DIY hifi source

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
The original image was derived from a .ts file saved on HDD. The 2nd image was derived from the .ts file after it was copied to a Corsair Voyager GT USB 2,0 memory stick powered by a very low noise +5V linear PSU. Both the + and - supply leads (red and black) were disconnected at the USB-A plug at the PC end and a 220 ohm resistor fitted across the + and - pins of the plug . This meant that there was no earth loop due to the screen of the USB memory being connected internally to the black 0Volts lead.
And were both files identical bitwise, if so they were exactly the same.
Ground loops and USB, what difference would that make, most digital layout and signal transfer involves as many ground connections as possible, if you were feading an analogue stage then isolation is best but for digital only transfer it dosn't matter.
 
Marce
MANY C.A. members will tell you that even the type and construction of a USB cable results in audible differences, although according to present theory that shouldn't happen. The better cables usually have greater separation between D+ and D- and the noisy SMPS +5V Vbus line.Many people are getting vastly improved SQ from their DACs etc.by using PSUs such as The iFi iUSB Power Supply
The iFi iUSB Power Supply - Neat Tweak & Upgrade for USB DAC
Note the earth lift provision in the photo posted in the thread.
There are quite a few other products available as well , such as the highly regarded SOtM USB PCI-E low noise +5V power supply which is capable of having it's highly filtered +5V switched off for DACs using their own external PSUs.
 
Jeremy
I did NOT say anything about images coming from the same disc.
The original image was derived from a .ts file saved on HDD. The 2nd image was derived from the .ts file after it was copied to a Corsair Voyager GT USB 2,0 memory stick powered by a very low noise +5V linear PSU. Both the + and - supply leads (red and black) were disconnected at the USB-A plug at the PC end and a 220 ohm resistor fitted across the + and - pins of the plug . This meant that there was no earth loop due to the screen of the USB memory being connected internally to the black 0Volts lead.

Did you see the message that I sent you ?
Alex

you missed the point, when you have a file on a disc, whether it be HD or SSD, its not all stored in a contiguous physical chunk of data, particularly with SSD. Enterprise level SSD (which has trickled down into quality consumer level) stores the data across multiple memory addressees and using wear leveling so that it spreads the wear of write and rewrite cycles evenly, it will be constantly taking from different areas in the memory. so if it sounds different from different sectors of a disc, then if I played a 1khz tone, it wouldnt be able to hold a tone, which is ridiculous.

only just saw your message now, will reply tomorrow
 
Last edited:
As fun as all this is, it would come to a screeching halt the moment anyone throws down a three word gauntlet:
Blind listening test.
In one corner the best Cookie Gold CD.
In the other, the cheapest available generic CDR media with the info from that CD after being sent from Marcie's to Sy's to Powerpan's to gk7's mailbox and whoever else wants in, and let them copy it back and forth to and fro any and all lossless formats that strikes their fancy. We can make it go around the world a few times. Maybe the Van Allen belt will come into play. Anyone live near Leh Ladakh? That should do something.

Finally, our tired, world-weary bits would get checked in their final resting place by a three person, bi-partisan panel for errors, finding none, it and the virgin original would go in identical CD players good enough to read them with practically no errors wired to the same amp and speakers, as unforgiving as possible. Gk7 rigs an A-B switch. Leaves the room leaving no hint. Enter all.

Know who the winner would be? It will be ...


Wait for it...


Drum Roll..........................................................
 
So when I buy a high resolution audio file the sound quality will be
completely destroyed once I have it downloaded, right ?

Imagine it has to pass miles and miles of cable, routers switches ...
If there is a proxy, even worse, most of them don´t have a real time OS...

It might have the same checksum but this means nothing... the same
way as 2 + 2 = 5 for large values of 2, right ?

Sorry for the sarcasm, but such irrational claims provoke it.

@SY I think we should rather go for the drink I owe you, than wasting time
with this. ;-)

first of all, please make things clear, network playback means play files in local network store in NAS, local playback means play files in USB disk locally.

download will have bit perfect by tcp/ip protocol.
 
