New free Room EQ plug-in for Foobar2000 from MathAudio

[vwwv]

A new free Room EQ plug-in for Foobar2000 can be downloaded from http://mathaudio.com

 

Features:

 

• Corrects deficiencies of room acoustics (multipoint compensation).
  
• Corrects acoustic imperfections of speakers.
  
• Doesn't delay the sound and, consequently, doesn't add unnatural pre-echo to the sound. Avoids the pre-ringing problem of conventional time-delaying automatic equalizers.
  
• Quells resonance peaks of frequency response while leaving the deep notches to prevent large excursions of the speaker diaphragm. Avoids the muddy sound of conventional linearizing equalizers.
  
• Allows one to choose a small amount of the room correction if the room is already acoustically treated.
  
• Supports high sample rates.
  
• Applies 64-bit signal path throughout.
  
• Works with USB measurement microphones (e.g. MiniDSP UMIK-1) or standard measurement microphones (e.g. NADY CM100 or Dayton Audio EMM-6).
  
• Supports microphone calibration files.
  
• Applies a patented method of frequency response correction.
  

Download: http://mathaudio.com/download.htm

[lazy]

I just tried this with my UMM-6 microphone, and it indeed made a huge improvement. I was a bit surprised that the flattest neutral target frequency response did not sound as good as a neutral target curve with some troughs still present, even when I played music at low enough volumes that driver excursion limits should not have been a factor. My room response between 2k and 6k is down 10 to 15 dB lower than the lowest bass trough, but flattening that out completely made the music sound tinny on the top end. I have B&W 804 speakers and tried lots of lossless music, so I can't attribute the result to bad masters.

 

What is the degree of precision in the frequency analysis and shaping, and how many EQ adjustments does that translate into from 20-20k? (1/3 octave?). Are these adjustments equally spread out over 20-20k, and if not, how are they grouped across 20-20k?

 

I have no idea what the L and R settings do for the microphone - I did not see or hear any difference with either setting. Overall, a pretty clever and effective tool!

[esldude]

Looks interesting. If it works well, the price is right.

 

Thanks for bringing this to our attention.

[vwwv]

I just tried this with my UMM-6 microphone, and it indeed made a huge improvement. I was a bit surprised that the flattest neutral target frequency response did not sound as good as a neutral target curve with some troughs still present

 

A “dead flat” frequency response is not always the best option. The optimum level of compensation may correspond to a higher position of the vertical slider. You can read a little more about it at http://mathaudio.com/room-eq.htm

 

What is the degree of precision in the frequency analysis and shaping, and how many EQ adjustments does that translate into from 20-20k? (1/3 octave?). Are these adjustments equally spread out over 20-20k, and if not, how are they grouped across 20-20k?

 

Sorry, I can’t answer all the questions It works like a WYSIWYG system: what you see is what you get. If you want to check the accuracy of the plug-in, you may want to use it to measure the frequency response of your sound card (use a short cable to connect Line In to Line Out). You should get a straight line.

 

I have no idea what the L and R settings do for the microphone - I did not see or hear any difference with either setting. Overall, a pretty clever and effective tool!

 

The L and R settings don’t do anything with the microphone. They just allow you to select your microphone input. If your microphone is connected to the left input of your sound card, select “L”.

[lazy]

I'm rather disappointed by your response to my questions to understand the limits and operating assumptions behind your product (your second group of responses above). Most of my listening is not with foobar2000, and I don't have time for games, cuteness or ambiguity where avoidable. So instead of spending $100 for a tool of unknown limitations, I just bought Acourate ($389 at current exchange rates), and they seem very forthcoming on what it does and its limits and assumptions for effective use.

[ironmine]

I just tried Math Audio as the Foobar Component (which is free, by the way) and it sounds pretty good, but it needs the function of volume adjustment as the processing results in digital clipping.

[CNoblet]

I'm having very good luck with MathAudio as an au plugin in Audirvana+. Not so much luck with Pure Music 2, where MathAudio resets to measurement mode between albums, thus muting the player. But it sure is nice to have room EQ with Audirvana+, and it doesn't alter the character of the A+ "sound." When setting up MathAudio, I used the same XTZ mike and placements as I had with Amarra Symphony w/iRC. I don't believe MathAudio provides the kind of sophisticated impulse response correction that iRC/Dirac does, but for the price it seems to me to be a bargain.

[fanohrde]

it needs the function of volume adjustment as the processing results in digital clipping.

[vwwv]

it needs the function of volume adjustment as the processing results in digital clipping.

The clipping problem is already fixed. Please download the latest version of Room EQ. By the way, the cliping was possible in old versions only with sinusoidal signals of maximum amplitude. Conventional music didn't cause the clipping issues.

[prerich]

Compared MathAudio to Audyssey MultiEQ on a TX-NR807 receiver...MathAudio gave me better results every single time. Audyssey seemed to suck everything away (depth, height - just squished it), Audyssey also seemed to exaggerate the highs. MathAudio Room EQ used with the neutral setting was very clear, and improved the musical soundstage and imaging. I'm using the paid version of MathAudio as a VST plugin in JRiver. Now if they only did an 8 channel version !!!!!

[ironmine]

So far, I like most of all the Convolver (Foobar component) when I use it with the filters made in the Room EQ Wizard. I convolve the signal after it is upsampled to 176.4 or 192 depending on the original signal. The frequency of the REW filters must be the same as the signal.

[ironmine]

The clipping problem is already fixed. Please download the latest version of Room EQ. By the way, the cliping was possible in old versions only with sinusoidal signals of maximum amplitude. Conventional music didn't cause the clipping issues.

 

No, it's not fixed. There is still digital clipping happening as a result of Math Audio processing. This software needs the Level Adjust feature.

 

Also, you are wrong about the real music not causing clipping issues.

[ironmine]

MathAudio has just released a new version of its Room EQ plugin which includes the Room EQ gain slider !!!

 

Thanks, MathAudio!