Uh oh, I beat my SD card trick; Bypass your Ethernet switch, and make your external drives sound close to RAM disk using an Apple Thunderbolt Ethernet Adaptor and a second network connection.

[Audio_ELF]

and a guy who codes a memory player who presumably knows more about this than we do says that's not crazy.

Actually from recollection Peter (XXHighEnd) says (or at least has said in the past) that it is crazy but it happens anyway!

 

Eloise

[wisnon]

Indeed, I would love to try out an RTOS if good player/library management s/w was available for one. Anyone know of such a combo that I could easily install on an i5 Intel Mac mini?

 

What's funny is that Windows CE--available to developers for a multitude of processor architectures--is now a real-time OS, and typically used for embedded applications. I wonder what it for take for PeterSt. or Miska (XXHigh-End and HQPlayer respectively) to get their products running under Windows CE? They might both shudder at the thought, but who knows...

 

Mhhh?

 

Windows CE running on minimalist hardware....sounding promising indeed.

[Audio_ELF]

But Jud, the typical muli-purpose computer is a sea of NOISE. Complete overkill too. That is the point.

Just a thought - but have you considered taking a BeagleBone or Wandboard, running a minimal Linux install (maybe even RT Kernel) and MPD which reads from a SD Card.

[wisnon]

Actually from recollection Peter (XXHighEnd) says (or at least has said in the past) that it is crazy but it happens anyway!

 

Eloise

 

Says WHAT is crazy?

[wisnon]

Just a thought - but have you considered taking a BeagleBone or Wandboard, running a minimal Linux install (maybe even RT Kernel) and MPD which reads from a SD Card.

 

Eloise,

 

That is above my pay grade. I am no engineer and most of the techie stuff is above me. I am only dealing in big picture ideas...

 

Having said that, what you propose sounds like a good direction. Can you do it?

[Audio_ELF]

Say WHAT is crazy?

Sorry ... it's crazy that the "source" of a file loaded into memory for playback matters.

[Jud]

Sorry ... it's crazy that the "source" of a file loaded into memory for playback matters.

 

Sounds like Peter.

[goldsdad]

Sorry ... it's crazy that the "source" of a file loaded into memory for playback matters.

 

A non-crazy explanation is placebo.

[Jud]

But Jud, the typical muli-purpose computer is a sea of NOISE. Complete overkill too. That is the point.

 

Yes, I agree.

 

Be careful, though, because some of this stuff can be counterintuitive. Is it a computer with greater capability running at a small percentage of its limits, or one built with minimal capabilities running nearer its limits, that produces more noise?

 

If you want the latter BTW, Miska has done it with his NAA, so the concept is available to try now.

[Jud]

A non-crazy explanation is placebo.

 

Sure, that's one. Another is that this is the way Peter expresses himself. Considering Peter has extensive instructions on his website regarding playing from RAMdisk, I think the latter is a reasonable conclusion. Of course Peter could be wrong, and if you have greater knowledge regarding how memory players work than he does, we should rely on your reasoning rather than his.

 

Edit: I don't want that last sentence to come off sounding at all sarcastic. I really mean that in a completely straight way, whoever is more knowledgeable is more deserving of our reliance. (There's very small cost to reliance in this case. A script to set up RAMdisks is trivial and costs nothing.)

[wisnon]

Sorry ... it's crazy that the "source" of a file loaded into memory for playback matters.

 

He could refer to embedded Jitter. The Porzilli Memory Player purports the same hypothesis.

[wisnon]

Yes, I agree.

 

Be careful, though, because some of this stuff can be counterintuitive. Is it a computer with greater capability running at a small percentage of its limits, or one built with minimal capabilities running nearer its limits, that produces more noise?

 

If you want the latter BTW, Miska has done it with his NAA, so the concept is available to try now.

 

But with a high speed non-data shifting SD Card?

 

Also is not so much the POWER of the computer, per se, but the amount of things its designed to do. it can be limited in its applicability but powerful in having lots of headroom for the single task.

[Jud]

But with a high speed non-data shifting SD Card?

 

That's just the storage. You have to consider all the requirements, including software/firmware and a CPU to run it.

 

Edit: Regarding the rest of your comment, sure, minimizing the tasks the player must perform is something we are all looking at various ways to accomplish. I can tell you though that I've run mpd from the console (no graphics) in a minimal FreeBSD installation and it was not nearly as good IMHO as HQPlayer on Linux, XXHighEnd on Windows, or A+ on OS X. So minimalism by itself is not a cure-all.

