Feature Request Minimum Phase Step Response

[user44455555]

Hello

Is it possible to add a minimum phase step response too ?. there is minimal Phase IR possible but this does not show comparable results on coaxial speakers (as the JBL 104 BT) is or without near field measure. step response is not so sensible of room reverb and it is enough to window to 800 µsec. when there is a minimum phase step response possible then the precision of speaker can better compare.

see screenshots. in sterep width the kali LP is the smallest and the JBL is the widest. but the JBL have much diffrence in minimal Phase IR.

there is explain ITD or better single-cycle interaural phase differences (IPDs) more exact how it work on ears and brain. there are a hear example for headphones. I mean this

B)

Example 2: same as in (A) but with a 1800 IPD for the sine signal. In spite of the level difference between noise and signal, the sine tone is now perceivable.

i can hear this with headphones easy at low volume and can use crossfeed plugin with upto 48% Crossfeed level and still hear on headphones. but on all of my speaker i hear not even when i put my head very near between speakers so crossfeed get less with the speaker.

http://acousticslab.org/psychoacoustics/PMFiles/Module07a.htm

so there need find speakers that are precise enough in the mid to produce correct single-cycle interaural phase differences (IPDs)

Or did other have speaker and can hear the example ?. if you can hear please tell me the speaker. best is when they are cheap and use 6.6 or 8 inch woofer mid.

 

[John Mulcahy]

The step response is the integral of the windowed impulse response.

 

[user44455555]

but the step response can read more easy. there can see rise time and decay time of system. and with minimum phase i think it is as normalised frequency. so frequency influence of step response is gone. or maybe can add a high pass filter for step response display. so when set this filter to 150 hz speakers results are even more better comparable. here is the FR with much bassboost but it change not so much step response.

with step response mic calibration do not work stand in manual too. maybe mic calibration for step response can work ?.

 

[John Mulcahy]

The step response is the integral of the windowed impulse response.

Please make an effort to understand this sentence, then you may also understand it makes no sense at all to talk about the step response as though it is somehow independent of the impulse response.

 

[user44455555]

you mean that step response is not need because impulse response show the same ?
I did not know how can see in impulse response the tweeter woofer delay and the raise time or decay time, or maybe you know a page that explain this ?. I find a german page https://www.fairaudio.de/lexikon/sprungantwort/
that explain wy step response display is usefull and they do some examples. I translate with google translate

From a mathematical point of view, the integral of the impulse response over time is nothing other than the step response. This means that it provides the same information, only in a different form - and for some purposes more clearly........ Breakouts downwards (below the zero line) indicate phase errors. If the measurement record shows clearly identifiable individual mountains or peaks in the course, it can be assumed that the response of the individual chassis is delayed and therefore not optimal. You can literally see the staggered reactions of the individual loudspeaker chassis. The following graphics clearly show how the individual chassis start with a time delay (the "spikes" in the rise). A phase error is also to be assumed in the first display. The second picture suggests a three-way system - the ascent is divided into three parts:

 

[John Mulcahy]

you mean that step response is not need because impulse response show the same ?

No, of course not. I appreciate that language may be a barrier here, but you appear to make little or no effort to understand replies. Per my earlier comments, and the section of text you posted, the step response is the integral of the impulse response. That means the step response is obtained by adding up the values of the impulse response samples. It does not exist as some independent property, it comes from the impulse response. As so often seems the case with your posts, we are going around in circles, so I shall not post further on this.

 

[user44455555]

That means the step response is obtained by adding up the values of the impulse response samples. It does not exist as some independent property, it comes from the impulse response.

my understanding is this.

the step response view contain lag errors of driver and filter (phase problems) and frequency problems Right ?. The frequency is known so when normalize the frequency for the step response then it shows phase errors better. whats wrong with the thinking ?

 

[user44455555]

No, of course not. I appreciate that language may be a barrier here, but you appear to make little or no effort to understand replies. Per my earlier comments, and the section of text you posted, the step response is the integral of the impulse response. That means the step response is obtained by adding up the values of the impulse response samples. It does not exist as some independent property, it comes from the impulse response. As so often seems the case with your posts, we are going around in circles, so I shall not post further on this.

Can you please increase the limit in EQ Target settings HF Fall Slope from the maximum of currently 6 db to 24 db ?. I use the target settings to make the FR on speakers same HP 100 hz 24 db LP 1 khz 6db and press button "generete measurement from predictet". And then the step response show how fast the driver is in compare to other with this target settings see the screenshots FR look near same after EQ but step response is still much slower on kali.

this is shiftet so kali start at same point as the iloud mtm to see the rise time between kali and iloud better

 

[John Mulcahy]

Use the Speaker driver target.

 

[user44455555]

Use the Speaker driver target.

speaker driver target do not work ?. please see the measure mdat attach and screenshot what happen. it use only few eq and very wrong. maybe i have a wrong setting ?. add room curve i have disable but it create same result.

 

[John Mulcahy]

The target level is much too low. EQ is not a level control.

 

[user44455555]

The target level is much too low. EQ is not a level control.

when i have higher level i notice same problem. see screesnhot. only 2 fllters are use. the predictet is highlight in this screenshot. I use 1/6 smoothing because i notice when use 1/3 or 1/2 smoothing it correct even more not. when use less smoothing. please can you try yourself with the attached mdat if you can get it good correction ?. in this screenshot i forget to switch on allow narrow filters. i switch it on and test, but it do not work too. most eq in generic are unused

 

[John Mulcahy]

If you want to apply crossover filters to a response, do it manually by choosing them from the selection the Generic equaliser offers.

 

[user44455555]

If you want to apply crossover filters to a response, do it manually by choosing them from the selection the Generic equaliser offers.

I notice when i set LP to 3000 freq(see in screenshot) it work automatic. change to 2500 work too not. I notice also strange behaviour when do normal linear correction too. maybe you can see in the test mdat easy wy it use only few filters and create very wrong result.
for me work much better now. I notice in the general mode when scroll down in EQ Filter set page there can set special crossover filter and i can LR4 choose. when set this 2 filters to LP and HP i like it correct ok in the middle too when freq use 1000