You will never stop sibilance by adjusting a cartridge, miss-tracking maybe. A cartridge can make sibilance worse by having a rise or peak in that region but it is upper midrange not treble. It is the highest frequencies that suffer with poor alignment and poor alignment does not increase high frequencies but decreases them and this is above 12kHz a full octave at the minimum above the problem. |
True sibilance is caused because our ears are most sensitive at certain frequencies mainly between 3 and 4 kHz. Many recording studio's use a house curve with a BBC or Gundry dip which drops these frequencies a little to try and prevent sibilance in their recordings. Some recordings without this correction are going to sound harsh on any system. You can stop this entirely by EQing this frequency band down just a dB or two.
I keep such a curve handy in my presets to deal with this problem. Sibilance occurs with instruments and voices that have a lot of energy in this region. Female voices and violins are classic. If your system can make it through an early Ricky Lee Jones recording without slicing your throat you have the problem licked. The problem is if you depress that band too much the music will start to sound... I guess muffled would be the right term. So, there is a trade off. |
@lewm , really? Some speaker manufacturers, to make their speakers sound better at low volumes and reduce sibilance tuned their speakers with a built in Gundry dip. Wilson did this with the Watt/Puppy. It is just another application of Fletcher and Munson's research. I'm not entirely sure but I think the BBC first started doing this in broadcasting or in their monitors maybe both.
It seems that many people here are talking about distortion not true sibilance. |
@lewm, yes, Wilson did. I know this for a fact as I ran a test on a pair. They just don't talk about it and very few audiophiles actually measure their speakers.
Loudness compensation you use to find in some preamps was not very versatile as it was only correct at one volume level. But if you tended to listen at one quiet level it worked fine, until the audiophiles trashed it.
I have a preset with a Gundry dip but for some strange reason I have not used it since I got the Sound Labs, haven't needed it. |
There seems to be a lot of confusion as to what sibilance is and means. It is not tracking distortion and it is not due to tonearm misalignment. It is due to the high sensitivity of our ears to frequencies in the 3-4 kHz range and systems or recordings that tend to emphasize these frequencies. It has nothing to do with distortion. You can eliminate any sibilance with a notch filter at 3500 Hz. You can not get rid of distortion this way. Sometimes all you have to do is reduce the volume a little. Rickie Lee Jones will always get sibilant if you push the volume to high on even the finest systems. |
jdjohn, sibilance and miss tracking distortion are two separate problems.
Even the best, perfectly maintained systems can have recording induced sibilance. Miss tracking distortion certainly sounds terrible and can be do to bad styli and cartridges but it is not sibilance even though it makes you squint just as bad. The easiest response to sibilance is just to turn the volume down a little. This will not get rid of miss tracking. |
@lewm , the dip is there. Now, you could argue as to whether or not it was intentional. Would Wilson do something that was not intentional?
Most loudness controls I am familiar with are just on/off switches with only one curve. It would be very complicated and expensive to rig an analog variable control. My TacT is the only unit I know of that has dynamic loudness compensation. It hops from one curved to the next with output level. It happens automatically. I love it. But, no other digital preamp with the power do do this has the programming. I suspect (but do not know for a fact) that Bozevick had a patent or copywrite on this programming. Others will institute it in time.
By the way, there is no one Fletchur-Munson curve. It is a continuum that is outlined by a set of curves that vary with sound pressure levels. Your ears do not hop from one curve to the next.
|
@rauliruegas, putting a 2-3 dB notch filter between 3 and 4 kHz is a very common approach to making certain program sources more listenable.
I suggest you take a very sibilant recording and try it. It is rather cool to hear the sibilance disappear (along with a little detail). How it got it's name as the "BBC" or "Gundry" curve doesn't really matter. I use those terms because that is what the industry seems to want to call it. I did not make it up.
As for Watt/Puppies, measured at one meter with modern computerized equipment they demonstrate a mild dip at those frequencies. I am absolutely sure those measurements were done correctly. The Maxx 2's are a totally different speaker and they were not measured near field individually. I have heard but have no proof that other manufacturers have done this. That graph, by the way has very poor resolution, is crude and you should note the measurement indicates a +- 2.5 dB variance throughout most of the midrange which the writer is calling "impressively flat." For an uncorrected speaker it is impressively flat but the bass is not good. It is down at 150 Hz, up at 80 Hz and falls off steeply below 50 Hz.
In order to get realistic low bass at reasonable levels the bass has to rise as you go down from 100 Hz. I adjust my system to be up 5 dB at 20 Hz.
The dip at 150 Hz is going to rob the bass of detail and impact. Pipe Organs can go down to 8 Hz. 16 Hz is no problem (but an extremely large pipe) The Maxx's bass response is, at least in part due to room nodes.
