Choosing a NAS Drive for Sound Quality

[kennythekey]

I'm looking hard at one of the two-bay Synology NAS Servers such as the 212+ or 712+. I am aware that there are some good choices for compatible HDDs, and there lies my questions.

 

Has anyone done any A/B analysis for sound quality? Are there recommendations?

 

I was ready just to pick one of the top three brands (2TB) when I stumbled across a review by Enjoy-the-Music's HiFi Critic. The review consisted of listening tests between a few NAS Servers and HDDs/SSDs. The review was an introduction to promote the realization that these devices also have their own sonic differences and some sound better than others.

 

[losingmyreligion]

well, it's been a few days, and no one else had responded so I'll bite.

 

At the worst case, this will serve as a 'bump' for your post.

 

I read the review, albeit quickly.

 

My position of the likely sonic performance of a NAS has been that of a downgrade (compared to direct connection to drive via Firewire), although never tested.

 

Compared to a single 2.5" hard drive (connected via Firewire, which means the bulk of the data processing is performed by dedicated Firewire chip), I didn't see how it was possible for a NAS NOT to sound worse.

 

This review reinforces my belief, for the following reasons:

 

If we prioritize the sonic qualities of the NAS(es) under review, the ones which require the most processing power (and thus are more likely to generate the most noise) sounded worse than the ones requiring less processing power (e.g. smaller processing chips in the NAS vs. NAS comparisons). For the comparison of the drives within the NAS, the HDD versus SSD comparisons were as expected, with the normal SSD sounding better. The exception was the Corsair SSD, which sounded worse, but this was likely due to it's design, about which the reviewers noted "SSD's microcontroller is exerting itself more than usual in order to maintain a high throughput of data".

 

So, I'd avoid use of a NAS if at all possible, if you want best sonics possible.

 

If not, I'd use damn good AC power filtering to prevent contamination of the AC circuit used by the other components.

 

If I were recommending use of a NAS to a friend, I'd recommend a Synology 411 model which allows use of 2.5" hard drives. If he were quite wealthy, I'd recommend that he use non-Corsair SSDs.

 

 

take this with a grain of salt, as I've never tried, listened, or even considered use of a NAS, so my views may be biased.

 

I use an Oyen MiniPro 1Tb with Firewire / USB

 

 

clay

 

 

 

[ted_b]

as fw or USB. The gigabit ethernet connection is faster, the NAS is acoustically and electrically isolated from my listening environment (in the next room on a different circuit), and the overall sonics are equal to or better than what I found with the Oyen. I will relook this, based on Clay's reputation (still haven't found a use for that damn Jameco btw ).

 

[losingmyreligion]

 

"electrically isolated from my listening environment (in the next room on a different circuit)"

 

 

It's likely that your electrical isolation has minimized / eliminated the potential detrimental effects. I've never actually listened to a NAS, so I'd go with Ted's actual experience versus my biases suported only by the review.

 

I'd make sure I employed the isolation Ted mentions though.

 

Ted, are you also isolating the power on the Oyen as well?

 

Apologies on the fit issue with the Jameco Linear PS versus the Oyen's DC input. A little cable surgery might be in order?

 

 

Kenny, please also keep in mind that everyone's environment is different, esp, when it comes to AC power and EMI/RFI, which are the two likely culprits for influencing sound whenever the bits are not changing.

 

Also note, that the reviewers were using an extremely high resolution system. The differences in sound quality might not be noticed in lesser systems.

 

The good news is, we found someone with actual listening experience for you.

 

cheers,

clay

 

 

 

[ted_b]

another; not sure if it's that easy?? Anyway, right now I'm copying some DIFF and DSF (DSD files) from my NAS over to the Oyen, and I'll a/b them tmrw or so. I'll write back...

 

Edit: found a 2.1mm to 1.3mm converter plug for $5 plus shipping.

 

[losingmyreligion]

 

 

What's your current configuration?

 

I'd suggest use of Firewire for data (with the Oyen) if possible, even if using a Firewire DAC. Putting USB processing on the Mac will likely make it sound worse, and I've never experienced any sonic penalty when using the Firewire bus in the manner it was intended (i.e. multiple devices).

