MacMini or Win7 Music server? Help a fence sitter get off the fence..

[Tom80112]

Hey all,

 

I've been researching music servers and can't decide which

way to go. I read some posts which say a Windows7 Music server

in WASAPI mode sounds every bit as good as a Mac Mini yet so

many people use the Mac Mini.

 

Some people claim the Mac Mini sounds better than a Windows7 box.

Some people claim Windows7 sounds warmer and that Macs sound

more analytical.

 

Has anyone done a shootout recently comparing the New Mac Mini

using Pure Music to a Windows7 machine using James River in

WASAPI mode??

 

Specifically I was thinking of either:

1. New Mini/4GB/SSD/Pure Music using USB to my DAC.

2. Dell Zino DualCore/4GB/SSD/JamesRiver in WASAPI mode using USB.

 

I have to admit that I prefer windows but I would buy a Mac if it

produced better sound.

 

 

Thanks,

The Fence Sitter

 

 

 

[mjb]

There are lots of posts on this subject here, but no actual "shoot out" that I've seen.

I think the Mac is generally preferred because of the way “core audio” works: it’s apparently easier to get bit perfect reproduction out of a Mac. My gut reaction would also be to recommend the Mac, but that’s a personal preference!

 

 

[Audio_ELF]

The general recommendation for Mac is a lot down to ease of use rather than superior sound quality. It also has FireWire which some use to connect DAC to the computer. At a general level, the Mac is more of a plug-and-play solution: because all it's audio goes via the operating system's Core Audio there tend to be less problems with compatibility of DACs - with Windows most USB DACs work fine with WASAPI but occasionally you'll find there is a problem and then there is a lot of work to solve these problems.

 

With Windows you also have the versatility of greater choice of software - J.River; Foobar 2000; Media Monkey; WinAmp ... that can be a blessing or a curse - I've (personally) never seen the point of a Mac setup unless you are using iTunes as the basis (with Amarra or Pure Music fine) as, when combined with iPod Touch remote, this is where the Mac excels (IMO).

 

If you prefer Windows, go for it. Either the Dell Zino or you can build similar for yourself - check out Chris' C.A.P.S. article. Another option becoming more popular is to use Linux with MPD as the playback device.

 

Eloise

 

P.S. please note these are my general impressions rather than set in stone test results.

 

[PeterSt]

Hi,

 

I think it is time that I say something about the WASAPI blahblah, because in here, nobody seems to know. Of course everybody will think this is about me myself and I, but if you have the guts to listen a bit, it may be helpful to you (all). More helpful than deleting this post ... I sincerely hope.

 

As some may know, I was the first to come up with a WASAPI player. People loved it, and throughout the years I improved it significantly and people loved it even more. All using the software, made no secret of it that it was the very best around.

At some stage though I started to fall back on older technology, maybe because I didn't know how to improve the sound of WASAPI further. From that point on WASAPI was history; I don't think I have one single user left who is using it.

 

Like ASIO has a character which can always be heard no matter the playback software, WASAPI also has a charachter, with the nasty habit of being "technical". Analytical if you want. It is totally without emotion;

Since I'm a technical listener, WASAPI allows you to just do that. You can focus on anything you like. But with a more emotional engine this just can't happen. The music is always in the first place ... that strong is the difference.

 

Chris, let this post stay. You could benefit from it too. If you want me to say I have a commercial interest, here you have it. But it is not about that. WASAPI is just a bad thing, which sucks you into details which brings nothing but those details. There's no music in it. The fact that hundreds and hundreds of people don't use it anymore, should tell you all sufficiently enough. I even had it out of order for a month or so, and didn't receive even one complaint.

 

I hope this stays as a sheer good advice, instead of hearing the opposite each and every day, which is just the *wrong* advice.

 

Peter

 

[Audio_ELF]

Peter ... a few specific questions for you.

