Cyclic debates and moderation Part II

[cooky1257]

I'm curious why it's taken us till post 100 to get the 0.05% threshold, maybe I missed it earlier, maybe Serge was challenged over his 0.1% number.
If the stock Berry has 0.08% are we saying a reduction of 0.03% would've made it transparent? I'm not up to mischief here, maybe the magic occurred much lower looking at Simons figure of .0005%, the benefits of the cleaner power supply are easy to see on the graphs too, I'm in no position to test stuff but do wonder if it is a combination of spec/measurements/improvements where the transparency threshold lies and not individual figures.

 

[pure sound]

Then my definition of transparency is different, as in my own blind A/B listening tests I've found the standard Behringer to have slightly audible distortion. Perhaps speakers have improved since the ones used for the tests? The most clear listening test for me was done at high volume levels with an active PMC MB2system.

This made me smile. What levels of distortion do you suppose the speakers were producing through which you could hear that 0.08%?

 

[sergeauckland]

Hmm just as an experiment I tried a 9KHz and 10KHz sine with a 1KHz sine at -60dB (0.1%) and I can hear it on headphones.

This is a similar result as IMD difference tones.

I can't hear it as a 2nd harmonic product.

My understanding has always been that the 0.1% figure was with regards to THD, not all types of distortion. Serge, can you find the studies you were taught from back then?
P.S. Dpes anyone else hate Adobe Audition? I used to use CoolEdit, but had to switch to Audition for Win7.

No, I wish I could. However, the 0.1% figure was always using programme material, i.e. speech or music. Tones are far more sensitive as there's no masking of the distortion products.

As we know, masking makes psycho-acoustic codecs like MP3 work, and is very powerful. If one uses tones with distortion added, I would expect that firstly, th 0.1% figure would need to be considerably tightened, but as what to, it would depend on the frequency, as our hearing is far more sensitive at mid frequencies, peaking at around 5kHz or thereabouts, and on level as our hearing's frequnecy response flattens out with increasing level.

I suggest that using speech and music is a valid limitation, as the audibility of an amplifier's distortion is relevant to its use playing music, unless the amplifier is being used for some instrumentation purpose.

S.

 

[sergeauckland]

What is your view of the audibility, and significance, of TIM or SID?

I don't know that these mechanisms actually exist in modern amplifiers, if they ever did.

TID / SID only exist in very poor designs where the amplifier's output is unable to follow the input signal due to excessive level and frequency. If the amplifier's frequency response is properly tailored to the amplifier's slew rate, that can't happen. In the past when slew rates were limited, it was possible for a poor design to exhibit SID, now, slew rates are so far ahead of what's needed for audio.

S.

 

[Tenson]

This made me smile. What levels of distortion do you suppose the speakers were producing through which you could hear that 0.08%?

Indeed, interesting isn't it. I've always guessed this might be due to one type of distortion being more objectionable than another. The active MB2 is a pretty low distortion speaker though

 

[pure sound]

Far enough, I will remove it. The truth is I think the above improvements are true, but subtle.

Wouldn't it be fairly easy to record the analogue outputs of a modified & un-modified unit, upload links of the recordings to your forum & then see whether a significant number of people can reliably discriminate between them by means of an ABX test?

 

[Yank]

The more I read this thread the more I think that I personally am not drawn to the piece of wire with gain sound.

Hard to judge, since in my view this is something that none of us have ever heard.

 

[Yank]

I don't know that these mechanisms actually exist in modern amplifiers, if they ever did.

TID / SID only exist in very poor designs where the amplifier's output is unable to follow the input signal due to excessive level and frequency. If the amplifier's frequency response is properly tailored to the amplifier's slew rate, that can't happen. In the past when slew rates were limited, it was possible for a poor design to exhibit SID, now, slew rates are so far ahead of what's needed for audio.

S.

What kind of time frame do you mean by "in the past"? I can think of some quite early transistor amplifier designs that are (1) still in production, and (B) have been listed by you as probably being transparent and indistinguishable from the competition.

 

[Robert]

This made me smile. What levels of distortion do you suppose the speakers were producing through which you could hear that 0.08%?

Guy the answer is to use those things that you don't particularly care for, ESLs

My regular listening loudspeaker is the ESL63 and that does have distortion comparable to clean electronics in the mid and top at least. It's pretty unique in that respect by some considerable margin.

But I think distortions can be cumulative. The age old listening sessions that settled on the original safe 0.1% were done throgh loudspeakers with distortion comparable to what is typical today. They were detecting changes down to that figure certainly it seems.

 

[sergeauckland]

What kind of time frame do you mean by "in the past"? I can think of some quite early transistor amplifier designs that are (1) still in production, and (B) have been listed by you as probably being transparent and indistinguishable from the competition.

Amplifiers of the late '60s, especially those with germanium output transistors.

However, even a germanium transistored amplifier, if it can output 20kHz at full output with low distortion, can't suffer from TIM/SID provided it's not asked to provide more than 20kHz at high powers. Slew-rate is V/uS, in other words, how many volts in how much time. A very high frequency can be OK if the Volts aren't too high, and the volts are fine up to clipping if the frequency isn't more than 20kHz.

It seems to me that when SS first came out, and the output transformer with all its limitations was eliminated, some designers forgot about loop gain and phase shift. This very effectively limited the amount of feedback which could be applied, and required the open-loop bandwidth to be correctly tailored. With SS amps, this wasn't anywhere near as onerous, and I think that in the excitement of being able to have large amounts of feedback, and wide bandwidths, the earlier lessons of stability margins was forgotten.

Nevertheless, TIM/SID is very easily avoided with one capacitor at the input!

S.