[goldsdad]

Sure, that's one. Another is that this is the way Peter expresses himself. Considering Peter has extensive instructions on his website regarding playing from RAMdisk, I think the latter is a reasonable conclusion. Of course Peter could be wrong, and if you have greater knowledge regarding how memory players work than he does, we should rely on your reasoning rather than his.

 

Placebo is a widely-accepted phenomenon and I'm only suggesting it is one possible reason. Does Peter actually present any reason counter to placebo?

[Daudio]

Indeed, I would love to try out an RTOS if good player/library management s/w was available for one. Anyone know of such a combo that I could easily install on an i5 Intel Mac mini?

 

What's funny is that Windows CE--available to developers for a multitude of processor architectures--is now a real-time OS, and typically used for embedded applications. I wonder what it for take for PeterSt. or Miska (XXHigh-End and HQPlayer respectively) to get their products running under Windows CE? They might both shudder at the thought, but who knows...

 

I left Windows CE off my list due to my general disdain for all things MS, and thinking that Linux based systems might be more compatible with OSX, but perhaps the driver situation might favor CE. I would guess that it should run on a Mini like any other Windows installation ?

 

I think this cage is worth rattling a bit more to see what might fall out

 

 

P.S Just won a 2010 Mac Mini w/ 8GB on eBay last night. Yeah !!

but lost out on a cheaper (?) one w/ 16 GB - ah timing...

[elcorso]

Sorry ... it's crazy that the "source" of a file loaded into memory for playback matters.

 

What is not crazy in computer audio?

 

The only logical explanation I found in some of the Alex Crespi (Superdad) multiple threads is the following:

 

 

This about storing music on an SD Card, but could apply to the OS & music player.

 

I have no experience with XXHighEnd, but it looks to me that it's 'computer process hungry' music player, then being in RAM it should help.

 

My experience is that it helps a lot storing music on an SD Card, the same experience with SSD under Thunderbolt.

 

Regarding if this placebo or not, I don't see anybody that is not happy when using this, but the contrary.

 

Happy listening,

 

Roch

[goldsdad]

I have no experience with XXHighEnd, but it looks to me that it's 'computer process hungry' music player, then being in RAM it should help.

 

The player will be in RAM, anyway. It needs to be in RAM to be run! It'll be run in RAM regardless of whether it has been stored on a RAMDisk (a virtual drive) or real hard drive.

 

The program instructions are first loaded from a drive (real or virtual) into RAM, then the instructions in RAM are processed by the CPU without repeatedly re-reading these instructions from the drive on which the program was stored.

[elcorso]

The player will be in RAM, anyway. It needs to be in RAM to be run! It'll be run in RAM regardless of whether it has been stored on a RAMDisk (a virtual drive) or real hard drive.

 

The program instructions are first loaded from a drive (real or virtual) into RAM, then the instructions in RAM are processed by the CPU without repeatedly re-reading these instructions from the drive on which the program was stored.

 

As I said before I have no knowledge at all in XXHighEnd.

 

Also, I'm not fan of RAMDisks, but I can't judge PeterSt recommendation because I never listened to his player.

 

But I'm fan (for music library only) of SSD on Thunderbolt & SD Cards, in this order.

 

Roch

[Jud]

Placebo is a widely-accepted phenomenon and I'm only suggesting it is one possible reason. Does Peter actually present any reason counter to placebo?

 

Hi goldsdad.

 

You know from PeterSt's comments here that his English is not always crystal clear, and I suspect he also wants to steer clear of any knowledge he feels might give a proprietary advantage. But as near as I can translate from what he's said on the matter, it has to do with speed and power draw. Timing is critical in Peter's ideas regarding how a player and computer setup can affect sound. In his player, the size of the chunks of a music file that are loaded into memory (and thus how often this occurs) is a tunable setting. John Swenson at the following link discusses how this can affect sound: Audio Asylum Thread Printer

 

CA forum member acg (Anthony), who uses XXHE and owns one of PeterSt's Phasure DACs, has commented regarding measurements he's made to correlate the RAMdisk setup with sound quality: http://www.computeraudiophile.com/f8-general-forum/my-deep-dive-media-storage-interfaces-musical-differences-heard-between-chipsets-firewire-400-800-usb-sata-flash-drives-sd-cards-and-network-shares-warning-may-cause-seizures-dbt-crowd-and-flat-earth-naysayers-18108/index2.html#post269177

[Miska]

RT-patched Linux - RT kernel only, or userland changes as well?

 

No need for user space changes. You can use eglibc to cut footprint, but it doesn't have practical effect on this particular area of performance.

 

Rest is of course about application's implementation... To get certain level of performance, it needs to be taken into account end-to-end.

[goldsdad]

Hi goldsdad.

