NOS1a and Q5 & W7 vs W8

[PeterSt]

Re: Orelo MK-II Imaging.

All right. I pinched my XXHighEnd PC together again yesterday and I started out with Windows 7 on it. Played an album of which I knew I did not play it often any more because "too difficult". Notice that this is about "too difficult" for W8. It is space-like synth stuff and no roar etc. would/will emerge in W8. Highs are too profound or something.

First off, I found no different perception under W7 than on my other W7 PC (with 13 disks attached to it and which can be regarded VERY noisy (also acoustically ). So I think the "machine" as a cause can be ruled out.
The album worked as intended (well, that is what I must assume) and I felt nothing much wrong with it. Do notice that I by now was rather used to the W7 sound, but with included "speaker tweaks". Also, I think if I had to survive with this forever, I could do it; maybe I wouldn't like all types of music for the best, but alas.

With this proof at hand that it is not the PC which matters, and also not forgetting that *I* personally can't imply differences with XXHighEnd settings, I finally went back to W8. And yes, I was more afraid that I would be disappointed than that I was looking forward to it after these 8-9 days of "struggle". The palpability of W8 has something and as said earlier, it can come across like LP (in comparison with W8).

So there I went ...

In 45 minutes of time I was not able to let that same album sound decent by any means; It was now cold, sterile, no roar anywhere and there we go again : made by robots.
D*mn.
Tried all the speaker settings I could think of, but nothing satisfied. Strangely enough the W7 settings satiesfied best.
And so, if we look at this (album) I must give Coen all the right in the world, because there's no comparison and W8 is awful. Still though, all I heard from the bass was infinitely better. There was no deep from down earth umphf boiling up, but eatch same from deep down earth hits had an attack, envelope and decay. So what do we want guys, perceive one big down earth umphf, or hear what the instrument really is about ? And again : d*mn it ! it has to be about the latter !

But this was such an album (but by now quite rare) of which I knew it didn't work any more since W8. So let's behave normal and play the stuff which "works".

First notice : all the flavours have gone. Read : W7 is one big pile of deep down earth flavour, with a sauce of grey on it on top. Interesting and it could sell, but even me should get annoyed by it after some time, never mind I just told I got used to it.

Further notice after playing somewhat longer (say two hours) :
All is so "normal" because natural. I supposed this is all related to the grey now lacking and the accuracy in place of it. I even forgot that cymbals could sound nice and interesting, because with W7 they're just not there (envision how grey mushes them into deep down earth nothing ).
A more strange thing is that again an hour further I could not imagine how this can ever go wrong. So, the W7/LP idea actually, but now 10 times emhasized. How contraditionary ...
Still there are those albums which will not work, but as far as I can recall they are always this synth sh*t. So let them be.

Or ?

It is very very dangerous to let such thing be. I mean, one of the worst regarding W8 robotics has been Vangelis' The Friends of Mr Cairo. More light than light weight and all completely nasty. This improved over (XXHighEnd) time, but with the NOS1a this went totally normal. Really so and all the examples I had which were so without ANY bass, now work (Heaven 17 is also a nice example). Many synth stuff - same story. But still not all.

So must I go back to W7 because of this handful not working in order to let 1000s not work optimally ?
No. I refuse.
I better sit down and work out the frequencies where W7 behaves totally different from W8. That seems doable to me. Next I will be able to tune the speakers for where W7 is the better one and apply that to W8, Mind you, if so at all.
But is such a thing useful to you out there ? No, to most it is not of course.

So I don't know what to do about it. I feel that W8 can be improved upon but at this moment I don't know how. I *know* that W7 is all over for the worse only and that plain distortion masks the robotic accuracy of W8. I personally can not proceed that route. Distortion is just that and it should not be the way to go.

More interesting of course is : how can it ever be. It shouldn't exist but it does. Bit perfect = just that for both. Well, we have talked about this (way) more often. Both just can-not be the same. Also, I don't know what actually happened that the difference is now so enormous (I am not exaggerating), which should only be because the DAC is again better in reproducing what it's offered; before we may not like the one over the other, but this is really out of space. If I only take my drumming recording and observe how that is TOTALLY messed up in W7, then what ?

