Indeed our brains "cheat" - that's why we can listen to a piece of music we're familiar with, on audio gear that's atrocious, and enjoy it - the cheat in this case is our memory, it fills in the gaps, masks over the distortions. And why if we listen to a completely unknown piece of music, done in a style we're not familar with, on that same piece of equipment it sounds appalling - we have no hints sheet for what's meant to be there, how it's supposed to sound, our brains have to rely only on the auditory information being fed to it, and it's too messy to unscramble. Obviously, this where ABX tests can come unstuck - our memory revs up, and starts cheating for us ...
As Scott implied, how are you going to determine whether the replay equipment behaves differently or not when ultrasonic signals pass through the chain - an assumption, perhaps, that it won't ... ?
I don't see that this tells you anything as a positive control. Try switching between them and see if anyone thinks they hear a repeatable difference. if they do, let them try it again blind.
Scott's confounder may be a serious one, though.
Griesinger found it wasn't the speakers', rather the electronics' artifacts that got 'folded down'.
T
Hi, ok so I have a few ideas about sound quality versus measurement......
I figure this thread a good place to write it.
Please correct me anywhere here if I'm lost in the forest.
______
To start with, tone deafness.
I realise tone deafness actually is pitch deafness and this condition of singing-in-pitch-deafness, to be linguistically correct, afflicts as much as 1 in 10 people.
The reason is, they are missing a large chunk of white or grey matter. Some have more white than grey, different topic.
As it were, they are missing a cable or chip in their head. Much like dyslexia. Except it's not the lexicon chip missing, it's the pitch chip or singing-in-pitch chip.
This neuron chip is called "superior branch of the arcuate fasciculus".
Extrapolation, is there instrument deafness as well?
You see, that is pitch. That pitch deafness has been known for aeons, since it can be clearly perceived via listening to someone else singing.
______
Next...... is tone.
Tone, colour, tonal colour, voicing, timbre, character and similar terms.
This is what differentiates a flute from a clarinet both playing the same note at the same volume, for instance.
Note: Are there individuals which can not differentiate this and need the visual or expectational cue?
Side-note: Tone is not tonality, which is related to FR and mathematics in instrument design, which may be present in speaker drivers as well, but nevermind this for now since tonality is a different topic.
The way I see it, tone is not related to frequency response, pitch, playback speed, volume, volume speed, echoic / anechoic, reverb effects, imaging, spatial vastness of sound or anything else I'm missing.
Tone is......
- Unique ADSR envelope, which includes speed
- Unique distortion spectrum or noise spectrum
- Unique overtone
- Unique non-/sinusoidal waveforms ..?
- Anything else
Afaik, in 2014, tone is still not replicated perfectly in any software, hardware or synthesizers.
The human voice sounds very synthetic still as well.
Now, this tone or likewise the lack thereof, is a lethal part of an audio system.
For instance......
driver X vs driver Y (Titanium vs paper etc.)
flute X vs flute Y
piano X vs piano Y
DAC chip X versus DAC chip Y
capacitor X versus capacitor Y
op-amp X versus op-amp Y
discrete X versus discrete Y
Now, I'm not sure about capacitors, but the Cyril Bateman papers indicate they have unique distortion spikes, so, I assume that's why they can sound unique and differ in sonic transparency.
I can only assume DAC chips have unique distortion spike / noise spike differences as well, which are not visible in the spec sheets.
Where is the different slew rate and settling times in the spec sheets?
DAC chips have different reconstruction filters and phase filters as well.
Then there's magnetic versus non-magnetic resistors and something called electro-magnetic interference rejection ratio, EMIRR, which could be present in these other components as well.
Then PSRR.
Then there's cables and their conductivity and shielding and such, which is a different topic.
Anything else I'm missing.
______
In summary, I think the internet is obsessed with FLAC and jitter, which I can't hear......
Instead of tone, which at times I think I can clearly hear.
Alright.
Hi, ok so I have a few ideas about sound quality versus measurement......
I figure this thread a good place to write it.
Please correct me anywhere here if I'm lost in the forest.
______
To start with, tone deafness.
I realise tone deafness actually is pitch deafness and this condition of singing-in-pitch-deafness, to be linguistically correct, afflicts as much as 1 in 10 people.
The reason is, they are missing a large chunk of white or grey matter. Some have more white than grey, different topic.
As it were, they are missing a cable or chip in their head. Much like dyslexia. Except it's not the lexicon chip missing, it's the pitch chip or singing-in-pitch chip.
This neuron chip is called "superior branch of the arcuate fasciculus".
Extrapolation, is there instrument deafness as well?
You see, that is pitch. That pitch deafness has been known for aeons, since it can be clearly perceived via listening to someone else singing.
