Kali Audio IN-8 Studio Monitor Review

[QMuse]

No need to tell me. I know I can be one. Again, I get caught up in these things. We each have a right to an opinion, apologies. Mine is related to the direct quote embedded above that not eveything gets combined within the first 20 or so ms after the direct sound. If you feel different, then that's your right.

This is a 2nd time you're apologising for misbehaving. I suggest you take some time and work on yourself as such behaviour can quickly lead you into unnecessary trouble.

 

[thewas]

Actually this question whether to correct only the direct sound above transition region or also include some reflections / sound power is imho one of the most interesting and not fully answered yet. Floyd Toole recommends the first while the experiments of his colleague Sean Olive kind of show the opposite, see for example slide 24 here http://petoindominique.fr/pdf/The S...ve Evaluation of Room Correction Products.pdf.
I am also experimenting since years with both approaches and sometimes (also depending on the target curve, measuring and EQ method, loudspeakers, music material and listening distance) I prefer more the one and then the other so I guess the truth like mostly lies somewhere in the middle. This is undermined from the fact that we perceive the tonal balance of loudspeaker with the same direct sound but different directivity differently. Thus I can only recommend everyone to test both approaches and finde the one that works best for you.

 

[Jon AA]

the fact that we perceive the tonal balance of loudspeaker with the same direct sound but different directivity differently.

Is that a fact though? Assuming both speakers have smooth directivity? Spaciousness would certainly be different, but tonal balance? That's not what Toole's research seems to show. If that was true, you'd want narrow directivity speakers to have a rising slope on axis anechoically, wider directivty speakers a falling slope, etc. Tool's research seems to indicate being flat on-axis anechoically gives the most preferred results accross a wide variety of speaker types.

 

[thewas]

Spaciousness would certainly be different, but tonal balance?

Smooth imho doesn't tell enough, I have compared few times good Hifi loudspeakers with wide and smoothly increasing directivity to good PA speakers with higher smooth CD directivity (thus increasing to the mid and then constant) and they sounded tonally different to me.
It would be interesting to have more details on that famous M2 vs. Revel comparison, to which target they were EQed and how they were perceived tonally.

 

[TimVG]

Actually this question whether to correct only the direct sound above transition region or also include some reflections / sound power is imho one of the most interesting and not fully answered yet. Floyd Toole recommends the first while the experiments of his colleague Sean Olive kind of show the opposite, see for example slide 24 here http://petoindominique.fr/pdf/The Subjective and Objective Evaluation of Room Correction Products.pdf.
I am also experimenting since years with both approaches and sometimes (also depending on the target curve, measuring and EQ method, loudspeakers, music material and listening distance) I prefer more the one and then the other so I guess the truth like mostly lies somewhere in the middle. This is undermined from the fact that we perceive the tonal balance of loudspeaker with the same direct sound but different directivity differently. Thus I can only recommend everyone to test both approaches and finde the one that works best for you.

Hi

This is of course one example using one flawed loudspeaker as a basis with the goal of testing different room correction products. We don't know whether the preferred curve was to the increase in overall low end (which acounts for about 30% of the rating) or not. One valuable comparison not performed, would have been to anechoically correct the listening window of the loudspeaker in question and correct only the area below the transition frequent to match the response of the 'preferred curve'. If Toole is correct, which I believe him to be, than that version should have beaten even the most preferred curve.

But it is speculation at this point of course. In my own experiences, multi-way loudspeakers will more often than not feature a dip somewhere in the power response simply because of vertical cancellations between the two (or more) drivers. Correcting these in the steady-state average diminishes the quality of the direct sound as also explained by @Jon AA in a previous post.

 

[thewas]

This is of course one example using one flawed loudspeaker as a basis with the goal of testing different room correction products.

Nope, the loudspeaker used was the very good JBL Monitor LSR 6332, its spinorama can be found here https://jblpro.com/en-asia/site_elements/lsr6332-spec-sheet
Its also interesting that some high end JBL synthesis EQ systems correct the full response for their own very good loudspeakers, which according to Toole they shouldn't, there is a discrepancy there between Toole and Olive and other Harman engineers.

If Toole is correct, which I believe him to be, than that version should have beaten even the most preferred curve.

