Mac Mini and Ayre QB-9--Amarra noticeably improving the sound?

[jay]

I'm on day 6 of the Ayre QB-9 and loving it. I have loved the Ayre house sound since first hearing their equipment 2 years ago (if that gives any indication as to my sound preferences), and I'm feeding a K-5xe, so I really wasn't too surprised that the QB-9's benefit was more of a good thing for my system.

 

Wanting to give it the best signal I can, I tried the Amarra demo, having read such positive reviews of that software. On my headphones (AGK 701s) and X-Can3 headphone amp coming directly from the QB-9 I heard some marginal improvement on a very few tracks, most noticeably a slightly smoother female vocal and on tracks with stand-up bass, a slightly better timbre and a slightly tighter feel as well for that instrument. But just slightly, and really no other noticeable differences on probably 10-12 songs I a/b tested the software with. None sounded worse with Amarra, btw. I should note that I struggled to hear any improvement through my Ayre pre/amp and speakers, even on the tracks where I heard some small improvement through my headphones.

 

I'm not doubting that folks are hearing significant improvements over the Mac/iTunes implementation in some systems, but I'm curious--are other people with Macs and QB-9s hearing a more pronounced improvement than I am? I realize there are a million reasons that others might be hearing different results, but my system is typically pretty resolving, and I am curious if I should continue to be giving Amarra serious consideration with my hardware combo. I still may buy the software, but I'm more curious as if I should be hearing more of an improvement.

 

Thoughts from folks with the same or similar set-ups?

 

[Purite Audio]

Obvious question but was Amarra switched on, and which files/ codec were you playing?

Keith.

 

[Mr.C]

I recall reading that when someone compared the Wavelength Cosecant to the QB-9...maybe it was different dacs.... with Amarra they found that the QB-9 was affected much less. Anyways, the gist was that amarra differences were somewhat dac dependent.

 

[Lars]

Much of what Amarra does is in the presentation of the soundstage. The differences between Amarra and iTunes can be more easily heard with good speakers. I have Stax 007 Omega II cans with the Stax SRM007T II amp. Compared to my Wilson Sasha W/P speakers, The Stax Omegas display less difference than the Wilsons.

 

 

 

[Dave Clark]

I think that with Amarra, there might be things going on that we do not quite have a grasp of... people who have top systems hear a dramatic difference and some with similar systems do not, those who have mid-type systems hear a dramatic difference and others with similar systems do not, and so on down the line. I am not so sure that anyone can make an absolute statement about what the system requirements are for it to be noticeable or not as there are too many conflicting reports. Here on my own it is not night and day (rather subtle though long term listening tends to tip its hat to Amarra versus iTunes - of which still sounds very good here), elsewhere it is night and day... is it an issue of the (mine or others) DAC, OS, memory, CPU, rfi/emi, jitter, ripping decisions, file decisions, etc.... got me. To say that it requires 'x' to be heard is rather questionable. Why some hear/experience differences and others do not, is not going to be an easy question to answer. In the end either it works for you or it doesn't. Mys system is very resolving and yet no one really hears that much of a difference between Amara and iTunes - though I do use and recommend Amarra as something you should try. What differences anyone hears here, tend to lie in the presentation of soundstage and overall smoothness with Amarra, but all quite subtle. Have 3189 here and my system is listed here... http://positive-feedback.com/staff.htm

 

 

[jay]

Yeah, I used the "on/off" a/b test feature with both AIFF and Apple Lossless.

 

[jay]

I also seemed to be reading that it "performed" at varying degrees with various DACs, and also seemed to be reading conflicting reports in some cases even with similar hardware systems.

 

I will listen to some more types of recordings and see if I hear obvious improvements in soundstage. I'm coming at that part from a very shaky frame of reference because the QB-9 in my system is doing such a much better job in soundstage than my last 3 years with the Benchmark that it all sounds much better to me in that regard, Amarra on or off. Lucky me I suppose.

 

I suspect that I was hoping to hear someone say "yeah, I have the same mac mini and same dac and ... and I hear a huge difference" as a frame of reference for the system I have downstream. Or my hearing. Or to your point Dave, 30 other things that we don't understand.

 

thanks folks!

 

[cfmsp]

 

very well said, Dave.

