Sound Quality Vs. Measurements

[VladimirK]

Originally Posted by Jay

We haven't discussed the other links such as the specs of the DAC chip (if one has NE5534 in the chain, he may miss so many opportunities ), and the speaker sub-system that has so many "non-scientific/voodoo" contents.

There is nothing perfect in this world, but with a system that I have put together, I may say, that I estimate quality of a CD recording. Any imperfection of a recording is listenable. I tend to listen mainly HDCD, XRCD, audiophile quality CD, etc. Maybe it is not good to have a CD testing facility at home, but that is a result of competition with high-resolution tube amps.
As for amp bandwidth itself, if I put RC chain at the input of amp (well above 100kHz cut-off), it affects sound substantially, irrespective of what theory says.
System components are: PS Audio Power Plant Premier, Furutech power cables 1,5m, tuned Shanling CD-T1500 player (EH6922 were replaced by matched NOS Philips Miniwatt SQ tubes, pairs selected from 10 tubes, sound became more "brilliant" and better articulated at bass, 6 pcs of tubes are remaining without use), Analysis Plus Silver Oval In 1m interconnects (bulk silver), self-made integrated amp, Stereovox 600 (top model) 2,5m speaker cables (bulk silver), PMC EB1i speakers, listening room is 36m2. Speakers resolve perfectly all mid and high frequencies, bass articulation depends on amp's topology. If topology is suitable, even 3W amp produces wery pleasant and articulated bass.

[a.wayne]

Originally Posted by dvv

Ooooooohhhhh, Pivetta! 20,000 Watts!

On a good day, even our own Wayne might be satisfied - for about an hour.

Ahhh, 1 kw/ch will suffice , makes for good bass , tonal balance , instrument size and dynamics and I love betting for dinner also , easiest bets are with politics and F1 ...

Politics because most still believe there is an option and F1 because it is as corrupt as politics but with speed , around here we call it WWF1...

[DF96]

Originally Posted by dvv

As for the NE op amps, careful there, they a religion in the UK audio business, and those who speak against deities tend to get burned at the stake.

I thought speaking against the NE5534/32 was the true religion. Now you have confused me. Should I use these chips everywhere, or nowhere? Or should I make engineering judgements based on facts, and ignore the 'fashion engineers' on both sides?

[vacuphile]

Originally Posted by DF96

I thought speaking against the NE5534/32 was the true religion. Now you have confused me. Should I use these chips everywhere, or nowhere? Or should I make engineering judgements based on facts, and ignore the 'fashion engineers' on both sides?

The time is ripe for fact free engineering.

[simon7000]

Originally Posted by scott wurcer

I thought I posted that link, they used it to measure the noise on batteries IIRC. In any case the auto-correlation amplifier is an old concept and I still don't see the use of quantifying distortion at these levels.

You very well could have. I came across that one when I was first using batteries in a test instrument. That was about 2007. There are some other examples. I find it gets about 20 db more range.

One issue I find at hand is the influence of out of band noise. Distortion masked by noise is interesting down to about 30 db below the noise. That should not be very difficult to measure.

What is more interesting is the ADSR envelope modulation effects.

The remaining issue is the comparison of amplifiers under load to unloaded.

[PChi]

It's possible to hear signals below the noise floor but it depends on the circumstances. With GSM phones the TDMA buzz / bumble bee noise caused by demodulation of the transmitter output is audible even though it can be below the audio band limited noise floor but is visible with an FFT plot but the noise floor is high. Typically in a car at speed on the motorway typical speech levels are similar to the background noise level but again the noise level is high.

I don't think it's worth obsessing about the last minute level of distortion when it is swamped by the non linearity of the loudspeaker. Though there is no harm in it as long as it is realised that it is going to have no effect on the sound quality / reproduction accuracy.

[nigel pearson]

Last night there was a question about significant hi fi . The only Japanese amp mentioned was Kondo . Kondo is one of many specialist amps in Japan . The man I list below doesn't particularly claim to be an amplifier designer , yet his 2A3 amp ranks as high as any I know ( a Monty Python joke in there somewhere , Life of Brian ) . His speakers are truly remarkable . They have strange measurements . They retrieve detail like no other I know .
The Kondo is said to be based on the PYE Motzart . It doesn't share many of it's features .
FAL Speaker

BTW . If thinking this is some very polite type of sound think again . It would work for home cinema even with his 2A3's . FAL standard demo is Japanese percussion piano . Also European classical . Meet this man if you ever get the chance .

One Germany guy said to me not worth hearing as he had seen the specs ( graphs ) .

日本の打楽器のピアノ


http://www.freepatentsonline.com/6895097.html

[vacuphile]

Actually, noise is the stuff of magic, even apart from dithering. Addition of noise can actually help to raise a signal that is below the threshold of detectability to above it. Your own auditory receptor may very well for that reason be the noisy system that it is. Therefore, what happens below the noise floor can be very relevant; the fact that there is noise may even contribute to the perception of what is going on at very low levels.

vac

[bcarso]

Originally Posted by vacuphile

Actually, noise is the stuff of magic, even apart from dithering. Addition of noise can actually help to raise a signal that is below the threshold of detectability to above it. Your own auditory receptor may very well for that reason be the noisy system that it is. Therefore, what happens below the noise floor can be very relevant; the fact that there is noise may even contribute to the perception of what is going on at very low levels.

vac

A not-too-bad short discussion: Stochastic resonance - Wikipedia, the free encyclopedia

[a.wayne]

Originally Posted by PChi

It's possible to hear signals below the noise floor but it depends on the circumstances. With GSM phones the TDMA buzz / bumble bee noise caused by demodulation of the transmitter output is audible even though it can be below the audio band limited noise floor but is visible with an FFT plot but the noise floor is high. Typically in a car at speed on the motorway typical speech levels are similar to the background noise level but again the noise level is high.

I don't think it's worth obsessing about the last minute level of distortion when it is swamped by the non linearity of the loudspeaker. Though there is no harm in it as long as it is realised that it is going to have no effect on the sound quality / reproduction accuracy.

Practical has shown this to not be true , such noise is percieved. , not as distortion , more as a masking of detail ....