Peter, I wonder why SFS values between 10 and 11 are allowed, but between 11 and 12 are not. What's the reason for that?

I am testing now between 10 and 11, with good results.

Kind regards
Richard

Hi Peter,

in your signiture I can see "USB3 from Mobo". What's the best way to have it like this with my Stealth Mach II?
I still have the USB cable attached to one of the plugs in the rear.

Kind regards
Richard

Hi Stanley,

that's funny. You and me posted quite the same in the same minute independently. I read your post after I had posted mine.

You: SFS=10
Me: SFS=8, SFS=12

This can't be accidentally.

Kind regards
Richard

Hi Peter,

I tried SFS=8 once more.
Yes, if you prefer cristal clear highs, then that's the better solution.
Mercedes Sosa's "Missa Criolla" for example sounds perfect and very real with this setting.

When I listen to the 24bit version of St Germain's album "St Germain", I prefer SFS=12. So the highs don't dominate the scene too much and there are more mids and a fuller body in the sound and stronger basses too.

What's good to know now: With such high buffers, the big changes of SQ happen with SFS between 8 and 14.
I will do more listening with different SFS settings within that range.

Kind regards
Richard

Hi Peter,

SFS = 10 is also worth a try. Already pretty more "analythical" SQ than with SFS = 12. So, between 14 and 10 SQ changes pretty much.
Below 10 I didn't find values with such good sound for my ears.
I am looking foreward to your experiences...

Kind regards
Richard

Hi there,

during the last days I listened a lot with release 2.09 using Q parameters like:
Q1 x xQ1 = 30x40.
Q3,4,5, = 1,1,1.
SFS = 20.
ClockRes = 15ms.
Appointment Scheme = 3-5.
OSD Text = Off.
Custom Filter = second highest or highest for 705600

Challenged by a pretty good new DAC, I borroughed from a friend, which is doing internally PCM to DSD conversion, I was trying different Q parameters.

The result:
SFS = 12    sounds magic in my setup!

There is an amazingly big difference to SFS = 20, where the highs were not so clear, as already stated in another post above.
I also tried very small SFS, like 0.9, but for me SQ becomes too thin then and there is kind of a bit of hashness with such small SFS values. The benefit of good highs is already there with SFS=12.

What's also good now:  There isn't any glich to hear any more after a few seconds now.

Kind regards
Richard

When I was writing my last report here Nov. 14th, I had my equipment running continously for 8 days.
The next day I rebooted the audio PC and switched the NOS1-a off and on with an interesting effect: The too hard, a bit annoying basses in some tracks were gone. SQ was well balanced over the complete frequency spectrum. There wasn't this kind of too much dynamics in the sound any more.
So, probably I had kind of low level distortion after running the systems for 8 days, what I could realise well in a few tracks only. I dont't know if it was something like Peter is describing above. For sure it was on a pretty low level and therefore a bit hard to dedect. And I changed the Q parameters a lot during these 8 days and therefore believed that this caused all the changes of SQ.

I had only time 2 or 3 hours to listen then and I don't know if it's the rebooting of the audio PC or the switching off/on of the NOS1-a what brought this positive effect or maybe both.
I will do some more investigation on this when I will be back home next week.

Kind regards
Richard

Peter, yes, it was the blocked file problem. It's working now (version X1). I could have known that myself.

Yes, I am using the "End" button pretty often. I will avoid it now and see if burst of noise still happens.

Kind regards
Richard

Peter, I now downloaded XXEngine3-209-X1.rar.
After I had copied this new XXEngine3 exe, in XXHE the output device is not found any more. Under setup this field then is empty and there is nothing to choose.
I tried it one more time after a reboot and got the same result. When I copy the older 2.09 engine3 exe over it, it's working fine again.

Kind regards
Richard

Today I have got this burst of noise again, scaring loud and this time it didn't stop itsself until I pressed the stop button.
I was playing the same song, the last one in the playlist, like 4 or 5 times before it occured.

Yesterday I had another phenomeon: After selecting new songs in the playlist and pressing "play" the last seconds of the previous song were played (out of the buffer?) and playback stopped after this without any error message.

