XXHighEnd

Ever since Marcin posted the reducing the split file size to the minimum produced very good sound for him, I've been experimenting with it as well. With vocals, Leonard Cohen, Dylan, Patty Griffin, Eva, Dianna,  etc., the "correct" setting changes for each singer depending on the volume level. Take a singer and turn the volume to a moderate level. Here, on my system, setting the split file size to 12 does sound very good, but, turn the volume up to very loud and everything sounds way too detailed and etched, like a too sharpened digital photo. Keeping the sound very loud, turn the split file size up to between 80 and 200 (depending on the vocalist) and it sounds very pleasing again, though it looses some detail it sounds very realistic.

Its interesting that reducing the size at loud volumes initially sounds "best" but it soon becomes fatiguing. In a way, the split file size acts like a master tone control: reducing it makes all ranges, low, mid and high,  sharper and more revealing, thus for moderate volume it sounds "best"; increasing it, at very high volumes, rounds all ranges off and makes them feel more pleasing. Give it a try........and let me know if I've lost my mind......

Can we let you know without trying ?

:grazy:

I have just thought of a Q6 knob for this ...
Could be more interesting than Q1 from Engine#3 ...

:yes::swoon::yes:

Good idea, but wouldn't it be less accurate from current form?

Having an easily accessible control, on the face of the gui, would be best, IMHO. I've started making a list of where each vocalist sounds best. Would be nice to be able to make this adjustment quickly.

Also, there's a lot more to unearth with this. I think the effect is different for high def files, they seem to like the lower settings at higher volume levels, though I haven't tested enough to be sure.

As Paul Simon so eloquently sang, Still Crazy After All These Years, yeah!

This post has nothing to do with volume, only split file size.

Firstly, Peter please NOT another Q knob for this - I'll go insane...

But the split file size clearly DOES change the sound. Lower values sound 'clearer', but perhaps too 'edgy'. Higher values sound 'fuller' but perhaps too 'woolly'. It therefore seems logical that there should be a good middle ground and that's what I've been trying to establish.

Again, I used my 24/176.4 recording from vinyl and compared the sound directly to the vinyl.

But everything is inextricably linked - all the other parameters in XX have an affect, so finding an absolute value for the split file size was going to be difficult. Without going into details, I found that my ADC/PC/DAC chain is most transparent when I use a device buffer size of 1024. (This has nothing to do with XX though.) With this set, I played around with the split file size.

And you know what, I got the best sound when the split file size was... THE SAME SIZE AS THE FILE BEING PLAYED!!! (For my 24/176.4 file, it was 330MB.)

Now, how this translates to multiple files and playlists with different formats, I'm not sure. But assuming there is something in my findings, wouldn't it be great if XX could set the split file size automatically, track by track, based on the size of each file being played.

This way we'd get the best sound... and remain relatively sane too... although it has to be said that all long-time users of XX are waaaay beyond this point already.

Mani.

Hmm ... Interesting ...

Although this has nothing to do with the eventual memory used (as you will know), it does make sense to me. Not how it works (for the effect of SQ) but indeed what causes it.
So, if true (and why not, for now) it would be a next thing which *is* under my control, without any notice why it would matter. On the other hand, I can imagine (at this moment) that when I dive into it, I may be able to reason out what's happening internally (the OS).

Warning :

It will *not* be so that anyone now can look at his file size and on a per track base set the Split File size;
With Mani it makes sense because his native file (size) is the same as what's been played (even that is not true 24->32 bits, correct Mani ?), so for that certain part Mani can see what to set. However, if this is about 16/44.1 becoming 32/176.4 there won't be a rule for you. Not even for me at this moment, and it is that what is to be sorted out by me.

But *if* others could find more "rules" in this area, I would be very happy to learn them; As  usual, I on my own won't be able to find all what you all togther can. So please do (explicitly for the good sake), and let's again make something better out of it.
It's been too long already. :)

Thanks, and thank you Mani (and please let me know your "DAC Needs" setting).
Peter

Peter, you're right of course. My DAC is set to 32 bits.

I started at 1000MB and then jumped to 12MB to hear the difference, which was as I described. I then continued to increase the split file size until the sound between the recording and vinyl 'equalized'. 330MB seemed to do the trick, but I'll try 495MB... I'll also try this with my Weiss interface, which does work at 24 bits.

Mani.

Nice discovery Mani! I confirm that it works and sounds awesome :) Now it is clear to me, why 'the best settings' that sounded great with one album, don't seem to work with the others...
It would be great to have auto split size for every track, but as Peter said, it won't be easy for upsampled files.

Marcin

PS
Mani, would you agree that buffer size at 4096 has more depth and air than 1024, but the latter is more saturated in the midrange? I would also like to know how you describe the difference between Scheme-3 and No-Appointment. It seems that we are the only guys here who prefer 'no-scheme' sound. For me, no-appointmen gives the most neutral, transparent and raw sound, which I personally prefer. Scheme-3, on the other hand, is smooth and it's very easy to distinguish and localise plans.

OK, there's not a lot in it between a split file size of 330MB and 495MB for this 330MB 24/176.4 file. But I prefer the 495MB - it really does seem to hit the 'sweet spot'. Pure coincidence? Maybe...

Marcin, my signature is wrong actually. I used to use 'no scheme' up until a few months ago when I switched from 4 to 2 cores in my BIOS. With 2 cores, I use 'scheme 3'. But I have to admit that I've always found it difficult to hear differences in SQ between the various schemes. If you're still using a single core, then I suspect 'no scheme' will be best for you, no?

I've updated my signature...

Mani.

No, I got back to default settings, because I need decent performance for apps like Photoshop. Either way, I hear the difference between schemes right away.

Hey Peter, forget these pleas for help... just threated to introduce yet another Q knob instead.

Mani.

Of course, it could just be that the 'sweet spot' for 16/44.1 material is always around 60MB irrespective of file size (with DAC set to 16 bits and assuming no extra processing in XX). In which case, one would expect the 'sweet spot' for native 24/176.4 files to be around 480MB (with DAC set to 32 bits and assuming no extra processing in XX).

But I kind of hope that the optimum split file size is actually related to the track size, and that this can in some way be automated.

Mani.

I agree, but in current state, there will be a lot of 'out of memory exceptions', at least I get those if I set the split size above 400 MB (16/44 with QAP).

