Jitter problem

[mansr]

Ok, I'll bite. So, what am I getting when I buy a hard drive, a thumb drive, RAM, etc? What do they store if not bits?

 

Let's also remember that bits, and all digital, are an abstraction. Bits are an encoding scheme devised by humans, not entites that exist in nature. However, certain naturally occurring materials have properties that lend themselves to the storage of two-state, binary bits in the design and engineering of digital systems.

 

A major example is magnetism, which can be polarized in only one of two possible ways. So, if you look closely at the magnetic domains on a hard drive, their polarization is unambiguously interpretable as either a zero or a one, nothing else, based on consistent standards established for the digital storage system. (Note that it is very difficult to completely "demagnetize" a previously magnetized material. "Degaussing" to "erase" magnetic domains and bit encoding is just really randomly scrambling the polarities of the magnetic domains in an externally applied, high intensity, variable magnetic field. It does not turn N-S or S-N polarities to a state of no magnetism at the magnetic domain level.)

 

An older, simpler example is the punched "IBM" card - remember those? The card was specified as having a format consisting of 80 columns and 12 rows. At each of those locations on the card, if it was punched out = a small rectangular hole, that signified a one. If it was not punched through at that location, that signified a zero. Those zero/one values could be sensed mechanically or optically or even visually. As long as the card was aligned properly in punching and reading operations, the individual binary bit values were clear and unambiguous.

 

The list of digital systems carefully engineered for the reliable, persistent, unambiguous storage of bits could go on and on.

 

Let me add to this that although there will always be some deviation from the ideal in any instance of a stored bit (magnetic domains not perfectly aligned, holes punched slightly off-centre, etc.), the retrieval mechanism still returns exactly a one or a zero. The stored bits are either retrieved correctly or they are not. That the entire damn point of digital systems.

[alfe]

Ok, I'll bite. So, what am I getting when I buy a hard drive, a thumb drive, RAM, etc? What do they store if not bits?

 

Let's also remember that bits, and all digital, are an abstraction. Bits are an encoding scheme devised by humans, not entites that exist in nature. However, certain naturally occurring materials have properties that lend themselves to the storage of two-state, binary bits in the design and engineering of digital systems.

 

A major example is magnetism, which can be polarized in only one of two possible ways. So, if you look closely at the magnetic domains on a hard drive, their polarization is unambiguously interpretable as either a zero or a one, nothing else, based on consistent standards established for the digital storage system. (Note that it is very difficult to completely "demagnetize" a previously magnetized material. "Degaussing" to "erase" magnetic domains and bit encoding is just really randomly scrambling the polarities of the magnetic domains in an externally applied, high intensity, variable magnetic field. It does not turn N-S or S-N polarities to a state of no magnetism at the magnetic domain level.)

 

An older, simpler example is the punched "IBM" card - remember those? The card was specified as having a format consisting of 80 columns and 12 rows. At each of those locations on the card, if it was punched out = a small rectangular hole, that signified a one. If it was not punched through at that location, that signified a zero. Those zero/one values could be sensed mechanically or optically or even visually. As long as the card was aligned properly in punching and reading operations, the individual binary bit values were clear and unambiguous.

 

The list of digital systems carefully engineered for the reliable, persistent, unambiguous storage of bits could go on and on.

 

If you buy a hard drive this is what you get: http://home.npru.ac.th/piya/Document_KMUTNB/Lec09.pdf

[alfe]

See: https://en.wikipedia.org/wiki/Optoelectronics

 

 

By periodically sampling an analogue signal and translating the measured voltage values to some binary format.

 

 

No.

 

 

That's actually the worst kind of jitter because it gets encoded in the binary format.

 

 

Coming from an ADC, yes. Coming from a storage device or a DAC, no..

 

Wrong! I think with more research you should do better.

[Abtr]

... The list of digital systems carefully engineered for the reliable, persistent, unambiguous storage of bits could go on and on.

Also, a harddisc or SSD uses a memory cache buffer where all data that is read from the device is first copied to. From there it is used by the program/process that requested the data (bit-stream). So even files that are physically stored on different disc sectors are equally copied to the same cache. There cannot be any significant electrical difference between two identical digital files being read from the same storage..

[Jud]

Ceterum censeo Carthaginem esse delendam.

[Abtr]

Quod erat demonstrandum.

[Cornan]

Non loqueris ? Ita vero! ;-) Hurray for Google translate. Always there for you if want to missunderstand something! :-)

[Jud]

Ceterum censeo Carthaginem esse delendam.

 

This wasn't just random Latin. From Wikipedia:

 

[A]ccording to several ancient sources the Roman statesman Cato the Elder frequently if not invariably ended his speeches in the Senate with a variant of this expression even when his speech had been totally unrelated to Roman foreign policy towards Carthage

 

Long threads here frequently if not invariably contain "discussions" of this bits identical/different sound topic, even where the thread originally was totally unrelated to the subject.

 

But I'm sure *this* will be the time we'll have everyone agreeing one way or the other, so please go right ahead....

[Abtr]

This wasn't just random Latin. From Wikipedia:

 

 

 

Long threads here frequently if not invariably contain "discussions" of this bits identical/different sound topic, even where the thread originally was totally unrelated to the subject.

