Mains Upgrade

[bigboss]

Statistical studies are a lot more complex than that. You have to arrive at a 'p' value, which assumes that the null hypothesis is true. The lower the p-value, the less likely the result is if the null hypothesis is true, and consequently the more "significant" the result is. If the p value < 0.05, then the study is considered significant.

So if these blind tests have been inconclusive, it either means that the study was underpowered to give a result (i.e, a larger sample size is needed) or that there is no significant difference between the power cords.

 

[aliEnRIK]

bigboss:

So if these blind tests have been inconclusive, it either means that the study was underpowered to give a result (i.e, a larger sample size is needed) or that there is no significant difference between the power cords.

Or the people simply were not 'golden eared' enough

Or the test wasnt very well done in the 1st place

If we followed what you said there and lets say 1 million people took the test. Lets say 10% got the audiophile cable right 100%, but statistically speaking overall the results were very poor (less than 20% managed to tell any differences over the 1 million people). What does that prove?

 

[Anonymous]

Why are people so scared of blind tests?

 

[aliEnRIK]

Tarquinh:Why are people so scared of blind tests?

Im not scared of them Tar'. Its just they seem very questionable when comparing hifi equipment (Which im sure you know when looking back through history)

 

[Anonymous]

Wasn't talking about you specifically, Rick. Just ruminating on the fact that on any forum the moment you mention blind-testing with reference to cables people start going into overdrive trying to deny their efficacity.

I don't think they're questionable at all, by the way, in fact I'd like to see more of them in hifi testing.

 

[bigboss]

aliEnRIK:

Or the people simply were not 'golden eared' enough

That's the case with the majority of the population. The purpose of a large sample size is to reflect the general population without bias.

aliEnRIK:

Or the test wasnt very well done in the 1st place

Yes, that's very much possible. That's why we should read the study in close detail (critical appraisal)

aliEnRIK:

If we followed what you said there and lets say 1 million people took the test. Lets say 10% got the audiophile cable right 100%, but statistically speaking overall the results were very poor (less than 20% managed to tell any differences over the 1 million people). What does that prove?

If the p value > 0.05, then the result is not significant (i.e, there's no significant difference). That's the law of statistics. All studies are based on statistical laws. If anyone tells you that, say, 32% of respondents identified the cable right, simply ask what's the p value. If the p value < 0.05, then the result is significant, otherwise not.

If only 10% of people out of a million get the audiophile cable right, that's pretty poor actually (worse than flipping a coin)

 

[aliEnRIK]

bigboss:

aliEnRIK:

Or the people simply were not 'golden eared' enough

That's the case with the majority of the population. The purpose of a large sample size is to reflect the general population without bias.

aliEnRIK:

Or the test wasnt very well done in the 1st place

Yes, that's very much possible. That's why we should read the study in close detail (critical appraisal)

aliEnRIK:

If we followed what you said there and lets say 1 million people took the test. Lets say 10% got the audiophile cable right 100%, but statistically speaking overall the results were very poor (less than 20% managed to tell any differences over the 1 million people). What does that prove?

If the p value > 0.05, then the result is not significant (i.e, there's no significant difference). That's the law of statistics. All studies are based on statistical laws. If anyone tells you that, say, 32% of respondents identified the cable right, simply ask what's the p value. If the p value < 0.05, then the result is significant, otherwise not.

If only 10% of people out of a million get the audiophile cable right, that's pretty poor actually (worse than flipping a coin)

But what would the statistics tell us if that 10% were tested over and over and got 100% correct?

 

[aliEnRIK]

Tarquinh:

I don't think they're questionable at all, by the way, in fact I'd like to see more of them in hifi testing.

If you were to choose between say 2 pairs of speakers, and in a blind test no one could discern ANY difference between them.

But when you yourself 'listened' to them you could easily tell a difference and therefore had a preference, would you still buy the prefered speakers if they cost double the price? Or would you buy the cheaper ones as you trusted the blind test without question? (Theoretically speaking of course

)

 

[bigboss]

aliEnRIK: But what would the statistics tell us if that 10% were tested over and over and got 100% correct?

That's a completely different study then...........a new study recruiting only these 10%. But then the results should be interpreted with caution, as the study population will not reflect the general population. A new set of characteristics of that population should be identified, so that people know if they fall into this group.

 

[idc]

Rick, google blind test, ABX test, audio, audiophile in various combinations and you will find that the majority of such tests have shown that the audiophiles cannot tell the differences. I go with the majority verdict here. If it was the other way around and such tests consistently showed cables do make a difference, I would accept that.

I suspect taht you would also be more inclined to accept such results as they are in your favour now. I also suspect that cable makers would also use such tests to prove their cables do really make a difference. But they do not.

Does that not ring any sort of alarm bell for you?