Marcin, I think this is related to Peter's 'warning' above. QAP will be introducing additional processing that will need to be factored in.

In any event, try what you did with a native 24/176.4 file and see what happens.

Mani.

All the testing I did with 16/44 files was also with 2x arc prediction. Here,playlists with various vocalists generally sounded best with the split file size set at 90mb. However, a particular artists voice on a particular album can be further "dialed in" based on the qualities of that artists voice. Lastly, at lower volumes, reducing the split file size generally sounds much better, particularly the bass. Is this not true for you Mani?

I agree that a new "Q knob" is not the answer because I don't think, given the location of Q knobs, all the different file formats, and  all the XX choices for playing those files, the split file size setting is something you set once and then forget about. Further, given these same considerations, I don't think it is likely (or even desirable given personal preferences) one can have XX automatically set the size per track. For convenience, I think a button for setting the split file size should be on the face of the gui down near the play button so one can quickly adjust it, much like a tone control on an integrated amp.

Just my 2 cents. 

Ofcourse a new knob is not desired not even 1 knob is disered, BUT, whe can learn from those settings and when we do,
I know for sure Peter will find a way to make "schemes" or "settings" depending on the genre or album.
We can not expect from him to test all options, there literary thousends!

I really DONT like knobs at all, not even check-boxes.
Peter needs to update the functionalty of the GUI together with the implementation of .ini files per album
when this is done you can put those commands under a remote button for eg, (or post them )
Also you can stay in your listening position and switch between settings or schemes you want to test.

I also dont like to "switch" everytime between adaptive and special mode when going from jazz to ambient for eg.

Roy

PS: hope this makes 4 cents
PPS: Dont forget Girder can learn all those functions and checkboxes when the xx form is fully "functional", it can do a string of command too.

Without meaning to stir the pot, I hope people realise how important this topic is. It's started to make me think that any comparisons that I've made in the past between, say, 16/44.1 and 24/176.4 files are invalid because I've not optimised the split file size for the two resolutions. For example, at 100MB, most 16/44.1 tracks sound OK. But 24/176.4 tracks sound too bright and forward. At 400MB, this brightness goes... but 16/44.1 tracks now sound too dark.

We've GOT to get to the bottom of what's going on here...

EDIT

Oh, and some people who use a large split file size may not agree that it sounds 'dark'. But I bet these people are of the 'ultra-low latency' brigade, which tends to sound brighter and more forward.

EDIT EDIT

Actually, this wasn't what I meant to say AT ALL.

It seems to me that the split file size has a simlar affect to the sound as the device buffer size - the larger the darker, the smaller the brighter. Now, for ultra-low latency you'll obviously have a low device buffer size, BUT... you're probably going to have to set the split file size quite low also to get glitch-free palyback. So, can you now be sure that it's the low latency that's causing the change in sound, and not the low split file size?

I think that's what I meant to say!

Mani.

Yes we have to, otherwise I'll end up in a psychiatrist institution, haha  :grazy:

But seriously, it does have a great impact on SQ. It would be great if Peter implemented 'auto split size' based on a track's file size (assuming that's the optimum).

Ah, and there's the rub, as a certain gentleman, who was born 10 Km down the road from me in 1564, would say.

But as Peter has said, he won't be able to determine this on his own. (I'm not sure if I trust his ears at his age anyway - he's started hearing 'rain'!)

Mani.

Yes,

to be - or not to be (confused)
that is the question

Per ;)

I'm afraid it may not be able to work at all, because the "memory stuff" is not something you can control dynamically. So, I can make those Split File sizes dynamic allright, but not the memory for it. If you try this for one track it will work, for that track. Do it a next time, and it will work again. So in the very far end it will need Engine3.exe to restart each time (doable) but it won't work at all for gapless etc.

But this all is hooked to your findings which may be too much of a derival from what's really (physically) happening.
I'll see.

Peter

Mani, you have all the necessary equipment (analog setup for LP) as well as A/D converter, so you'll be able to find 'the best' values - the most transparent compared to vinyl. I don't have any reliable point of reference, only my ears and assumption what sounds right or not. I think we all should contribute and try different values, especially for 16/44 with QAP.

I think it depends what you mean by 'sounds better'. The bass seems to tighten up with lower split file sizes., but can lose some of its weight and fullness. For example, a grand piano can start sounding like a smaller stand up. Generally, the sound can get too thinned out.

But I haven't really listened out for differences between various volumes - if the volume becomes yet another variable, I really will lose the plot and might just descend to using WMP.

Mani.

Unfortunately, I don't think this is possible. You see, if I swap my RME interface with my Weiss interface, everything changes. So yes, I think I can find the best settings for my setup, but I really doubt they'll be the best for anyone else. Hey, but who knows? I'll certainly do this and let you know what I come up with...

But for now, I'm finding that if I play 16/44.1 with a split file size of 65 (my favourite 'fixed' setting), then I need to play QAP with 70 or so to take a slight edge off and bring the bottom end back. Not a big difference in split file size at all - certainly no where near the increase that native 24/176.4 seems to require to sound right.

From the Schumann thread:


Interesting Jack.

My understanding is that the higher the split file size, the more 'stress' is placed on the PC. So, could it be that the changes in SQ that I'm (we're) experiencing are all due to different stresses placed on the PC with different split file sizes? Maybe.

An easy test would be for me to set the split file size and then increase the CPU clock rate and/or CPU voltage. I'll do this and report back...

EDIT But what's particularly interesting about different split file sizes is that you can hear the change in the sound immediately, within the first second or two of the music playing. I mean, you don't have to wait until the data is transferred internally in the PC, when a smaller split file size should place less stress on the system. And the sound seems to be quite consistent - it doesn't seem to change between the points where data is being transferred internally.

Does this make sense?

Mani.

I've just tried this, and am quite surprised by the results - not what I expected at all...

Just to give some background, I can vary the base CPU rate of 133MHz by between 9x-21x using the mobo's BIOS utility. I normally have it set to 9x. But I've just tried 15x and 21x also, keeping the split file size fixed at 12MB. I used a single track for evaluation throughout; David Sylvian's 'When Poets Dream of Angels', which is beautifully recorded and starts with three acoustic guitars at left, centre and right.

9x 133MHz = 1.2GHz
The sound is as I described it before - very detailed, but edgy and fatiguing after a while.

