A revolution in audio rendering - classic from 10 years ago!

[formdissolve]

"A revolution in audio rendering"

Oldie but goodie. Pretty funny how far people go to think they are hearing something better.

"...found that a function called memcpy was the culprit, most memory players use memcpy and this is one of the reasons why memory play sounds worse ie digital sounding. Fortunately there is an optimised version of memcpy from http://www.agner.org/optimize/, using this version removes the hard edge produced by memcpy. the other thing I did was to close the file after reading into the buffer."

Some quotes:
"Sounds awesome, the previous version had a slight tendency to defuse the treble, but with VS2012 compile it is a much more complete sound with absolutely no digital harshness, some 16/44 albums I could hardly play before without getting a headache are now rendered in their full glory."
"It's just lazyness on the part of the player developers that they rely on the old methods, I guess they think bits are bits."
"Goto also sounds better than anything else I have tried." - the "Goto" in the quote means a goto in C used to replace a loop.
"also most players use malloc to get memory while new is the c++ method and sounds better."

 

[MaxwellsEq]

 

[fatoldgit]

Well.... about 3 years ago I "discovered" the Computer Audio forum (now renamed Audiophile Style) and there were numerous voodoo threads on there about tweaking your computer for better audio.

There were threads on different RAM memory modules that made massive differences (ECC memory from a specific brand was best)

Threads on different motherboards, CPU's, Sata cables, network cards, usb cards, LPS power supplies, SSD disk drives etc making massive differences (all cumulative of course!!!)

And of course threads on all types of network tweaks: reclocking routers and switches, specific routers/switches, different ethernet cables, power supplies attached to routers, switches and media convertors, OS level network changes etc making massive differences.

All of this stuff is expensive... check out the stuff from JCAT.

Its bloody hard to do a real unsighted test/instantaneous switching for comparsion when your have to spend several minutes swapping out RAM, hard drives etc.

The only "obvious to anyone" tweaks that might make a difference would be media conversion (i.e. ethernet -> fibre -> ethernet) and an LPS on the endpoint computer (i.e. that which plugs into the DAC).

I am now 100% PC/USB DAC based and aside from implementing the above two tweaks (to soothe my paranoia, less than $US 300 spent) the other simple thing is to have a two PC system: a big grunter storing your files/upsampling/EQ'ing not in your room and a low powered, silent diskless PC that boots into memory off a USB stick that is the end point into your DAC.

So people ended up with topologies like the one below to get music from their router to the DAC (including 7 LPS's, reclocked switches, high end ethernet cables, PS audio perfectwave, Chord DAC's etc... the last two being ASR favourites!!!).

 

[antcollinet]

Well.... about 3 years ago I "discovered" the Computer Audio forum (now renamed Audiophile Style) and there were numerous voodoo threads on there about tweaking your computer for better audio.

There were threads on different RAM memory modules that made massive differences (ECC memory from a specific brand was best)

Threads on different motherboards, CPU's, Sata cables, network cards, usb cards, LPS power supplies, SSD disk drives etc making massive differences (all cumulative of course!!!)

And of course threads on all types of network tweaks: reclocking routers and switches, specific routers/switches, different ethernet cables, power supplies attached to routers, switches and media convertors, OS level network changes etc making massive differences.

All of this stuff is expensive... check out the stuff from JCAT.

Its bloody hard to do a real unsighted test/instantaneous switching for comparsion when your have to spend several minutes swapping out RAM, hard drives etc.

The only "obvious to anyone" tweaks that might make a difference would be media conversion (i.e. ethernet -> fibre -> ethernet) and an LPS on the endpoint computer (i.e. that which plugs into the DAC).

I am now 100% PC/USB DAC based and aside from implementing the above two tweaks (to soothe my paranoia, less than $US 300 spent) the other simple thing is to have a two PC system: a big grunter storing your files/upsampling/EQ'ing not in your room and a low powered, silent diskless PC that boots into memory off a USB stick that is the end point into your DAC.

So people ended up with topologies like the one below to get music from their router to the DAC (including 7 LPS's, reclocked switches, high end ethernet cables, PS audio perfectwave, Chord DAC's etc... the last two being ASR favourites!!!).

View attachment 272357

Nuts.

