Pure Music

[cfmsp]

 

Ted,

 

I am truly sorry to hear this.

 

Take care of your family - and don't forget to let them also take care of you when needed.

 

Bless you,

Clay

 

[chrille]

Hi again Paul,

good to hear you were not offended.

I have somtimes offended people here .

I too paid the 79 dollars while the offer was still valid.

Since there are still a lot of valid discussions going on here regarding both PM and other software players,let me just add that all I am saying is that PM sounds clearly better than Itunes.

I am not saying it is in every respect ,as good as listening to a masterfile from a Pyramix a Meidtner DSD or DAD /DXD at sessions.

But unlike Itunes it sure makes hi rez digitally recorded music sound very nice indeed.

 

We are IMHO, entering a very exciting era in home audio, with more and more recording companies actually letting us hear their recordings as they were meant to be heard, without the compression,low resolution and other digital "nasties" the CD era imposed on music lovers.

My main concern now is that I want more, much more,,and I want it now!

 

With the exception of the real Pioneer Morten Lindberg at Norwegian label 2L, whose own website has not only got some of the best recorded music for download at the absolutely fastest speed.

I download at speeds up to 10mbs from 2L.

Some others are still embarrassingly slow.

Notably Classical,net where some titles from Chandos are available in stunning hi res quality, but ridiculosly slow speeds.200-300kbs.That is really almost a joke by today´s standards.

It may have been ok for mp3 downloads but certainly NOT for hi rez files which sometimes ,with classical material are in the region 2 to 4 Gigabyte for a full album.

 

Linn has improved a bit recently and I am downloading on average at speeds around 700-900kbs from them.

But there was one instance this past weekend when I was getting 55000kbs !

I downloaded a 67 minute Channel Classics title in about ten minutes .

I don´t now what happened?

But it sure was good.

The site with the most hi rez downloads of all, is still limited to downloads within the USA and that is very irritating.

 

But I am hopeful that in all respects things will just improve from here. And even as things stand today, it is definitely a very good start.

When I first started listening to music via my Macbook Pro it was with lots of reservations and only when travelling. Now I sometimes even prefer it to SACD or LPs at home.

All the best Chrille

 

 

[bdiament]

Hi clay,

 

"For a data file to be bit perfect (as compared to the file received as input), it must have been rounded back (e.g., integer to float and then back to integer) to the same value it had prior to the 'processing'. IOW, any decision making on the part of the developer as to when to round (that resulted in anything other than bit perfect output) would be what I referred to as 'additive' in nature, i.e. not completely accurate to the input. It might sound better, but it wouldn't be the same as the input."

 

Yet we have both encountered apps that provide "bit perfect" results, yet sound different from each other. There would seem to be more at play.

 

To be clear, my own reference for "better" is an output that sounds like the input. (I don't personally like hardware/software that "improves" the sound of the original, though I know folks that do and feel it is just as valid a path to take, since the goal is deriving enjoyment. It just isn't the path I want to take.)

 

 

 

 

 

"Here's a far out idea - suppose that improper (ie. unnecessary) 'rounding' somehow resulted in very subtle differences in signals between left and right channels, or the equivalent of an increase in the difference / side signal, or similar?"

 

I don't know whether rounding (proper or not) would be the reason but the change in sonics in precisely this regard is what one of my acquaintances has mentioned.

 

Well, it sure will be interesting to see where this still nascent world of music servers is going to take us.

 

Best regards,

Barry

www.soundkeeperrecordings.com

www.barrydiamentaudio.com

 

[bdiament]

Hi ted_b,

 

So sorry to hear this terrible news.

Please take care of your family and yourself.

Our hearts are with you.

 

Best regards,

Barry

www.soundkeeperrecordings.com

www.barrydiamentaudio.com

 

[timztunz]

Dear ted_b,

 

Please accept my most sincere condolences for you and your families tragic losses. You and your family will be in my thoughts and prayers and if there is anything I can do for you please don't hesitate to ask.