 

What API do you think should be used for a Windows system then? Are you saying ASIO is much better? What about interfaces where there is no native ASIO drivers - is ASIO4All preferable to WASAPI in this case too?

 

Eloise

 

[Tipper]

I’m fascinated.

So let me understand this; you have said its very difficult to tell the difference between say a 320 mp3 and a 44.1 but you can definitely hear the difference between WASAPI and ASIO and whatever the white fruit company use?

 

 

[PeterSt]

Haha, I can imagine you don't understand a f*ck of this, but yes, that is so. Or let's say this was so, because I have learned what to listen for regarding MP3's ... But forgetting this for now :

 

There is so much more going on when the "sound" leaves the player than many of us can believe, that the difference between, say, 320 and lossless (1411) is totally overrulled by it. Mind you, this really is "within one second" work.

 

It may sound new to you, but there's a whole world out there (nowadays) judgeing players for their merits, which is well beyond what we see in here (CA) between e.g. Amarra and Pure Music. Never mind this "well beyond" for now, but please see the difference between any MAC product (which are a relative few only) and Windows which (again, in here) seems to be "the other one only". And this is so far from the truth. But the point is, CA has grown to a MAC site, with hardly any counterweight from the Windows side of the game. As a matter of fact, you just saw how this goes : after so many months of reading about WASAPI and shutting my mouth because there's little use in it, I now at last had to say something, because I now saw that the OP tries to judge Windows and MAC, but doesn't care about all the players on the Windows side *and* using WASAPI on the Windows side to judge the both.

 

Each player sounds different. Most by now know it, for many this is not understandable (because we talk in the bit perfect realm in the first place). How to recognize what, is merely a matter of experience than good ears or equipment. Like the character of ASIO I mentioned. No matter what player, and no matter what exact ASIO implementation (and a few exist), they carry a similar sound (I would call it gray-ish). Such a thing can't be said from MP3 vs. WAV (etc.), where the differences ly in a very different area.

 

For me it is not so difficult to explain why it is so difficult to hear the difference between MP3 and WAV. But this is all very technical stuff and also very equipment related. It would be much more difficult to explain why it is so easy to perceive differences between players. And trust me, the differences are major. If you were here I could point it out, and right after that you could to it too. It is a matter knowing what to listen for.

 

Regards,

Peter

 

 

 

 

[Tipper]

Oh thats okay PeterSt, I've got a degree in Maths and Physics; I was an R&D engineer for an Avionics/Acoustics company for many years.....

before I turned into the unworldly halfwit you find before you.

Have a go, explain it to me.

 

Edit. Just in case you're still worried that i wont be able to keep up given I'm getting on a bit now I'll mention I've been building Hi Fi for 25 years now.

 

[PeterSt]

Hi, always good to see you asking questions, knowing that you will add it to good knowledge :-)

 

No, this is not ASIO. If you read around a bit you can see that it is not just my personal taste which tells it is no good. By now relatively few people are using it, but I guess the existance of WASAPI contributes to that. Why ? because WASAPI doesn't have a colouring character. ASIO does, and that is funest.

 

It is Kernel Streaming.

First off, I have always found this sounding the most neutral. But, together with that it lackes the detail of WASAPI. Funny thing is, together with *that*, it carries the emotion in music. As if both won't go together. Those who followed me a bit, know that I brought the first KS version as "the sound engine with emotion". And everybody agreed. Back then, at hopping back to WASAPI, I was always thinking what I had been missing the past weeks listening to KS. The additional detail of WASAPI is so inmense (at least in my implementation of it) that is seemed a waste not to listen to it.

 

But KS is very versatile, and operates right in the kernel of the OS. So, many things can be done with it, and many approaches can be taken;

I scratched the first version for technical reasons, but by now I have 3 very different KS versions (yea, differentiated within a seond again), and at least one of them is able to combine emotion with detail. So, it doesn't exclude eachother at all, but just takes much time to get it under control, so to speak.

And I am the most serious : if I compare this with the WASAPI version everyone raved about only 6 months back ... well, it is total rubbish. It really is.

