New REW measurements, hope knowledgeable people will help me analyse them.

[gorman]

This is just for illustration purpose of medium cary wage ones for reference:
https://www.amazon.com/IsoAcoustics-Orea-Equipment-Isolators-Bronze/dp/B07KGCRZ9V You can find similar in a package and for less money. Goal is that it transfer it to very low frequency and more contained, plane little better silicone and cuple of coins does the trick.

Got it. I'm currently using this SVS product on all three subs.

 

[ZolaIII]

Got it. I'm currently using this SVS product on all three subs.

Well seams those work judging by reviews at least so you already done that. Windows are hard to deal with.

 

[gorman]

If I remember correctly there's an option to NOT set up delay and levels with Dirac (in the PC program, not on the AVR). I could try activating that and see if, then, I can set them on the AVR (normally they're greyed out, with a message stating that Dirac handles them).
I could take note of current settings and copy them all, except for subs delays. Try that, measure, and see if things get better. It's just subs delay that I would need to act upon, right? That was alex-z suggestion, initially. I'm not sure I'm remembering correctly, though.

I remembered correctly, there is a setting like that in the program. What it does... and what use it could have... a mystery to me.
It does not set delays or levels BUT they both remain greyed out. Useless for my case.

 

[ppataki]

The AVR can do that, sure. But in manual mode. Not with Dirac running. So all this would be... experimental? Just to understand if it could be fixed that way. But then, without Dirac, I'd have a much, much harder time correcting the three subwoofers and, as I said, I don't use my HTPC to listen to music. So... the experiment would be just for science, without a practical outcome for my case. Or am I missing something?

The point of this experiment would be to see if it can be done better than DLBC
That would further prove that DLBC is doing something incorrectly
+ you will hear how your system sounds when set up properly

 

[gorman]

The point of this experiment would be to see if it can be done better than DLBC
That would further prove that DLBC is doing something incorrectly

I have taken measurements with Audyssey vs Dirac DLBC. I will post the file later. There's certainly a difference in group delay.
But it's not the only one and, looking at graphs, I think DLBC remains preferable.

I have also tried placing a piece of absorbing material in front of the center speaker, plus closing one or both curtains (kinda heavy curtains, btw). Measurements will be in the file but, to my eye, they don't seem to make the slightest difference for refractions.

+ you will hear how your system sounds when set up properly

Not sure I want that, as long as I don't have a viable way to achieve it permanently for my setup.

 

[ZolaIII]

@gorman DSP can't correct that not even by hand, you need to move it from the side wall it's the only way. At least their new CEO (Dirac) thinks much as I do that licensing fees where crazy and cut them down. As you can see there is not much of a difference and Audiseey still has a cuple of extras.

 

[gorman]

@gorman DSP can't correct that not even by hand, you need to move it from the side wall it's the only way.

Since I cannot, I'll live with it.

As you can see there is not much of a difference and Audiseey still has a cuple of extras.

I think there are differences, considerable ones, I'd say.

See FR for left speaker, var smoothing.

Dirac

Audyssey

FR for Center speaker.

Dirac

Audyssey

FR for Right speaker
Dirac

Audyssey

Going back to group delay...

This is Audyssey right speaker:

And this is Dirac right speaker

Measurements file is too big to attach here (I've done just LRC, with different combinations of curtains and the absorbing material in front of the center speaker), downloadable from here (link valid for 30 days): https://easyupload.io/u408za

 

[gorman]

@ppataki I would appreciate if you could confirm that refractions are unchanged when drawing curtains in the above measurements.

Also, I've taken a measure for LFE balancing volume at 80db, as for the other channels.
(it's in the same file above)
Distortion seems better to me but... let me know if I'm wrong.

 

[ppataki]

Distortion seems better to me but... let me know if I'm wrong.

Yes, distortion is way better indeed!

@ppataki I would appreciate if you could confirm that refractions are unchanged when drawing curtains in the above measurements.

