Computer Audio Asylum

Music servers and other computer based digital audio technologies.

Return to Computer Audio Asylum

Message Sort: Post Order or Asylum Reverse Threaded

Do Hard Disk Drives (HDDs) add timing distortions to music?

71.146.8.173
Posted on September 7, 2014 at 22:32:09
Jaundiced Ear
 HDDs are mechanical devices. They have physically spinning platters and physically moving read/write heads.

I know regular turntables have "wow," or slow speed timing variations. Do hard drives have something analogous?

I know tape decks have "flutter," a high speed timing variation caused by friction between the tape and the tape heads. Could the vibration of the read/write head in an HDD cause something similar to this?

I have to admit all of my musical archives are on HDDs. Perhaps this is why I am unable to hear the things folks report here? What are the recommended HDD brands? Or should I instead move my music to Solid State Drives? In which case too, what are the recommend brands?

JE

 

Hide full thread outline!
    ...
RE: Do Hard Disk Drives (HDDs) add timing distortions to music?, posted on September 8, 2014 at 00:10:51
fmak
Audiophile

Posts: 13158
Location: Kent
Joined: June 1, 2002
 Hard drives typically have access times of up to 20mS. I have had brand new ones that exceed this signicantly. In constrast SSD have access times one or two orders of magnitude less.

Hard drives, when accessed, feed back large impulsive noise due to fast current demands into the computer voltage rails.

Both SSDs and spinning HDDs need to be powered independently for better audio performance. In addition, both the Sata and Power cables emit noise that can be picked up by other components. I have seen Sata cables that are not shielded properly. Large spinning HDDs may also generate significant magnetic fields around them.

Effect on sound quality? - variable and who knows unless the effects can be separated and assessed. They are likely to be different from system to system.

 

RE: Do Hard Disk Drives (HDDs) add timing distortions to music?, posted on September 8, 2014 at 09:21:06
Presto
Audiophile

Posts: 5957
Location: Canada
Joined: November 10, 2004
 I don't profess to be an electronics engineer or circuit design guru but conventional wisdom says that since a digital signal is clocked, you're not going to get any "wow and flutter". But, of course, you can have time-related issues with digital such as jitter and various kinds of waveform distorion ("actual" square waves look more like truncated triangle waves on a scope...)

Theory says that a buffer would handle the anomalies of timing issues with the data coming into the buffer, but from what I've read it all depends on the clocking scheme.

Cheers,
Presto

 

RE: Do Hard Disk Drives (HDDs) add timing distortions to music?, posted on September 8, 2014 at 10:16:30
Roseval
Audiophile

Posts: 1846
Joined: March 31, 2008
 As HDs are analog devices they will have wow, flutter, vibrations (read a CD at speed 40 and you know what I mean).
But does it matter?
I don’t think so.
As much as a HD delivers the data faithfully e.g. a word document to a program, as much it will do with any file including the audio files.
Inside the digital domain jitter is simply not an issue, we are transferring data and this is all ways buffered hence not real time stream.

The most likely explanation is that all electronic components generated some dirt.
Be it EMI, RFI or a ripple on the ground plane.
If this creeps into the DAC e.g. the ground plane is modulated hence the reference of the analog out, it will affect sound quality.

Basically it is about:
silencing the source
isolating the DAC ( async protocol, galvanic isolation)
both

Sometimes the impact could be demonstrated by measurements
An examples can be found here: http://thewelltemperedcomputer.com/Intro/SQ/SoftwareInducedJitter.htm
I do think this is typical for a badly designed on board audio.
Archimago couldn’t measure diffrences.

Jitter level of a Oppo when playing and stopped
http://thewelltemperedcomputer.com/Intro/SQ/Toslink_Coax.htm

I know swapping a HD for a SSD has been reported as an improvement in SQ.
But before you spend your money on it I can’t resist to cite some very wise words with sums up my feeling about all these tweaks nicely.

I used to try out all sorts of tweaks and other nonsense such as making my own cables. I was always thrilled when I first plugged in the new goody (it almost always "improved" the sound), but I always noticed that "new sound" effect didn't last more than a few days. Indeed, often after a few days I was glad to pull the tweak out of the system and go back to the way it was. Which ironically also "improved" the sound.