One of the things I love/hate about DIYaudio forum is the ease with which professional audio people, including manufacturers, can be provoked into proudly exposing their astonishing ignorance about the technology which provides their living! Even funnier/sadder, their fans sometimes chip in and agree with them!

When I was young I assumed that people who made and sold things were smarter than me. Then I got a job in industry and realised that I too could work at a 'world class' level with bright colleagues; I was the equal of others. Now I find that, as an audio amateur, I seem to know more than some of the professionals.

As I said, same bits plus same timing means same sound. The bits are guaranteed by modern IT systems (apart from random cosmic rays!!). The timing ought to be guaranteed by any serious audio system - we certainly don't want our digital audio sample timing to be reliant on variable things like interrupt latency in a non real-time system like a Microsoft OS. Where the bits came from is completely irrelevant. Anyone who thinks otherwise is just ignorant and/or gullible.
 
One of the things I love/hate about DIYaudio forum is the ease with which professional audio people, including manufacturers, can be provoked into proudly exposing their astonishing ignorance about the technology which provides their living! Even funnier/sadder, their fans sometimes chip in and agree with them!

When I was young I assumed that people who made and sold things were smarter than me. Then I got a job in industry and realised that I too could work at a 'world class' level with bright colleagues; I was the equal of others. Now I find that, as an audio amateur, I seem to know more than some of the professionals.

As I said, same bits plus same timing means same sound. The bits are guaranteed by modern IT systems (apart from random cosmic rays!!). The timing ought to be guaranteed by any serious audio system - we certainly don't want our digital audio sample timing to be reliant on variable things like interrupt latency in a non real-time system like a Microsoft OS. Where the bits came from is completely irrelevant. Anyone who thinks otherwise is just ignorant and/or gullible.

HIFI manufacturers sometimes do mislead people, studio equipments won't, we use professional studio equipments as the reference, that's all.
 
I remember the old days when 1's were an honest five volt signal (more or less). How can anyone expect this newfangled 1.5V memory to sound good when it's not ones and zeros anymore, it's more like 1/3s and 0s???? TTL forever! ;)

Sorry - this thread has degenerated into silliness so I figured I'd just pile more on.
 
Studio equipment manufacturers are not entirely free from audiophile nonsense. You can buy stuff with a few valves sprinkled in (probably badly biased) to add 'tube warmth'.

By 'professional audio people' I did not mean people who make or use studio equipment but anyone who makes money from any aspect of audio: manufacturers, retailers, authors, journalists, consultants.
 
ahh yes that old technique, calling in armies of anonymous professionals to aid your rapidly fading credibility. sorry but said professional designers would be finding this as ridiculous as us, perhaps moreso.

how do you suppose this works properly to produce movie sound tracks for Hans Zimmer?

do remember, we are not talking about a room full of spdif devices, some on wordclock, some on USB, some on Spdif and synchronization issues, even that would again be erroneous application, since the above is quite possible
 

Attachments

  • Hans studio wires.jpg
    Hans studio wires.jpg
    489.4 KB · Views: 143
Last edited:
And the Objective side doesn't suffer from "expectation bias" ?
Pigs might fly!!! They are far more likely to be affected by expectation bias due to their training. Many believe that they can look at their test equipment, and know EXACTLY how something is going to sound. :p
The scientists believe in "magic" too, you know, ;). Get an impressive looking test instrument, add a stick on label that overlays whatever brand is there, that proclaims AP, Tektronix, HP, etc, and pixie dust is in the air. The wise have spoken when a result is flashed up, the measurements are indisputable, "truth" is in the room ...

Of course, there is no such thing as a faulty bit of machinery, or poor testing technique; being in the cathedral and close to the high priests always guarantees that all is well ... ;)

Frank
 
Last edited:
...
Of course, there is no such thing as a faulty bit of machinery, or poor testing technique; being in the cathedral and close to the high priests always guarantees that all is well ... ;)

Frank

Frank you are mixing up physical properties with mathematics. If eg. two files
have the same checksum they are identical by definition. You don´t have
to "prove" or "measure" (with limited resolution) that 1+1=2.
 
