384khz?

[the_doc735]

Comes back to price/performance issue. Expenditure is always additive: a system with components costing 2k, 3k, and 4k costs 9k. Performance is not necessarily additive: a poorly matched system will perform less than the sum of its parts and vice versa. Performance/price ratios are rapidly subject to diminishing marginal returns so that a 10k system will not be twice as good as a 5k system (subjective measurement issues aside). I guess for most of us, the goal is to define our performance expectations and then decide how much we are willing/able to spend in achieving that level of performance. The real art is finding that combination of components that best meets those performance goals at a particular price point.

My primary (i.e. streaming) system costs significantly less than £5k. I think it would be difficult for me to improve upon this system within a 5k budget. I suspect that I would have to spend significantly more than 5k to achieve worthwhile improvements and for me that would not represent value for money.

All highly personal and subjective. As it should be!

Geoff

"Performance/price ratios are rapidly subject to diminishing marginal returns so that a 10k system will not be twice as good as a 5k system", my point entirely Geoff. I agree with your entire statement though! Maybe I should go the other way and try some S.M.S.L?

[Sherwood]

"Performance/price ratios are rapidly subject to diminishing marginal returns so that a 10k system will not be twice as good as a 5k system", my point entirely Geoff. I agree with your entire statement though! Maybe I should go the other way and try some S.M.S.L?

I would go so far as to say the key challenge in the pursuit of audio nirvana today is to find the most cost-effective amp/speaker combo, at least for digital/streaming systems. The performance of even modestly priced streaming systems (e.g. an RPi with add-on audio boards at less than £100) is now so good that one needs to spend considerably more to get substantive improvements. The challenge for analogue systems is more complicated as good engineering cannot be achieved at low cost. I think many people start off with a short list of speakers or speaker types and then try to find an amp that is a good match. This was how I arrived at my current digital system after a little trial and error. In hindsight it might have saved me time and money if I had looked at the speaker/amp purchase with a greater concern for synergies from the outset.

Geoff

[the_doc735]

I'd love to compare say: The Chord Dave to a SMSL M9 DAC XMOS Optical Coaxial USB Asynchronous 768KHZ/32Bit DSD. Anyone got a dave? (to compare). This whole '384' business seems a bit strange when most audio files are 44.1/16 bit anyway! Even the Graphice EQ studio software only allows U to select 44/48, and it appears non of my gear runs at 32 bit, only 24! There is wasapi/asio DS for foobar but that won;t run through the EQ though.

Then there's the VB Cable software that allows the EQ Studio to work; that always seems to output at 44/16 as well (even though it will input at far higher ratio's)? Yes, the synergies should be a major consideration I think. My old NAD 3020 was excellent for £50! but the speakers weren't that good really.

People DON'T want to hear me or anyone else say: "The SMSL M9 is as good as the Chord Hugo/2Qute for example". If you base your buying on financial grounds i.e. more expensive MUST BE better - right? "BUT!" ~ the chord is GB manufactured whereas the SMSL is Chinese manufactured; the trade off is not in quality but in manufacturing costs! If the Chord was manufactured in China I dare say it would only cost a third of what it does in the UK? If you are a 'badge snob' then you have won the bragging rights with the Chord, it's the old BMW/Kia thing all over again. I may take the SMSL challenge, it's certainly cheap enough, be prepared for a shock verdict if I do!

[Rothchild]

Infinite sampling rate is really what you need - now what technology can do that, Analogue?

Interesting aside here (further to my previous comments on RIAA curves) is that in practical terms it seems that the Redbook spec was aligned with 'standard' vinyl production. I've started reading on RIAA (in the hope of building a phono pre with my engineer father) and found this:

The RIAA curve can also be broken into two sub-curves, which when one cascades into the other will define the complete RIAA curve. The RIAA curve nicely breaks into a shelving network (50Hz and 500Hz) and a low-pass filter (2122Hz). The two circuits below embody the desired functions.

http://www.beigebag.com/case_riaa_1.htm#riaa5

LPF at ~20kHz sounds somewhat familiar, right?

[JimC]

Interesting aside here (further to my previous comments on RIAA curves) is that in practical terms it seems that the Redbook spec was aligned with 'standard' vinyl production. I've started reading on RIAA (in the hope of building a phono pre with my engineer father) and found this:

The RIAA curve can also be broken into two sub-curves, which when one cascades into the other will define the complete RIAA curve. The RIAA curve nicely breaks into a shelving network (50Hz and 500Hz) and a low-pass filter (2122Hz). The two circuits below embody the desired functions.

http://www.beigebag.com/case_riaa_1.htm#riaa5

LPF at ~20kHz sounds somewhat familiar, right?

