[Sisyphos]

Recently I discovered a new way to improve the sound of my speaker setup.

It aims at the fact that each room has its own acoustical traits that somehow color the sound in an unforeseeable manner.
There are different ways to compensate for these colorations (see link below for an overview).
One very promising way is to measure your room acoustics and use a software to calculate a filter to balance the sound.
I tried a software called Automatic Room Correction (ARC) and it was very easy to handle and the result was really stunning.

If you want to know more about ARC and room correction in general please have a look at my review (posted several weeks ago):
www.head-fi.org/forums/showthread.php?t=251313

More information about ARC (so far only in German):
http://www.high-end-manufaktur.de/arc.html


And just the usual disclaimer because this review might sound overly positive:
I am in no way affiliated with the developers of ARC - I just stumbled across this software on a German hifi discussion board, bought it and the necessary equipment and now I am really pleased by the results and now want to know what other people think about this approach ...


Marco

 

[islewind]

I'm not familiar with the software you mention, but the room is one of the most neglected areas in achieving satisfying reproduction - and only recently have there been reasonable solutions accessible to the average enthusiast. Of course, equalization is still a bad word, and that's half the equation (room treatments can only do so much)

 

[sejarzo]

I bought a used TacT RCS2.0S on Audiogon a couple of weeks ago that arrived in less than full working condition--didn't do the correction correctly or consistently--but I will report back on the results when it comes back from TacT's inspection/repair process. My problem is an asymmetrical listening space with much more reflective surface close to the right channel compared to the left, and the frequency sweeps done with Room EQ Wizard show some substantial differences that must wreak havoc with the imaging.

Dennis Sbragion's DRC program isn't very straightforward to use. I spent a couple of days going through the 70-ish steps to do measurements/transforms and set up Foobar to use Convolver and run my speaker rig from my 0404 USB with room correction......and it sounded like a bad boom box. Something went massively wrong, and no wonder--it is a complex process to do it with that program plus others required to produce the impulse response.

I figured that the TacT processor would work better, and I am confident it will once it returns in good shape.

 

[bigshot]

Automatic room correction is great. But you should continue to adjust through listening and tweaking the equalization curve. Automatic correction can only adjust for one position and even then it's an approximation. You can improve it even more through careful listening and figuring out how to tweak it.

See ya
Steve

 

[Sisyphos]

Quote:

Originally Posted by islewind /img/forum/go_quote.gif Of course, equalization is still a bad word, and that's half the equation (room treatments can only do so much)

That's true, in the world of hifi equalization seems to be regarded as some kind of improper trick, but digital room correction software is in fact the most sophisticated way to eq and does the job quite well.

When implemented correctly, room treatments can really be the most powerful way to tune the sound of a room. But in order to achieve really good results a lot of efforts are necessary. Just have a look at how the control rooms of professional studios are built - that's perfect but almost unachievable under normal home conditions.

 

[Sisyphos]

Quote:

Originally Posted by sejarzo /img/forum/go_quote.gif I bought a used TacT RCS2.0S on Audiogon a couple of weeks ago that arrived in less than full working condition--didn't do the correction correctly or consistently--but I will report back on the results when it comes back from TacT's inspection/repair process. Dennis Sbragion's DRC program isn't very straightforward to use. I figured that the TacT processor would work better, and I am confident it will once it returns in good shape.

The ARC software I reported about is in fact built upon Sbragion's DRC but it is much easier to use. Before I discovered ARC I tried to get DRC working but I just did not have the time to read all those guides and acquire all the additional software that's necessary. And if you have the necessary hardware (a soundcard with proper mic in and a measurement microphone) ARC is really a 'one click solution'.

Some ARC users that are also familiar with the TacT solutions reported on a German hifi board that the ARC results were even more convincing than those that they had reached with their TacT.
So it seems to me that ARC is at least a much cheaper alternative to the TacT.

 

[Sisyphos]

Quote:

Originally Posted by bigshot /img/forum/go_quote.gif Automatic room correction is great. But you should continue to adjust through listening and tweaking the equalization curve. Automatic correction can only adjust for one position and even then it's an approximation. You can improve it even more through careful listening and figuring out how to tweak it.

First I found the results of ARC a bit too bass shy. Fortunately the software gives you the opportunity to change the desired value and to recalculate the FIR based on the last measurement. So I set a value of +2db between 20 and 120 Hz and now enjoy this slight 'bass boost' without having any problems with room modes ...

Of course it's true that you can only adjust for one position, but if you perform different measurements for different listening positions (as I did) you can have FIR for each of your favorite listening spots
.

 

[robm321]

I use CARA. It's a CAD program where you have to put all of the dimensions and materials in (lots of work), but once you do it will basically tell you how to best tweak your room without having to use EQing. It still takes some trial and error, but...

 

[bigshot]

Quote:

Originally Posted by Sisyphos /img/forum/go_quote.gif Of course it's true that you can only adjust for one position, but if you perform different measurements for different listening positions (as I did) you can have FIR for each of your favorite listening spots .