 

Kind regards,

Tim

 

[cfmsp]

 

"There would seem to be more at play."

 

Totally agreed.

 

I'm no expert on the topic of noise (AC-related & otherwise) - indeed, I can barely make it down the blue slopes - but those who know about these things believe energy is better spent in this area than in wondering which software player 'sounds' better. Sadly, they no longer post here.

 

I believe that the processing 'footprint' of an application can impact the(ir) sonics. But here, we get some conflicting results from Macs versus PCs, which doesn't make sense to me. Heck, even with Amarra, it sounds different/better and yet it's increasing the processing footprint by requiring that iTunes be running simultaneously. Perhaps this allows the opening for PV/PM to sound 'better'? Even PV is not the lightest weight app out there.

 

I can see how hard disk access could have an impact on sonics, but there have been many reports of RAM disk, and 'memory play', having a positive impact on sound even when SSD's are installed, which also doesn't make a lot of sense to me either.

 

Lots to sort out it seems.

 

Clay

 

 

 

 

 

 

 

[Lars]

Ted,

 

I'm sorry to hear of your loss. This audio stuff doesn't seem that important when things like this happen.

 

Best regards,

 

Steve

 

[ted_b]

I really didn't mean to hijack this thread with my personal issues but I've come to realize that this CA forum is really a close community, and I should have expected as much. Thank you, thank you. It's been very tough and this kind of support from, in most cases, people I've never truly met, is quite something.

 

Now let's get this back on track.

 

[bdiament]

Hi clay,

 

"...and yet it's increasing the processing footprint by requiring that iTunes be running simultaneously...

 

I believe the "processing footprint" only becomes an issue when it exceeds the capability of the hardware. Keep in mind PM has iTunes running simultaneously too.

 

In the end, I don't think the load on the CPU, presented by these two apps is significantly different.

 

Best regards,

Barry

www.soundkeeperrecordings.com

www.barrydiamentaudio.com

 

[Guest]

Actually the processing footprint is but one variable in the whole equation of the "computer" portion of Computeraudio. Processing footprints of the music programs as well as every other process running on the machine can affect the outcome. This also includes other variables such as RAM timings and CPU power consumption. Much of this has been laid out in Cics's research and verified by many computer audio enthusiasts.

 

It is advised to disable and/or optimize as much as possible any process that is not critical to the operation of the computer. Programs like WLAN Optimizer and Vista Anti-Lag for example do a great job at reducing CPU latency spikes due to WiFi on Windows machines. WiFi probably being the worst because of its nature in Windows to cause spikes in CPU usage.

 

With Mac, there are similar tools but the Mac OS being much like Linux has not been explored and/or exploited as much as Windows yet. However incremental steps are being made by the PureMusic folks to bring Mac into music Nirvana. For example just recently they announced they are working on being able to bypass the 32-bit floating in Mac and send the Raw PCM data directly to your DAC. Those who follow the details surrounding the controversy of processing 16-bit data to a floating 32-bit might understand how removing this step should make Pure Music sound even better.

 

Until then, I suggest that people first consider and understand every variable involved in playback....from and including the power source up until the sound reaches its final destination. Remember every aspect is but a link in the whole chain, and optimizing or even making perfect "one" link will not necessarily make the whole chain stronger, but most likely weaken the whole chain. Keep a holistic / synergistic / realistic mindset.

 

 

 

[Paul R]

There is a lot more to it than just hardware capacity Barry. MacOS is, for all intents and purposes, BSD Unix tuned to provide a responsive user interface and generally smooth operation. Windows is a completely different animal, and though both of them look superficially similar, they are not even close cousins under the covers.

 

Take for example, how the two different OS's multi-task. MacOS (nee BSD) has a very complex scheduler and dispatcher, one designed originally to do two different tasks - support hundreds of users on (by today's standards at least) very modest hardware, or to support graphics intensive work by a single user on a single machine, but still handle background tasks like remote file sharing (i.e. NFS). Windows has multi-user / multi-tasking "bolted on" as it were, and it still shows that heritage. I really wants to be a single user / one task at a time OS. That it does so well as a multi-tasking / multi-user OS is testament to the genius at Microsoft.