As reported I didn't have any of these problems for aboit 16 month before I upgraded t o 2.09 and changed the Q settings a lot.

Kind regards
Richard

Hi Peter,

I read your so detailed and straightforward report from last Sunday a couple of times and did some intense testing yesterday.
As you described, all seems to be a bit crazy now. And that's the way I see it too. Maybe there are several innovations and changes that are interacting now, building up something special - predominantly in a very positive way, especially with your actual Q settings.

I try to recall my experiences during the last two months:

- First, the 3G upgrade brought already significant changes like more precise, "detached" sound, stronger basses, nice highs. But all this in a more decent way, not as extremly as it is now together with 2.09 and the latest Q settings.
- Release 2.9 brought the biggest impact on this, espacially for highs and basses. As I stated before, I switched a dozen of times between release 2.07 and 2.09, because it wasn't so easy to find a judgement about the "new" SQ.
- Finally your so different new Q parameters brought amazing effects.

I never had activated OSD text since more than a year. So, for sure all this mentioned above does not depend mainly on it. I just tried OSD text yesterday once, for a short time only, because with the actual setup it really has a horrible impact on SQ.

An interesting point probably related to this: For me, as a NAS user, cover art never shows up in unattended mode anyway. Instead of seeing it on the screen I get a short glitch a few seconds after the song starts (when the cover  art should probably show up).
With lower buffer size (like 14x5 or below), there is nothing like this to hear, no glitch at all.
The higher the buffer size the stronger the negativ influence of this kind of disturbing activity obviously is. That is perhaps the reason for the behaviour with OSD text too.


Peter, with your new settings most of the tracks I listen sound incredibly good now. Trumpets don't sound hard and harsh any more, but airy and very "real" instead. Voices are tremendously present and realistic.
With these new settings it's again a new experience, amazing clear, airy and natural sound, with acoustic instruments too (what's always a good indicator). So that's another big step foreward.

What's so impressing is this extremly dynamic sound. For years I was searching for more dynamics in the tunes. Now, sometimes with some songs I am thinking... "Isn't that maybe too much now, especially for the basses?"
Stong, precise highs usually bring precise and punchy basses. So probably the +6dB in the high frequencies is one of the reasons for that.

It's only the basses what still give me to think now... they become so strong and punchy. That was already my first impression, when I got 2.09 to run on my audio PC. And with the new settings the basses get even more punchy, maybe the additional 16dB you described above.
Maybe that's the point Juan was described above.
For sure, all this depends on the type of music, on the recording, on the acoustics in my living room (I have got stong basses anyway due to the geometrics of the room)., So perhaps it's just the room modes which exagerate now and sometimes become annoying.
Let me say it like this: about 90% of the songs produce spectacular SQ now. It's he minority that causes too strong, umpleasant, sometimes annoying basses. Maybe there is no pleasant way to listen to these songs any more with higher volume (in my room?).

Therefore I was trying different filters yesterday. For the first time, ArcPrediction is an option for me, specially for easy listening. And the second highest custom filter (25010) always works fine. That's the one I was using most of the time anyway. And now again in prooves to be the best for me.

You are right, we have to re-adapt ourselfs too, to a new style of listening. I am on this path since 2.09 has been installed. What makes this a bit difficult are the so huge changes of SQ in such a short period.

Thanks a lot for all and especially for your detailed explanations last weekend.
It's great how openly all these issues are discussed here.

Kind regards
Richard

Hi there,

maybe I can contribute a bit finding the reason for this burst of noise.
I had this problem a few times about one and a half years ago, right after I got my new audio PC, NOS1-a and XXHE running for the first time.
Peter, as I remember, I reported it to you in some emails.
I remember that it also happened with highres files (maybe with 16/44 files too, what I don't remember).

Later on, for some still unknown reasons, I never had this burst of noise problem again for about 16 month. Not even once.

Recently, since about two weeks only, I got it back. It happened 3 times that this burst of noise frightened me, always for just about 1 second only and music stopped playing right after it. In my case it happened when playing a whole album in the e.g. during the 5th track being played.