What happens if you have friends over?

It's not about finding the correct spot in the room. It's about making it sound good in the whole room.

See ya
Steve

 

[Sisyphos]

Quote:

Originally Posted by bigshot /img/forum/go_quote.gif What happens if you have friends over? It's not about finding the correct spot in the room. It's about making it sound good in the whole room. See ya Steve

Hi Steve,

Just run some test tones and walk through your room.
You will experience that it can sound very different even if you move only 3 or 4 feet, particularly in the bass frequencies.
Therefore it's a difficult task to make it sound good in the whole room. Probably only heavy room treatment can make this possible.

But even then there is another thing to consider: speakers are not omnidirectional. This means that if you are listening off axis you will never experience perfect sound - no matter how much you tune your system.

However if you cure problems like the specific room modes of your room (by using something like ARC or a parametric EQ) it will sound much better in the whole room because there is no more overly boomy bass.

In order to achieve the best sound when my friends chill out on my sofa I performed a measurement with the mic at ear level and at the middle of the sofa. This seems to be the best compromise - and the guys don't tend to be that critical when they are just lounging ...

Ciao,

Marco

 

[Sisyphos]

Quote:

Originally Posted by robm321 /img/forum/go_quote.gif I use CARA. It's a CAD program where you have to put all of the dimensions and materials in (lots of work), but once you do it will basically tell you how to best tweak your room without having to use EQing. It still takes some trial and error, but...

I also know some people who use Computer Aided Room Acoustics (CARA).
It's fine but you normally have to buy at least some (ugly) absorbers to make real use of it.

 

[sejarzo]

Quote:

Originally Posted by Sisyphos /img/forum/go_quote.gif First I found the results of ARC a bit too bass shy. Fortunately the software gives you the opportunity to change the desired value and to recalculate the FIR based on the last measurement. So I set a value of +2db between 20 and 120 Hz and now enjoy this slight 'bass boost' without having any problems with room modes ... Of course it's true that you can only adjust for one position, but if you perform different measurements for different listening positions (as I did) you can have FIR for each of your favorite listening spots .

The big question is "What target response are you seeking?" The TacT system computes nine different corrections with various bass boost and hi roll-off so you can get an idea of what works right for you. Indeed, a flat system does sound "flat" (meaning, sort of lifeless/cold), because of the non-linear response of the ear and the fact that recordings are mastered to sound "right" to someone, somewhere, who doesn't have a flat room either.

The newest TacT gear uses the Fletcher-Munson data to adjust dynamically based on playback volume. On the other hand, there have been some studies after Fletcher-Munson that show a different population didn't agree very well with the F-M curves.

No doubt, ARC is cheaper. OTOH, you can get a used TacT RCS 2.0s for $1200 or so that includes a DAC and preamp, so it's not that large of an investment.

Now, I will say the treatments in this room have made a world of difference, but some problems remain. And I don't find the absorbers all that ugly, if I might say so myself!

EQ got a bad rep when it was implemented with 15 or 31 analog filters per channel, which cause problems that (and don't ask me to explain why, I am taking the word of EE's for it) don't exist in DSP implementations.

 

[sejarzo]

Quote:

Originally Posted by Sisyphos /img/forum/go_quote.gif But even then there is another thing to consider: speakers are not omnidirectional. This means that if you are listening off axis you will never experience perfect sound - no matter how much you tune your system.

Interestingly, TacT suggests that the user investigate less toe-in of the speakers at some point to optimize the size of the "sweet spot" post-correction. The claim is that there is a broader range of more consisent response a bit off-axis than directly on-axis for most tweeters. While it might not measure as flat per se, it's a wider window, and it's going to be corrected, anyway.

 

[bigshot]

Quote:

Originally Posted by Sisyphos /img/forum/go_quote.gif Just run some test tones and walk through your room. You will experience that it can sound very different even if you move only 3 or 4 feet, particularly in the bass frequencies. Therefore it's a difficult task to make it sound good in the whole room.

It isn't difficult at all. You just start from where the automatic equalization or your test tones tell you is flat. That's your jumping off place, not your end destination- machines are never totally perfect at this sort of thing. Then you listen carefully to a bunch of different music under a bunch of different conditions. Make minor adjustments and see how they work. It probably isn't going to take a big tweak, but with EQ, 2 or 3dB here or there can make a BIG difference. Tweak for a few weeks and you'll find a really nice place.

See ya
Steve

 

[bigshot]

Quote:

Originally Posted by sejarzo /img/forum/go_quote.gif EQ got a bad rep when it was implemented with 15 or 31 analog filters per channel, which cause problems that (and don't ask me to explain why, I am taking the word of EE's for it) don't exist in DSP implementations.

That's sales pitch. You can get to the exact same place with a 31 band equalizer. It's a lot cheaper than the fancy systems but it requires thought and careful listening to make it work properly. It isn't just pressing a button and bingo.

See ya
Steve