15x 133MHz = 2.0GHz
The sound is fuller. It's almost like the split file size has been increased.

21x 133MHz = 2.8GHz
The sound is fuller still. Rich and warm. Again, it's like the split file size has been increased further.

So, increasing the clock rate has a similar effect to increasing the the split file size!!! This is not what I expected at all. Certainly, with the CPU rate set to 21x, I have no problems listening with a split file size of 12MB - it sounds so much more refined than with the CPU rate set to 9x.

The only explanation I can offer is this: With a smaller split file size, data is transferred more often (though in smaller chunks). Maybe more CPU power allows things to settle down more quickly, well before the next data chunk is sent.

Of course, God knows what happens to other things, such as RFI, with increasing CPU rates...

Peter?

Mani.

I think again, its all about speed,

Did anyone proved that underclocking is better vs more RFI ?!
The faster a pc can handle things, the more time there is to handle the music.

Speed, speed, speed.

Roy

Just a couple more cents......adjusting the split file size has, again, revolutionized my experience of the sound quality that XXHighend is capable of producing. After adjusting the split file size(hereafter called the SFS) to 165mb, I played Patty Griffin's, "Icicles" -from her Impossible Dream CD-for my wife. Her comment was either Patty was somewhere in the entertainment console or she, my wife, was having an LSD experience, something that hasn't happened in 30 years.  :)

The increase in SQ is really significant, though after spending a few more weeks with it I'm sure area's in need of improvement will emerge too.

I have started to keep track of where the SFS for particular vocalists on  particular cds sounds best,but I only find it necessary to adjust the SFS for that album, not for each track on it. For playlists with multiple vocalists I find a compromise size that works best for all the tracks in the playlist. Though a bit of a compromise it's waaaaaaay better than anything I was listening to ten days ago!  

Has anyone else found that significantly lowering the SFS at moderate and lower volumes helps SQ?

It seems that with Mani's latest discovery increasing the cpu clockrate allows for the converse regarding the SFS. It allows one to lower the SPF and get the same result. However one gets there it is a welcome result and it will be very interesting to see how Peter sorts all this "good stuff" out.

Did I say anything about a tone control recently........ :)

I think changing the default SFS from 100mb to 60mb would be a good thing,
especially for gapless playback.

What if high cpu speed IS important together with a low SFS.
XX is designed to run on 1 core, thats good if you have a dual core, one for #3Engine, one for windows.
but if you have a quad core can xx run on 2 cores instead of one.

Peter, is it possible to let xx run on 2 cores ?!

I think Boleary AKA SFS moniker is right ... sort of. It is volume related ... kind of. :thankyou:

I tried to mimic something of what you all claim here, and at some stage I had something working. I continued the Dread Zeppelin album where I quit the day before, and immedietaley noticed

a. an emphazation of distortion (guitars hehe);
b. some kind of crazy dynamics for the more rough parts. Like hitting a snare (drum) without the snare activated.

Next I ran that Album from Prince of which I never know the title (I call it "P" because it's in cold (a P) on the sleeve), and of which the first track is called "My name is Prince". It is no coincidence that I wanted to try this album, because it has been one of those 5 CDs I never could play without distortion for 10 years long. This was history already for a year or so, but still.
This first track never sounded so mean to my ears as ever before, but for the first time I could hear why that is;

First there's a bass synth which forms out of very fast off/on and which is very difficult to notice because if the (bass part of) a system isn't fast enough, it isn't off/on at all, but just a mere continues tone (things can't follow). Btw, the 2nd track on Musicology contains such a bass which may excurse your woofer to tearing dimensions when your system can't follow (and which may be a matter of the proper Q1 setting !).

Now, with this "My name is Prince", this time, I also noticed that they applied the same on/off thing to his voice. Actually the effect was very similar to the distortion guitars and I appreciated it as distortion. Very strange, because only his voice exhibited it. Also, all of the next tracks didn't have it, and didn't have any of the emphazation effect at all.

To me it looks like that a very "square" wave now can be followed better. For Dread Zeppelin I didn't like it, and the whole lot received a flare hence character. Worse, I heard that character back in the first track of P. But then listening carefully, I coulld only appreciate it, because it just shows something which seems not to be there otherwise. But still. And it didn't bother me at all anymore after the first track.
3 tracks before the end I got an an out of memory, changed the program for that, ran a nect album with smaller tracks, but kept on having the out of memory. So I reinstalled the original program to be able to play along.

If this is emphasizing the more square stuff, the first to notice will be sibilance in women's voices. You may like this better because it will show better the unique character of the voice. I mean, if it's in there it just is, and then better hear it. But if you play this louder, things may go very wrong, because this is so square that things won't be able to follow, and you'll have distortion instead. So this *is* volume related ...

I'm not sure what to do with it;
So far I couldn't find any means to make it decently working in the program, and with my 3GB of memory there's hardly any headroom to play with. So, this needs more memory, but it just isn't there. I must add to this that all is right in the middle of my mangled "Dot-net" memory management, which is fragile to begin with, because not official(ly used). So if I tweak only a little, I'll get an out of memory. And like I just implied, with the small tracks ! Two years back I have been working on this for two months or so, and I better leave it alone ...

But it sure looks there is this other dimension. Now how to normally utilize it.

Peter, what SFS were you using during your listening?

What's of particular interest to me is whether changing the SFS causes a consistent change in SQ. For me, starting at 12MB and increasing the SFS makes the sound fuller, but perhaps too woolly if increased too much. Do other people experience the same?

What I am now convinced of though is that we CANNOT come up with a universal 'best SFS' number for everyone. When I had my CPU rate set to 9x, I preferred a SFS of ~60-70 (for 16/44.1 material). But set at 21x, I'm perfectly happy with a SFS of 12MB.

Mani.

My SFS has been 70MB for ages. CPU is at 9x btw. :)

During my testing yesterday we can't speak of "split File size" as such, because I mangled the program.

Mani
Sorry about that new 'can of worms' I just gave you!
However, did you up the voltage at each change, or just once, or not at all? Just to complete the picture for me.
Thanks
Jack

and the more distortions goes to an audio interface  :bored:

... but it is more likely it goes the other way around, though the distortions will be of a differend kind then (think like skips).

As far as I'm concerned, it must be concidered a myth that undervolting etc. will really help in a "controllable" fashion. *If* something is interfering with sound at all, it should be removed and not attenuated. Thus by more physical means like better grounding.

