• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Is big better? (A data into DAC musing)

Headchef

Active Member
Joined
May 1, 2021
Messages
152
Likes
42
Location
In a jar, by the door
My system sounds great, I’m very happy with it, but as is our collective affliction, I can’t just leave the bl**dy thing alone…

so rather than swapping things I figured I’d wonder about this for a bit, whilst actually listening to my hifi of course.

I use an Apple MacBook Pro as my digital source, that’s currently feeding (the superb) Aune S8 Reference DAC.

Now the Aune S8 is a very capable DAC being able to digest and expel very high rates (32bit 768k DSD512) but as we’re all aware such source material is very thin on the ground.

Now whilst I was goofing around actually making music I found myself in the OSX Audio MIDI Setup corner of my MacBook where I noticed a whole slew of options for my Aune S8.

Obviously when you set something like Roon to use your DAC it just does it, and the DAC will automatically switch between rates as per the encoding of the file you’re intending on propelling towards your ears. Automatically, *click* done, no dramas, exactly as it’s presented.

But what’s this then? In Audio MIDI set up you can set the rate that’s coming out of your Mac into your DAC, it states quite clearly that it’s using the DAC’s XMOS clock (which is going to be superior to the one inside the Mac) and in the limited OSX Audio MIDI Setup documentation it (almost) clearly states:
  • If your Mac supports the hardware sample rate converter, the Hardware Rate Converter pop-up menu is available in the Input pane. When you choose Automatic from the pop-up menu, the hardware sample rate converter is turned on if the audio samples coming in are PCM and formatted according to the international standard IEC 60958-3. The hardware sample rate converter is turned off if the input stream is encoded. You can also choose Enable or Disable from the menu, to keep the hardware sample rate converter turned on or off.
I can’t claim to understand this, which is why I’m petitioning your great collective hive mind, I’m assuming (possibly incorrectly) that the Mac is using the DAC to upsample the original data which it then converts to beautiful noises?

So am I right in thinking the Mac us uncompressing the file/stream, then sending the data to the DAC (FPGA) which then gets upsampled in the DAC and turned into that lovely analog goodness, or is the Mac uncompressing the file then using IEC 60958-3 to upsample it then send it to the DAC?

is it better to feed the DAC a 2ch 32-bit Integer 768.0kHz file from the Mac or just let it get fed straight from the software “transport“ at what ever the rate it comes as?

No I haven’t done any testing be it blind or otherwise, I’d just noticed this functionality on the Mac/DAC and began wondering about it.

Anyone know? Thanks in advance for your input be it high or low res :D

And just in case you were curious my set up is as follows:

Macbook Pro (various high res masters, all streaming platforms and also Roon)
Aune S8
EAR / Yoshino 859
ProAc Response 1SC
x2 Rel Quake

TellleriumQ Black II Interconnect & Speaker cable.
 

Vincent Kars

Addicted to Fun and Learning
Technical Expert
Joined
Mar 1, 2016
Messages
781
Likes
1,555
All most all operating systems work the same.
They are designed to handle multiple audio streams and as a consequence, run at a fixed sample rate. (You can’t mix different sample rates, neither send different rates to a DAC).
Special software like Roon or Audirvana can do automatic sample rate switching.

I don’t expect the AUNE USB driver to provide automatic sample rate switching.
If the Aune can display the sample rate, you can easily check this.
Play a 16/44.1 and a 24/96. If the incoming rate remains the same, it is the Mac converting everything to the sample rate as set in the Midi panel.

It has asynchronous USB so the XMOS part will do the buffer management.

Most DAC’s do up- or oversampling. Looks like the driver interface allows you to defeat this.

Nothing wrong by setting the bit depth to 32.
IMHO this is not about the dynamic range of the recording but about the arithmetic precision of the data path to your DAC.
Sample rate I should set to what is common in you collection.
You might have a look at a simple tool as Bitperfect.
 

