The most neutral monitors for mix?
Soundhaspriority
there any opinion as to candidates?
Please don't sidetrack onto the premise. I understand it's dubious, but it
would nevertheless help me personally to hear your opinions.
Bob Morein
(310) 237-6511
Mike Rivers
> Pretending for a moment that there is such a thing as a neutral monitor, are
> there any opinion as to candidates?
Have you looked at ATC?
--
If you e-mail me and it bounces, use your secret decoder ring and reach
me here:
double-m-eleven-double-zero at yahoo -- I'm really Mike Rivers
(mriv...@d-and-d.com)
drichard
Full size or nearfield? On the mastering boards I hear names like
Lipinski, Duntech, Dunleavy, ATC, but those are not nearfields. I have
no experience with any of them, but they enjoy good reputations.
Dean
Scott Dorsey
>
>Full size or nearfield? On the mastering boards I hear names like
>Lipinski, Duntech, Dunleavy, ATC, but those are not nearfields. I have
>no experience with any of them, but they enjoy good reputations.
Nearfield monitoring is sort of a cheesy compromise that has become
increasingly popular as people have had to work in poorly treated control
rooms. For the most part, you aren't going to get great imaging with a
nearfield system. And when you do, you will get it only in a small place.
I'm still a fan of big planars myself, but they're going to require more
space than the monkey boxes.
A good compromise are bass cabinets with ribbon tweeters, as used by
the Stage Accompany monitors and the big Griffins.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Ethan Winer
> Pretending for a moment that there is such a thing as a neutral monitor,
> are there any opinion as to candidates?
Mackie haters don't like to hear this, but Mackie 824s are about as flat as
it gets.
> Please don't sidetrack onto the premise.
But I must. :->)
Loudspeakers vary very little throughout their useful range compared to what
a room does with no bass traps and other acoustic treatment. Would you buy a
speaker whose response has half a dozen peak/null pairs spanning 20 to 35 dB
all below 200 Hz? That's what most rooms are like. I'd rather have a pair of
$80 Behringer Truth passives in a well treated room than Lipinskis in the
typical untreated bedroom studio.
--Ethan
Badmuts
Flat as a ruler on my spectrum analyzer in my new-and-treated room, AND to
my ear. Mixes translate well.
Bm
Scott Dorsey
>Bob,
>
>> Pretending for a moment that there is such a thing as a neutral monitor,
>> are there any opinion as to candidates?
>
>Mackie haters don't like to hear this, but Mackie 824s are about as flat as
>it gets.
They do measure pretty damn flat on-axis in a chamber. They have some other
issues, and they don't have deep bass extension, but they are still a hell of
a deal for what they are. They are way too forward for my personal taste,
but they are good enough that you can hear what is going on, and that's
saying a lot at that price point.
It was very interesting to listen to the Hafler monitor next to the Mackie...
the Hafler definitely had a very constricted and nasal sound in comparison.
>Loudspeakers vary very little throughout their useful range compared to what
>a room does with no bass traps and other acoustic treatment. Would you buy a
>speaker whose response has half a dozen peak/null pairs spanning 20 to 35 dB
>all below 200 Hz? That's what most rooms are like. I'd rather have a pair of
>$80 Behringer Truth passives in a well treated room than Lipinskis in the
>typical untreated bedroom studio.
This is true, but presumably anyone buying the Lipinskis has already spent
the money for the room.
This statement, by the way, can easily be borne out by listening to any of
the high end monitors on a trade show floor, where they all invariably sound
pretty dreadful no matter how good they really are.
Soundhaspriority
Bob Morein
(310) 237-6511
"drichard" <DRic...@wi.rr.com> wrote in message
news:1f8e1be8-5dda-44e2...@a70g2000hsh.googlegroups.com...
Soundhaspriority
"Ethan Winer" <ethanw at ethanwiner dot com> wrote in message
news:Fo6dnbhuDc5cfavV...@giganews.com...
Plinius SA100MKIII class A amp. The room is semi-treated, the speakers
spanning the room diagonal. There is Sonex, and a heavy carpet and underpad.
On prerecorded commercial material, the Thiels sound marvelous. And one oper
singer who heard the speakers reproduce my recording of her pronounced them
very fine. That is my opinion, as well. The problem that sparked my request
has to do with the recording of a guitar.
One kind individual, who also participates in this group, and is
accomplished in the music field (and who I will not further identify, though
that person is free to do so) remarked that a recording of guitar had little
in the way of lows and highs. I know a little about that individual's hifi
equipment; it's decent, but not at the level of my setup. This raises the
question: By mixing to the Thiels, am I creating a mix that cannot be
enjoyed on a modestly competent hifi system?
The Thiel CS2.3 is a three-way first order crossover time aligned
floorstanding speaker, which has three metal drivers, and an original cost
of around $4K per pair. In the hifi marketplace, they considered to have
enduring value. But they sound different from my other speakers, because all
speakers sound different. They are brighter than my Kef Reference III's,
which have fabric domes, and are less optimally placed in the room.
Bob Morein
(310) 237-6511
Soundhaspriority
"Scott Dorsey" <klu...@panix.com> wrote in message
news:g16mod$4fc$1...@panix2.panix.com...
Bob Morein
(310) 237-6511
Soundhaspriority
"Mike Rivers" <mri...@d-and-d.com> wrote in message
news:m4zZj.13366$HJ5.10146@trnddc01...
No, but I will.
Bob Morein
(310) 237-6511
Ethan Winer
> There is Sonex, and a heavy carpet and underpad.
That can help the higher frequencies, but does nothing for bass peaks and
nulls and ringing.
--Ethan
Ethan Winer
> They are way too forward for my personal taste
But that's my point. They are very flat. Most speakers have an intentional
response dip in the "harsh" range between 2 and 4 KHz. Even many pro
speakers. So they sound "smooth and airy" and by comparison Mackies sound
harsh. But the Mackies are more accurate.
> presumably anyone buying the Lipinskis has already spent the money for the
> room.
I wish. :->)
--Ethan
Soundhaspriority
"Ethan Winer" <ethanw at ethanwiner dot com> wrote in message
of checking for it. We look for "tuneful" bass, as opposed to "one-note
bass." In the past, I have used 1/3 octave warble tones to check for bass
anomalies, but I've been so pleased with this setup, I haven't bothered. It
is the best sounding setup, with the most tuneful bass, that I've had in my
house.
Perhaps I should be more analytical, but I have a feeling this is a
"producer problem." One successful recordist with whom I have a
conversational relationship has mentioned his reliance on a third
individual, who he refers to as a "producer", to equalize for the intended
audience. It is interesting that even though he is himself a purist, he
finds the need to compromise for other, presumably purist individuals who
consume the product.
Bob Morein
(310) 237-6511
Soundhaspriority
"Ethan Winer" <ethanw at ethanwiner dot com> wrote in message
with nearfield monitors, precisely to avoid the difficulty of perfecting a
room. I do have a set of Kef RDM1, a small acoustic suspension design with
concentric drivers. I could eq them flat for the normal position of my head.
Thoughts?
Bob Morein
(310) 237-6511
Ethan Winer
> I have apparently missed the dogma of some that mixing should be done only
> with nearfield monitors, precisely to avoid the difficulty of perfecting a
> room.
Unfortunately, nearfield listening does not avoid room acoustic problems.
> I could eq them flat for the normal position of my head.
EQ does not work for many reasons. The three big reasons are:
1) EQ can lower peaks but can't correct typical 20-30dB nulls
2) EQ can't reduce ringing, which is just as damaging as peaks and nulls
3) EQ is highly positional - 4 inches away the response is very different
Much more here:
http://www.realtraps.com/art_audyssey.htm
--Ethan
Ethan Winer
> Perhaps I should be more analytical, but I have a feeling this is a
> "producer problem."
It never hurts to measure, just for a sanity check. One problem with
assessing room acoustics bass problems by ear is how the musical key
interacts with the room modes. So some music may sound great if few bass
notes excite modes, but music in a different might sound terrible. Or one
key aligns with deep nulls at the listening position and another key misses
those frequencies. This can lead one to conclude wrongly that the mixing or
mastering is at fault on this tune or that, when in truth the variation is
caused by the room.
--Ethan
Chris Hornbeck
ethanwiner dot com> wrote:
>Scott,
>
>> They are way too forward for my personal taste
>
>But that's my point. They are very flat. Most speakers have an intentional
>response dip in the "harsh" range between 2 and 4 KHz. Even many pro
>speakers. So they sound "smooth and airy" and by comparison Mackies sound
>harsh. But the Mackies are more accurate.
Another thing complicating this range is that a crossover pretty
much *has* to fall here. The crossover itself adds a big change
in directivity and (usually) a significant tilt in the radiation
pattern through the crossover region.
All the crossover issues are just exaggerated by our hearing.
Yikes! but not a lot of getting around it; dimensions are
pretty much fixed by mechanical constraints (linearly moving
pistons, yada yada) and they fight against the desire to
make all drivers very small WRT wavelength for best radiation
patterns. Good numbers fall into this range.
Possible valid but mutually exclusive crossover designs might
have flat on-listening-axis magnitude response, or flat "power"
(summed response of all sound in the room) response, or some
compromise, or even some other design goal. A perfect solution
doesn't exist. Bummer, but that's the deal.
But to respond to Bob's OP, if you've got the dough-re-mi, try
to find a pair of Nearfield Acoustics Pipedreams to listen to.
Possibly next to unobtainium now, and more expensive than my
house when current (2000, 2001, etc.) but they made the big
Wilsons sound like loudspeakers. And they're "nearfield" at
10 feet away. NOT portable (without a crew of guys with prison
tat's, anyway).
Much thanks, as always,
Chris Hornbeck
"I'm walking out, between parked cars, with my head full of stars"
-Elliott Smith
Soundhaspriority
"Chris Hornbeck" <chrishornbe...@att.net> wrote in message
news:0soh34hcmk5f6u53v...@4ax.com...
wide, in my office. They are pure electrostats with no crossover. I could
set up displays between them and mix to them in near field. The problem
isn't finding the ultimate loudspeakers. The problem is that at over $4K a
pair, my Thiel CS2.3's are atypical even as concerns an average hifi setup.
