Room Response and Preference
- Thread starter Ambio
- Start date
Those are actually spectrograms of your room's decay time.
Without seeing any other measurements, the second would be the one to strive for. The decay is steep (falls off quickly) and there is less fluctuation in volume level (fewer/smaller peaks & dips).
Yes they are. I couldnt find the impulse and the response chart to go along with spectrogram. But the FR can be analyzed from the spectrogram.
You are right the second chart is what one should strive for and very difficult to get it in our listening room.
However, for some heavy bass tracks which are usually used as audiophiles' reference CD, such as Jennifer Hunter's The Hunter -Low Down Deep, the incorrect response of the first chart was preferred. Including myself. That was the dilemma I was facing whether to do the correct thing or setup how I would prefer it to sound.
What would you do?
[ I just don't get how the images are displayed in this forum. I could see the image after signing off but not when I am signed in. Anyone?]
Last edited:
Always go by what you hear - if you know what you want to do.
FWIW measurements can be inherently flawed because of the complexities of mic placement and the basic drawbacks in the technology. Floor bounce, driver layouts, driver type etc all can throw off measurements. Unless you really know what you are doing these can be sometimes be plain wrong.
My 0.02!
Preference cannot be incorrect. There is no right or wrong in personal preference. As long as you understand that your preference is to hear more of the room and less of what was on the CD (the peaks & dips you prefer in the first chart are not part of the original recording). As long as you like it, don't let anyone convince you otherwise.
BTW, the CD is by Jennifer Warnes (not Jennifer Hunter) and the song is Way Down Deep (not Low Down Deep), in case anyone reading this wants to get the CD (one of my all time favourites).
My personal preference is to hear less of what the room is adding and more of what was in the recording itself. So the long decay times and the peaks & dips in the first chart would not be my cup o'tea. I've always preferred the type of response you see in the second chart, with less decay (overhang) and less unevenness in the response.
without knowing SPL levels - It may also depend upon how much bass was there to begin with. If the speakers aren't putting out much bass, then even peaks may be perceived as good bass.
Many times we don't know what is good vs bad till we know what good is. We don't understand what good sound is till we hear a proper setup. So, enjoy the sound as you like it.
Many times we don't know what is good vs bad till we know what good is. We don't understand what good sound is till we hear a proper setup. So, enjoy the sound as you like it.
Always go by what you hear - if you know what you want to do.
FWIW measurements can be inherently flawed because of the complexities of mic placement and the basic drawbacks in the technology. Floor bounce, driver layouts, driver type etc all can throw off measurements. Unless you really know what you are doing these can be sometimes be plain wrong.
My 0.02!
The measurements were accurate and repeatable. Everything calibrated properly.
LOL...it has been more than 5 years and I am still stuck with Low down deep. One of the local self declared audiophiles guru used this track to judge my system. He kept emphasising the low deep bass in the track and my system couldn't reproduce them. So somehow the "low" got stuck to name.
Out of curiosity, I ran a spectrogram test on the track. All the bass that is worshipped by audiophiles was not in the lows but rather around 90hz to 120 hz. Coincidently, these frequencies are also one the main room boom culprits.
If you look at the first chart you will see an over emphasis boast in the 90 to 120hz region but with a nice decay. There is some ultra low fq boast too. Now suddenly my room received the stamp of approval when it was setup according to image 1.
I went on to do more analysis and decided to record the sound from the speakers and compare with the original digital spectrogram. Guess what? When the setup is according to second image the spectrogram is much closer to the original file but the next one matching the first image carries a 10 over dB in the 90s hz region.
I decided to stick to the flattest response. At least, I don't have o complain that my system only play certain genre well. we have heard this remark quite often in many high end setups
Last edited:
all measurements taken from 70 to 90dB in the steps of 3 dB. The difference were not too great. You can see from the image the reference volume of each. The lowest and the highest.
Yup, the opening drum hits of that song are in the 60Hz region. That local audiophile guru that thought there was low bass in that track doesn't know what he's talking about.
LOL...it has been more than 5 years and I am still stuck with Low down deep. One of the local self declared audiophiles guru used this track to judge my system. He kept emphasising the low deep bass in the track and my system couldn't reproduce them. So somehow the "low" got stuck to name.
Out of curiosity, I ran a spectrogram test on the track. All the bass that is worshipped by audiophiles was not in the lows but rather around 90hz to 120 hz. Coincidently, these frequencies are also one the main room boom culprits.