So, what did I change recently:
1) Mid of September I began using my 3G upgraded NOS1-a. But that's too long ago. Therefore I don't believe that this is the reason.
2) About two weeks ago I installed release 2.9
3) About two weeks ago I changed the Q settings pretty much.

It seems logical that 2) or/and 3) probably have caused the reincarnation of "burst of noise" in my system.

Ad 3): my changes of the Q settings:
My old settings during the 16 month without this problem had been (more or less never changed much):
Q1: between 14x1 and 14x10, Q3/Q4/Q5: 1/1/0, SFS: between 20 and 120, Clockres: 0.5ms, Coustom Filter: High (2nd in the list).

My new settings now are:
Q1: 14x40, Q3/Q4/Q5: 0/0/0, SFS: between 0.7 and 2.6, Clockres: 1ms, Coustom Filter: High (2nd in the list).

The biggest difference between those two sets of settings are:
- Buffer size Q1xQ1x
- SFS

Maybe that helps to find the culprit...
Kind regards
Richard

During the last days I did some testing with different Q parameters and got amazing results.
The values I prefer now with release 2.09 are:
Q1: 14x40, Q3/Q4/Q5: 0/0/0, SFS: 0.7, Clockres: 1ms, Coustom Filter: High (2nd in the list).

I tested also with very high buffer 30x40 and got the same effect, Peter described above when listening Frank Zappa's "Jo's Garage": very detailed and clear sound.
But, other tracks with acoustic instruments, like e.g. Clark Terry's trumpet, sound too hard and harsh with these settings, it almost hurts. Using the AP filter smothes the sound pretty much, too much in most cases. Therefore I prefer the costom filter.

As I can see, all that is pretty different to other reports here. The reasons could be:
- I am using a tube preamp and tube amps (brand Ayon). I have to try it without the the preamp once more, what I didn't do since I received the 3G upgraded NOS1-a.
- My Lush is 3m long (I am the first and maybe still the only one using such a long Lush USB cable).

Peter, I believe I should try a shorter Lush just for a test. What do you think?
If this will bring a significant improvement I will have to regroup completly my hifi units.

Once again I have to state:
With 3G upgrade, release 3.09 and these new Q-settings SQ made such huge steps foreward, I could not have imagined before.

Kind regards
Richard

Hi Peter,

I have got a question about the obviously big differences in SQ between die XXHE releases 2.07 up to 2.09:
Would it be possible that these factors which influence SQ in that way, could be adjusted by the user in a future release? I am thinking about a few more Q parameters.

Kind regards
Richard

"What a difference a release makes..."

I got version 2.09 installed and running yesterday. I pasted it over version  2.07. So, I skipped 2.08. At first I didn't read the post above in this topic and therefore didn't expect much impact on SQ.

But... listening to the first song... what's that?! - powerfull, strong basses, incredibly clear sound and perfect soundstage.
The basses were even too strong for my ears and I was thinking: "Isn't that maybe a bit too much?". Especially electronic basses sounded to hard for my ears. To compare I changed between 2.07 and 2.09 a few times: incredible, what a difference a release makes!

Later a read the post in this topic and changed my settings to Peter's values. Yes, much better, not so hard and harsh sound any more, especially when I used Arcpredic.
In general I am not a fan of AP, because in my system it always caused too much smearing what also happens a bit with version 2.09.
I was impressed when I went to sleep last night, but not 100% convinced yet about naturalness of the sound, especially for acoustic instruments.

This afternoon... another day... and it's perfect. Somehow all what sounded a bit too hard and harsh yesterday was gone. Maybe the hardware like NOS1a and amplifiers had to get used to the new tunes too, not just my ears 

There was no doubt about great SQ for electronic music, but for acoustic instruments? Therefore I played acoustic tracks, like from Stephane Grapelli, Rob Wasserman, etc., compared SQ my amazing Platine Verdier turntable and my other DACs and I have to say:
YES! 2.09 brings amazing sound, for sure far the best I ever got out of a DAC. And it's very natural sound.

I wasn't sure about AP, switched it on and off pretty often and I prefer now to listen without AP because of the so clear and precise tunes.

Great job, Peter!

Kind regards
Richard