Dynamic EQ recommendations

[spreston]

Hi All:

Thank you for helping to equalize my 2.1 channel system with the REW software; it sounds quite good at moderate to higher volumes (fantastic with classical and jazz music - getting better with popular music). However, I am troubled by the system's response at low listening volumes, where lower frequencies somehow need to be boosted.

This low volume issue is a common problem as evidenced by the ubiquitous loudness button incorporated into many receivers. The loudness button on my system just sounds muddy and has a too much gain. I have just not been able to find much useful Dynamic EQ information online and thought someone here could help out.

I would like to use the Dynamic EQ function included in the Behringer DEQ2496 to resolve this issue, but have no idea where to start with respect to baseline settings to be used at home. REW could be a great tool but it does not seem to cooperate at low volumes, presumably due to the small difference between ambient noise and testing volume. Could anyone help me get started with some advice relating to Dynamic EQ such as:

1. What are typical levels of boost (called M-Gain on the DEQ)?
2. What is a typical gain threshold in db used to initiate boost?
3. Ratio settings?
4. Mode (L6 or L12) - what do they mean?
5. What frequencies are boosted (or is the right question below what frequencies should be boosted)?
6. Other issues?

Thanks,

:p

[Wayne A. Pflughaupt]

Hi spreston,

I'm moving your thread to our BFD Forum. It's the best place for questions related to equalization and the Behringer products.

Regards,
Wayne

[spreston]

Hi Guys. Attached is some preliminary information for educational purposes relating to what is often called the Fletcher-Munson Effect or Equal Loudness Contour (helps explain why bass frequencies need to be increased with decreasing gain/volume).

BACKGROUND

The 2003 DYNAMIC BASS BOOST APPARATUS AND METHOD patent filing by MARTIN, BURLINGAME, DAVID and ROGGE of Boston Acoustics is educational, and it would be interesting to see if this is used in any modern products. I would recommend reading the whole document linked at the bottom of the page, but here is a sampling:

“The Fletcher-Munson effect is the name given to the well known fact that human perception of sound in the bass frequency band (frequencies less than about 500hz) is influenced by the level of the sound. In particular, the sensitivity of a human listener to bass frequencies decreases as volume level decreases, wile the sensitivity to higher frequencies decreases to a lesser degree. Accordingly, music played at low volume levels is perceived as lacking bass frequencies. The Fletcher-Munson effect can be compensated for by increasing the sound energy at low frequencies relative to the sound energy at higher frequencies.

To achieve such compensation, audio filter circuits have been designed to increase the bass frequency band of an electrical audio signal used to produce an audio output, relative to higher frequencies as volume level decreases. Such systems are sometimes referred to as bass circuits.

Compensation may be provided using a manual, two-level compensation scheme, such as a conventional loudness feature on a stereo, or compensation may be dynamic. Dynamic filter circuits produce sounds with bass frequencies that are automatically controlled relative to the higher frequencies as a function of the volume of the sound produced.”

Dynamic Bass Boost Curves by gain & frequency (Patent filing figure#3) graph:



Some sample figures, mentioned in this document:
- At low input signal of -40dBv - high Q (bass boost) in the 50-150 hz range
- At higher input signal of -10dBv - Q substantially decreased
- Cutoff frequency relatively constant regardless of input signal level
- Gain at higher frequencies relatively constant regardless of signal input
- Attack time of approximately a constant 2 milliseconds

BRIEF HISTORY OF FLETCHER-MUNSON EFFECT

* Fletcher-Munson Effect - documented in 1933
* Robinson and Dadson Study - updated in 1956 and became basis of ISO 226
* ISO 226:2003 - major overhaul of ISO 226 following 18 years of research

The recent overhaul of ISO 226:2003 likely explains why the loudness button on my older equipment is so unrealistic.