DSJR

Major Contributor
Joined
Jan 27, 2020
Messages
3,314
Likes
4,427
Location
Suffolk Coastal, UK
I'd get rid of the tellurium cable nonsense and maybe have a look at a more neutral amp myself using (shock horror) transistors. The Response 1sc's are too sparkly too up top as was Pro-Acs design choice (customers loved 'em when demmed while I was recoiling away on cymbal crashes, but the 859 may smooth this out - I loved the V12 amp but found the others I heard (509's and 899 in more recent years) smothered the highs a bit - again, another design choice by Tim when he was alive as he could design anything to order).

There! You mentioned a need to tweak, replace and fiddle with our stereo's as is our wont, so leave the source alone and start to 'normalise' everything else towards neutrality :D
 

Daverz

Major Contributor
Joined
Mar 17, 2019
Messages
1,294
Likes
1,451
I think the general attitude here would be that the noise and distortion of the tube amp would swamp any concern about Apple's resampling.

I would set the Audio-Midi settings to something like 32-bit/96 kHz. Even with Benchmark amp levels of SINAD, it might not be an audible issue. You might want to look into a streaming solution that doesn't have this limitation (and no umbilical from the macbook to the DAC).
 
OP
Headchef

Headchef

Active Member
Joined
May 1, 2021
Messages
152
Likes
42
Location
In a jar, by the door
All most all operating systems work the same.
They are designed to handle multiple audio streams and as a consequence, run at a fixed sample rate. (You can’t mix different sample rates, neither send different rates to a DAC).
Special software like Roon or Audirvana can do automatic sample rate switching.

I don’t expect the AUNE USB driver to provide automatic sample rate switching.
If the Aune can display the sample rate, you can easily check this.
Play a 16/44.1 and a 24/96. If the incoming rate remains the same, it is the Mac converting everything to the sample rate as set in the Midi panel.

It has asynchronous USB so the XMOS part will do the buffer management.

Most DAC’s do up- or oversampling. Looks like the driver interface allows you to defeat this.

Nothing wrong by setting the bit depth to 32.
IMHO this is not about the dynamic range of the recording but about the arithmetic precision of the data path to your DAC.
Sample rate I should set to what is common in you collection.
You might have a look at a simple tool as Bitperfect.

the Aune will switch on the fly dependant on what’s being sent from any source be it the optical in from the (Apple) TV or any of the (software) transports, and it shows the change in rates on the screen, it works without any issue.

I was curious though that Audio MIDI Setup allows the option to fix the rate, using the DAC’s XMOS clock, which suggests to me that the upsampling is being done between the Mac & the Dac “according to the international standard IEC 60958-3”, which is probably utterly pointless because if there’s no data there there’s nothing to upsample.
 

Vincent Kars

Addicted to Fun and Learning
Technical Expert
Joined
Mar 1, 2016
Messages
781
Likes
1,555
IEC 60958-3
Looks like this is the current standard for AES3 and SPDIF.
Normally you send PCM using this protocol but it might also be DTS

Maybe the following is the case:
AES3/SPDIF have in common that the send rate of the bus is the sample rate.
The send rate is generated by a clock at the sender, in your case the Mac.
This won't be one with femto precision.
As they mention the clock at the side of the DAC this might be about re-clocking the AES3/SPDIF input to get rid of the input jitter.
 
OP
Headchef

Headchef

Active Member
Joined
May 1, 2021
Messages
152
Likes
42
Location
In a jar, by the door
I'd get rid of the tellurium cable nonsense and maybe have a look at a more neutral amp myself using (shock horror) transistors. The Response 1sc's are too sparkly too up top as was Pro-Acs design choice (customers loved 'em when demmed while I was recoiling away on cymbal crashes, but the 859 may smooth this out - I loved the V12 amp but found the others I heard (509's and 899 in more recent years) smothered the highs a bit - again, another design choice by Tim when he was alive as he could design anything to order).