 

[bigboss]

idc: I also suspect that cable makers would also use such tests to prove their cables do really make a difference. But they do not.

Does that not ring any sort of alarm bell for you?

I respect your views on this, idc; someone who has dabbled with making cables yourself.

 

[AL13N]

idc:In medicine and other ares where ABX/blind tests are used, if there are a series of tests that show no difference between say a medicine and a placebo, the tests are not flawed, the medicine does not work. Why should that not be the case with cable or other audiophile blind tests?
Because medicine is science. Audiophilia is passion.

How about this. If you're interested in a cable then purchase it, making sure you have minimum 30 days to return it (if not, look elsewhere). Take it home and do nothing with it yourself.

Give it to someone you trust and give them free reign to do whatever they like whenever they please without you knowing.

Do all your listening at the same time each day, preferably in the evening/night with the lights off so you cannot visualise any changes. Use old favourites and new music. If you hear anything different, whether better or worse, make a note of the exact time and date. The 'cable switcher' will also make a note of their changes to your system.

At the end of the 30/60/90 days compare notes. Then either keep the cable or return it.

Just my two pence worth (and I've never spent one on mains cables

).

 

[idc]

Thanks bigboss.
I have not blind tested my DIY OOTW Solar Mk I or II with the SHB or Choseal interconnects that I have. But I have been swapping cables about a lot. The one thing that was clear was the different cables have different volume sweet spots. So with each I need to make adjustments with the volume. Certainly there is a larger sweet spot with the OOTW MkII than the SHB.
I can only assume because there will be a different resistance/impedance between the cables that there is a variation in the voltage between the source and the amp. Attenuators can make a sound difference by adjusting the voltage and so giving greater volume control. On my old setup a tiny volume adjustment could make a huge difference to the sound. In the sweet spot it opened up the music by improving clarity, detail and dynamics. When I added attenuated interconnects that sweet spot opened up and became larger. So I can see different cables making the volume setting different which in turn makes them sound different.
That is now my theory as to why cables can sound different, but in actual fact are not different. So if you equalise the volume in any set up (so the actual volume is the same, not where the volume knob sits) you will not be able to tell the difference between cables.
If you do get to play about with the volume, but do not know which cable is which, you will not be able to tell which cable is which.

 

[aliEnRIK]

idc:

Rick, google blind test, ABX test, audio, audiophile in various combinations and you will find that the majority of such tests have shown that the audiophiles cannot tell the differences. I go with the majority verdict here. If it was the other way around and such tests consistently showed cables do make a difference, I would accept that.

I suspect taht you would also be more inclined to accept such results as they are in your favour now. I also suspect that cable makers would also use such tests to prove their cables do really make a difference. But they do not.

Does that not ring any sort of alarm bell for you?

I completely agree with you in some ways. Ive found some cables to be utter rubbish (I'll not post them here in case I get into trouble).

I suspect some manufacturers dont wish to get into these tests as they could easily find their cables are seriously overpriced or whatever. Im suprised ive never seen some of the cheaper ones showing their findings mind (That said, if they did then theyve set a 'standard' and pricing wars might start. Good for the consumer, not so good for the manufacturers)

But again, I want all the 'golden eared' to be tested first and formost. I'll leave the deaf till last

 

[aliEnRIK]

bigboss:

aliEnRIK: But what would the statistics tell us if that 10% were tested over and over and got 100% correct?

That's a completely different study then...........a new study recruiting only these 10%. But then the results should be interpreted with caution, as the study population will not reflect the general population. A new set of characteristics of that population should be identified, so that people know if they fall into this group.

I completely agree with you. But would you not then agree that this would indeed prove that differences must then exist and are perceptable? (If only to a certain percentage of the population)

As ive stated before, my mate cant tell the difference between SD and HD on his own tv (Which I can with ease). That is NOT the sort of person id want blind testing video cables for example

 

[idc]

The majority of blind tests are with audiophiles, who else would want to take part!? So it is reasonable to assume they know their stuff and yet............they cannot reliably tell the differences.

 

[aliEnRIK]

idc:The majority of blind tests are with audiophiles, who else would want to take part!? So it is reasonable to assume they know their stuff and yet............they cannot reliably tell the differences.

Heres what I know ~

When I first tried a mains cable it was on my (then) Arcam A85 in place of the stock cable (Which id been using for years). The affect it had was immediate ~ bass was so much more controlled and powerful.

Since then Ben Duncans measureably proven that braided mains cables in the right setup can make a meaureable difference at an amps output

With all the testing ive done since im absolutely positive that 'certain' cables in the 'right conditions' can make a system sound better. Ive even had them make my (now replaced) Pioneer tv look better

I recall a guy on an american site taking part in a blind test. He swore he could tell a difference between some amps (If I recall, at least I think it was amps). Anyways, he couldnt. He failed to tell differences in the blind test and due to that he actually sold HIS amp (Which was in the test). Months later he realised it was a mistake and that he should never have sold it. Why? Perhaps a few minutes listening to a certain part of a track doesnt tell you the whole story...............

