#42241 - 10/13/02 06:15 PM
Re: A Listening Challenge
|
Desperado
Registered: 12/11/01
Posts: 1054
Loc: Santa Clara, CA
|
i have modeled every speaker in my database and found that implementing a high pass filter before the amp (at 40, 60, 80 and 100hz.) causes a huge 6db hump at the crossover point. Huh? When I went through the process of eq'ing my sub, I plotted my sub alone, my mains alone, and the sub + mains, from 20 Hz to 98 Hz (Autosound 2000 test CD) through the 60 Hz crossover point. I never saw a 6 dB rise. In fact, most crossovers I'm familiar with, set the high pass filter for the mains at -3 dB and the low pass filter for the sub at - 3dB at the crossover freq so that you get *flat* freq response through the crossover point...
_________________________
If it's not worth waiting until the last minute to do, then it's not worth doing.
KevinVision 7.1 ... New and Improved !!
|
Top
|
|
|
|
|
#42242 - 10/13/02 06:55 PM
Re: A Listening Challenge
|
Desperado
Registered: 08/19/02
Posts: 430
Loc: charlotte, nc usa
|
[QUOTE]Originally posted by soundhound: [B]Bosso:
I wonder what your take is on this: what ever happened to 'acoustic suspension' speakers in the transistion to Home Theatre? Those speakers were certainly capable of going down to 30-40hZ easily and cheaply, and with modest cabinet dimensions. _________________________________
acoustic suspension, or sealed box speakers, are everything you say except the 'cheaply' part. the driver has a low Fs due to the greater cone mass required, super-high excursion, a large magnet structure, long voice coil and a cone that won't distort when subjected to the radical forces necessary at low freqs. it costs a pile-o-dough to build a good one. _________________________ Can you think of any reason why they would not have been a better solution than a vented box, which can flail it's woofer if presented the wrong frequency and level of bass? __________________________
vented enclosures aid the driver to acheive a response lower than Fs. and, because the vent takes over at the lower freqs, the driver is less taxed. it is usually more efficient with less distortion due to less excursion...and...is cheaper to build. yes, it can unload at lower freqs, but this is easily overcome by inserting...yes...a high pass filter before the amp. (the filter Fx and Qx must be specific to the driver, is the catch). bottom line is, vented drivers/enclosures are cheaper to produce with better results. the car audio people use acoustic suspension more often because most vehicle interiors have an acoustic response that rises at 12db/ octave from about 50hz, giving the sealed box (12db/octave roll off, as bstan said) a flat response down low. (plus, they seem to have unlimited funds for drivers, caps, batteries, alternators, amps, etc.)
_________________________
"Time wounds all heels." John Lennon
|
Top
|
|
|
|
|
#42243 - 10/14/02 12:06 AM
Re: A Listening Challenge
|
Desperado
Registered: 01/14/02
Posts: 1176
|
One big problem is deciding what to measure when looking for or trying to avoid cross-over 'humps'. Do you want constant voltage, constant power, etc.
_________________________
Charlie
|
Top
|
|
|
|
|
#42244 - 10/14/02 12:29 AM
Re: A Listening Challenge
|
Gunslinger
Registered: 05/24/02
Posts: 279
Loc: Mountain View, CA, USofA
|
One would want constant spl (sound pressure level), which would correspond to constant power (assuming reasonably flat speaker frequency response). I remember years ago a discussion about active crossover design. One design used active circuitry to get a high pass or low pass signal that was used for the appropriate driver(s). This same signal was subtracted from the full range signal to get the complimentary low or high pass signal. Example: take a full range signal and using Rs and Cs and whatever active device of choice, get, for example, a high pass filter which can be used for a tweeter or a low pass limited speaker system. Take this same low pass limited signal and subtract it from the full range signal and you have a (derived) low pass signal which can be use appropriately.