Side note because it hardly has been a subject in here :
W8.1 too was like a drag; I now thus have no reasons to believe that that suddenly changed (with the NOS1a). But I can fairly easily try it. Who knows ... it can be in between W8 and W7 ...

All I can reall say is
But we do this for over 18 months already (W8/W7).

Peter

[fralippo]

Yes, please try W8.1 and report.

[PeterSt]

Please notice : I already have tried W8.1 and it totally s*cks. No need to try that again for you.
What can be different now though, is how it sounds through the NOS1a.

Of course I was not clear about that much, and it requires the "knowledge" of this (my) judgement in the first place (maybe 7 months or so ago).

I will do it anyway, but just in case it works out for the better over here ... for you it will be for the worse. I mean, nothing changed !
But ask away when it is not clear, ok ?

Regards,
Peter

[manisandher]

So I don't know what to do about it. I feel that W8 can be improved upon but at this moment I don't know how.

Hey Peter, try taking hyperthreading OFF with W8. In my setup, this brings the LF back up quite a bit and tends to smooth the sound overall. (I'm using processor scheme 'Core 3-5' with a 6-core CPU.)

Don't over-think this one... just try it!

Mani.

[esimms86]

So I understand that the PC's processing power matters, being able to upsample with XXHE matters and filters matter, but I thought that, otherwise, the NOS1a was immune to software settings. It looks like that is not the case?

Esau

[PeterSt]

Hey Peter, try taking hyperthreading OFF with W8. In my setup, this brings the LF back up quite a bit and tends to smooth the sound overall. (I'm using processor scheme 'Core 3-5' with a 6-core CPU.)

Thank you Mani. But I don't think this can go "just like that". So look at my sig (W8 part) ans see the XTweaks settings. Those are still mine, and I am pretty sure I can't let stay them like that when switching off Hyperthreading. So what were yours and what are yours ?

Cheers,
Peter

[PeterSt]

Hi Esau,

So I understand that the PC's processing power matters, being able to upsample with XXHE matters and filters matter, but I thought that, otherwise, the NOS1a was immune to software settings. It looks like that is not the case?

I yet have the first change to see (hear). But let's say that indications exist which now show how things come in from the back door (mains). Let's just recap what I all had to do to let W8 perform : remove all switching power supplies.

This is not an easy task if you incorporate the PC itself and envision all sorts of devices connected to it, including USB disks with SMPSes into the wall somewhere. Long story short, for me this resulted in :

- No such devices attached to the audio PC (watch the monitor !);
- The audio PC only containing a small OS disk and nothing else (but a USB connection to the NOS1);
- That audio PC in the same mains ring as the audio gear;
- A "Server" PC containing 12 music disks, including USB connected, that PC connected by means of a LAN cable to the audio PC;
- That "Server" PC also connected to the same mains ring;
- All the USB disks etc. SMPSes in a normal house ring.

The latter two are logically wrong because
a. The disks etc. connect to another ring with different ground potential;
b. USB (etc.) most 99.99% probably not isolating, so again I was connecting 2 mains rings (for PE hence different ground potential);
c. That "Server" PC does not need to be in the audio ring - but it just was.

Ad c.
The LAN connection between the two PC's *IS* isolating.

Crucial side note : The W7 I was using the past week was that "Server" PC, while I did not perceive any difference with the W7 on the Audio PC. This can be summarized to a most noisy "Server" PC compared to a completely lean audio PC, both sounding exactly the same for W7.
Still I have the clear experience that the audio PC had to be made that lean in order to let W8 perform.
Let's keep in mind : This has all been in the realm of the USB connection from audio PC to DAC because it was the most logical path to think about. And maybe it was or maybe it was partly. We can not say "not at all" because of the huge difference accomplished with the now isolated NOS1a which also is clearly measurable (jitter).