Of course this is about (relative) evolution, and what I had 6 months ago still is infinitely better than what I had myself 24 months ago. It goes on and on, and I must add that KS allows this (much more than WASAPI).

 

So I can only say to everyone, no matter what player you use, use it with a Kernel Streaming engine (Foobar, MM, anything), and don't fall into the pitfall of detail. Besides this is the nature of WASAPI, there's nothing more easy than creating detail (tweak your crossovers and there you have it). But it is fake. It interests, but it's fake. Better catch yourself on being foottapping.

 

Peter

 

[PeterSt]

I see you are on the edge of being ironical, and I guess that's ok. But since you were wondered how both (the MP3 thing vs the player thing) could go along within me in the first place, why don't you take a shot at it yourself ?

 

And may you wonder how technical things can get, well read my posts. In here or in my own forum. I'm sure you will disagree with 80% of it, but I guess that implicitly tests me. No need to do that here again.

One thing you may perceive as special (FWIW :-) : there's not one single word copied from somewhere else. But this is how I proceed with audio.

 

I hope this suffices. Maybe another time ...

Peter

 

[Audio_ELF]

Follow up question Peter... Is Kernel Streaming possible with Windows 7? I was under the impression it was Windows XP.

 

Eloise

 

[PeterSt]

Already for Vista MS "claimed" that KS wouldn't be there anymore, or otherwise it would be emulated. Not true ...

It is still there, but what has changed is that -because MS claimed this- manufacturers don't have to comply to KS drivers anymore. For example, the "High Definition Audio Device" as it comes with nearly all motherboards doesn't support it. So, it is not part of the "Signing" procedure anymore, and with some bad luck newer sound cards just don't support it (because they don't need to). But, so far I haven't found one (via all the users) which doesn't support it under the hood (but that HD Audio Device).

 

Also notice that MS never officially supported KS, and it always has been like "if things break we are not responsible". And indeed, if you see how complicated things are in the internals, this would be a good advice.

Btw, this is why MS came up with WASAPI as the low latency official alternative to ASIO (for musicians hence the Pro world). But bow start lauging :

 

Even KS which is appreciated as very low latency, is guaranteed by MS only to 20ms. This, while people manage to play in XXHighEnd with ... 1 sample latency. Believe it or not (just pick your calculator and see how much lower that is than 20ms :-).

This is why I responded to some thread in here -that thread stating that Windows can't be near anything like a real time system- this certainly can be done because I do that myself beyond imagination. It takes some effort though (but Chris coincidentally hit the delete button at that post).

 

I think KS is alive as never before, partly because I reintroduced it as a new engine (what fool does that), but certainly also because the HiFace couldn't work otherwise at the start (which also was half a year ago).

 

Kind regards,

Peter

 

[Tipper]

So was that a yes or a no Peter?

 

[PeterSt]

Oh, I am sorry. A clear YES.

 

Btw Tipper, I am heading for a post you might like, in here : http://www.computeraudiophile.com/content/Hi-Res-article-Sterophile

 

Not only because of the subject of mutual interest, but also to let our before posts end somewhat more satisfactory. Ok ?

 

All the best,

Peter

 

Edit : I guess I will be making a new thread for it.

 

[Tom80112]

Thanks for all the info so far. I know this might seem

like a controversial subject but I have to think there are

other people like me who can't make up there minds on

Windows vs Mac music servers.

 

This is very interesting about the Kernel Streaming. I spoke

of WASAPI and JamesRiver for a music server only because they

seem to be the most talked about.

 

It seems really odd that bit-perfect from a Mac could sound

any different than bit-perfect from a Windows7 box if both

are fed async to a USB DAC but since I hear great differences

in power cords, I am ready to cast all notions aside.

 

[PeterSt]

Tom, our "problem" is that *everything* sounds different. But I guess it will be our new hobby. So, the same program on a (e.g. Win) laptop sounds different than it does on a desktop. But, its character will be the same.

