Member
Joined 2009
Paid Member
First off, we can talk about testing, where the utmost in controls need to be taken. Then, yes, nearfield in a set environment is one way to get there. Reasonably good room isn't going to get us to ~ -90 dBFS resolution.
But my point was translational: the average person needs to worry far far less about their DAC (unless pathological) than they do their speaker placement; and managing resonances and reflections. Or a need to southern Utah in order to minimize environmental noise. It's one thing to say that various dithering protocols in 16/44 or 24/96 are audible under some test condition, which obviously must be contrived. It's another thing entirely to say that it translates meaningfully to the end user, i.e., effect size matters.
In other words, all this talk of ultra-low distortion/noise/0.5lsb dithering/etc is a distraction to much, much, much bigger problems in a reproduction chain. It's not that I want to censor the discussion, but it seems very much a forest/trees situation.
Similarly, not ignoring a complaint by testees, just accepting it as a limitation of testing. In one light we can interpret Bob Ludwig's comment as a thinly veiled "it can't be tested" when you start adding up the number of confounding factors at play. Let me offer this as an extreme example: I just got back from a talk on a wide scope, (and N!) multi-center collaboration to tease out factors in colorectal cancers. I'm sure the cost of the overall research effort stretches well into the billions USD. Their best predictive ROCs based on genetic and environmental factors were still mediocre.
Humans are hard to work with.
Attempts at clarification of "why" might end up turning into a philosophical argument.
Maybe we should just measure to see if it is? If we find out it is, then it would make sense to try to figure out why. I don't understand why resort to philosophy, imagination, arm-waving, etc., to produce an answer of dubious quality when one can just measure it?
Regarding anxiety, some people can be in a car accident, physically uninjured, but end up quite stressed out about what could have happened, and for a very long time. For other people, it's like its over and I'm okay, so that's it, forget it. People vary a lot in what produces stress and to what degree that stress affects their functioning. There is no logic to it, the reaction occurs in System 1. System 2 with it's logic and reasoning has little or no say in the matter.
Last edited:
Hey, we agree!
Regarding the testing, speaking for myself, I only know what I can hear on my system, and very much in the near-field. If someone makes assumptions that produce a less accurate listening environment than what I have now, I would expect to perform worse. And maybe vice versa. Anyway, for me, the NS-10 speakers matter, as although they have lousy frequency response, they have very good time domain response. If some experimenter assumes that a test setup with good frequency response is fine, and time domain response shouldn't be a big deal, that could adversely affect my performance. I don't know, but I would like to listen to what they come up with and see if I can hear well on it. If not, testing me on that system might not be productive from my standpoint. And it might not be productive from the experimenter's standpoint either, although they might not know it. At least, until somebody comes along and does a better job with the experimental setup.
Thanks John, yes an open mind is required and does indeed lead to progress.
Many/most around here are concerned with vanishingly low levels of distortion/noise etc.
My approach is in a sense quite the opposite in that my research is toward retrofit improving existing equipment and 'making audio great again'.
In my growing up days we had a hallway valve radio with DC magnet that propagated smooth flowing music and voice throughout the house.
My solution has essentially recreated this smoothness but also with pleasing clarity and intelligibility and no nasties.
In other words any piece of audio gear can be transformed from harsh, irritating and shouty, to clean, clear, musical and pleasing and in the long term....once experienced you can't go back.
From bathroom radio to $100k home systems through to multi million PA systems (Fleetwood Mac, AC/DC and others) the same clarity and ease is induced, leaving no signature of the system but instead the music only propagates.
I believe that subtly bad sound as experienced by all of us daily is actually neurologically/physiologically harmful, same for RF transmissions.
I also find that this harm can be ameliorated, and indeed turned to beneficial.
With the filtering I have installed at the power entry to my house, my home environment feels like I am out in the country, relaxed, calm and peaceful and very unlike other houses/shops etc that I visit.
This knowledge of the potential harm of magnetic disturbances due to electricity/RF is old knowledge, I reckon there is a way to economically cure this problem.