 

But I'm sure *this* will be the time we'll have everyone agreeing one way or the other, so please go right ahead....

Well, personally I'm open to anything, except this 'identical files can sound differently when streamed from the same storage device' argument. And I think it has been proved to be nonsensical..

[alfe]

This wasn't just random Latin. From Wikipedia:

 

 

 

Long threads here frequently if not invariably contain "discussions" of this bits identical/different sound topic, even where the thread originally was totally unrelated to the subject.

 

But I'm sure *this* will be the time we'll have everyone agreeing one way or the other, so please go right ahead....

 

Jud,

 

My friend, I'm just giving some information about jitter and media storage.

And don't forget Cogito ergo sum

[Fitzcaraldo215]

If you buy a hard drive this is what you get: http://home.npru.ac.th/piya/Document_KMUTNB/Lec09.pdf

 

 

Right, it is something that stores and retrieves bits, as I said. Your point?

[sandyk]

Recently I uploaded the last pair of comparison tracks that appeared on the CD that I sent esldude several months ago, and from which the CD-R was burned .

All 3 C.A. recipients had no problems identifying audible differences between the tracks. There was however disagreement on which track was which, with 2 reporting that the cannon at the end of the track sounded not quite right. The 3rd recipient accurately identified it as thunder.

I had neglected to tell them that the English translation of the track was "Thunder and Lightning Polka."

The "explosion" at the end , and during the track, was actually a real thunderclap recorded from indoors according to the booklet.

IIRC, at least 2 of the 3 recipients were using Phasure NOS DACs.

 

Alex

 

(Unter Donner und Blitz Polka, Op. 324 from the Telarc CD "EIN STRAUSSFEST")

[sandyk]

Jud,

 

My friend, I'm just giving some information about jitter and media storage.

And don't forget Cogito ergo sum

 

Jud

What makes you so damn sure that Jitter/timing inaccuracies isn't involved (or reduced Jitter ) as at least part of the explanation for what an increasing number of members are reporting hearing when using Linear PSUs etc ?

The title of this thread is "Jitter Problem"

Alex

[jabbr]

Ok, I'll bite. So, what am I getting when I buy a hard drive, a thumb drive, RAM, etc? What do they store if not bits?

 

Let's also remember that bits, and all digital, are an abstraction. Bits are an encoding scheme devised by humans, not entites that exist in nature. However, certain naturally occurring materials have properties that lend themselves to the storage of two-state, binary bits in the design and engineering of digital systems.

 

A major example is magnetism, which can be polarized in only one of two possible ways. So, if you look closely at the magnetic domains on a hard drive, their polarization is unambiguously interpretable as either a zero or a one, nothing else, based on consistent standards established for the digital storage system. (Note that it is very difficult to completely "demagnetize" a previously magnetized material. "Degaussing" to "erase" magnetic domains and bit encoding is just really randomly scrambling the polarities of the magnetic domains in an externally applied, high intensity, variable magnetic field. It does not turn N-S or S-N polarities to a state of no magnetism at the magnetic domain level.)

 

An older, simpler example is the punched "IBM" card - remember those? The card was specified as having a format consisting of 80 columns and 12 rows. At each of those locations on the card, if it was punched out = a small rectangular hole, that signified a one. If it was not punched through at that location, that signified a zero. Those zero/one values could be sensed mechanically or optically or even visually. As long as the card was aligned properly in punching and reading operations, the individual binary bit values were clear and unambiguous.

 

The list of digital systems carefully engineered for the reliable, persistent, unambiguous storage of bits could go on and on.

 

See, this discussion is about hanging chads.

 

All of the media which transmit bits are ultimately analogue and analogue is ultimately digital though digital is ultimately analogue (probabilistic) ... in any case there is scant discussion of the transfer functions of the "extra bit" signals. Digital circuitry as designed to strongly amplify the bit signals wrt the "non bit" signals to the point where they do not survive.

[davide256]

That's why any half-decent DAC these days has a buffer and a local clock.

 

The problem occurs before the DAC; its what all of us are struggling with in selecting/implementing a high quality source to feed the DAC.

[Jud]

Jud

What makes you so damn sure that Jitter/timing inaccuracies isn't involved (or reduced Jitter ) as at least part of the explanation for what an increasing number of members are reporting hearing when using Linear PSUs etc ?

The title of this thread is "Jitter Problem"

Alex

 

Yes, it was about the OP trying to get his system to sound as good from a computer drive as it does from a spinning disc. If the problem is the quality of the disc spinning system during the rip, why does that disc spinning system sound so good when it's playing?

[sandyk]

Yes, it was about the OP trying to get his system to sound as good from a computer drive as it does from a spinning disc. If the problem is the quality of the disc spinning system during the rip, why does that disc spinning system sound so good when it's playing?

 

Jud

Did the OP rip files to the PC using perhaps coax SPDIF to his PC , or did he rip the files using a computer optical drive, whether internal or external ? It's a P.I.T.A. ripping CDs via a normal CD player at normal playing speed..