There! You mentioned a need to tweak, replace and fiddle with our stereo's as is our wont, so leave the source alone and start to 'normalise' everything else towards neutrality :D

Yes yes, I’m well aware that there’s a predominant school of thought that believes that coat hangers sound just as good as fancy cables and if that’s what tickles your pickle then so be it! :D But whilst we’re on the subject the TelluriumQ cables do have a pronounced effect on the sound that comes out of the speakers, and it pleases me to hear it. As a musician I know that changes to the signal chain can and do have an effect on the sound that comes from the amp and replacing my old 0.5m interconnect with the 1m TQ interconnect had an undeniable and distinct improvement in the sound output. You are of course welcome to disagree but you’re wrong! :D

As for the amp, it sounds lovely, great presentation, soundstage, blah blah it’s a joy to listen to and the ProAc’s not at all splashy or sharp on the top end, I do think that the art to making a great sounding system is the matching of the components. After all music should be a joy to listen to and having spent tens of thousands in the past 30 years, and ended up with things that whilst analytically revealing and no doubt exceptionally well measuring they would fail to convey the very essence of what music is all about. And that’s not graphs or perfect sine waves but a sound that cups your bollocks and makes you want to dance around the living room (rather than rub your thighs under a desk at a distortion free plotted line!).

of course science, measurements and benchmarks are all useful tools in the pursuit of ‘a perfect sound forever” it’s important not to discount entirely the reason that music exists!
 
OP
Headchef

Headchef

Active Member
Joined
May 1, 2021
Messages
152
Likes
42
Location
In a jar, by the door
I think the general attitude here would be that the noise and distortion of the tube amp would swamp any concern about Apple's resampling.

I would set the Audio-Midi settings to something like 32-bit/96 kHz. Even with Benchmark amp levels of SINAD, it might not be an audible issue. You might want to look into a streaming solution that doesn't have this limitation (and no umbilical from the macbook to the DAC).

have a read up on the noise/distortion figures of the EAR / Yoshino 859 ;)
 
OP
Headchef

Headchef

Active Member
Joined
May 1, 2021
Messages
152
Likes
42
Location
In a jar, by the door
Looks like this is the current standard for AES3 and SPDIF.
Normally you send PCM using this protocol but it might also be DTS

Maybe the following is the case:
AES3/SPDIF have in common that the send rate of the bus is the sample rate.
The send rate is generated by a clock at the sender, in your case the Mac.
This won't be one with femto precision.
As they mention the clock at the side of the DAC this might be about re-clocking the AES3/SPDIF input to get rid of the input jitter.

I think you’ve ”clocked” exactly what I was wondering about, the Aune S8 says the following in their marketino:

“For sake of listening joy, for sake ASRC
Thanks to the digital processing technology from Titans Audio Lab, the S8 can completely discard the ASRC inside the DAC chip and present music more vividly. Authentic femtosecond clock reproduction: the S8 adopted the testing standard AABT (Aune Audio Band Test) that exceeds AES-12id- 2006 (r201 7). Strict standard brings natural and musical sound.

Authentic femtosecond oscillators, makes USB distinguished
USB has basically been a minor role for traditional DACs. The S8 specifically enhanced the USB design: adopted the solution of the combination of XMOS and FPGA, in which XMOS does USB data analysis and FPGA does audio data synthesis. Moreover, the customized Thesycon driver enables the S8 to support music files up to 32bit 768k and DSD512. Plus the two high-performance authentic femtosecond clocks, this time, USB has become a reason why you want to buy a DAC, too.”

which means about as much to me as a helicopter does to a cow. What I’m attempting to figure out is if by setting the Mac’s (audio MIDI Setup) that the data files are then being fed through the DAC’s clever bits and pieces or if the host PC (the MacBook) is doing the heavy lifting and then just sending the upsampled files to the DAC, given that it’s showing that the DAC’s XMOS is being used in the settings I think it suggests that the MacBook is sending the decompressed files via AES3/SPDIF/PCM to the DAC which is then doing its fancy upscaling?
 

Jas0_0

Active Member
Forum Donor
Joined
Oct 3, 2019
Messages
285
Likes
513
Yes yes, I’m well aware that there’s a predominant school of thought that believes that coat hangers sound just as good as fancy cables and if that’s what tickles your pickle then so be it! :D But whilst we’re on the subject the TelluriumQ cables do have a pronounced effect on the sound that comes out of the speakers, and it pleases me to hear it. As a musician I know that changes to the signal chain can and do have an effect on the sound that comes from the amp and replacing my old 0.5m interconnect with the 1m TQ interconnect had an undeniable and distinct improvement in the sound output. You are of course welcome to disagree but you’re wrong! :D