I think I have seen how such things can change the jitter level impeeded by something we call noise, but in the end is always some (recognizeable) signal coming through which shouldn't. For example, I can let the "clock" run faster because of it. So, if the clock signal coming from the clock (steadily running by itself) gets polluted with another nice signal, once in the x cycles of the both signals another nice "top" of the wave comes up, which is seen as the proper rising (or falling) edge within the DAC. By this means 1000Hz can become 1030Hz easily !

This was an example of the most sound influencing kind, and it is totally unrelated to something like CPU speed "hence" RFI or anything. It is wrong design and out of your control.

The (for me) strange thing is, that while I never saw prooven that RFI does something to a signal we are talking about here, the "grounding wire" itself *does*. So, when there is this unwanted signal (could be the clock visible on the RCA output) and all is officially grounded already, connecting a next ground wire to some point (you expect to be good), moving it around changes the amplitude of that unwanted signal all over. Move your hand around that wire also shows big changes. Connecting it to one side of the same cabinet may remove the signal, while connecting it to the other will amplify it.

Maybe a more understandable example :
Connect your interlink to your preamp (or DAC) and at the other end connect the scope (all signals off). The cable has two wires, one for ground and one for signal. Both are connected to the plugs. Now connect another single wire to the outside of the plug (which is connected to the ground) and also connect it to the plug at the scope's side. The noise will be totally different from this, and while you had some "hum" before, now it may be totally gone.

So, no matter this is beyond our comprehension most of the time, it is these things which matter. All that other hocus pocus ? I can't see it on a scope ...

All together, if you find that undervolting etc. really makes a difference, better investigate whether you did all to avoid the influences. This is where I started with a network player to be the "soundcard" and some later the FireFace (USB should do the same, but it never cut it for me).

That software can still influence SQ is quite another matter, for the largest parts a secret for myself too. But it influences in the DAC, not in the PC.

Peter

True, but my sound card works as a DAC. For digital transport the influence is smaller and proper ground should do it.

For digital transport the influence is larger ...
:yes:

Peter, so what you're implying is that it should be easier to get a decent sound from a sound card (analog outs) than from a transport+DAC combo assuming that both sound card and transport are fed from PC? This is sth new to me, I always thought that it's the other way...

Regarding our topic, after few sessions and multiple combinations I came to a conclusion that 'file size=split size' rule applies for 16/44 upsampled tracks as well. There is no need to increase it further.

Hi Marcin - I guess I must read back some posts to understand this exactly. So, what do you mean "increase it further" ?

On the soundcard ...


No, that is not what I wish to imply. Both will be subject to the same sh*t, assumed the s* is there, but the influence on digital data is more devistating.
Notice that both digital and alogue data are as analogue as can be, and thus both are subject to the same influences. But, for example, if a most significant bit becomes 1 while it's intended as 0, the magnitude is 32768 (for 16 bit data) "wrong" for the analogue result of it. And while this is about a "data" error, it can also be about clock errors. This is all when the data gets too bad to be processed digitally, and depending on where it goes wrong you may not even have a lock. So *this* won't happen so easily, or it must be very very bad. But :

[and here I just gave the finest description how jitter emerges, but I decided to scratch it away; I just realized that even the best "jitter engineers" here in Holland don't realize that it works like that and I started to think a bit commercially; my DAC should remain the best there is, and I just layed out a large deal of how it happens. Sorry !]

But ... Jitter emerges under the influence of noise, which changes the square wave implying the clock signal. So, the falling (or rising) edges which trigger the samples to (and out of) the DAC will change per "clock cycle" and and the timing of that trigger will vary because of it. This is jitter. So, the less noise, the better it is.

When that same noise would be in the analogue signal, it would be inaudible.

Peter

I meant that in my opinion there is no need to increase the split size over the track's file size, even when we play 16/44 track which is upsampled (QAP).

Now try the same with your Weiss interface :) I think that maybe your motherboard is 'prepared' for higher clocks, providing more stable voltages with default CPU speeds (which now are 2.5-3 GHz) rather than underclocked and undervolted. I'm just thinking out loud, I'm not an electronics engineer.

Allright. A week back or so, I set my Split File size (SFS) to 12. I don't know anymore whether I did that to just listen and judge it, or that I started to work on the ticks at track part boundaries, but in either case it combined with having another amp to test, different LS cables and more confusing stuff. Next I got a rather severe cold and my ears wouldn't do what they're supposed to.
In the mean time, I received complaint over complaint that I played too loud, which by itself could have been related to the higher gain I have in the DAC which also happened somewhere last week, and I lost my volume reference a bit (including mis behaving ears that is).

Even yesterday, at testing some not more than 20 seconds pieces, I was asked why this all must be that loud. Why ? otherwise I can't hear the ticks.

Last night, those "you play too loud" pestering was out, so I grabbed the opportunity to play without the nagging. Still being bugged by stucked ears, I measured a 90dB as minimum level (@4m), so that should be quite normal.
I played a familiar album, but could not much recognize it; tried to hear through the stucked ears, but couldn't imagine much other than "this is different".
I set down behind the PC to select another album, which was just too late for a new track (not) to start. It started with some deeper drums, and it not only sounded strange to me, but also felt strange (hands on the desk). And *then* I thought, wait a minute, this won't be that SFS of 12, will it ?
So I set it back to my normal 70, and indeed it appeared that the drum now was normal again. Switched back and forth a couple of times, ending up with "70 is better".
I selected Yello - Flag, and went on with what I was doing (which was cooking).

Really within seconds I received such a silky sound from not such an at all silky performer, that I thought my improvements from the last months must have been vast, knowing that I hadn't listened to this album for quite some time. I must add that I was totally ignorant here, and never thought of that SFS = 12, now changed to 70 setting. I was merely overwhelmed by this beautiful sliky sound, which reminded me of switching from SPDIF to i2s.
I listened and listened, and even more ignorantly I started to think what could be wrong that my wife says it sounds harsh and too loud; what can't I hear that she can, despite my little hearing problem ?

Believe it or not, but it really took 30 minutes before I thought of the SFS setting again. It took another 5 because I wouldn't let go of the "beauty", but then I set it back to 12 ...

