Audio Asylum Thread Printer
Get a view of an entire thread on one page
|For Sale Ads|
This is an audiophile question rather than a strictly musical one but, since it applies mostly to classical music, I thought I'd post it here.
I am experiencing excessive harshness, brightness and what I'd describe as distortion in higher passages for violin sections in listening to my system. The harshness becomes present around the F-A (2nd above middle C, so freq of ~750-1000Hz). It is an issue for string sections only. Individual violins sound great up there, as do high winds, so it isn't a problem with my treble.
For reference, here is my system: CD player -> Audio GD Digital Interface re-clocker --> Metrum Octave DAC -> Passive preamp (Lightspeed Attenuator) -> W4S ST-500 amp (Class D) --> Magnestand (Gunn'd) MMGs.
Here are my leading theories:
1) Digital recording of high string sections is still a problem.
2) The Class D amp adds noise at higher frequencies. (I'm channelling morricab here.)
3) Room interactions. (My room & set-up is normal and symmetric although there are two intrusions into the room.)
4) Jitter. (The re-clocker should help a lot with this issue but the Audio GD is not a state-of-the-art unit like an Off-Ramp or such.)
What are your theories about what is causing this problem? Any hints on diagnosing or solving the problem?
Have you ever tried vinyl?
that particular facet because it is more pronounced. In the name of high-end, many "modern" systems boost the upper end to allow listeners to "hear into" the music. Like shining a bright light onto a particular part of a film screen, it distorts everything.
It is hard to pinpoint the specific component(s) of your system since all are different in their interactions. One fact: you cannot impact one area w/out impacting the whole.
Thoughtful suggestions. I will be analytic and try to confirm or disconfirm the various ideas presented, from tinnitus to excessive digital processing to recording technique.
I don't really know where to begin other than by saying that it seems you have a system guaranteed to create the kinds of problems that you are describing.
Let me start by saying that live strings, no matter where in the audience that I sit (for that matter even sitting on stage) never sound harsh to me...that is live string sounds are never overlaid with the kind of digital distortions that you are describing.
Class D amplification sounds extremely wrong to my ears and seems to distort that sound that it reproduces. I don't know why, but Class D amplification alone can create the kind of problems you are describing. Compare the sound of Class D to Class A amplification. I think going to Class A will make a world of difference.
Furthermore, lets not overlook the digital source. These add digital distortions that overlay the sound. I find SACD to be the least distorted and closest to analog of the digital sources. When done right, it can fool me quite easily into thinking that I am hearing live music. I use SACD for my most critical listening, especially my most critical listening to classical music.
Moreover, when I have auditioned Magnepan speakers I found that they had a narrow band peak or break-up that added shriek and glassiness around 1-2 kHz, the problem area that you are describing with regard to orchestral strings.
What you have is a recipe for disaster...Class D amplification, a lower resolution digital source and speakers that act up at problematic frequencies.
First thing I would do is to try Class A amplification, then add an SACD source. I think these two steps alone might cure the problem.
The final step would be to address the speakers. Truth be told, I have not found many that do not have the same kinds of problems.
The Dynaudio line might be one to investigate as their abiding characteristic seems to be an exacting neutrality that gets out of the way of orchestral string sounds.
Thanks for the thoughtful post. I will point out that the treble in my system is actually one of the best things about it as long as violin sections aren't involved. Last night, I listened to one of the offending recordings - an EMI disc with the London Philharmonic - on my second system which, while not as resolving as my main rig, is a) class A amped, b) box speakers (B&W); c) CD source but little processing. And the sound had the same level of irritation as on my Magnepan rig, except less detailed.
I'm taking your comment about Class A vs D seriously, but an initial comparison leads me to be skeptical on that point.
B&W = metal tweeter + aramid fibers (kevlar) that stiffen when exposed to ultraviolet light (sunlight) resulting in a rasping, papery, gritty sound at the top of the drivers range. Same kinds of problems as the Magnepans.
I no longer use speakers with metal dome tweeters because the (metal) tweeters have a tendency to ring and shriek in the lower part of their range resulting in steely sounding high strings.
Ideally the driver should not overlay the sound of the recording with spurious sounds (resonances) of its own.
If you tap or scrape a driver, does it make sound? If it does, then those resonances will overlay the sound being reproduced.
For that reason, I like drivers that self-damp, that resist ringing...such as fabric dome tweeters and paper or plastic coned woofers (even some the newer space age foam sandwich materials like rohacell).
Electrostatic drive units also seem to be good at self-damping.
Going to class A amplification usually (but not always) means removing the kind of gritty and edgy sounding distortions that Class D, Class B and (to a much lesser extent) Class A/B type amplifiers produce.
