About two years ago I decided to make the plunge into computer audio. My goals were to lessen the clutter in my listening room, i.e., CDs scattered about, and to gain easier access to my music. It was my belief that computer-based audio could sound as good, if not better, than that from optical drives, though I knew that achieving this was not trivial (more on this below); as something less than a computer expert, this worried me a bit. As is my custom, rather than jumping in blindly, I read everything I could get my hands on, and asked numerous questions of those more knowledgeable. See for example, my Interview with Chris Connaker, and Rob Robinson. Along the way there were of course numerous decisions that had to be made:
1. I opted to use a computer hooked up directly to a DAC, rather than streaming the music wirelessly, or via Ethernet from a computer to a Squeezebox, which would then pass the signal to a DAC.
2. Because the computer would be on my equipment rack, it needed to be quiet. I knew that so-called silent PCs are commercially available, but that they tend to be considerably more expensive than traditional models.
3. I opted for a PC rather than a MAC, both because of my familiarity with Windows operating systems, and because of PC’s lower costs relative to MACs.
4. I decided to go with a computer with a flexible, open architecture, rather than a music server with a closed (i.e., proprietary) operating system. While some of the closed systems are reputed to have superb sonics, I didn’t like the idea of being locked into their system, especially in light of the rapid advances in software for computer audio; the closed systems with proprietary software are also often quite expensive.
5. As I noted above, I was aware that for optimal sonics, a computer must be optimized, which largely entails selecting certain options, and shutting off superfluous functions. Because my computer skills are mediocre at best, this aspect worried me; I didn’t want to purchase a good computer but have its performance as a music server compromised by improper set-up.
6. A last issue that remained to be resolved was how I would handle back-ups, once my CD collection had been ripped.
I thus found myself in the position of understanding in theory what I needed, but unsure of how to actually implement it. As is the case for most of us, difficult economic times was an additional fly in the ointment. One possibility of course was to build a computer; for example, Chris Connaker has published the “recipe” for his CAPS (Computer Audiophile Pocket Server) (http://www.computeraudiophile.com/content/405-computer-audiophile-pocket-server-caps-v20). While this may appeal to some, it is well beyond my abilities, and thus not something I could even consider; I assume at least some readers are in the same boat. So what to do?
I was introduced to Rich Hollis by our mutual friend, speaker-designer Danny Richie of GR Research. Rich knows a lot about digital audio and computers, and is very down to earth in his approach, so I began to pick his brain as to what sort of computer system would meet my needs. As luck would have it, not long after we initiated these discussions, Rich produced his first commercial product, the HAL (“Hollis Audio Labs”) MS-1 music server. It soon became clear that the MS-1 was precisely what I had been looking for.
In producing the MS-1, Rich’s goal was to provide an affordable, plug ‘n play music server, that provides high-quality music playback. As a starting point for the MS-1, Rich sought out a small-profile computer (8” wide x 5” deep x 2.5” high) with adequate processing power, suitable connections, and low electrical noise. For the latter reason, Rich eschewed laptops and notebook computers, as the attached screens tend to create considerable electrical noise. Once Rich identified a suitable computer, he replaced the stock fan with one which is very quiet. Having sat in rooms with noisy computers, I can attest to the importance of a quiet computer. Because the quiet fan was deeper than the stock fan, Rich had to create a new top for the device; he chose to make it out of wood, which I find visually attractive.
The next piece was the built-in hard drive, for which Rich uses a solid state drive (“SSD”). All the machine programs, including Windows, are stored on the solid state drive. SSDs have very fast access, as a result of which the MS-1 boots much quicker (i.e., in about 25 seconds) than spinning drives. Programs also load much faster. A further advantage of SSDs is that they are low powered devices; by putting less load on the MS-1 internal voltage regulators, they generate less electrical noise. Of course, because they lack any moving parts, SSDs are also sonically quieter than spinning drives.