 

By the way, BSD works even better on a PowerPC than on an Intel chip, because the PowerPC chip architecture is closer to the original design - which was based upon DEC machines. A VAX in the case of BSD 4.x, which is a direct ancestor of MacOS.

 

All of which is to say, the OS is going to try and smooth out the processor performance to a very high degree, and even manipulating process priorities is not going to change that a whole lot - not with these programs. I mean that to say, none of these programs runs the processor to 100% utilization, they spend too much time in I/O, graphics routines, and other system calls.

 

Of course, if you instrument a machine to record this, you change the load and processing paths. Not enough to radically change the measured usages, but enough to change the sound that a program reproduces. Catch-22 there.

 

-Paul

 

 

 

[cfmsp]

HI Barry,

 

Perhaps my terminology was not the best.

 

By processing footprint, I mean the sum total of the actions occurring in the computer that are required by the application - NOT just the processing itself.

 

E.g., this would include disk access, AC draw, etc. Any of these might impact other events occurring within the computer simultaneously, e.g. the clock being used by the device drivers, etc.

 

Not the least of which is the noise component being introduced, and dare I say it, then modulated by the processing.

 

 

EDIT:

"In the end, I don't think the load on the CPU, presented by these two apps is significantly different."

 

Totally agree with you on this point, and also that modern computers have more than enough processing capability that it is unlikely to be the limiting factor.

 

Interestingly, however, it's been posited that Macs somehow sound better with MORE capabilities, whereas PCs sound better with LESS capabilities. For example of the latter, refer to the work of cics (which Dynobot mentions). For the former, there are claims (unverified by me) that increasing the amount of memory will improve the sound of, e.g., Amarra EVEN with the prior installation of an SSD (which, of course, eliminates need for hard disk access, and has much of the inherent advantages of memory).

 

 

 

clay

 

 

 

[bdiament]

Hi Paul,

 

My post was a short one and perhaps oversimplified the point I was trying to make.

 

I understand what you're saying. (As an aside, I've done some programming myself, though much of it was mainframe -COBOL, JCL, CICS, IMS, DB2-, with some AS400 -yechh-. Did some C and VB too. Though I'm sure you can tell from this, it was a while ago. Now its just XHTML, CSS, PHP and a little Javascript for the BDA and SR web sites. Interestingly, to me, much of the code in my beloved MH devices is done in Assembler. !)

 

Best regards,

Barry

www.soundkeeperrecordings.com

www.barrydiamentaudio.com

 

[Paul R]

Hey Barry -

Yep. My day job here involves millions of lines of HLASM, but since it is the 21st century, it is running on a zSeries under Linux with a z/VM Hypervisor.

 

MMM... an old z800 would make one HECK of a music server wouldn't it... ?

 

-Paul

 

 

[Paul R]

After chatting with Rob in e-mail last night (and leaving him perhaps a little puzzled as to why I didn't understand what he was saying), I think have finally *got* what he has been saying about NetSend. It *is* a way to send high resolution audio around - and it looks really sweet!

 

I've barely scratched the surface, but I think this is going to make a LOT of people happy. It seems to be a very nice step up from Airport Express streaming music.

 

If this capability is as good as I think it is, then Pure Music is a VERY good deal indeed - worth much more than they are asking for it.

 

More as I figure it out. If anyone else has it working and can comment on simple setup instructions, I am all ears!

 

-Paul

 

 

[Mr.C]

I've done this successfully. The critical piece is finding software that can function as an audio generator/ the netreceive function. Audiofile Engineering's Spectre does this I believe, or you can use AULab as a part of XCode, available free from Apple. Once you figure out where and how to setup netreceive (ie. in Spectre or AULab), then you just start netsend in PM and point the Netreceive portion to that name (Turntable I think).