Morevoer, gritty and edgy sounding (digital) distortions are a more or less permanent fixture of the CD standard of digital encoding. The better (CD) players can supress these but to my ears only higher resolution formats are more or less free from these issues (SACD being the most natural sounding to me).
In my system(s) I use speakers with fabric dome tweeters (scanspeak, dynaudio), and plastic coned woofers (dynaudio, rogers). The speakers do not sound edgy or harsh and do not shriek or ring. My main amplifier is Class-A. It sounds silky, refined and detailed and free of the hard, edgy distortions of many of the amplifiers that I have used in the past. My source is usually SACD but the Marantz SACD/CD player does a very good job with CD too. I am quite pleased.
the B&W speakers I was listening to have a polyester tweeter, not metal.
Also, you're now pointing out that your Marantz does pretty well on CDs.
I'm sure your system sounds wonderful but your explanation of my problems (class D, Magnepan, CD) is not fitting the fact pattern.
I ask as that is my problem with massed strings in orchestras (violins and violas) is I don't like how string instruments sound arco style (played with a bow). Back when I liked classical music I preferred music that stressed the winds, brass and percussion sections of the orchestra.
I was never a big fan of orchestral music live, preferring acoustic jazz especially big band and dixieland and blues. I also have a problem with rock music live, even though rock is my favorite style of music as it is often highly distorted live. Oops getting off subject.
In short if you don't like massed strings live in a concert hall, you may just not like massed strings.
I like the violin solos in "Sad Lisa" and "Into White" from Cat Stevens' "Tea For The Tillerman" both played arco style.
On the other hand, if you do like how massed strings sound live it could be something in your system. I second the suggestion of playing some massed strings in an analog format to compare. Or if you have access to any high resolution digital (SACD, DVD-Audio or 24 bit downloads) see if you like massed strings on them as I find them smoother in the highs.
Everyone's free to do as they please, but I just hate when try to impersonate women.
What a rather rude question.
I know there are not many of us in audiophile land. Teresa is my real name and I am a paid writer with checks made out in my name. Here is one of my articles at PF.
Could attitudes like yours be one of the reasons there are not more female audiophiles?
I had to sell most of my SACDs during the past year to pay rent and necessities. However first I recorded them to my computer at 24/96 using Audacity. They don't sound quite as good as pure DSD from my SACD player however they still sound excellent. I figured it was better to have 24/96 versions versus being homeless and living on the street.
I am waiting on my disability to kick in, it takes forever. In the meantime my income is $50 every two months, I do get food stamps. Rent is all that I worry about.
The Paavo Jarvi recordings were among the last ones I transferred to 24/96 and even though I have deleted them from my hard-drive with all the other classical works, I still have them on backup DVD-R's if I ever start to liking classical music again. So my remaining Telarc's are jazz SACDs to 24/96 AIFF computer music files.
Currently I have 129 albums on my hard-drive and they breakdown as follows:
I still have 5 SACDs - all classical and all for sale in the Audio Asylum Trader.
4 24/96 DVDs - Jazz and world music and not for sale.
3 Movie DVDs - all for sale in the Audio Asylum Trader.
In the past I have never went back to music I no longer liked, I have tried though. As you may know I quit liking most symphonies over 5 years ago, with few exceptions such as half of the Mahlers, and a handful of others. Next I couldn't stand to listen to any Mahler Symphony and that was over a year ago. I found myself deleting more and more classical music, I seriously don't think I'll ever like classical music again, however if I do, you can tell me "I told you so".
Here is something else I should say, I have and love every single Rock and Jazz 24 bit music file on my computer that I either downloaded or copied from SACD or DVD-Audio. And before my complete disgust with classical music, I deleted or sold over 90% of what I bought as either not good enough sound, too boring or too atonal. With that high of a rejection rate at first I thought maybe I was too picky. However if one really loves the music such as in my case with rock, jazz and world music one can overlook less than perfect sound.
I now think my 38 years of exploring classical music was a complete waste of time. Today I listened to Duke Ellington's "Blues In Orbit" (SACD to 24/96 music file), "The Body Acoustic" (24/96 Chesky download), and Emerson, Lake & Palmer's "Brain Salad Surgery" (DVD-A to 24/96 music file). I am enjoying music more than I have ever before. Dumping classical (for me) has been liberating!
... to hear about your health problems and financial troubles. Best of luck to you as you continue to push forward.
"He was one of those men who live in poverty so that their lines of questioning may continue." - John Steinbeck
Well, for that, Teresa, I certainly feel for you because I know about the rent, lack of funds thing. I still buy an occasional disc, but am selling a bunch of stuff with hopes finances turn soon.
