Setting up Speakers with an Equaliser

Gronich

Member
Jun 18, 2000
145
0
0
I would like to setup up my speakers to sound as accurately as possible to the original sound.

The setup:

Sound card: Fortissimo III 7.1
Speakers: Logitech Z-680

I'm connected using the optical connection.

I intend to use the supplied equaliser with the Sound card to set the baseline for all output i.e. DVD-Video and Music.

I have invisaged that the I must create a tone for each frequency the equaliser uses i.e. 60, 170, 310Hz etc. at a specific Decibel and then using a sound level meter adjust the equaliser until it reaches the desired volume - repeat the process for each frequency.

I don't know where I can get hold of a sound level meter but I have wonder if I could use my second PC with a microphone to measure the volume.

Any help would be appreciated or If you could point me in the direction of a good audio related site...
 

f95toli

Golden Member
Nov 21, 2002
1,547
0
0
It would be better to use a microphone with a reasonably flat response such as the Behringer ECM8000 to record the response.
However, it still won't work; equalizers can sometimes help but can not do what you want; at best you can reduce the effect of the room. Google "Behringer DEQ2496" or read a few threads on a audio-forums (diyaudio.com, audioasylum), that should give you some ideas about how this is done and the limitations of this technique.
 

Matthias99

Diamond Member
Oct 7, 2003
8,808
0
0
If you have a DVD drive in your system, there are a number of audio-video calibration DVDs available (the most popular being AVIA and Digital Video Essentials). Most include sound calibration utilities with appropriate test tones. You can 'eyeball' the sound levels (so to speak), but you really need a sound meter (available at Radio Shack, or basically any electronics or music store) for accurate results. You could use a second PC, but you'd need a good mic, and the sound meter's probably still more accurate.

And as for sites, www.avsforum.com has more audio and video information than you'll know what to do with.
 

Eddieo

Senior member
Nov 17, 2004
329
0
0
Not possible to get it accurately close with that setup. You probably would want professional studio equipment using a phantom powered, condenser mic.
 

f95toli

Golden Member
Nov 21, 2002
1,547
0
0
The ECM8000 I recommended IS a phantom powered condenser mic, meaning you also need a mic amplifier. Unfortunately this also means it is quite expensive (something like $150 for the mic+amp) for it is still a bargain compared to pro equipment.

 

Tiamat

Lifer
Nov 25, 2003
14,068
5
71
Originally posted by: oneshot47
If you were that serious about your sound you wouldnt be listening to Logitech Z-680's.

Thats true, however he is probably just trying to "maximize" what he was able to buy for the money he spent.

In any case, OP, it takes quite a while to make sure the speakers are calibrated correctly due to their positioning in your room. Creative cards have a feature to help with this. With the quality of your equipment, it might be more feasible if you just "do it by ear" on test tones. It would obviously be more advantageous if you could find an equilizer with more adjustments so you could get a better "flat line" across the spectrum.
 

Howard

Lifer
Oct 14, 1999
47,982
11
81
Originally posted by: PumpActionWalrus
Originally posted by: oneshot47
If you were that serious about your sound you wouldnt be listening to Logitech Z-680's.

B&W all the way
You don't necessarily need B&W speakers to get good sound.
 

oneshot47

Senior member
Aug 6, 2004
435
0
0
Originally posted by: Howard
Originally posted by: PumpActionWalrus
Originally posted by: oneshot47
If you were that serious about your sound you wouldnt be listening to Logitech Z-680's.

B&W all the way
You don't necessarily need B&W speakers to get good sound.

But you do need something better than some crappy computer speakers. Hes not going to get any appreciable benefit by messing with an equalizer on that. If it were that easy to tweak an equalizer to get any speakers to reproduce the original recording exactly then we'd all be using the internal speaker on our computers (slight exaggeration).
 

itachi

Senior member
Aug 17, 2004
390
0
0
in my opinion.. there's really no point. since your speakers are so small, moving your head to the left/right a few inches will change the curve significantly.. and i'm guessing you're no more than 2 feet from your speakers, so.. running the samples at a single point won't be sufficient, the distance from the center of the listening point and either ear is significant. good chance that the waves are interferring destructively at the place you measure (lower volume), and constructively where your ears would be (higher volume).

but if you wanna go ahead and do it anyway, no stopping you. if you do.. when you run the tones, run it at every frequency once then again at -x db. calculate the general difference, and factor that in when you're equalizing. otherwise, you might be boosting a band too much.

if you do decide to go with it.. and need the tones.. just tell me the frequencies you need.
 

KoolAidKid

Golden Member
Apr 29, 2002
1,932
0
76
Radio shack sells a cheap sound level meter that is probably adequate for this purpose. I would offer to measure the frequency response of your speakers for you, but I doubt that you'd want to ship them to me.
 

