Best External Hard Drive for music storage.

[Bob Stern]

If you want to minimize fan noise, or even mount the drive in a silent, fan-less case, low current draw (low heat dissipation) will be a consideration. WD Red had lower current draw than other drives I found when I researched this a year ago.

[chauphuong]

Do you folks think it is any useful to use linear PS for NAS and external drives?

[alsterfan]

Do you folks think it is any useful to use linear PS for NAS and external drives?

For years a Seagate Backup Plus 4 TB was my choice. Though it was reliable and nearly noiseless I always thought that using USB to feed it with 5 V wasn´t a good idea. As Alex´s JS-2 has two DC outputs I bought a LaCie 3,5´´ (USB B) drive to feed it with 12 V. I was surprised how much better the SQ was.

I then displaced the standard USB cable by a decent Supra USB 2.0 cable showing how big an impact the cable architecture can have on SQ. For the time being I´m very pleased to use a Oyaide Continental 5S that easily topped the Supra cable, revealing more details without loosing its musical flow.

So far so good. Inside of the LaCie HDD a Seagate 3,5´´ platter was doing its job but it was too noisy for my ears. So I bought a Western Digital My Book and put the WD Blue 5TB HDD, that was inside, into an external case (USB B). This is a really quiet solution as the Blue line (the Green line no longer is offered) rotates with 5800 rpm only. It is for me for the first time that

a 3,5´´ HDD can compete with its smaller counterpart (2,5´´) noisewise.

At first I was a bit disappointed showing a bass without its contour that I had liked. But about 10 hrs of burning in (using a cascade mp3 white noise file) the soundstage became deeper and so was the bass fundament without loosing its contour.

So for me it was worthwhile to displace a 2,5´´ HDD by a 3,5´´ platter, using the advantage of feeding it with 12V by an external linear power supply and employing an "audiophile" USB cable.

Regards,

Uwe

[ElviaCaprice]

In my experience with USB streaming, cables make a difference when there is some electrical ground loop/AC noise being transmitted by the cable. Get rid of those interferences and the cable shouldn't matter as much other than impedance.

Question, has anyone tried to eliminate the 5Vbus and/or ground/shield from the HDD to the computer? Wondering if the ground/shield can be cut after handshake? Or would the ground be necessary for constant back and forth communication with the HDD? What we are trying to do here is eliminate any AC contamination of the Mobo.

[YashN]

Question, has anyone tried to eliminate the 5Vbus and/or ground/shield from the HDD to the computer? Wondering if the ground/shield can be cut after handshake? Or would the ground be necessary for constant back and forth communication with the HDD? What we are trying to do here is eliminate any AC contamination of the Mobo.

I have the means to try it here, and it is planned, unless someone comes forth saying this should absolutely be avoided for HDDs.

 

Currently enhancing my Linear Regulated Power Supply though.

[plissken]

In my experience with USB streaming, cables make a difference when there is some electrical ground loop/AC noise being transmitted by the cable. Get rid of those interferences and the cable shouldn't matter as much other than impedance.

Question, has anyone tried to eliminate the 5Vbus and/or ground/shield from the HDD to the computer? Wondering if the ground/shield can be cut after handshake? Or would the ground be necessary for constant back and forth communication with the HDD? What we are trying to do here is eliminate any AC contamination of the Mobo.

 

Why not go with a NAS (Toshiba even makes a 3TB unit for ~$70) that goes into a wireless access point and then you are free of any 5Vbus or grounding issues.

 

For a customer I did this NAS, a D-Link AC router and it was less than $100 all in. This way the electronics can be totally housed elsewhere.

 

My wireless setup gets 50MB/s and for 24/192 it's like not breaking a sweat.

[ElviaCaprice]

Why not go with a NAS (Toshiba even makes a 3TB unit for ~$70) that goes into a wireless access point and then you are free of any 5Vbus or grounding issues.

 

For a customer I did this NAS, a D-Link AC router and it was less than $100 all in. This way the electronics can be totally housed elsewhere.

 

My wireless setup gets 50MB/s and for 24/192 it's like not breaking a sweat.

 

Wireless would eliminate any contamination from the HDD's, but now you introduced further electrical needs at the mobo to receive that wireless signal and also create further OS operations that in turn can effect SQ. No matter what we do, moving data will create OS operations, guess it's just a question of which create the lest impact to SQ. Is worthy of a test. Might be hard to pinpoint SQ differences, but should be easy to see how much it impacts OS operations.

[plissken]

Wireless would eliminate any contamination from the HDD's, but now you introduced further electrical needs at the mobo to receive that wireless signal and also create further OS operations that in turn can effect SQ. No matter what we do, moving data will create OS operations, guess it's just a question of which create the lest impact to SQ. Is worthy of a test. Might be hard to pinpoint SQ differences, but should be easy to see how much it impacts OS operations.

 

Since both wired and wireless are all buffered and not a zero copy stack. That along with a decent DAC that is going to isolate external noise. I doubt there is going to be a difference detectable once sighted bias is accounted and controlled for.

[ElviaCaprice]

Since both wired and wireless are all buffered and not a zero copy stack. That along with a decent DAC that is going to isolate external noise. I doubt there is going to be a difference detectable once sighted bias is accounted and controlled for.

 

 

As duly noted, Plissken, your in the strict technical aspect of digital applications. But for us, listening observers comes first camp, science has not kept up with the discrepancies observed, in this case heard. So we continue on with this experiment.

[plissken]

As duly noted, Plissken, your in the strict technical aspect of digital applications. But for us, listening observers comes first camp, science has not kept up with the discrepancies observed, in this case heard. So we continue on with this experiment.

 

Science, measurements, math, have made your entire audio world possible.