The hifi system of the musically accomplished person who gave the mix a
thumbs down doesn't have the extension of the one I mixed to. Few systems
do. Even though I am a purist, I lose my audience if the mix pleases only
listeners with exotic all-metal driver speakers with first order crossovers.
> Much thanks, as always,
>
Knowledge is power. So it occurred to me that perhaps the next step should
be some MLSSA analysis. I'm looking at Professor Angelo Farina's software
http://www.aurora-plugins.com/Aurora_XP/index.htm
This is the plan: I'll install it on a laptop and get samples of different
hifi setups. The samples may point toward a golden mean. I'll eq the Thiels
to approximate it.
This is the tentative plan. Perhaps a producer will step forward with
insight that will allow me to sidetrack it.
Bob Morein
(310) 237-6511
Chris Hornbeck
<now...@nowhere.com> wrote:
>I have a pair of Acoustat 2+2's, seven feet eight inches tall and 20 inches
>wide, in my office. They are pure electrostats with no crossover. I could
>set up displays between them and mix to them in near field.
Holey shirt Sherlock. You're set. Done deal. The big Acoustats
can tell ye anything ye wanna know. They're great.
> The problem
>isn't finding the ultimate loudspeakers. The problem is that at over $4K a
>pair, my Thiel CS2.3's are atypical even as concerns an average hifi setup.
>The hifi system of the musically accomplished person who gave the mix a
>thumbs down doesn't have the extension of the one I mixed to. Few systems
>do. Even though I am a purist, I lose my audience if the mix pleases only
>listeners with exotic all-metal driver speakers with first order crossovers.
There are no first-order crossovers. But for a serious response
about how to translate to Bose speakers, you'll need someone
else's input. Never believed in the concept, and prolly never
will.
And I doubt that you will either, if seriously attempted, but
that's a whole different critter.
hank alrich
> Knowledge is power. So it occurred to me that perhaps the next step should
> be some MLSSA analysis. I'm looking at Professor Angelo Farina's software
> http://www.aurora-plugins.com/Aurora_XP/index.htm
> This is the plan: I'll install it on a laptop and get samples of different
> hifi setups. The samples may point toward a golden mean. I'll eq the Thiels
> to approximate it.
>
> This is the tentative plan. Perhaps a producer will step forward with
> insight that will allow me to sidetrack it.
NS10's; Auratones. <g>
--
ha
Iraq is Arabic for Vietnam
Chris Hornbeck
wrote:
>Soundhaspriority <now...@nowhere.com> wrote:
>> MLSSA analysis.
>> This is the plan: I'll install it on a laptop and get samples of different
>> hifi setups. The samples may point toward a golden mean. I'll eq the Thiels
>> to approximate it.
>>
>> This is the tentative plan. Perhaps a producer will step forward with
>> insight that will allow me to sidetrack it.
>
>NS10's; Auratones. <g>
But seriously, Hank, what's your real recommendation? I'd bet
good money that your actual, personal method is just pure
cubic inches; you listen to your recording in an environment
that you're comfortable working in, and make all adjustments
in wetware (horsepressure between the ears).
Sanity checks are another matter, and maybe I'm all wet.
(First time for everything!)
Soundhaspriority
"hank alrich" <walk...@nv.net> wrote in message
news:1ihgvo1.1o4sg5aio399rN%walk...@nv.net...
So I've heard. I'll give ten bucks to the first person who drops a pair
off at my door :)
Bob Morein
(310) 237-6511
hank alrich
> On Sat, 24 May 2008 22:58:35 -0700, walk...@nv.net (hank alrich)
> wrote:
>
> >Soundhaspriority <now...@nowhere.com> wrote:
> >> MLSSA analysis.
>
> >> This is the plan: I'll install it on a laptop and get samples of different
> >> hifi setups. The samples may point toward a golden mean. I'll eq the Thiels
> >> to approximate it.
> >>
> >> This is the tentative plan. Perhaps a producer will step forward with
> >> insight that will allow me to sidetrack it.
> >
> >NS10's; Auratones. <g>
>
> But seriously, Hank, what's your real recommendation?
In Austin my first real monitors were JBL 4320's. After onion audio was
no more and I was working with the Lonestar Recording guys I tracked and
mixed on JBL4315's, until they blew them up. I bought a pair of 4430's,
which I really liked, at least in that little Storyk-designed room.
At onion audio and Lonestar we drove the JBL's with MC2100's, one per
speaker, bridged. I have not been as pleased since with any other amps,
though I've gotten by with Crest PL400's, a PL300, and a Hafler P3000.
After leaving Austin I ran out of space for the 4315's. I then found a
good used pair of Tannoy NFM8's, and ran them until I had space and time
to reinstall the JBL's. I was a happy clam then, as I had back the low
end info that was hard to discern sometimes over the Tannoys. Then one
day last year the Jerbils started making funny noises. I pulled the
grill cloths to discover that almost all of the surrounds had given up,
at the same time.
I'd been using these since 1982, and present financial circumstances
prevent repairing them properly and completely right now. A friend of
mine gathered thousands of bucks worth of fine gear not so long ago and
then had his life get redirected. He offered me the use of his Genelec
8040's for what will probably be a couple of years. They seem to work
well. The low end extension is impressive, assuming that one gets the
settings right for a given room. I have mixed one product on them, some
of which was tracked here and some of which was done at Two Street in
Arcata CA. (Mike Kapitan, an independent engineer who moved there some
years ago from LA, was great to work with.)
The raw tracks from Two Street, when brought back here, demonstrated the
value of intelligent, even if minimal acoustical treatment, and the
wetwear you reference. They had the newer powered versions of the NS10's
from Yamaha, and while the bottom sounded okay in their weird little
room, here over the Gennies it was obviously a mess. I massaged it all
as best I could, and then I took it all to Jerry Tubb at Terra Nova in
Austin, and he did a masterful job of mastering. I was pleased that all
the problems I heard here showed up at Terra Nova, but obviously over
the Duntechs they were revealed in greater detail. My wetwear was not
preprogrammed for Two Street's monitoring, and did poorly there under
the circumstance.
This project was for a first-time recording artist. (I've done that a
lot, and seem to be good at helping folks deal with their apprehension
and ignorance, in the positive sense of that word.) The budget was
ridiculous but these folks made some key decisions that most
money-challenged beginners do not accept. They hired me to produce and
they bought Terra Nova's services. I offered them a payment plan that
helped them get this done, and cut them considerable slack.
The final result is not overly impressive-sounding, but this is one of
those projects where the material is the really interesting aspect, and
the arrangements are also quite cool, if simple. The performing will
grow to match.
I've known the guy since he was a teenager, and the primary artist, his
wife, Melissa Ruth, writes interesting indie-chick kind of stuff. I'd
never worked with that style of music, and really enjoyed it. They're
already planning the next CD and have digested the wisdom of many things
I suggested that they didn't accept the first round. They will allocate
more time, there will be more rehearsal, we'll have more energy put into
preproduction.
Her CD will be out in a few weeks. They've lined up a series of gigs for
the summer. They're both high school music teachers who will take a
mutli-year sabbatical from teaching if her performing career goes well
enough to sustain them. I'm nearly through recording other folks, but I
will work with Melissa as long as she wants me to, or I'm alive.
Here's a link to her site: http://www.melissaruthmusic.com/
She demonstrates the value of research, persistence and hard work. She
took a very rough demo recording that I'd given her (she had the flu, a
huge rainstrom came through and we wound up tracking in the control
room, with the laptop whining in the background - this was so she could
hear her songs and send 'em to the copyrighting machinery) and put
together a nice little tour behind it. There are lots of lessons here
for other beginners, but few will pay attention. That she does is one
more reason I look forward to working with her again.
> I'd bet
> good money that your actual, personal method is just pure
> cubic inches; you listen to your recording in an environment
> that you're comfortable working in, and make all adjustments
> in wetware (horsepressure between the ears).
Correct, and when the environment is unfamiliar and not well put
together, and my wetwear isn't programmed for it, things are likely to
turn out interestingly, in the sense of the Chinese curse. I cannot
overstate the value of truly professional mastering services. Jerry Tubb
is extremely versatile and a completel pleasure to work with.
If I had the money I would buy a pair of Sound Anchor stands and put
Klein & Hummel O300's on them.
> Sanity checks are another matter, and maybe I'm all wet.
> (First time for everything!)
It's raining here this morning.
Soundhaspriority
> Chris Hornbeck <chrishornbe...@att.net> wrote:
>
>> On Sat, 24 May 2008 22:58:35 -0700, walk...@nv.net (hank alrich)
>> wrote:
>>
>
> After leaving Austin I ran out of space for the 4315's. I then found a
> good used pair of Tannoy NFM8's, and ran them until I had space and time
> to reinstall the JBL's. I was a happy clam then, as I had back the low
> end info that was hard to discern sometimes over the Tannoys. Then one
> day last year the Jerbils started making funny noises. I pulled the
> grill cloths to discover that almost all of the surrounds had given up,
> at the same time.
>
Hank, that was great! - thanks for sharing.
There is a local guy who has a national reputation for driver repair -- all
kinds. He works on extremely exotic stuff. It's all he does. You take the
drivers out, and send them to him, and it's quite reasonable. Let me know if
you want to be hooked up.
Bob Morein
(310) 237-6511
hank alrich
Thanks, Bob. Eventually I will get the JBL's fixed. But the direction
I'm headed is to turn the control room into a personal all-purpose
tracking and mixing room for eveything that doesn't need the space or
natural reverb of my music room. I'll want to use something smaller than
the 4315's in there. The room is 9' h. (with a whole lot a acoustical
'glas and cotton batting in the foot-deep spaces between ceiling
joists), 11' w. and 17 l.
So space is a bit of a premium. I'll add more acoustical treatment to
the ceiling and some to the walls. I recently sold the old A80 1", which
gave me a bit more room. It was the VUT (transportable model) and sat on
a large cabinet we built for it to ride at onion audio.
Witness: http://mail.nv.net/~hank/Other_Photos/138_3854.JPG
Soundhaspriority
Bob Morein
(310) 237-6511
hank alrich
> "hank alrich" <walk...@nv.net> wrote in message
> news:1ihhs07.nnhwc21vl13jhN%walk...@nv.net...