Guys I deal with are talking about installing low powered transmitters on every street pole for purpose of building a parallel data network for purpose of providing data traffic capability to support IOT and driverless cars, so in future we will be bathing in RF such as we have never experienced. 😱 .
Re 'digitalis', I have solved 'digitalis' since a very long time ago.
Recent comment from sound engineer was "you made my digital desk based PA sound analog, cool man, cool."
Back when the AD712 was novel I built a little line level unity gain switchable polarity inverter stage.
It all worked perfectly fine, however it did introduce an odd 'metallic' character to the sound.
This is an experiment that I will revisit, along with a bunch of other AD devices supplied by Scott, thank you Scott.
Dan.
> Yet so many stories abound about professional classical musicians
> not being bothered about having a nice stereo because they can't
> really enjoy the music as they are listening to the performance
> and its such a poor facsimile of being on stage in the midst of it all
I would say that the truth is closer to that professional musicians don't
bother with extravagant equipment because they hear the ' musicality '
rather than the ' fidelity ' of the performance . Musicality being timing,
harmony, and nuance in a context far transcending minute bits
( or lack of ) fidelity .
> not being bothered about having a nice stereo because they can't
> really enjoy the music as they are listening to the performance
> and its such a poor facsimile of being on stage in the midst of it all
I would say that the truth is closer to that professional musicians don't
bother with extravagant equipment because they hear the ' musicality '
rather than the ' fidelity ' of the performance . Musicality being timing,
harmony, and nuance in a context far transcending minute bits
( or lack of ) fidelity .
ABX test boxes will by definition add/subtract/alter the testing signal/music.
This will cause sufficient masking to obscure subtle details/differences and render such testing less definitive.
Another factor I find on my system is that the playback system subtley changes according to the source material and stabilises subsequent to the first high level peak signal amplitude.
In other words, the programme induces a 'set' into the system and this renders ABAB testing invalid and needs to be AABBAA etc.
Those of you who build a new amp or line stage should be aware that these systems change upon passing the first high amplitude peak.
In my experiences with servicing all manner of amplifiers, after major/blanket resoldering the amp sounds harsh/all wrong on initial operation at low amplitude and requires running up to brief clipping events to stabilise the behaviour/signature.
For example run the amp at -20dB for extended period to allow thermal stabilisation, initial burn in etc and take note of the sound.
Next, run the amp at say -14dB briefly and then return to -20dB operation and in my experience the sound changes, and for the better.
Listen for this difference.
Dan.
Mark, I'm not sure why my question offended you and why you felt a compunction to respond rudely -- if you haven't played with the chip, then maybe a "hey, no I haven't, but curious what you find" would be a better response than a full accusatory stance? I obviously haven't used it (yet) otherwise I wouldn't have asked. Yes, I'm aware of your efforts in snubbers and diodes. Sheesh.
Perhaps it's outrageous and inconsiderate that I wanted to build off whatever was out there before embarking on said chip. There are some very well-versed and experimental people on this very thread whom might have some tribal knowledge not found on the DS or from LTSpice they're willing to confer. And searches around this site and the net don't yield any alarms or surprises. I wasn't asking anyone to do my legwork, just trying to be thorough in my own.
Last edited:
NOFI required here, loud and obnoxious just like Japanoize will do fine. I think there is a fundamental problem folks think AC/DC is different from Melt Banana or the Boredoms but not to me.
As I'm sure people are bored of hearing me spout, I was part of the Richard Clark Amplifier challenge staff, and I totally agree with this point. I saw trained engineers overlook audible differences under the stress of being tested. Gross differences can be reliably heard under many circumstances, but the mental condition of the test subjects is a major component of performance in ABX. Calling the mental condition a (testing) system noise is appropriate.
We are creatures of relative sensory processing, not absolute, so returning to mental baseline before analysis is a prerequsite, especially for subtle phenomenon.
It is for this reason I can agree with Bob Ludwig that one better way may be to relax into a recording over a longer period of time and take notes on aspects to be compared. I used to do this with customers in our mastering suites during transfers. Once you have done that you can go back and highlight those areas you suspect have issues. Then short segment comparisons may be easier. Also, a specific audible artifact may not be excited by all parts of a recording...
Cheers,
Howie
Whew.
This thread reminds me of Willie Dixon's "I Can't Quit You Baby."
Gonna have to put you down for awhile...
This thread reminds me of Willie Dixon's "I Can't Quit You Baby."
Gonna have to put you down for awhile...
Thanks for the input, hhoyt. I am not surprised that this happens. There are a number of other problems as well. That is why I ignore ABX testing.
Exactly. If it is thought that you need to settle in and listen to a whole piece to hear the difference, then DBT that...
I still don't get the NS-10 thing. You angst about teeny issues in the ADC and use speakers that were considered lousy for domestic use. I get that once you lose something in the conversion its lost but sometime over a beer I would love to understand the love of this speaker when far more accurate speaker in both frequency and time domain exist. I don't expect everyone to use wilson WATTs on the bridge as skywalker sound used to, but why chose something clearly and audibly flawed still confuses me. It looks like otherwise sane people having a herd mentality as it gives 'cred'.
See impulse and waterfall plots vs. other speakers here: http://dt7v1i9vyp3mf.cloudfront.net/assetlibrary/n/ns10m.pdf?jQWj8tYIeZeymRCNXitG9Qfwq9mLf1t0=
If you have a suggestion for a speaker that measures better in the time domain, I'm interested to hear more about it.
Also, some other info:
The Yamaha NS10 Story |
By the way, there are a few amplifiers that are said to mate well with NS-10s, primarily the Bryston 4B. I went from another brand to a Bryston and it was an improvment. I also have a story about someone else's NS-10 and amplifier selection. While I have not seen a study on this aspect of NS-10s, so far I haven't found any evidence to contradict Bryston as good choice. Would be interested to try a Benchmark AHB2, though.
Then there are these things I use with them that also help: Recoil Stabilizers | Primacoustic
And of course, they are used as near-field monitors. I know I can't hear as much detail several feet away as I can from 2-1/2 or 3 feet away. Room reflections and HF loss with distance are some of the problems if not sufficiently near-field. I also make a practice of listening from mid and far field positions, which can be more informative for some things.
Having said all that, the setup I have now works pretty well. I wouldn't be surprised if today I could buy some better speakers that do everything I want. However, no doubt they would be rather expensive. Maybe $5k per box, or more. Any contributions towards that would be gratefully received.
Getting back to the subject of ABX testing, if an experimenter has a reproduction system that allows me to discern as much or more than my present system, then that should be fine. I just want to listen to help verify if that were in fact the case. Measurements would be good too, but if I am being testing for my ability to hear, then I want to first hear the speakers that I will be required to use, not just see some specs. Again, if the experimental setup is as good or better than what I am used to, that would be quite satisfactory.
EDIT: While I do use my system for various applications, I wouldn't suggest it would be a good choice for some one else wanting to listen for pleasure only. In that case, a system that makes everything sound good would probably be more suitable. But, I don't think a system that makes everything sound good would likely or necessarily be a suitable choice for critical ABX testing.
Last edited:
Very few domestic speakers there. I still don't get it, claiming a fast fall off in the bass when the speaker is 10dB down at 200Hz over 1.5kHz is wrong to be. There is no real bass energy so of course it will decay fast. I can see a couple there I would rather have unless you are lord-alge and just want something to sound great in the car so it sells millions in which case fidelity is tertiary concern.
And why no LS3/5a? the original nearfield.🙂
And why no LS3/5a? the original nearfield.🙂
The tight bass may be more because it is sealed instead of ported. Porting extends low frequency response, but by reducing damping and/or increasing resonance. That slows down time domain response. Of course, it is possible to make sealed LF cabinets with good low frequency response and good time domain response, but in that case the trade off is usually lower maximum volume. My impression would be that speaker manufacturers don't find a lot of demand for speakers that are good at everything, except can't get very loud.
Have to agree with you that more bass extension would be helpful, though. Some people switch on a subwoofer when they want it. But, weak bass is an issue NS-10 users have to deal with, and a good reason to look around for something better. Those better speakers would just need to be as good or better at other things besides bass than NS-10 are, and be affordable within my budget.
Last edited:
- Status
- Not open for further replies.