 

You have no idea what the impact of these sub-micro details is on stuff further down the line. I say this (sub-micro etc.) because we are always talking about bit perfect operations, which leaves only jitter as the property which can influence. Or the carrier of it anyway.

 

When I talk about 1 second differences I mean that, with that only indicating again how huge the impact is.

But please notice that it takes some "better" equipment in the first place, although I'm sure not talking about 6 figure numbers.

 

Anyway, good luck with it, and if you can't decide you can always go for the comfort options (see Eloise her post).

 

Peter

 

 

[Audio_ELF]

Tom

 

At the end of the day, one thing you must ask yourself is how much any minutiae differences matter to you.

 

Most people you are looking to listen to music and the difference in quality between different software is (if perceived at all) similar to the difference between a Stradivarius and an-other professional quality violin. (Thanks to iamimdoc for this analogy) That's not to say they are not there, but only through detailed listening will you truly notice them - others may disagree by the way. Whether you go with Mac & Pure Music or Windows & J.River I'm sure (given good upstream equipment) you will get good results.

 

Eloise

 

[PeterSt]

Tipper,

 

Although I spent some two hours on my nice little story, I didn't dare to post it. Too pretentious, and maybe just a few weeks too early. Besides that, no matter what my stories are, they are without exception about my own products, and I don't think Chris likes that. It is the only way to proove my statements though, and you don't want to know how many dozens of posts I wrote but never posted.

 

If you like it, I may consider to PM it.

 

Anway, what I wanted to tell you today is that you actually got me thinking about the MP3 vs WAV and the player differences. Or better, you surprised me a bit with your question.

I don't have much more to add than I did yesterday in the earlier post, but it urges (me) to find better answers than what I wrote (which *is* what I think at this moment), but which can't be the real explanation. I mean, if the player differences can be incurred by jitter only, how can an MP3 with the bytes different all over have less impact. That this is so is definite (for me), but the why is another matter.

 

At this moment I must stick to the phenomenon "character" of music, where a constant same influence (the player and PC behind it) will create resonances which are devistating, that implying the character. This can't be the case with MP3, where the difference opposed to the original is a deviation of the original itself. No (strong) resonances applicable here. Just another version of it, and hard to distinguish.

 

However, the means to quite easily recognize MP3 from WAV is funnily enough the bass. How ? it has a strong "on-off" vibration. Aha, that smells like resonance again (now think how MP3 compresses). Sadly it isn't that easy, because to my experience, the less jitter is fed to the analogue system, the more this same on-off exhibits with uncompressed music, while you can't even recognized what is better (more real). This is because it implies a long wavelength, and the more you hear the on-off from it, the better your system will behave.

 

Related to my unposted story (but if you'd read that, you'll understand and will agree up to 60% or so), is that I have technical reasons to believe that MP3 may let perform your system better than uncompressed music. Mind this "let your system perform", because at the technical level it is all about that. Just look at it the other way around, with this example :

 

When I am capable of pushing through high transient data at an unexpected (for the ready built electronics) rate of 352.8 (4 x 44.1), it may happen that your capacitors can't cope. Of course, they have specs, and the designer calculated in everything, but forgot to look at square waves, which is the representative of high transients. This story is much longer because it starts which NOS DACs and proper filtering, but with what I do (in electronics) these high transients remain as they were in the original wave form, and thus the caps choke (or the chokes cap(itulate) haha).

The other way around again, and we have MP3. Not that the sample rate changes, but it is all more flattened because of the compression process, and it *will* sound better. It lets your system behave better ... if that isn't on par in the first place.

 

Lastly for now, and totally unrelated, a teaser :

We need much current to let the bass come trough properly, right ?

Wrong.

It is the high frequencies which eat the current for the bass.

Yea, if you think about it (with calculator) you might even agree.

 

Peter

 

[PeterSt]

Whether you go with Mac & Pure Music or Windows & J.River I'm sure (given good upstream equipment) you will get good results.

