"Official" Audyssey thread (FAQ in post #1) - Page 2031

Yes, Dynamic EQ might be implemented as a multiband compressor.
Edited by urwi - 3/20/13 at 3:47am

I'd guess that a LOT of work goes into crafting it so that the "adaptation" is NOT heard. I remember pumping in the early NR schemes ...

Jeff

Well, maybe that's the rub? IMO, what it does sorta kinda of fits the definition of compression, but the technology doesn't compress everything into a smaller dynamic range. And I'd bet that the people that worked on DEQ and it's competitors would give a big eye roll , acknowledge the similarity and then move on.

Jeff

 

It isn't working all that well (the crafting) if Roger and other 'careful listener' can hear it working, is it?

Nor, do most people have steel eardrums that would be useful to listen at such loud volumes.
Do you really listen to "live" & 'reference" levels in your home. That's pretty loud dude.

 

Yes. Well not at Reference, as I said. -5dB. That is loud and I have the subjective feeling that it seems to be about as loud as when I go to a movie theatre (which is of course 0dB). In a good, properly calibrated system, in a treated room, with plenty of headroom and little distortion, these levels do not 'sound loud'. So 'steel eardrums' aren't needed, fortunately :)  What do you do when you go to a movie theatre?  You can't ask them to 'turn it down' so perhaps you use musician's earplugs? (Serious question, I am not having a go at you).  EDIT: ITYF that quite a few people in this thread enjoy systems that play cleanly at or close to Reference levels - I don't think I am all that unusual ;)

 

If my system sounded 'excruciatingly loud' (that's where I think I’d need 'steel eardrums) at these levels, I'd be fairly certain something was wrong. Assuming that none of the equipment was faulty, or being driven beyond its designed parameters, the first suspect would be the room and untamed reflections.

 

For music, where there are no standards, I try to listen at what is my recollection of similar music played live. I mostly listen to Jazz and Blues, Soul etc. I have plenty of experience at listening to this sort of music live and I just try to approximate that at home. It is subjective - I have never taken an SPL meter to a gig :)  To me, the point of 'hi-fidelity' is to recreate the original sound, or at least get as close to that goal as possible (As Quad used to say ; "the closest approach to the original sound"). I realise others have different objectives of course, but that is mine.

Edited by kbarnes701 - 3/20/13 at 6:15am

Yes, I was kidding with the "steel eardrums" remark .
I've seen many people post about listening at reference and I was curious. My system in my room is just too loud for me at reference. That being said I probably shouldn't make any comments about volume as my hearing is not the best due to many loud concerts when I was a kid.........way, way long ago.

Well last night I adjusted my AVR (it has a setting to increase the volume per input to maintain equa-volume when changing inputs) back to "0" for the volume "boost" when watching Cable and watched a few shows at -1db and noticed an improvement (to my ears) though it was still louder then before. My better half complained to 'turn it down' but I pretended to.

I know for most True HD movies from my Popcorn Hour I'll be able to listen at 0db but DTSMA movies will be something to test (when the better half is out of the house)!.

Spot on, actually. I don't know who you are but all 5 of your posts so far have been, Cheers.

Not a chance, not "instrument" but rather that instrument's frequency range. DynamicEQ is simply an algorithm that does two things, applies your good 'ole loudness curve, that is the EQ part of the "tier" or whatever. The "Dynamic" part is that it accentuates this loudness curve the further from reference you get as you turn down the volume knob on your AVR. listening at -5dB, there is little additional boost that dynEq is applying to the low end, but at -20 it is bringing up the low end much much more to still attempt to keep those freq's audible.

Think of it using a pipe organ, which is easy to get REALLY low notes on, down to 16hz to be exact, but lets just say the pipe organ is playing a 20hz fundamental, and a 1000hz fundamental at the same time with your AVR's volume at say -20dB. With DynamicEQ on, the spl difference between the two is going to be less (Call it compression if you like, but I think we need to look for a more useful term) as DynEQ is boosting the 20hz tone/note quite a bit to still make sure you are able to hear it, but needs not boost the 100hz tone at all. Now lets crank the volume to say -5 on the AVR. Measured you will see the 1000hz note has (hopefully) increased exactly 15dB's according to the 15dB increase on the volume knob. Now you will see the 20hz tone has NOT increased the full 15dB as now that you are closer to "reference" that note doesn't need that much correction as it did at 20hz to make it still audible. I will pull these numbers as merely hypothetical, but it should maybe explain it a little better at each AVR volume level and corresponding SPL level:


WITH Dynamic EQ on
AVR: 0.0 -5 -20
20hz: 85dB 81dB 75dB
1khz : 85dB 80dB 65dB

WIth Dynmic EQ off
AVR: 0.0 -5 -20
20hz: 85dB 80dB 65dB
1khz : 85dB 80dB 65dB

This is all assuming the organist was playing the same fundamental notes at the same level to start with btw.



