Help me WmAx; you're my only hope!

[sploo]

Hi Buckle-meister,

Just like Ethan, I didn't have much luck with ETF's mixer (in fact I toasted the audio setup on a nVidia Soundstorm based motherboard ). I'd strongly suggest using the standard Windows mixer.

If memory serves me right, the image I've just put together should help you get your cabling setup right.

I wouldn't bother doing the levels check, I had so much 'fun' with it I found it was much better just to setup in the working layout, keep your amp volume level low, and start actual recordings (1 second, sweep). Just play with the levels till you get it right.

The attached pic (hopefully) shows that you connect L out to L in (via a double female connector), R in goes to the SPL meter, and R out goes to your amp.

On Ethan's advice, I use a splitter at the amp to ensure I get the same sound from both speakers (it does work much better).

As for SPL settings, I did read something about the RS meter that indicated that both the weighting and response speed matter - I set to C, and FAST.

 

[Steve1000]

This is GREAT info. Fascinating. Thanks. Makes a lot of sense to me subjectively -- I often notice with live music it seems to me that the high end is way over-the-top (i.e., extremely bright) as opposed to a typical hi-fi set-up. Also, I've noticed that as the volume gets louder, the higher-amplitude high frequency (i.e., relatively bright) speakers can go from seeming to sound comparatively better to me to seeming to sound not so wonderful anymore.

Why do some people like the color blue? That's why I recommend that you try to correct it with the reciever equalizer and see if it's an improvement. I, for example, despise a flat response into the treble, when measured at the listening position, for most music, but that's just me, although this view does see to be supported by at least some prior perceptual research [1].
-Chris

[1]

5 study conclusions, of the average in-room response of speakers considered to be 'flat' in response, and that measure a flat power response in close proximity.

http://www.etymotic.com/ephp/er4-ts.aspx

 

[sploo]

Hi Steve,

I certainly found that I could 'get away' with playing simple, quiet music with no room treatment. Anything complex, certainly a live concert with crowd noise, would be truly terrible.

I'd like to get the first reflection absorbers working down a little further, but they certainly have a very positive impact on the sound right now, so I'm just grateful for what I've got!

 

[Buckle-meister]

Just like Ethan, I didn't have much luck with ETF's mixer...I'd strongly suggest using the standard Windows mixer.

I agree. I just couldn't seem to get it working. After a lot of messing about, I find that using WAVE volume control, I can adjust the 'Level check' dB figures to between -10 and 0 as per the 'manual'. I'm guessing that WAVE is internally generated, because even when I pull the cable loop out completely, it still seems to work.

That aside, the graphed response (refer to attachment) seems to be very good (I think). At least, it matches what the 'manual' says it should, so I think I can assume that everything is ok....

By the way, how do you attach 'thumbnails' as opposed to 'attachments'?

EDIT: Doh!!!

The attached pic (hopefully) shows that you connect L out to L in (via a double female connector), R in goes to the SPL meter, and R out goes to your amp. On Ethan's advice, I use a splitter at the amp to ensure I get the same sound from both speakers (it does work much better).

1. Is the attached photo 'adapter' what you mean by 'double female connector'?

2. If the answer to 1 above is yes, then can you confirm that I have interpreted things correctly?;

I'll plug the 1/4" into the computer's headphones jack and the L/R RCA end of that cable will plug into both jacks of one side of the adaptor. From the other side, the Left can be connected with a single cable to my computer's Left-in jack, and the RS SPL meter will be connected, again by a solitary cable, to my computer's Right-in jack. This would only leave the connection to be made, via the remaining Right from the adaptor to the amp.

This is where I'm a little confused. If I'm taking a mono signal (the Right) to the amp, why am I splitting it into two mono signals to put through the amp's stereo inputs? I know you wrote:

On Ethan's advice, I use a splitter at the amp to ensure I get the same sound from both speakers (it does work much better).

...but why do I want the same sound from both speakers? I'll want to record the frequency response from the L/R speakers independantly won't I? And assuming that I do need to do as Ethan and yourself advise, how do I split the Right cable's signal into two? What kind of adapter? Do you have any pictures?

I set to C, and FAST.

Yeah, I suspected it would have to be 'fast'; the frequency sweep would never be recorded accurately if set to 'slow'.

Regards

p.s. I'm getting there! Slowly but surely! Thanks for all your help so far everyone.

 

[Buckle-meister]

I certainly found that I could 'get away' with playing simple, quiet music with no room treatment. Anything complex, certainly a live concert with crowd noise, would be truly terrible.

I find this incredible. Even in my own room, or any room I've set the hi-fi up in for that matter, I've never thought the sound was so absolutely appalling. And as for friends feeling physically ill, well, I just cannot comprehend how bad it must have sounded.

Regards

 

[WmAx]

Very odd. I can open the attachment no bother, on two different computers.

For the 20Hz tone, the attachment shows the sine wave centred along the horizontal axis at -90dB, and it's peaks and nulls reaching -3dB (the vertical axis increases from the horizontal axis each way).