> > So space is a bit of a premium. I'll add more acoustical treatment to
> > the ceiling and some to the walls. I recently sold the old A80 1", which
> > gave me a bit more room. It was the VUT (transportable model) and sat on
> > a large cabinet we built for it to ride at onion audio.
> >
> > Witness: http://mail.nv.net/~hank/Other_Photos/138_3854.JPG
> >
> It's smaller than the 8 track 350 conversion I used to service!
But it's a little larger than the MIO.
Ty Ford
(in article <1ihhkvj.kqlddezsfup9N%walk...@nv.net>):
> In Austin my first real monitors were JBL 4320's. After onion audio was
> no more and I was working with the Lonestar Recording guys I tracked and
> mixed on JBL4315's, until they blew them up. I bought a pair of 4430's,
> which I really liked, at least in that little Storyk-designed room.
>
> At onion audio and Lonestar we drove the JBL's with MC2100's, one per
> speaker, bridged. I have not been as pleased since with any other amps,
> though I've gotten by with Crest PL400's, a PL300, and a Hafler P3000.
>
> After leaving Austin I ran out of space for the 4315's. I then found a
> good used pair of Tannoy NFM8's, and ran them until I had space and time
> to reinstall the JBL's. I was a happy clam then, as I had back the low
> end info that was hard to discern sometimes over the Tannoys. Then one
> day last year the Jerbils started making funny noises. I pulled the
> grill cloths to discover that almost all of the surrounds had given up,
> at the same time.
Too many years ago, I walked into a "hi-fi" store in Annapolis, MD. I went
into the speaker room and the salesmman punched the buttons on the speaker
switcher.
"Stop!", I said, "What's that?"
"Those are JBL 4311 studio monitors", he said.
"How much?", said I.
(don't remember, but somewhere around $375 each...maybe)
"Ouch!" said I, and I walked out.
About a year later I walked back into the same store, into the speaker room,
to the same panel and began pushing unlabeled buttons.
At one point, I hit a button (a year later, now) and turned around to find I
was listening to the 4311...again.
Some time later I found a used pair of L100 (the orange foam covers
disintegrated a long time ago) , the consumer version JBL came out with
between the 4310 and 4311. I use them with no covers and had to dial back the
"Presence" and "Brilliance" adjustments when I put them up on stands as "just
further than nearfields." Still using them, these days with a Hafler P4000.
Turns out, I stumbled into what a lot of people consider some really nice
monitors. If the JBL weren't working for me, I'd try some Dynaudio, Adam's,
Meyers HD-1's, Klein & Hummel or Genelec.
Regards,
Ty Ford
--Audio Equipment Reviews Audio Production Services
Acting and Voiceover Demos http://www.tyford.com
Guitar player?:http://www.youtube.com/watch?v=4RZJ9MptZmU
William Sommerwerck
> floorstanding speaker, which has three metal drivers, and an
> original cost of around $4K per pair. In the hifi marketplace, they
> considered to have enduring value. But they sound different from
> my other speakers, because all speakers sound different. They are
> brighter than my Kef Reference III's, which have fabric domes, and
> are less optimally placed in the room.
I'm not a fan of Thiels. They're a classic example of speakers with a flat
response that nevertheless sound "insipid".
What kind of music are you monitoring? If it's classical -- or even jazz --
you might want the highest-quality speakers you can get -- including
electrostatics.
But 'stats aren't good for other material, because they mislead you about
what the sound will be like on commonly used speakers.
I would go with a pair of B&Ws. They're neutral without being vapid. You
could probably get a pair of 801s for less than $2000.
Soundhaspriority
"William Sommerwerck" <grizzle...@comcast.net> wrote in message
news:DYWdnSrqoNAONKfV...@comcast.com...
>> The Thiel CS2.3 is a three-way first order crossover time aligned
>> floorstanding speaker, which has three metal drivers, and an
>> original cost of around $4K per pair. In the hifi marketplace, they
>> considered to have enduring value. But they sound different from
>> my other speakers, because all speakers sound different. They are
>> brighter than my Kef Reference III's, which have fabric domes, and
>> are less optimally placed in the room.
>
> I'm not a fan of Thiels. They're a classic example of speakers with a flat
> response that nevertheless sound "insipid".
>
at the whole thread, because your remarks touche on the major issue, but in
the "opposite" direction.
[snip]
Bob Morien
(310) 237-6511
Soundhaspriority
"Ty Ford" <tyre...@comcast.net> wrote in message
news:R6ydnfnDSvYMAafV...@comcast.com...
> On Sun, 25 May 2008 11:50:27 -0400, hank alrich wrote
> (in article <1ihhkvj.kqlddezsfup9N%walk...@nv.net>):
>
>
> Turns out, I stumbled into what a lot of people consider some really nice
> monitors. If the JBL weren't working for me, I'd try some Dynaudio,
> Adam's,
> Meyers HD-1's, Klein & Hummel or Genelec.
>
I don't want it to turn out like my mattress buying experience, so I'm
setting up an MLSSA rig to study it further.
Bob Morein
(310) 237-6511
William Sommerwerck
> as to glance at the whole thread, because your remarks
I tried to contact you directly, but your address is invalid. If you want to
discuss this off-line, I'm game. Contact me directly.
mark steven brooks
the musical key
> interacts with the room modes.
You mean nodes.
Don Pearce
Chris Hornbeck
wrote:
<snipped cause you've already read it>
Thanks, Mon. Don't make me get all weepy just because
it's Memorial Day. But thanks. As always.
Soundhaspriority
Let's not get sidetracked on the characteristics of the Thiels. Reading on
the web, I found a repeated assertion, that seems reasonable to me, that the
mix will tend to be attenuated in bands accentuated by the mix speakers.
Conversely, if the mix speakers attenuate in certain bands, the
corresponding mix will tend to be accentuated in those areas. The human
"wetwear" tends to make an inverse tranformation to convert the mix to
whatever sounds neutral over a particular pair of mix speakers.
A musically accomplished person auditioned this mix on this person's
personal stereo system, and found it lacking in highs and lows. Since this
was not intended by me, it is a result of the difference in portrayal
between the Thiels and this person's system. It implies that the Thiels have
more highs and lows than that system.
The question of this thread is not about getting phenomenal speakers.
Lapinskis are, I'm sure, very enjoyable to listen to, but would not connect
well with the experience of listening on mid-fi, or even entry level
high-end. Hank Alrich suggested, with good foundation, that Yamaha NS-10M's
might have something to add, because they are, by aesthetic standards,
deficient in highs and lows. Had I used NS-10M's to mix, it is likely that
the person who noted the absence of highs and lows would have a different
observation.
There is no shortage of speakers at this location. The current inventory of
full range, floorstanding speakers is large and diverse. I also have a large
collection of bookshelf speakers. Ty Ford mentioned some of the well
accepted choices. But it's a bit like buying a mattress. I do not want to go
out and buy another speaker, unless it is a well supported choice.
Your ball.
Bob Morein
(310) 237-6511
William Sommerwerck
> seems reasonable to me, that the mix will tend to be
> attenuated in bands accentuated by the mix speakers.
That goes without saying, assuming one is trying to produce a mix with a
subjectively neutral tonal balance, and one is making aggressive use of EQ..
> Conversely, if the mix speakers attenuate certain bands, the
> corresponding mix will tend to be accentuated in those areas.
Ditto, for the same reason.
> A musically accomplished person auditioned this mix on this person's
> personal stereo system, and found it lacking in highs and lows. Since
> this was not intended by me, it is a result of the difference in portrayal
> between the Thiels and this person's system. It implies that the Thiels
> have more highs and lows than that system.
It _implies_ that, but Thiels are basically "flat" speakers. There might be
other reasons for this person's subjective impression -- the mix might be
reasonably neutral, but he _prefers_ lots o' bass 'n treble.
There is another issue we haven't raised. "Neutrality" (in this context)
imples a basically flat frequency (or power) response. But the flatness of
response is only indirectly related to one's impression of detail. The
speaker with poorer subjective detail might encourage the mixer to process
the signal in ways the better speaker might not.
In an amazing coincidence, I was just speaking with Dr. Clayton Barclay,
whom I used to work for. He told me that, when mixing, he listens to the mix
over a wide variety of speakers, to make sure he gets the basic effect he
wants on all of them. See below.
> There is no shortage of speakers at this location. The current inventory
> of full range, floorstanding speakers is large and diverse. I also have a
> large collection of bookshelf speakers. Ty Ford mentioned some of the
> well accepted choices. But it's a bit like buying a mattress. I don't want
> to go out and buy another speaker, unless it is a well-supported choice.
I understand your confusion. But there's no way to produce a "universally
acceptable" mix listening to just one speaker.
Here's a suggestion. Pick a speaker you like and are familiar with. Create
the best mix you can using that speaker. Then, listen to it on lots of other
speakers, and see what you think.
I don't know if this will resolve the issue for you, but it will certainly
give you a better idea of what the appropriate questions are.
Soundhaspriority
[snip]
>
> In an amazing coincidence, I was just speaking with Dr. Clayton Barclay,
> whom I used to work for. He told me that, when mixing, he listens to the
> mix
> over a wide variety of speakers, to make sure he gets the basic effect he
> wants on all of them. See below.
>
salons in Bryn Mawr. I used to go over there to ogle the Crown stuff. In
what capacity did you work for him?
Bob Morein
(310) 237-6511
William Sommerwerck
news:56idnez2Pv9ds6bV...@giganews.com...
> "William Sommerwerck" <grizzle...@comcast.net> wrote in message
> news:bOSdnVB24YVytqbV...@comcast.com...
>> In an amazing coincidence, I was just speaking with Dr. Clayton
>> Barclay, whom I used to work for. He told me that, when mixing,
>> he listens to the mix over a wide variety of speakers, to make
>> sure he gets the basic effect he wants on all of them.
> Clay Barclay? The neurosurgeon? He used to have one of the early
> high-end salons in Bryn Mawr. I used to go over there to ogle the
> Crown stuff. In what capacity did you work for him?
I was a salesman and a kind of engineer. I worked in the retail store from
late '77 through '78, then switched to the "engineering" department for '79.
We were one of the few dealers who sold the Plasmatronics speaker. Ahhh...
"Everything else is gas light", as Fluffy was wont to say...
Soundhaspriority
What is Clay doing these days?