______
Next...... is tone.
Tone, colour, tonal colour, voicing, timbre, character and similar terms.
This is what differentiates a flute from a clarinet both playing the same note at the same volume, for instance.
Note: Are there individuals which can not differentiate this and need the visual or expectational cue?
Side-note: Tone is not tonality, which is related to FR and mathematics in instrument design, which may be present in speaker drivers as well, but nevermind this for now since tonality is a different topic.
The way I see it, tone is not related to frequency response, pitch, playback speed, volume, volume speed, echoic / anechoic, reverb effects, imaging, spatial vastness of sound or anything else I'm missing.
Tone is......
- Unique ADSR envelope, which includes speed
- Unique distortion spectrum or noise spectrum
- Unique overtone
- Unique non-/sinusoidal waveforms ..?
- Anything else
Afaik, in 2014, tone is still not replicated perfectly in any software, hardware or synthesizers.
The human voice sounds very synthetic still as well.
Now, this tone or likewise the lack thereof, is a lethal part of an audio system.
For instance......
driver X vs driver Y (Titanium vs paper etc.)
flute X vs flute Y
piano X vs piano Y
DAC chip X versus DAC chip Y
capacitor X versus capacitor Y
op-amp X versus op-amp Y
discrete X versus discrete Y
Now, I'm not sure about capacitors, but the Cyril Bateman papers indicate they have unique distortion spikes, so, I assume that's why they can sound unique and differ in sonic transparency.
I can only assume DAC chips have unique distortion spike / noise spike differences as well, which are not visible in the spec sheets.
Where is the different slew rate and settling times in the spec sheets?
DAC chips have different reconstruction filters and phase filters as well.
Then there's magnetic versus non-magnetic resistors and something called electro-magnetic interference rejection ratio, EMIRR, which could be present in these other components as well.
Then PSRR.
Then there's cables and their conductivity and shielding and such, which is a different topic.
Anything else I'm missing.
______
In summary, I think the internet is obsessed with FLAC and jitter, which I can't hear......
Instead of tone, which at times I think I can clearly hear.
Alright.
Last edited:
If there's anything I have learnt from my headphone amps, it's that headphones can sound way better than most will ever be aware of once they are driven by dedicated electronics cooked up for them.
Even models you would never suspect of being that capable.
If it's detail you want to hear, headphones are the way to go. However, if you want musical realism, speakers are better at simulating a concert hall, they give you music all around you.
This shouldn't surprise anyone who has taken a peek at how headphone are deriven from at least 90% of the electronics. Typically via a resistor voltage divider from the power amp. Drive them directly from a dedicated amp and they start to come alive like you wouldn't believe.
I read in an interview with Dan d'Agoastino (Mr Krell) that he designs the sound of his amps by listening to them via a special pair of headphones, which has been tweaked and tailor made to his hearing. He says that's the only reliable way to hear is there's any dirt, grime and noise left in the amp which is still audible.
...No that's just women.
Ah, if it only were so.
Actually that may be the mechanism by which one can hear 'above 20 kHz sound'. We did an earlier experiment with an FM mux filter that cut off everything above 15 kHz and could be reliably identified. Although the difference was easier to hear with HD800's rather than a speaker, so that goes against that suggestion.
Jan
Last edited:
Economy in lying is to do it when absolutely necessary. Jesus was very evasive which comes close. I wasn't there so go on the evidence of others. I don't know enough of Buddha to say if he also? For my safety I dare not go further. If you follow my train of thought the unmentionable one is almost by definition of what happened before a lie or at best superfluous. Hi fi often seems to suffer from the traits of religion. Old habits die hard. I heard recently that kindliness is the only mortality and all other good things can come from it. I have even seen kindliness to be the thing that makes people financially richer proved by set theory. That's why communism failed, it was far from kind. I guess the Scandinavian countries get it mostly right?
SACD, CD, MP3. As I said before I like MP3. It often has been recorded from CD. I doubt it is a simple filtering effect. When the radio version the FM via Nicam sound is marginally better despite having been through FM processing compared with MP3 like process for radio via TV. On the whole BBC radio is OK. My friend John says quoting his old boss David Mate that the brain is very good at processing wrong sound. Alas it is fatiguing even when seemingly perfect. Davids main thought being timing errors. He also believed speakers must attempt to reproduce square waves if to be minimally fatiguing. Douglas Self wrote to me saying would we not get vision fatigue more easily if either exist? After some thought I suggested that the visual world is not as different as the sound world from when we were forest people. Also the vision world has rather better equipment than the ears.
