Blind Listening Tests & Amplifiers

Status
Not open for further replies.
Bratislav

"Can you really say this with a straight face ?"

Had no mirror in front of me writing that but I think my face looked about the same it use to.

"Or do you really understand what are you talking about here ?"

Yes I know very well what I talk about, do you?

"Any idea how much residuals will be there that are introduced by speaker cable ? "

Not in exact numbers, relative small errors though but still audible without a second of doubt. Skin effect is relatively high in normal cables and Ben Duncan and Tommy Jenving has prooved (by measuring) that there are clear diffrences in the harmonic series (THD)of speaker cables driving real loads.

"How about this experiment : disconnect your left speaker and connect it between amp's right 'hot' terminal and right speaker input 'hot' terminal."

Interesting suggestion, maybe I´ll try that for fun when I have some time. However, the test doesn´t at all have any correlation with detectable distortion on normal listening levels, as I´m sure you understand.

Happy listening!

/Peter
 
Hold it and sit down, nw! Sit sit sit!!!!

nw,
How can you say I suddenly came with blind tests now?
If you read my posts I gave various examples.
I even talked about my wife's reaction to a component change in my amp.
You seam to go more for the technical explanation of why does it sound good, and it seams you don't know how to HEAR.
Take a look at this link:

http://members.chello.nl/~m.heijligers/DAChtml/Project description.htm

Please read it, at least until the first picture.
Interesting that man has the same oppinion I have.
I'm not a techno-junkie, and I think Audio is not such an exact science.
You can design a crossover for a speaker on a computer software and it may end sounding like crap if you don't fine-tune it by ear.
You can measure more resistance, capacitance, etc. on a cable and it still sounds better than the other.:devily:
Please trust your ear to detect good gear.
Oh well... I feel like praying in the desert.:bawling:
 
Doug's objectivity

All good amplifier stages ignore imperfections in their power supplies, op-amps in particular excelling at power-supply rejection-ratio.

As an objective statement the above would seem to show that Mr. Self has a serious lack of understanding of basic electronics!

No amp I've ever measured ignores it's power supplies - the degradations are clearly measurable. And if he thinks op-amps all have excellent PSRR then I can only say his thinking is a little to DC for me.

Like almost all op-amps almost every parameter degrades at 6dB / oct above a certain point. This point is often before the start of the audio band.

Hence by the time we're a good way through the band PSRR looks very poor indeed. Since we are trying to create a system capable of dynamic ranges of the order of 120dB, it's easy to see, hear and measure the degradation.

Unless Mr. Self has managed to design an amplifier with over 90-100dB PSRR throughout the entire audio band?

I'm fence-sitting here, as both measurement and listening ARE crucial to audio design. Without determining the effects of changes by both methods you cannot then determine the best use of resources to progress the design. I therefore support both technical and subjective views, but Doug's arguments are very poor and blinkered in their outlook.

Andy.
 
AX tech editor
Joined 2002
Paid Member
Pan said:
[B[snipWe can not hear a difference if there is no measurable difference, that is clear as a bell. I know some people beliewe in "magic" in hifi equipment, plaese do not put me in that camp.
[snip]/Peter [/B]

That is what I also thought, but recently I have begun to change my mind. It is important to distinguish between hearing as in activating the hearing sense system, and in hearing as in perceiving something. If you really dig into it, it almost is like the two are totally separate.

I am convinced now that you can have two systems measurably identical and still being perceicved a ssounding differently. It is a long story, but perhaps I can give an example from the visionary system.

If you ask two or more people who witnessed a traffic accident what exactly happened, I bet you get different stories. Yet, the scene was the same; there was just one accident (hopefully). I have been in a court case several months after an accident, and it almost looked like there really were two accidents, so totally differing were the witness descriptions.

It is my conviction that most of the cases of reported change of sound with break-in can be attributed to hearing the same, but perceiving something else. Not all; I am aware that for instance speaker drivers change there physical properties when being used.
One thing that has always bothered me is that everyone always reports an improvement from break-in. With all the variables involved, if there was a break-in effect, somebody (say 50% of the cases) would also report a worsening effect. I have never heard someone say: "I tried this cable over a few days, it started to sound worse and worse, and at day 3 I threw it out, it's crap".

Jan Didden
 
Break-in

One thing that has always bothered me is that everyone always reports an improvement from break-in

Jan,

That's quite thought provoking - I have a theory that may explain it, see what you think.

It's entirely possible that burn-in is primarily a lack of change, or a slowing of the rate of change to inaudible levels.

Any change in a circuit that shouldn't be there is, by definition, an error signal. We can debate the audibility, but the effects are real, and physical in most components, from metal-film resistors to caps to semi's.

Whilst it doesn't explain the mechanisms in cables (dielectric effects?) it's a feasible theory for everything else. I can also find plenty of people who've experienced worse performance during break-in but better results after, hence the theory.

Andy.
 