Best

Vincent

The Well Tempered Computer

 

No, posted on September 8, 2014 at 14:34:01
AbeCollins
Audiophile

Posts: 46302
Location: USA
Joined: June 22, 2001
Contributor
  Since: February 2, 2002
 HDDs are mechanical devices. They have physically spinning platters and physically moving read/write heads. I know regular turntables have "wow," or slow speed timing variations. Do hard drives have something analogous? I know tape decks have "flutter," a high speed timing variation caused by friction between the tape and the tape heads. Could the vibration of the read/write head in an HDD cause something similar to this?

No. Even though disks are electro-mechanical in nature and may have ever so slight rotational variation they are not like LP's and tape decks where the 'music' is being 'played' in an analog fashion off the media. If this were the case, files 'recorded' to a 5400 rpm disk would sound like Alvin and the Chipmunks if played off a 7200 rpm disk. ;-)

In a correctly functioning computer, files are being read bit perfect and in this case bits ARE bits. If they weren't bit perfect the disk controller or file system would either correct the bit or report errors. Plus, the data is being cached in the onboard HDD cache and in the operating system buffers. So they aren't really being played directly off of disk anyway. And then there are players that support 'memory play' where the music file(s) are entirely cached in system DRAM.

I have to admit all of my musical archives are on HDDs. Perhaps this is why I am unable to hear the things folks report here? What are the recommended HDD brands? Or should I instead move my music to Solid State Drives? In which case too, what are the recommend brands?

Doubtful but try an SSD for yourself and see if it improves your music. In my case I heard a very slight difference with the SSD but it was no better than the HDD, just different. Most folks here have their music stored on spinning disks especially if they have a large collection.

There are only 3 (spinning rust) disk drive manufacturers left: Seagate, Western Digital, and Toshiba. All are very good, all have had problems, and all disks are susceptible to data corruption and failure. Pick your poison and make sure to archive & backup regularly.

For SSD my choice would be Intel or Samsung, and there's no good reason or benefit to choose horrendously expensive SLC SSD for a music player.


 

So, posted on September 8, 2014 at 15:02:02
fmak
Audiophile

Posts: 13158
Location: Kent
Joined: June 1, 2002
 a number of inmates are saying that an access time of 20mS (and varying) makes no difference in the final outcome of a time variable audio signal outputted from a group of chips in the signal path, as opposed to access times of 0.15 to O.2 mS. These chips also have inbuilt time constants as well.

Does this sound reasoned and reasonable?

 

Thanks for the replies, all!, posted on September 8, 2014 at 17:37:44
Jaundiced Ear
 fmak: Thanks for responding. I've assembled my own PCs for years now, but I don't ever recollect seeing a case or motherboard set up for more than one power supply. How would I get separate power to my hard drives? And to be honest? very little of the stuff I read here sounds reasoned or reasonable to me which is why I keep asking questions!

AbeCollins: Thanks for responding. Point taken about data drives differing from analog drives. I seem to remember wondering about the effects of helium on a drive. I guess that's the same principle.

"In a correctly functioning computer, files are being read bit perfect and in this case bits ARE bits." Fair enough. But where in the system do the bits start being something else besides bits? I always thought the entire point of digital audio is that it is more robust than analog audio and that tiny changes to the bits shouldn't matter as the intrinsic "bitness" or on/off nature of the signal would overwhelm all but the grossest changes.

Vincent: Thanks for responding. Whoa! I'm so used to getting beat up for the things I post that I never dreamed I'd see myself quoted back to me with approval, especially by a guy who runs as cool a site as yours. Thank you so much!

By the way, I love your site. It was incredibly helpful to me when I was first getting started out in computer audio. Please keep up the good work!

Presto: Thanks for responding. I'm curious about the waveform distortions you mention: ""actual" square waves look more like truncated triangle waves on a scope..." Could you point me to any sites that have examples of this? Thanks!

All the Best!