SandyK, the USB cable has been discussed before in the past, possible sound differences could be down to noise transfered from one system to another, bit transfer should be correct if the cable is engineered for the job, I think that discussion is and has been looked at on other threads.
Powerpan, I asked what benefits you think teflon PCBs are going to provide over FR4 based PCBs for these frequencies, and are there any other aspects of PCB design in realtion to digital signals you think will affect sound?

Hmmm ... case A: really crap, low end power supply feeding DAC, no RFI issues considered - case B: superbly engineered interference resistance -- glad to know those same bits are going to protect me from any sound issues ...
Same a SandyK and USB cables you are bringing two factors to the argument, noise is a different problem to bit transfer. Your fist point of a low end power supply! why If you want to engineer the best repreduction would you use a crap power supply, that is bad engineering as the power supply is the most critical part of the system. Secondly how much of a problem do you think RF is and how much comes from your power supply, it is a sweeping statement again often used, but with no measured data, and when it comes to RF you have to measure so you know whether you have a problem, what the frequency range is if there is a problem and finaly how to engineer a solution.
Generaly people who use measurements, physics and good engineering practice dont generaly believe in pixie dust, we have hard core audiophiles for that:)
Ying and Yang
Marc
Fas42, two points, you cannot design any equipement without measurments and the tools to perform those measurements, yes you have to listen if its audio based, look if it is picture based etc, but contary to what some gurus would have us believe, you cant design by ear alone. I think some loose touch with the fact that audio equipement designs is like any other branch of electronics it is engineering, some seem to forget this (or like to create a mystique) thinking they are a Stradivari creating an instrument, they are not they are engineers (you would hope, though wonder some times) creating some electronics to provide the best fidelity in playback equipement.
 
You listen to the result to see if there is some large factor you havent measured for, or have measured incorrectly; that is audibly abhorrent. you also listen to see if the compromises you have made were the right ones, or to see if you need to change the focus of your measurements. you dont listen to see if there is something you have measured successfully to not be there that you can hear.

then also you can listen to relax and be happy with what you have created.

all of that is off topic completely and bares no relation whatsoever to bit perfectness, if its bit-perfect, its bit-perfect, there is no grey area... you dont have 'a little bit-perfect' :D
 
Frank you are mixing up physical properties with mathematics. If eg. two files
have the same checksum they are identical by definition. You don´t have
to "prove" or "measure" (with limited resolution) that 1+1=2.
They are identical within themselves, but it's impossible for them to be identical in spatial and temporal senses, otherwise they are the one and same file in every sense.

In other words, the fact that they are stored in different locations means that there is a different pattern of electrical activity when they are replayed. And the fact that one is played after the other "changes things". This might seem all terribly subtle, but I've dealt with headaches along these lines over many years; the physics of everything does come into it, unfortunately ...

Frank
 
True:D
It is also rare these days to have any problems with data, even with complex 3D models in Solidworks, 30M size photos, large PCB files, and often having more than one CAD package open at a time, I cant remember when I last had a crash or any file go wrong! And audio data is not that taxing for a PC these days, the technology has moved on in leaps and bounds since the conception of red book audio.
One reason why so many like to use the smoke screen that digital is analogue, it allows analogue problems to attributed to digital signals, to give rise to incorrect beliefs that noise can be transfered and recorded onto a digital signal as it passes throughy the replay network.:( Low level broadband noise!
 
lets say for arguments sake that you are correct, it still has nothing to do with the file, nothing at all; it would have everything to do with the playback mechanism.

I repeat, it has nothing at all to do with the file, or 'energy' stored within the file, just as the 'energy' that transmits the file does not directly relate to the playback of the file. what matters to the playback is the data represented by the file, which is identical in both cases. the only possible difference would be the impossibly small subatomic difference in the playback mechanism at the 2 different times

but can we leave out the metaphysical codswhollop? its crap like this that quite rightly makes reasonable people think audiophiles are insane, as more problems are solved with the technology and diagnostics allow us to see more and more clearly into the mechanisms at play, more and more dubious and elaborate excuses are made for there being differences where there are none.
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.