For music storage, the MS-1 comes with an external 1 TeraByte eSATA hard drive. Rich chose an eSATA drive, as that interface sounded better to him than a USB drive; his assumption is that this is a function of the eSATA drive having its own power supply, which keeps the motor drive noise out of the MS-1. During music playback, the MS-1 takes the data from the eSATA hard drive and buffers it in main computer memory. Rich explained that Foobar 2000 (see below) has very big memory buffers to do this.
The MS-1 comes without a keyboard, mouse and monitor; a wireless keyboard and mouse are available as an option. The MS-1 will work with just about any Microsoft-compatible USB keyboard and mouse, which many customers may already have lying around. The purchaser must provide his or her own monitor; Rich chose this route because whereas some end-users might want a small monitor, others might one much larger. In this way, the end-user gets to choose a monitor that best serves his or her needs. Current (as of September 2012) pricing is $699 for the MS-1, $150 for the 1TB eSATA/USB2 drive, and $40 for a Logitech K400 wireless keyboard and track pad. Shipping is not included in the pricing.
The user must also provide a disc drive for music ripping, though this can be avoided if one opts to rip from another computer equipped with a disc reader. The MS-1 has an ethernet port for connectivity to the internet and home networks, as well as an antenna and receiver for wireless connectivity. If one chooses to rip via the MS-1 and an external disc drive, connectivity to the internet allows one to obtain relevant metadata, as well as downloading music. Wireless connectivity also allows for remote operation of the MS-1; a point to which I will return.
The MS-1 comes with Windows 7 installed. Rich also pre-installs Exact Audio Copy (“EAC”) for ripping, as well as the music playback program FOOBAR 2000. FOOBAR is elegant in its simplicity — much like the MS-1 itself. FOOBAR lacks the graphical interface of more sophisticated and more expensive programs, operating instead on a Windows-like Tree directory format. I subsequently purchased JRiver but for beginners, FOOBAR can’t be beat.
To operate the MS-1, one makes all the obvious connections: Power, Monitor, wireless dongle (for keyboard and mouse), and ethernet (not necessary, but recommended). One then connects one’s DAC to one of the MS-1’s USB ports. If the DAC requires a separate driver, as does my AMR DP-777, it may be recommended that the driver be loaded before the DAC is connected to the music server. In my case I went to the AMR website, and followed the instruction for downloading the driver. Note that it may then be necessary to go into the FOOBAR menu, and select the AMR DAC driver. Once the download was completed I connected the DAC to (1) the MS-1 via USB, and (2) my preamp via the analogue outputs, then powered it up. I selected the music I wanted to hear via FOOBAR (Rich provides some sample tracks with the MS-1) and easy as pie, I was listening to music.
For remote operation, one needs a wireless router (I already had one installed from Verizon FIOS), and any sort of mobile device, for example a SmartPhone or Tablet. I use a Netbook, as I had one around that I wasn’t otherwise using. One opens up a web browser and enters a URL from either Foobar and JRiver, along with the network address of the MS-1, and voila, instant remote access. Each remote software has its own plusses and minuses; the MS-1 is a passive partner in this regard, simply taking and implementing the commands.
Ripping one’s CD collection is a time-consuming process. The last thing one wants is to lose the data, so it is essential to have them backed-up. There are sophisticated backup strategies, such as RAIDS, but these are expensive, complicated to implement, and for many of us, overkill. Rich suggested a system that is both simple and inexpensive; I will describe the process for those who have not yet made the plunge into computer audio, to show how easy it can be.