 

[MaxwellsEq]

Well.... about 3 years ago I "discovered" the Computer Audio forum (now renamed Audiophile Style) and there were numerous voodoo threads on there about tweaking your computer for better audio.

There were threads on different RAM memory modules that made massive differences (ECC memory from a specific brand was best)

Threads on different motherboards, CPU's, Sata cables, network cards, usb cards, LPS power supplies, SSD disk drives etc making massive differences (all cumulative of course!!!)

And of course threads on all types of network tweaks: reclocking routers and switches, specific routers/switches, different ethernet cables, power supplies attached to routers, switches and media convertors, OS level network changes etc making massive differences.

All of this stuff is expensive... check out the stuff from JCAT.

Its bloody hard to do a real unsighted test/instantaneous switching for comparsion when your have to spend several minutes swapping out RAM, hard drives etc.

The only "obvious to anyone" tweaks that might make a difference would be media conversion (i.e. ethernet -> fibre -> ethernet) and an LPS on the endpoint computer (i.e. that which plugs into the DAC).

I am now 100% PC/USB DAC based and aside from implementing the above two tweaks (to soothe my paranoia, less than $US 300 spent) the other simple thing is to have a two PC system: a big grunter storing your files/upsampling/EQ'ing not in your room and a low powered, silent diskless PC that boots into memory off a USB stick that is the end point into your DAC.

So people ended up with topologies like the one below to get music from their router to the DAC (including 7 LPS's, reclocked switches, high end ethernet cables, PS audio perfectwave, Chord DAC's etc... the last two being ASR favourites!!!).

View attachment 272357

I can understand why people might need to tweak turntables and speakers. I sort of get analogue electronic tweaking. But this just demonstrates a lack of understanding of the underlying technology.

 

[FrantzM]

Thing is , before ASR, there weren't many places where an audiophile could gather decent, honest information. About the same time 12 years ago, I, went to that website for info and .... spent a good amount of money, on their fool errands. I did build one their recommended computers/ My memory of it, is that I spent a lot of time and money, finding the perfect motherboard, then the perfect power supply (linear of course) then the best headphones amps, then the DACs.. On those I spent a bundle, and on ESL headphones too ...
.
Underlying all these sites is sales. that is all that there is to these. Cynical. They know it is all BS, but they also know it makes them money. Those who frequent those sites spend, and a lot. There is always up there a violently unobtainium headphones or amp or DAC or PC or cables (!!!)), to acquire as the ultimate.. of course numerous tweaks and settings on computer ... and lest your forget software. It never ends.
Here in my quest for better sound through headphones. I landed, thanks to ASR. It seems, almost anticlimatic. You were so wanting the latest and willing to up your disbursements ...and here at ASR, you are proven that you can't hear past 15 KHz ... Can't recognize 3% THD from 0.001% (that one was a blow for me on the Kippel Test) . That 50 dB of SINAD is quite OK, that you can't distinguish 320 mp3 from lossless (another blow for many) , that Hi-Rez is Highway robbery... That you have been fooled for so many years First is a sense of being humbled, even of humiliation. Then the calm sets in: you are grounded. You have a footing. You EQ, your various headphones, while they don't sound the same but the better (in term of FR and THD and lack of resonances) sound similar, eerily similar .. The TOTL do continue to sound better but not by a large margin... Liberation! Joy! Now you discover that there was sooooooo much music to listen to, to enjoy. And the search is over ... or almost: You are an audiophile and do know (search maybe) there is something better.. it has measured better but now you do ask yourself the question.. Do I need to spend that much? Or change what I had? OR even can I hear those differences... Buying or staying put would depend on your pocket depth, or amount of discipline but often ... You stay the course. FOMO has disappeared.

I am in that place ... for headphones .

Peace.

 

[FrantzM]

Hi
I did a bit of search on the products on the chart below:

the SOtM gigabit swich is $800.00
The Mutec Ref 10 Clock is ... GASP!! DAMN ... etc.. $4999.99
The Paul Hynes are the stuff of Legend and there is a controversy on the person/business, those (Linear, of course) power supplies vary from $500 to 4000.oo..I didn't bother to research ..
I'll stop there ...

What a pile of Bovine Manure!!!


Peace.

 

[AdamG]

Nuts.

Most excellent response Sir! I seem to recall that one word response from history.