ISO 226:2003 Equal Loudness Contour graph:



BASS BOOST CHARTS USED IN INDUSRTY

* New Japan Radio Co chip chart (p.10)
http://pdf1.alldatasheet.com/datashe...C/NJM2706.html

* Nippon Precision Circuits chart (p. 9)
http://pdf1.alldatasheet.com/datashe.../SM5852FS.html

* Toshiba chart (p. 9)
http://pdf1.alldatasheet.com/datashe.../TC9470FN.html


KEY SOURCES

* http://www.iso.org/iso/iso_catalogue...csnumber=34222

* http://patft.uspto.gov/netacgi/nph-Parser Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.htm&r=1&f= G&l=50&s1=7171010.PN.&OS=PN/7171010&RS=PN/7171010

* http://www.wipo.int/pctdb/en/wo.jsp?...DISPLAY=STATUS

* http://en.wikipedia.org/wiki/Equal-loudness_contour

It is entirely your responsibility to be mindful of any legal protection relating to above information

[terry j]

will be watching this thread...for the life of me I've never been able to work out the dyn eq on either the dcx or the deq...great thread idea (I hope!)

[spreston]

Thanks Terry. After a few hours of tinkering with the Behringer's graphic and parametric equalizers, I found the DEQ menu to be rather familiar territory. That said, there are a few DEQ functions I am still a bit murky on, especially attack and release (anyone?).

For the first experiment, I set up a curve, which (should) dynamically boost bass as gain (volume) is lowered, with roughly the following characteristics:

- max of 10db bass boost at peak frequency and very low volumes
- peak frequency around 83hz
- positive (right) skew, where boost trails off to 0db boost around 60hz and 300hz, respectively

I don't see how to implement a right skew on the Behringer so I just overlapped two DEQs. I did this by eyeball for now but may sit down later and calculate a more precise curve (any recommendations would be welcomed). Preliminary settings follow:

......................DEQ1.......DEQ2
M-Gain.............+10.........+4
Threshold (db)....?..............?
Ratio................1:2.5.......1:2.5
Attack (ms).......2.05........2.05
Release (ms).....105.4......105.4
Mode................BP..........BP
Frequency.........82.4.........126
BW (Oct)..........1/2...........1

==> System sounds great at low volumes with this DEQ curve!

You may have noticed that the Threshold figures have not been specified above. Since I connected the Behringer via the tape-in/tape-out loop, the DEQ does not see any change in gain (volume). So for the time being, my DEQ is not fully automated (e.g. when listening at lower volumes, I manually turn on the DEQ, set the thresholds to 0db and adjust the M-Gains based on volume). I will need to rethink how to connect the Behringer so that it can read volume levels and automatically implement DEQ when volumes are low; probably will connect the Behringer between the preamp out and poweramp in and that will be resolved.

I enjoyed working on this experiment. Given that I spend a lot of time listening to music at low volumes, the project is quite useful. All comments are welcomed.

[spreston]

I reconnected the Behringer and got the Dynamic EQ (DEQ) "bass boost" working; this is a really cool option as bass boost is automatically and Dynamically adjusted with volume. System sounds very good at all volumes now!

I hope you can do the same with the Behringer Feedback Destroyer (BFD).

The Behringer seems to be made to deal with relatively high volumes (it is a professional unit after all); to knock out the "bass boost" at higher home volumes I set the Threshold setting to the extreme minimum (-60db) and bumped the Release up to about 1000 ms. However, even at medium listening volumes, the DEQ still injects a bit of unwanted "bass boost," so I think I need to boost the Release setting more (what does the release function actually do?)

......................DEQ1.......DEQ2
M-Gain.............+10.........+4
Threshold (db)...-60..........-60
Ratio................1:2.0.......1:2.0
Attack (ms).......2.05........2.05
Release (ms).....1015.1......1015.1
Mode................BP..........BP
Frequency.........82.4.........126
BW (Oct)..........1/2...........1


By the way, I reconnected the Behringer between preamp and poweramp with $1 Y cables I found at home:

Preamp=>Behringer=>Power amp
...........................=>Subwoofer

Strangely the system sounds much better (less veiled) this way than it did with the Behringer connected through the tape-in/tape-out, so there must be some problem with either the previous "expensive" Y cable (don't waste your money on cables!) or the tape monitor circuitry in my system. I will test that next week and let you know how it goes.