Oh man ...
This is much much worse than you can imagine ... I think. But the quote from Mani above is there for a reason of course. And yes, I recall his referring to this topic in between the lines a few days back ....
At 12, at least here, the sound becomes totally harsh, things come forward which shouldn't, and there is NO normal bass anynore. Haha, NOW I knew what my wife was bothering this last week. But was I right ?
Put SFS back to 70, and right in the last track of Flag she came home. Uhohh ... not the best example track ...
But nothing happened.

Moved over to a next album, and I noticed she was rather shouting towards me in order to be able to make her words rise above the music level. Uh-ohh.
But still nothing happened.
Then already within the next album, I dared to ask : "So you still hear harshness or anything ?" and the answer was "no, nothing so far".

Allright, I always present you with these kind of "ABX" stories, because they work the best for me, but also hopefully show the realism / facts of matters.

In the mean time, I now know this is so severe that I will go to the bottom of it, because indeed it is so important. It is more important than any Q setting, and it even plainly destroys a well behaving DAC.
At this moment I have no clue as to what can be happening that this influences so much, but regarding the character of it (frequencies come forward which shouldn't) I can even imagine it can me measured.
So, nothing else to do than to "solve" this. If I can, beware. Ok, most probably we have listened to the optimum already, assuming it is some border which is crossed somewhere. But if it goes further and really would be related to the track size or anything (how ???) a real vast improvement must be possible again.
You know I already tried some things, and alhough audible, it wouldn't work stable. Also, this was without clue what to actually achieve. So I guess that's first now.

Help ...
:)
Peter


PS: I have read about the suggestion of the 2nd level cache being 12MB, but I haven't and most people haven't. Still it may be in that area somewhere.

Did we know that we can control the L2 Cache Size in Windows

HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Session Manager\Memory Management

SecondLevelDataCache = 0 on standard

256KB: 100 hex
512KB: 200 hex
1024KB: 400 hex
2048KB: 800 hex
etc

You can speed up the PC giving it a value, well I think, didnt try it yet.

Roy

PS: probably it will be set automaticly already by the HAL.

It's great when 'our' developer finally starts listening to what we're saying :)

I think this is important:


But let me be clear - when I say "perfectly happy", I'm really only talking about 16/44.1 material with no upsampling in XX. Hires and 4x upsampled material seems to require a larger SFS to sound right. (BTW, I wrote this before I knew what SFS Peter uses.)

The minimum SFS seems linked with the CPU power in some way...

Mani.

It's the "coming forward" of those frequencies and, conversely, their slow "disappearance" (and emergence of others....) when increasing the SFS, that has sorta obsessed me lately. So far, it seems there are no hard and fast rules for it, only generalizations. Most 16/44 material sounds pretty good (settings as signature below) between 70-100; however, SOME material, Oscar Peterson's "Night Train", 24/96 HD Traks version, sounds better with a setting around 20, while Karen Carpenter, 16/44 home rip (WMP unfortunately, would rerip it but can't find the disk!), becomes silk at 120.

Its because I find myself frequently fiddling with this setting that I ask for the convenience of a button on the face of the GUI. If you guys can make XX automatically adjust to the optimal setting per trak, that would be great too, though I think you'd need to employ magic, or maybe just adjust a cpu power setting or two.....to get that happening.  :)

For anyone who's interested, after quite extensive listening, my 'new' SFS values are:

35 - for non-upsampled 16/44.1 material
50 - for non-upsampled 24/176.4 material
50 - for QAP upsampled 16/44.1 material

These seem to be the 'Goldilocks' values for my system as it currently stands.

If you haven't played around with SFS yet to see how it affects the sound, then you should... asap...

Mani.

What happened to file size related values? :)

Haha, this was only ever a hypothesis and I made it clear that it could just have been a total coincidence. In any event, I shunned the hypothesis as soon as I discovered there seems to be an inverse correlation between the optimum SFS and the CPU power.

Now it could well be that larger files do actually require larger SFSs, but testing this with all the variables at play might prove really difficult.

Mani.

Yeah, a hypothesis which can be very easy verified and for me it's true. Try playing very short tracks, which for 16/44 would have track size of around 15-20MB and compare the SFS of that 15 or 20MB to your 'Goldilocks' values'.

I've just had a session with Stan Getz - The Girl From Ipanema (24/96). The track played upsampled (double) sounds really bad with the SFS at 50. The voice and instruments are very unnatural and the bass is somehow missing. The one thing that in my opinion seems better is high, although I don't think it is meant to be so clear in that recording (reconrded in 1963). Anyway, what I'm trying to say is, that the optimal SFS maybe doesn't 100% equal the track's size, but I'm sure it is quite close and each time I play a small track (file size), the small SFS value sounds best, and the same formula applies for large tracks.

Marcin, what would have been the best size for that track you think ?

The track is 109 MB and I played it with 109, 50 and 25MB SFS. I don't know if that's the optimum, since I can't listen to analog source and compare, but for me it sounds pretty damn good. It would be nice if more people contributed - played around with various values and came here to post their findings. If we find the 'golden' rule then I guess you could implement it to XXHE next to manual SFS setting, as it is right now.

I have been experimenting a bit with split file size. SQ definitely changes and it really depends on your system and your preference.
In my system, larger file size give warmer, may be a tad darker, fuller sound whereas smaller file size is leaner and moving toward drier side
as split file size gets smaller. I definitely does not want to change split file size regularly depending on the file being play.
For me, I find 350mb to be the sweet spot. Anything bigger, the sound is a bit too sluggish and fat, but smaller split file size is too lean for my taste.
So for now, I have to accept the occasional out of memory error. At least now iPad/VNC make things a bit more bearable!

350 MB? That can't work well with upsampling... Maybe with DAP it could work, but certainly not for QAP.  I encourage you to try some very short track, check its file size and set the same SFS. Then compare to your 350MB setting.

I haven't noticed a specific connection between music file size and the appropriate SFS setting. Here things like powerful female vocalists and violins totally benefit from a higher SFS, whereas classical guitar and piano, generally,  sound much better, richer overtones and tighter treble, with a lower SFS.

I'd be interested in knowing whether your preferred SFS values come down with a more powerful CPU.

Oh, and welcome to the modern world!

Mani.

My new Goldilocks value is 36 for all material.

