USB audio cracked... finally!

[Jud]

1 hour ago, mansr said:

Could you give an example of two settings in your software giving distinctly different sound?

 

Not @PeterSt, and speaking purely subjectively, I thought the sound changed with adjustments to the size of the portion of the converted file loaded into memory before being sent on to the DAC.

 

Edit: You asked for two.  Another would be the choice of kernel streaming or WASAPI (I thought, again subjectively, that the former sounded considerably better).

[Jud]

3 hours ago, mansr said:

It's plausible in the same sense as the existence of a teapot orbiting Pluto.

 

Thank you, Sir Bertrand. 

[barrows]

OK guys, I am unsubscribing.  I see nothing of substance here about the LUSH cable anymore, just the usual bickering.  When I get a LUSH cable here I may choose to re-engage here with some listening impressions.

I accept already that different USB cables give rise to different sonics, although in my current system the implementation of the USB transmitter and receiver sections (and isolation of USB on the DAC side) is good enough to reduce those difference to very small levels.  I am looking forward to trying the LUSH and seeing what it brings, I am also interested in hearing of others' listening experiences, vs. other USB cables...

[PeterSt]

1 hour ago, mansr said:

Could you give an example of two settings in your software giving distinctly different sound?

 

OK, if you think this is going to work for you, look here :

 

 

This is from someone without NOS1 DAC who follows the advice to change the SFS of 120.00 to 4.00 (you can not see that, but you will believe me). N.b.: It is useless to explain what "SFS" is because it would require the explanation of everything and all (and way more than shown here, in consistency) but anyway the SFS is the chunk size of read (music) file data in "some" unit.

 

Of course the SFS is related to how the memory is organized to begin with, how the disk is a HDD or an SSD, whether all is remapped to memory, whether the OS has been booted from memory and no disk is to be seen anywhere, or whether maybe the file is on the LAN somewhere (like NAS etc.) and not proxy'd (cached in full locally). Then of course a (for audio) "Minimized" OS behaves totally different than when this has not been done and logically (ahum) it requires an XXHighEnd license to Minimize the OS in the first place.

 

There you have it. One parameter (and the others mentioned are related to this specific setting or otherwise the effect is different).

 

PS: I forgot to mention the dimension of 4 different OSes on a "RAM-OS Disk" which can each be booted in 4 very different(ly sounding) means, which makes the permutations of that alone something like 16. Still we talk about one parameter only (the SFS).

Oh, the minimizing itself can also be set (to what degree and what) with the notice that especially with Windows 10 builds it can come to you that playing with GUI sounds better than Unattended (no GUI anywhere) ... which is not meant to be but sh*t happens.

[PeterSt]

8 minutes ago, barrows said:

When I get a LUSH cable here I may choose to re-engage here with some listening impressions.

 

@barrows See you day after tomorrow !

 

Regards,

Peter

 

[austinpop]

4 minutes ago, barrows said:

OK guys, I am unsubscribing.  I see nothing of substance here about the LUSH cable anymore, just the usual bickering.  When I get a LUSH cable here I may choose to re-engage here with some listening impressions.

I accept already that different USB cables give rise to different sonics, although in my current system the implementation of the USB transmitter and receiver sections (and isolation of USB on the DAC side) is good enough to reduce those difference to very small levels.  I am looking forward to trying the LUSH and seeing what it brings, I am also interested in hearing of others' listening experiences, vs. other USB cables...

 

Don't blame you! I myself am hanging in there via liberal use of the "Ignore User" feature. The difficult ones to ignore are those who usually have sensible and useful things to say, but then get caught up in the noise. Ah well, nothing's perfect.

 

The listening impressions do trickle in - I posted mine a few days ago. And Mani does an excellent service by posting a summary every few days.

[mansr]

21 minutes ago, Jud said:

Not @PeterSt, and speaking purely subjectively, I thought the sound changed with adjustments to the size of the portion of the converted file loaded into memory before being sent on to the DAC.

 

Edit: You asked for two.  Another would be the choice of kernel streaming or WASAPI (I thought, again subjectively, that the former sounded considerably better).

Perhaps I wasn't clear. I want two sets of values to punch into various controls that will likely result in a noticeable difference. Something along the lines of "setting A at X vs Y" or "setting A at X and B at Y vs setting A at Z and B and W." Exploring every combination would take far too long time.

[Jud]

32 minutes ago, mansr said:

Perhaps I wasn't clear. I want two sets of values to punch into various controls that will likely result in a noticeable difference. Something along the lines of "setting A at X vs Y" or "setting A at X and B at Y vs setting A at Z and B and W." Exploring every combination would take far too long time.

 

Do you have XXHE up and running?  The WASAPI vs. kernel streaming setting should be obvious.  The other setting I mentioned that had to do with amount of a file loaded into memory was SFS.  Look at the Phasure forum to get Peter’s SFS setting from his sig and try that.  Then choose some arbitrary other number along the wide range of possible SFS settings that is significantly different than the first.

[manisandher]

6 minutes ago, Jud said:

The other setting I mentioned that had to do with amount of a file loaded into memory was SFS. 

 

I would try SFS=1 vs. SFS=120. Best to have memory set to 'straight contiguous' first (see http://www.phasure.com/index.php?topic=2127.0 if necessary).

 

There is a strong consensus among XXHighEnd users on how SFS affects the sound. Hopefully, you'll hear the same... (I suspect it'd be easier if you had XXHighEnd fully optimized, which would require a licence.)

 

Mani.

[mansr]

Just now, manisandher said:

I would try SFS=1 vs. SFS=120. Best to have memory set to 'straight contiguous' first (see http://www.phasure.com/index.php?topic=2127.0 if necessary).