Sorry for being really thick here, but I still have no idea what to do.

Am I comparing the -3dB peak/null value with the peak/null values from other frequencies? And if so, what frequency do I make them relative to?

Regards

I don't have Word on this computer, instead I am usingw Atlantis Nova to view the documents[it opened the last ones with no problem], and it simply refuses to open the last document.

What I wanted to know was the peak dB values compared in the different files. I just wanted to verify that the test CD was compensated for C weighting of the SPL meter. It sounds as if it is not, if the peak value is -3dB at 20Hz. If you look at 100Hz or higher frequency signal file, is should be roughly 6db higher in amplitude than the 20 Hz file, if it's C weighted.

BTW, here is a weighting cross-reference calculator if you need such:

http://www.measure.demon.co.uk/Acoustics_Software/a_weight.html

-Chris

 

[WmAx]

Hi Steve,

I certainly found that I could 'get away' with playing simple, quiet music with no room treatment. Anything complex, certainly a live concert with crowd noise, would be truly terrible.

I'd like to get the first reflection absorbers working down a little further, but they certainly have a very positive impact on the sound right now, so I'm just grateful for what I've got!

This certainly depends on the placement/distance/room geometry and speaker polar radiation patterns.

You can potentially destroy the fidelity[so far as spatial realism] in some cases[though these cases are not common, involving speakers with an exceptional polar response, and a sufficient room and placement/distance from the walls to begin with] by significantly reducing the amplitude of the front wall and side wall reflections.

-Chris

 

[Steve1000]

Interesting remarks, certainly. Maybe I wasn't so clear in my post, though.

Actually, what I meant was, when I go and listen to live music, it seems much brighter to me than what I usually hear from a good he-fi set-up. For example, the drum cymbals, etc., in a good rock or jazz concert strike me as much much brighter than what I hear on a normal good hi-fi setup.

I am also thinking of my experience with my Infinity Primus 150s, which apparently measure pretty flat. I end up turning up the treble 3-5 db to get what I consider realistic sound. When I do that, the sound seems very, very realistic to me (with the help of a cheap subwoofer) even at pretty high volumes. This is in line with the etymonics site that WmAx linked to said about a typical flat-measuring loudspeaker transducer losing a perceived 4 to 10 db at 10,000 hz due to room interactions. So it was an epiphany of sorts, to me, as to why without turning up the treble a little those speakers sound just a little on the dull side, at least to me. Or, as another example, the flat-mesuring Sony V6 headphones (with no room interactions, obviously) often sound quite bright by subjective standards, yet they are reportedly quite realistic if fed with a flat microphone signal. It's just a number of interesting things seeming to fit together under this concept.

Hi Steve,

I certainly found that I could 'get away' with playing simple, quiet music with no room treatment. Anything complex, certainly a live concert with crowd noise, would be truly terrible.

 

[WmAx]

I
Actually, what I meant was, when I go and listen to live music, it seems much brighter to me than what I usually hear from a good he-fi set-up. For example, the drum cymbals, etc., in a good rock or jazz concert strike me as much much brighter than what I hear on a normal good hi-fi setup.

Where do you find unamplified 'live' rock or jazz? Or do you mean the amplified 'live' performance?

In far-field position[normal seated perspective], live unamplified instruments and voices could hardly be considered in any way 'bright', due to the absorbtion of high frequencies by the air. This is probably why the V6[MDR-7506], for example, sounds 'bright' to many people when playing back most music: music which was recorded at a distance much closer than would happen in a live unamplified performance, relative to an audience member.

-Chris

 

[Steve1000]

Amplified, of course, in parks and churches and museums around DC during lunch time mostly.

The cymbals, brass, voices, etc., at these live performances seem to sound way bright to me. Could be due in part to the live performance volumes. But I always note it sounds very bright, cymbals crashing are a much bigger component of the music than in a normal recording on good speakers, for example, IMHO.

Where do you find unamplified 'live' rock or jazz? Or do you mean the amplified 'live' performance?

In far-field position[normal seated perspective], live unamplified instruments and voices could hardly be considered in any way 'bright', due to the absorbtion of high frequencies by the air. This is probably why the V6[MDR-7506], for example, sounds 'bright' to many people when playing back most music: music which was recorded at a distance much closer than would happen in a live unamplified performance, relative to an audience member.

-Chris

 

[WmAx]

Amplified, of course, in parks and churches and musems around DC during lunch time mostly.

Thank you for the clarification.

-Chris

 

[Buckle-meister]

On Ethan's advice, I use a splitter at the amp to ensure I get the same sound from both speakers (it does work much better).

I have just been to Maplin (electronics store) and they do not have an adaptor which splits a mono into two monos. They don't know how this would be achieved either.

Help!

Regards

 

[Ethan Winer]

Buck,

> but why do I want the same sound from both speakers? <

For low frequencies anyway, you want both speakers playing because almost all music is recorded with bass instruments panned to the center. So testing both speakers together more closely represents what happens when you play real music. Pro acousticians usually also test each speaker separately. Otherwise a problem that exists in only one may not be seen. For example, if one speaker has a blown woofer the combined response would be only 3 dB down, and so could be missed in the sea of peaks and nulls and generally terrible LF response all rooms suffer from.