Oh, brother. Now you don't like classical music, but in 6 months, after you've sold everything, you'll love it again. No one takes you seriously Teresa.
(well, not in the sense that you were thinking :-) )
For years, Ms. Teresa has been railing against the RBCD format and how it makes violins sound harsh. NOW, we learn that she doesn't like the sound of real violins because they sound harsh! Not only that, she talks about "formerly" liking classical music. When did she stop liking such music? She continues to talk about it on other fora as if she is an expert (BTW, it's arco not argo - and you were told that several weeks ago in another thread).
So let me get this straight - live orchestral violins sound harsh; recordings of orchestral violins on RBCD sound harsh; ergo, the problem is with RBCD. Seems quite logical to me. Bwa ha ha ha ha ha!
Spell check always misses this everything as it's not a misspelling but an incorrect word use. Since I have been calling arco argo for many decades it may take many years for it to sink in that arco is the correct word.
BTW while is not as offensive with musical instruments I like the sound of, a combination of violins played with a bow and RBCD is painful for me to endure. With other forms of music such as big band jazz or rock RBCD is merely just not enjoyable.
Real violins heard live are not really harsh just not enjoyable, as long as there is brass, winds and percussion to balance them out they aren't too bad live. In other words I never liked string quartets, string orchestras and violin concertos live. However it is RBCD and only RBCD where violins sound harsh.
It was about two months ago when I woke-up and discovered I didn't like classical music anymore. The weirdest thing that every happened to me, since classical music was my favorite style for 38 years despite the use of string instruments played with a bow.
I'm not allowed to post on Hi-Rez Highway, DVD-Audiobon, Digital Drive or Computer Audio Asylum. I mostly post in the General Asylum.
"NOW, we learn that she doesn't like the sound of real violins because they sound harsh!"
Where did she say that?
I did in answering his post but not in relation to violins played live, I just said I don't like how massed strings sound live in a concert hall.
In response to krisjan's post I said "Real violins heard live are not really harsh just not enjoyable, as long as there is brass, winds and percussion to balance them out they aren't too bad live. In other words I never liked string quartets, string orchestras and violin concertos live. However it is RBCD and only RBCD where violins sound harsh."
In her above post
I very much like the sound of live strings and solo strings sound absolutely great in my system.
As a minimalist, I have no expertise with the type of system you have configured, but it occurs to me that the complexity of your front end my make it vulnerable to the kind of problem you describe. I suggest you borrow a simple, integrated CD player of decent quality and see how things sound.
An equalizer might alleviate the symptoms, but it won't cure the underlying problem, and will introduce anomalies of its own.
I am also suspicious of class D amps, but doubt that that is the source of this particular problem.
Screechy violins are common in poorly recorded orchestral performances. The problem occurs because the radiation pattern of a violin is directional -- different frequency ranges radiate in different directions. So if violin sections are miked close and overhead, they sound screechy.
Unfortunately, there's no good solution: you have either to stick to well made recordings or equalize, and after the recording has been mixed if you EQ the violins you'll damage everything else. Some recordings are so offensive that I find I have to do that just to make them listenable.
I have a Charles Mackerras Haydn recording on Telarc which is single miked which I will use as a test case. A lack of screechiness from a mike far away wouldn't resolve the issue, but it would support your argument.
I've never heard that in simply miked recordings. Even most simply-miked recordings are too hot, since they use forward-facing cardioids, and put them at the front of the stage, rather than in the audience. But they don't seem to suffer from the screech effect.
BTW, I have to give credit where credit is due -- this isn't my own theory, it's something I read about elsewhere. Not sure if I first saw it in Toole, or first read about it somewhere. But if I recall correctly there's an AES paper on it somewhere. The problem *can* be avoided with even a closely-miked recording, by equalizing the strings, which is why not all of them are screechy. It's possible that some mixes are better than others, since an LP mix forex might be made with EQ on the violins but the CD mix without, it's going to depend on the knowledge of the mixer and the desire of the producers.
Another thing that bothers me about some recordings is that the strings are often pushed up in level so that they outshout the rest of the orchestra. Composers knew that the strings weren't the loudest instruments and brought in the more raucous ones when they wanted a climax. IMO, pushing up the strings distorts the music, but it seems that some conductors prefer the sound that way, just as some prefer a podium perspective.
My Metric Halo DAC has parametric EQ.
I find that a dip of 2 dB centered at 3.5 to 4 KHz tames screechy violins but does not rob the music of air like the high frequency shelf of a typical treble control.
In practice, I use different values for the amount of dip (dB) and the center frequency for different recordings. After a while, you gain a knack for choosing the best values.