 

BTW, Airfoil/ Airfoil speakers does this as well in a way that works with more programs. I haven't compared the two to see which sounds better.

 

[Paul R]

Thanks for the tips, I've been using AULab from xCode, and once one penetrates the obscure paradigm it uses, it opens an amazing set of capabilities I didn't know how to access on the Mac. I only got it working last night, then had to stop to get some sleep. It's exciting though!

 

Airfoil / Airfoil Speakers is a really great application- but when used with Pure Music I find that there are dropouts caused by mismatched bit rates. At 16/44.1 or 24/44.1, if everything matches, it works fine. If it doesn't match, then unexpected dropouts occur when transmitting to an Airport Express. AirFoil Speakers seems better able to handle that, but it is probably because AirFoil speakers are running on a PC or Mac with gobs more processing resources than an AirPort Express.

 

I'm suitably embarrassed that I didn't know about AULab, as I have been coding software on Macs for ages. Indeed, I have my original Lisa still, in case someone needs something coded for a 128K Mac. Ah well, one thing about my line of work, it keeps one humble!

 

-Paul

 

 

 

 

[Paul R]

By the way, if anyone didn't notice (ahem... like me for instance) the Pure Music License allows you to use one copy on multiple computers legally.

 

This is cool.

 

-Paul

 

 

 

[Phil Townsend]

One of the things I like about Amarra is the EQ.

Pure music does not have it... But on the other hand Amarra does not do digital xover and that for me is most important!

I love to be able to chase down the right xover point. The "Overlap" feature is so cool I can hardly stand it.

I'm thinkin the Channel D folks will develop some form of EQ as well. With a calibrated mic input to help zero out the room.

I use a Lynx Aurora with firewire so the interface would be easy at least for me.

 

I have waited to buy the Pure Music software until it was more expensive...

 

 

[coot]

Give me PM w/PEQ and a visual display of settings, please!

Oh,we're such a demanding bunch...

 

[cfmsp]

 

Phil,

 

Steve/Lars posted that Pure Music will soon accept plug-ins, which will allow use of your preferred EQ tool.

 

Perhaps this will meet your needs?

 

Clay

 

 

[Lars]

I would like to share Rob's discussion about floating point and hog that you should find very interesting. Also, with the next release of Pure Music, the activation code for Pure Vinyl should also work with this program for the impatient guys like myself.

 

Rob's e-mail to me:

 

I think there is some confusion regarding "floating point conversion"

in CoreAudio. Nothing is really "converted" per se; the audio is just

placed into a different container (provided that the audio device

driver does this properly, which isn't necessarily the case). This is

the same notion as lossless audio compression (though that's not the

goal here), in that what comes out is exactly the same as what went

in. This is the responsibility of the audio driver, and floating

point, being the canonical format for CoreAudio, is what's normally

handed to the audio software by the driver, and vice versa, unless the

software takes special efforts to access the integer stream (see below

about Hog mode).

 

Pure Music has been verified to be bit transparent / perfect by a

poster in the CA forum thread, and independently by at least another

two other prominent, technically oriented people in the industry (one

of whom is mentioned in the Pure Music documentation; during his

testing, he found a problem in the Pure Music upsampler, which was

corrected immediately.)

 

That said, there are folks who might think that something "happens" to

audio when converted to a lossless format and then back again, even

though the two can be compared and proven to be identical. Maybe it's

related to physics that someone like Stephen Hawking could explain,

because this, at least to me, borders on acknowledging the influence

of mystical or supernatural phenomena (which can seem as strange or

impossible as certain topics in particle or cosmological physics).

Don't get me wrong; I don't mean to be sarcastic at all, only

interested in finding an explanation, and maybe there's a reason.

There are a lot of things we don't know about our universe. But for

the time being anyway, most folks can accept that audio "works"

according to conventional Newtonian physics, and that there's very

little mystery involved.

