OK, sadly, John is willing to chat about this issue for 70 pages but won't agree to the test. At times you really do give the impression that you care about more than only what works for you, when you talk about teaching and such, but then... I have to say that I'm slightly disappointed. And I know, you don't owe anybody anything, at least not me, not even a reply, much like syn08.
Maybe Andre will do it? Anybody?
Maybe Andre will do it? Anybody?
John,
As this thread was useless, thought I'd hijack it for the chance to ask you a couple of questions that don't require magic to explain. Just engineering. Searching the forum gives me only thousands of non-results.
I have been trying to understand why such a diference between otherwise measured "good" amplifiers. (no, not a discussion of "if") I am looking for dynamic things that can explain some of it. Things that I can measure. Remembering back to my lab days, a big issue in digital was instanious curent starvation. Caused ugly things inside chips as well as slower switching and heat. Is the proximity and instantaneous impedance to all the devices active devices a reasonable possibility? Both power and ground of course. I know this is now addressed in better executed circuits. What measurements would highlight this?
I have seen some odd things on an instanious pure sine pulse. Has any work been done on harmonic or amplitude distortion in a big delta condition? I have been doing 3 cycle pulses of 4K and seen some odd waveforms coming out of the speakers. These effects seem to only be in the first cycle and a half. Looks like a matter of the driver "catching up". Amps seem to effect this slightly.
Is the trend for very wide bandwidth, your amps are notable for this, working to push the time and phase delays in the feedback above the audible range? Wondering if this is why tube amps, with their very limited bandwidth, tend to not sound very smooth with gnfb.
The topic of constant voltage vs constant current has been going on for ever. Thinking back to Quad. What is the current thinking on this? Leaning one way or the other?
What ever happened to the "ACE" concept as presented by Karl Stahl to the ASEE circa '78. Sounded reasonable on paper.
Enough. Thanks.
A bit late and OT too, but my favorite acoustical illusion is to listen to a ca. 70Hz square wave for a while (1 minute or so, and even low volume will do), and then switching it off, listening to music or just any environmental sounds -- you'll think your ears are broken, the sound appears to be very "chopped" and phasy for minutes...
Not at all. Simply show an OBJECTIVE result, rather than SUBJECTIVE opinion, as you insist upon.
What you don't seem to get is that OBJECTIVE, does not depend on the viewer, listener, etc. to exist.
SUBJECTIVE includes the observer's point of view. Without you telling us so, there is no independent observation here yet that an audible difference exists. We can only take your word for it.
Wow, Amazing. You have two positive occurrences out of two attempts. Meaning you flipped a coin and it landed heads-up TWICE IN A ROW! STOP THE PRESSES! TWO OUT OF TWO!
I'd only add that the measurement would need to be objective - I.E. what is the color temperature in the room before and after, recorded using a light meter.
Subjective would be John and Andre telling us why changes in the light switch's wiring - caused by the temperature change left by the finger which flicks it 'on' or 'off' - actually change the color, but that only they can see it, due to their vastly superior 'golden eyes.' - Or something like that.
I looked around, I wasn't able to find anything that said the reward and offer weren't as I said they were. Maybe you could post a link?
A correctly rendered DBT does not require belief. It gives only a probability of a positive or a negative. If the probability is extremely high that you may be able to tell a difference, and several people also independently find the same high probability, there is a concurrently low inverse of doubt left. 'Belief' then, is irrelevant to fact. The whole point of DBT's is to remove 'belief' from the equation.
No where have I DENIED it's existence, I deny anyone can prove they hear the difference.
This is not a DBT. It isn't even a blind test
10 times is sufficient, and generally accepted. Both of you would do well to at least understand what a DBT is, by reading about it.
LOL, after skimming the last day or so, that is my consensus as well.
Not quite. What defines 'dirty' water? Is it contaminated with bacteria only? Without knowing which is which, could you tell the difference? You might later, if you got sick, but your sense of taste couldn't tell.
Lord... Now you're just being silly. The 'Golden Ear' syndrome again, as if you have some super-power of perception that the rest of us mere mortals lack. 'learn to see with your ears'?? Shall we now test you for perfect pitch, as well as the ability to watch a sunset and predict the weather with your ears??
Like how to 'see' with your ears, I've no doubt.
X2.
As pure as I can get. Not intentionally modulated. I use software generators to record a tone, run it through various software filters and normalizer, then use an editor to snip a nice clean piece where the wave starts within one bit of zero. I then add a bit of space before and aft. Perfect, no. Seems to be close enough. I really wish I had a way to do this at higher than CD rates. 192 would be nifty.