 

I'm all for discrepancies heard but it needs to be more than the sighted variety.

[ElviaCaprice]

Science, measurements, math, have made your entire audio world possible.

And here I thought it was observation first?? Which comes first the chicken or the egg?

[plissken]

And here I thought it was observation first?? Which comes first the chicken or the egg?

 

 

In what regard specifically? If we are talking say TCP/IP the observation was "I need to get this bit of information to a remote host". The rest was math (DUAL, Belleman-Ford, Dijkstra etc).

[YashN]

Wireless would eliminate any contamination from the HDD's, but now you introduced further electrical needs at the mobo to receive that wireless signal and also create further OS operations that in turn can effect SQ.

 

And the wireless circuit near the D/A process has its own inherent noise profile as well.

[plissken]

And the wireless circuit near the D/A process has its own inherent noise profile as well.

 

What? Where is this wireless circuit at? The D2A could be happening 6 feet away.

[YashN]

What? Where is this wireless circuit at? The D2A could be happening 6 feet away.

'Near the DAC' was as opposed to the other end, i.e. near the sending device.

[plissken]

'Near the DAC' was as opposed to the other end, i.e. near the sending device.

 

Any technical papers that speak to this profile of noise?

[YashN]

Any technical papers that speak to this profile of noise?

Read John Swenson's 3-part interview on Audiostream.

 

I wonder how technical papers will help you: you didn't find anything wrong in Amir's "measurements"...

[plissken]

Read John Swenson's 3-part interview on Audiostream.

 

I wonder how technical papers will help you: you didn't find anything wrong in Amir's "measurements"...

 

I'm curious how BenchMark was able to run a 100 foot cable for their DAC, with just horrendous amounts of noise and yet it was able to 100% extract pristine audio.

 

Which of Amir's measurements? I hope they aren't any where 3-4 other EE's were able to replicate them in their entirety and form the same conclusions.

[YashN]

I'm curious how BenchMark was able to run a 100 foot cable for their DAC, with just horrendous amounts of noise and yet it was able to 100% extract pristine audio.

 

They probably did a great job on isolation. Are you assuming all DACs are built the same?

 

Which of Amir's measurements? I hope they aren't any where 3-4 other EE's were able to replicate them in their entirety and form the same conclusions.

 

You have a selective or defective memory - Amir messed up his measurements completely and since you aren't competent in that field, you didn't find fault in them contrary to people who do have proper experience in the relevant fields. Besides, what are your credentials to gauge anything having to do with Engineering? Are you an Engineer yourself?

[plissken]

They probably did a great job on isolation. Are you assuming all DACs are built the same?

 

 

 

You have a selective or defective memory - Amir messed up his measurements completely and since you aren't competent in that field, you didn't find fault in them contrary to people who do have proper experience in the relevant fields. Besides, what are your credentials to gauge anything having to do with Engineering? Are you an Engineer yourself?

 

Again. Which of Amir's measurements are you referring to?

 

Which DAC's do you feel will exhibit odd behavior if the computer they are attached to is feeding it via a wireless vs wired connection.

 

I have a feeling if I propose a $99 DAC you will say it's not 'High Resolution' enough. If I propose a $2000 DAC then it will be 'well designed' and therefore not susceptible.

[YashN]

Again. Which of Amir's measurements are you referring to?

I'll let you find your own posts about it on this very forum. We had this conversation before on this very subject... The rest is useless speculation and side-tracking.

[rando]

Thus having barely tasted a trickle of blood, the saber was abruptly sheathed until a more imminent threat arrives.

 

/thread

[plissken]

I'll let you find your own posts about it on this very forum. We had this conversation before on this very subject... The rest is useless speculation and side-tracking.

 

So any idea on what DAC's are going to exhibit 'sonic anomalies' when fed with a computer that is connected wireless vs wired?

 

I seem to remember a conversation where I proved you wrong because you were using a certain persons (JK) post as some form of bizarre support for your howling at the moon.

[YashN]

So any idea on what DAC's are going to exhibit 'sonic anomalies' when fed with a computer that is connected wireless vs wired?

 

I seem to remember a conversation where I proved you wrong because you were using a certain persons (JK) post as some form of bizarre support for your howling at the moon.

You haven't proved anything, especially not that you understood the flaws in Amir's measurements or that you have any Engineering credentials at all.

[plissken]

Read John Swenson's 3-part interview on Audiostream.

 

So I read the 3 part interview. Where does Swenson provide a point that supports your unsubstantiated claim?

 

** Thanks for pointing out the 3 part article*** What was very interesting is the part 3 comments section where an SGI engineer that has chops writing device driver software takes the article to task for the mis-understanding about software and bit streaming.

 

Cache killers....

Submitted by John Sully on December 30, 2013 - 7:14pm

 

As a former engineer for SGI, I've done a fair amount of cache performance analysis, and have done my fair share of hardware design work and very low level hardware support code. I've also written several hard disk controller drivers, ethernet drivers, serial drivers and some more exotic hardware drivers.

 

The real cache killer is the sequential access to the DMA buffers of your data. You will NEVER get a cache hit on the data your are decoding/transferring, so worrying about cache locality in the executable is worthless. The noise on the computer from code cache misses will be swamped by those from the data buffer misses.

 

There is a lot else wrong with this article on the computer side but a lot of it depends on specific hardware implementations. Bearing in mind that I know very little about the USB protocol, I even have a hard time giving much credence to "packet jitter". Such a problem can be solved with a minimal amount of buffering in the DAC and as long as the system is not bogged down (that's a technical term) interrupt latency should be on the order of microseconds.

 

Read more at http://www.audiostream.com/content/qa-john-swenson-part-3-how-bit-perfect-software-can-affect-sound#WfEQbgl4S5Eht2XS.99