Yep, I did, and no, most theaters don't play at reference to begin with as most people still, even in a commercial environment find it too loud.

 

Well that isn't what Roger says. He says: "If the horn is loud and tends to mask the piano, then the process will adapt based on the piano. When the horn ceases, the process will adapt to the piano."  I suspect he meant 'horn' when he said the first 'piano' BICBW.  I dont argue with Roger, ever ;)

 

 

 

According to those who claim to know, it does more than that. (THX LP does what you describe). Apparently, DEQ 'listens ahead' and when it spots loud and soft passages 'coming up' it adjusts for them on the fly, regardless of the setting of the MV. IOW, it adjusts for the input level as well as for the output level. The fans say that this is the 'dynamic' part. IDK if this is true or not but it is being touted as the huge difference between DEQ and the other forms of EQ which purport to do the same thing.

 

I actually don't really care what it does or how it works because, for me, it doesn't actually do very much at all because of the levels I habitually listen at. But it's interesting to read all the different theories, for that is all they are.

 

I permanently damaged my hearing at a King Crimson gig many years ago, before there were controls on the SPLs allowed. The 2 hour gig gave me tinnitus in my right ear and I have had it ever since of course. Fortunately, I am usually able to 'tune it out' so it doesn't bother me too much, except when I am very tired and then I seem to find it difficult to tune it out.

I can't remember the last time I heard some one mention King Crimson. Unfortunately you must be damn near as old as i am....

(sticks fingers in ears) la la la la la

So the AVR always plays the sound at a delay when DEQ is on? How else can it read ahead?
Fraction of a second delay is still a delay, and I'd have never thought an AVR would do such a thing.(unless you manually set an audio delay which i think can be set upto 300ms on the Denon)

Some of their catalog recently became available in high resolution multichannel audio.

Jeff

With any kind of digital precessing, there are delays introduced. This applies to the audio and video streams.

Jeff

I think that somehow parts of "Dynamic Volume's" characteristics have been tossed in to the DynEQ theories, in Roger's case as well perhaps. Because overall compression of the signal is EXACTLY what dynamic volume does, but by no means is that what Dynamic EQ does. I really need to dump some more graphs on here to really show EXACTLY what dynamicEQ does. Urwi did that, with plain loopback levels and no one really took it for what it was worth, but I can perhaps get some on here that shows exactly what dynamicEQ does engaged on my system. Simply put, if dynamicEQ was bringing the levels of the piano and the horn closer together to make them each stand out a little better, that is the definition of dynamic COMPRESSION, the antithesis of true dynamics...Maybe if this is true, Audyssey should have called "CompressionEQ"

I also agree that there seems to be some conflation of Dynamic Volume and EQ here. For example, I always though it was Dynamic Volume that is doing the "listening ahead" and monitoring the fluctuations in loudness a few milliseconds in advance to predict when to turn down/up the volume. Dynamic EQ just needs to know the input signal level at any point in time, I don't see why it needs to "listen ahead".

Well, again as with the prior discussion I think there's a large semantics issue here. The root problem seems to be the negative connotation with the word "compression". If you don't view it as inherently negative it's not so scary -- and I think your description of it as compensating for the "expansion" of our hearing at lower volumes to make things *sound* normal again is a very elegant way of expressing it.

As you note (and beastaudio mentions above) it's not a "compressor" in the way that something like Dynamic Volume works to squash dynamic range, but there is some compression at certain frequencies to maintain the perception of proper tonal balance. I don't think it's a bad thing as it makes my system sound better

Seems to me all this concern about DEQ "distorting" (for want of a better word) the artist's/musician's original presentation, particularly if using examples of a horn or piano, is overblown. We all know that DEQ really affects only the lower bass and, to a much lesser degree, I thought, the high upper frequencies. Most of what you hear from horns (continuing with that example) IMO isn't in those frequency ranges, nor is most of what you normally hear from a piano. I'm not saying ALL but most. It seems like many of the recent posts on this subject seem to be indicating that everything they hear from a recording using DEQ will be greatly affected when (again IMO) the greatest portion of what you hear just WON'T be. The greatest difference I hear using DEQ is only in the lowest bass region mostly well below 100Hz (and my SVS PB13 Ultra reproduces well below 20Hz) and I listen to most music and movies at around -15 to -12 on the MV. When I have friends over for dinner and just want background music turning my AVR to around -30 to -35 makes a noticeable difference again only in the lower bass and not in all the instruments. Granted I'm not a musician nor a sound engineer but DEQ just doesn't affect enough frequencies to distort the original recording, at least IMO. (Though I will admit listening to just music in true multichannel I often don't appreciate the boost to my side surrounds that DEQ does. But that is another subject entirely.)