Disabled Account
Joined 2002
Re: Re: ...to continue pan,

Pan said:


To repeat myself, if two amps have truly identic behaviour under all practicall situation as reactive loads and different levels, then off course there would be NO audible diference. Since there are no two amps that are identical, there will never be two amps measure exactly the same. Now can we hear that is the question we have been beating to death here :) and off course that depends on how much the measured difference is.
Listening tests indicates that we hear much smaller errors than some have been made to believe and/or our measurments fails to get the whole picture (so to speak) when an amp drives a complex reactive load with complex program material.

This amps I have tried I can tell apart in a couple of seconds, that big a difference there is.

Audio Research VT100
GamuT D200
Pass Aleph 5
Zapsolute
Patriot V100
Gainclone
Holfi Integra and Audis
Luxman
Arcam

Maybe these manufacturers lies about the THD levels and the reason I can hear the difference so easily is because off high distortion. I suspect not though, but simply that a high resolving rig can and do differentiate between such low distortion amps as above.

Happy listening :)

/Peter

...pan, you will not be capable of telling the difference between these amps for calibrated, identical voltage swing across the same loudspeaker in turn....that is the point you repeatedly, and doggedly insist on missing...:nod:
 
Re: Re: Re: ...to continue pan,

mikek said:


...pan, you will not be capable of telling the difference between these amps for calibrated, identical voltage swing across the same loudspeaker in turn....that is the point you repeatedly, and doggedly insist on missing...:nod:

Yes I would be able to do that in seconds without a doubt.

As a matter of fact the difference between some of these amps are so big, you could play them at different levels and I not only hear the difference, but I can point out which is which.

You could play one amp at 0dB, -3dB and +3dB, swap for the other amp and do the same with that one, and I would tell you which was which.

Ther´s a big difference in how these amps handle the highs, the extreme lows, the midbass and the mids.
The FET amps (Aleph5 and GamuTD200) have rising distortion at about 1k and that is clearly noticable because they can not reproduce a cymbal or metal strings the way the Zapsolute can (BJT class A).

Neither of Aleph5 or GamuT has the same level in the deepest octave as the Zap, but both have subjectively a fuller midbass.

Comparing the Zap with Aleph5 was so easy (telling them apart that is) that I could listen to only the recorded reverbtails and not the "direct" sound itself, and the reverb sound of Aleph5 was centered more around the mids while the Zap had more emphasis on the highest octave.

Zap compared to GamuT, Zap sounds more extended in both extremes while the GamuT sounds a little more "midrangy" in a warm positiv way. One could say that the GamuT is somewhere between Zap and Aleph5.

How come you believe that those amps would sound the same, driving complex loudspeaker load? CDP/SACD players have magnitudes lower distortion than power amps, and also have a much easier task driving a high, mainly, resistive load, and still there is a clear difference is sound between those sources.
Heck, only changing the clock in my SCD-XB940 and my friends CD63KI-SE made such a big difference we could hardly believe our ears.

Don´t mean to sound rude but guys, you are missing out someting if this is what you believe.

Peace and happy listening! :)

/Peter
 
One more thing

My friend needed something better than his 70´s BO so we decided to try the LM3886 simplistic solution. Having heard the Rowland amps I knew that there was a potential in this chip.

We put together a 200VA toroid with schottkys and some surplus caps 2x 2000uF or so and hardwired the chips to this on a heatsink without a box.

We than compared it to my Zapsolutes and I got a real surprise there :) The bass was less dynamic (lacked power and drive) on the LM3886 but heeey, the Zap´s have dual mono 500VA and 260.000uF each side, so that would be excpected.

Also in the highs the Zap was much better. Cymbals sounded like white noise were added on the LM3886 while the Zap could sort out complex HF sounds much better for a much more realistic result. One other thing the LM3886 didn´t manage so well was low level resolution. Hearing the ambient clues and reverb tails of natural halls or digital reverbs was possible down to a certain level, then it was like the signal was clipped, it simply vanished in the noise floor of the LM3886. What was a 2s tail on Zap was only 1.5s or so on the LM3886. Of course we didn´t measure it but the difference was very obvious.

NOW, what was it that made me surprised then..... MIDS, MIDS, MIDS, this little LM3886 amp had a very good handling on the high level part of voices and acoustic instruments. So good that I said to my friend "heck, this track is better on your LM3886 than on my class A monos which cost about ten times more.
The result was so good and so surprising that I will build a "all out" IC chip amp with better parts and batteries. If I can manage to better the highs, lows (I´m positive the lows can be fixed) and low level resolution, then I maybe give up my Zap´s for the main rig.

Friend called me after listening to his new amp on his cheap speaker and he was very pleased as there was a clear improvement even in his rig, which I later could confirm.

/Peter
 
Jan:

>One thing that has always bothered me is that everyone always reports an improvement from break-in.<

Based on how I do listening tests, which is with multiple units that may or may not be identical inside, are visually identical and sealed, I would say "not necessarily". I find that sometimes burn-in results in no perceivable differences before and after, sometimes there are sonic differences that are not perceived as improvements, and sometimes the perceived results are worse after burn-in.