I recall the discussion of this derived ciruit: the general consensus was that when the active signal was summed with the derived signal, proper voltage was the result. One wants proper (flat) power, which this derived approach does not give.
I have my ICBM between my power amps and pre amp. The only bass management done in my system is done by the icbm is it is used to send the bass below 60 or 80 Hz (I am still playing around with it) to the subs.
Paul
------------------ the 1derful1
_________________________
the 1derful1
|
Top
|
|
|
|
|
#42245 - 10/14/02 12:58 AM
Re: A Listening Challenge
|
Desperado
Registered: 08/19/02
Posts: 430
Loc: charlotte, nc usa
|
Originally posted by Kevin C Brown: Huh? When I went through the process of eq'ing my sub, I plotted my sub alone, my mains alone, and the sub + mains, from 20 Hz to 98 Hz (Autosound 2000 test CD) through the 60 Hz crossover point. I never saw a 6 dB rise. In fact, most crossovers I'm familiar with, set the high pass filter for the mains at -3 dB and the low pass filter for the sub at - 3dB at the crossover freq so that you get *flat* freq response through the crossover point... ___________________________________________ Questions: 1.) your sub+mains were flat from 20-98hz? 2.) did you move the mic around the room for each plot point? 3.) is the room acoustically corrected? 4.) did you use a 1/3 octave real time analyzer, or a rs spl meter? points: 1.) we're talking about a low frequency high pass filter, not a crossover network. 2.) depending on the Q of the filter, which to me at least is unknown, a hump or a hole may be the case, but NEVER flat when the filter frequency is above Fs. 3.) my software contains only info for 2nd order active high pass filters. i am unfamiliar with digital filters (and would like any info or links that might clue me) 4.) the 950 specifications list only analog filter info, i can find not one word about 'dsp precision bass management' in the manual...i would also appreciate any info anyone may have on this subject. 5.) i am basically clueless to any test or plot anyone may have made on there setup at home. i only know what my software tells me, and it virtually never lies. i would surely like to know the answer to this part of the bass management puzzle. 6.) in the meanwhile, i do not use bass management at all. i plays it da way dey mixes it. also, if i fail soundhound's 12bit vs 16 bit test, i'm gonna listen to only AM radio for the rest of my life, and all of this will be moot. (i think he rigged the disc)
_________________________
"Time wounds all heels." John Lennon
|
Top
|
|
|
|
|
#42246 - 10/14/02 01:22 AM
Re: A Listening Challenge
|
Gunslinger
Registered: 05/24/02
Posts: 279
Loc: Mountain View, CA, USofA
|
Well, I've listened to the sound challenge .wav file a few times now.
I have not been much of a jazz fan, but I enjoyed this music very much. Maybe it is time to get a new/better CD player, one with SACD capability perhaps and to start exploring jazz. Then I will listen again and see, oops, hear what I can hear.
Some general comments/questions.
Was this taken from a vinyl record?
I ask this because at times, especially on trumpet "blasts", I can hear what sounds like to me as either record damage or cartridge mistracking. Also, in the very last few seconds, as the music is fading out, I can hear what sounds like to me as line induced hum (60Hz or 120Hz hum). Bings I do not know when this recording was recorded, I have no idea what limitations the recording equipment.
I have the most confidence in identifying the transitions from 8 bits to 16 bits and almost as confident when going from 16 bits to 8 bits. 8 bit transitions. Of course these would be more noticeable than four bit transitions, of which I am less confident.
The background hiss level seemed higher and less random in the 8 bit sections. The decay of tapped cymbals was harsher and truncated. Yada-yada-yada.
(results were here but I removed them so as to not influence others)
To describe things with impercise terms:
The main sounds of this music were not all that much changed when going from one bit depth to another. The sounds around the main sounds, however, were changed. Fewer bits, more blockier and harsh sounding with a less smooth decay. To me it sounds possible that with fewer bits, a lot of the low level detail is lost. Instead, there is false details created by the cruder 12 bit or 8 bit word depth.