 

Eloise, excuse me. Of course you will be right, but only with your reference of what "good result" is.

You could try to qualify the distance to live music and your good results. Even if you'd say "50%", you will be far from what can be achieved. But better say 5%, which is what I expect from a moderate good playback chain.

 

Is this important ? that depends how crazy you are, and whether you are obsessed with achieving the "live music level". This answer is rather clear too : we will never be able to achieve that, so don't bother and let's have our own reference. Good for the money too.

 

But now I tell you in between the lines that what I have here *is* just that live music level. No wait, I am not telling you, I am only knowing it. If I now read your post it doesn't make sense to me.

So you see, it is always dangerous to write these things, because you can't know the reference of others.

And might we have achieved that live level, it is very comfortable, because it is an absolute measure. I don't need to explain it to you.

 

Not intended as nagging, but maybe to think over.

 

All the best,

Peter

 

[Audio_ELF]

You are correct Peter talking about "good results" is a relative term.

 

To my mind, recorded and live music are two completely different things. In many cases the absolute "quality" of live music falls far short of the possible quality of recorded music (contravercial I know). Live music is about the atmosphere much more than absolute sound quality. Recorded you can pick up nuances more in the playing. To use my earlier analogy, you are probably more likely to be able to tell if someone is using a Stradavarius on a recording played back on moderate high end equipment than during a live performance (excluding the site aspect of live of course).

 

So my reference to good results would be compared with the best equipment I've heard. The original poster however didn't give any reference to the level he is expecting. Is he using NAD and Cambridge Audio level equipment, or more Weiss and ATC (sorry those are aribary thoughts not to put down any equipment).

 

You are, if I may say so, at the crazy obsessed level - you are designing your own DAC IIRC. The majority of people here (even for an audiophile site) are the more casual, sane people who listen to music and want the best within certain level of faffing (IMO). Maybe this last thing is something you also need to consider when you answer questions.

 

Eloise

 

[PeterSt]

You are, if I may say so, at the crazy obsessed level - you are designing your own DAC IIRC. The majority of people here (even for an audiophile site) are the more casual, sane people who listen to music and want the best within certain level of faffing (IMO). Maybe this last thing is something you also need to consider when you answer questions.

 

Haha, you are right of course, but it is not as simple as that. Thousands of communities like in here exist, and one of them is my own, were it for forums. Now, I could try to be as dumb as possible and contribute nothing in the end, but I think it is more logic and helpful (in the very end) if I use the standards from "overthere". Of course I don't need to remind you that the site here is called Computer Audiophile, and not Computer Playback, and I try to use this to the best of my knowledge, while it is my job in the mean time.

 

What I meant with "live music" was my means to express how the artists sound at taking the recording. So I explicitly did not mean a live performance as such. It is about the 1:1 representation of whatever it is we never will know, but which is good because it gives us some headroom, just in case :-).

 

Eloise, please try to understand;

At some stage things may really be happening when it comes to music reproduction through loudspeakers. If this were so, and you yourself caused this major step everybody will profit from ... are you going to keep it a secret, or will you be sharing ?

 

Peter

 

 

 

 

[Audio_ELF]

Peter... You are right that this site's primary aim is the pinnacle of music reproduction using computers as the source/transport. However we are at times in danger of alienating people who find the site looking for lower end solutions and we need to moderate our advise accordingly.

 

I would ask you to consider if you have really added anything but confusion to the original poster's "Mac OS X or Windows" question. It maybe the OP is looking for the pinnacle, but without knowing his reference points we don't know - hense my asking him in one of my last few posts.

 

Eloise

 

[PeterSt]

I would ask you to consider if you have really added anything but confusion to the original poster's "Mac OS X or Windows" question

 

I understand what you mean, and certainly feel the deviation from the original question. But you have to understand (ok, if you like) that there can't be a simple one liner from my hand because it won't be accepted by most. So, what I always do (as you know) is giving as much context as is within my power, in order to next let the reader decide for himself.