[JohnM]

To get a picture of what the settings are doing it might be interesting to connect the unit in a loopback to the soundcard, play a Pink Periodic Noise test signal and look at the RTA display at a few different signal levels. Make sure the RTA window is set to rectangular and see if you can get a reasonable display with averaging set to none.

I imagine the release setting will relate to the time period over which the signal level is measured for the purposes of adjusting the EQ and/or how long the unit waits after a level change before altering the EQ settings.

[sfdoddsy]

I apologize for bring this thread back from the dead, but just wondering if anyone else has played around with this. I have a DCX2496 which I use for crossovers and have just enough processing power left to one Dynamic EQ setting.

Currently I am using a 15dB shelving filter at 6dB octave centred at 60Hz with the threshold at -15dB and 1.2 ratio. This gives a slight but pleasant boost.

Would I be better off trying the more parabolic shape linked in the first graph?

[Andysu]

I find it to be funky device to tend automatically adjust sound/frequencies on films practically complex soundtracks that may have soft sounds in the centre not just speech but music and effects that will in someway get masked dramatically due to left/right or even the surrounds!

I only have one DCX2496 as of this time, and will look forward to getting a few more at some later time.

I’ll run some crude tests with passing a single output to the REW and some Spectrumlab tests to see what I can see in terms of its action on sound/frequencies.

[Andysu]

This is direct line patched centre HF

I’ve used 1/24 smoothing to make it look a bit softer.

I had to make a few db changes in the output on the second graph as the Dynamic EQ was activated and the REW couldn’t see it as clearly due to Dynamic EQ automatic action of reducing the frequency.

AVR set at 0db

Centre setat 0db

The Dynamic EQ was set at 805Hz HP -10.0db
DCX 0.0db
Centre HF +2.0db
Threshold -40.0db
Attack 3m/s
Release 105m/s

I have no idea if I the other test would work with REW that is simple frequency capture from a film?
I might have to try Spectrumlab and even that is going to look like Predator thermal-heat-vision.

[Andysu]

A few spectrumlab tests with dynamic EQ OFF and ON

This moment is from chapter 25 as commander Farrell (Kevin Costner) runs for his life so that he can prove the manhunt for Russian spy Yuri is bogus and to get David Brice (Gene Hackman) to call off the search while holding evidence that he was present at Susan Atwell (Sean Young) apartment at the time she was killed.

(Oscar winner Maurice Jarre of, Lawrence of Arabia) score adds a trilling techno tension to (No Way Out 1987) with plentifully of bass moments.

Sadly I just leaned last night that Maurice Jarre passed away early this year after a fruitful life of scoring for films and shows. His music is quite memorable on many of scores.

http://www.imdb.com/name/nm0003574/
A few spectrumlab tests with dynamic EQ OFF and ON

This moment is from chapter 25 as commander Farrell (Kevin Costner) runs for his life so that he can prove the manhunt for Russian spy Yuri is bogus and to get David Brice (Gene Hackman) to call off the search while holding evidence that he was present at Susan Atwell (Sean Young) apartment at the time she was killed.

(Oscar winner Maurice Jarre of, Lawrence of Arabia) score adds a trilling techno tension to (No Way Out 1987) with plentifully of bass moments.

Sadly I just leaned last night that Maurice Jarre passed away early this year after a fruitful life of scoring for films and shows. His music is quite memorable on many of scores.