As for the amp, it sounds lovely, great presentation, soundstage, blah blah it’s a joy to listen to and the ProAc’s not at all splashy or sharp on the top end, I do think that the art to making a great sounding system is the matching of the components. After all music should be a joy to listen to and having spent tens of thousands in the past 30 years, and ended up with things that whilst analytically revealing and no doubt exceptionally well measuring they would fail to convey the very essence of what music is all about. And that’s not graphs or perfect sine waves but a sound that cups your bollocks and makes you want to dance around the living room (rather than rub your thighs under a desk at a distortion free plotted line!).

of course science, measurements and benchmarks are all useful tools in the pursuit of ‘a perfect sound forever” it’s important not to discount entirely the reason that music exists!

Suggest reading post 1 of this for a sense of the climate on this forum:

https://www.audiosciencereview.com/...iles-posting-at-asr-for-the-first-time.17598/
 
OP
Headchef

Headchef

Active Member
Joined
May 1, 2021
Messages
152
Likes
42
Location
In a jar, by the door
Suggest reading post 1 of this for a sense of the climate on this forum:

https://www.audiosciencereview.com/...iles-posting-at-asr-for-the-first-time.17598/

I’ve read that, and I understand it, but I’ve never been one to follow the herd so to speak. There’s absolute value in the science aspect and I enjoy it, but then I’m mindful not to lose sight of my ultimate aim which is to play music in a way that delivers the most pleasurable experience. Cables or rebar withstanding :D
 

Vincent Kars

Addicted to Fun and Learning
Technical Expert
Joined
Mar 1, 2016
Messages
781
Likes
1,555
I think it suggests that the MacBook is sending the decompressed files via AES3/SPDIF/PCM to the DAC
Older Mac's had a headphone out doubling as a Toslink out.
Is your Mac connected over Toslink or over the USB?
I assume USB.

Personally I prefer to play everything at its native sample rate simply because most DAC's are up- or over-sampling.
Hence if the PC does re-sampling and send this to the DAC, it will be resampled by the DAC again.
Resampling as done by OSX is not a crime, in fact it is pretty good https://www.thewelltemperedcomputer.com/SW/OSX/AudioMidi.htm
 

DSJR

Major Contributor
Joined
Jan 27, 2020
Messages
3,314
Likes
4,427
Location
Suffolk Coastal, UK
Yes yes, I’m well aware that there’s a predominant school of thought that believes that coat hangers sound just as good as fancy cables and if that’s what tickles your pickle then so be it! :D But whilst we’re on the subject the TelluriumQ cables do have a pronounced effect on the sound that comes out of the speakers, and it pleases me to hear it. As a musician I know that changes to the signal chain can and do have an effect on the sound that comes from the amp and replacing my old 0.5m interconnect with the 1m TQ interconnect had an undeniable and distinct improvement in the sound output. You are of course welcome to disagree but you’re wrong! :D

As for the amp, it sounds lovely, great presentation, soundstage, blah blah it’s a joy to listen to and the ProAc’s not at all splashy or sharp on the top end, I do think that the art to making a great sounding system is the matching of the components. After all music should be a joy to listen to and having spent tens of thousands in the past 30 years, and ended up with things that whilst analytically revealing and no doubt exceptionally well measuring they would fail to convey the very essence of what music is all about. And that’s not graphs or perfect sine waves but a sound that cups your bollocks and makes you want to dance around the living room (rather than rub your thighs under a desk at a distortion free plotted line!).

Of course science, measurements and benchmarks are all useful tools in the pursuit of ‘a perfect sound forever” it’s important not to discount entirely the reason that music exists!

I am sorry for slipping into dealer mode... It's just your stereo looks to be a very toppy speaker (a Pro-Ac thing as current domestic PMC's have) smothered by the amp and cables (high inductance speaker cables 'designed' to have a sonic character with suspect amps) and a valve amp doing all manner of things to eq the sound and with the characteristic E.A.R high frequency 'softness' which is neither realistic nor required with better balanced speakers. Like I said, the V12 doesn't do that and sounds really nice to me. Go listen to your Response 1sc's with a more neutral amp and you'll notice that tweeter, I bet ;)

I do appreciate you obviously love the tones your system gives you, but the vibe here on ASR is more - 'by all means colour and distort the sound in the recording and mixing stage to get the artistic effect, but for goodness sake keep the playback system as transparent and honest as possible so the artistic intent of the recordings comes through without being stamped on further by the playback system.' Nothing to do with sine waves and other such, just a feeling for as much distortion-free playback as possible - and that includes the room as well :)

My reaction was that you have a system made of such careful sonic balances one against the other (how audiophiles seem to get off on their hobby), that who knows what a source change is going to do - and I can guarantee you'll have different results to many here when making these changes.