 

If I would not have been on tiptoes (which is necessary in here without exception, just because it is me), my first post in this thread would have been :

 

Go Windows. It has infinitely more solutions for audio playback than the MAC, while the MAC solutions seem to be an endless debate without conclusions whatsoever. The MAC solutions are so few that you really may ask yourself -just by looking at statistics- how you coincidentally could meet the best product for SQ in there. Besides that, there is no single proof to me (and not only me, but many others tried) that explicit reasons exists for the known MAC players to sound better. It is all coincidence, just because *all* players sound different. Not so on Windows. There, a few players exist that -for many years now- explicitly work on sound quality in the bit perfect realm, while the means they achieve are known and discussed.

Lastly, I know of no person who had the choice, who hopped over from one these products to a MAC product. However, the other way around situations are numerous.

Of course I am only talking about the sound quality, knowing that at least one of these Windows players I am referring to, doesn't even have a user interface at all.

 

This is what is in my head, this is what I really mean to say, and it is as honest as can be. But now think. Everybody -without a single exception- can write this. But not me.

 

I am here on this forum to help as you do. Now look what fuzz I have to make (my first post) in order to more or less say the same. It is as true, as honest, but not what I really wanted to say, because it is an indirect way from here to Tokyo. It takes huge amounts of time, and then to think that Chris may delete my post anyway.

 

Peter

 

 

 

[Tipper]

@ PeterSt

I have to be honest here Peter and say I’m having trouble distinguishing the information form the static in your posts

 

I will hazard a guess that English isn’t your first language (?)

 

I can’t help but admire your enthusiasm for your subject whether I agree with your opinions or not.

You remind me very much of some game coders I know.

I also have to agree that it is possible that different music players have their own sound.

 

For example, I prefer foobar2000 to Winamp. I have absolutely no rational explanation as too why this is, it’s totally subjective but none the less quite definite. Given that foobar2000 isn’t the most attractive UI available and getting it to work with various components I have found to be a bit of a nightmare, the perceived sonic benefits must be considerable.

I do not know much about computer sourced audio and I’m still looking for scientific explanations as to why different players can sound different to each other.

You obviously know a great deal about this subject from your own project development so I was rather hoping you could provide some rational explanation, preferably one that doesn’t involve too much subjective stuff…….and a lot less static

 

The capacitor debate.

This is as confusing and mystified as the cables debate.

I try to stick to what is and isn’t measurable here.

I have no doubt that there are sonic benefits to using polypropylene caps in speaker crossovers for example, but given they can have 50X the response speed of an electrolytic this is hardly surprising.

The audible effects of construction materials and wrapping density etc is an interesting topic but possibly not one to go into here.

 

With regard to this site and helpful contributions.

I’m fairly new here and the reasons I keep reading and contributing is because in all the rubbish there are contributions from people such as Eloise who obviously try to not only differentiate the subject from the objective but also consider what is relevant to the posters financial circumstances and performance requirements.

(Way too many posts implying you need to go and spend so many thousands of dollars to appreciate music and join the true audiophile club)

As for the prospect of having your posts deleted by Chris. I have to say that unless a contributor is blatantly advertising their own products in my experience so far he is remarkably tolerant about post content.

 

 

[Tom80112]

Some wondered what sort of system I would be using so as to

gauge the level of performance I might be expecting.

 

My system includes a Havana Tube DAC, Naim XS integrated amp,

B&W Matrix 803 speakers and Analysis Plus Solo Crystal cabling.

I prefer a warmer "analogue-like" sound with good rhythm, pace,

and timing. I prefer emotion and good decay times in music over

detail. IE: I have no issue giving up some detail for emotion.

 

I mentioned the Dell Zino and Mac Mini setups such that the

folks here could guide me to the Zino or Mini based upon my

above preferences. Thanks for all the suggestions.