(13 September 1924 – 28 March 2009)
http://www.imdb.com/name/nm0003574/

In all I’m using 9 Dynamic EQ modes that is the one that is on ABC and the rest on the impendent outputs LF/HF LCR for the matching modified JBL Control 5. Each LF/HF is powered separately via two Alesis RA300 for LF and two Martantz 1050 for the HF.

The more filters that are used for crossover networks the free percentage gradually reduces same with parametric EQ.

The rest of the EQ is being handed via additional EQ
Technics SH-8058 for left/right
Technics SH-8055 for centre

This is to only trim a few frequencies for possible flat response while using the SH-8055 RTA along with the Soundlab UD236 microphone.


The spectrumlab shows some differences with Dynamic EQ ON.

The Dynamic EQ was set at 805Hz HP -10.0db
DCX 0.0db
Centre HF +2.0db
Threshold -40.0db
Attack 3m/s
Release 105m/s

Additional Dynamic EQ
200Hz LP +10.0db/12db
Attack 3m/s
Release 150m/s
Threshold -30.0

[spreston]

sfdoddsy wrote:

Currently I am using a 15dB shelving filter at 6dB octave centred at 60Hz with the threshold at -15dB and 1.2 ratio. This gives a slight but pleasant boost.

Would I be better off trying the more parabolic shape linked in the first graph?

Hi SF,

Thanks for reviving this post. I really enjoyed playing around with the dynamic EQ. The dynamic EQ really does make low level listening almost as involving as higher level listening!

The scientific charts are far from scientific and the curves posted from the patent office are examples of what a curve might look like. So you will need to use the information for a rough guide then experiment based on your equipment, room and ears. Like any EQ, it is always best to err on the conservative side.

Good luck!

:p

[Andysu]

I can’t stand even bare to listen too high shrilly sounds anymore that bend my ears! It’s uncomfortable in an environment where we can tailor the sound to be somewhat tamer and controlled.

Take the opening of Alien 3 (1992) has some really neat sounds of the alien encounter egg and its face hugger crawling around the spaceship! The glass on the cryo-tube casing cracking is very high.

The graphs are not to my liking. I tried TrueRTA but couldn’t get it to do what I wanted! I even tired spectrumlab but could only stretch its frequency response to around 5 KHz.

REW looks a bit iffy I can’t find a (Peak Hold) so I had to screen capture it literal after the sound occurred.

Threshold
-30db
Gain +15db
Frequency 1 Khz to 20 Khz
HF Amplifier set accordingly within a comfortable level

[Andysu]

I did these quick in-room sound tests videos of Alien 3 (1992) centre channel discrete only taken from first edition region 2 DVD six-track Dolby stereo.

Note the differences with the dynamic EQ threshold setting.

You Tube

You Tube

You Tube

You Tube

[Andysu]

I tried to get a direct reading on the computer but the lead is stubborn so I placed the ECM up close within a few inches HF only playing.

I pressed REW to record or frequency sweep with no direct line to the AVR as it’s only listening for Alien 3 HF only.

I can see the difference as well as hearing the difference while looking at my other RTA and seeing the db difference on the SPL meter.

[Andysu]

I’ve done this few videos in-room with ECM8000 between last night and this afternoon.

Nancy Scream from season 1 Star Trek “The Man Trap”

Nancy Scream Star Trek The Man Trap LCR

You Tube

You Tube

Nancy Scream Star Trek The Man Trap LCR dynamic EQ ON

You Tube

You Tube

I’ve activated the dynamic EQ on the ABC input plus the six dynamic EQ filters on the outputs for a smoother relaxing Nancy scream!

T-Rex VS super dinosaur is JBL! So I looked at the problem how to improve on the softly spoken dialogue as well as soft Foley effects.

Jurassic Park III centre channel DME Normal level

You Tube

You Tube

Jurassic Park III centre channel DME Dolby audio compressor ON

You Tube

You Tube

Jurassic Park III centre channel DME Dolby audio compressor with dynamic EQ ON

You Tube

You Tube

[robcart65]

It is so useful