Not sure what you mean by this
If you mean the Wavelet diagram, I have checked that for measurements 1-5 in your above file and they are all the same
The difference is with Audyssey (measurement nr. 15) the Wavelet looks way better with that (35ms delay below 80Hz instead of DLBC having 75ms delay below 200Hz)

 

[gorman]

Not sure what you mean by this

Sorry, I meant early reflections for L and C.

I thought curtains on the windows and (temporary) absorbing material in front of the center speaker could make them better. But absolutely nothing seems to change in the different measurements with different conditions (it's in the name of the single measurements).

 

[ppataki]

Sorry, I meant early reflections for L and C.

I thought curtains on the windows and (temporary) absorbing material in front of the center speaker could make them better. But absolutely nothing seems to change in the different measurements with different conditions (it's in the name of the single measurements).

ah OK I see

The method that I personally use to see reflections is to use the IR curve and check the peaks, those are all reflections

If we take a look at your center channel with and without the absorber:

The black arrows show you the reflections and the yellow highlighted peak is somewhat reduced with the help of your absorbers - I am not sure though how audible that is.....

Tip: you can even calculate the distance of those reflections by CTRL + Right mouse click + dragging your mouse from the impulse main peak:

The Waterfall diagram gives you some hint about your room's echo or ringing properties - the sooner the waterfall waves 'die out' the less ringing you have in your room
Just compare those same two Center channel measurements, you will see that with the absorber there is a tiny improvement at around 250Hz

I also have absorbers in my room (2 pcs of 60 x 20 x 240 cm of Caruso Iso Bond, unfortunately not in strategic locations, ie. first reflection points) but they do help a lot with taming the echo

 

[gorman]

ah OK I see

The method that I personally use to see reflections is to use the IR curve and check the peaks, those are all reflections

[CUT all the rest of most useful information]

Thanks a lot! Now I understand better many things. You've been most kind and helpful. Much, much appreciated.

Not sure there's much I could do with my current equipment and given the room limitations. This is probably the best I can get. The difference in refractions appears minimum to me and you said you're not sure it could even be perceptible.
The group delay thing, I'll write to Dirac and I'll report back in case they provide an answer.

PS
The only real doubt I remain with is if it would make sense to put serious bass traps behind the front L and R speakers. That's one place where I could substitute the absorbing panel with something far thicker (there would be the room to do it). But looking at all measurements, I'm not sure if they could provide better results. This is the data sheet for the bass trap I could get: https://gikacoustics.co.uk/wp-conte...-Trap-w-FlexRange-Technology-Datasheet-UK.pdf
Not sure if, given my measurements, I would benefit more from the full range option or from the one with limiter installed. Or if I would benefit at all...

 

[ppataki]

Thanks a lot! Now I understand better many things. You've been most kind and helpful. Much, much appreciated.

You are welcome!

Let us know what Dirac will answer

I don't think a bass trap would make any difference but I will let others chime in too

 

[gorman]

Let us know what Dirac will answer

I've just written to them, attaching the measurement file which clearly shows Audyssey not exhibiting the same problem.

 

[Keith_W]

Since you asked me to reply to your question about room volume and RT60, I have taken your RT60 measurement and drawn some lines on top of it.

First I calculated your Schroder freq (calculation in the diagram). Then I drew in the zones in your room. The RT60 is not interpretable below the transition freq of 496Hz because there is no reverberant field, there are room modes instead.

The RT60 target varies according to room volume and intended application. Here it is calculated to DIN 18041 with a music target. We can see that your measured RT60 is way below the target. This would create a really "dry" sound and can affect speech intelligibility. Having reverb too low takes away the sense of spatiality and immersion. It would also extend the critical distance (the point where direct sound SPL equals reflected sound SPL) by quite a bit. If you are sitting closer than the crit distance, you are in the "direct sound predominant" field as Toole calls it. Other people call it "nearfield listening" but we avoid that term because it is a misnomer.

I'll ask you a question - when you sit down and listen to music, do you have to turn it up so loud that it makes normal conversation difficult? Does your wife have to shout at you over the music so that she can be heard? This is a symptom of reduced intelligibility. I don't think this observation is published in any acoustics book, but I have noticed that I have to do this in every overtreated room I visit.