JE

 

RE: Thanks for the replies, all!-HDD PS, posted on September 8, 2014 at 23:58:45
fmak
Audiophile

Posts: 13158
Location: Kent
Joined: June 1, 2002
 Just disconnect the sata power plug and connect 5V or 5V+12V high quality dc supply Info here:

http://www.computeraudiophile.com/f10-music-servers/powering-solid-state-drive-17170/

For spinners make sure that the current available is adequate

 

RE: Thanks for the replies, all!, posted on September 9, 2014 at 00:34:24
Roseval
Audiophile

Posts: 1846
Joined: March 31, 2008
 Thanks for the kind words about my website.

I always thought the entire point of digital audio is that it is more robust than analog audio and that tiny changes to the bits shouldn't matter as the intrinsic "bitness" or on/off nature of the signal would overwhelm all but the grossest changes.

Completely true.
The big difference between analog and digital is of course that both are analog signals (we don’t have digital electrons). In case of analog signal transmission we use the absolute value of the signal hence any distortion is one and no way to reconstruct the original signal.
In case of digital we use the rise and fall of the signal hence as long as we can detect the zero crossing properly we can reconstruct our block pulse.

This explains why when transmitting “values” digital is far more robust than analog.
This explains why our bits are in general de right ones, indeed bits are bits.

What happens if we start to do something silly like using the send rate of the signal as a piece of information (indeed SPDIF, it uses the send rate for sample rate).
Now all of a sudden we are back in the analog world.
Our sample rate is as accurate as the clock of the sender.
Any deviation (no clock is perfect) will be a change in sample rate
Any distortion of the transmitted signal will result in a small shift in time of the zero crossing.
Yep we are talking jitter, rare on audio forums :)

Basically, our bits are perfect and our timing imperfect by design.


Vincent
The Well Tempered Computer

 

Ir sounds reasonable to those who understand how computers work., posted on September 9, 2014 at 07:31:08
Tony Lauck
Audiophile

Posts: 13629
Location: Vermont
Joined: November 12, 2007
Tony Lauck

"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar

 

RE: Ir sounds reasonable to those who understand how computers work., posted on September 9, 2014 at 08:47:10
fmak
Audiophile

Posts: 13158
Location: Kent
Joined: June 1, 2002
 some of those who claim that they know how computers work have no concept of time dependency of events including audio replay - a 2D picture of the process, in fact.

In this myopic world everything is expressed in one dimensional bits?

 

Thanks for the link! nt, posted on September 9, 2014 at 18:58:08
Jaundiced Ear
 nt

 

RE: Thanks for the replies, all!, posted on September 9, 2014 at 19:06:16
Jaundiced Ear
 "Thanks for the kind words about my website."

Hey, my pleasure! Thanks for the great website!

JE

 

RE: Thanks for the replies, all!, posted on September 10, 2014 at 12:07:43
AbeCollins
Audiophile

Posts: 46302
Location: USA
Joined: June 22, 2001
Contributor
  Since: February 2, 2002
""In a correctly functioning computer, files are being read bit perfect and in this case bits ARE bits." Fair enough. But where in the system do the bits start being something else besides bits?"

Most folks believe it is at the interface from the computer (USB, Toslink, SPDIF) to the DAC in the form of jitter.

 

So jittery bits..., posted on September 10, 2014 at 18:00:19
Jaundiced Ear
 turn in to jitter bugs?

(Sorry, I couldn't resist.)

JE

 

RE: So jittery bits..., posted on September 10, 2014 at 18:34:12
AbeCollins
Audiophile

Posts: 46302
Location: USA
Joined: June 22, 2001
Contributor
  Since: February 2, 2002
 Apparently so, but mostly in certain crippled systems. I have not experienced jittery bits degrade to jittery bugs. ;-)



 

RE: So jittery bits..., posted on September 10, 2014 at 18:56:55
Jaundiced Ear
 I just had visions of the bits in my system coming to life and having themselves a good old time!

JE

 

RE: So jittery bits..., posted on September 11, 2014 at 06:36:31
b.l.zeebub
Audiophile

Posts: 9361
Location: 52deg 28'N,1deg56'W
Joined: April 17, 2006
 You can always use the jitter bugs to feed the lizards.

 

Page processed in 0.023 seconds.

[ Top Top of Page ] [ Contact Us ] [ Support/Wish List ] [ Copyright © 2024, Audio Asylum, all rights reserved ]
[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]