Kali Audio IN-8 Studio Monitor Review

[TimVG]

I'm not sure I understand context of this, can you please clarify?

We do our best to make accurate quasi/pseudo anechoic measurements in our homes/garages with modest means. It is not a perfect system but not terrible either. The main cost being time and effort.

 

[QMuse]

Yeah, Dr Toole and the gang at Harman, Paul Hales of QSC fame, etc, are known for not having any valid arguments.....

It seems that Harman engineers are not following that philosophy as C52 and M16 seem to be optimised for flat LW.

I'm quite aware of what I did. Please learn to read.

Ditch the arogance if you want to continue conversation.

 

[TimVG]

In any case my own in-room response measurements have always correlated well against my quasi/pseudo-anechoic measurements and subsequent correction. In the Genelec for example, if I were to correct merely on the in-room response I would correct the crossover dip which only shows in the vertical plane and sound power output leaving me with a bump in the direct sound field.

 

[QMuse]

We do our best to make accurate quasi/pseudo anechoic measurements in our homes/garages with modest means. It is not a perfect system but not terrible either. The main cost being time and effort.

Aha, I understand now.

I tried to do correction of my speakers based on pseudo anechoic measurements in the range upper of 900Hz (above the transition region where room doesn't count anymore) and here is how it looked:

The problem was that, while looking perfect and all done by the book, it sounded worse than when I did correction of that region based on steady state measurement.

 

[TimVG]

Aha, I understand now.

I tried to do correction of my speakers based on pseudo anechoic measurements in the range upper of 900Hz (above the transition region where room doesn't count anymore) and here is how it looked:

View attachment 53194

The problem was that, why looking perfect and all done by the book, it sounded worse than when I did correction of that region based on steady state measurement.

Which speaker was this (for reference)? It is a process of trial and error. If this occurs in my own course of action, my first response is always to look at my methodology and look for the weak links in the chain. Is my microphone calibrated? Is my sound card/output calibrated? Am I using the correct acoustical axis and how does it relate to summing in the far field? Are there artifacts in the measurements not caused by the speaker? I usually start of with -very- nearfield measurements of the individual drivers to spot any potential issues before making rougher measurements at 50cm on the acoustical axis. It can be quite a process but also rewarding.

 

[QMuse]

Which speaker was this (for reference)? It is a process of trial and error. If this occurs in my own course of action, my first response is always to look at my methodology and look for the weak links in the chain. Is my microphone calibrated? Is my sound card/output calibrated? Am I using the correct acoustical axis and how does it relate to summing in the far field? Are there artifacts in the measurements not caused by the speaker? I usually start of with -very- nearfield measurements of the individual drivers to spot any potential issues before making rougher measurements at 50cm on the acoustical axis. It can be quite a process but also rewarding.

Revel Performa3 F206. Mic is UMIK-1, calibrated. I used very mild filtering to achieve this curve, even milder thatn filters I built for Kali IN-8 as F206 was already linear. The point here is that I don't suspect anything doen wrong here - it simply didn't fit best to my own personal taste. The sound resulted from filters I built based on steady state curve sounded much better to me. And every day I tried to compare them I ended up with the same result.

 

[TimVG]

Revel Performa3 F206. Mic is UMIK-1, calibrated. I used very mild filtering to achieve this curve, even milder thatn filters I built for Kali IN-8 as F206 was already linear. The point here is that I don't suspect anything doen wrong here - it simply didn't fit best to my own personal taste. The sound resulted from filters I built based on steady state curve sounded much better to me. And every day I tried to compare them I ended up with the same result.

Since we have the same speakers, feel free to try mine if you'd ever find yourself in an experimental mood.

 

[Absolute]

Here is pure On-axis EQ (gated) measured in listening position vs raw response fixed in listening position with Audiolense.
Klipsch RP160M + two subs. Ignore the dip(s) at high frequencies as that's a result of cross-talk cancellations when measuring both speakers when the microphone is not placed perfectly in center.

Subjectively I prefer the on-axis EQ by far. Checking the listening window before and after it looks like this;

To my ears at least it seems like Dr. Toole knows what he's talking about; If possible, EQ the speaker to become better, but don't EQ it to become worse in order to make the in-room response look prettier.

 

[QMuse]

View attachment 53200

Here is pure On-axis EQ (gated) measured in listening position vs raw response fixed in listening position with Audiolense.
Klipsch RP160M + two subs. Ignore the dip(s) at high frequencies as that's a result of cross-talk cancellations when measuring both speakers when the microphone is not placed perfectly in center.

Subjectively I prefer the on-axis EQ by far. Checking the listening window before and after it looks like this;

View attachment 53204

To my ears at least it seems like Dr. Toole knows what he's talking about; If possible, EQ the speaker to become better, but don't EQ it to become worse in order to make the in-room response look prettier.

Measuring response of both speakers playing with sweep is not really a good idea. If you want to do it like that I suggest you do it with MMM RTA.

You also have to be aware that both speakaers play together up to app 80-100Hz, after that the localization starts and it is each speaker for itself, so better to measure them separetely.