Even is one is not able to consciously recognize a lower or higher bit depths, I supect listener fatigue will set in sooner with a lower bit depth.
I personally find that when I visit a high-end audio shop (or a low end), I usually start getting a headache in a few minutes in most cases. The times I don't is when I am listening to a good (good for me, anyway ... was it good for you?) system. I have gotten headaches from some very expensive and highly regarded equipment.
Paul, stilespj@mindspring.com
------------------ the 1derful1
[This message has been edited by Paul J. Stiles (edited October 14, 2002).]
_________________________
the 1derful1
|
Top
|
|
|
|
|
#42247 - 10/14/02 02:19 AM
Re: A Listening Challenge
|
Desperado
Registered: 12/11/01
Posts: 1054
Loc: Santa Clara, CA
|
BoB- 1) As flat as I could get them by eq'ing out room induced peaks. Know how I knew there weren't any 6 dB rises due to the interaction at the crossover point? Because there weren't any major differences between sub alone, vs mains alone, vs sub+mains. Just the rises and falls due to the slopes on the filters applied to each. (Really neat to see in the real world, btw.) 2) I averaged the responses from where I sit, and the 2 other best listening positions. (Yes, I know what I'm doing.) 3) Very accidentally, I have a "good" room for acoustics. One side wall is open to the kitchen, so no reflections there. The other side wall is dominated by a sliding glass door, with a curtain in front of it. Carpeted floor. I even went so far as to put a make shift absorber in the corner of the room with the sidewall. One of those cardboard fabric bolts, covered by a few layers of fabric, propped in the bottom of the corner. 4) RS meter, every Hz from 20 to 98 Hz. With correction. Although to be honest, all the correction does is to change the tilt of the readings, doesn't hide any strange perturbations in the measurements. Maybe instead of spending so much time trying to model these things with software, you should get out and do some real world measurements...
_________________________
If it's not worth waiting until the last minute to do, then it's not worth doing.
KevinVision 7.1 ... New and Improved !!
|
Top
|
|
|
|
|
#42248 - 10/14/02 03:51 AM
Re: A Listening Challenge
|
Desperado
Registered: 04/10/02
Posts: 1857
Loc: Gusev Crater, Mars
|
Kevin:
Just wondering. Do you have access to, or do you have an RTA program on your computer?
|
Top
|
|
|
|
|
#42249 - 10/14/02 09:33 AM
Re: A Listening Challenge
|
Gunslinger
Registered: 04/02/02
Posts: 32
Loc: Kansas City, MO, USA
|
Well, I fully expected to post my results today, but unfortunately I could not get the cd to play. I have two dvd/cd player but both are Panasonic based (one is a Panasonic and the other is an EAD). I guess I could try it on my computer, but that I don't have very revealing speakers on my computer.
Let me know who to forward the CD to at rbessinger@deloitte.com
|
Top
|
|
|
|
|
#42250 - 10/14/02 11:21 AM
Re: A Listening Challenge
|
Desperado
Registered: 01/14/02
Posts: 1176
|
Originally posted by Paul J. Stiles: One would want constant spl (sound pressure level), which would correspond to constant power (assuming reasonably flat speaker frequency response). Actually there is a lot of discussion of this and related topics. Apparently it's not quite as cut and dried as that. Keep in mind the 'frequency response' of a speaker will say something like '1w/1m' but it's really a fixed voltage, not a fixed power, measurement. In designing crossovers for custom speakers there is quite a lot to take into consideration and the 'perfect' crossover seldom is really perfect.
_________________________
Charlie
|
Top
|
|
|
|
|
|
0 registered (),
53
Guests and
1
Spider online. |
Key:
Admin,
Global Mod,
Mod
|
|
|
8,717 Registered Members
88 Forums
11,331 Topics
98,703 Posts
Most users ever online: 677 @ 09/27/24 06:41 PM
|
|
|
|