Which means that both test series potentially show there are problems with short-term ABX at revealing subtleties...but both tests certainly reveal the rhetoric surrounding marketing amps is, well, overblown, one might even call it distorted? (runs from the thrown tomatoes...)
Just my 0.02 worth.
Howie
Partially.
Howie rightly pointed out the problem that comes with 'blindness'.
I could add some remarks.
ABX is confusing. It is hard to compare more than two sources at once.
(Often, those blind tests looks like flawed, oriented to demonstrate that something do not exists.)
This kind of test add stress, specially when you are not alone and quiet in an unfamiliar environnement. No good for an accurate listening.
Sources have to be chosen carefully for this purpose. Not often the case.
- They have to be well known by the listener and agreeable to them. No way to listen to a track you dislike or have to discover.
- Sources have to be chosen FOR what you are searching for. No way to try to find distortion differences with horns or some electric guitars ;-)
Worse when the source is yet awfully distorted during the recording, like in one test I have seen on this forum, with a terrific track of Neil Young used to track differences between compression ratios.
- No way to analyse instant dynamic with violins or organs etc.
- No way to compare things that cannot be compared, like switching from A to B while a tune is playing: you compare two different musical parts.
- The same musical sample you use to compare has to be long enough to allow the brain to analyse-it, short enough for it does not forget in between.
- The sample has to be edited for the switch happens without a break in the rhythm, that can had un unwanted stressing event and distract attention.
In the same time, it is good to let the listener free to chose the moment he want to switch and the number of time he want each thing to be played.
Of course, the all test has to be short enough to not add fatigue, long enough to have some meaning.
All this is boring, complicated, time consuming. Reason why I prefer *for me* Simple Sighted Subjective Tests.
Anyway, I love to fool myself and convince myself of fairy tails that I already believe in hard.
With the blessing of the many, just to feel less alone with my certainties ;-)
Oh, and I love to buy expensive snake oil and to be manipulated by
hypnotists or, how they call them in the TV, mentalists ? ;-)
Finally I let this kind of tests to people that want to make researches on Psycho Acoustic, like Fletcher and Munson. We need more of them.
Tryphon Tournesol
Attachments
Last edited:
Howie, could you roughly define what you mean by short term and long term? Some people might think sitting at home with foobar for 30 minutes instead of 5 minutes should be long enough for revealing whatever subtleties there may be. From your experience, should we be thinking in terms of minutes, hours, days?
No, because the goal was to find audibility of frequency response modulation by speaker impedance curve for defined amplifier output impedance.
No. Not really; appears you do not understand how that process helps with accuracy.
THx-RNMarsh
Hmmmmm. comparing a high power amp with a low power one but below the 2% level of the low power amp? You dont see anything wrong with that? Really.
THx-RNMarsh
Last edited:
Great response ! I would add being put into an acoustic listening space you have never been in before and short term comparison listening. Best to do test in ones own home with familiar music and only switched to compare at sufficiently long periods of time. Weeks minimum.
If you have to compare a LOT of amps/products to each other this could take a LOT of time !
-Richard
Last edited:
You seem not to understand your definition of "accuracy" is subjective....
We'll have to agree to disagree.
YouTube
Richard, just please stop, honestly. Accuracy also must mean something wildly different to you than to me*. At this point everything you've written is heresay about "accuracy" or "better" or what have you. If you have a case to make with some actual data, then please be forthcoming. And not data/studies that are not germane to your own point.
Here's what I'm working with (and the established scientific community) on accuracy and precision: Accuracy and Precision
Now, when it comes to audio playback, it's going to be hard to separate these two. But the moral of the story, if we're going to talk accuracy, it's basically output = input*gain.
I like the contrast between this view and Earl Geddes' view. Earl seems to have a distinctly non-subjective view of accuracy while at the same time thinks THD is almost meaningless and many competently designed amplifiers (probably both VFA and CFA) are indistinguishable.
Wrt the clark challenge, we discussed it occasionally in the past and pointed out that there might have exist some problems with gathering of the data as the reported number of trials with no one getting more than 6x% correct (even due to chance) is extremely unlikely.
And of course it is known that the ABX protocol might lead to inferior results compared to (for example) an paired comparison test (i´ve cited some experimental numbers from the 1960s). Back then the difference was attributed to the more complex mental processes involved during an ABX test.
And of course it is known that the ABX protocol might lead to inferior results compared to (for example) an paired comparison test (i´ve cited some experimental numbers from the 1960s). Back then the difference was attributed to the more complex mental processes involved during an ABX test.
That's a good question, just for me I would say it takes hours of listening with different material I am very familiar with in a familiar space...once again, in a subjective manner only. I think the emotional and stress aspects of ABX could certainly inhibit clear identification it even if one was to spend hours.
Here again, I think the RC Challenge was designed to figure out if two amps sounded so different that someone familiar with one amp could pick the place, time, source material, etc. and reliably tell the difference in a few hours of listening. This does not mean in all setups there is no difference. It is a very specific test.
Richard, I was merely quoting Richard Clark's test parameters. I dislike the single number distortion spec (or single number spec for anything, really), and wish we could report each harmonic rising above 0.1% or some other number, or a weighted single spec giving the first few odd harmonics heavier weighing. In regards to the power difference, a 1kw amp for PA use with higher gain will often have higher noise than a 10 W one in a test like this.
I chimed in on the RC test to clarify what we were actually doing, and thought I made a point of saying it is a very specific test which doesn't say everything about an amp. What it does say is interesting about amps in general, ABX testing and people's expectations as to the potential differences. If the differences were as huge as people thought, it should have been easy to identify one over the other...which never happened.
PLease keep in mind the impetus for this challenge was people saying there were huge differences between amps, so Richard said, okay what are they? Can you point them out to me reliably? He even told a mfgr of a jaw-droppingly expensive tube amp that if it sounded better or even different he would purchase two for each mastering suite at our factory...and we kept using Crowns...once again, I am not advocating for Crowns...please focus on what this challenge was and not on my reporting of it.
Cheers!
Howie
Last edited:
Not only when evaluating but also when just enjoying a listening session to ones owns music collection I use to some times turn of all light and have all windows optically sealed completely blocking out any light from outside, it's in my opinion a magnificent experience sit there in the complete darkness more or less able to perceive just the acoustic information I have noticed others simply can't walk into a dark room and relax, this is maybe some of the most common things many people have difficult with in these days hectic society, they aren't able to relax and use the senses to the fullest.
Last edited:
R2R tape. That is what I would use for a heavy test. Could splice the same musical passage over and over. Make a simple tape yourself with good mics... Then one generation down transfer to make the splices.
This would be logical.
I just wonder when two amps sound the same if it was because of a so so CD being used as the source. I ran into this scanario at Harman a few times.
This would be logical.
I just wonder when two amps sound the same if it was because of a so so CD being used as the source. I ran into this scanario at Harman a few times.
- Status
- Not open for further replies.