Thanks, that's the kind of suggestion I was looking for. All other settings at defaults?

[PeterSt]

Wouldn't it be possible to go elsewhere with this ?

 

15 minutes ago, manisandher said:

I would try SFS=1 vs. SFS=120

 

Mani, please cut this crap. My example is a good one and this won't even work (system will be too slow and no Straight Contiguous stuff anywhere and surely also no license to solve it.

 

Take this to somewhere I won't see it. OK ?

 

[manisandher]

5 minutes ago, mansr said:

Thanks, that's the kind of suggestion I was looking for. All other settings at defaults?

 

A few screen shots...

 

Press the S button to go into Settings. Scroll down to 'Output'. Make sure the 'Device Buffer Size' reflects the correct buffer size in your DAC:

 

Scroll further down to 'Memory and Disk Utilization'. Probably best to set 'Mx' to 120, and make sure 'Max the same as SFS' is unticked. Set 'Memory Organization' to 'Straight Contiguous' for best sound.

 

Finally, go out of Settings by pressing P or L, and go down to Q. Just use these settings to start with:

 

Press the X button to close.

 

Good luck!

 

Mani.

[manisandher]

2 minutes ago, PeterSt said:

Wouldn't it be possible to go elsewhere with this ?

 

 

Mani, please cut this crap. My example is a good one and this won't even work (system will be too slow and no Straight Contiguous stuff anywhere and surely also no license to solve it.

 

Take this to somewhere I won't see it. OK ?

 

 

Huh?

[PeterSt]

4 minutes ago, manisandher said:

Huh?

 

Nothing "huh", Think.

 

Edit : or just read what I just said. Your settings can not work. That is, for Mansr.

 

And it is offtopic as h*ll, but alas your thread.

[manisandher]

 

1 hour ago, mansr said:

Something along the lines of "setting A at X vs Y"...

 

25 minutes ago, manisandher said:

I would try SFS=1 vs. SFS=120.

 

Peter, this thread is all yours. See ye...

 

Mani.

[manisandher]

4 minutes ago, PeterSt said:

And it is offtopic as h*ll, but alas your thread.

 

FFS, you started it!

 

11 hours ago, PeterSt said:

 

I do something that lets the CPU behave in some (erratic) way, and the sound changes. This turns out to change the sound in similar fashion for everyone, regardless the DAC. *That* it would work out the same for everyone, I could not guess in advance. But today we (all the software users) empirically found that it works out like that and someone with a newly found nice settings combination can post it, and everybody hears the same kind of thing from it.

 

Also notice that there has been a time that (for testing the users themselves) I told about a setting, told with it that I did not like it (but tempting them to try it in sneaky fashion) and next they did not agree with me and all (or many as far as I could see) started to use it.

Some times I need such tests myself or otherwise it could be me hearing things, or indeed else pushing people to hear the same (that is easy to do and NOT the way to go for).

 

My customers will also know that in about each major sound change situation and in 100% of cases, I ask my wife what she perceives for differences. It can happen that she says "none". In that case I ask as long as it takes until there is one answer (can take more than an hour). And that answer eventually always comes and is ALWAYS right.

 

Mani.

[PeterSt]

Mani, please ... READ. YOUR SETTINGS CAN-NOT WORK.

 

You aren't deaf, are you ?

 

You imply a million more questions (because what you pose can't work) and you do this in an unrelated thread.

 

Let's turn it around ... anything wrong with what I suggested ?

Yes, according to mansr there is. READ.

 

So cut this crap.

[mansr]

21 minutes ago, PeterSt said:

Wouldn't it be possible to go elsewhere with this ?

 

Mani, please cut this crap. My example is a good one and this won't even work (system will be too slow and no Straight Contiguous stuff anywhere and surely also no license to solve it.

Why don't you suggest something that will work?

[PeterSt]

3 minutes ago, mansr said:

Why don't you suggest something that will work?

 

I F*ing did so. Didn't I ?

And exactly the two opposite values you asked for. But I suppose I am in your ignore list.

Or at least I should be.

 

Or there is another misunderstanding, but I certainly don't get it.

[manisandher]

3 hours ago, mansr said:

Could you give an example of two settings in your software giving distinctly different sound?

 

1 hour ago, PeterSt said:

 

OK, if you think this is going to work for you, look here :

 

 

 

1 hour ago, mansr said:

Perhaps I wasn't clear. I want two sets of values to punch into various controls that will likely result in a noticeable difference. Something along the lines of "setting A at X vs Y" or "setting A at X and B at Y vs setting A at Z and B and W." Exploring every combination would take far too long time.

 

And all I'm suggesting is that he change the SFS to determine if he can hear a difference. I really don't understand your issue with this.

 

(My mentioning the Straight Contiguous memory and the licence was simply to say that there's a lot more optimization possible in XXHighEnd.)

 

Mani.

[Jud]

1 minute ago, mansr said:

Why don't you suggest something that will work?

 

It would involve significant preparation of a Windows machine (which I imagine you could do with some reading on the Phasure forum), plus a license (which you would have to sort out with Peter).  Otherwise, at least so far as my subjective impressions suggest, Peter would be correct - the differences would be minimized, so as he says, this wouldn't "work" for you.  But frankly, I cannot see any regime under which you would feel these settings changes *would* work for you to hear differences, as the time to make the changes would be far too long.

[rando]

[manisandher]

Yep!

 

Mani.

[austinpop]

What just happened?

[manisandher]

Dunno. But at least any conspiracy theories of collusion between Peter and me should be out of the window now 

 

Mani.