I believe by now that the truth is somewhere in the middle, anyway if a loudspeaker has great directivity behaviour you see that also in the room curve (if you measure accordingly, a single sweep isn't appropriate) and if not, like Toole says, things get unpredictable.

In my own experiences, multi-way loudspeakers will more often than not feature a dip somewhere in the power response simply because of vertical cancellations between the two (or more) drivers. Correcting these in the steady-state average diminishes the quality of the direct sound as also explained by @Jon AA in a previous post.

I agree there (although my multi-way loudspeakers don't have such problems as they are coaxials ), but this rather means that correction should be done with care and not just to a "random" predefined curve. Broad corrections based on room measurements can make sense though as our perceived tonality depends also on the room reverberation characteristics and according to it the same loudspeaker can sound more or less treble heavy, which according to Toole it shouldn't.

 

[TimVG]

Nope, the loudspeaker used was the good JBL Monitor LSR 6332,

In my understanding it was the B&W 802N as on page 8 of the document you linked

 

[Jon AA]

Its also interesting that some high end JBL synthesis EQ systems correct the full response for their own very good loudspeakers, which according to Toole they shouldn't, there is a discrepancy there between Toole and Olive and other Harman engineers.

The high end JBL Synthesis systems do, in fact, have the ability and are often set up to correct the speakers full range. But, the system has data on all their speakers built into it--basically providing a speaker-specific target curve. They aren't just randomly correcting to a generic in-room response.

 

[thewas]

In my understanding it was the B&W 802N as on page 8 of the document you linked

I checked again and you could be right there, the photos of the slides showing the JBL could misleading.

On the other hand even if they used that 802N the results would mean that by correcting the sound power (which has a dip at mid/tweeter transition of those B&W as otherwise direct sound would need to be raised to fill) they got better results by "deforming" the direct sound.

Also interesting is the comment of Olive in http://seanolive.blogspot.com/2009/11/subjective-and-objective-evaluation-of.html
Since the JBL LSR6332 is already a good loudspeaker, we are mostly doing room correction at below 300 Hz.
which coincides to my experiences.

Also it is interesting that all monitors of highly regarded brands like Genelec, Neumann, JBL have high shelf filters to compensate for different room absorption which also contradicts to the just direct sound above transition theory.

 

[thewas]

The high end JBL Synthesis systems do, in fact, have the ability and are often set up to correct the speakers full range. But, the system has data on all their speakers built into it--basically providing a speaker-specific target curve. They aren't just randomly correcting to a generic in-room response.

Of course as I wrote above you shouldn't use a generic curve but one matched to the directivity and reverberation characteristics of your loudspeakers and rooms. On the other hand if only direct sound would matter, why wouldn't JBL synthesis just engineer it to be optimal so that their room correction product needs to correct only till the transition region?

 

[TimVG]

I checked again and you could be right there, the photos of the slides showing the JBL could misleading.

On the other hand even if they used that 802N the results would mean that by correcting the sound power (which has a dip at mid/tweeter transition of those B&W as otherwise direct sound would need to be raised to fill) they got better results by "deforming" the direct sound.

Also interesting is the comment of Toole in http://seanolive.blogspot.com/2009/11/subjective-and-objective-evaluation-of.html
Since the JBL LSR6332 is already a good loudspeaker, we are mostly doing room correction at below 300 Hz.
which coincides to my experiences.

Also it is interesting that all monitors of highly regarded brands like Genelec, Neumann, JBL have high shelf filters to compensate for different room absorption which also contradicts to the just direct sound above transition theory.

But it also has a (albeit shallower) dip in the on-axis/LW reponse around the same area. I'm just saying we shouldn't draw conclusions too soon based on possibly a false premise. The result after the correction would be a slight boost in that area, but in my opinion possibly offset due to the heftier low end and thus, preferred. The fairest comparison would be the preferred curve vs the same curve below the transition frequency, but above it equalized flat anechoically - but that is off the table since the test has long passed.

My original protest was against what I've seen @mitchco state more than once, namely that there is a definite in-room curve which he himself prefers in his room across al sorts of speaker systems. PS: this is not meant as an insult, or to cause offense, I am merely claiming that as @Jon AA has stated, there cannot be a single universal curve due to the various reasons named.

 

[thewas]

there cannot be a single universal curve due to the various reasons named.