That all said, yesterday I finally moved the "Server" PC to the basement (a task I should have done 10 years ago because the house was explicitly made for that), where I moved ALL the mains connections to a normal house ring. So the whole lot of PC itself plus USB disks etc., now is in one house ring, while the audio PC is in the audio ring with nothing else but the OS disk in there, and just saying : the monitor not even connected to it, which also is crucial (so I access it through RDP). The LAN cable connect the two PC's, and this is isolated.

Result : Way better again.

Now, those who can follow, can see that this is not anywhere related to the USB connection; nothing changed there. All I did was separating the audio ring from the house ring better (or fully now if all is correct) and *that* thus does a few things.

What it does is very indirect because all it *can* have done is providing a better mains to the audio gear (DAC + amps). And well, now we are in an area which is vague to begin with.

Still there ?
When any "Server" PC can molest the mains as a whole for the audio gear because connected, this just as well works the other way around : the audio PC molesting that same mains as a whole because connected. I deliberately say "connected" and not "connected to ..." because we must try to derive that from my text above and it can be different for everybody. But point is : without notice we connect two rings and whether the one element influences the one ring and the other element the other, it is about the result of both which is the same because ... connected. The influence may not be as impacting of either, but the result ends up in the audio ring.

This is of course when we *have* two rings, but let's say that most do already by means of filters, isolation transformers etc. etc. And, might you put really all the (also switching) supplies in the one ring you use for audio, then the result is also bad (because all switching supplies just are).

So what did I all say ?
That I used the "Server" PC to influence SQ via the mains, and that I see no reason why the audio PC wouldn't be able to do the same - via the mains. Whether the latter is still possible through the fragile XXHighEnd settings - I did not hear a thing of that yet, but with the notice that I already did quite OK with the separation of the rings (which you wouldn't say after reading this post, haha).

Peter

PS: Please don't blame me for showing some enthusiasm for the path of "tweaking is still possible". I keep on feeling a quite dead hobby without it. Also it won't be a coincidence that I just now finally made this move, meant to be for 10 years ago. When nothing else is to be done ...

[CoenP]

Hi Peter,

The importance of these "tweaks" cannot be underestimated and I did not take serious notice about them being crucial to win 8. I realize that there is a lot of work to do on my suppy lines after reading your last post. In other rooms for instance there are SMPSs connected to my audio ring. Also I can improve upon PC leanness by moving disks to a server and the disabling the monitor.

Still it is very possible that the audio PC, which has an SMPS which causes hard to kill HF on the line, influences the power amps.

regards, Coen

[PeterSt]

Still it is very possible that the audio PC, which has an SMPS which causes hard to kill HF on the line, influences the power amps.

Sure Coen. And the more we can proove that this sh*t matters, the more the one remaining (that of the audio PC) will become profound (compared to when it would not be there).

Peter

[CoenP]

Sure Coen. And the more we can proove that this sh*t matters, the more the one remaining (that of the audio PC) will become profound (compared to when it would not be there).

If we assume the USB link between pc and DAC is sufficiently isolated wrt noise, only the PC "contamination" of the powerline remains including the hypothetical win 7 and win 8 "signatures". Imho this can be dealt with by a fat battery pack and an inverter or better some DC-DC converters that power the PC directly. I'd rather put the audio PC on batteries than either the DAC or amplifier. Along the road I've looked at this several times, it shouldn't be hard to do, my PC is consuming less than 50 watt at all times.

Did you do something special to isolate the LAN of the server pc or is it standard Ethernet practice?

regards, Coen

[PeterSt]

Is standard Ethernet practice.

If we assume the USB link between pc and DAC is sufficiently isolated wrt noise, only the PC "contamination" of the powerline remains including the hypothetical win 7 and win 8 "signatures". Imho this can be dealt with by a fat battery pack [...]

No. If it would be the OS implying now the mains noise differences, all the XXHighEnd dials previously implying clear differences would still do that. But they don't.
So see ? This might be the case with you (with good reasons for it) but here not, or not perceivable by me, while the difference between W7 and W8 is out of space.