Does anyone have an explanation as to why the SFS affects the sound so much? How is it that we're all pretty much hearing the same thing (irrespective of our preferred SFS value)? Reducing the SFS below the Goldilocks creates a 'thin' and 'shouty' sound. Increasing it beyond the Goldilocks creates a 'thick' and 'too mellow' sound. What the hell is going on here?

Mani.

On the matter of that boundary ... no clues here. But, I tried to avoid anything which can matter, and in 0.9z-3 you will have 3 options :

- The old one;
- A new one with certain organization which IMO can't be influenced by the Size;
- A new one with certain different organization which IMO can't be influenced by the Size.

At least this is a piece of cake to re-implement after I lost the source of it yesterday. Took a month to think about it only. :)

Peter

Peter said, that his DAC is totally resistant to PC :P Joke, we know that sth is going on on a software side rather. I personally don't believe in speed/robustness corelation to SFS value.

Since this thread got started, I have been following a bit and tried various SFS from about 50-400 and I kept coming back to 350.
Generally I use 4x AP with 16/44.1, 2xAP with 88/96 and 1xAP with 176/192. My computer is a dual core 3Ghz with 4GB RAM.
 I don't neccessarily disagree with generalized description of the sound at higher ad lower SFS but how the sound come out will also
depend on the rest of the system as well.

Don't know if its appropriate to ressurect this thread, but there were a lot of interesting ideas in it so what the hell:

Am home today and had a chance for a nice session with XX. Don't know why but it occurred to me (again) that SFS and volume are very interrelated. For the heck of it I set my SFS to 500 and increased the volume between 4 and 6 db depending upon the music. These volume settings were in a range I rarely go cause all gets distorted when playing this loud. However, with a SFS of 500, these high levels sounded very, very good. Reducing volume to a more usual level, but keeping the SFS at 500, things sounded  too woody or muted. Of course, lowering the SFS with the lowered volme brings back that "great sound."

This little exercise got me to wondering if there isn't a "correct" (or approximately correct) SFS for any particular volume level or range of levels, and if there is, or at least seems to be, wouldn't it be great (for those who dislike "fiddling" with too may buttons) if the SFS could be automatically adjusted as volume was increased or decreased? This "setting" would have to be based both on the SPL the music was originally recorded at and the volume you want to use (whether or not wife is in the room, etc) when listening in your living room. The feature could be bypassed (with a button, of course) for those who can't stop fiddling with buttons).

If implemented properly, this could be the "solution" to the SFS "problem". I mean, if volume and SFS are related, there ought to be an accurate SFS setting for each cd and the volume its being played at. Or instead of a specific SFS for each volume it could be set for a range and would change at a certain point that constitute a new range.

Sorry if I've got a bit too much time on my hands and you've just spent a couple of minutes reading a bunch of crazy talk!
 

Probably you are right. I use a SFS of 430 and I noticed that it covers very well a volume range between -36db and -27db. That´s the range I use almost all the time for all my music. With that in mind I realized however that most of the time -33db  is the perfect volume level, but is that due to the acoustical condition of my listening room and to my system (amps/loudspeakers)?, or it is due only or mostly to the SFS setting?. Probably all the parts are important and related.


I really don´t know if that would be easy to do or even if that would be possible, but wouldn´t it be easier to choose the best sounding SFS setting for the most usual volume level and vary the volume a little depending on the album?. For me at least this way works well.

However I must agree with you that this is a relevant issue and that there are important differences in SQ with a certain SFS at different volume levels.

Best regards,
Juan

Hi there,

Although this topic is somewhat longer and it perfectly anounces what could be going on in the first post, I don't even recall I ever dove into this, or really put some attention to it for that matter. Most probably this is because the SFS was a sort of mystery at that time, me not having a clue what was going on with it. Okay, that didn't change all the way per today, but at least we have spades of more experience now and me too understand some things more.

I don't know ...
:swoon:

As implied above, judging the technical merits of the SFS already is a difficult task, and combining it with the amount of attenuation - it is hard to dig for me. However, that you will be perceiving what you do is without doubt.

I never noticed anything with this (combination as described), but that could well be because I am not changing levels so much. I tend to play as loud as comfort depicts, and for example this is never late night listening people being asleep. So this may matter.

Reading about this, my theories would go in the direction of "distortion" being there in the first place, which can be influenced by the SFS. So, whatever its merits are, it should be a smoothening factor when set to lower values. So, there's logic in that to me;

There's also logic in the lower Volume settings encouraging for higher distortion. This is pure theory and is talked about more often, then thinking about the more digital attenuation the more bits we loose etc. etc. *If* that is true net is a very different matter, but for argument's sake let's say that is so.

Now, next it would be obvious and logic that when the lower SFS values encourage for more smoothing - but which is distortion by itself, the additional distortion caused by more digital attenuation is a sort of counter attacked (by the smoothing). Thus, looking at it from this angle, thing would nicely fit.
But then of course I made up a story around what exhibits; not much scientific that is ...

When I only look at myself and what I notice from things, I don't see it happening, but I also don't try. All I know is that you have to have a feeling for an SFS which can to be too high (that squeezing the OS which runs out of memory without you telling) and when you stay under those virtual limits, sound is the best. So, from this comes a 430 while 500 would be possible. Exceed this limit and sound will get stressy, and don't ask me whether that gets different when changing the Volume. I'd rather not know, and I wouldn't like the tweaking. But ok, when this would be an automized job things would be different regarding this, of course.

But as I say it all the time lately, it *has* to be a moot thing all over, when first other distortions are there, and you have those. I can't guarantee that they get out of your way with 0.9z-7, but at least you should try. The worse what can happen is that the less distortion behavior is 10 x better than your worst setting of now. And next the same "issue" can emerge. But, that seems unlikely when whatever smoothing factor is currently applicable for counter attack, when the distortion is inherently there.

Btw, today still (0.9z-7), I'm sure that the higher SFS is the better one, but it is a sort of overruled by "that" other new setting which exaggerates much more of something else which already was good for NOS1 users. By 40 times to be more precise. So, think like being able to set the SFS 40 times higher than now, and imagine it would extend the good exhibits of the higher value SFS from today.

So, wait a bit again ? :yes:
But sure don't forget to come back on this one, when 0.9z-7 is sorted out somewhat on your side.

Happy waiting !
Peter

In the mean time though ...

I have been listening to about the same songs (nicely short, always Beatles) at testing stuff and ended up with figuring out the limits of certain settings and what I should prevent you to dial in (because of anomalies might happen). Then ...