That sounds about right. The precise frequency presumably depends on where they put the mic:
Unfortunately, you can't bring the violins down without affecting the other instruments. But you're probably getting better results than I am since I do it with a graphic rather than a parametric EQ so I don't have precise control over the center frequency. I'll have to try it your way.
That would be where I would start.
- different than the ones I would use for jazz, for example. The biggest mistake one can make is to use overly bright sounding speakers for symphonic music. Large venue acoustic music sounds best to me with speakers or headphones that are voiced slightly on the mellow side (slightly rolled off highs and deep blooming bass). For jazz recordings I prefer slightly zippier highs and tighter bass. For rock recordings I like punchy midbass and an emphasis on leading edge transients across the spectrum. I always try to own at least two sets of speakers and two or three sets of headphones that can be switched in or out of my system, depending on the recording.
"He was one of those men who live in poverty so that their lines of questioning may continue." - John Steinbeck
Visit www.regonaudio.com to read several articles about the problems and pitfalls of recording classical music. They are very enlightening and may help you determine what's going on with your system and what to do about it. Hint: The problem probably is not entirely with your system and room.
Bob - Thanks for the link. I found Greene's site to be very interesting.
I've been able to eliminate most of the high frequency nastiness in my system by reducing high frequency interference from RFI/EMI. The cheapest and easiest way to make a dent in this noise is to place crystals (quartz, amethyst, topaz, etc.) in strategic places. Another cheap way of reducing EMI is to aim an ionic generator at the chassis of your power amp. This turns the chassis into a farraday cage which repels stray electrical fields. I use the one linked to below and it works surprisingly well. The evidence of reduced EMI is a larger soundstage, better definition of instruments and a more "relaxed" musical presentation.
Could be the room. Could be cancellation/edge losses from close mono-miking.
I have one Hyperion disc of Handel's Messiah where a soprano breaks up / clips. Everything else is fine.
I've got ~6 watts driving the dome tweeter which is padded to 91db/w and is 3rd-order from 3.5kHz. 6 watts is about 7dbw so it shouldn't be an issue.
And, it may NOT be the fundamental that's breaking up.
Note that a post in response is preferred.
The Skyptical Mensurer and Audio Scrounger
And gladly would he learn and gladly teach - Chaucer. ;-)!
'Still not saluting.'
"Here are my leading theories:
"1) Digital recording of high string sections is still a problem."
This is true, but it would not be audible with *all* recordings...... Especially if the problem sounds similar with all recordings.
"2) The Class D amp adds noise at higher frequencies. (I'm channelling morricab here.)"
The problem isn't adding noise, the problem is the noise modulating with the music. This is how RFI corrupts the sound. You can't hear RFI and switching noise in itself, but the modulation with the music (in the analog domain) is where the problem occurs.
That said, I've yet to hear a Class D amp that I thought sounded right......... But the effect would not necessarily cause strings to sound "bright" at certain frequencies....... It's more a fatiguing effect after listening to music (of any kind) long-term.
"3) Room interactions. (My room & set-up is normal and symmetric although there are two intrusions into the room.)"
Do the clap test..... If you hear "slap echo", there could be modes that introduce audible peaks at the frequencies where the problem is occurring.
Also try some cheap, temporary room treatments (taping a blanket to a wall), and check if the problem is mitigated in any way.
"4) Jitter. (The re-clocker should help a lot with this issue but the Audio GD is not a state-of-the-art unit like an Off-Ramp or such.)"
I read up on the re-clocker. It has a straight pass through mode and an upsampling mode. If you're running the latter mode, that could be the problem, try running in straght pass through mode. Like Class D amps, I've never heard an upsampler that I thought sounded right.
I'm also not a fan of re-clocking in any shape, matter, or form..... I'd also try running the system without it. (I think the best sounding D/A preserves the original jitter signature of the recording, only because no recording is A/D'ed "jitter free".)
A partially blown driver or drivers? If it's not fully cooked, thus being a dead short, it'll exhibit selectively distorting behavior, just as you've described.
many of them I have objective reasons for thinking are pretty good. When I began noticing the problem, I attributed it to the engineering of individual discs, but it manifests itself across many labels, time periods, ADD vs DDD, etc.
By any chance could you have tinnitus?
I have mild tinnitus, but it doesn't flare up with most treble sounds played on my stereo nor do I experience harshness when listening to a live orchestra playing high passages.
I have mild tinnitus, too, and if it flares up, it's ALWAYS triggered by recordings of high strings, even a string quartet. Other high-pitched sounds don't trigger it, nor does an orchestra or a quartet in live performance.
Post a Followup:
Post a Message!
This post is made possible by the generous support of people like you and our sponsors: