TOPIC:

HI,
i use a SHD setup with active crossover 2-way speakers. I dialed in a delay of 0.4 ms (the speaker's manufacturers recommendation) to timealign woofers and tweeters.

When i later run Dirac live Optimization i get an Impulse response latency value of 8.02.

The Dirac Live® User Manual states:
"… and the corrected impulse peak is positioned a few milliseconds later—typically around 7ms. This is the true latency of the filter introduced to the system and is needed to correct for the mixed-phase behavior of any speaker …"

Now my assumption: the closer the value gets to 7ms the less the Dirac algorythm has to alter the signal. I tried different Values in the crossover section from the SHD in 0,2ms increments and found the "best" value to be 1ms resulting in a Dirac live Impulse response latency value of 7.47.

What do you think, is my approach reasonable - or am i getting something wrong?

...here a corresponding screenshot

I haven't got the answer.

The delay suggested by the speaker manufacturer, is that for just one of the drivers so to align woofer and tweeter? 

The impulse response graph that Dirac shows you I always thought was a bit like the overall delay to get the impulse of both L and R speakers time aligned. I didn't think it was showing us individual driver alignments because the graph is for the entire channel. I assumed dirac was doing frequency alignment in the background somewhere and it was showing the result of their frequency time alignments plus overall speaker delay.

I'd be interested in reading some other thoughts on this one. 

Duplicate removed

thank you,

yes you*re right the tweeters are delayed 0.4ms to the woofers.

And yes, a good point, i accidently forgot that the Dirac processing takes only 2 channels into account, my effort seems to be worthless,
even if there is a 'significant' change in the result (7.5ms vs 8ms)

Maybe i have to find another way to measure if 0.4ms is the absolute best value. Maybe i can create a signal via REW that sends an impulse containing
eg. 500Hz for the woofer and 2000Hz for the tweeter. Could such a measurement show if the delay is 'on point' ?