For convenience, I do my ripping at my desktop computer, though as mentioned, one could do so directly to the MS-1; all one needs is an external disc drive and an internet connection. I used EAC, which I downloaded to my desktop computer (http://www.exactaudiocopy.de). As I noted above, the MS-1 comes with EAC already loaded. I save the data on a portable 1 TB USB drive (approximately $100). I first created two folders in that drive, one named “MUSIC,” and the other “MUSIC”. At each ripping session, I save the ripped CDs to the “MUSIC” folder. When the ripping session is over, I remove the external drive, plug it into the USB port on the MS-1, and copy the data to the “MUSIC” folder on the e-SATA drive. The MS-1 comes with this folder already present. I then return the portable drive to my desktop, and move the data from the “MUSIC” folder to the “MUSIC” folder. I thus have two copies of the data: One in the “MUSIC” folder on the MS-1’s e-SATA drive, and another on the “MUSIC” folder on the portable USB drive, which serves as a backup.
For my next ripping session, the first thing I do is rename the “MUSIC” folder with the current date. I save the newly ripped music to that folder, then remove the portable drive, connect it to the MS-1, and copy those files to the “MUSIC” folder on the MS-1. I then return the portable drive to my desktop computer, and move the files in the “MUSIC” folder to the main “MUSIC” folder. The “MUSIC” is thus empty, and ready to be renamed and refilled at my next ripping session. This is actually far simpler than it sounds.
As this was my first music server, I cannot tell you how the MS-1 compares sonically to other music servers. I did however compare it to my reference Sony X707ES CD player used as an optical transport, the transport system of which was modified by Alex Peychev of APL Hi Fi. Because my AMR DP-777 has both USB and S/PDIF inputs, I inserted a CD into the Sony and selected the ripped version on the MS-1, then hit “play” on both at the same time. I could then switch back and forth by simply switching inputs on the DP-777 (and correcting for volume changes between the two). My hope was that the MS-1 would equal the modded Sony. I would be lying if I said the differences were huge; they were however easily discernable, with the MS-1 bettering the modded Sony in virtually all sonic attributes. The same result was obtained with every disc I tried, such that I no longer use the Sony, except for example, to play a CD that a friend brings over. And not surprisingly, hi-rez downloads far surpass CDs.
Like everyone before me who has made the switch to computer audio, the ease of selecting tracks, and of searching by artist or album, has allowed me to re-discover music in my collection that I hadn’t heard in years. And of course, I achieved my goal of having far less clutter in my listening room.
More out of curiosity than anything else, I decided to try JRiver, in lieu of Foobar. JRiver has more features than does Foobar, in particular a graphical interface that allows one to see album covers, and select them for playback simply by clicking on them. To my ear, JRiver sounds better than Foobar, though Foobar recently added implementation of WASAPI Event Driven output, which may narrow that gap, though I have not yet tried it. Irrespective, I prefer JRiver’s interface. But what deserves mention is that even this dinosaur was able to download JRiver, which costs approximately $100, and get it running with no help. Okay, I exaggerate slightly. I initially couldn’t hear any music and contacted Rich for help, who told me to go into the JRiver menu and select the AMR DAC for the output. Alas, some of us learn slowly. With my newfound confidence, I am going to try J-Play, a new JRiver add-on program that is reported to improve the sonics.
Computer audio can be complicated and overwhelming for the uninitiated. In contrast, the MS-1 is simple to operate — one simply connects a monitor and DAC of one’s choosing — and provides sonics equal to or better than many optical transports. It is Windows-based, and thus should pose no problem to anyone who has even moderate familiarity with that operating system. Because it has an open architecture, one can easily, and reversibly, experiment with various playback software programs. Because the storage is external, one can simply add extra hard drives as the need arises. One can also use it for ripping or, as I do, one can rip on another computer to a portable hard drive, and then simply copy the data to the MS-1’s eSATA drive. It is also small, quiet, and very reasonably priced. The MS-1 allowed me to become a convert to computer audio, and I have no buyer’s remorse whatsoever. As such, I enthusiastically recommend it. Rich does not have a website but can be contacted via E-mail (email@example.com), at Facebook (www.facebook.com/HollisAudioLabs), or at Audio Circle (http://www.audiocircle.com/index.php?topic=98467.0).
- (Page 1 of 1)