The auto EQ in the DEQ2496 is unusable. (I have one too...)
The main problem is it is a steady state measurement, which means its measuring the effect of all the room reflections which introduce narrow band peaks and dips in the response which it then attempts to "correct", which is totally the wrong thing to do. (Why its wrong to measure the room response to EQ a speaker >300Hz is a very long involved discussion way off topic for this thread...just take my word for it, or read the thousands of posts long thread http://www.diyaudio.com/forums/multi-way/172806-flat-not-correct-stereo-system.html
)It was designed for applying "house curves" in large auditorium/theatre sized rooms, but that technique simply doesn't work in living room sized rooms. So, forget about the auto EQ function, unless your speakers are terrible to begin with it always makes things worse, even with a good microphone.
I agree the DEQ2496 is a good piece of kit though, I've had one in constant use since 2004, and I do use it for correcting some driver resonances, baffle step correction, room bass response correction etc. The parametric EQ's are perfect for driver resonance correction since you have fine control of centre frequency and width/Q.
Discrete resonances should always be corrected with a PEQ, the GEQ is useless for that. What I find works best is rather than just measuring the frequency response and attempting to generate a complimentary GEQ curve, is to manually identify discrete, troublesome resonances, and correct them with individual PEQ's, and use the shelving filter (not GEQ) for BSC etc.
The less overall coefficients the better, (don't over complicate the correction, just correct what matters, not every tiny bump in the response) and usually you can do everything with a few PEQ's and shelving filters, which have much better resolution than the GEQ.
If you do use the GEQ, make sure its in "True Response" mode on the Utility page.
Last edited:
The auto EQ in the DEQ2496 is unusable. (I have one too...)
(Why its wrong to measure the room response to EQ a speaker >300Hz is a very long involved discussion way off topic for this thread...just take my word for it, or read the thousands of posts long thread http://www.diyaudio.com/forums/multi-way/172806-flat-not-correct-stereo-system.html
I don't need to take your word for it and I don't need to read anything about it.
It sounds crap. Equivalent to a $10 walkie talkie. Hahaha
Gentlemen,
I think you need to take a look at the charts in the other thread and the one provided by Fostex. It has a terrible problem at 6500. Unless you are putting it in a sealed box, no way should you be doing any bass eq. Go look at xmax. This is a 100 or in a fantastic box, 80 Hz driver.
If you are using them away from the wall, they need some amount of BSC, which I could take you are saying as bass eq, but it is upper bass, not lower. It then could use a tad of eq to bring the top end up just a tad.
We all agree, auto eq does not work. The DEQ is designed for PA systems, not living rooms. They assume there is a professional sound mix man who understands the room and can start adjusting after the auto baseline. It is a good tool, just not magic.
See what happens with about a 4 dB shelf around 300 or so, deep notch at 6500, and a bit of boost at 10K. Use your ears from then on.
Hiya, OP here (yea you know me, who's down with OP? maybe Ron Howard is, lol).
Thanks scott.
I don't mind the ultracurve discussion. It's an option for all of us with full range drivers. Can you hook the digital stream (from cd/dvd player) straight to it ?
HTGuide Forum - large speaker design candidates?
"Scan Speak/ Seas Excel are in another league, but the W4 1337 is a good value for the money. It has an airy, almost electrostatic presence that is hard to come by in a metal cone driver in this price range." (on the 1337sa)
and the quick waterfall from 1khz - 10khz. (again sa)
(click on comparison information, then tb w4-1337sa, then csd)
Zaph|Audio
To me an electrostat would mean detail, or clarity. Sometimes my wifey says sharpness.
I was this close to ordering a pair of an15's put in the huge phy open baffle.
I figure 9 x 4" is about the same area as a 12". I've been working on a focal array for maybe 7 years now. I was originally going to use 16 x w3-871s but by the time I was ready, it was discontinued. Then I had to find a builder. My last guy who built my 4 x 4" array vanished with 18 x b3s and 2 aplair12 gen 2. Now I'm blessed.
I've seen a few posts by different people saying how a small driver breaks up at a higher volume than a larger driver. But the larger driver goes louder and deeper. My buddy who is 50+ said that makes sense. He was raised on them (15's with whizzers) and backs up that they are in breakup mode losing detail starting at a lower frequency than smaller drivers. I'm also not keen on whizzers.