 

[thewas]

To my ears at least it seems like Dr. Toole knows what he's talking about; If possible, EQ the speaker to become better, but don't EQ it to become worse in order to make the in-room response look prettier.

Am quite sure also your MMM in-room response will look quite prettier after the correction you are doing, of course overcorrection should be avoided though.

 

[Absolute]

Measuring response of both speakers playing with sweep is not really a good idea. If you want to do it like that I suggest you do it with MMM RTA.

You also have to be aware that both speakaers play together up to app 80-100Hz, after that the localization starts and it is each speaker for itself, so better to measure them separetely.

What is MMM RTA? Yesterday I used RTA listening area averaging with pink noise pn to get a better bass response around the seating area for the first time. Previously I've just used single point measurement to EQ the sub 100 hz area.

I know that you should measure speakers separately, but these measurements are really only taken to look at the bass range and to check the relative spl levels between bass and higher frequencies

I do have multi-seat measurements with the Audiolense, though. They all look very much the same, implying that Mitcho is correct in how Audiolense works.

 

[QMuse]

What is MMM RTA? Yesterday I used RTA listening area averaging with pink noise pn to get a better bass response around the seating area for the first time.

That is MMM - Moving Microphone Method using RTA mode with pink noise.

 

[thewas]

What is MMM RTA?

Just RTA with pink noise and moving the microphone in a sphere around your listening position:

 

[thewas]

@QMuse, haha posted the same link in just a distance of few seconds

 

[Absolute]

Thanks both, that's the video I was inspired by yesterday!

 

[QMuse]

Thanks both, that's the video I was inspired by yesterday!

You may also find this article about MMM interesting.

 

[Jon AA]

It seems that Harman engineers are not following that philosophy as C52 and M16 seem to be optimised for flat LW.

And now you seem to be confusing the LW with the steady state in-room curve, or changing your story. The explanation given to how the brain separates direct and some early reflections from later arriving reflections was an explanation why you don't want to correct to a steady state in-room curve (at higher frequencies) and screw up the direct sound in the process--which is what you were advocating. You seem to be the only one here not recognizing the potential problems with that. Even Mitchco, who advocates for room correction, explained the way his room correction works is much different than what you are advocating.

The minute difference for most speakers between the LW and direct on-axis (if your contention is even true) doesn't change any of that.

Ditch the arogance if you want to continue conversation.

Oh no. That would make me so sad.

 

[QMuse]

And now you seem to be confusing the LW with the steady state in-room curve, or changing your story. The explanation given to how the brain separates direct and some early reflections from later arriving reflections was an explanation why you don't want to correct to a steady state in-room curve (at higher frequencies) and screw up the direct sound in the process--which is what you were advocating.

You failed to understand the experiment where listeners were given a choice to adjust the LF and HF levels to their preference (which most of them did) ignoring the relation of PIR (calculated from LW, ER and SP) and your simplifcation of the direct sound response represented by on-axis pseudo-ancechoic measurement. Btw, once you put your speaker in the room and let it play anything you should be aware that on-axis, LW, ER and SP are not there - there is only PIR, which correlates with steady state response, and your brain, interperting the sound collected by your ears in your own way.

Oh no. That would make me so sad.

Maybe it would be polite to say I'll say I miss you, but that's simply not true.

 

[Jon AA]

You failed to understand the experiment where listeners were given a choice to adjust the LF and HF levels to their preference (which most of them did)

Ignoring these?

And the end result:

To the contrary, all of my philosophy is based upon the science that has been presented to me (I'm not claiming to be any trailblazer with new information that people should believe over the science, as many here are).

Maybe you missed it, but I have stated several times that I am in agreement with Toole that there is nothing wrong with broad, smooth, tone controls users may adjust to their preference and/or make up for program material that is lacking in some way. That is something completely different than what you are advocating--EQing to steady state measurements that include late arriving reflections. They aren't the same thing.

ignoring the relation of PIR (calculated from LW, ER and SP)

Ignoring the relation of PIR to ....? You didn't finish the thought.

and your simplifcation of the direct sound response represented by on-axis pseudo-ancechoic measurement.

You mean the exact same way Olive and Toole both did in the pictures above? Forgive me for agreeing with them. Like I said, I'm not the one to pick apart their research, contradict them or call them wrong until I have enough science of my own to back it up and feel qualified to do so.

Btw, once you put your speaker in the room and let it play anything you should be aware that on-axis, LW, ER and SP are not there - there is only PIR,

Again, this comes from the basic misunderstanding of what was explained above. The direct sound and indirect sound (even late arriving) ARE all there. Your microphone may not distinguish without specific gating techniques, but your brain will.

which correlates with steady state response, and your brain, interperting the sound collected by your ears in your own way.

If you believe your ears interpret the sound in you own way, different from what the science shows for normal human beings...OK. Upon what do you base this conclusion?

View attachment 53245

Maybe it would be polite to say I'll say I miss you, but that's simply not true.

I'm not sure what to do with that. I can't remember the 8th grade very well.

 

[Ilkless]

Sorry for the OT, but where did you get this graph? I've been looking for it for some time.