Bob Morein
(310) 237-6511
Scott Dorsey
No, he doesn't.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
david correia
"William Sommerwerck" <grizzle...@comcast.net> wrote:
> I understand your confusion. But there's no way to produce a "universally
> acceptable" mix listening to just one speaker.
>
> Here's a suggestion. Pick a speaker you like and are familiar with. Create
> the best mix you can using that speaker. Then, listen to it on lots of other
> speakers, and see what you think.
Very good advice. Listen on your fave headphones as well.
The game is to make your mixes sound good/acceptable on everything.
David Correia
www.Celebrationsound.com
Ethan Winer
> You mean nodes.
No, I mean modes. Mode is short for "mode of vibration," and in a room these
are the resonant frequencies determined by the wall-wall and floor-ceiling
spacing. A node (or anti-node) is a physical place in the room where waves
collide creating peaks and nulls.
--Ethan
Scott Dorsey
>
>> They are way too forward for my personal taste
>
>But that's my point. They are very flat. Most speakers have an intentional
>response dip in the "harsh" range between 2 and 4 KHz. Even many pro
>speakers. So they sound "smooth and airy" and by comparison Mackies sound
>harsh. But the Mackies are more accurate.
I dunno, I think my objection has to do with the radiation pattern and not
a presence area dip. But could be! It would be interesting to see.
The Altec 604 is forward as hell and I can't stand listening to the things,
but they sure were popular and a lot of people made some great music using
them.
>> presumably anyone buying the Lipinskis has already spent the money for the
>> room.
>
>I wish. :->)
It's a weird world.
William Sommerwerck
news:g1hna6$2b7$1...@panix2.panix.com...
>> Most speakers have an intentional response
>> dip in the "harsh" range between 2 and 4 KHz.
Most? I doubt it.
Scott Dorsey
Attribution check, please.
Chris Hornbeck
<now...@nowhere.com> wrote:
>Knowledge is power. So it occurred to me that perhaps the next step should
>be some MLSSA analysis. I'm looking at Professor Angelo Farina's software
>http://www.aurora-plugins.com/Aurora_XP/index.htm
>This is the plan: I'll install it on a laptop and get samples of different
>hifi setups. The samples may point toward a golden mean. I'll eq the Thiels
>to approximate it.
>
>This is the tentative plan.
I ride the short bus, so most often things take a while to
filter in. The above has been knawing at my brainstem, but
with maybe a different slant.
The webpage is refreshing and anarchistic, but I haven't tried
the software (already got something else). Of course you
already know that windowing gives a lower frequency limit
to measurement, and that the low frequencies are the most
difficult to monitor.
But I wonder if applying some analysis to your existing
monitoring mightn't give you another coupla data points.
You can't really measure anything "definitive", but it's
sometimes interesting anyway. You'll *always* have to
make some translation from your monitoring, so maybe a
few curves from the monitoring might mentally help the
translation.
Interesting thread; much thanks, as always,
William Sommerwerck
> William Sommerwerck <grizzle...@comcast.net> wrote:
> >"Scott Dorsey" <klu...@panix.com> wrote in message
> >news:g1hna6$2b7$1...@panix2.panix.com...
>>>> Most speakers have an intentional response
>>>> dip in the "harsh" range between 2 and 4 KHz.
>> Most? I doubt it.
> Attribution check, please.
Ditto, Scott.
Ethan Winer
> Most? I doubt it.
Heh, glad to see everyone is paying attention. :->)
The following is from a recent forum post I made where this same subject
came up.
--Ethan
----------------
A few years ago a fellow named David French measured ten popular pro
speakers at very high accuracy using a technique that's even better than an
anechoic chamber. The problem with acoustic lab anechoic chambers is they
are anechoic only down to 100 Hz, or maybe 80 Hz if it's a really good lab.
What David did was fly all ten speakers high up over the stage in a huge
auditorium. By keeping the speakers at least 50 feet (as I recall) from all
boundaries, a technique known as Gating can be used to avoid the effects of
all reflections down to a very low frequency.
I have a photo of the test setup and it was valid. I also have the graphs,
which were made using the same ETF software I use and trust. I've tried for
the past year or two to reach David for permission to post his graphs on my
own site because they are so incredibly telling. Without his permission I
can't post the actual graphs, but I can tell you what they say. :->)
Understand that frequency response is only one part of loudspeaker
performance. But it's probably the most important, and frequency response is
clearly the main contributor to adjectives such as "boomy" or "scooped" and
so forth. But these graphs don't show distortion which is typically very
high in speakers compared to other "gear," nor do they show what happens
off-axis which is a big factor for good imaging and sound stage. Nor do they
show ringing which is very important, especially at bass frequencies.
The speakers tested were Tannoy 800A, Genelec 1031A, Event Studio Precision
8, Dynaudio BM15A, Mackie HR824, Alesis M1 Active MkII, M&K S-150PK THX
Ultra, Yamaha NS10, ADAM S2A, and Event TR8. In order:
The Tannoy was reasonably flat from 60 Hz to 14 KHz, with a broad 3dB dip
centered around 4 KHz.
The Genelec was reasonably flat from 60 Hz to 20 KHz with a broad 4 dB dip
centered around 4 KHz.
The Event was a little bumpier between 60 Hz and 20 KHz with a narrow 4 dB
dip at 400 Hz and a broad 4 dB dip between 4 and 6 KHz.
The Dynaudio was pretty terrible having a narrow 9 dB dip at 3.5 KHz and a
sharp rolloff above 8 KHz.
The Mackie was by far the smoothest of the lot being flat within 3 dB (total
span, not +/-) from 60 Hz to 20 KHz, with a broad 2 dB dip between 1.6 and 4
KHz.
The Alesis response went out to 18 KHz but had a narrow 5 dB peak centered
at 60 Hz.
The M&K fell off sharply below 100 Hz, with narrow dips of 3 and 4 dB at 3
KHz and 6 KHz respectively, and a narrow boost at 12 KHz.
The NS10 was as expected with very little bass below 120 Hz, a narrow 4 dB
boost at 1.6 KHz, another narrow peak at 12 KHz, and a sharp fall-off above
12 KHz.
The Adam was pretty lumpy with a steep LF roll-off starting at 80 Hz, a
broad 3 dB boost at 1 KHz, a broad 6 dB scoop centered at 4 KHz, and then
was flat to 20 KHz.
The Event was similarly lumpy, but with a 9 dB narrow dip at 4 KHz, a narrow
peak at 13 KHz, and a sharp roll-off above that.
Again I want to stress the importance of nearly all these speakers having
dips in the "irritating" range between 2 and 4 KHz. As anyone who has ever
EQ'd music knows, this is where music sounds harsh and unpleasant. So when a
speaker has an intentional dip there, it can be marketed as sounding smooth
and silky and all those other audiophile adjectives we all know and hate.
Likewise with peaks in the "sparkle" range above 6 KHz or the "fullness"
range below 100 Hz. I expect such trickery with speakers sold to the hi-fi
market, but I find it very upsetting to see this used to market speakers
that are supposed to be pro quality.
--Ethan
Scott Dorsey
>After leaving Austin I ran out of space for the 4315's. I then found a
>good used pair of Tannoy NFM8's, and ran them until I had space and time
>to reinstall the JBL's. I was a happy clam then, as I had back the low
>end info that was hard to discern sometimes over the Tannoys. Then one
>day last year the Jerbils started making funny noises. I pulled the
>grill cloths to discover that almost all of the surrounds had given up,
>at the same time.
>
>prevent repairing them properly and completely right now. A friend of
>mine gathered thousands of bucks worth of fine gear not so long ago and
>then had his life get redirected. He offered me the use of his Genelec
>8040's for what will probably be a couple of years. They seem to work
>well. The low end extension is impressive, assuming that one gets the
>settings right for a given room. I have mixed one product on them, some
>of which was tracked here and some of which was done at Two Street in
>Arcata CA. (Mike Kapitan, an independent engineer who moved there some
>years ago from LA, was great to work with.)
Go to Parts Express. Get the generic Waldom kit which will fit the JBLs.
Install it yourself, using the supplied directions. It will cost you about
$20 for two speakers, and it will actually be fine.
You'd think that changing the surround would affect the driver Fs a whole
lot... but actually most of the stiffness is supplied by the spider and
even if the stiffness of the surround is substantially different than the
original, it doesn't make much difference. I was really blown away by this
when I found out... I was altering the surrounds to try and reduce the Fs of
pro audio drivers and found I could make only minimal changes.
If you don't like the end results, you can always send them off and have
them done by the experts at Cardinal Sound and Motion Picture with OEM
surround kits. But I bet you decide not to bother.
>The raw tracks from Two Street, when brought back here, demonstrated the
>value of intelligent, even if minimal acoustical treatment, and the
>wetwear you reference. They had the newer powered versions of the NS10's
>from Yamaha, and while the bottom sounded okay in their weird little
>room, here over the Gennies it was obviously a mess. I massaged it all
>as best I could, and then I took it all to Jerry Tubb at Terra Nova in
>Austin, and he did a masterful job of mastering. I was pleased that all
>the problems I heard here showed up at Terra Nova, but obviously over
>the Duntechs they were revealed in greater detail. My wetwear was not
>preprogrammed for Two Street's monitoring, and did poorly there under
>the circumstance.
The Genelecs are damn good, though... I think you will find you like them
more than the JBLs in the long run. But really, try doing the JBLs yourself.
The worst you can do is be out the cost of a surround kit.
>If I had the money I would buy a pair of Sound Anchor stands and put
>Klein & Hummel O300's on them.
A customer of mine is currently replacing a large array of Altec 604s with
K&H 0300s and I'm looking forward to seeing how the thing sounds. We actually
rescued one of the old 604s out of the dumpster, but the other fifteen had the
magnets removed with crowbars and salvaged to use as fridge magnets by the
demolition guys. While I hate the way 604s sound, I also am looking at a lot
of money going down the tubes too.
jakdedert
>> wrote:
>>
>>> Soundhaspriority <now...@nowhere.com> wrote:
>>>> MLSSA analysis.
>>>> This is the plan: I'll install it on a laptop and get samples of different
>>>> hifi setups. The samples may point toward a golden mean. I'll eq the Thiels
>>>> to approximate it.
>>>>
>>>> insight that will allow me to sidetrack it.