I tried some "under limit" which comes down to a 0.9z-6 setting you can apply today, and it is (at least for NOS1 users) a Device Buffer Size of 4096, Q1 = 15 and SFS = 2.
Suddenly John Lennon began to sing the most musical ...

Not all will be the same at your side, but maybe you should try it. It's not such an obvious combination ...

And oh, might it matter, I have the OSD Running Time active as well (not the Bar type).

And ehh ... this should be a pure smoothening factor as how I meant it. Maybe of the type which is not equal all the time (think of how "musicality" emerges).

Have fun,
Peter

Hi Peter and All ,

I have a question regarding the buffer size .
The device ( M-Audio , Fast Trackpro ) has the latency slider seletcable from 128 to 4096 .
From a forum I did read a thing about a 2x "effect" on the latency settings ( which I didn't quite understand ) so after that trying the XXHighEnd trik , I did try to verify the latency of the device by the means of that method .
Then to 128 buffer size , Adaptive Mode and Q1=0 there is some buffer error ( audible ) but without XX telling me .
Letting the M-Audio buffer to 128 and encreasing the buffersize on XX to the next level , 256 , it does plays without audible errors .
This until reaching say 1024 or less , I believe .

Should I conclude that I have to consider the buffer size of my device with a "2X" on the XX side?

Thanks

stefano

PS : will 09-z7 include the button Arc Predict and PeakExtend ?

 :thankyou:

Hi Peter ,

did try your setting last night , and it was very much more musical . I did listen to "steely dan - alive in america " and " Ed Motta - Poptical " among other albums .

Only one thing was that it didn't play the complete album , instead it stopped it randomly at the end of the second track or fouth track and so .
My settings were in unattended , minimized , and running time .

s

Stefano,

Hoping that I understood the question well :

Whether the Device Buffer Size is set to 128 and Q1 to 2 or 256 and Q1 to 1 makes no difference; the result is the same.

What I would not do though, is setting Q1 to 0 or even lower. The result will be unpredictable and actually this is not made for Kernel Streaming but for WASAPI (I now realize that maybe there should be a protection for using it with KS).


2MB is rather low and the PC must be somewhat faster to cope with it. The allowed 0.2 works sometimes with my own PC, but sometimes not. I think 1MB is reliable for me, but it's a long time ago that I really tested it.
Furthermore I can't compare the situation of 0.9z6 with z7 because so many things changed, but I played with 2MB for many (happy) hours yesterday without any glitch.


Of course !

Regards,
Peter

Great , so is understandable that we will not use KS and Q1=0 for the latency test .


Yeah !

:-)

Btw, especially for our American friends :

At some stage I learned that the term "musical" or maybe "Musicality" is understood differently in the US than it is in the rest of the world. We, overhere, know it from being 2 years old that this is about how sound turns into music. Hard to explain, but maybe when something sounds more musical it sounds more vibrant. Also we dedicate more "performing for you" to it.

Maybe others can explain it better.

Peter

I've been amazed by this quote since you wrote it! Please, please, please provide for my American ears the definition of the word bit. Cause over here it means the briefest passage of time...... :)

Haha, you audio freak. The definition for us is simple :

A time period of a bit is defined by the sample rate. Look :

One stereo sample through your NOS1 consists of 2 x 32 bits. So, that is 64 bits (btw, this is equal to 8 bytes). All your samples are this long.
But what is the sample rate you use ? That's up to you.

Mine is up to me.

Clear ?

:swoon::swoon:

I've tried these settings and yes found them to be more musical but suffered from the same result as stefanobilliani.

I have changed SFS to 20 and it now plays all the tracks on an album.

I never went back. See my sig ...

:)
Peter

I guess all the *...* are new features ... :pleasantry:

Nah, just significant changes compared to previous settings (thought to be important).

Peter your settings are much better after doing lots of comparisons. I've gone back to 4096, 14, 2 SFS and its definitely more musical and detailed. Now I need to make a decent computer and throw away the laptop, buy a Phasure Dac and hope I can afford the next software version "z7".

Under Pressure as Freddie Mercury once sang!!!!

Peter are you stating that having the music on a seperate usb3 drive is a significant change? (for the 'better' ?) Is this from a convenience or also from a sq perspective?

I have limited sata connections so having no or a negative drawback  ;) when migrating to an usb3 hub would suit me very well.

Regards, Coen

Nope.




:swoon:

When I apply 4096, Q14, SFS 2 it stops after 4 tracks and a box comes up "No Track Data". I can only push out the stopping by raising the SFS to say 20. Disappointing as it sounds better at this setting.

Several times I've tried Peters settings and everything but the SFS of 2 works for me. I'm staying with my SFS of 450, though I've changed my signature regarding the other settings. Voices are eerily natural on my system at this SFS setting. Maybe I'll change my mind when, geeze, whats that new thingy called, oh yea, 9z-7 is released.

Peter, please try and change my mind soon.  :P

Brian

You are right Brian we all need to be put out of our misery and load up z7. The problem is once I've heard a sonic improvement I don't want to go back.

Its a worry we are needlessly wasting time on settings for  z6 that may be redundant with the new version.

Thanks to Peter for his latest updates on the inner workings of XXhighend it does help to understand what is happening.

Robert

What seems to be working best for me with a Wyred DAC2 is buffer=2048, Q1=14, SFS=2 and yes its definitely more musical and detailed.  An important difference is that I'm using Normal mode of Engine #4.

The 2048 buffer size may have to do with my use of 2x or 4x upsampling (rather than 8 or 16 as others are using).  Also, I'm now using PE and AP or at least AP + Minimize OS.   None of this is reflected yet in sig (at the time of this post).

So this works on a non-Phasure DAC.

I've gone back to SFS430 but no not as good soundwise. I'm currently at 40 SFS so good so far.

I forgot to mention that with the settings I posted two posts ago, that playing does not stop after a few songs.  I've queued up over 15 tracks and it played through all of them.  This is with 44.1.   With higher sampling rates and unattended mode, I have trouble sometimes even with the first track.  But HiRes sounds good attended, with the advantages in convenience of attended mode.

So I recommend, if you are having trouble playing more than 4 or 5 songs with SFS=2, try using the Normal mode of Engine#4.

