With no disrespect intended, I see no point in discussing what you have said as it is nonsense. I don't mean that to be derogatory, it's just not anything that relates to physical reality.
I've been using Thuneau's Allocator for a few years. Whenever I've messed with it, I've found the 'Phase Arbitrator' part of it does very little.
I've recently moved away from Allocator for a few reasons - it is really buggy. I can't tell you how many hours I've been annoyed by it (and all the other software/hardware you have to connect to it). The Phase Arbitrator has a large latency time that you can't turn off, and prevents you syncing things like video or mixing. Third, it is not as versatile as what I've moved on to.
I'm currently using Reaper, which is software meant for recording/mixing music. It is open source, and has a really large community of people who develop and use it. I find I have way more control, and is much easier to use. You have to setup a routing chain, and load the DSP effect module you want, and then you're off. I use ultrafunk:fx EQ. I haven't started working on linearizing the system's phase response yet.
REAPER | Audio Production Without Limits
I've recently moved away from Allocator for a few reasons - it is really buggy. I can't tell you how many hours I've been annoyed by it (and all the other software/hardware you have to connect to it). The Phase Arbitrator has a large latency time that you can't turn off, and prevents you syncing things like video or mixing. Third, it is not as versatile as what I've moved on to.
I'm currently using Reaper, which is software meant for recording/mixing music. It is open source, and has a really large community of people who develop and use it. I find I have way more control, and is much easier to use. You have to setup a routing chain, and load the DSP effect module you want, and then you're off. I use ultrafunk:fx EQ. I haven't started working on linearizing the system's phase response yet.
REAPER | Audio Production Without Limits
If we design for coherent phase throughout the system, using a FIR (or some form of phase filter), what are we compromising in the process?
I understand the idea that absolute phase coherency may not be physcho-acoustically important past a certain frequency. But I'd like to know what we're compromising in the pursuit of phase coherency. If we implement a system that has minimal magnitude effects, what else are we affecting?
I understand the idea that absolute phase coherency may not be physcho-acoustically important past a certain frequency. But I'd like to know what we're compromising in the pursuit of phase coherency. If we implement a system that has minimal magnitude effects, what else are we affecting?
you don't like my 3 AM logic ? 😀
i do see the point of framing the problem in such a way but i agree my "solution" was nonsense.
i'm gonna have to put this issue on hold because i don't think i have time to give it proper treatment now. i will need to do some reading on how FIR actually works before i can re-examine it.
ok i have a question for John ( or anybody who can answer it ).
is it possible to have a filter into which a waveform consisting of two impulses goes in and a waveform consisting of just one impulse comes out ?
is it possible to have a filter into which a waveform consisting of two impulses goes in and a waveform consisting of just one impulse comes out ?
I think John's point is there is only one impulse response. Its shape depends on the magnitude and phase response but talking about 2, 6 or 12 impulses doesn't make any sense.
i'm gonna start a new thread about this.
please don't respond to my last couple posts in this thread.
continued here:
http://www.diyaudio.com/forums/multi-way/154653-some-dsp-questions-fir.html
as John said my example was nonsense.
but the question remains unanswered.
as far as i understand what John said only is true if the combined response of the two drivers is minimum phase.
however if it is not then more questions must be answered, and that is what the new thread is for.
You are ignoring the whole concept of directivity. Because you can only "fix" one point by messing up other points. What you hear is as much these "other points" as it is the one you supposedly "fixed". So what you are "giving up" is your money, buying something that doesn't improve the sound quality - probably makes it worse - and just gives you some pretty pictures to show your friends.
Aha! maybe...
So phase is no more uniform then magnitude when looking at the polar response? It may not be as severe, I do not know, but is this along the lines? What does phase response look like on a polar graph?
I need a measurement setup! 😀
So phase is no more uniform then magnitude when looking at the polar response? It may not be as severe, I do not know, but is this along the lines? What does phase response look like on a polar graph?
I need a measurement setup! 😀
I'm sure it's no surprise to you ..but I don't think this supposition is *overall* correct.
Granted, if only one point is "correct" (..and I don't think the program goes quite that far in most instances ..looking at the manual), then it won't be *correct* at other off-axis points (..though it might come "closer" with a very good concentric design).
The problem I have however is that you are necessarily "messing up" other points off-axis.
In other words, by fixing one point - you may well be *improving* others off-axis, at least from the original design. In this respect it may well be rendering an *over-all* measurable improvement (..though again, if it's perceptual improvement or not - that I'll leave to each individual's experiments.)
On the other hand with certain designs you may well be correct that the "single point" improvement renders the points off that axis over-all worse.
What I don't think can be stated however is that "probably makes it worse".
Rather than being "down on the technology", why not give it a try for yourself? If you find that it actually results in something that is over-all worse in any respect, then by all means - please provide the results. 🙂
It has nothing to do with the phase response off axis, it has to do with using filters to correct the magnitude response at any single point. If the polar response is not uniform then any correction at one point will necessarily make it worse at others. It's not complicated, just conveniently ignored.
An all-pass filter would not affect the magnitude response, correct?
edit: I should have noted in the earlier post that I was considering modifying only the phase using an all-pass filter. This would leave the magnitude response up to other design criteria.
edit: I should have noted in the earlier post that I was considering modifying only the phase using an all-pass filter. This would leave the magnitude response up to other design criteria.
Last edited:
]I was the Chief Technician at Dunlavy Audio Labs for several years all the way through it's demise. Every speaker that went out the door was of true minimum phase calibrated to +/- 1 dB which could only be accomplished by using first order crossover network designs. There was a problem blowing tweeters and drivers. So I got started on my own research to find a solution to the problem while maintaining the loudspeaker in minimum phase. I have successfully designed and engineered a High Power True Minimum Phase Loudspeaker system with a flat excess group delay over the entire audio range which is an extremely accurate excellent measurements and capable of reproducing high sound pressure levels with out damaging low voltage transducer drivers. It can also reproduces square waves and plays music at very very loud levels.
That's great! Exactly what we are looking for over here... Any further details on how you were able to achieve that?
Pretty remarkable claims. Any data to support them? What about polar response?
- Status
- Not open for further replies.