I listened to the master tape of Led Zeppelin Stairway to heaven via itself, 33 1/3, 78 and SACD ( Sony ). The SACD sounded like a tribute band, voice quality not similar. 78 was a very good AB with 33 1/3 not far off ( 45 no better than 33). I was totally surprised because I didn't think Vinyl could get that close. I have 2 expensive volumes of the AES on the subject. It is as near an impossible concept as you could care to dream up. I was listening to 1927 Bob Crosby and 1929 Bing Crosby last night. Already in the 2 years the cutting was getting to be quite good , 1938 is in some ways the best it would be for many years. Saxophones arguably better than modern recordings. Sometimes a 78 can be pressed from good quality stampers on to vinyl. 15 kHz is not unknown, 32 kHz when FFRR.
Doubtless the SACD player I heard was not the best. All the same the pick up we were using was a cheaper Shure. Shure pick ups don't suit 47 K very well. They like higher ( 1 M even ) with a tweak of 75 uS to do some restoration. 75 uS is best when passive. This also allows EQ tweaks as most records are rarely exactly 75 uS. The other parameters are complex ( 3180/318 and variants). The treble is easy. Passive EQ also helps ease the slewing issues down the chain. If you have active EQ it might be worth fitting 2 uS passive output filter. Scratches often are elongated if active EQ used. A video op amp using double inverting inputs might offer something. I would make Rin 25 R and Rg 1K6 with whatever gives 75 uS ( 47 nF ) active. The next stage whatever op amp you fancy. Denon 103 or 103R. MC loading is rarely about the pick up unlike MM . It is mostly to make the pre amp work. If a transformer is 100 : 1 and feeding 47 K that will be lower.
One digital designer uses a Garrard 401 as a reliable repeatable source. If the digital system sounds different to it he knows it didn't work.
SACD, CD, MP3. As I said before I like MP3. It often has been recorded from CD. I doubt it is a simple filtering effect. When the radio version the FM via Nicam sound is marginally better despite having been through FM processing compared with MP3 like process for radio via TV. On the whole BBC radio is OK. My friend John says quoting his old boss David Mate that the brain is very good at processing wrong sound. Alas it is fatiguing even when seemingly perfect. Davids main thought being timing errors. He also believed speakers must attempt to reproduce square waves if to be minimally fatiguing. Douglas Self wrote to me saying would we not get vision fatigue more easily if either exist? After some thought I suggested that the visual world is not as different as the sound world from when we were forest people. Also the vision world has rather better equipment than the ears.
I listened to the master tape of Led Zeppelin Stairway to heaven via itself, 33 1/3, 78 and SACD ( Sony ). The SACD sounded like a tribute band, voice quality not similar. 78 was a very good AB with 33 1/3 not far off ( 45 no better than 33). I was totally surprised because I didn't think Vinyl could get that close. I have 2 expensive volumes of the AES on the subject. It is as near an impossible concept as you could care to dream up. I was listening to 1927 Bob Crosby and 1929 Bing Crosby last night. Already in the 2 years the cutting was getting to be quite good , 1938 is in some ways the best it would be for many years. Saxophones arguably better than modern recordings. Sometimes a 78 can be pressed from good quality stampers on to vinyl. 15 kHz is not unknown, 32 kHz when FFRR.
Doubtless the SACD player I heard was not the best. All the same the pick up we were using was a cheaper Shure. Shure pick ups don't suit 47 K very well. They like higher ( 1 M even ) with a tweak of 75 uS to do some restoration. 75 uS is best when passive. This also allows EQ tweaks as most records are rarely exactly 75 uS. The other parameters are complex ( 3180/318 and variants). The treble is easy. Passive EQ also helps ease the slewing issues down the chain. If you have active EQ it might be worth fitting 2 uS passive output filter. Scratches often are elongated if active EQ used. A video op amp using double inverting inputs might offer something. I would make Rin 25 R and Rg 1K6 with whatever gives 75 uS ( 47 nF ) active. The next stage whatever op amp you fancy. Denon 103 or 103R. MC loading is rarely about the pick up unlike MM . It is mostly to make the pre amp work. If a transformer is 100 : 1 and feeding 47 K that will be lower.
One digital designer uses a Garrard 401 as a reliable repeatable source. If the digital system sounds different to it he knows it didn't work.
FAL Super-Tweeter. I have heard them, also as headphone made by a Swiss company who claim to supply FAL. Amplifier was 2A3. Very close to being a real sound and very loud in the positive sense. No obvious distortion. A Krell and Martin Logan speakers was totally different and to me nothing like real music. A real disappointment. I am here and the hi fi is there. The FAL, I am here and the room is alive with music. I would guess the latter poorly set up. All the same I was allowing for that. mostly when I have heard the Krell I have been left unmoved. I have always thought the buyer did not understand the amp and had more money than ability to create a system. The dealer who sold it should be told off.