If you look at the first chart you will see an over emphasis boast in the 90 to 120hz region but with a nice decay. There is some ultra low fq boast too. Now suddenly my room received the stamp of approval when it was setup according to image 1.
Way down deep is the best song to judge the texture of the bass. I started hating foobar after comparing that with song played from cPlay / JRiver. 80-120hz is the lower midbass region. Plays a crucial role in the integration of your sub with the mains. If the Fs of your mains is around 90Hz and you have crossed your sub at 80Hz, then you will have a weaker/incoherrent 80 - 120Hz region. If the Fs is 60Hz, then you get much better crossover at 80Hz. Male vocals can go as low as 100Hz. So, cleaner reproduction at this range helps immensely to get thick male vocals.
For me, the best bass lies around 35-80Hz. My Neumann KH120 speakers have Fs of 60Hz. Pretty flat from there. Still i feel it is very lean to play "Adiyae" from Kadal. Brilliant song to test that region. The drum slaps will be weak and the entire emotion of the song is lost a bit if 35-80Hz is not good in that song. Pretty difficult for any bookshelf speakers to do that. The same song springs to life when i play it in my car where i had HAT unity midbass (Fs 80Hz) and Sinus Live sub (Fs 35Hz and can go all the way to 800Hz) crossed over at 80Hz @12db slope.
Last edited:
Thad E Ginathom
Well-Known Member
Those of us with small speakers that just don't do low probably welcome a certain amount of proximity enhancement.
In my listening space, otherwise known as an alcove, however, it can get a bit boomy. I haven't measured this properly, but I've found from test tones that the main problem is around 160Hz. Quite high, but it seems to affect deeper sounds.
In my listening space, otherwise known as an alcove, however, it can get a bit boomy. I haven't measured this properly, but I've found from test tones that the main problem is around 160Hz. Quite high, but it seems to affect deeper sounds.
Thanks Ambio and sdurani for the pointers to this wonderful album.
Her "Famous Blue Raincoat" is a really good album as well.
Adiyae sounds pretty good in my system with full bodied sound. And AR Rahman's Chinese album - Heaven and Earth (sic/I don't really remember names) - the drums sound really dramatic and powerful. I am using Rythmik Audio F12SE subwoofer crossed at around 60 - I think. Speakers - Harbeth SHL5. There used to be a JAVA based program for accurate calculation of crossover point and phase adjustment for a given placement but the link is no longer available. Somehow, 60hz matches going by ears too.
Not sure what you mean by Fs. Care to explain because web search say it is frequency resonance and I am not sure where to find that info for my speakers.
The only way to know if your sub integrates with the mains is by measuring sound at your sweet spot. Guess work is not reliable because it changes with your test material. Correct bass for The Hunter's track may not be correct for another track. What I am saying is if you want to measure a foot precisely then you have to use a ruler.
I used to be a car SQ enthusiast but IMHO a car sound will never match a high end home stereo. Write about it in another thread.
just finished listening to Kadal. I think moongil thootam contains better deep bass than Adiyae. Listen around 1m 55s for the low bass. 127 Hours is nice too. You need good bass handling system. I wouldn't recommend Adiyae for bass test in a system.
typo- in my previous post I meant to say I wrote about in another thread.
typo- in my previous post I meant to say I wrote about in another thread.
Last edited:
You can say that a speaker plays well 'after' its resonant frequency. A driver with 55-5000Hz with Fs of 80Hz will play well after 80Hz. It is better to cross them at 80 rather than at 55.
Adiyae is for mid bass. In fact, i hate deep bass in stereo which causes rumble.
I think you are referring to the point where the loudness starts to drop. Drivers specs looks like this - measured in free space without enclosure. Note the Fs point.
http://www.madisound.com/store/manuals/EW638.pdf
http://www.madisound.com/store/manuals/UW1058.pdf
Loudspeakers specs looks like this.
Attachment 1. See the "40Hz25kHz (6dB points)"
From the independent measurements.
see attachment 2.
It is around 45Hz, where there is a 6dB loss.
It is usually suggested to set your crossover point around 45Hz. The Fs for the driver should be around 30Hz and speaker designer would make sure that the crossover network cut off any frequencies reaching that stage.
In practice it doesn't work like how your are suggesting due to room mode. See my first post's charts.
In most cases, it happens due to pushing the drivers too hard beyond their limits. But it shouldn't be in a good speakers or subwoofers. Unfortunately, you have to pay good money to get the last few octaves in pristine condition.