 

[Anonymous]

 

[idc]

Rick, my volume theory explains why some cables make a 'difference' in some systems.

Ivor Tiefenbrun the founder of Linn failed a blind ABX test switching between digital and analogue.

The chap who sold his amp and then regretted it is explained by the loss of the feel good factor and looks of an amp he obviously really liked.

 

[aliEnRIK]

idc:

Rick, my volume theory explains why some cables make a 'difference' in some systems.

Ivor Tiefenbrun the founder of Linn failed a blind ABX test switching between digital and analogue.

The chap who sold his amp and then regretted it is explained by the loss of the feel good factor and looks of an amp he obviously really liked.

The chap who regretted losing his amp bought it back (or another one). The amp hed compared it with in the blind test is the one he replaced it with and ever since then he knew it had been a misstake. He couldnt tell differences in the blind test, but he sure noticed differences over 'long term' listening

(Im going to find this again just to refresh my memory)

All im saying is that from what ive read on the net, blind tests have proven to be dodgy at best where hifi is concerned (Perhaps purely because its so difficult to truly recall what one sounds like to the other in the time scale given)

The Ivor guy you just spoke of also goes some way to suggesting that blind tests are perhaps not the best way to determine differences between hifi components

 

[Anonymous]

OK Rick, I'll bite. Point me to these 'dodgy' blind tests. I've never found any.

 

[b33k34]

aliEnRIK:
All im saying is that from what ive read on the net, blind tests have proven to be dodgy at best where hifi is concerned (Perhaps purely because its so difficult to truly recall what one sounds like to the other in the time scale given)

The Ivor guy you just spoke of also goes some way to suggesting that blind tests are perhaps not the best way to determine differences between hifi components

All a properly conducted blind test does is remove the knowledge of what you're listening to - the same as testing a sugar pill against an aspirin or a £5 wine against a £500 wine.

There is no reason that blind tests would 'prove to be dodgy' unless the differentiating factor is knowing which cable is which before you listen to it.

You're obviously convinced but please don't get into a position where you're ever making decisions about spending public money or affecting anyones health.

 

[aliEnRIK]

Tarquinh:OK Rick, I'll bite. Point me to these 'dodgy' blind tests. I've never found any.

Ive a few bookmarked at home I think

I'll see what I can find wednesday if I remember

I didnt really been dodgy in that sense. I ment more 'dubious' in that theres a few 'possible' reasons why results were so random

 

[aliEnRIK]

b33k34:aliEnRIK:

All im saying is that from what ive read on the net, blind tests have proven to be dodgy at best where hifi is concerned (Perhaps purely because its so difficult to truly recall what one sounds like to the other in the time scale given)

The Ivor guy you just spoke of also goes some way to suggesting that blind tests are perhaps not the best way to determine differences between hifi components

All a properly conducted blind test does is remove the knowledge of what you're listening to - the same as testing a sugar pill against an aspirin or a £5 wine against a £500 wine.

There is no reason that blind tests would 'prove to be dodgy' unless the differentiating factor is knowing which cable is which before you listen to it.

You're obviously convinced but please don't get into a position where you're ever making decisions about spending public money or affecting anyones health.

Thats just not the same though is it? Tasting wine is using your mouth for example. Also, im pretty sure ive seen at least one example of wine testing that also had random results. I dont know about you, but if I kept putting 2 different wines in my mouth im pretty sure id lose all sense of taste for a while and fine it almost impossible to discern differences between the 2

And a sugar pill against an aspirin? Well id taste sugar over aspirin for certain. But if you mean actually affecting the body in some way, then thats easy to test for. If you mean if id 'feel' changes, I dont believe id know about either personally.

But again, NOTHING to do with listening for differences are they?

 

[bigboss]

aliEnRIK: But again, NOTHING to do with listening for differences are they?

Well, you're not even believing the studies of listening for differences. The only study you seem to believe is the one sponsored by Russ Andrews. There's an obvious conflict of interest in that study. Can you imagine Russ Andrews sponsoring a study only to find that cables don't make a difference? Do you think Russ Andrews would ever publish such a study?

With regards to the Ben Duncan study, large RF signals coming down the mains at the levels they were testing at just doesn't ever happen and therefore isn't an issue. RF breakthrough is invariably caused by direct radiation and you know when you have it because you can hear it.

Russ Andrews commissioned the test solely to deflect the ASA and he succeeeded in that, because he could back his claims for RFI reduction in his mains leads, albeit a rather small reduction, but at what price? A Tacima is far cheaper.