- Jul 21, 2018
- The Neitherlands
Had the same ideaIt would be interesting though to see difference in spectrum
If there is too much pre-ringing you will recognize it. So e.g. a simple drum hit = tok may sound like whoop-tok.
Had the same idea
DeltaWave should reveal that.
Uli's track splitted into two section (1st four, 2nd four) gives this in DeltaWave:
View attachment 47706
Not sure whether looking at the spectra makes sense under this circumstances - anyhow, see both spectra here:
- the split was not perfect - DeltaWave should handle that
- while to 1st four (blue) have max. value to the negative side, the 2nd four (white) have that in opposite direction
- there are people claiming that polarty can be detected by listening
- looks like the clips in the second have are a bit longer than in the first half
View attachment 47708
- any proposal what is caused be the pre-ringing?
Indeed. Easily heard here:
EQ applied was aggressive but could be in the realm of possibility for heavy handed FIR eq of a narrow room mode.
That is another good example, but is there a way for us to look at the pre-ringing of an impulse response and predict whether music will have an audible whoop? The threshold of audibility...how much pre-ringing is allowed to precede the peak, what amplitude is allowed, etc.
That would be step 1, and step 2 would be practical rules of thumb e.g. screenshots of an app like rePhase where we say it's OK to correct one wiggle back to flat but not another because the correction will be audible.
@Krunok Now that you have removed all pre-ringing, can you hear a difference? Either
- It sounds worse because you have removed a lot of corrections
- It sounds better because you did not realize the pre-ringing was detrimental
- It sounds no better or worse (but maybe different) because you lost some corrections, but the pre-ringing had not been excessive
Don't know if this adds anything to this thread but Mathaudio Room EQ avoids pre echo (pre-ring). I really don't know if this has any effect of measurement results however Mathaudio stated that the absence or pre echo ensures the neutrality of the sound . https://mathaudio.com/room-eq.htmFor quite some time I was using minimum phase room EQ filters tha I generated mannually with rePhase, but recently I decided to apply full scale time domain correction and go for linear phase response. As a result I got pretty nice looking phase, GD and step response graphs (measured at LP, 4m away from speakers):
View attachment 47650
View attachment 47651
View attachment 47652
As a side effect I also got some pre-ringing in my filters which is visible in IR:
View attachment 47654
View attachment 47655
That pre-ringing visible before IR got me worried a little, especially as I was not able to find any relevant data on the Netabout audibility of pre-ringing artefact. And no, I cannot hear any nasty effects after I adjusted phase - subjectively I would say sound got better, but of course it's impossible to tell that without a proper blind test.
Anyway, what I did found was an IR response of Topping D10 posted on this forum:
View attachment 47657
What surprised me is how similar it looks to IR of my speakers, both in terms of shape and in terms of pre-ringing. Based on that my conclusion was that pre-ringing with my room EQ filters is very acceptable, so nothing to fear off.
What are your thoughts/comments on this?
Apparently Krunok has disappeared, but as another reference point, here are the impulse and step responses I ended up with for my recent 3 way build. I am using FIR LR2 crossovers at 300 and 2.5kHz with only 1024 taps each, and no additional FIR corrections. I do see a pre...let's say disturbance, but not a pre-ringing problem.
View attachment 67186
View attachment 67187
What kind of all pass filter does this have? What I am getting at is how severe is this preringing?The preringing is caused by an "excessive" excessphase treatment. The excessphase has an all pass behaviour, also after treatment. Thus the spectrum remains unchanged.
Here is a simulated example of a step response of 3-way speakerWhat kind of all pass filter does this have? What I am getting at is how severe is this preringing?