Very occasionally I will use the notch filter. I dislike loosing detail and you can frequently cut the sibilance by just turning down the volume a bit. |
@rauliruegas , somehow I think we are dealing with a language barrier/misunderstanding problem.
Speaker measurements vary dramatically. The same speaker will measure very differently in different situations. The only measurement that is really valid is nearfield in an anechoic chamber. Very rarely if ever is this the case with speaker reviews.
Speaker designers can vary the presentation of a speaker in numerous ways by intension manipulating a lot of factors and sometimes, to get what they are looking for requires a lot of electronics. The crossovers of the Dahlquist DQ 10 and the Apogee Diva's are great examples. What they had in mind is rarely known by the public. Some designers leave it up to the owner by giving them controls to adjust the relative driver volume or in my situation the contribution of one transformer vs the other transformer. Do any of them purposely boost bass and/or treble? You bet!
Most companies only want to sell loudspeakers and they do so but giving the public what they want and many will go for the speaker that is brightest or has the most bass. Companies will generally do anything to sell speakers. Those of us that buy expensive speakers are generally more critical and the companies that make them are more careful to keep the speaker as flat as possible in the average room. Who has an average room? What is an average room?
Dropping the volume between 3 and 4 kHz is a well known way to diminish sibilance and there are many people who like an easy going loudspeaker that does not make them squint. Boosting bass and treble a little achieves the exact same result which is an easy going loudspeaker that sounds better at the lower volumes most people listen at.
My system is adjusted for playback at 95 dB. At 85 dB without loudness compensation it sounds a bit dull and bass light and people will notice this until I turn the volume up (or kick in the loudness compensation)
Because my front end is digital and managed by an advanced processor I can alter the frequency response +- 30 dB at any frequency from 0 Hz to 20 kHz either individually or as a group. Over the years I have experimented with hundreds if not thousands of curves which is why I can look at a frequency response curve and have a good idea what a system will sound like. I adjust my system to get the tonal balance and imaging I like. I have one preset loaded with my favorite curve but with a broad dip centered on the 3 to 4 Hz region for the occasional recording that gets sibilant at the volume I like which is why I and anyone who listens to my system know darn well that this works.
As for what any speaker manufacturer had in mind? I have no idea. You would have to ask them.
|
@rauliruegas, maybe you missed this but computerized impulse testing of older Watt/Puppies in the near field produced similar curves for each speaker that rose in the treble and bass leaving a depression centered at 4000 Hz. The Watt/Puppy had a reputation for being a very easy to get along with loudspeaker and put Wilson on the map. I suspect this was the reason why. You have not shown me any evidence to the contrary.
There are plenty of recordings with sibilance issues. While it is true that some systems are more resistant to sibilance than others there are some recordings where you just can not get away from it. If there is no circumstance under which you get some sibilance with female voice or violin your system has a problem, probably a curve similar to the Watt/Puppy's! Since I measured the Watt/Puppys even better technology has come along where you measure the speakers with a sine sweep in real time. Lots of fun. You can program the computer to do the sweep but I prefer playing the sweep on a CD leaving the computer totally free to measure. One warning. If you get into this you are going to drive everyone else in the house NUTS. This is best done while you are alone but you might want to do a gummy first that way it will just sound like a new Trent Reznor song:-) |
Raul, I did not start this thread. I was just responding to the OP trying to shed some light on the situation. I am not sibilance crazy. As far as Rickie Lee Jones goes, who I have seen in person twice "it's Like This" was recorded later in her career when her voice had lowered a bit. Listen to any of her first three albums particularly "Pirates"
I have no idea what Wilson intended. All I can tell you is what I measured which matches up perfectly with the speaker's reputation. Obviously, you do not trust my measurements which s your prerogative. But, that is your problem not mine.
I must be a very lucky guy. I hardly ever have to reach for my sibilance preset and I did not have to do near the amount of....stuff to my system to get it to my liking. I see you have this thing for silver wire. Impressive. |
@lewm , Yes, that may have been a bit pretentious. However the response of the speaker being what it was does lead one to wonder. As I said before I doubt Wilson did things by mistake although it might have been an end they were happy with. Just by listening it is going to make a more natural sounding speaker at lower volumes and it may have been by listening that they arrived at this result. Wilson was a young company at the time and certainly did not have the resources they have now.
As for the anechoic chamber, quite right. You can get a feel for a speaker's behavior doing near field measurements. The largest errors are going to be in the bass. I do believe there are now computer programs that with impulse testing can ignore reflections and give a pretty accurate curve without an anechoic chamber. I certainly do not have one.
This was not my own idea. I had heard manufacturers did this for two reasons, to lower sibilance and to make the speaker more natural sounding at lower levels. Both are very true and can be easily demonstrated.
@rauliruegas , Sibilance and distortion are two very different problems with different solutions. Distortion might sort of sound like sibilance but it is not. |