 

You know from PeterSt's comments here that his English is not always crystal clear, and I suspect he also wants to steer clear of any knowledge he feels might give a proprietary advantage. But as near as I can translate from what he's said on the matter, it has to do with speed and power draw. Timing is critical in Peter's ideas regarding how a player and computer setup can affect sound. In his player, the size of the chunks of a music file that are loaded into memory (and thus how often this occurs) is a tunable setting. John Swenson at the following link discusses how this can affect sound: Audio Asylum Thread Printer

 

Hi Jud,

 

As far as I see, none of that is related to characteristics of the storage device, e.g. hard drive or RAMDisk, on which the program has been stored prior to loading of the program into RAM.

 

Copying the program from hard drive to RAMDisk then loading the program from RAMDisk into RAM will give exactly the same program in RAM as if it were simply loaded from hard drive into RAM. A given tuning of the running program's reading or processing of music files isn't going to produce results that differ in some way because the program was loaded from one particular storage device and not another device.

 

CA forum member acg (Anthony), who uses XXHE and owns one of PeterSt's Phasure DACs, has commented regarding measurements he's made to correlate the RAMdisk setup with sound quality: http://www.computeraudiophile.com/f8-general-forum/my-deep-dive-media-storage-interfaces-musical-differences-heard-between-chipsets-firewire-400-800-usb-sata-flash-drives-sd-cards-and-network-shares-warning-may-cause-seizures-dbt-crowd-and-flat-earth-naysayers-18108/index2.html#post269177

 

Anthony was knowingly making changes to the system then listening, so placebo is one possible reason for hearing a difference after loading the program from a RAMDisk.

[Jud]

Hi Jud,

 

As far as I see, none of that is related to characteristics of the storage device, e.g. hard drive or RAMDisk, on which the program has been stored prior to loading of the program into RAM.

 

Copying the program from hard drive to RAMDisk then loading the program from RAMDisk into RAM will give exactly the same program in RAM as if it were simply loaded from hard drive into RAM. A given tuning of the running program's reading or processing of music files isn't going to produce results that differ in some way because the program was loaded from one particular storage device and not another device.

 

Yep, I was speaking mainly with regard to music files in RAM (thus the reference to John Swenson's correspondence on Audio Asylum about the way a player program loads music files into memory). As far as putting the app itself in a RAMdisk, the situation there remains status quo:

 

- I don't know any reason it should sound different but believe I do hear it make a consistent sound quality difference

 

- PeterSt recommends putting both app and music in RAMdisks, and all I can deduce about his reasoning is what I mentioned above.

 

Anthony was knowingly making changes to the system then listening, so placebo is one possible reason for hearing a difference after loading the program from a RAMDisk.

 

Yep, certainly always a possibility. I would not contend we're doing anything like lab quality testing of hypotheses. On the other hand, as a late friend of mine used to say, "Don't cost nuthin'," so what it amounts to is some harmless playing around. Meanwhile we're enjoying the music, and if it would turn out there's actually something to this, then I guess we're enjoying it for a reason.

[goldsdad]

Yep, I was speaking mainly with regard to music files in RAM […]

 

A completely different matter than the storage of an application on a RAMDisk, which is what my message to Roch was about.

 

Regarding Anthony's comments to which you linked. I misunderstood them and so my response there can be ignored. On re-reading, I see he was talking about benefiting from the XXHE Playback disk being a RAMDisk. It's entirely reasonable that characteristics of the Playback disk, to which files are written and read during operation of the program, could genuinely impact sound quality. That is a separate matter from the question of whether there is a physical difference in sound waves as a result of a program itself having been loaded into RAM from RAMDisk instead of from hard drive.

[Jud]

A completely different matter than the storage of an application on a RAMDisk, which is what my message to Roch was about.

 

Regarding Anthony's comments to which you linked. I misunderstood them and so my response there can be ignored. On re-reading, I see he was talking about benefiting from the XXHE Playback disk being a RAMDisk. It's entirely reasonable that characteristics of the Playback disk, to which files are written and read during operation of the program, could genuinely impact sound quality. That is a separate matter from the question of whether there is a physical difference in sound waves as a result of a program itself having been loaded into RAM from RAMDisk instead of from hard drive.

 

Why should the response be any different? It is quite possible for the cause to be placebo whether or not there is another explanation we're aware of. And of course it is also quite possible there is an explanation we're not aware of.

[PhilipAC]

He could refer to embedded Jitter. The Porzilli Memory Player purports the same hypothesis.

 

Can you have "embedded jitter"? I understood that jitter was timing errors introduced during the movement of a file from place A to place B, but that once the movement was finished the file (now "stationary" in place B) was identical to how it was in place A.

 

Have I got this wrong?

 

Philip