I've now moved to buffer=3074, Q1=14, SFS=2, Engine#4:Normal, 4x AP+PE, Unattended, MinimizeOS (will update my sig once I've settled on this experiment, if I do).

Otherwise, does anyone else trying SFS=2 notice a degradation in SQ with use (as memory fragments)?

What happened last night is that I started with pristine, detailed, and musical results with SFS=2 on a known "bad" recording (Wilson Pickett's Midnight Hour CD).  Then I switched over to JRiver to listen to high res DSD.   After an hour I exited JRiver and tried to play with SFS=2 and the same Wilson Pickett song, Mustang Sally, sounded awefully distorted.   HuH???   Yes, it happened.

Maybe SFS=2 SQ depends on how fragmented the memory has become over time.

I can report , that I fairly noticed that the stops playing various albums at SFS=2 does happen "always" at the same passage to the next song and "always" at the same song . For example one album stops after the first song . One album stops after the fifth song .
At SFS=4 all albums flow to the final song ....

Charlie,

This *is* something I am concerned about myself. It is not for the exact reasons or way it will work out as you describe, but it's certainly in the same area and phenomena.

Because I know about this, I am the most careful with selecting the programs I need to fire up before a playback session, which btw always is preceded by a reboot. This is not the case (read : I don't feel like needing that) with the higher SFS amounts (like 350-450).
It is not that I ever proved that it works out like you suggest, but maybe you just did. So, no matter the way you describe it can't happen in my view, let's take it for granted that for net result something like that will be in order - when not careful.

I did not ever notice degradation during a playback session, but in my case this will never be longer than 3-4 hours, plus I always reboot in advance (I have to, because I use the same PC for developing the lot and this requires Normal OS Mode).

Thanks a lot for sharing Charlie - it could be important ...
Peter

Stefano, you too thanks for sharing.

I play with 2 all the time and never anything stopped anywhere. This is a bit dangerous, because things *did* change in this area (for the version I use), so maybe that's it. Otoh, what sure can happen to me is that I hear 1-2 seconds of a track and then all stops again. This is important to the kind of same issue I think, because it cannot be repeated. So, let's say that one out of 15 album-starts this happens, and when it happens it can even happen 3 times in a row before it finally plays.
Btw, most probably the same will happen to you if you try 0.2 or so. It needs that "flow" like a water lever.

Supposed this won't change for you or anyone bugged by it in 0.9z-7 - I will never attempt to solve it. It is a crazy complicated thing, since this is still official memory playback, but then at micro level so to speak. So, if 4 works, then please stick to that and hope it won't make a difference too much with 2. Small problem : it most probably will from theory, and in fact I am waiting for the day to try 4 myself. But I didn't so far because I like so much what I'm hearing and I don't want to change.

Regards,
Peter

Welcome Peter ,
I am trying right now to the set of SFS=2.2  and 2.4 . It helps a lot . wait .... wonderful and relaxed sound with 2.4  :yes:

Edit: my XX 0.9-z6 defenitely plays all songs in a flow at SFS=2.8 .

s

I knew that there was someting to look at , regarding the stops after track ending at SFS=2.0
Once went back to the Memory and Disk Utilization part of the Setting Area did find out that " include garbage collect " button was active . I did suddenly remember some implication from the time I was using a laptop with very little Ram .
It was stopping after a song when " include garbage collect" was ticked .

Now SFS=2.0 plays without stops , when "include garbage collect" is NOT active .

Well yeah . Nice Find for today . :)

Shall I add more decimals for you ?
:rofl:

LOL!!!

 :dancing:

WHAT ?!
(I didn't see this post before)

Well it is not true . It did play completely one album that was stopping before .

And *it stops* at another album , exactly where it was stopping before .  :scratching:

Isn't it strange ?   :fishy:

By the way I can ear a distinct sound quality change between 2.0 and 2.8 SFS .

I am sure of that.
SFS=2.0 has a certain "collapsing" character of sound to it , which goes away at 2.8 . And that is it for now

Best Regards

s

Hope this may provide more data on the SFS = 2 playback stopping.

Playback stops here as well between tracks often at the end of the first track, sometimes as many as four tracks will play. In my case I decreased SFS to 1 and have continuous playback. I do like the new low SFS with Q1 at 14. It provides a very nice presentation as people have described.

I prefer SFS = 1 to a value of two, will try slightly higher values and check the garbage collect setting with SFS = 2 to see if the playback stops are cured.

Nick.

I must say I prefer the good 'ole settings for engines 3 and 1 over the low sfs. Engine 3 actually sounds more 'musical' (the Peter Dutch style) to me.

The low sfs seems to add a bit something, but for some reason after some listening I feel a lack of something else. Ok, this is vague but I dunno what it is. It somehow wears fast.

I'm going to wait for v07 and apply the settings in their proper fashion.

Regards, Coen

Well putting the music disc in a seperate usb3 housing has been a more fulfilling experience. It deliveres sq results that are of the kind that you want to stay a little longer with. No wear yet  ;)!

Regards, Coen

Hi Peter and all of You ,

Glad to hear that 0.9-z7 is near soon !

By the way yesterday I did reorganized my music disk  carefully putting the folders as described in the very helpful section of the forum called " Guidelines and Manual " .
Then I had various hours of listening in the famous SFS=2 settings .
Suddenly the quality of the music began better and for a long time , with just some occasional stop here and there .
Also a connection of the (sata)disk through and external USB box  port was in the tests , it did seem good to me .

Sometime I only wish I could listen through a NOS dac .  ;)

But soon who knows  ... :)
 

Oh, but Stefano, you already have quite some parts ! The software, the PC, some dongles and something like a Smartphone of which I only can hope you can work with it well (I can try it myself in a few days time - just ordered one today - yea, I'm behind a bit).

haha

Great  :)

Nope.  Like at least one other, I'm moving back to my original settings with SFS=72 (as in my sig at the time of this writing).

I found that I was moving my buffer size to larger and larger sizes to get smoother results.  In other words, I was hearing the eventual distortion that SFS=2 was bringing with Normal mode of KS.

BTW, the driver for my DAC supports buffers as large as 8192, but xxHighEnd only goes as high as 4096.   Can 8192 be added?

With KS that's just a matter of increasing Q1. So, Q1=1 equals one time the Device Buffer Size. And so on. That's for Adaptive Mode. For Special mode it works with samples (which are way under the Buffer Size for "length"), but the principle is the same.

Peter