Anyway bottom line is IMO we're getting too far out there with this concern over DEQ distorting or compressing sound. Frankly that sort of thing is more a function of Dynamic Volume which we all know is different than DEQ.

Just my two cents.

Edit to add: Looks like two others above me are saying some of the same things and beat me to it.

 

hehehe - now you definitely can't hear it working :)

 

:)  Their albums still sell well so I am told... the thing about being old is that it is a damn sight better than the alternative ;)

 

To play it at realistic levels, you need a system that can sustain 120dB for 2 hours. Well, that's what they apparently did when I saw then :(

 

I agree with you. You need to ask Feri - he's the one who knows how it works. And IIRC he says it does not compress. 

That is exactly what I - and I'd bet Feri - take issue with: Making it "compression" carries a particular connotation and seems a bit agenda-driven.

Jeff

I think it is "getting" old that beats the alternative. "Being" old isn't better than the "alternative."

Jeff

 

 

Yes, there may be some conflation. Nonetheless, if it doesn't 'know' the input level has changed and all it does is adjust the EQ based on the position of the MV (the output level), and vary the EQ amount, of course, by reference to how far the MV is moved from 0dB (the 'dynamic' bit in my feeble understanding of it), then all it does is what THX LP does. I mentioned that in a poorly worded post a day or so ago and sparked off this entire heated debate (unfortunately). It HAS to do more than what I just said here, or it is the same as THX LP. And those who say they 'know' say it isn't like THX LP at all. Harrison gave the most clear explanation IMO of how it is alleged to work. He said this (and Feri agreed with him (my bolding):

 

"As THX explains, you turn down the volume and they dial in a fixed EQ curve that provides tier one (I like that characterization btw). That would be all one needs if every track had just one amplitude from beginning to end. With Audyssey, you get that first tier correction and on top of that you get more or less EQ adjustment in accordance with the equal loudness contours as the amplitude of the track content rises and falls. For instance, a bass drum hit that is at the reference level would be corrected via tier one at your listening level (let's say -20db) correctly by THX or Audyssey. Another softer hit of the same bass drum that is 10 or 20db below reference level would not have extra correction in the THX implementation but would have more EQ in accordance with the equal loudness contour at that -10 or -20db level in the Audyssey implementation. I.e., with the THX implementation you hear the bass less as the amplitude of the track drops".

 

Kal chimed in with:

 

"My recollection about Audyssey's DEQ is that they were the first, and maybe still the only company that adjusts the EQ every millisecond or two as the level of the content varies. That surely deserves to be called "dynamic.""

 

"And that is what makes me leery of it.  If the quieter notes are more EQ-ed to sound as if they are less quiet while the louder notes are less EQ-ed, that is a musical distortion, imho.  Linking the correction to overall level setting makes more sense even if there is the capability to do "dynamic" EQ."

Edited by kbarnes701 - 3/20/13 at 2:59pm

I suppose that could happen if you pass out inside a bass bin.

And you play movies at - 5 db from reference with no RLO?

Good thing you never listened to Mr. Bungle.....
http://en.wikipedia.org/wiki/Mr._Bungle

Exactly my point and seems others are starting to see the light.

On another note, the compression thing SHOULD carry a bad connotation as speaker compression is VEWY VEWY bad, signal compression is obviously not desirable, from the main source, and often times you can get either of both of these types of REAL compression in play. Let us also not forget that many don't have those capable systems that can and often do distort/compress on their own, regardless of DynamicEQ or anything else for that matter. I know my first theater setup years ago would do some seriously bad things when I wanted to lay the coal to it, and I vowed to not stop working on my HT until I got to a point where my threshold was within the range of my speakers, for the entire spectrum, and I am here now. Even if I want to venture into +5dB territory for movies OR music, which I rarely do and if so it is for a few seconds, I know my system can handle it...

Anyways, for this discussion we are discussing an intentional compression of sorts, or something I like to call "Signal Shaping!!" There, see? I fixed it!