Attachments
f/s is the critical frequency below which the SPL starts to taper off. It also depends on other parameters, enclosure volume and so on. ideally an octave above the f/s is recommended for drivers. With subs higher slopes are generally better.
My understanding is Fs is the point where the drivers moves in sync with its natural frequency. As from the above specs even a 10in driver starts to drop off the flat frequencies at around 100Hz. However, the enclosure of the speakers make it possible for designers to maintain flat frequencies few decades below that.
My point is Fs is not relevant because crossover network would avoid the point. The important point to start off is look at the loudspeakers specs(NOT DRIVERS SPECS) and use that as the crossover point for the sub setting. In a perfect room it should be ideally lower than the lowest point of FR of the Loudspeaker (NOT THE DRIVERS). Why? because say at 40Hz the loudness is 77dB down from the 80dB. So if you cross your subwoofer slightly lower then you should have to get the right volume to lift the volume at 40Hz to 80dB. Remember this is the theory but in a real room all these can be thrown out and you have to rely on actual measurements.
In reality, even with a 24dB crossover slope the effect of subwoofer could be felt one or two octave above. The best way to know is to measure using a room frequencies analyzer. Otherwise, we would forever be chasing perfection without knowing the root cause.
Fs is the free air resonant frequency of a driver. Meaning, if the driver has no air resistance, this is the lowest frequency it can produce. The air in speaker cabinet acts like a spring and does not allow the driver to move freely, so it can't achieve Fs. This happens more so in a sealed box than vented. Driver manufacturers also specify Vas (Accoustic Suspension volume). This is the volume, beyond which air spring effect is non existent or very negligible. If one makes the speaker box bigger, less spring effect and driver tends to go towards Fs. When box size is equivalent to Vas or more, there won't be any spring effect, driver will be able to produce output at Fs. Another big advantage is - No air resistance, means the driver moves freely and very quickly with lowest power possible, compared to a box. If the box size is multiples of Vas and there is no backwave, this type of setup is called "infinite baffle". Another way to achieve free air movement is open baffle. No box, no resistance.
Bottom line is - No driver will be able to produce lower frequencies than Fs in an an-echoic chamber. Although, in actual rooms, room gain comes into picture and some speakers will give lower output.
Fs is usually specified by driver manufacturers. Speaker manufacturer's will not specify it because it is of no consequence. Rather they specify F3, which is the frequency where speaker starts to roll off. This is the -3db point on the frequency curve. Below this frequency, the speaker will have output, but not as loud as its rated. If one was setting cross-over, it should be set at least equal to this frequency. Ideally it should be set higher than this. Because F3 will already have -3db output and in audio terms, it means half loud. If my speakers have F3 at 60 hz, I would set the crossover at 70 or 80.
Bottom line is - No driver will be able to produce lower frequencies than Fs in an an-echoic chamber. Although, in actual rooms, room gain comes into picture and some speakers will give lower output.
Fs is usually specified by driver manufacturers. Speaker manufacturer's will not specify it because it is of no consequence. Rather they specify F3, which is the frequency where speaker starts to roll off. This is the -3db point on the frequency curve. Below this frequency, the speaker will have output, but not as loud as its rated. If one was setting cross-over, it should be set at least equal to this frequency. Ideally it should be set higher than this. Because F3 will already have -3db output and in audio terms, it means half loud. If my speakers have F3 at 60 hz, I would set the crossover at 70 or 80.
I am no speaker designer and I am have no idea of the the technical terms you are referring to.
Just curious, are you confusing "free air resonant" to "air free resonant"? The simplest way to determine Fs is to tap the driver with your finger and frequency where it resonates. That will be the Fs of the driver. This is is done in free air. NOT in air free vacuum.
Now taking the same woofer/driver in a cabinet, the resonance goes higher as it is no longer able to vibrate freely in free air but need to vibrate with the resistance of the air in the enclosure. Here the resonant frequency will be higher than what it was in free air. That value is called Fc.
I was just trying to reiterate that neither Fs or Fc matters. All the guide of the correct subwoofer crossover point and phase is just a guideline. Your room is a huge loudspeaker. Or imagine yourself inside your loudspeakers and listening to the sound. no one can predict how the soundwaves behave at your listening space.
If you don't do a room measurement for the subwoofer, you will compromise the sound. Try run a frequency sweep and you will be surprised to see that some frequencies can be completely in audible at your normal listening level and some may be too loud.
IMHO.
Purchase the Audiolab 6000A Integrated Amplifier at a special offer price.