 

I'm sure you know this, but I'll post for others who may be lurking.

 

Firewire has a peer to peer protocol managed by the Firewire chipset which was designed for multiple data sources over the same bus. [note: this is what allows daisy chaining (of Firewire drives) rather than needing a plethora of ports, such as with USB.] The Firewire chip does the vast majority of data handling. This is pretty much the best case for audio/video data handling (the latter of which processes more data then even DSD). USB, OTOH, relies on the main processor for it's data handling, and so works counter productively to players as they try to minimze the processing by the main CPU for best sound.

 

Therefore, IMO and IME, the seeming intuitive advice to use USB for data when using Firewire for DAC can actually make things sound worse than using Firewire for both.

 

YMMV,

clay

 

 

[Guest Caner]

I have recently upgraded to Synology DS212 and I use it as my media server.

 

I have 2 different set-ups:

 

1) Critical listening: MBP with dual OCZ Vertex 3 SSD (optical drive replaced) -> Decibel -> Oyaide Neo FW Cable -> Weiss DAC2

 

2) Non-critical listening: Synology DS212 (2 x 2TB Hitachi Deskstar 7K3000 600 MBps drives)-> Squeezebox server -> SB Touch -> SPDIF -> Weiss DAC2

 

I can not make a comparison between different NAS brands (my last NAS was a Synology unit too) but I can make a comparison between the above 2. Not surprisingly there is a palpable difference between them in the favor of #1 (hence the critical listening).

 

I realize this does not address your inquiry, but hope it helps.

 

[Julf]

Are we talking about using the NAS just as a storage server, or also as a streaming server?

 

[Audio_ELF]

Caner, with respect your comparison (as stated) is as more comparing Squeezebox Touch via SPDIF vs MacBook via FireWire than comparing NAS to direct storage.

 

Have you tried playing files on your NAS via MacBook vs the internal drives of the MacBook.

 

Eloise

 

PS I agree with Julf that NAS as storage vs streaming server is also a consideration.

 

[ted_b]

My DAC is firewire right now (USB driver is messed up for the moment). Do I go out of the Oyen into both (Oyen has two fw ports, Mac and DAC only one). Too bad my best cable (Granite Digital) must go to Mac cuz it's the 9 pin end.

 

The NAS is on gigabit ethernet (with special attention paid to router, etc to make sure the whole path is gigabit).

Ted

 

[ted_b]

but instead, as Eloise stated, a comparison of the different systems entirely, with the biggest variable your Squeezebox vs Mac! I have no doubt they sound different. Different DACs, different connections, different players, and different music storage. And as Julf says, SB is streamed (TCP/IP packets) and other system is not. in fact, those two systems couldn't be more different unless the music files were different formats.

 

Julf, in my NAS setup (and I assume the OP's) I am not using any NAS services, any UPnP streaming apps, etc. Just NAS as remote storage.

 

[Julf]

in my NAS setup (and I assume the OP's) I am not using any NAS services, any UPnP streaming apps, etc. Just NAS as remote storage

 

That's what I would recommend too, as the streaming, UPnP etc. stuff is just "well, let's chuck that in too" add ons for the NAS manufacturers - probably not really optimal in any way.

 

[kennythekey]

Thank you all because I thought I posted another sleeper!

 

I'm just putting together a digital system, but I got bit by the bug back in 1966. Right now, I have an EE DAC Plus, a Squeezebox Touch, and some hi-rez albums from HD Tracks on my internal HD (Windows). I want to move these off my computer.

 

I plan on building a C.A.P.S. or doing something else later to replace the SBT. For now, I'm using coax from the SBT to the DAC. My goal is to rip my CD collection to a NAS and add new music as well.

 

I now require storage, so getting a NAS is my next step and for now I would be using it for streaming. I found nothing on-line about the sound differences between HDDs, except for the review article.

 

The information that some of you provided about handling the AC properly makes the most sense to me and I can do that. I also believe that if HDD/SSD comparisons are made then the rest of the chain must be identical, so no apples with oranges.

 

[Julf]

I found nothing on-line about the sound differences between HDDs

 

Probably because there aren't any. Most decent players buffer the data, so the NAS doesn't contribute to the sound at all.