 

Although Lars probably comes closest to pointing to the reason why the 'difference' is not heard as dramatic by all, i.e. soundstaging.

 

As opposed to my 'youth', I no longer place huge emphasis on so-called imaging aspect of the overall soundstaging, despite that one of the strengths of my speakers is imaging (Audio Physic Virgos). However, I am quite interested in the contribution of ambience (i.e. reproduction of the overall acoustic environment) to overall 'realism', and this is where Amarra makes a contribution ot my listening experience, although it is admittedly subtle in my system (perhaps DUE to my system).

 

I believe this to be (at least part of) the reason that I don't have an "OMG, I can't listen to iTunes without Amarra" experience.

 

In my view, most decent audiophile systems (at least that I've heard) can provide 'imaging' that is superior to to what I experience with 'live' music, even though they do NOT exhibit the 'you are there' realism that I desire.

 

Pinpoint imaging is rather low down the list in my priorities, actually most sudiophile characteristics pale in comparison to reproduction of 'you are/were there' realism.

 

cheers,

clay

 

 

 

 

[borderdog]

Jay,

 

Amarra and Ayre dac.

I have the same mac mini and the same Ayre dac and a very resolving system and I hear

a very, very, very slight difference. I have not listened to Amarra through other dacs, so

I won't comment on what other people are hearing. But as far as Amarra through an Ayre,

the difference is not enough to warrant the price of admission.

 

I noticed in another thread that you are hearing differences as your new mac mini is breaking in.

If you can hear that difference but no noticeable difference with Amarra, I would not suspect

your hearing, but rather suspect Amarra. I am not willing to pay $1000 for an automatic sample-rate changer. That is the only benefit I see. Maybe the three band parametric equalizer - used for bass room modes. Some day I'll pull out my test equipment and do measurements and play around with the equalization settings in Amarra. But for now, I'll keep the demo on my mini and use it as a sample-rate changer.

 

Aaron H

 

[jay]

Aaron--

 

Thanks, that's very helpful, I appreciate your time and feedback.

 

BTW, for anyone interested, the new Mac Mini 24 hours of play later is definitely back to where the old one was. However, for those on the Apple website ready to buy a new one, I'm not hearing any noticeable improvement (yet, anyway) over the old one despite 4G of RAM (vs. 2) and a dual-core (and faster) vs. single-core and slower. I had read various threads on this site primarily suggesting "stronger" Macs (i.e. more memory, more/faster processors) may have been displaying better playback characteristics. Between Mac Minis I'm not yet hearing an improvement (but darn, does it sound great paired up with the QB-9!). I don't have a Mac Pro loaded up to compare to, so who knows there. With an 8-core and 10G of RAM I can't imagine it would be worse.

 

Back on this topic, if different DACs are seeing various levels of improvement from Amarra, could it be (pure conjecture) that Amarra may be eliminating jitter as one of it's accomplishments, and DACs that may be more sensitive to jitter may be getting more noticeable improvement than others that are less susceptible? That's again is just a guess as I don't have a bunch of DACs to test with or tools to measure jitter. I'm certain their digital volume control, parametric features, etc., are also far superior to iTunes (I'm ashamed to admit that I was unknowingly suffering quality loss from leveraging the volume control in iTunes when necessary because I didn't have a remote for my pre-amp). However, I'm not using any of those features now that I have upgraded to a better pre-amp.

 

Anyway, thanks folks!

 

[CG]

" I am not willing to pay $1000 for an automatic sample-rate changer. "

 

Then don't. >> http://sbooth.org/Play/

 

"Unstable" revision 1105 - isn't all software unstable? - does automatic sample rate changing. That's if you want it to.

 

After all my bleating about this *free* software, I'd like to read of someone actually trying it. To me, it does the soundstage thing better than iTunes does. Not quite as sexy looking, but it works well and sounds good. Of course, YMMV, as they like to say.

 

[CG]

"could it be (pure conjecture) that Amarra may be eliminating jitter as one of it's accomplishments, and DACs that may be more sensitive to jitter may be getting more noticeable improvement than others that are less susceptible?"

 

I'll add my own pure conjecture.

 

Many, OK all, digital audio processors of any kind filter the audio stream in one way or another. That's what they do in order to process the audio. In the case of Amarra, nobody outside of the authors verifiably knows what kind of filtering is being done by Amarra. Certainly, some filtering is performed under some circumstances like when dither is added. At some level the bits *are* being changed; where and how is the real conjecture.

 

How this filtering interacts with the hardware based filter within your own DAC is a variable. In the case of the QB9, Ayre has gone on record on their web site describing the unique filtering they use. It has also been detailed in Stereophile magazine. Other DAC vendors use their own filtering algorithm, whether it is a custom program performed in an FPGA or DSP, or the filter built into the DAC chip they are using. There is a big difference in the characteristics of these filters.

 

Although not the same filtering as done in the DAC, http://src.infinitewave.ca/ shows some examples of filter comparisons.

 

So, when you do a little processing in the computer and some more in the DAC itself, you get the concatenated results at the DAC output. Not to mention all the filtering that was performed in the mastering chain. In some cases, the second filter may effectively over ride the effects of the first. The results you get in your system will be the multiplicative effects of all the digital filters used in the chain between the microphone and your preamp. Then your preamp, amplifier, and speakers will add their own analog filtering. This is all before the acoustics of the room do their bit. To really confuse things, then you have your own auditory processing system that will add biological filtering.

 

 

[CG]

Man, can I kill a thread or what? I figure that if I commented in every thread over the next two days, Chris would be shut down before the solstice.

 

I think it's time for winter hiatus. Have a good holiday season everyone!

 

[Dave Clark]

thread still alive... I am not so sure what is going as there are simply so any variables...

 

[Mr.C]

So, to take this apply it, the filter on the Ayre dac overrides the benefits of the Amarra filtering and the "detriments" of the iTunes filter to produce a relatively software independent sound.

 

[church_mouse]

Going off tangent slightly, CG asked for any feedback about sbooth's Play.

 

I have been using this software ever since I was first loaned, then bought my Weiss Dac2. My primary reason is that all my lossless is in Flac, which Play plays without issue. Initially I did compare it to Songbird and then to i-Tunes (using Fluke, and also converting some tracks to AIFF). For me Play, in both my speaker and headphone systems, was noticeably better - seemingly less rolled off at the top and bottom.

 

I must qualify my favourable sonic impression of the Play software by saying that I have found the recent "unstable" versions rather...well, unstable. I have experienced problems with tracks in the playlist simply disappearing before getting played. There is also an issue with dragging and dropping an album in to the playlist and the track order getting reversed. My solution has been to switch back to the last stable version (I have only 16/44.1 FLAC at the moment).

 

To use a popular acronym I have seen on this site (as an aging Englishman, I had to look up what it meant when I first saw it) YMMV.

 

I have not tested Amarra - waiting to see if they produce a FLAC reading version.

 

 

[cfmsp]

 

 

CG,

 

I totally respect your ability to waltz in and out of here like this was just an internet forum or something, but please understand that your signal-to-noise ratio is very, very high.

 

until next time, enjoy,

 

clay

 

 

[Part-Time Audiophile]

I have been trying out (demo mode, with all the annoying breaks) Amarra for about a month now, and I've stopped using it based on my experience. That is, my lack of experience -- my system simply cannot discriminate the difference w/ and w/o Amarra.

 

Chatting with Steve Nugent from Empirical, he mentioned as an aside that this is probably due to my DAC (as opposed to the rest of my system) -- the PS Audio DLIII is an upsampling DAC, so all pre-DAC processing is likely lost or diminished. As an aside, he also mentioned that there are very few DACs out there that actually can discriminate those differences (his being one, of course). But what that implies is that there is a range of DACs that are sensitive enough to showcase those upstream differences.

 

I'm loathe to conjecture as to why there is that stratification and why it might be that the Ayre might be in that non-discriminating class. My suspicion is that the Ayre performs an EQ similar to what the Amarra software does. Maybe it's that MP filter.

 

Regardless, I will cheerfully propose that just because the Ayre doesn't make a lot hay out of Amarra does not in any way imply that it isn't a great DAC.

 

[Eric2]

Peter, Amarra is designed to be used with the iTunes interface and Apple don’t support FLAC so I can’t see this happening anytime soon.

 

[cfmsp]

 

 

"As an aside, he also mentioned that there are very few DACs out there that actually can discriminate those differences (his being one, of course)."

 

Of course, he does.

 

 

"But what that implies is that there is a range of DACs that are sensitive enough to showcase those upstream differences."

 

The fact that a DAC sounds different via Amarra could as easily imply that it is overly sensitive to the input it receives from a computer and that Amarra resolves the issues that the DAC could not, e.g., jitter.

 

The point is, since no one can explain how/why Amarra sounds different, we can assume little with regard to why it appears to interact differently with some DACs versus others.

 

clay

 

 

 

 

 

[Part-Time Audiophile]

You know, no matter how many times I see your icon, it makes me scratch my head. ;-)

 

You're quite right of course. But while it's hard to know *why* some things sound better than others, there is that pesky fact that, ceteris paribis, some nonetheless do. Take Shakti. I have NO idea what those stones are doing. LOL.

 

Ok, that said, I think we can make some guesses as to why some things create the perception of improvement, no?

 

/Begin Speculation/

 

As for Amarra, I can only assume that what it's doing isn't reducing jitter. How could it? Given that its a shim for iTunes, I just don't see where it could add either clocking or significant buffering to address jitter in any meaningful way (not to say that it couldn't, but I don't believe that's what its doing, given that it's still using Core Audio -- at least in the Mac -- as well as iTunes). I think its (in my case, alleged) sonic improvements have more do to with it being a filter, kind of like a phono pre-amp applying an RIAA curve. If so, it follows that with DACs that apply similar filters, either explicitly or implicitly via circuit design, Amarra is sonically redundant.

 

Is Amarra actively applying a sonic filter, even when Sonic EQ is turned off? I think its reasonable to assume that there is a default filter, with Sonic EQ providing a mechanism to tweak that default in directions that might be interesting or pleasing to a given user. But it's also reasonable to assume that iTunes processing is likewise like applying a filter (again, whether this is explicit or implicit in its design -- I have no idea if either is actually the case, but the example seems to work even if only as a metaphor) -- a filter that gets bypassed with the use of Amarra.

 

/End Speculation/

 

[Dave Clark]

meaning that when some says you got to have x to hear y I am always a bit skeptical... like with Amarra. You have to have this DAC and that RAM or whatever if you want to hear it. And yet people who do, do not hear it... and then people who don't, do don't.... and naturally vice-versa. In the end it either works for you or it doesn't... either way enjoy whatever it is you are listening to...

 

[Dave Clark]

I have had similar conversations with Steve as well. I doubt that my Playback MPS-5 is not good enough to hear Amarra, nor that it is doing 'something' to make Amarra moot. Why? Because I know of a few MPS-5 owners who do hear significant differences with Amarra and my MPS-5 allows me to hear all sorts of changes done to not only it but elsewhere in the system (cables, feet, etc.).

Ditto the Ayre QB-5. Some owners hear differences and some don't... so it ain't the DAC. I do like Clay's response though... well said.

The fact that I am not experiencing night and day differences here (more subtle in nature and it ain't just me as no one else who has heard it here hears much to report about either) does not mean that Amarra is not a product for one to dismiss. The fact that it has so many success stories means that it does work... but where and why is the 64 dollar question.

 

[cfmsp]

 

Scot,

 

My icon is a Picasso image called 'Acrobat', if memory serves.

 

I use it so that people won't take me seriously - the image conveying, as it does, someone trying to put their head up their own ass.

 

I've been meaning to switch for a change of pace, but now, perhaps I won't.

 

clay

 

 

 

 

[cfmsp]

 

Anyone have a good icon for 'devil's advocate'?

 

"Some owners hear differences and some don't... so it ain't the DAC."

 

hmmm... I'm not sure that follows. allow me to use a more familiar example as illustration. imagine that multiple instances of Amarra were feeding a preamp directly from the computer with the same inconclusive results as you describe above. would/could/should you say that it's definitely NOT the preamp when in fact it might easily be a function of the input impedance (of the preamp) producing variable results based on the output impedance of the upstream component?

 

I'm not suggesting that we know what the source of the mystery is, only that it may be due to some characteristic of the DAC and how it interfaces with the computer(s) on which Amarra is running.

 

does that make sense?

clay