I tossed my 4 x 4" focused array because I moved, the 2' x 2' x 19" would be too big. I'm betting the 9 array will cover about a couch wide at 11' away (focus @ 12'). But I'm one of the few who can tolerate the massive drawback of 3 listening seats in a room. Everywhere and I mean everywhere else, it sounds like you are underwater. It is almost disorienting if you stand up. Your brain freaks out thinking that you just got a head cold or something.
Thanks Melo Theory.
And thanks melon head for going out on a limb getting the an15's and talking about them. I'm surprised more have not.
Norman
Thanks scott.
I don't mind the ultracurve discussion. It's an option for all of us with full range drivers. Can you hook the digital stream (from cd/dvd player) straight to it ?
HTGuide Forum - large speaker design candidates?
"Scan Speak/ Seas Excel are in another league, but the W4 1337 is a good value for the money. It has an airy, almost electrostatic presence that is hard to come by in a metal cone driver in this price range." (on the 1337sa)
and the quick waterfall from 1khz - 10khz. (again sa)
(click on comparison information, then tb w4-1337sa, then csd)
Zaph|Audio
To me an electrostat would mean detail, or clarity. Sometimes my wifey says sharpness.
I was this close to ordering a pair of an15's put in the huge phy open baffle.
I figure 9 x 4" is about the same area as a 12". I've been working on a focal array for maybe 7 years now. I was originally going to use 16 x w3-871s but by the time I was ready, it was discontinued. Then I had to find a builder. My last guy who built my 4 x 4" array vanished with 18 x b3s and 2 aplair12 gen 2. Now I'm blessed.
I've seen a few posts by different people saying how a small driver breaks up at a higher volume than a larger driver. But the larger driver goes louder and deeper. My buddy who is 50+ said that makes sense. He was raised on them (15's with whizzers) and backs up that they are in breakup mode losing detail starting at a lower frequency than smaller drivers. I'm also not keen on whizzers.
I tossed my 4 x 4" focused array because I moved, the 2' x 2' x 19" would be too big. I'm betting the 9 array will cover about a couch wide at 11' away (focus @ 12'). But I'm one of the few who can tolerate the massive drawback of 3 listening seats in a room. Everywhere and I mean everywhere else, it sounds like you are underwater. It is almost disorienting if you stand up. Your brain freaks out thinking that you just got a head cold or something.
Thanks Melo Theory.
And thanks melon head for going out on a limb getting the an15's and talking about them. I'm surprised more have not.
Norman
Last edited:
I am not sure of the "electrostatic sound" either. I know my wife has liked only two speakers in a store. Martin Logan's were one, 2Ce the other. One an electrostatic, the other a soft dome. She does not talk about fast, sharp, or clean, but about not causing pain from the distortion on violins and trumpets. She has accepted the Seas 27TBFC/G in the pair I am building now.
These will sound far more accurate than an electrostat.
Electrostatics along with planars are line sources, they sound "musical" because they attenuate high frequencies due to comb filtering.
These will radiate as a point source.
These will truly have headphone clarity along with an appropriate frontal soundstage, as long as room nodes and reflections are delt with.
These will actualy be more accurate than headphones because headphones themselves radiate as a line source in all directions because having an 1.5 inch driver right against your ear is like having an infinite sized driver at distance.
That's why if you look at the FR of headphones, they all have a downward response at high frequencies.
The only reason that focused arrays are not more popular is because only 1 person can be in the sweet spot, but.......how many audiophiles live in your house? The answer is usually 1 unless there is an audiophile commune somewhere
Electrostatics along with planars are line sources, they sound "musical" because they attenuate high frequencies due to comb filtering.
These will radiate as a point source.
These will truly have headphone clarity along with an appropriate frontal soundstage, as long as room nodes and reflections are delt with.
These will actualy be more accurate than headphones because headphones themselves radiate as a line source in all directions because having an 1.5 inch driver right against your ear is like having an infinite sized driver at distance.
That's why if you look at the FR of headphones, they all have a downward response at high frequencies.
The only reason that focused arrays are not more popular is because only 1 person can be in the sweet spot, but.......how many audiophiles live in your house? The answer is usually 1 unless there is an audiophile commune somewhere
In my specific setup I have dual 15s taking care of bass duties and I only listen at less than 1/8 Watt, so exceeding xmax is not an issue for me. This is all just for fun based on ra7 recommending trying it out.