>>> NS10's; Auratones. <g>
>> But seriously, Hank, what's your real recommendation?
>
> In Austin my first real monitors were JBL 4320's. After onion audio was
> no more and I was working with the Lonestar Recording guys I tracked and
> mixed on JBL4315's, until they blew them up. I bought a pair of 4430's,
> which I really liked, at least in that little Storyk-designed room.
>
> At onion audio and Lonestar we drove the JBL's with MC2100's, one per
> speaker, bridged. I have not been as pleased since with any other amps,
> though I've gotten by with Crest PL400's, a PL300, and a Hafler P3000.
>
> good used pair of Tannoy NFM8's, and ran them until I had space and time
> to reinstall the JBL's. I was a happy clam then, as I had back the low
> end info that was hard to discern sometimes over the Tannoys. Then one
> day last year the Jerbils started making funny noises. I pulled the
> grill cloths to discover that almost all of the surrounds had given up,
> at the same time.
>
> I'd been using these since 1982, and present financial circumstances
> prevent repairing them properly and completely right now. A friend of
> mine gathered thousands of bucks worth of fine gear not so long ago and
> then had his life get redirected. He offered me the use of his Genelec
> 8040's for what will probably be a couple of years. They seem to work
> well. The low end extension is impressive, assuming that one gets the
> settings right for a given room. I have mixed one product on them, some
> of which was tracked here and some of which was done at Two Street in
> Arcata CA. (Mike Kapitan, an independent engineer who moved there some
> years ago from LA, was great to work with.)
I second Scott's recommendations on retreading the JBLs. It's
time-consuming and a 'little' exacting, but not really all that
difficult (or expensive) in the long run. Budget a couple of days to do
all the surrounds in a pair of 4315s. Not that it will take all that
long...hour or so per driver, less if you do more than one at a time. I
assume the tweets are still okay, but they're the easiest to recone,
having removable diaphragms...most expensive, though.
I suggest doing one to get up to speed, then a couple more (or how ever
many at a time you get comfortable doing). Most of the time is consumed
cleaning the old material off the edges of the cones and speaker frame
(and waiting for glue to dry).
I'm usually cautious about recommending surround retreads, because often
people are doing them without any knowledge of what damage that might
have occurred to the voice coil, or the connections to it. In your
case, you were listening critically when the problem exhibited, and
surely you discontinued using them immediately....
Careful googling will locate the bare surrounds (not 'kits') for much
less than Parts Express prices. The 2105 mid drivers might be a little
difficult to find good matches for....
jak
Glenn Dowdy
help you. I don't get the ditto.
Glenn D.
Soundhaspriority
"Ethan Winer" <ethanw at ethanwiner dot com> wrote in message
news:18-dnZJpm8yH_6DV...@giganews.com...
> William,
>
[snip good stuff]
>
> Again I want to stress the importance of nearly all these speakers having
> dips in the "irritating" range between 2 and 4 KHz. As anyone who has ever
> EQ'd music knows, this is where music sounds harsh and unpleasant. So when
> a
> speaker has an intentional dip there, it can be marketed as sounding
> smooth
> and silky and all those other audiophile adjectives we all know and hate.
> Likewise with peaks in the "sparkle" range above 6 KHz or the "fullness"
> range below 100 Hz. I expect such trickery with speakers sold to the hi-fi
> market, but I find it very upsetting to see this used to market speakers
> that are supposed to be pro quality.
>
> --Ethan
>
It is disappointing, because it could incline us to make mixes accentuated
in the "irritating range." Which of the test subjects would be your choice
to mix with?
Bob Morein
(310) 237-6511
Harry Lavo
Makes a good case for continuing to use your Thiels....and then maybe a pair
of Radio Shack "7"s to simulate cars and home radios.
Soundhaspriority
"Chris Hornbeck" <chrishornbe...@att.net> wrote in message
news:kffp34pt23t8j10ts...@4ax.com...
Maybe nothing at all. But as far as windowing, permit me to run this by you:
1. To see the "first arrival" mid and high response, window as necessary.
2. In the bass region, there is no such thing as "first arrival", so there
really is no need to window smaller than the maximum length sequency.
So, make two separate runs, one curve relevant only for the bass. If this
makes sense to you, what frequency would you suggest to drop the window?
And what package do you use?
Bob Morein
(310) 237-6511
Soundhaspriority
"Harry Lavo" <hl...@comcast.net> wrote in message
news:GKmdnQy--ebicqDV...@comcast.com...
refer to, but I have an ample quantity of the RS diecast aluminum A/V
speakers. The false bass hump is so extreme as to make them a bit too
unique.
Bob Morein
(310) 237-6511
drichard
Your note confuses me. I have the Event TR8's, and to me they don't
sound that bad. I can mix on them. But after reading your note, I
listened to a sweep tone from 200 hz to 6 Khz, and according those
test results would have expected to hear a huge drop around 4K. But I
didn't. I didn't hear much of a change at all, in fact they sounded
very consistent. No drop out at all centered around 4 Khz. I'm
listening in a nearfield triangle position, only about three feet from
the speakers, with the speakers about three feet apart.
Could the test you're referring to have been flawed? While tests in an
anechoic chamber might have problems below 100 hz, shouldn't they be
accurate above that? If so, why do other tests not show these same
anomalies around the 3K center as that particular one?
I hold your opinions in high regard, but what I'm hearing here doesn't
coincide with those test results.
Any thoughts?
Dean
On May 28, 7:44 am, "Ethan Winer" <ethanw at ethanwiner dot com>
wrote:
Soundhaspriority
"drichard" <DRic...@wi.rr.com> wrote in message
news:3932a632-3381-4bdc...@26g2000hsk.googlegroups.com...
Hi Ethan,
Your note confuses me. I have the Event TR8's, and to me they don't
sound that bad. I can mix on them. But after reading your note, I
listened to a sweep tone from 200 hz to 6 Khz, and according those
test results would have expected to hear a huge drop around 4K. But I
didn't. I didn't hear much of a change at all, in fact they sounded
very consistent. No drop out at all centered around 4 Khz. I'm
listening in a nearfield triangle position, only about three feet from
the speakers, with the speakers about three feet apart.
Could the test you're referring to have been flawed? While tests in an
anechoic chamber might have problems below 100 hz, shouldn't they be
accurate above that? If so, why do other tests not show these same
anomalies around the 3K center as that particular one?
I hold your opinions in high regard, but what I'm hearing here doesn't
coincide with those test results.
Any thoughts?
----------------------------------------------------------------------------------
You're not using an instrument, and the ear is not reliable. It's very easy
to acclimate. And I would guess you're getting some reflected sound. When
designing, one can aim for
1. Flat on-axis response.
2. Flat response integrated over some angle, or, practically, flat including
some reflections.
It is known that the ear recognizes the timbre of direct and reflected sound
differently. This probably has something to do with the "Law of the first
wavefront", which means that the ear determines location of the sound source
from the direct, not reflected, sound.
Bob Morein
(310) 237-6511
Chris Hornbeck
<now...@nowhere.com> wrote:
> But as far as windowing, permit me to run this by you:
>2. In the bass region, there is no such thing as "first arrival", so there
>really is no need to window smaller than the maximum length sequency.
>So, make two separate runs, one curve relevant only for the bass. If this
>makes sense to you, what frequency would you suggest to drop the window?
My poor understanding would express the issue differently:
There is still a first arrival for the bass, but it's not
possible to measure bass by windowing the listening mic, because
not enough of a wavelength fits into the window. Window
size is limited by room dimensions', testee and mic spacing
from room boundaries, etc., natch.
To my understanding, this doesn't just cause just a slope
to the measured response, but rather a "ringing", similar
in shape to the low frequency response of an acoustical horn
when dimensions fall into the wavelength's range. A window'd
sample of a wavelength (with fixed window size) will give
a response that could also be compared to a sampling error.
Enough BS; I hope the issue is clearer to you than my
explanation.
And I'd greatly love to be corrected if this is too far off
the reservation.
Another way to say a large window time is no windowing at
all, and this is the way bass measurements are *still*
done. Ya stick the mic up real close, or dig a whole in the
back yard and point the speaker up, or climb up on the
grandkids' swingset and mount the speaker up there, and
then climb yourself up a *really* scary extension ladder
and mic from there.
After yesterday and today's day-job exploits on the end of a
Genie holding two guys at 40 feet up and sometimes (too often....)
20 feet OVER SIDEWAYS, 'm becoming almost bored with
dangling way up in the air for no good reason at all. So
I don't discount the swingset approach anymore.
>And what package do you use?
I still use the ancient WinAIRR and with an even more ancient
Mitey Mike (personally calibrated by Joe D'Appolito himself!)
Not reference quality, but still way too good for speakers
and rooms.
Much thanks, as always,
drichard
I don't agree with your assessment. First, I'm three feet from the
speakers in a room heavily treated with 703 and corner panels. It
could use more bass traps, and it is small, but it's not bad - a bit
on the dead side. I'm several feet from three of the walls, and not
getting very much reflected sound at all. If anything, reflections
should cause more anomalies, not fewer. What I hear is an even, steady
sweep. I would typically expect room reflections to throw that sweep
askew, not smooth it out.
The quoted test says these speakers are 9 db down at 4 Khz. That's
incredibly inaccurate. Using a sweep like the one I'm using, one
should be able to hear a frequency response deviation that gross from
the next room! But I don't. As I'm sitting right here running it, it
sounds smooth and even.
Dean
On May 28, 8:13 pm, "Soundhaspriority" <nowh...@nowhere.com> wrote:
> "drichard" <DRich...@wi.rr.com> wrote in message
>
> news:3932a632-3381-4bdc...@26g2000hsk.googlegroups.com...
> Hi Ethan,
>
> Your note confuses me. I have the Event TR8's, and to me they don't
> sound that bad. I can mix on them. But after reading your note, I
> listened to a sweep tone from 200 hz to 6 Khz, and according those
> test results would have expected to hear a huge drop around 4K. But I
> didn't. I didn't hear much of a change at all, in fact they sounded
> very consistent. No drop out at all centered around 4 Khz. I'm
> listening in a nearfield triangle position, only about three feet from
> the speakers, with the speakers about three feet apart.
