This review page is supported in part by the sponsors whose ad banners are displayed below

Reading Paul Candy's report on JPlay had me suddenly realize that I owed our readers a quick follow-up on Audirvana. I was unexpectedly able to over a few weeks deepen my comparisons against Pure Music thanks to a far more resolving full Ocellia system I happened to have on review in my music room. Over the previous pages I'd come to the conclusion that the two software players were sonically too close to truly separate; and that a choice should be based on the feature set you desired and the budget you felt was acceptable. This conclusion still stands. For the most part. Yet in a more revealing system there indeed are nuances between the two programs that could become slightly more meaningful.

All in all both are tremendously resolved and can reproduce the concert venue and all its acoustics realistically. Audirvana though will offer slightly looser upper bass (think slightly warmer and more rotund with a little less impact and detail) while offering more realistic (read closer to a real instrument) timbre accuracy and tonal colors. Pure Music on the other hand will give you a slightly tighter presentation with a tad more punch while falling back just a little on tonal richness, at times showing a remnant of digital hardness. On this last point I doubt anybody listening to Pure Music without a comparison would ever accuse it of sounding digitally sterile—not even close—but when facing Audirvana both Ocellia's Samuel Furon and I felt that Audirvana went just that little further in closing the gap to master tape or superior vinyl playback.

So I don't think my conclusion changes at this time. For most systems the choice should really be driven by the need for Pure Music's advanced features or not. But for somebody who was ultimately looking for the most sophisticated reproduction of nuance in tone and body of acoustic instruments, Audirvana would get my nod. As such it is the player I use as my reference until something even better displaces it.

One of the advantages of a far more resolving system with a finely tuned computer audio source is that I can now throw in my two cents on fascinating issues such as ALAC vs AIFF playback (also known as FLAC vs WAV for the Windowphiles amongst us). In all fairness until now I'd heard no difference between the two file formats. Zip, nada, nothing. But I was wrong. Pulling my trusted Chesky audiophile high-resolution disc off the shelf once again, I added an AIFF rip via XLD (Max Paranoia mode thrown in for good measure) to my existing ALAC version ripped with the same settings. This guaranteed as much as is technically possible today that both copies were bit identical, the only difference being that one was in a lossless compressed format (ALAC), the other uncompressed. All computer experts will tell you that there cannot possibly be any audible difference. I absolutely agree. The data is identical. The only difference is an unzipping process for the ALAC file. This is mundane low-level on-the-fly stuff for any modern processor. It really should result in no difference whatsoever. But it does. To put minds at ease, it is far from night and day. It's subtle at best. Yet once identified, it is surprisingly consistent. The difference I heard was with the lowest-level information which determines the next strata of realism.

On the Chesky disk the easiest way to hear that difference was to actually listen to the narrator who introduces each track. In ALAC his voice is extremely present and three-dimensional in the center between the speakers but in AIFF you actually hear his lips part. The slightly wet and very low-level sound he makes when his lips open can be consistently heard on the uncompressed tracks but not on the compressed (zip) files. Granted I have to strain to hear the difference and one could categorically question its relevance on musical playback. But the fact that a difference that ought not to exist does albeit quite small puzzles me enough that I will henceforth rip my CDs in an uncompressed format. When I get to it, I will also convert back all my ALAC files to AIFF just to be safe. Storage is cheap. Not all systems will be capable of reproducing that low-level data to worry about it but when I listen to gear that can, I'll be glad to know that I am feeding it the best signal I can...   

24/352.8kHz DXD file downsampled to 24/192, integer/direct mode, absolute polarity inverted to account for inverting preamp

Publisher's comment: With the most current versions of Amarra, Pure Music and Audirvana on my iMac, the Integer and Direct mode features of Damien's still beta version—at the time of publication—widen the gap to Pure Music. On my iMac (27" quad-core fully maxed out top model, OSX Lion) previous betas suffered instability issues. These ran the gamut from first black-screen freezes with forced reboot command (Kernel crash) to interruption of playback with nasty steady-state noise during later versions. has banished these gremlins in the context of an Audiophilleo 2 USB-to-S/PDIF converter. Direct mode requires integer mode to operate (but the latter can work without the former engaged). Integer mode requires a copasetic D/A converter (both my Audiophilleo and the XMOS module of the April Music Eximus DP-1 are).

One should also know that direct mode circumvents certain core safety protocols of Apple's OS. This then becomes a high-performance mod that's not fully street legal. In our context it simply means that you shouldn't use your computer for anything other than music listening whilst in direct mode. Surf the web, answer emails and you're liable to have accidents sooner or later. Those reluctant to engage this mode might prefer to wait for the official version until final bugs have been exterminated. Based on my own experiments, I'll loudly join Frederic in naming Audirvana the as yet best-sounding Mac-centric music player I've tried. Except that in direct mode it's even better than he's already called it!
Audirvana website
BitPerfect website
Decibel website
PureMusic website