 

[kennythekey]

Sorry, I did not provide the url to the article.

 

http://www.enjoythemusic.com/hificritic/vol5_no3/listening_to_storage.htm

 

[Julf]

While this additional test was in no way scientific

 

No s*t, Sherlock!

 

 

 

[Guest Caner]

"Have you tried playing files on your NAS via MacBook vs the internal drives of the MacBook"

 

That I haven't done also...

 

To me, using NAS as a wireless streaming server is so convenient that I would not consider it usng as a storage since my "select" music library is around 240GB which can be easily handled by the 2nd SSD I've installed to MBP.

 

But since you've mentiond that, now I wonder if there would be a difference in the SQ using NAS as such.

 

[kennythekey]

Julf -

 

Your opinion is well taken. I did not consider this to be a factor.

 

The review article admitted to being somewhat inconclusive...pending a test of other NAS combinations. This is the reason I'm asking the forum.

 

[Mark Powell]

Yes. I use JRiver. By default it buffers 6 seconds, but also you can play the whole track from memory. Therefore no possible differences.

 

I suspect that SQ differences are (1) a conspiracy theory, (2) a need to fill the magazine with something or other.

 

If you hear a difference between drives, NAS or any other sort, including SSD, you MUST be playing direct from disk, which is a 'bad' way to do it.

 

EDIT: Just disabled lots of things in Windows 7, including the continual security scan which I installed.

 

Voila!

 

Play from memory, the lights indicate the the disk is not accessed at all. After a while it stops going round completely (sleeps).

 

Buffer only, disk is accessed evey few seconds.

 

[Guest Caner]

that my comparison did not address the OP inquiry. And I can not comment on the sonic qualities of different NAS brands, as I can not do a comparison.

 

You've also correctly identified that there are a multitude of factors/differences that can/does affect the outcome between the 2 setups that I choose to use (FW vs SPDIF, SSD vs HDD, SBS vs Decibel, Oyaide Neo vs wireless).

 

Using NAS as a server (like I do using software mentioned) vs storage is also a valid consideration.

 

 

In essence what I've tried to say is that there is a difference between these 2 setups for even a novice like me.

 

Now I realize that I am off the mark.

 

 

 

[kennythekey]

Mark -

 

This makes complete sense and seems to debunk any differences.

 

At the same time, Chris Connaker has made A/B listening tests with a number of people who seem (majority) to prefer the sound of SSDs.

 

Does this mean that those tests are being conducted direct from disk?

 

[Mark Powell]

Don't know what it means. People will always search for villains, even non existent ones

 

Personally I didn't hear any difference either way. The same people will say I have cloth ears or my equipment is not as expensive as theirs

 

[Julf]

Chris Connaker has made A/B listening tests with a number of people who seem (majority) to prefer the sound of SSDs

 

A/B tests, or ABX tests? With something like this, it is important (and very easy) to do a blind ABX test to be as objective as possible.

 

Another thing is that statistics are important too. If you test with 4 people,

and 3 of them prefer A and one B, that doesn't prove anything. The probability of that happening by random even if there is no actual difference is actually 50%. The probability that all 4 will prefer the same is 12.5 %, or 1 in 8 - not very unlikely.

 

 

 

[Mark Powell]

Some will always claim to hear a difference. How often have you seen here the phrase "I have sensitive ears"? If this was a video site they would claim to be able to see into the infra red and the ultra violet. They are no doubt also taller and better looking than the rest of us. There may some so blessed in all such aspects, but not many.

 

The true test is, of course, whether they can reliably repeat the difference they claim to hear And also pick the same 'best' one the next day, week, or whatever. And do it many times, being right very very often.

 

And given my 'no disk access whatsoever' test, internal disk, USB, NAS, or SSD, it is all nonsense anyway, AB,ABX, or any methodology. A disk you are not accessing and that has turned itself off cannot influence the sound.

 

[losingmyreligion]

Ted,

 

Ideally, the Myetek would have two firewire ports, but hopefully will work with the Oyen. I prefer to put the hard drive on the end of the chain.

 

Just connect both of the single port devices to the one with dual ports (Oyen, in this case).

 

 

clay