>
> Could the test you're referring to have been flawed? While tests in an
> anechoic chamber might have problems below 100 hz, shouldn't they be
> accurate above that? If so, why do other tests not show these same
> anomalies around the 3K center as that particular one?
>
> I hold your opinions in high regard, but what I'm hearing here doesn't
> coincide with those test results.
>
> Any thoughts?
> ----------------------------------------------------------------------------------
Soundhaspriority
Bob Morein
(310) 237-6511
"drichard" <DRic...@wi.rr.com> wrote in message
news:22b41650-d965-45e4...@a70g2000hsh.googlegroups.com...
Chris Hornbeck
<DRic...@wi.rr.com> wrote:
>I don't agree with your assessment. First, I'm three feet from the
>speakers in a room heavily treated with 703 and corner panels. It
>could use more bass traps, and it is small, but it's not bad - a bit
>on the dead side. I'm several feet from three of the walls, and not
>getting very much reflected sound at all. If anything, reflections
>should cause more anomalies, not fewer. What I hear is an even, steady
>sweep. I would typically expect room reflections to throw that sweep
>askew, not smooth it out.
>
>The quoted test says these speakers are 9 db down at 4 Khz. That's
>incredibly inaccurate. Using a sweep like the one I'm using, one
>should be able to hear a frequency response deviation that gross from
>the next room! But I don't. As I'm sitting right here running it, it
>sounds smooth and even.
Perhaps I could butt in and offer a few comments. First and
most obvious is that we human beans aren't really much good
at *quantitative* measurement. *Qualitative*, sure, and the
final arbiter, but we're so finely tuned, with sliding scales,
log responses, threshold effects, and lots, lots more still-
poorly-understood inherited (from ancestors that *had* those
characteristics, in contrast to neighbors who *didn't* and
so didn't become our ancestors) quirks. Generations of
learning await.
Second is that speaker measurements, when expressed as a single
curve, are only one dimension smaller than a room's response
expressed as a single curve. Speakers at least sit still;
listeners migrate. Both are several dimensions too small
to matter to us.
Third is that the vagaries of loudspeakers' responses interact
with the particular measuring room, even in the simplest case
of a speaker who's response is itself independent of the room.
The speaker is *NOT* is simple point source, and its response
is *NOT* directionless, dimensionless, or frequency-independent.
God and the Devil dwell in the details, because one or the
other (both, if you ask me) has (have!) conspired to make the
real world an endless mystery. Cool though.
Thanks, as always,
Soundhaspriority
>
> God and the Devil dwell in the details, because one or the
> other (both, if you ask me) has (have!) conspired to make the
> real world an endless mystery. Cool though.
>
>
> Thanks, as always,
>
> Chris Hornbeck
> "I'm walking out, between parked cars, with my head full of stars"
> -Elliott Smith
It is rather diabolical, isn't it? Ethan comes up with the perfect
experiment, and Drichard shows up to refute it ;)
Bob Morein
(310) 237-6511
Chris Hornbeck
<now...@nowhere.com> wrote:
>> God and the Devil dwell in the details,
>It is rather diabolical, isn't it? Ethan comes up with the perfect
>experiment, and Drichard shows up to refute it ;)
My contention is not only that the experiment isn't perfect,
but that *no* experiment is perfect. But yeah, irony lives,
fersure, fersure.
But seriously, we're at peril if we depend on experimental
evidence here. We're conditioned to give it weighting that it
seldom deserves. But we're *always, without exception* traveling
mostly blind in the acoustic world. Makes it a tough call, but
that's the cards as dealt. Absolute truth is reserved for the
gods and goddesses. (I know one goddess, but she's somewhat
reticent about the Eternal Mysteries, and that's another story).
We're still in the modeling stage, and our current models
are still fatally flawed. In the case of speakers and rooms
though, I'd say that our (best) models are true enough but
too complex to allow a simple weighting to be applied, or
for a simple weighted "number" or "curve" to be derived.
I'm truly sorry to get so preachy about such an innocent
topic, but it's something that trips up so many younger
folks coming up. Our need for information and our desire
for linear solutions are our (human) greatest strengths
and greatest frailties.
Much thanks, as always,
Chris Hornbeck
"They told me he was bad,
But I knew he was sad,
That's why I fell for ...."
-Barry, Greenwich, Morton
"Look out! Look out! Look out!"
david correia
"Ethan Winer" <ethanw at ethanwiner dot com> wrote:
And all you'll need to hear them all exactly that way is a 50' chair ;>
Lilly Tomlin may still have one.
I think an interesting add-on to the above test would've been to have
tested 2 pairs of each model. Preferably 1 pair out-of-the-box and
another pair already broken in.
David Correia
www.Celebrationsound.com
David Correia
www.Celebrationsound.com
Ty Ford
(in article <18-dnZJpm8yH_6DV...@giganews.com>):
>
> Again I want to stress the importance of nearly all these speakers having
> dips in the "irritating" range between 2 and 4 KHz. As anyone who has ever
> EQ'd music knows, this is where music sounds harsh and unpleasant. So when a
> speaker has an intentional dip there, it can be marketed as sounding smooth
> and silky and all those other audiophile adjectives we all know and hate.
> Likewise with peaks in the "sparkle" range above 6 KHz or the "fullness"
> range below 100 Hz. I expect such trickery with speakers sold to the hi-fi
> market, but I find it very upsetting to see this used to market speakers
> that are supposed to be pro quality.
>
> --Ethan
>
Ethan,
Nice work. Thanks for your efforts.
Very helpful to have data that correlates to what my ears are hearing.
That dip in the 2-4 k range could be just what the heck to do with the
crossover, though, right?
Regards,
Ty Ford
--Audio Equipment Reviews Audio Production Services
Acting and Voiceover Demos http://www.tyford.com
Guitar player?:http://www.youtube.com/watch?v=4RZJ9MptZmU
Scott Dorsey
>Your note confuses me. I have the Event TR8's, and to me they don't
>sound that bad. I can mix on them. But after reading your note, I
>listened to a sweep tone from 200 hz to 6 Khz, and according those
>test results would have expected to hear a huge drop around 4K. But I
>didn't. I didn't hear much of a change at all, in fact they sounded
>very consistent. No drop out at all centered around 4 Khz. I'm
>listening in a nearfield triangle position, only about three feet from
>the speakers, with the speakers about three feet apart.
You won't hear much of a drop out... it's very hard to hear things on a
sweep tone test, and we are talking about effects that are a whole lot
more subtle than bass issues in the room.
But if you want a good test, record a female vocal with an accurate
measurement-style microphone, then play it back on the speakers. Compare
the live sound from the performer's mouth with the sound from the speakers.
Listen, and decide in what way they are different.
I am not sure I would characterize the Event as having a presence dip,
per se, but it does have a lot of midrange nastiness going on.
Richard Webb
Chris Hornbeck wrote:
> But seriously, we're at peril if we depend on experimental
> evidence here. We're conditioned to give it weighting that it seldom
> deserves. But we're *always, without exception* traveling mostly
> blind in the acoustic world. Makes it a tough call, but that's the
> cards as dealt. Absolute truth is reserved for the gods and
> goddesses. (I know one goddess, but she's somewhat
> reticent about the Eternal Mysteries, and that's another story).
WHich is why when all's said and done we still have to use those two
measurement tools on the sides of our heads to see waht works for us.
> We're still in the modeling stage, and our current models
> are still fatally flawed. In the case of speakers and rooms
> though, I'd say that our (best) models are true enough but
> too complex to allow a simple weighting to be applied, or
> for a simple weighted "number" or "curve" to be derived.
THen there's the factor of being mislead by all those charts and specs.
Manufacturers want to show you the numbers that put their products in the best
light. when you're done reading the manufacturer's spec sheets, there's still
those aforementioned tools on the side of your head.
> I'm truly sorry to get so preachy about such an innocent
> topic, but it's something that trips up so many younger
> folks coming up. Our need for information and our desire
> for linear solutions are our (human) greatest strengths
> and greatest frailties.
we want quantifiable criteria, something we can put on a paper that tells us
whether this is right or that's a better fit for us. I find less and less in
this world that I can buy to get work done by just looking at the specs and
other paperwork. SPeakers: I need to listen to them at the showroom, take them
to my working environment, use them, both with known sources and my own work.
IF I'm getting better at making my work translate on them, then they're a fit
for me. IF not ...
HEll I can't even buy a damned microwave oven these days without seeing one not
in the box. yEs my criteria are a bit different than most. I have to know if I
can even operate the damned thing. that's not something that the box or the
text in the catalogue is going to tell me <g>.
Regards,
Richard
... use elspider at bellsouth dot net to email ...
--- timEd 1.10.y2k+
* Origin: Radio REscue net operations BBS (1:116/901)
--- Synchronet 3.15a-Win32 NewsLink 1.85
* Derby City Online - Louisville, KY - telnet://derbycitybbs.com
Ethan Winer
> I listened to a sweep tone from 200 hz to 6 Khz, and according those test
> results would have expected to hear a huge drop around 4K. But I didn't.
A better test is to simply measure with a small diaphragm omni microphone
placed a foot in front of one speaker. Listening to high frequencies with
two ears is not a good test because of standing waves in the room. Even if
you have all of the reflection points treated with absorption, the
difference in arrival times from both speakers creates additional comb
filtering that can confuse what you hear.
If you email me from my web site www.ethanwiner.com I'll send you the graph
for that speaker.
> Could the test you're referring to have been flawed?
Anything is possible, but I don't think so. If all of the speakers David
tested had a common error, or similar shape, that might suggest an error
with his test. But all of the speakers were different, with different
midrange "scoop" frequencies and shapes.
> While tests in an anechoic chamber might have problems below 100 hz,
> shouldn't they be accurate above that?
Sure. In this case David had free access to an auditorium, but not to an
acoustics lab.
> If so, why do other tests not show these same anomalies around the 3K
> center as that particular one?
By other tests do you mean published graphs? A lot of speaker makers use
third octave (or even octave!) averaging to hide the true response. If you
have a link to published graphs for this speaker, please post it and I'll be
glad to have a look.
--Ethan
Ethan Winer
> That dip in the 2-4 k range could be just what the heck to do with the
> crossover, though, right?
I'm not a speaker or crossover expert, but I have to think that whatever the
cause of the dips, it's on purpose to make the speakers sound less harsh.
