Any difference in SQ from USB drive storage to NAS playback?
Not sure if there is any difference in theory but.....
Should there be any difference in SQ of files played back through a USB hard drive or flash drive plugged directly into back of my streamer. Vs the same ripped files accessed by internet from a NAS?
Interesting question. Though I am no expert, I think there might be a difference. If I recall correctly, Sam Laufer's The Memory Player outputs files to memory before they are played. The result being higher sound quality. So, it would seem using either memory, UDB or NAS, will result in differing playback. Come to think about it, Aurender caches songs to a SSD before playing music. I'll put the question to Sam when I speak with him. Bob
Bob I have the Aurender N100 and my understanding is it only caches streaming music to it's SSD? It maybe my imagination but I swear music played from the nas sound fuller than from a USB hard drive plugged into back of the n100.
I have been using a Melco N1A as my NAS for the last 2+ years now, and it definitely makes for better SQ than when the music I played was stored on a WD My Cloud or Seagate Backup Plus device. So in my experience, the quality of the storage implementation definitely makes a difference. Melco, Aurender, Innous, and such are purpose built audiophile devices, equipped with better power supplies and quieter storage mechanisms, all making for a lower noise floor and yielding better SQ, and as expected, you have to pay for the priviledge.
I researched this on my Mac mini and found that the file transfer speeds were significantly faster with the NAS versus the USB drive. I believe the Aurender N100 only provides USB 2.0 data ports. This provides 480 Mbit/s versus the Gigabit Ethernet port. You also need to take into account the usb controller chip(s) used by your streamer. On my mac mini there is only one controller chip so it splits time between bringing the data in from USB drive and sending it out to the USB dac. I believe this is why music sounds better from the NAS than USB drive on my system
A quality NAS running a fair distance via Ethernet sounds better than local UsB spinning disc storage.. I suspect this is caused by higher noise drive and limitations on USB protocol/ switcher power supply close to preamp... the best sound is USB 3.0 stick plugged right in front of NAIM DSD128 especially good.
As long as there is enough bandwidth to transfer the data fast enough ( there should be with most home network connections or with most any properly functioning usb drive with cd resolution files ) there should be no difference.
If not, some streaming software is designed to be able to reduce streaming resolution accordingly, but in most cases the music will simply stop playing while the data rebuffers up enough to stream again at the source resolution.
Behavior here could vary streamer to streamer.
Otherwise, there should be no difference. The storage device is either capable of delivering the data fast enough or not.
I can't see USB2 being a bottleneck for music. The average uncompressed music file is about 50- 60 MB, a CD is around 680 - 740 MBs, USB2 runs at 60MBs a second so if everything could process it, theoretically you could play an uncompressed song in 1 second over USB2.
For Melco (and I believe for Innous as well) a good part of the game is jitter reduction. There is both an ethernet line in as well as an ethernet player port out. Everything gets reclocked by the Melco. So in effect devices such as the Melco function as an ethernet line conditioner.
Jitter is relevant to the digital to analog sound conversion process between music streaming source and DAC downstream but not here in the case of either usb disk storage or NAS. Neither has anything directly to do with sound quality, just digital data transfer from computer storage to computer memory, audio being just another kind of data file accessed. Bits are assured of being transferred accurately in all cases if things are in proper working order else nothing loaded from a usb drive or network adapter could work reliably.
This is why I started this thread. We have plausible reasons both for and against a possible difference in SQ from the two sources but it seems that nobody truly knows definitively.
Mapman With all due respect that is your opinion and great that you have shared it but please do not become like some here who assume their opinion is law and gospel and cannot be argued against.
You could well be correct but an open mind never hurt anyone.
One possible problem scenario with a usb attached drive might be if the computer is both a server and streamer and the USB drive attached results in the computer producing more noise that introduces jitter in the streamer to DAC interface.
That is a unique case where a single device serves as both music server and streamer. I can only speculate that might possibly be a factor in some cases but even if so should be manageable with a good design.