I fully agree there and wrote it even above post, on the other hand I am also more critical to Tooles imho too simplifying statement that above transition frequency only the direct sound matters. I also don't think that @mitchco claims that there is one curve for all, but he has found one which works well for his room and listening distance, although its not very different to mine, but that can be a coincidence of course.
I also agree that a lot more research should be done there and I find it a bit sad that the audio industry doesn't seem to care about such important matters and just keeps "improving" their products on unimportant or untested or even wrong premises.

 

[TimVG]

I fully agree there and wrote it even above post, on the other hand I am also more critical to Tooles imho too simplifying statement that above transition frequency only the direct sound matters. I also don't think that @mitchco claims that there is one curve for all, but he has found one which works well for his room and listening distance, although its not very different to mine, but that can be a coincidence of course.
I also agree that a lot more research should be done there and I find it a bit sad that the audio industry doesn't seem to care about such important matters and just keeps "improving" their products on unimportant or untested or even wrong premises.

I'm certain he feels that the off-axis curves must be smooth (free of resonances or other irregularities) but he seems to consider the direct sound field to be of greatest importance indeed.

 

[TimVG]

That being said I feel we as a community, despite our differences from time to time, at least are asking the right questions.

 

[thewas]

That being said I feel we as a community, despite our differences from time to time, at least are asking the right questions.

I fully agree and I don't even see them as differences as I even myself have done EQ above transition frequency in the past also based just on direct sound, these are very complex issues and possibly there even can't be a generic answer for them that always is optimal, so its very interesting to see the opinions and experiences of others.

 

[TimVG]

I fully agree and I don't even see them as differences as I even myself have done EQ above transition frequency in the past also based just on direct sound, these are very complex issues and possibly there even can't be a generic answer for them that always is optimal, so its very interesting to see the opinions and experiences of others.

To put it into perspective, at gearslutz, where the creators of content are centered, one can now find questions such as whether amplifier a sounds a bit warmer than amplifier b, and if cable x is worth the cost.

 

[thewas]

To put it into perspective, at gearslutz, where the creators of content are centered, one can now find questions such as whether amplifier a sounds a bit warmer than amplifier b, and if cable x is worth the cost.

Fortunately most real professional creators (and not wannabe producer kiddies ) have no time or lust for such forums but rather spend their time and money on creating music and on new techniques and important sound creation/manipulation hardware and software.

 

[Jon AA]

On the other hand if only direct sound would matter, why wouldn't JBL synthesis just engineer it to be optimal so that their room correction product needs to correct only till the transition region?

I have no data on how often they do what, sorry. However in my personal opinion, there can be much value in EQing (based on in-room response at the MLP) well up into the "Transition Zone" particularly to attempt to compensate for boundary effects (SBIR, etc).

the same curve below the transition frequency, but above it equalized flat anechoically

That's exactly what I've been trying to do lately (but a bit higher than the actual transition frequency). I figure, based on all the research I've been able to absorb at this point, that can't be wrong. It also has the potential of actually improving your speakers (very, very, few passive speakers look like they couldn't benefit from a little EQ when you see their anechoic response) by flattening the on-axis/listening window, giving them a neutral tonal balance and fixing resonances. The DSP horsepower required is already there in many room-EQ systems, they just aren't set up to do it that way (you have to fool them into doing it).

 

[thewas]

I have no data on how often they do what, sorry. However in my personal opinion, there can be much value in EQing (based on in-room response at the MLP) well up into the "Transition Zone" particularly to attempt to compensate for boundary effects (SBIR, etc).

Yes, up to the transition zone correction is most important, Genelec GLM also mainly corrects up to 1-2 kHz.

 

[QMuse]

IMO listening window curve is an abstraction and something that exists only in anechoic measurements. As soon as you put speaker in a real room what you will hear will be according to the predicted in-room response (PIR) where LW will participate with only 12% according to the method described in CEA-2034.

In a hypothetic situation where LW is perfect but ER and SP are bad PIR would also end up bad. Is it a good speaker or not?

In another hypothetic situation we have not so good LW but perfect ER and SP so PIR ends up really good. Is this a good speaer or not?

So, which one is a better speaker according to the CEA-2034 standard?
@MZKM , which one would end up with better Olive's score?
Which one would sound better in a real room? Btw, if that is not the one with perfect PIR and not so good LW then what is the purpose of PIR???