 

[Robert]

I think that striving for an industry standard for use of a word like transparent is a very positive thing. It may well be that all ‘transparent’ amps do in fact sound nigh on identical; if this is the case then the purchaser can look at its features rather than benefits. Does it have enough inputs, remotes, whatever?

The more I read this thread the more I think that I personally am not drawn to the piece of wire with gain sound. Of course many will say, but there is no ‘sound’ this is what is on the recording, and this may well be the case.

I am now 55 years old and have been surrounded by music and musicians for nearly all my life. I know how music can move me and I know what I like to hear. I am more than happy to understand that my preference in musical reproduction involves some sort of built in distortion. I would personally view this distortion as musicality and emotion.

Again I know that the science lead enthusiast here will say that hi-fi can’t add musicality and emotion and again I agree with this. What I personally feel is that some ‘transparent’ hi-fi actually strips and bleaches out the latent musicality and emotion that is within the recording.

A friend of mine who works in the hi-fi industry has often said that it is pretty easy to make things read better but that will often lead to the product sounding worse. He believes, as I do, that there is a skill set beyond the science to unlock the musicality and emotion that is latent within recordings.

The only way that I can describe these latent emotions is that there are a couple of recordings that if the system is set up correctly and using the right equipment will actually make me cry and make the hairs stand up on my neck. On other equipment there is not the same access to the emotion that is latent. It's like I'm hearing it all but I'm not feeling it all.

I’m quite prepared to believe that all the equipment that I have heard these recordings on would be transparent by today’s standards, yet they deliver the soul of the music in different ways.

I can prove none of the above but I know what I know. This is obviously a personal viewpoint but I know that I am not alone.

You make some very fair points.
But I don't see the conflict between trying to achieve the transparency that you mention, plus suggesting ways to reduce bias on the one hand, and enjoying whatever you want on the other. Nobody is saying anyone has to like the idea of 'straight wire with gain', but it is the goal of many manufacturers to produce electronics that preserve the source quality wholly intact as their priority.

I noted your post recently about hearing the benefit of adding a Hicap to a pre amplifier and hearing the sound improve. To suggests that you, for example match levels and aren't visually aware of when the unit is operating doesn't in any way conflict with enjoying music and it isn't telling you what to like or that some notion of transparency cooked up on this forum is the only way.

Look at it a different way.
It is effectively a second chance check, a means of saying 'are you absolutely sure'.

 

[Brian]

This is my last post on this forum.

I hoped this would be a place where folk could say how their kit sounds and ask others for advice on the sound upgrades would bring but thread after thread just gets crapped on with it all sounds the same and you couldn't tell blind.

I'm sorry but this place will turn into zero gain rapidly and i'm gone.

Look,

You made a claim that resistive attenuators sound different. That's an outrageous claim that defies all logic. I pulled you up on that, and now you're flouncing off! Think about what you said for a moment. What possible mechanism could there be for a resistive attenuator to have any sort of sound. It's a resistor! It's linear! It doesn't have a sound.

S.

Thought I'd mention that while I can't make out what a Naim Hicap does when added to a Naim pre/power amp combination, I've heard clear differences between volume controls.

Using my lo-fi amplifier I've tried and found differences between :

Alps Blue

a TKD

and a DACT Stepped Attenuator

The passive TVC is better again...

 

[TheDecameron]

Brian, stop teasing, which was best? Received wisdom would say the dact...

 

[cooky1257]

"Just a moment...just a moment...I've just picked up a fault in the AE-35 unit. It's going to go a hundred percent failure within 72 hours. "

"Yes, and it will stay that way until it fails. "

"Yes, that's a completely reliable figure. "

"Yes. It's puzzling. I don't think I've ever seen anything quite like this before. I would recommend that we put the unit back in operation and let it fail. It should then be a simple matter to track down the cause. We can certainly afford to be out of communication for the short time it will take to replace it. "

"I hope the two of you are not concerned about this. "

"Are you quite sure? "

"Well, I don't think there is any question about it. It can only be attributable to human error. This sort of thing has cropped up before, and it has always been due to human error. "

 

[srh]

Serge, with regard to slew rate, what do you consider to be an acceptable value for, say a 100W/ 8R SS amplifier?
Regards,
Steve.

 

[Brian]

Brian, stop teasing, which was best? Received wisdom would say the dact...

In my opinion the TVC is best, it uses Silk transformers though and is not cheap even as a kit. The DACT is a very close 2nd, the TKD is 3rd and the Alps Blue is 4th. I found some nice comment about the TVC here

However, the best value for money is the DACT imo.

 

[Sue Pertwee-Tyr]

We seem to have reached a point where we might adopt a dictum from a certain Mr Blair*

"all amplifiers are transparent, but some are more transparent than others".

We seem to have moved away from Serge's claim that once an amplifier reaches his (50 year-old) threshold of 0.1% THD it is (to all intents and purposes) transparent, to one in which 0.05% might be transparent, but 0.08% probably isn't. So what we're arguing about now is where exactly the thresholds of audibility lie?

This, it seems to me, is the position adopted by many on here, who argue (repeatedly) that they hear differences which don't correlate to the measurements. Or if, as seems more likely, they do correlate to measurement, the counter-claim is that the difference is inaudible. We now seem to have a credible challenge to that counter-claim, at least in some respects.

Might we hope for a reduction in overall TUD (Totally Unhelpful Dogma) as a result, do you think?


*not that one.

 

[Tony L]

Might we hope for a reduction in overall TUD (Totally Unhelpful Dogma) as a result, do you think?

Needs to be at 0.1% or less IMO.

 

[Fox]

Some of you guys need to get laid. Seriously. Really badly.

 

[Pete the Feet]

It's all quite tragic.