vent. High Q filters result on high group delay
https://people.eecs.berkeley.edu/~lewis/Ken3.pdf
When we discussed gd before we said the room dominates
The room still has worse GD than headphones so the difference can be heard.
https://people.eecs.berkeley.edu/~lewis/Ken3.pdf
When we discussed gd before we said the room dominates
The room still has worse GD than headphones so the difference can be heard.
Last edited:
Well the frequency response of a horn is actually a filter. It's exactly the same thing.
Last edited:
That may well be. It takes an impulsive/transient signal (i.e. with a rich spectral content) to hear phase distortion, I would assume.
The trend looks to reverse depending on high vs low gd
The test where gd was measured in the negative showed a type of reversal in behavior by the 2 cest, which then tapered into a lower amplitude
The test where gd was measured in the negative showed a type of reversal in behavior by the 2 cest, which then tapered into a lower amplitude
Last edited:
Camplo - since your test was of a horn, which is minimum phase. if there is GD then there is also a frequency deviation from flat, probably a peak. Both will cause the transient modifications. How do you know which it is that you hear?
Mabat - if a linear system is feed with a pure tone then only pure tones can exist at the output. There will be a transient, but its frequency content will still only be at the input frequency. In an FFT, it will have a broadened peak - the spread due to leakage - but the center tone will always be at the input frequency. The width of the peak will depend on the signals duration, i.e. the FFT window.
Mabat - if a linear system is feed with a pure tone then only pure tones can exist at the output. There will be a transient, but its frequency content will still only be at the input frequency. In an FFT, it will have a broadened peak - the spread due to leakage - but the center tone will always be at the input frequency. The width of the peak will depend on the signals duration, i.e. the FFT window.
Earl, there can never be a single frequency in any real scenario - it would have to take infinitely long time. A finite "tone burst" itself has a pretty rich spectral content. That was the core of my comment - that much more frequencies are actually at play on those pictures.
Last edited: