Measurements are fine, I do them all the time and am very interested to know how to do better. But they are only a means to an end - for me. That end is that it sounds right. For you, the measurements are the end. Different goals, deferent tastes. Pretty simple.
But believing that changing one's opinion of something after more experience and exposure is justification for a lack of objectivity - well, that's just denying that we can learn and perceive. Pity us poor humans, locked into a world of dark, never changing ignorance. 🙁
But believing that changing one's opinion of something after more experience and exposure is justification for a lack of objectivity - well, that's just denying that we can learn and perceive. Pity us poor humans, locked into a world of dark, never changing ignorance. 🙁
The most valuable information I have gained from listening sessions were subjective. But generally I am in the same listening session to understand what the individual has to say. Sometimes, I ask the question, what do you think can be improved to better express performances. I can get feedback in many ways. Some can totally go against my impression, some actually teach me something new that I had missed, some are long the lines of my impression. No matter what the response, I can generally understand why each response is so. The key is how does the subjective response change when you change the characteristic after measurements or technical reasoning. This is when I try to keep my opinions to myself until the listener has formed an opinion. I think this is a more constructive approach to making listening an important part of development process. Of course the people you select is very critical, they generally have to have the same impassioned for good music, and in some way their listening capability or knowledge about music sort of complement the team as a whole.
Yes, it *is* OK.
People can tell "good from bad" just by listening - it is their experience after all.
-and obviously they can change their opinion about it (.."wind" or not).
THAT'S WHY IT'S OPINION, and obviously - "a totally biased subjective experience".
Measurements are fact - within the confines of the measurement. (..of course from there it's opinion on if the measurements were conducted properly and if they were meaningful.)
Why do you keep equating the two, and then refuting it?
"Hmm, this apple and orange are round - yet they are not the same." (..It's like having a flash-back to an early child-hood memory of an Electric Co. episode - which seemed absurd at age 3 as well.) 🙄
Berating peoples opinions and then exclaiming the superiority of measurements is functionally "tilting at windmills".
Earl doesn't trust his ears, or anyone else's for that matter. That's his thing. I do trust my ears to a large extent, but find measurements very useful when I'm lost.
Earl has achieved results that satisfy him without using his ears. (Tho oddly, he advertises with claims of "best sounding"). It works for him, he's happy with it. It's not everyone's cup of tea, tho.
Earl has achieved results that satisfy him without using his ears. (Tho oddly, he advertises with claims of "best sounding"). It works for him, he's happy with it. It's not everyone's cup of tea, tho.
Funny that Dr. Geddess should mention religion a couple of posts back! No, I'm not gong to argue religious beliefs but I am going to say that many people believe that the lack of any faith necessary to believe in a religion is because of a man's arrogance or, if you like, his belief in his own superiority. Is not this what is coming across in the good doctors posts in this and many other threads. I may be about to burn in hellfire!!!!! 🙂
jamikl
jamikl
What does the above statement mean? Are you filing small shavings off resistors while listening to opera music over the speakers? Only if too big a bit chips off, do you have to resort to looking at a 'scope?
Does the same mistrust of measurements also apply to those detestable calculations? Aren't they really just two sides of the same coin? The measurements confirm the calculations, nothing more. If you distrust measurements, then you also distrust calculations..? But without calculations and using only your ears, you can design an amplifier?
Last edited:
If I have a passive speaker with certain deviations from 'straight' in terms of phase, frequency response, group delay etc. and I set out to create the "best sounding amplifier", tuning it by ear alone, I might end up with, to some extent, an amp with the inverse of the speaker's characteristics. But the measurements would look bad on paper, and the amp might sound terrible with another speaker.
Hopefully my amp would still be stable and all those other boring things that amplifiers have to be.
Has anyone ever designed an amplifier that way? I don't think so. I think everyone starts out wanting to design an amp that is neutral and, as mere confirmation, measures well. Perhaps, rather than consciously tweaking the design away from neutral, they choose a topology with 'character' like a SET...
Hopefully my amp would still be stable and all those other boring things that amplifiers have to be.
Has anyone ever designed an amplifier that way? I don't think so. I think everyone starts out wanting to design an amp that is neutral and, as mere confirmation, measures well. Perhaps, rather than consciously tweaking the design away from neutral, they choose a topology with 'character' like a SET...
To answer the original question, the main things that I've noticed cause me listening fatigue (can't speak for others) are the following:
1) Those old MB Quart dome tweeters. My high school girlfriend's dad had a pair of Chapman speakers for a while, and during that time MBQ was hugely popular in car-fi. I couldn't go half an hour without literally getting a headache. Turned me against metal domes for years! Better metal domes don't do that, though. (And "better ones" were around even back then, from, inter alia, Vifa and Seas.)
2) Unattended cone breakup. A modern offender here is the KEF Q900, which sounds pretty good when set up to my tastes, but is tiring to listen to for long. The breakup of its 8" aluminum midrange isn't as bad as some, due to the lossy attachment of cone to former and other tricks, but it's still there and basically untreated in the crossover. Many "full range" speakers are fatiguing for the same reason.
3) Listening too loudly for a given set of speakers. The easy rule of thumb here is, if it "sounds loud" (i.e. you can just tell it's loud, rather than trying to turn to talk to someone and realizing that you need to raise your own level to compensate for the loudspeaker SPL) then it's going to be fatiguing.
I suspect that the #1 and #3 are relevant and material, #2 doesn't really play into fatigue (unless a vent is really badly misaligned), and for the latter two it depends far more on how they're used than their mere presence.
As for the bass differences between vented and sealed, they're mostly swamped by the room. Assuming competent design of both, at least. Yes, the vented box will be more efficient above tuning, and not be able to pressurize the room below tuning. But we're talking small effects of a few dB, compared to 20+ dB swings imposed by the room. High fidelity means multisubs.
The DSP+multiamp approach can do things that a passive cannot (easily) in terms of time delays, bass EQ, relative levels of drive-units, etc. Also, modern DSPs may lead to better sound in the end because they can allow for rapid switching between two different transfer functions for comparison purposes.
But I've yet to see anything credible that, if a given transfer function can be done with active or passive, the two will sound any different given equivalent power into both. If anything, the passive crossover may have a minor edge, because with so many fewer electronics boxes there's less opportunity for noise (ground loops, extra gain stages, some would argue AD/DA loops, etc.) to present itself as a problem.
I can think of two. Baby boomer nostalgia, and a general culture of retro fetishism.
Minidsp, Behringer DCX, there are probably units from Ashly, ElectroVoice, QSC, Rane, BSS etc. that do their job but otherwise get out of the way too.
For people who need to spend more money (and sometimes get more advanced features) there's Dirac, Trinnov, or DEQX.
If "passive biamping" changes the sound at all, the cause is either psychogenic or user error/incompetence, not the "parallel inputs"...
1) Those old MB Quart dome tweeters. My high school girlfriend's dad had a pair of Chapman speakers for a while, and during that time MBQ was hugely popular in car-fi. I couldn't go half an hour without literally getting a headache. Turned me against metal domes for years! Better metal domes don't do that, though. (And "better ones" were around even back then, from, inter alia, Vifa and Seas.)
2) Unattended cone breakup. A modern offender here is the KEF Q900, which sounds pretty good when set up to my tastes, but is tiring to listen to for long. The breakup of its 8" aluminum midrange isn't as bad as some, due to the lossy attachment of cone to former and other tricks, but it's still there and basically untreated in the crossover. Many "full range" speakers are fatiguing for the same reason.
3) Listening too loudly for a given set of speakers. The easy rule of thumb here is, if it "sounds loud" (i.e. you can just tell it's loud, rather than trying to turn to talk to someone and realizing that you need to raise your own level to compensate for the loudspeaker SPL) then it's going to be fatiguing.
I suspect that the #1 and #3 are relevant and material, #2 doesn't really play into fatigue (unless a vent is really badly misaligned), and for the latter two it depends far more on how they're used than their mere presence.
As for the bass differences between vented and sealed, they're mostly swamped by the room. Assuming competent design of both, at least. Yes, the vented box will be more efficient above tuning, and not be able to pressurize the room below tuning. But we're talking small effects of a few dB, compared to 20+ dB swings imposed by the room. High fidelity means multisubs.
The DSP+multiamp approach can do things that a passive cannot (easily) in terms of time delays, bass EQ, relative levels of drive-units, etc. Also, modern DSPs may lead to better sound in the end because they can allow for rapid switching between two different transfer functions for comparison purposes.
But I've yet to see anything credible that, if a given transfer function can be done with active or passive, the two will sound any different given equivalent power into both. If anything, the passive crossover may have a minor edge, because with so many fewer electronics boxes there's less opportunity for noise (ground loops, extra gain stages, some would argue AD/DA loops, etc.) to present itself as a problem.
I can think of two. Baby boomer nostalgia, and a general culture of retro fetishism.
Minidsp, Behringer DCX, there are probably units from Ashly, ElectroVoice, QSC, Rane, BSS etc. that do their job but otherwise get out of the way too.
For people who need to spend more money (and sometimes get more advanced features) there's Dirac, Trinnov, or DEQX.
If "passive biamping" changes the sound at all, the cause is either psychogenic or user error/incompetence, not the "parallel inputs"...
Last edited:
My beliefs are pretty much in line with Pano's. Measure and listen and find the best blend of the two.
Lets make a distinction between divining the proper balance of a speaker, and detecting if subtle changes are significant. When it comes to whether connectors, interconnects, capacitors, amplifier types, etc make a difference (when measurements show they don't) I would say go straight to blind ABX testing. In my experience that makes those type of differences evaporate into the ether. Expectation bias is everything there and strongly influences "the faithful" in always confirming their beliefs.
But when it comes to evaluating speakers we aren't talking about an ABX situation. Speakers are very imperfect devices and proving they are different isn't necessary. Choosing between different flavors of flawed is a subjective endeavor. Measurements keep us honest and let us throw out the truly bad. They also let us converge in the design process quickly on something that will be pretty good. Still, I would never sell something that had only been measured, never listened too. We are still too far from knowing exactly how to evaluate our measurements to rank order similarly worthy units. (At the same time I have never seen measurements to be grossly misleading: good sounding speakers measure well. Bad sounding speakers measure poorly in some manner.)
I have found blind testing to be essential in subjective evaluation. We had a situation at psb where friction between designers was becoming obstructive. I instituted tests behind curtains and it was very enlightening. For the most part the audible differences between designs didn't change but your preferences becames much less firm. When you don't know "which is the new one", "which is mine", "which has the new tweeter", then the confidence of your choices, "which one is best", drops dramatically.
David
Lets make a distinction between divining the proper balance of a speaker, and detecting if subtle changes are significant. When it comes to whether connectors, interconnects, capacitors, amplifier types, etc make a difference (when measurements show they don't) I would say go straight to blind ABX testing. In my experience that makes those type of differences evaporate into the ether. Expectation bias is everything there and strongly influences "the faithful" in always confirming their beliefs.
But when it comes to evaluating speakers we aren't talking about an ABX situation. Speakers are very imperfect devices and proving they are different isn't necessary. Choosing between different flavors of flawed is a subjective endeavor. Measurements keep us honest and let us throw out the truly bad. They also let us converge in the design process quickly on something that will be pretty good. Still, I would never sell something that had only been measured, never listened too. We are still too far from knowing exactly how to evaluate our measurements to rank order similarly worthy units. (At the same time I have never seen measurements to be grossly misleading: good sounding speakers measure well. Bad sounding speakers measure poorly in some manner.)
I have found blind testing to be essential in subjective evaluation. We had a situation at psb where friction between designers was becoming obstructive. I instituted tests behind curtains and it was very enlightening. For the most part the audible differences between designs didn't change but your preferences becames much less firm. When you don't know "which is the new one", "which is mine", "which has the new tweeter", then the confidence of your choices, "which one is best", drops dramatically.
David
Basically what David says in his post that starts with this:
His full post sums it up well.
Shaving resistors. Now you're just being silly. 😛
I can agree with Dave as well, but that is not the way many, if not most, come across here. If the measurements or physics disagree with them then they are wrong - hearing is the final judge, as if it is not flawed.
One must have a correlation between what is measured and what is heard. Once that is achieved then the measurements will trump the subjective because they are not prone to bias. I believe in my measurements so listening is no longer necessary.
And as Dave said "At the same time I have never seen measurements to be grossly misleading: good sounding speakers measure well. Bad sounding speakers measure poorly in some manner."
One must have a correlation between what is measured and what is heard. Once that is achieved then the measurements will trump the subjective because they are not prone to bias. I believe in my measurements so listening is no longer necessary.
And as Dave said "At the same time I have never seen measurements to be grossly misleading: good sounding speakers measure well. Bad sounding speakers measure poorly in some manner."
Friend of mine from school use to make speakers, with the best advise above being followed. His blind testing was something that was a hard pill to swallow tho. He'd get so bloody trashed he wouldn't remember the changes made and thus thought this would make a good blind test parameter. Stopped by one day for the heck of it and heard this gawd aweful sound coming from his garage. What you up to Nick? That's rather narly speaker you've got there. Yeah I couldn't figure out what I had done so terribly wrong and just can't put my finger on it. The problem is called a drunk design with a hangover solution,... Dude! 😀
Actually...
When I first started building my own stuff when I was about 12 I used to use big old carbon resistors (terrible I know) that I'd pulled out of old radios. I didn't have many so whenever I needed a value that I didn't have I used to take a lower value and scrape or file off as much from the middle as I needed to increase it's value to the value I needed! I suppose you could call it shaving 😛!
Precursor to precision laser trimming.
Maybe there's a market for "lovingly hand trimmed" resistors?
Anyhow, thats dedication.
LOL - yeah. I'm sure you could market them as being able to "better define nuance, resolution and timing as making things by hand is more natural and therefore in more tune with the natural tones of the music when placed in an SET amp" and lots of other subjective woo...
I remember my Mum telling me off for using the cutlery to cut them down the middle and getting the filings on the bedsheets - they still have black smudges on them!
I remember my Mum telling me off for using the cutlery to cut them down the middle and getting the filings on the bedsheets - they still have black smudges on them!
Dunno Dave, John Krutke was building a custom crossover with a stepped attenuator in 1dB increments awhile back. Cost something like $12k. So yeah there could be a demand (but not for carbon 😉
If there's a market for tube amps with >2% THC then there must be a market for carbon resistors...
Surely they must sound more natural after all as they are more simply made, thus providing the shortest path for the audio signal which might be tainted by more complicated components😀...
Surely they must sound more natural after all as they are more simply made, thus providing the shortest path for the audio signal which might be tainted by more complicated components😀...