Howard

Lifer
Oct 14, 1999
47,982
11
81
Originally posted by: itachi
in my opinion.. there's really no point. since your speakers are so small, moving your head to the left/right a few inches will change the curve significantly.. and i'm guessing you're no more than 2 feet from your speakers, so.. running the samples at a single point won't be sufficient, the distance from the center of the listening point and either ear is significant. good chance that the waves are interferring destructively at the place you measure (lower volume), and constructively where your ears would be (higher volume).

but if you wanna go ahead and do it anyway, no stopping you. if you do.. when you run the tones, run it at every frequency once then again at -x db. calculate the general difference, and factor that in when you're equalizing. otherwise, you might be boosting a band too much.

if you do decide to go with it.. and need the tones.. just tell me the frequencies you need.
Generally, the off-axis response of smaller drivers tend to be more linear compared to larger drivers, given that both kinds have flat on-axis response, unless you meant phase cancellation in your first point, and it's basically negligible in-room (IIRC).
 

f95toli

Golden Member
Nov 21, 2002
1,547
0
0
Ths size of the speakers has absolutely nothing to do with this. Neither the size of the speker nor the size of the elements can be directly used to draw any conclusions about what a speaker can or can not do.

 

jdpal

Junior Member
Nov 24, 2004
3
0
0
If the original poster is just trying to determine an EQ setting that would theoretically give the speakers a flat response curve, why not just look up the response curve of the speakers, and work from that with EQ to compensate and make it a flat line? Assuming the speakers themselves can handle something close to 20-20 (which may not be a safe assumption at all, I obviously haven't checked), a flat response would be possible, albiet probably only at a low volume level.
 

oneshot47

Senior member
Aug 6, 2004
435
0
0
Originally posted by: jdpal
If the original poster is just trying to determine an EQ setting that would theoretically give the speakers a flat response curve, why not just look up the response curve of the speakers, and work from that with EQ to compensate and make it a flat line? Assuming the speakers themselves can handle something close to 20-20 (which may not be a safe assumption at all, I obviously haven't checked), a flat response would be possible, albiet probably only at a low volume level.

I highly doubt logitech publishes a response curve for their computer speakers. Also, its not nearly that simple. In theory, yes that sounds like a good idea, but in reality its much more complex. For one thing, that doesnt even take into account the acoustics of the room or the performance of the amplifier or the source. A lot of money and research has gone into developing transparent loudspeakers (though you wouldnt know it by walking into bb or greggs) and its not something that messing with an eq can do. The op's time and money would be much better spent on a real stereo system than in trying to find the magic panacea to some crappy computer speakers.
 

itachi

Senior member
Aug 17, 2004
390
0
0
Generally, the off-axis response of smaller drivers tend to be more linear compared to larger drivers, given that both kinds have flat on-axis response, unless you meant phase cancellation in your first point, and it's basically negligible in-room (IIRC).
hmm.. didn't know that. then wouldn't his speakers be less linear than a pair of 2-way speakers? it lacks a tweeter.. and its driver is a cone, so the sound dispersion for the higher freq. is more directional.. no?
Originally posted by: f95toli
Ths size of the speakers has absolutely nothing to do with this. Neither the size of the speker nor the size of the elements can be directly used to draw any conclusions about what a speaker can or can not do.
sounds like bose propaganda.. you may not be able to draw conclusions on timbre, but you can about the band that the speaker can produce. a 3-in driver just wont cut it for freq. under 200 hz.
 

f95toli

Golden Member
Nov 21, 2002
1,547
0
0
For the record: I do not like Bose and I no fan of "designer" HiFi, personally I own several B&W speakers including a pair of 803, they are large floor-standing speakers (they are also quite expensive, I bought them used).

Anyway, you are right that it is EASIER to get good bass reponse from large speakers but that does not mean that it is impossible to get good low bass from small speakers. The problem is that small speakers with good bass-reponse usually have a very low sensistivivty which means you need a LOT of power.
A small "High-end" speaker is en extremely difficult load for a power-amp, floorstanders are usually much easier (for some large floorstanders it is enough with a few watts, they are often used in combination with tube-amps)


Some of you might have seen the new line of Beolab speakers from B&O (it is the first series of "real" high-end speakers from B&O in a long time, usually B&O is just expensive because of the design).Beolab 2 is a small speaker with relatively good bass resonse. This speaker would have been impossible to build just a few years ago.
The key here is to use a good class-D power amps, the sensitivity is very low but that is not a problem as long as you have enough power. As far as I understand they also use DSP compensation in order to extend the bass response (a more advanced version of the active compensation used in dipole speakers).


 

jdpal

Junior Member
Nov 24, 2004
3
0
0
Originally posted by: oneshot47
Originally posted by: jdpal
If the original poster is just trying to determine an EQ setting that would theoretically give the speakers a flat response curve, why not just look up the response curve of the speakers, and work from that with EQ to compensate and make it a flat line? Assuming the speakers themselves can handle something close to 20-20 (which may not be a safe assumption at all, I obviously haven't checked), a flat response would be possible, albiet probably only at a low volume level.

I highly doubt logitech publishes a response curve for their computer speakers. Also, its not nearly that simple. In theory, yes that sounds like a good idea, but in reality its much more complex. For one thing, that doesnt even take into account the acoustics of the room or the performance of the amplifier or the source. A lot of money and research has gone into developing transparent loudspeakers (though you wouldnt know it by walking into bb or greggs) and its not something that messing with an eq can do. The op's time and money would be much better spent on a real stereo system than in trying to find the magic panacea to some crappy computer speakers.

Absolutely, it wouldn't take into account the acoustics of the room. But, if you just wanted to give the speakers a close to linear response, couldn't that be achieved by EQ?

I'm quite new to all of this, so if I'm completely wrong, please excuse me, although I'd love to learn.