In the 90's when I could afford to play around with gear, I went to many jazz gigs despite this genre not really being my thing. I was knocked flat by a close up drum kit played 'vigorously' and trumpets and saxes pushing me into the back wall with their intensity. I went the active ATC route and loved every second of their ownership, as in my small room, I was able to recreate a very passable facsimile of this when the neighbours were out and using well recorded 1950's jazz discs. I had to sell it all for marriage, combining two small homes into one and I weep still at the loss of that quality of sound I had, despite perhaps being of its time.

Coat hanger speaker 'wires' - YEAH - bring it on!!!!! Inline resistance may be a bit high I admit, but it was done once and nobody could tell the difference until shown (I don't think the runs were very long though). Apart from that, unless the capacitance or inductance of the cables is artifically skewed you can safely choose a speaker wire by the gauge of the copper cable and the colour of the jacket (been there and done it, so quite happy).


So, you carry on enjoying your stereo, just please don't expect to have a smooth ride here - and it's not about conforming either :) I know exactly where my particular 'sound' wants to be - and no amount of source or amp changes will do it, despite my trying....
 
Last edited:
OP
Headchef

Headchef

Active Member
Joined
May 1, 2021
Messages
152
Likes
42
Location
In a jar, by the door
Older Mac's had a headphone out doubling as a Toslink out.
Is your Mac connected over Toslink or over the USB?
I assume USB.

Personally I prefer to play everything at its native sample rate simply because most DAC's are up- or over-sampling.
Hence if the PC does re-sampling and send this to the DAC, it will be resampled by the DAC again.
Resampling as done by OSX is not a crime, in fact it is pretty good https://www.thewelltemperedcomputer.com/SW/OSX/AudioMidi.htm
site (click through on your link) that the OSX upsampling seems very good. I wonder what effects presenting different DAC’s with the upsampled (of dac upsampled) data would give and if it would give different readings when testing the DAC’s?
 
OP
Headchef

Headchef

Active Member
Joined
May 1, 2021
Messages
152
Likes
42
Location
In a jar, by the door
I am sorry for slipping into dealer mode... It's just your stereo looks to be a very toppy speaker (a Pro-Ac thing as current domestic PMC's have) smothered by the amp and cables (high inductance speaker cables 'designed' to have a sonic character with suspect amps) and a valve amp doing all manner of things to eq the sound and with the characteristic E.A.R high frequency 'softness' which is neither realistic nor required with better balanced speakers. Like I said, the V12 doesn't do that and sounds really nice to me. Go listen to your Response 1sc's with a more neutral amp and you'll notice that tweeter, I bet ;)

I do appreciate you obviously love the tones your system gives you, but the vibe here on ASR is more - 'by all means colour and distort the sound in the recording and mixing stage to get the artistic effect, but for goodness sake keep the playback system as transparent and honest as possible so the artistic intent of the recordings comes through without being stamped on further by the playback system.' Nothing to do with sine waves and other such, just a feeling for as much distortion-free playback as possible - and that includes the room as well :)

My reaction was that you have a system made of such careful sonic balances one against the other (how audiophiles seem to get off on their hobby), that who knows what a source change is going to do - and I can guarantee you'll have different results to many here when making these changes.

In the 90's when I could afford to play around with gear, I went to many jazz gigs despite this genre not really being my thing. I was knocked flat by a close up drum kit played 'vigorously' and trumpets and saxes pushing me into the back wall with their intensity. I went the active ATC route and loved every second of their ownership, as in my small room, I was able to recreate a very passable facsimile of this when the neighbours were out and using well recorded 1950's jazz discs. I had to sell it all for marriage, combining two small homes into one and I weep still at the loss of that quality of sound I had, despite perhaps being of its time.