I use USB drives on my music server PC but I have separate streamer devices that connect wirelessly to the server. This is a very good configuration IMHO because the streamer is far away from the home computer with no vulnerable wired connection and home computers can certainly introduce noise capable of producing jitter on the feed to a DAC.
So it is always possible in this case one connection behaves better than another due to noise differences but even if so that does not mean that one approach is better than the other categorically. That would be jumping to conclusions.
Also note that most good quality modern DACS are designed to be jitter resistant so the likelihood of an audible issue is even less in that case.
...so Uber looking back at your original question your configuration with uSB connected direct to streamer fits the bill. Your streamer is serving as both a music server and streamer with usb attached so it could produce more noise and jitter that impacts the digital to analog conversion process than a NAS. The reverse could be the case as well, especially with a wired internet connection to NAS, or there could be no difference at all. Could vary case to case, streamer to streamer. The devil is always in the details. One would have to try both ways to know.
That also sounds very plausible too. Could very well explain why to my ears played from the nas has more body and dynamic than from an attached USB HDD or even USB flash drive.
I am always receptive to trying it out but sometimes also interested in just why something would act as it appears to.
Are the media files the same Flac or Wav on both the NAS and USB drive? Is the drive a small 2.5 device that is being powered by USB or does it have it's own power supply? I actually have a similar problem using my Node2i. Streaming from Tidal sounds a little better to me than streaming from my hard drive attached to my router by USB. I can't imagine the N100 has a poor USB implementation I imagine it has noise reduction galvanic isolation.
Djones if the streamer is accessing files on a remote device directly via a network file share, the file share protocol may be a bottleneck that could limit bandwidth. Better to run music server software on the device hosting the files and allow streamers to connect to to music server as clients.
Media files are identical flac files ripped with same software at same time. All I did was download them all to USB HDD drive and some to USB flash drive for a comparison.
Can't really tell any difference between the two USB drives but from the nas it sounds slightly preferable to my ears.
I should point out as others have that jitter and noise vary by format. For any USB ( 2.0 and 3.0 ) I use the Audioquest Jitterbug.
nice story : I was visiting an unnamed robotic machine tool supplier to check progress with final acceptance testing of a $100 M fleet of machines. Open the panel and it is chock a block full of Jitterbugs. I saw the MTBF and bit error rates with and without. Ordered enough to plug all the ports that evening.
Please forgive my ignorance but if software such as Audirvana loads songs or albums into RAM before playing, doesn’t that eliminate any lag or latency? Doesn’t that mean that SSD storage vs NAS storage both become irrelevant and both would be better than streaming?
Uber to keep this less contested, when you visit Seattle this summer remind me to do a test listen with you of Jitterbug in the Unitiserve with SP-DIF into the DAC there will be Pinot and smoked salmon, of course the Trollers ( pardon the pun ) will claim no double blind differences between farm ( ugh ) and wild fish ( yum )
Oh well. At least he tested the jitterbug in the right place where a reduction in jitter might matter. Unfortunately there appears to be no useful effect.
Tgris,
Most streamers do buffer data in memory to isolate the D to A process from any minor temporary variations in bits per second but jitter can and usually does still get introduced to the signal between streamer and DAC for various reasons so its up to the DAC to deal with any jitter introduced into its input signal and more and more modern DACs do just that by re-timing the digital signal just prior to conversion to analog.
Uber, Friday Harbor is a beautiful place. It would be great to meet up. I don't get to Seattle too often tomic, but it would be great to meet you as well. You're in Redondo if I remember right?
For many years I used a late 2012 Mini Mac with an internal 1TB hard drive. A few weeks ago the HD died. I'm back-up crazy so nothing lost. I have a NAS which was used for streaming movies and back-up's for multiple computers. I re-tasked the NAS for music playback. The NAS uses standard hard drives, not SSD. I have been streaming from the NAS for about a week and have not notice any change at all.
I knew what you meant Uber. Tomic, we have a music club here in Olympia that meets monthly. Tonight actually. If you would like to check it out sometime, let me know and I'll send details.
You must have a verified phone number and physical address in order to post in the Audiogon Forums. Please return to Audiogon.com and complete this step. If you have any questions please contact Support.
