The essence of lobing could be described as the result of the physical offset of drivers, and the delay/phase differences at different (usually vertical) angles. The result being a variation of power relative to the direct sound.which introduces a phase shift of 360 degrees around the crossover point, the system at least does not produce anything different being ON axis or OFF axis. Is my assumption incorrect?
And your messurements show a sharp phaseshift off axis that gives a big phase delay of axis
Not my ideal design
Haha...i don't think i ever showed any off-axis measurements in this thread, other than a 500Hz wavelet...
....you can tell a big off-ax phase delay from that? , LOL
Really dude? Adios.
Last edited:
If we have a perfect point source, seems to me the on and off axis phase response would be the same.EDIT: for this thought experiment, let's assume a mathematical perfect system, or maybe the next closest physical analog, a coaxial speaker system.
So if point source has a IIR LR4 on-ax phase rotation, same for off-ax.
And if linear on, linear off.
Center-to-center driver distances of course muck up off-ax phase summations, and hence off-ax mag, ....be it IIR with phase rotation or linear phase.
So I'm a believer in minimizing c2c...along with lin phase xovers😀
Get rid of all muck, acoustic design or electric filters.!!!
Why can't I measure pre-echo in my circuit? According to you, in order to identify the FIR filter on the signal path and see pre-echo, you need to apply a pulse, I did, and I do not see the FIR signal processing on the pulse.That is the expected result.
An idea came to my mind that should solve the problem of pre-ringing in FIR. In fact, as I understand it, the reason for the presence of ringing is that in the initial conditions of the FIR filter is empty, there is no data in it, and when the input signal appears, the filter needs to be completely filled with numbers, which the filter wastes each cycle, as the filter is filled with numbers, it forms this pre-ringing, and when the entire filter is filled with numbers, the ringing ends. So, it seems to me that at the software level such a problem is very easily solved, you just need to disconnect the filter output from the DAC while it is filling, and when the filter is full, connect its output to the DAC, and voila, there is no pre-ringing at the DAC output. After-ringing is not so important, and You don't have to do anything with it.
Yes, that's true, but my idea solves the problem of the crossover pre-ringing, which is easy to hear when the speakers in a multi-band speaker system are not located at the optimal distance according to their acoustic centers and the crossover frequency.isnt fir pre ringing going on continously while music is playing? With every impulse reproduced. So you cannot turn the signal off and on at will.
As I understand from the reviews, the problem is caused by the pre-ringing, which can sometimes be heard when the speaker system is not optimally designed and the FIR filter is long.
Good point.I have a feeling this isn't how it really works..?
I suspect that different cases have different shortcomings. For example, people often FIR phase-align the low-frequency band, and in this case it is very easy to make many mistakes that will eventually lead to an audible distortion of the perception of the low-frequency band, not because of filter pre-ringing, but because of incorrect phase correction. But in general, I think I should agree that with a very long filter, its "inertia" will manifest itself more clearly.
You are correct; that's not how it really works. Any change in the signal from one or more constant-amplitude, infinite-duration sinewaves, triggers the impulse response of any filter. The impulse response of a FIR filter is equal to its coefficients. The impulse response of an IIR filter must be computed.I have a feeling this isn't how it really works..?
D
Deleted member 375592
would you care to explain the graphs? what is the trade off between pre echo and ‘better inversion’ ?
cheers
Lars
cheers
Lars
I did not read into the script as it looked like a tiresome thing to do, but anyone interested please note that all MATLAB functions have online documentation on the Mathworks website.
To anyone who couldn't open the files attached by mikets42: These're MATLAB files, open with Wordpad or equivalent.
To anyone who couldn't open the files attached by mikets42: These're MATLAB files, open with Wordpad or equivalent.
D
Deleted member 375592
The graph shows the original spk+room IR as blue, The correction filter (which is indeed a reverse of the original IR) as red, and the result of correction as yellow. The closer the peak of correction is to the middle of the correcting filter, the lower ||err|| will be, if everything else is the same.would you care to explain the graphs? what is the trade off between pre echo and ‘better inversion’ ?
cheers
Lars
I apologize for the pathetic explanations, it is easier for me to do things than to explain how and why I've done it.
Let me show a less abstract example, and I hope it will help. Let's take a planar tweeter (AMT-920), and measure it in a room from 0.5m. Then, could we prefilter the signal in DSP by an FIR so that on the output it behaves as its idealized version? Yes, almost. a picture is better than a thousand words:
Here you can see that giving ample lookahead to IR inversion allows for longer but lower pre-echo (-110 dB) and some room echo dereverberation.
The last picture may also explain why Dirac is a scam. It's a lot of noise about nothing, totally ignoring the fact that an ear is not an omni.
The code is 20 lines... but I am afraid they are cryptic to the uninitiated.
hi Michael, very interesting!
I just downloaded your matlab files into a folder an ran the tst1.m. It seems to reference class called fsaf which was not included. there was a class class fsaf_room but it does not have the same methods. Are we missing the fsaf class?
cheers,
lars
I just downloaded your matlab files into a folder an ran the tst1.m. It seems to reference class called fsaf which was not included. there was a class class fsaf_room but it does not have the same methods. Are we missing the fsaf class?
cheers,
lars
D
Deleted member 375592
https://www.mathworks.com/matlabcentral/fileexchange/83363-fast-subband-adaptive-filtering-fsaf
the file specific to the AMT-920 is attached, together with the IR to be corrected.
the file specific to the AMT-920 is attached, together with the IR to be corrected.
Attachments
D
Deleted member 375592
In your example with HF speaker correction, its IR is in the rir.mat file?Let's take a planar tweeter (AMT-920), and measure it in a room from 0.5m.
thanks Michael, I can reproduce your plots now 🙂
Need to figure out what goes on - just got back from vacation so my brain is slow
Need to figure out what goes on - just got back from vacation so my brain is slow
- Home
- Loudspeakers
- Multi-Way
- Digital Signal Processing - How it affects phase and time domain