 

[charleski]

IBM did an ad for this many years ago, since there's a lot more money in commercial servers that's where you want to put the magic:

 

[fatoldgit]

Hi
I did a bit of search on the products on the chart below:



the SOtM gigabit swich is $800.00
The Mutec Ref 10 Clock is ... GASP!! DAMN ... etc.. $4999.99
The Paul Hynes are the stuff of Legend and there is a controversy on the person/business, those (Linear, of course) power supplies vary from $500 to 4000.oo..I didn't bother to research ..
I'll stop there ...

What a pile of Bovine Manure!!!


Peace.

I wont bother going thru all the available JCAT stuff but that network card in the diagram above is EURO 435 and you can, for example, get a 320GB SSD drive (noted as FEMTO enhanced) for EURO 1025.

Also note in the above diagram that we go from in-wall CAT 5e to a CAT 8 cable into two CAT 7 cables (all expensive models)

I am sure that there is some logic (provable of course) to this configuration.

 

[FrantzM]

I wont bother going thru all the available JCAT stuff but that network card in the diagram above is EURO 435 and you can, for example, get a 320GB SSD drive (noted as FEMTO enhanced) for EURO 1025.

View attachment 272415

Also note in the above diagram that we go from in-wall CAT 5e to a CAT 8 cable into two CAT 7 cables (all expensive models)

I am sure that there is some logic (provable of course) to this configuration.

There is no logic to this configuration. Well ... except to have people spend money...


Peace.

 

[AdamG]

There is no logic to this configuration. Well ... except to have people spend money...


Peace.

Bingo was his name….Oooo

 

[formdissolve]

I can understand why people might need to tweak turntables and speakers. I sort of get analogue electronic tweaking. But this just demonstrates a lack of understanding of the underlying technology.

High level analogue tweaking definitely offers immediate differences - i.e. cartridges, needles, loading, stuff like that. That's all audible and measurable in the end. But all this digital tweaking mentioned in the article, and especially that chart that was posted is so absurd as to look like a ChatGPT AI post of nonsense. It borders on perversion and exploitation! Of course someone who spends all that time and money will go VOILA, THIS was what I was missing! The brain is a curious piece of flesh.

 

[dorakeg]

Well.... about 3 years ago I "discovered" the Computer Audio forum (now renamed Audiophile Style) and there were numerous voodoo threads on there about tweaking your computer for better audio.

There were threads on different RAM memory modules that made massive differences (ECC memory from a specific brand was best)

Threads on different motherboards, CPU's, Sata cables, network cards, usb cards, LPS power supplies, SSD disk drives etc making massive differences (all cumulative of course!!!)

And of course threads on all types of network tweaks: reclocking routers and switches, specific routers/switches, different ethernet cables, power supplies attached to routers, switches and media convertors, OS level network changes etc making massive differences.

All of this stuff is expensive... check out the stuff from JCAT.

Its bloody hard to do a real unsighted test/instantaneous switching for comparsion when your have to spend several minutes swapping out RAM, hard drives etc.

The only "obvious to anyone" tweaks that might make a difference would be media conversion (i.e. ethernet -> fibre -> ethernet) and an LPS on the endpoint computer (i.e. that which plugs into the DAC).

I am now 100% PC/USB DAC based and aside from implementing the above two tweaks (to soothe my paranoia, less than $US 300 spent) the other simple thing is to have a two PC system: a big grunter storing your files/upsampling/EQ'ing not in your room and a low powered, silent diskless PC that boots into memory off a USB stick that is the end point into your DAC.

So people ended up with topologies like the one below to get music from their router to the DAC (including 7 LPS's, reclocked switches, high end ethernet cables, PS audio perfectwave, Chord DAC's etc... the last two being ASR favourites!!!).

View attachment 272357

Lol, all the snake oil crap.....

These folks have absolutely no idea how data is transferred via network and USB.

While I do agree that noise is importantissue , it can be easily done via RJ45/USB isolator. The referece clock does nothing at all since network and USB transfer data in asynchronous mode (unlike SPDIF).

 

[fatoldgit]

. The referece clock does nothing at all since network and USB transfer data in asynchronous mode (unlike SPDIF).

And lets not forget all the network buffers along the way (which totally invalidate clocks within and between network devices):

- in the switches
- in the PC NIC cards
- in the TCP/IP stack
- in the playback software

And if your streaming... all the Internet plumbing to get it to your router.