In the end, what matters is whether you like the sound. Some people prefer dry sound like that, others don't. I'm not quite a subjectivist, but I strongly believe in preference and personal taste and i'm not going to rag on you if you like it. But if it was me, I would be removing some treatment. You might like it more.

 

[gorman]

First I calculated your Schroder freq (calculation in the diagram). Then I drew in the zones in your room. The RT60 is not interpretable below the transition freq of 496Hz because there is no reverberant field, there are room modes instead.

Ok, interesting. I did a calculation for Schroeder frequency some days ago, it was not vastly different from yours (I don't remember what the exact value was. Supposedly one should not EQ past it? Or past its value times 4? In any case... I tried not using Dirac past 800Hz... sound was worse. Problem with Dirac is that you cannot equalize with a simple high frequency roll off. So if you stop at 800Hz, you keep highs... high.

The RT60 target varies according to room volume and intended application. Here it is calculated to DIN 18041 with a music target. We can see that your measured RT60 is way below the target. This would create a really "dry" sound and can affect speech intelligibility. Having reverb too low takes away the sense of spatiality and immersion. It would also extend the critical distance (the point where direct sound SPL equals reflected sound SPL) by quite a bit. If you are sitting closer than the crit distance, you are in the "direct sound predominant" field as Toole calls it. Other people call it "nearfield
listening" but we avoid that term because it is a misnomer.

What would be this distance? My ears are approximately 2.44 meters from L and R speakers, slightly less for C.

I'll ask you a question - when you sit down and listen to music, do you have to turn it up so loud that it makes normal conversation difficult? Does your wife have to shout at you over the music so that she can be heard? This is a symptom of reduced intelligibility. I don't think this observation is published in any acoustics book, but I have noticed that I have to do this in every overtreated room I visit.

Hmmm... I did several tests to judge my average listening level... with slow response, C weighted, my SPL meter more or less hovers between 75 and 82db. If I can't listen that high, I avoid turning on my main setup. When I listen there, I want to *listen* and just do that. So people talking are, quite frankly, not happening. As a sidenote, both a friend of mine and my wife, when I had them listen to my setup wanted volume higher than what I use. There's also an issue of safety for hearing. It's not that I dislike higher than 82db but I'm getting older and hearing declines by itself without further injury.

In the end, what matters is whether you like the sound. Some people prefer dry sound like that, others don't. I'm not quite a subjectivist, but I strongly believe in preference and personal taste and i'm not going to rag on you if you like it. But if it was me, I would be removing some treatment. You might like it more.

Hmmm... the only panels I could more or less easily remove are the ones behind the two front speakers. I could substitute them with bass traps with diffusers, maybe? The others are decorative, on top of being acoustic panels... and the ones on the ceiling are glued. The whole back wall, in front of L and R is a bookshelf with books.

EDIT: a huge thank you for all the help and time!

 

[Keith_W]

Ok, interesting. I did a calculation for Schroeder frequency some days ago, it was not vastly different from yours (I don't remember what the exact value was. Supposedly one should not EQ past it? Or past its value times 4? In any case... I tried not using Dirac past 800Hz... sound was worse. Problem with Dirac is that you cannot equalize with a simple high frequency roll off. So if you stop at 800Hz, you keep highs... high.

No, the relevance of the Schroder freq is that different strategies are used above and below. Problems above Schroder or transition freq: very gentle DSP if required, preferably no DSP. Or as Toole says, "broad, low Q, tone-control type equalisation". Above Schroder, room treatment is better than DSP. Below Schroder, DSP is your best bet.

This is one reason I am not a fan of Dirac - not enough user control.

What would be this distance? My ears are approximately 2.44 meters from L and R speakers, slightly less for C.