 

If we are talking the use of CoreAudio "Hog" mode and native integer

support and supposedly "bypassing CoreAudio" (and one can't do that

without first creating a device driver, which requires thorough

knowledge of the hardware details of EVERY conceivable piece of audio

hardware / DAC, etc. that it will be used with; and I doubt that audio

hardware manufacturers would care to release the necessary

information, even under an NDA arrangement), we are incorporating Hog

Mode, but whether or not it has the potential to make any sonic

difference, at least from a technical perspective (same argument as

lossless compression vs. linear PCM) is open to discussion. It may

slightly reduce the overhead of the audio driver (except that with

float to int conversions, we are talking changing the CPU footprint by

a miniscule amount). My impression is that "Hog" mode (exclusive

driver access would be a better term, but Hog conveys the idea more

succinctly) is included in the CoreAudio API to keep a developer from

trying to hack a similar functionality. Hog Mode is as old as

CoreAudio (OS X 1.0, nearly 10 years ago), and isn't something that's

generally used. Hog Mode's behavior is what it sounds like: not a nice

citizen on the computer! But it's now an optional setting in Pure

Music 1.1 (Pure Vinyl imminently) for the folks who want to use it

with compatible devices (Hog mode can optionally be enabled now;

tapping into integer streams is coming very soon).

 

Regarding the CPU footprint... it depends it you have lots of options

turned on or not. The CPU footprint also peaks when tracks are being

loaded for memory play... this is to be expected, but doesn't matter

because no audio is playing at that time (unless Pure Music's Hybrid

Memory Play mode is enabled, and you are willing to accept the

tradeoffs of that). With the proper settings, and these are explained

in the documentation, on a 2 GHz Mac Mini, the footprint is under 2

percent (playing redbook CD from memory at native sample rate), with

iTunes adding less than another 1 percent. This is even better than

the posts on our website for versions of Pure Vinyl, from last year.

It is very cool to watch Disk Activity in Activity Monitor and see it

at zero (after loading tracks, and nothing else happening on the

computer, and enough RAM to avoid VM paging, etc.) We've taken a lot

of effort to tune and balance the CPU load of the various internal

processing threads, and priority is given to the important ones.

Priority defined as the ability of an important thread to temporarily

shut down or pre-empt a less important thread; this is done in the CPU

hardware, and can be manipulated by the developer. For those folks

finding that handshaking between Pure Music and iTunes isn't as

instantaneous as they'd like, we may provide an option to adjust that

balance for themselves (similar to renice, except on a per thread

rather than per application basis). This capability was in earlier

builds of Pure Vinyl and it was removed to reduce complexity, but it

can be returned if necessary.

 

Finally, we've decided to allow those with a Pure Vinyl activation

code to be able to use it in Pure Music, as well (this will be enabled

in the next release of Pure Music). The reason is because Pure Music

launches noticeably faster than Pure Vinyl; some initialization and

"preflight" audio validation needed for reliable recording can be

skipped over. While there is tight integration between the many

features in Pure Vinyl and the iTunes Music Player mode, and it's easy

to instantly switch between iTunes, recording, or playing back

archived "flat" vinyl recordings (and soon high-res editing), it can

be more convenient to just launch Pure Music if you only want to

listen to tracks from iTunes.

 

Thank you,

 

Rob Robinson

 

Channel D Support

 

 

 

[Phil Townsend]

Oh yes Sir, That will do me just fine...

Clay thanks for the info!!!

I am, all atwitter with delight.

 

Yes, My Mom was right... Go to school... learn something.

 

Since I have had the good luck to get what I have wanted since joining this Group...

Here is one more thing I need...

 

Atomic Clock...

who has one for me...cheap!

 

 

[Lars]

"Steve/Lars posted that Pure Music will soon accept plug-ins, which will allow use of your preferred EQ tool."

 

I did ????

 

[Jim L.]

Rob,

Thanks!! I appreciate allowing us Pure Vinyl owners to also have access to Pure Music. Very kind of you!

 

Jim