As long as the playback software buffer is never starved... all is good

I have reasonable knowledge of TCP/IP programming (Unix/Linux 40+ years) but very little about the internals of USB buffering aside from the simple point it is asynchronous relative to USB audio but I guess there is at least two buffers: one in the USB receiver and one in the USB OS stack.

But in the end, as has been said many times... its just a hobby... its their money and probably no one gets hurt.

KInda like the $15 supermarket steak thats sold for $400 in a restaurants cause its covered in edible Gold. Doesnt change the fundamental nature of the steak but the person eating it gets to feel good about themselves and gets bragging rights.

Peter

 

[dorakeg]

And lets not forget all the network buffers along the way (which totally invalidate clocks within and between network devices):

- in the switches
- in the PC NIC cards
- in the TCP/IP stack
- in the playback software

And if your streaming... all the Internet plumbing to get it to your router.

As long as the playback software buffer is never starved... all is good

I have reasonable knowledge of TCP/IP programming (Unix/Linux 40+ years) but very little about the internals of USB buffering aside from the simple point it is asynchronous relative to USB audio but I guess there is at least two buffers: one in the USB receiver and one in the USB OS stack.

But in the end, as has been said many times... its just a hobby... its their money and probably no one gets hurt.

KInda like the $15 supermarket steak thats sold for $400 in a restaurants cause its covered in edible Gold. Doesnt change the fundamental nature of the steak but the person eating it gets to feel good about themselves and gets bragging rights.

Peter

I am not exactly sure about USB buffering. However, you are right about software playback. The main buffer is using either part of your RAM or SSD/HDD space to store incoming data. ITs size is only limited by how much RAM/storage space the computer has.


The Roon server has this crap too (itself is jsut an Intel NUC).

NET Card FEMTO - JCAT . next-gen computer audio

1st ever audiophile ethernet network interface for music streaming. Get better sound from TIDAL & Qobuz. FREE WORLDWIDE DELIVERY

jcat.eu

Jitter was never an issue for TCP/IP protocol.
Enterprise level network controller does nothing to improve audio quality.
Femto clock does nothing. Even if you connect it to an atomic clock (most accurate clock available to man), it does nothing as well.
1Gbps does nothing too, audio streaming consumes very little bandwidth
"audiophile" grade gold plated network ports?? Even network port has gold plated contacts. While gold plating sounds impressive. IT actually does nothing for audio. ITs electrical conductivity is even worse than copper. The only thing good about gold is that its highly resistance to oxidation and corrosion.

This is way too much even for me.

 

[IAtaman]

"...found that a function called memcpy was the culprit, most memory players use memcpy and this is one of the reasons why memory play sounds worse ie digital sounding. Fortunately there is an optimised version of memcpy from http://www.agner.org/optimize/, using this version removes the hard edge produced by memcpy. the other thing I did was to close the file after reading into the buffer."

If you needed proof that any sufficiently advanced technology is indistinguishable from magic (to those who don't understand it), this would be it.

This kind of stuff always worries me. To which topics I am this ignorant to believe BS of this level?

 

[fpitas]

I can understand why people might need to tweak turntables and speakers. I sort of get analogue electronic tweaking. But this just demonstrates a lack of understanding of the underlying technology.

On the bright side, they can't screw up the digital too much without it not functioning at all. So they can get pride of accomplishment without suffering through degraded audio.

 

[JP]

The only "obvious to anyone" tweaks that might make a difference would be media conversion (i.e. ethernet -> fibre -> ethernet)

Nope.

 

[JayGilb]

I've worked on ASIC designs that operate with greater than gigahertz clock speeds and point to point timing requirements less than a picosecond. The jitter specs on some of these audio word clock generators (oven controlled, etc) is tighter than we use when designing these chips.

Even the highest audio sampling rates do not require such a level of accuracy, yet many companies put a huge amount of R&D into designing and manufacturing devices that do not make an audible difference.
The engineers of these femtosecond accurate audio word clock devices, like the Mutec Ref 10 Clock mentioned above, must actually believe it makes a difference or why would they design these units ?
I cannot imagine an architecture meeting where they are proposing specs knowing that this is a multi-thousand dollar waste of time.