2122Hz is 2k (2.122k to be exact)though, not 20k.....?
Or am I missing your point?

Jim

[Rothchild]

2122Hz is 2k (2.122k to be exact)though, not 20k.....?
Or am I missing your point?

Jim

No, you're right. Sorry, a long and arduous weekend has caused me to add a digit that wasn't there! Off to bed now for me.

[m10]

Fascinating thread. Not very technical here (I feel I understand, but probably don't really). I have been reading interesting things about temporal resolution as a justification for higher sampling rates, and all the Bob Stuart/MQA stuff around pre-ringing in digital systems (caused by that steep roll-off) and how that can be tackled.

Regardless of the science - or whether it's just down to nice (re)mastering - I have been having a nice time with MQA with Tidal/Roon and a Meridian Explorer 2. I'm so pleased that the Explorer 2 has replaced my Cambridge Audio Stream Magic 6 permanently. Not for MQA, though - but for the way it has had me rediscovering my mostly CD-derived collection (with a smattering of MP3). The MQA stuff really is nice - the Maria Callas remasters, several lovely Karajan EMI albums and some lovely Joni Mitchell. And as it's not HD Tracks at £20/£25/£30 per album, I'm quite happy to use it and enjoy it.

[the_doc735]

Fascinating thread. Not very technical here (I feel I understand, but probably don't really). I have been reading interesting things about temporal resolution as a justification for higher sampling rates, and all the Bob Stuart/MQA stuff around pre-ringing in digital systems (caused by that steep roll-off) and how that can be tackled.

Regardless of the science - or whether it's just down to nice (re)mastering - I have been having a nice time with MQA with Tidal/Roon and a Meridian Explorer 2. I'm so pleased that the Explorer 2 has replaced my Cambridge Audio Stream Magic 6 permanently. Not for MQA, though - but for the way it has had me rediscovering my mostly CD-derived collection (with a smattering of MP3). The MQA stuff really is nice - the Maria Callas remasters, several lovely Karajan EMI albums and some lovely Joni Mitchell. And as it's not HD Tracks at £20/£25/£30 per album, I'm quite happy to use it and enjoy it.

The WTA filter algorithm has taken twenty years of research to develop. It solves the question as to why higher sampling rates sound better. It is well known that 96 kHz (DVD Audio) recordings sound better than 44.1 kHz (CD) recordings. Most people believe that this is due to the presence of ultrasonic information being audible even though the best human hearing is limited to 20kHz. What is not well known is that 768 kHz recordings sound better than 384 kHz and that the sound quality limit for sampling lies in the MHz region. 768 kHz recordings cannot sound better because of information above 200 kHz being important – simply because musical instruments, microphones, amplifiers and loudspeakers do not work at these frequencies nor can we hear them. So if it is not the extra bandwidth that is important, why do higher sampling rates sound better?

The answer is not being able to hear inaudible supersonic information, but the ability to hear the timing of transients more clearly. It has long been known that the human ear and brain can detect differences in the phase of sound between the ears to the order of microseconds. This timing difference between the ears is used for localising high frequency sound. Since transients can be detected down to microseconds, the recording system needs to be able to resolve timing of one microsecond. A sampling rate of 1 MHz is needed to achieve this!

However, 44.1 kHz sampling can be capable of accurately resolving transients by the use of digital filtering. Digital filtering can go some way towards improving resolution without the need for higher sampling rates. However in order to do this the filters need to have infinite long tap lengths. Currently all reconstruction filters have relatively short tap lengths – the largest commercial device is only about 256 taps. It is due to this short tap length and the filter algorithm employed that generates the transient timing errors. These errors turned out to be very audible. Going from 256 taps to 1024 taps gave a massive improvement in sound quality – much smoother, more focused sound quality, with an incredibly deep and precise sound stage.

The initial experiments used variations on existing filter algorithms. Going from 1024 taps to 2048 taps gave a very big improvement in sound quality, and it was implying that almost infinite tap length filters were needed for the ultimate sound quality. At this stage, a new type of algorithm was developed – the WTA filter. This was designed to minimise transient timing errors from the outset, thereby reducing the need for extremely long tap lengths. The WTA algorithm was a success – a 256 tap WTA filter sounded better than all other conventional filters, even with 1024 taps. WTA filters still benefit from long tap lengths; there is a large difference going from 256 taps to 1024 taps.