Member
Joined 2002
definitely, but not just one person can do it for it to be 'valid.' beyond that, neither the tester or the test-ee can know which are which.
Check this out, It's is as close to an instruction manual as I've found.
It's fine if the claimant does the burn in - as they see fit in any way they like, and be certain it's to their own satisfaction. Once it is submitted for testing in a DBT scenario, no one can know the identity of a given cable.
John could then claim SY didn't do it correctly when he couldn't make an identification. The claimant should be able to test the 'tweek' using his own equipment and methodology.
exactly
there is no need to conceal the identity. the 'tweek' (i.e. the burned-in cable is compared against a non-burned in cable - or any other quality cable for that matter. As long as they and the 'X' are randomized, and no one knows which is which until after a selection is made, it's fine.
Read Here
Another great thing to do is try it yourself on, say, audio codecs or something else easy to get a feel for how it works. Foobar, and a couple other tools have some really easy to use ABX comparators which are easy to use. Try it, it's fun! - If you're geeky like me.
The more results - I.E. the more people - who take the same test, the bigger the sample size and more accurate - i.e. irrefutable - the results.
HERE IS HOW IT'S DONE, not just in audio, but in the scientific community at large.
I'd throw in some cash... We as a group buy a set or two of cable, let the claimants burn them in to their heart's content. Then we do a double-blind survey test with them, against a $20 set of same length and gauge 'Best-Buy Specials.'Id help
That'd be a good round 1.
If they could in fact tell a difference, we could take the one they scored well with, and test it against a non burned in sample of the same cable, i.e. brand new.
If they are right, they should be able to hear a difference "easily".
It would be a good idea to have as many others who think they can hear a difference do the tests as well. The more, the merrier.
Do you mean something like depicted below? If that's the case, it's no pure sine wave because of the sudden onset and infinite derivatives (around the red circle).
Attachments
I have never said it was a DBT. Just checked quickly, since when is an ABX test the only allowed blind test?
Don't mind, thought it would be too much for some.
To be honest, i'm still using the copper cables and dreaming about the silver ones, just too much for my wallet but I believe in miracles.
I will do the test.
What cables would be suitable? Anything that could be DIY'd without a lot of expense?
se
Now you are just flaunting your continued ignorance. An A/B/X test can be conducted as a double blind test, OR a single blind test. A double blind or single blind test can also however be conducted without using an A/B/X technique.
You can also do a simple blind A/B test, (not the same as a 'single blind' A/B test) but these are not generally accepted as valid for aforementioned reasons. The tester can, even unintentionally, influence the results (think 'Pepsi Challenge'). None of these is an exclusive definition of the other.
I think you need to do some more research before you continue on here.
Woody Allen; said:
I will pay for 1 pair of cables in a group buy. I would prefer that we pick from the lo-cost types and I think we should go for something off-the-shelf for convenience and speed, but I'm prepared to be flexible about it in order to facilitate agreement. DIY is obviously OK with me too. Come to think about it, how about CAT 5?
w
Unfortunately the better cables (also more expensive) seems to make a larger difference, so it depends how easy or hard you want to make it for me.
I've listed the cables I've tried earlier, we can also try others but of course I would like to listen to them first.
Yes, and quite correct. Even though this is the test signal from the PC, it has to pass the integration in the D2A and the integration through the amp and the mother of all integrators, the motor. So it would not have the infinite series as you suggest. I wonder, if I looped this pulse quick enough if the spectrum analyzer would be able to show these components. I leave a pretty long gap now as it lets the ringing die out.
I am not looking for something easy, but for something where the amp/speaker reaction shows a difference that could correlate differences in character. Acceleration from rest is not an easy feat.
I am searching for ideas on what to measure. Steady state numbers in the .001% range are well below our hearing. At least at any level I listen to, well below the ambient and below the threshold of hearing period. Hafler showed no measurable difference in to out. Not simple bandwidth, overhead and recovery, TIM, SID, THD, noise profile, level matching etc. All important, but all measurable and can be well controlled in even budget designs. As I am an engineer of sorts and don't believe in magic or essence, that suggests to me we should be looking in a dynamic domain. This was my first guess. I guess I am hoping that not being an expert, I wil l look in the obscure areas.
For those cable testers, I have run RG8 for years. It was designed for high current at low frequencies. Sounds like a speaker to me! Got that tip from a paper at AES many years ago. It seems to not damage the sound to any determinable degree. It does not add color, texture or character ( aka distortion) either. At least not enough to mask diferences in amps and DAC's
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.