When the disc is playing in a CD/DVD Rom it will be at normal playing speed. When ripping the rip speed is likely to be > 10 times that speed. Imagine the vibration etc. of a flimsy piece of polycarbonate spinning around at a huge rate of knots. Even with the LG GGW H20L Blu Ray writer that Alfe designed, at least with mine, it hums quite audibly at higher extraction speeds and this couples through the case of the PC as well.

In any event, I am talking about improved power supplies to not only the Optical drive itself, but improved low noise power to internal SSDs etc. Quite a few members have fed their internal SSDs via low noise voltage regulators, such as the low noise 40uV offering from ebay and reported an improvement in SQ. I would be very surprised if someone who had ripped a CD to a lossless format on a standard Mac Mini to say USB memory stick, or even burned a CD from the ripped files, then ripped the same CD again to the USB memory stick after a JS2 replaced the internal SMPS, didn't find that the more recent copy sounded better. It's not just about a PC with an improved PSU exporting a cleaner output via USB , it's also about the ripped files having an improved signal integrity to start with.

BTW, in case you were wondering who the recipients of the files were, the 3rd listener came from Queensland.

Regards

Alex

[Abtr]

I would be very surprised if someone who had ripped a CD to a lossless format on a standard Mac Mini to say USB memory stick, or even burned a CD from the ripped files, then ripped the same CD again to the USB memory stick after a JS2 replaced the internal SMPS, didn't find that the more recent copy sounded better.

Well, I'm away for a few days, but I'll try that. Still, it would not be the same as hearing a difference between e.g. files copied to a memory stick with and without a Regen in the USB path (I may try that too).

 

BTW, the Regen seems to (largely) cure audible USB problems at the DAC end of the USB chain (obviously) without changing any bits. So the problem must be analogue noise and digital timing errors i.e. not the files themselves.

[sandyk]

Well, I'm away for a few days, but I'll try that. Still, it would not be the same as hearing a difference between e.g. files copied to a memory stick with and without a Regen in the USB path (I may try that too).

 

That test is best done into the USB input of a decent media player unless you are able to play from the USB memory stick without also using USB out to a DAC.

Gordon Rankin recommends not to use 2 lots of USB at the same time.

It's weird how the file saved without the Regen in line doesn't seem to really improve when played with the Regen in line. but the file saved with the Regen in

line further improves.

You will need something better than the supplied Meanwell SMPS to power the Regen though.

[alfe]

Yes, it was about the OP trying to get his system to sound as good from a computer drive as it does from a spinning disc. If the problem is the quality of the disc spinning system during the rip, why does that disc spinning system sound so good when it's playing?

 

Did you read my questions? my links ? OEIC, ADC,PLL are the answers to the OP question.

I'm not trolling Jud, I was just trying to help with my experience of designing storage buckets for bits.

[Jud]

Did you read my questions? my links ? OEIC, ADC,PLL are the answers to the OP question.

I'm not trolling Jud, I was just trying to help with my experience of designing storage buckets for bits.

 

Hi alfe. I didn't have time yet. But I always learn from your links, so I will. Meanwhile I have partially accomplished my "secret mission" to get people out of the rut of uselessly going through the same old debate in the same old way and move them on to considering fresh information like what you have brought up.

[sandyk]

Hi alfe. I didn't have time yet. But I always learn from your links, so I will. Meanwhile I have partially accomplished my "secret mission" to get people out of the rut of uselessly going through the same old debate in the same old way and move them on to considering fresh information like what you have brought up.

 

Jud

If Alfe hadn't returned to posting here again where would you have obtained this additional information ?

It comes across to me, more as you not wishing to see further discussion on this issue, partly because you are sick of it, and partly because you appear to refuse to accept what a few of us are reporting, and a couple of high profile and well respected members have also confirmed with the comparison files that I uploaded recently. I am not prepared to name and drag them into this silly debate, but I am happy to let you know privately if you wish to verify this with them.

My apologies if I have misread your posts on this issue.

Alex

[mansr]

Jud

If Alfe hadn't returned to posting here again where would you have obtained this additional information ?

 

It comes across to me, more as you not wishing to see further discussion on this issue, partly because you are sick of it, and partly because you appear to refuse to accept what a few of us are reporting, and a couple of high profile and well respected members have also confirmed with the comparison files that I uploaded recently. I am not prepared to name and drag them into this silly debate, but I am happy to let you know privately if you wish to verify this with them.

My apologies if I have misread your posts on this issue.

Alex

 

How about you provide everybody with those files?

[sandyk]

How about you provide everybody with those files?

 

The files have now expired, and I have no intention of providing them to members with such an exhibited high amount of Expectation Bias/mediocre equipment , that they wouldn't have a snow flake's chance in hell of hearing the differences.

I have already provided a couple of comparison CDs to Dennis to pass around among those with suitable equipment and who don't have a welded shut brain on this issue. I get the distinct impression however, that's it's far easier to shoot your mouth off in a forum than actually listen and pass them on to the next recipient. You would need to get off your arses to do that !

Alfe is a very clear exception here, as he has done much further research, and I greatly appreciate his assistance.

[alfe]

deleted