Coat hanger speaker 'wires' - YEAH - bring it on!!!!! Inline resistance may be a bit high I admit, but it was done once and nobody could tell the difference until shown (I don't think the runs were very long though). Apart from that, unless the capacitance or inductance of the cables is artifically skewed you can safely choose a speaker wire by the gauge of the copper cable and the colour of the jacket (been there and done it, so quite happy).


So, you carry on enjoying your stereo, just please don't expect to have a smooth ride here - and it's not about conforming either :) I know exactly where my particular 'sound' wants to be - and no amount of source or amp changes will do it, despite my trying....

oh I’m not worried about a smooth ride! :D no forwards motion without a little turbulence :D

i must admit I do enjoy the audiophile vice of trying different things, and I’m a great believer in the rubbish in rubbish out school of thought (hence my original question as to what is the optimal way of “driving’ the dac), I have a friend threatening to pop a Chord Qutest around to see how that plays out, happy to give an unlearned opinion based review and then dive back into a valve heated Anderson shelter as the comments pile on! :D
 

DSJR

Major Contributor
Joined
Jan 27, 2020
Messages
3,314
Likes
4,427
Location
Suffolk Coastal, UK
oh I’m not worried about a smooth ride! :D no forwards motion without a little turbulence :D

I must admit I do enjoy the audiophile vice of trying different things, and I’m a great believer in the rubbish in rubbish out school of thought (hence my original question as to what is the optimal way of “driving’ the dac), I have a friend threatening to pop a Chord Qutest around to see how that plays out, happy to give an unlearned opinion based review and then dive back into a valve heated Anderson shelter as the comments pile on! :D

I believe Chord digital stuff (and some of their amplification) perform really well - their amps suit me a lot better than an equivalent Naim for example, but that's me... The ONLY rider on Chord gear really is the price charged - and I remember overhearing a conversation with the boss and someone else and he explained to the other party how they felt they needed to dress the gear up a good bit and increase the prices to cover it. Now, many enthusiasts wouldn't bat an eyelid at the prices (and don't, even for the little sweet-box dacs and so on), but I'm way over that kind of vibe these days (I look to used gear now as and when I've needed to change).

I can't offer anything (you'll be pleased to know) regarding streaming, as I'm at the Chromecast Audio level and nearly crushed the bloody thing a few days ago when I couldn't stream to it (VLC player) and when I tried using my phone, I had hang-ups and glitches (thank Gawd I've kept my CD's and LP's). After major uninstalls and re-installs, plus a couple of network resets I finally got it working. Roon? Do leave off, I'm not minted as many of you lot seem to be ;) My next dac will be a basic Topping or Schiit (the latter does have some kind of support in the UK and I like the story they tell, especially now they're taking properly good performance a bit more seriously which isn't doing their business any harm either, while leaving the wacky stuff there if anyone wants to spend more for the extra tales told ;) )

My feeling only - and thanks for being so good humoured about my Aspergic posts above - is for you to *try to eventually* gradually bring the pendulum of one aspect heavily balancing another towards a more central position. I swear you'll suddenly find 'differences between cables and gear' a heck of a lot less - and you could save yourself a hell of a lot of money. Pros use wires to connect stuff together and as long as they don't get hum or a degradation in the signal passing through it, I bet you'd find they couldn't give a s**t what make it is :D Valves can keep you warm in winter but I firmly believe an anonymous little box with a modern Class D powerhouse inside is really where it's at and some of them are superbly made inside with carefully considered cable layouts and so on.

Good luck and keep with the ride here, even if you don't always agree with it... ;)
 
OP
Headchef

Headchef

Active Member
Joined
May 1, 2021
Messages
152
Likes
42
Location
In a jar, by the door
Are there measurements published somewhere?

there’s some explanation here: http://www.allegrosound.com/EAR_859_AllegroSound.html

obviously not up to the critical analysis you’d get from this site it’s more your traditional wiff-waff but it’s the best I can offer given the age of the thing, that said if anyone can suggest a method that I can gather the data I’d be happy to do it :)
 
Top Bottom