ROB WATTS. REF: CHORD DAC/64.

[Macca]

The WTA filter algorithm has taken twenty years of research to develop. It solves the question as to why higher sampling rates sound better. It is well known that 96 kHz (DVD Audio) recordings sound better than 44.1 kHz (CD) recordings. Most people believe that this is due to the presence of ultrasonic information being audible even though the best human hearing is limited to 20kHz. What is not well known is that 768 kHz recordings sound better than 384 kHz and that the sound quality limit for sampling lies in the MHz region. 768 kHz recordings cannot sound better because of information above 200 kHz being important – simply because musical instruments, microphones, amplifiers and loudspeakers do not work at these frequencies nor can we hear them. So if it is not the extra bandwidth that is important, why do higher sampling rates sound better?

The answer is not being able to hear inaudible supersonic information, but the ability to hear the timing of transients more clearly. It has long been known that the human ear and brain can detect differences in the phase of sound between the ears to the order of microseconds. This timing difference between the ears is used for localising high frequency sound. Since transients can be detected down to microseconds, the recording system needs to be able to resolve timing of one microsecond. A sampling rate of 1 MHz is needed to achieve this!

However, 44.1 kHz sampling can be capable of accurately resolving transients by the use of digital filtering. Digital filtering can go some way towards improving resolution without the need for higher sampling rates. However in order to do this the filters need to have infinite long tap lengths. Currently all reconstruction filters have relatively short tap lengths – the largest commercial device is only about 256 taps. It is due to this short tap length and the filter algorithm employed that generates the transient timing errors. These errors turned out to be very audible. Going from 256 taps to 1024 taps gave a massive improvement in sound quality – much smoother, more focused sound quality, with an incredibly deep and precise sound stage.

The initial experiments used variations on existing filter algorithms. Going from 1024 taps to 2048 taps gave a very big improvement in sound quality, and it was implying that almost infinite tap length filters were needed for the ultimate sound quality. At this stage, a new type of algorithm was developed – the WTA filter. This was designed to minimise transient timing errors from the outset, thereby reducing the need for extremely long tap lengths. The WTA algorithm was a success – a 256 tap WTA filter sounded better than all other conventional filters, even with 1024 taps. WTA filters still benefit from long tap lengths; there is a large difference going from 256 taps to 1024 taps.

ROB WATTS.

Marketing bollocks.

It solves the question as to why higher sampling rates sound better. It is well known that 96 kHz (DVD Audio) recordings sound better than 44.1 kHz (CD) recordings Odd then that when all else is equal no-one is capable of distinguishing between a 16/44.1 and a 24/192 file in blind testing then.

transients can be detected down to microseconds, the recording system needs to be able to resolve timing of one microsecond. A sampling rate of 1 MHz is needed to achieve this! There is nothing to back up this assertion. it is generally considered that humans are not capable of hearing such miniscule timing differences, and there is nothing to suggest that this can apply to music either. This is just made up on the spot.

Going from 256 taps to 1024 taps gave a massive improvement in sound quality

Really? A massive improvement? How is this being quantified? What comparisons were done? No, no need to go into any detail, I'm sold.

Just made up, back of a fag packet nonsense. Might flog some DACS though.

[StanleyB]

Odd then that when all else is equal no-one is capable of distinguishing between a 16/44.1 and a 24/192 file in blind testing then.

More bollocks from you. The correct information is that those subjected to the music and equipment selected for the test were unable to distinguish between the two formats.
To hear the difference you need to play the type of music that is dynamic and rich in a variety of information within the track. Playing a bunch of tedious notes will sound the same at any bit level. You also need something better than an Amstrad setup. These rumoured test claims are more often than not stripped of any information on the music and equipment used.

I had to end my typing of the above abruptly due to a visitor at the door. So now back on track, I am now able to add the following to my comment:
When people do hear a difference, the first excuse that is offered against that is that the two formats were recorded differently, which is why a difference can be heard. The excuses sceptics come up with to debunk the possibility of hearing a difference are staggering.
One reason why I still listen to my CD collection going back to 1984 is because with improvements in DACs, noise levels in electronic components, and improved power supplies those old discs are slowly releasing additional information that could not be detected previously.