--Ethan

 

[sploo]

The splitter I use is one of these:

http://cpc.farnell.com/jsp/endecaSearch/partDetail.jsp?SKU=CN06928&N=411

However, you could buy another of the dual female-female connectors and butcher it (solder the ground from each channel together, and do the same with the signal lines too). This would result in a single RCA plug in one end producing the same signal out of both sockets at the other.

EDIT: Oh yea, to answer your other questions:

A double female connector should be fine (I just bought two singles as the two pairs of plugs don't need to be next to one another).

Your description of setup looks fine (unless I've missed something).

 

[Buckle-meister]

For low frequencies anyway, you want both speakers playing because almost all music is recorded with bass instruments panned to the center. So testing both speakers together more closely represents what happens when you play real music. Pro acousticians usually also test each speaker separately.

Thanks Ethan. I'll go ahead and order the splitter Sploo has linked to (thanks Sploo). I gather from the above that I'll need three plots; a response from the Front L/R combined, and another two from the L/R separately (even though I sure hope none of my drivers have blown!)?

Also, by recording each speaker separately, do I get to call myself a Pro. Acoustician?!

Regards

 

[Buckle-meister]

You're always fighting a losing battle by doing EQ first. As the frequency response of the listening position will change depending on where you are in the room. This of course means that a boost or cut that's good for one location could be really bad for another. Oh, and an EQ won't help you with comb filtering (it's not going to remove reflections).

WmAaaaaaaaaax!!! Where are you man? You never mentioned any of this!

Actually, I'm not too bothered that the response might not be perfect at positions other than the listening position; I always sit in 'the hot seat'!

I don't have Word on this computer...

Don't have Word! Where have you been man?! This is the 21st century! Bill will be mighty displeased that one little fishy got away.

What I wanted to know was the peak dB values compared in the different files. I just wanted to verify that the test CD was compensated for C weighting of the SPL meter. It sounds as if it is not, if the peak value is -3dB at 20Hz. If you look at 100Hz or higher frequency signal file, is should be roughly 6db higher in amplitude than the 20 Hz file, if it's C weighted.

I'm pretty confident that the Rives test CD (2) is compensated for the RS SPL meter, seeing as it says on the inlay card that it is, but I checked the 100Hz tone just to confirm: -8.5dB, which is approximately 6dB different from the 20Hz tone...but lower, since both plots were centred about the -90dB horizontal axis.

I know you may not be able to open the attachment again, but I have included a screenshot of the 100Hz tone so that others that can open it may be able to explain why it is lower rather than higher.

Regards

 

[sploo]

...I gather from the above that I'll need three plots; a response from the Front L/R combined, and another two from the L/R separately (even though I sure hope none of my drivers have blown!)?

If your goal is improving sound quality, I wouldn't worry about measuring each speaker seperately.

I'm sure the pros would be able to gain something from doing this, but you can get great improvement from measuring both together (and tweaking room acoustics / EQ settings based on your results).

Also, by recording each speaker separately, do I get to call myself a Pro. Acoustician?!

*LOL* As I get older, I learn that the true meaning of a pro is best defined as someone who has an understanding of what he doesn't know .

 

[Buckle-meister]

If your goal is improving sound quality, I wouldn't worry about measuring each speaker seperately. I'm sure the pros would be able to gain something from doing this, but you can get great improvement from measuring both together (and tweaking room acoustics / EQ settings based on your results).

I'm afraid that I don't see the logic of this. If using room treatments, it is unlikely (I think) that one would put it up anything less than symmetric about the listening position (looking forwards), but if I was to use a PEQ, then surely I'd want to adjust each channel independantly? My (crude) existing plots already show that each speaker is giving a different 'sound', so why wouldn't I require the response from each speaker on its own to determine what to do with the PEQ for each channel?

Regards

 

[WmAx]

WmAaaaaaaaaax!!! Where are you man? You never mentioned any of this!

I did not mention this, as this thread was about measurement and equalization, not room treatment(s), specifically. As the other person stated, bass equalized response will only be optimal at the position at which the correction measurements were made. You have three options if you want better response at all positions:

(1) Use di-pole bass modules.
(2) Use a high number of subwoofers, placed around the room.
(3) Use of many bass trap devices.

-Chris

 

[WmAx]

Ibut if I was to use a PEQ, then surely I'd want to adjust each channel independantly? My (crude) existing plots already show that each speaker is giving a different 'sound', so why wouldn't I require the response from each speaker on its own to determine what to do with the PEQ for each channel?

Regards

Since this sound is a combination of direct and reflected sound, by using an assymetric equalization[on midrange and treble], you will likely cause even more problems[most probably in the imaging/soundstage perceptions, since you will then have assymetric direct sound at the expense of symmetrical total sound response].

-Chris