So "here I am, the one that you love, asking for another day"
Last edited:
Really? How does this apply to the many examples of line sources that do not suffer from attenuated HF, either through employing a column of ribbon / planar tweeters (which effectively form a single radiating column), or by using active or passive Eq to provide a balanced FR?
I guess that some planar speakers are line sources but the original Quad Electrostatic speakers, EL57, are 3 way and are not a line source. They do not attenuate the treble and have low distortion and low stored energy and sound good. But they have some disadvantages such as a limited bass response, down to about 50 Hz depending on room. Also the treble panel is tall and thin so the horizontal dispersion is OK but the vertical listening window is limited.
Ribbon tweeters display comb filtering as much as any line source would, unless measured in the far field, or if measured in the near field...they must have a natural rising response, but they will still exhibit comb filtering due to delayed time arrivals.
Comb filtering isn't limited to driver spacing, it's mainly due to the fact that it's a line source to begin with.
Well, of coarse anything can be eq'd.
line arrays to me (and e-stats / panel) sound like a 6' tall guitar. Or they sound like a closeup of an actress's lips in a movie. Even 30' away from a 6' tall one. It had a big top sound to it.
Check out this jagged step response of a kuze flat line array (32 x 2" drivers).
from this project.
Parts Express: Project Showcase
Especially check out the unequalized array response compared to a single driver. Even equalized, you still have a signal smeared in time.
So, yea, combing is bad.
Norman
Check out this jagged step response of a kuze flat line array (32 x 2" drivers).
from this project.
Parts Express: Project Showcase
Especially check out the unequalized array response compared to a single driver. Even equalized, you still have a signal smeared in time.
So, yea, combing is bad.
Norman
Last edited:
Yup.
Ah, so you're now resorting to changing the criteria and invoking the old delay-time argument between array ends. It doesn't seem to matter how many times this is exploded, it still crops up again.
Human hearing does not function like a microphone and a frequency response graph. When working with line sources, you have to account for temporal masking. As Blauert demonstrated, and was well-known to researchers in acoustics / psychoacoustics decades earlier (notably Haas) the precedence effect dominates, and you can have significant delay times between otherwise identical signals without audible problems since the later timed signal is suppressed. See:
-Jim Griffin's paper http://www.diy-audio.narod.ru/litr/nflawp.pdf
-The Murphy Corner Line array (note the measurements once Eq'd) The Murphy Corner-Line-Array Home Page
-Roger Russell's several articles etc. My Experience with Column Systems (you can find measurements for these two arrays in the AudioXpress articles referred to) Why be so overly concerned about comb filtering in column speakers when you probably get this all the time with the speakers y
With certain exceptions, indeed it can. But you may wish to recall that you were the one claiming line sources are inaccurate because certain types can suffer HF rolloff (to claim they all do is outright twaddle & completely refuted by measrements of multiple examples, some of which I've referred to above). So a competent designer accounts for this in their design. Therefore, since the HF is now back, by your own criteria, it is now 'accurate.' Quad erat demonstrandum.
The impulse response of a line source is always going to measure in a non-linear fashion; that's inherent to the configuration. However -see above regarding its actual relevance in practice (ditto the FR).
I found that too. Until it occured to me I was too far away from it. When listening to a line source within the fresnel zone, you only localise from what is at or close to, ear height. As you move out into the Fraunhofer field though, it does become a problem. So you either need to sit closer, or build a longer array.
Or a focused array & stick with point-source of course, although the advantages of a line source are sacrificed. YMMV. Personally, I like both, albeit for different reasons. Your TB project looks to be interesting.
Last edited:
Yes scottmoose, I completely understand what you are saying, and I've read hrough all the examples that you have posted several times. I know there has been debate over this for a long time, I've read through all the forums several times. I like line sources, I've owned esl's and planar speakers. Line sources are very useful, for parties and music venues, anywhere that requires broad, deep and even coverage. For me, they are not for critical listening. Like Norman stated, they present a larger than appropriate image, (and the fresnel zone changes with frequency). which is proof that there is smear due to time incoherence, that's what I hear psychoacousticly. And this phenomena is backed up with measurements. I'm sorry I brought it up, I should have known better. This debate will never die.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.