Listener's perceptions can also change over time, despite there being no significant changes to the device(s) under test. For example, I have tried some things that I thought were promising at first listen, but subsequently rejected. When I listen, I tend to focus on how the music is expressed rather than on individual sounds, and it may take me some time to build up an assessment of the balance of merits and demerits.

The above reflects my personal findings and perceptions. YMMV.

regards, jonathan carr
 
Re: pan...

mikek said:
this should help your learning process:nod:

http://www.dself.demon.co.uk/subjectv.htm
Since Pan is a big fan of the differences in cables, he may also want to read this:

http://www.verber.com/mark/cables.html

It's written by an audio designer (John Dunlavy) who confesses that he'd love to design cables that genuinely sounded better in blind tests, but it doesn't seem to be possible. He goes on to talk about a carefully controlled blind test where GoldenEars were brought in to evaluate speaker cables. As expected they heard "big differences" between them but they never told them they were listening to 12 gauge zip cord in all cases! They thought the biggest, sexiest looking cable sounded best. They never told them for fear "they would become an enemy for life" if they knew the truth!

This is just like the Candid Camera episode where wine tasters were asked to evaluate 4 glasses of wine, each sitting in front of a different bottle, and everyone tasted big differences between them. Then they revealed all four were poured from the same bottle. Pan: You are vastly understimating the psychology behind non-blind listening.

The above link is referenced in this excellent article that talks more about blind testing along with several other references regarding cable myths:

http://www.svconline.com/magazinear...zineid=49&siteID=15&releaseid=3221&mode=print
 
Re: pan...

mikek said:
this should help your learning process:nod:

http://www.dself.demon.co.uk/subjectv.htm

I do not understand what you mean.
What is it that you feel that I need to learn???

BTW, I have read Self´s site (some of it) and he seems to have things to learn about high performance audio.
Is he your guru and guidance in life since you feel I need his findings?

/Peter
 
Disabled Account
Joined 2002
Re: Re: pan...

Pan said:


I do not understand what you mean.
What is it that you feel that I need to learn???

BTW, I have read Self´s site (some of it) and he seems to have things to learn about high performance audio.
Is he your guru and guidance in life since you feel I need his findings?

/Peter



...:D...no...he's not my 'guru'.....but on the basis of what you've written in this and other threads, there is much for you to learn......:nod:
 
nw_awphile,

"Since Pan is a big fan of the differences in cables, he may also want to read this:"

You are just to funny. What does mr Dunlavys test mean?
Nothing at all, well it means that in that particular test that was the result, so what?

Besides, I´m not close to be a "fan of the diff...". Cables is what I care the least for in my rig, it´s what I have paid the least amount of money for (about $15) and I can assure you that the cable issue is not anything that keeps me awake at night :)
Most of the effort and money should be concentrated to speakers and room acoustics if one want to experience high performance audio.

About Dunlavy, his loudspeaker is not exactly what I would call SOTA, and in my opinion it may be necessary with higher quality parts to show the difference. I never heard cable differences when I had a speaker with Vifa drivers and normal lively room acoustics. Moving to better drivers and acoustic tools made some difference I can say.

Besides, I don´t know how the set up was in that test, maybe I´ll go to his site and check out for fun.

"Pan: You are vastly understimating the psychology behind non-blind listening"

No I do not. I have done blind tests and the result is the same for me no matter if blind or not. I have a good pair of ears and a very critical mind. You are wrong. I guess some persons get biased by the press and such, I´m not such a person though.

Oh, BTW this may be a laugh for you;

Did yet another blind test a year ago or so, after I moded my SCD-XB940 with LClock and Zapfilter. This mod made this player a very good one from have being one I did not listen to becasue it did sound so bad. Told my friend about the success and we decided to compare my SACD to his $200 carousel Japanese CDP in his set up. His set up is far from my in terms of resolution but it use to sound ok though. He did the swapping and I had no clue which player that was spinning the disc. I could tell the two apart in every instance he played same track on both players, BUT sometimes I mixed them up and simply prefered his cheap CDP to mine. This was depending on which kind of tune was playing. his CDP sounded a little brighter and my a little warmer, and on his not so highly resolved rig, the CDP that sounded best was the one that best matched the tonality of the track playing.
Now, think about this, I take his player to my place and find out that his player sounds better on my rig..... huuuuuuuuhh scary thought :D

My point: both these players have very low THD and have flat fr. response, I had no problem at all tell them apart in a BLIND LISTENING TEST though.

/Peter
 
"The above link is referenced in this excellent article that talks more about blind testing along with several other references regarding cable myths:"

Yea, I´m sure this guy has all the answer... :D he seems to have a problem understanding Ohms and Volts though judging by reading his thought about interconnects and speaker cables. Hehe!

/Peter
 
Status
Not open for further replies.