Firstly, the critical distance is frequency-dependent because speakers are more omnidirectional at low freqs and more directional at high freqs. More omni means more sound energy is thrown into the room as reflections, so the ratio of direct sound to reflected (the DRR, or Direct to Reflected Ratio) is less, so the critical distance is shorter. More directional is the opposite, the DRR is higher, so the crit distance is further away.

Smaller rooms have a shorter crit distance because reflections arrive earlier. Room treatment lengthens the crit distance because you remove reflected sound by converting sound to heat.

Remember that the critical distance is where direct sound SPL equals reflected sound SPL. You measure this by playing a test tone. Start by standing at a point equidistant between the two speakers, at 0m with an SPL meter or RTA. Then walk backwards in a straight line. You will observe the SPL begin to drop. When the SPL stops dropping, this is where direct SPL = reflected SPL. This is the crit distance for that particular frequency. Obviously this has no meaning below a certain wavelength because you get room modes instead of reverberant fields.

IMO the DRR is a less appreciated aspect of speaker/listener placement that does not get much discussion on ASR.

Hmmm... I did several tests to judge my average listening level... with slow response, C weighted, my SPL meter more or less hovers between 75 and 82db. If I can't listen that high, I avoid turning on my main setup. When I listen there, I want to *listen* and just do that. So people talking are, quite frankly, not happening. As a sidenote, both a friend of mine and my wife, when I had them listen to my setup wanted volume higher than what I use. There's also an issue of safety for hearing. It's not that I dislike higher than 82db but I'm getting older and hearing declines by itself without further injury.

As I hinted in my post, overtreatment affecting clarity is a subjective phenomenon because it has to do with psychoacoustics. It works like this: overtreatment reduces subjective clarity. This is well-known, you can read about it in Toole. The rest is my theory: we turn up volume to compensate. Turning up the volume means we can't hear people speaking to us because the music is so loud that it drowns out speech.

Hmmm... the only panels I could more or less easily remove are the ones behind the two front speakers. I could substitute them with bass traps with diffusers, maybe? The others are decorative, on top of being acoustic panels... and the ones on the ceiling are glued. The whole back wall, in front of L and R is a bookshelf with books.

I keep saying that room treatment should not be installed by people who don't know what they are doing. The effect of these things is profound. Toole says that we can achieve our RT60 target in most rooms without any room treatment. He cites studies. I cite anecdotes since i'm not an acoustic scientist like him. I can tell you he's right.

Forget the bass traps with diffusers. No room treatment unless you know specifically what you are targeting, please. And remember what I said in that other thread - don't get too hung up on measurements. If you like the sound, then there is no problem. The vast majority of people in our hobby are subjectivists, and they somehow assemble decent sounding systems with no measurements. Even I am impressed by what they can achieve. Measurements are a way to get to your goal faster and more predictably, it is not an end in itself!

 

[gorman]

Forget the bass traps with diffusers. No room treatment unless you know specifically what you are targeting, please. And remember what I said in that other thread - don't get too hung up on measurements. If you like the sound, then there is no problem. The vast majority of people in our hobby are subjectivists, and they somehow assemble decent sounding systems with no measurements. Even I am impressed by what they can achieve. Measurements are a way to get to your goal faster and more predictably, it is not an end in itself!

Thank you for all the help and information. The treatment I installed was in fact targeting the room "echoing" in a clearly perceptible way, even without measurements. It did not sound good. Sounds far better now. I might have overdone it, though. I'll try to remove the panels behind the front speakers and check whether that measurement gets better. The bass traps, especially the "tuned" one, was targeting a specific problem I had. Then I got the third sub and now... bass seems decent.

I'm tempted by trying to enjoy the system a bit before. Since I've gotten the new KEFs I've spent far more time measuring than listening to music... :-/

 

[gorman]

Forget the bass traps with diffusers.

One question regarding measurements if I try removing the absorbers behind the front speakers.

If I remove them and see RT60 values getting closer to suggested values, will I see that improvement when I rerun Dirac? Or is RT60 part of what Dirac tries to "correct"? If you know or anyone knows that, that is.

 

[Keith_W]

DSP can not correct reverberation.