John heard my Magneplanars last night. He is going to lend me Crown DC 300. Old prototype Crimson amp made for his old company by Crimson and C-Audio. I will do DC checks as those amps have had a rough life. His friend Martyn a Quad 405. The C-Audio is the one I want to try. I should build up a Hitachi clone. I dare say a Krell would be fun. I heard the DC 300 the other week with Musical Fidelity The Preamp. It was ultra charming if a little rounded. I've used it's big brother to test body scanner gradient coils ( 3 phase powered ).
FAL Speaker
John heard my Magneplanars last night. He is going to lend me Crown DC 300. Old prototype Crimson amp made for his old company by Crimson and C-Audio. I will do DC checks as those amps have had a rough life. His friend Martyn a Quad 405. The C-Audio is the one I want to try. I should build up a Hitachi clone. I dare say a Krell would be fun. I heard the DC 300 the other week with Musical Fidelity The Preamp. It was ultra charming if a little rounded. I've used it's big brother to test body scanner gradient coils ( 3 phase powered ).
FAL Speaker
I am trying to build a headphone amp. I find the presentation of stereo via headphones the most unnatural sound possible. Even proper recordings not very good.
As yet some cheap Sony ear buds almost seem best. I have Stax to compare.
What I have considered is a simple complementary feedback single pair and gain stage. The amplifier would have no loop feedback. If I do use loop feedback 3 transistors would give me about 0.05% THD ( 0.013% from the output part ). I tried an op amp design and found it unpleasant.
I tried a stereo blend circuit. Not my cup of tea.
As yet some cheap Sony ear buds almost seem best. I have Stax to compare.
What I have considered is a simple complementary feedback single pair and gain stage. The amplifier would have no loop feedback. If I do use loop feedback 3 transistors would give me about 0.05% THD ( 0.013% from the output part ). I tried an op amp design and found it unpleasant.
I tried a stereo blend circuit. Not my cup of tea.
I wonder which pair.
I'm not very familiar with Krell, but I've tried the overpriced Bakoon headphone amplifier in Current-out mode.
I wonder if there are Current-out speaker amplifiers?
For portable audio, I think iBasso are practical.
The discussion of them is at head-fi or at Erji.net.
Those forums are only advertising and discount reviews though, like a magazine, so you have to read with a kilo of salt.
Last edited:
Sony headphones might be MDREX15LPBX6.YG
They improve with use and are very comfortable. For some reason the presentation is less confusing which makes up for mid fi quality. Certainly not lo fi. The Stax have their moments, somehow me and them don't get on.
Beyer DT 990 is OK. Not the studio version.
Krell amps was nothing to do with the amps. They seem to attract clueless customers and dealers who somehow do not have the honesty to resist the money.
I remember years ago a young lad said his grandfather left him some Quad ESL 57. I told him to send them to Quad to be serviced. The cost was quite high. His challenge to me was to put him a system together for as little money as possible that would do them justice. We found a secondhand Naim Nait 2 and new Rega Planar 3 Goldring PU. The Naim works with the Quads as it was a reference speaker for Naim. None of the typical Naim defects are heard. I would say the difference compared with NAP 250 not vast. That system made me very jealous.
They improve with use and are very comfortable. For some reason the presentation is less confusing which makes up for mid fi quality. Certainly not lo fi. The Stax have their moments, somehow me and them don't get on.
Beyer DT 990 is OK. Not the studio version.
Krell amps was nothing to do with the amps. They seem to attract clueless customers and dealers who somehow do not have the honesty to resist the money.
I remember years ago a young lad said his grandfather left him some Quad ESL 57. I told him to send them to Quad to be serviced. The cost was quite high. His challenge to me was to put him a system together for as little money as possible that would do them justice. We found a secondhand Naim Nait 2 and new Rega Planar 3 Goldring PU. The Naim works with the Quads as it was a reference speaker for Naim. None of the typical Naim defects are heard. I would say the difference compared with NAP 250 not vast. That system made me very jealous.
Portable Headphone Amplifier - RED - Page31
3 transistor headphone amp.
I was rather taken by this idea. Not least it works with 3.7V. The bootstrap to make a CCS I like. I will have to build it to see what the DC offset is. I suspect being that it doesn't send feedback to Tr1 emitter is quite a nice thing. There will be no crossover distortion and battery life is OK. Mini JLH cum HC Lin.
3 transistor headphone amp.
I was rather taken by this idea. Not least it works with 3.7V. The bootstrap to make a CCS I like. I will have to build it to see what the DC offset is. I suspect being that it doesn't send feedback to Tr1 emitter is quite a nice thing. There will be no crossover distortion and battery life is OK. Mini JLH cum HC Lin.
Last edited:
- Status
- Not open for further replies.