--Ethan
Ethan Winer
> Which of the test subjects would be your choice to mix with?
Most of my friends have Mackie 824s and I think they're great. I have large
JBL 4430s in my home studio, but I also check things on the Mackie 624s in
my living room system. So you can call me a Mackie fan. They were the
flattest of the ten speakers David tested, and flat is important to me.
--Ethan
GregS
No good speaker builder would have any kind of peak in that range. Some dipping
is sometimes preffered. Flat may also be too harsh.
greg
Soundhaspriority
"GregS" <zekf...@zekfrivolous.com> wrote in message
news:g1mkpe$4u2$1...@usenet01.srv.cis.pitt.edu...
if a monitor has the same dip, it could cause the mix to be accentuated in
that region, which would nullify the purpose of the dip designed into the
listener's speakers.
Bob Morein
(310) 237-6511
drichard
First off, a hearty dose of humble pie for me. Ouch. Mea culpa. My
ears are nowhere near as sensitive as I thought they were. Nowhere
close. I sit here much more humble, but hopefully a little more
knowledgeable.
Last night I did some testing - not very scientific but still
interesting to me. I have several sets of nearfields here: Tannoy
Reveal, Auratone T6 (no, not the little cubes), JBL L19's, and Event
TR8's. My first test was with each set of speakers from about three
feet. I put up an ECM8000 through a fairly neutral preamp, used
generator software to do some sweeps, then recorded them into
Audition, and used the frequency analysis tool to look at them. There
were rather large fluctations in response. Then, I moved in to about a
foot from one of my TR8's and did another full range sweep. Still
rather large fluctations: some 6-9 db or so. So much for the "smooth"
response I thought I heard.
However, there is one interesting thing, which still conflicts with
the aforementioned test results. The TR8 had a low dip centered around
2.5K or so, which is the crossover point. Seems reasonable. But it was
pretty good at 4K. According to that test it should have been down 9
db at 4K. In my test, it wasn't.
Ethan, I think I might be ordering some bass traps soon.
Dean
On May 29, 9:53 am, "Ethan Winer" <ethanw at ethanwiner dot com>
wrote:
> Dean,
>
> > I listened to a sweep tone from 200 hz to 6 Khz, and according those test
> > results would have expected to hear a huge drop around 4K. But I didn't.
>
> A better test is to simply measure with a small diaphragm omni microphone
> placed a foot in front of one speaker. Listening to high frequencies with
> two ears is not a good test because of standing waves in the room. Even if
> you have all of the reflection points treated with absorption, the
> difference in arrival times from both speakers creates additional comb
> filtering that can confuse what you hear.
>
> If you email me from my web sitewww.ethanwiner.comI'll send you the graph
Chris Hornbeck
<DRic...@wi.rr.com> wrote:
>Last night I did some testing -
> I moved in to about a
>foot from one of my TR8's and did another full range sweep. Still
>rather large fluctations: some 6-9 db or so. So much for the "smooth"
>response I thought I heard.
OTOH, these kind of numbers aren't the least bit surprising with
real-world speakers, even measured up close. After you've
measured a few expensive speakers, you'll wonder how anybody can
hear anything at all. Speakers are *all* terrible; it's just
such a tough gig that the imperfections are really big.
For monitoring use, the speakers' imperfections are half of
the problem. Our human ability to listen "through" the speakers
is the other half. Forgiveness is a virtue, but it makes the
translation from listening to our monitors in our monitoring
rooms to imagining the sound in another listener's speakers
and listening room (IMO) more, rather than less, difficult.
We're almost too smart for our own good. Others will likely
feel very differently about this.
>However, there is one interesting thing, which still conflicts with
>the aforementioned test results. The TR8 had a low dip centered around
>2.5K or so, which is the crossover point. Seems reasonable. But it was
>pretty good at 4K. According to that test it should have been down 9
>db at 4K. In my test, it wasn't.
Measurements this close will greatly exaggerate issues of
radiation pattern tilt through the crossover range. You really
can't give them too much importance.
There's a way to measure speakers in this frequency range and
at representative distances, but it involves software that
"windows" the measurement mic's listening time. You send the
speaker a noise burst that's designed to be easily convertible
(by computin' magic) to an impulse (for later use by your computer).
After an appropriate amount of time (distance from speaker to
measuring mic divided by 1130 ft/s) the mic is window'd "on".
Then, after an amount of time that excludes all reflected sounds,
which, natch arrive *after* the direct sound from the speaker, the
measuring time window is closed. This works OK in the frequency
range that you're interested in, and in ordinary rooms. It'd take
an auditorium sized room to to the same trick in the bass range.
Thanks, as always,
Chris Hornbeck
"He sorta smiled, and kissed me goodby,
And tears were beginning to show.
As he drove away, on that rainy night,
I BEGGED him to go slow..."
"Whether he heard, I'll never know...
(NO NO NO NO, NO NO NO)
Look out! Look out! Look out! Look out!"
We Can Do It
> On Thu, 29 May 2008 15:01:37 -0700 (PDT), drichard
> <DRic...@wi.rr.com> wrote:
>
>>Last night I did some testing -
>
>> I moved in to about a
>>foot from one of my TR8's and did another full range sweep.
>>Still
>>rather large fluctations: some 6-9 db or so. So much for the
>>"smooth"
>>response I thought I heard.
MOve the mic a couple inches and repeat. You will see a new
set of fluctuations at different frequencies. Sweeping
speakers in a room is rather useless other than to open you
eyes to how badly this kind of stuff is and how we humans are
not sensative to it . Point being if the musicians were in the
room the same fluctuations would be there. Humans filter out
this "noise" they hear when they listen to what they need to
understand. Microphones dont.
peace
dawg
Scott Dorsey
>Last night I did some testing - not very scientific but still
>interesting to me. I have several sets of nearfields here: Tannoy
>Reveal, Auratone T6 (no, not the little cubes), JBL L19's, and Event
>TR8's. My first test was with each set of speakers from about three
>feet. I put up an ECM8000 through a fairly neutral preamp, used
>generator software to do some sweeps, then recorded them into
>Audition, and used the frequency analysis tool to look at them. There
>were rather large fluctations in response. Then, I moved in to about a
>foot from one of my TR8's and did another full range sweep. Still
>rather large fluctations: some 6-9 db or so. So much for the "smooth"
>response I thought I heard.
Life is like that. And _because_ of those large fluctuations, it's
hard to say that there is necessarily a particular dip in the presence
region.
Some of those fluctuations are going to be room-related. Some of them
are going to be speaker related.
>However, there is one interesting thing, which still conflicts with
>the aforementioned test results. The TR8 had a low dip centered around
>2.5K or so, which is the crossover point. Seems reasonable. But it was
>pretty good at 4K. According to that test it should have been down 9
>db at 4K. In my test, it wasn't.
This would pretty much agree with what I think I hear when I listen to
the Events. Note that you could have room stuff going on at 4K which
counteracts a dip there too. This is why making any kind of generalization
about speakers is usually not so good.
>Ethan, I think I might be ordering some bass traps soon.
Always a good idea, in any room with any speakers.
Chris Hornbeck
<richard.webb@1:116/901.remove-6rz-this> wrote:
>we want quantifiable criteria, something we can put on a paper that tells us
>whether this is right or that's a better fit for us.
NASA calls this the Figure-of-Merit Fallacy. It's almost
tragicly common in modern design and very hard to avoid.
> I find less and less in
>this world that I can buy to get work done by just looking at the specs and
>other paperwork. SPeakers: I need to listen to them at the showroom, take them
>to my working environment, use them, both with known sources and my own work.
>IF I'm getting better at making my work translate on them, then they're a fit
>for me. IF not ...
The very definition of a good tool.
Much thanks, as always,
Chris Hornbeck
"He sorta smiled, and kissed me goodby,
And tears were beginning to show.
As he drove away, on that rainy night,
I BEGGED him to go slow..."
"Whether he heard, I'll never know...
(NO NO NO NO, NO NO NO)
Look out! Look out! Look out! Look out!"
William Sommerwerck
> to do with the crossover, though, right?
A "good" woofer (the KEF B200 being the prototype of such) can get up to
about 3kHz. In a 2-way system, the crossover is going to fall at roughly
this frequency. So one might very well have a dip in this region -- though
whether it's intentional or due to bad design is hard to say.
We Can Do It
"William Sommerwerck" <grizzle...@comcast.net> wrote in
message news:J5ednUkHVfGI1qLV...@comcast.com...
There are a lot of speaker design people who think that a nice
gentle dip in the 2-4k region sounds subjectivly better and
less harsh that a flat response. Either way one must consider
the power response of the speaker over a wide range of angles.
A dip on axis may be evened out by peakes off axis and vice
versa....
peace
dawg
Don Pearce
You never want a crossover to turn over at the same frequency the
woofer is starting to fall because you end up with an undefined
combination of the two. Woofer and tweeter should ideally extend a
good octave beyond the crossover frequency. Even then with a first
order crossover they will only be 6dB down when they start to mess up,
which is perhaps still too loud for comfort.
d
--
Pearce Consulting
http://www.pearce.uk.com
Paul Stamler
news:483f9692...@news.plus.net...
> >
> >A "good" woofer (the KEF B200 being the prototype of such) can get up to
> >about 3kHz. In a 2-way system, the crossover is going to fall at roughly
> >this frequency. So one might very well have a dip in this region --
though
> >whether it's intentional or due to bad design is hard to say.
> >
>
> You never want a crossover to turn over at the same frequency the
> woofer is starting to fall because you end up with an undefined
> combination of the two. Woofer and tweeter should ideally extend a
> good octave beyond the crossover frequency. Even then with a first
> order crossover they will only be 6dB down when they start to mess up,
> which is perhaps still too loud for comfort.
Which is one of the reasons a lot of speaker designers prefer higher-order
crossovers. Less overlap.
Making a 2-way speaker with an 8" woofer is a dicey proposition. They do
tend to roll off beginning at around 3kHz, and their cones start to break up
not far above that. Unfortunately, crossing over a domed tweeter much lower
than 3kHz is asking for trouble in terms of distortion and blown tweeters.
I've done 2kHz, but only with tweeters that had a chambered mid. Even so, I
made sure the crossover was 4th order so it would be down about 40dB at the
tweeter's resonance frequency.