So something doesn't match up and the only thing I can think of (but it apporoaches two years that I am saying this) is that both are not "bit perfect" at the point in the chain where they should be. Which is at the end of the USB cable, so to speak.
But seriously, things get too complicated. Or let's say that I don't have sufficient time for it all, like setting up the measurements, learn what to look at, how to interpret it and what I don't think of at the moment.

Complicated ? ... First we (or me) think that W8 is totally out of order and W7 is very OK, and a year further and a LOT of tweaking (XXHIghEnd software plus mains stuff) ... and almost without notice ... now I think the exact other way around ?
It must be me myself out of order.
And Coen, remembering about your "off and on" with W7 ... didn't we all have the same one day OK, next day pure sh*t with W8 while nobody complained about that with W7 ? Well, if I observed correctly even that now happens to me with W7. One day OK, next day unlistenable. This while W8 is super stable now here regarding to that.
So that too exchanged position, as it seems.

In itself it is a good thing to think (or know) that the DAC isolation is in order, which removes quite a lot (at first I thought all) out of the equation. But I can't really say it became more easy thinking because of that because now we must try to find the means how for example your Q5 makes a difference. If you are correct on it of course, which I for now must assume. Mani's idea not different.
And Yes, *that* should be able to be counteracted with mains isolation but FIRST please arrange for the things I layed out (I am sounding like a wise guy, but 24 hours ago I didn't do that well myself).

At least it gives us something to do.
It would be nice that this "to do" is detected by means we can apply to learn that it only gets worse, like in my case (I am really satisfied). Anyway, say this is part of my job and I hope I can learn from that in order to help people like you, Coen, who can get really annoyed by the need "to do" because otherwise sound is poor. Still it would be my advice to start doing things with W8 in order, because while thinking I have all quite OK over here, W7 shows the annoyance all over (and all is still as OK of course). So I think it is the better path.
Meanwhile - and just a hunch - find yourself a longer USB cable and move away your NOS1a as far as possible from the other audio gear. If that changes things, move it back with the same cable so you'll know it's not that.

Peter

[CoenP]

Hi Peter,

Thanks for thinking along. First things first of course, and those can be done at any time!

I might be back on the forum for the RDC which I never put to work (got the dongels though).

regards, Coen

[PeterSt]

Coen, just use WiFi. No need to use Bluetooth with the NOS1a.
Says me.

[CoenP]

Coen, just use WiFi. No need to use Bluetooth with the NOS1a.
Says me.

Just to be shure: that is wifi to the access point router and then via the LAN to the music PC or does it assume a server PC that is not part of the LAN? I guess Wifi dongels do not fit in the minimal audio PC setup...

regards, Coen

[PeterSt]

Correct.

But before you ask, derived from my previous post(s) today :

I coincidentally control the "Server" by it, that one communicating with the audio PC. Actually the same procedure as proposed with Bluetooth but now for another reason. So :

Tablet (RDP) -> WiFi -> Access Point -> Router -> LAN -> Server PC (RDP) -> Direct LAN between Server PC and audio PC.

The coincidence here is that I use the LAN the other way around : Audio PC must get the music from somewhere. This is form the "Server" PC. This is through direct LAN connection (crossed cable).
Since this connection is there anyway, I can now use the "Server" PC to remotely control the audio PC.

Thus, RDP on tablet remotely controls the Server PC, while that one itself remotely controls the Audio PC. So I (on tablet) take over the screen from a PC which takes over the screen of the audio PC.
The advantage of this could be that now really never the monitor of the audio PC is needed because I use that other ("Server") PC for that. Meanwhile I can use the tablet whenever I want.
Would the tablet connect directly to the audio PC, then I couldn't do all on the tablet or at least not in convenient fashion (just think general maintenance stuff).
Also and actually the same : Many people will work with RDP already, from a laptop or other desktop PC, to control the audio PC. So what people should do is just remotey control that laptop/other desktop.

It's not even counter-intuitive, but still not the first thought - I suppose ...

Peter


PS: So I should finally take out the video card from the audio PC (the whole PC consumes 49W at this moment, during playback).