One way around this problem is to use the 3kHz rolloff of the woofer
deliberately, combining with a 12dB/octave rolloff in the crossover for a
total of 24dB/octave (you'd use a 24dB/octave crossover on the tweeter).
That works fairly well, but requires very predictable performance from the
woofers, which boosts the price..And the off-axis response is sometimes more
squirrelly than usual.
Another way out is to use something besides a domed tweeter. Electrostatic
tweeters, or Heil-type tweeters, can have lower resonance frequencies, which
let you do the crossover lower. (See the Adam line for examples.)
Good 2-way speakers are tough. Personally, I usually prefer a 5" two-way
with perhaps some kind of sub.
Peace,
Paul
William Sommerwerck
> >about 3kHz. In a 2-way system, the crossover is going to fall at roughly
> >this frequency. So one might very well have a dip in this region --
though
> >whether it's intentional or due to bad design is hard to say.
> You never want a crossover to turn over at the same frequency the
> woofer is starting to fall because you end up with an undefined
> combination of the two. Woofer and tweeter should ideally extend a
> good octave beyond the crossover frequency. Even then with a first
> order crossover they will only be 6dB down when they start to mess up,
> which is perhaps still too loud for comfort.
Can you say "acoustic Butterworth"?
Don Pearce
Hang on, I'll try.
Yup, I can.
Since we are dealing with masses and springs, which are effectively
equivalent to inductors and capacitors, then of course we can. Bu a
woofer at the upper extreme of its range doesn't behave like that. The
cone is starting to break up and all sorts of modes are ringing away.
You want the crossover to be happily attenuating well before that happens.
d
Arny Krueger
Paul Stamler's comments resonate with me, as I recall my last efforts to
build a high performance 2-way speaker system. Implementing an
acoustic/electrical Linkwitz-Riley crossover was the formal goal. Many
real-world situations will work together to rain on this parade.
A Linkwitz-Riley crossover is formed of 4 butterworth filters, two high-pass
in series, two low-pass in series. The first cut at this problem would be to
form each pair of Butterworth filters from an acoustic section based on the
roll-off of the driver, and an electrical section, composed of a capacitor
and a inductor.
First off, as a rule tweeters have low end response that goes well below the
limits of their low end power handling. So, you don't want to try to cross
them over anywhere near their low end roll-off point. Thus, your
acoustical/electrical plans are dead - you'll have to do it all
electrically. The remaining problem is that the driver is anything but
resistive, so it will take even more work to actually create an effective
cascade of 2 Butterworth filters.
Secondly, there are few woofers that haven't started breaking up by the time
they are rolling off. Then, few of the woofers that aren't breaking up too
badly at that point, have a true Butterworth response. The remaining problem
very familiar. Again the driver is anything but resistive, so it will take
even more work to actually create an effective cascade of 2 Butterworth
filters.
The first simplifying assumption is to bi-amp which instantly solves the
driver impedance curve problems...
GregS
Many speakers make use of all effects. Using the combination of natural response, impedance, and
crossover rates is real. The only thing undefined is variations in drivers. 6 dB down might be not so much
of a problem as phase shifts and off angle response.
greg
Scott Dorsey
In general, all speaker systems will be at least a little screwy in the
crossover region. And in general having screwyness at 1KC is a whole lot
worse than having screwyness at 3KC....
The Altec A-7 system has the crossover point at 500 Hz, and it is really,
really screwy at the crossover point. But it's not all that offensive,
really, because of the frequency where they picked it. It gets still less
offensive with a higher order crossover too. Of course, maybe all of the
horn honk and clang masks it...
Ethan Winer
> First off, a hearty dose of humble pie for me. Ouch. Mea culpa.
Not at all! I didn't know this stuff 10 years ago, and most people - even
pros - have no idea how severe the peaks and nulls are. And how bad the
off-axis response is. It puts a whole new light on how silly some people
are to obsess over stuff like clock jitter or a microphone preamp being 1/4
dB down at 22 KHz. :->)
You already got great explanations from Chris, dawg, and Scott. If I didn't
link this before, here's a short article (about a different issue) that
shows how problematic it is to measure the response in a room at mid and
high frequencies:
http://www.ethanwiner.com/believe.html
> Ethan, I think I might be ordering some bass traps soon.
Great. You know where to find me. :->)
--Ethan
Paul Stamler
news:7GN%j.2573$102...@bgtnsc05-news.ops.worldnet.att.net...
> Making a 2-way speaker with an 8" woofer is a dicey proposition. They do
> tend to roll off beginning at around 3kHz, and their cones start to break
up
> not far above that. Unfortunately, crossing over a domed tweeter much
lower
> than 3kHz is asking for trouble in terms of distortion and blown tweeters.
> I've done 2kHz, but only with tweeters that had a chambered mid.
But he meant to write:
"...but only with tweeters that had a chambered back."
Sorry for the error.
Peace,
Paul
Paul Stamler
news:8Z-dnWLb4dVSZqLV...@comcast.com...
>
> First off, as a rule tweeters have low end response that goes well below
the
> limits of their low end power handling. So, you don't want to try to cross
> them over anywhere near their low end roll-off point. Thus, your
> acoustical/electrical plans are dead - you'll have to do it all
> electrically. The remaining problem is that the driver is anything but
> resistive, so it will take even more work to actually create an effective
> cascade of 2 Butterworth filters.
Difficult with a tweeter, but doable if your tweeter has its resonance at
least 2 octaves below the crossover point.
> Secondly, there are few woofers that haven't started breaking up by the
time
> they are rolling off. Then, few of the woofers that aren't breaking up
too
> badly at that point, have a true Butterworth response. The remaining
problem
> very familiar. Again the driver is anything but resistive, so it will
take
> even more work to actually create an effective cascade of 2 Butterworth
> filters.
Fixing the non-resistive character of a woofer's not all that hard; a Zobel
RC network across the terminals will flatten it reasonably well. You need to
create the Zobel by trial and error, but you can get the impedance to stay
within an ohm across the range where it matters.
The break-up problem's more serious, but there are woofers that don't start
breaking up until an octave or so above the rolloff frequency. They ain't
cheap, though. KEF made some, Focal still does.
Then there's the beaming issue...
Peace,
Paul
William Sommerwerck
> in the crossover region. And in general having screwyness at
> 1KC is a whole lot worse than having screwyness at 3KC....
Agreed. This is why I prefer (for cone-type systems) 3-way speakers.
Scott Dorsey
Yeah, but then you have screwyness in two different places. Sometimes
that's even worse. Sometimes it's better, too.
Chris Hornbeck
>William Sommerwerck <grizzle...@comcast.net> wrote:
>>> In general, all speaker systems will be at least a little screwy
>>> in the crossover region. And in general having screwyness at
>>> 1KC is a whole lot worse than having screwyness at 3KC....
>>
>>Agreed. This is why I prefer (for cone-type systems) 3-way speakers.
>
>Yeah, but then you have screwyness in two different places. Sometimes
>that's even worse. Sometimes it's better, too.
I pretty much fall into the camp that says you need to make
the voice range work first, and then to build out from there.
This argument leaves you with two crossovers (for lil' round
pistons, anyway) and allows you to deceive yourself that they're
in innocuous territory.
Self-deception is probably *the most important* part of
loudspeaker design ( "All else is gaslight" ) and not to
be lightly discounted.
William Sommerwerck
news:g1qa5u$saa$1...@panix2.panix.com...
> William Sommerwerck <grizzle...@comcast.net> wrote:
>>> In general, all speaker systems will be at least a little screwy
>>> in the crossover region. And in general having screwyness at
>>> 1KC is a whole lot worse than having screwyness at 3KC....
>> Agreed. This is why I prefer (for cone-type systems) 3-way speakers.
> Yeah, but then you have screwyness in two different places. Sometimes
> that's even worse. Sometimes it's better, too.
The theory (as suggested by the following poster) is that a three-way system
uses a single driver to cover the most-important part of the sound -- the
midrange. And you're operating it over the "good" part of its range, not at
the upper end of a large driver.
People sometimes quote Henry Kloss as saying that two-way systems are better
than three-way. He did, but I believe he was talking about the need to
design at a particular price point.
William Sommerwerck
> loudspeaker design ( "All else is gaslight" ) and not to
> be lightly discounted.
Which is one of the reasons I own Apogees. Their designer -- Leo Spiegel --
is clinically deaf.
Teddy D'Bear
were not selected for a transparent transition between speakers. Or the
knees of the crossover was not calculated properly, so while the low starts
to fall off at, for example, 500 Hz, the mid does not kick in till 600,
leaving a dip in the response curve.
I've also seen older speaker crossovers change due to the aging of
electrolytic capacitors. One of the reasons I have renewed interest in
building a new set of speakers.
Quoting "Ethan Winer" <ethanw at ethanwiner dot com>:
>
>I'm not a speaker or crossover expert, but I have to think that whatever the
>cause of the dips, it's on purpose to make the speakers sound less harsh.
>
>--Ethan
>Ty,
>
>> That dip in the 2-4 k range could be just what the heck to do with the
>> crossover, though, right?
We Can Do It
William:
Do you have a cite on Leo's deafness? I thought I was the
first deaf speaker designer and would like to start a club.
But really I am very interested. Google was not of much help
here.
peace
dawg
>
>
William Sommerwerck
news:SYGdnVrykoHQ7NzV...@earthlink.com...
> "William Sommerwerck" <grizzle...@comcast.net> wrote in
> message news:U9idneFrsYOc39zV...@comcast.com...
>>> Self-deception is probably *the most important* part of
>>> loudspeaker design ( "All else is gaslight" ) and not to
>>> be lightly discounted.
>> Which is one of the reasons I own Apogees. Their designer --
>> Leo Spiegel -- is clinically deaf.
> Do you have a cite on Leo's deafness? I thought I was the
> first deaf speaker designer and would like to start a club.
> But really I am very interested. Google was not of much help
> here.
I don't know where I heard it. It might have been Leo himself -- I'm pretty
certain he or someone else told me his hearing was damaged during WWII.
I have spoken with him on the phone several times, and there was indication
he had any trouble understanding me.
William Sommerwerck
> was indication he had any trouble understanding me.
Sorry... no indication.