My first Windows 8 experience

[Jud]

The way to go is a different one. It looks a bit like this NAA from Miska, but the approach is different. My way. You will see, hopefully soon ...

Peter

Always the tease.... 

[PeterSt]

Will give a more detailed response, but for now let me just say that whatever is behind the "Wow!" factor of W8 is front and center on the Hellborg track. Geeze it sounds alive and tonaly/harmonically rich with each and every instrument.

Brian - Not to encourage you to come up with further details faster than you can cope with, but is it possible to say this in Dutch ?
I really can't tell whether this is good or bad and whether this is W7 or W8.
And that is a pitty.

Peter

[boleary]

This is in W8 and I mean to say that the sound is very, very good.

[manisandher]

But it is no virtue.
All you do is make your playback "PC" independent of any control. As if control weren't needed.

Yep, these things are difficult to talk about.

I'm all for control, if we have an idea of exactly what we're controlling. I mean, can anyone tell me why changing the SFS changes sound the way it does? I mean, universally we find that a lower SFS sounds 'brighter' and more 'airy' and a higher SFS sounds 'richer', 'thicker' and more 'dense'. This is pretty much consistent across the board.  Does anyone (you Peter?) actually understand the mechanism driving the sound here? There's obviously a 'sweet spot' and maybe a simple Linux installation hits it first time?

The other part of the "issue" is that you now blindly assume Linux is any good. Of course, it is actually the same topic as in the previous paragraph, but it really is not (good) at all. And so ...

I'm pretty ambivalent towards the exact OS being used. My initial search was actually driven by trying to eradicate all switch-mode power supplies from my dedicated mains ring. I tried an Atom PC with a linear PS, but it wasn't a good fit with XX. I then came across the NAA concept. I could use a high-powered PC on the house ring connected (with galvanic isolation) to the NAA, which itself was powered by a high-quality linear PS. The biggest downside I found was not being able to use XX or the NOS1. I have a lot of respect for Miska, but I prefer XX over HQPlayer for a number of reasons. And as for the DAC, I simply refuse to use anything other than the NOS1 in my main system.

As for Linux, I have no idea whether it's better or worse sounding than W7/W8. But remember, I'm not talking about using a Linux software player, but rather using a Linux buffer.

But if I look hard I can quote text from you where you explicitly state that this NAA doesn't cut it either.

Yeah true. When I first tried Miska's NAA, I could detect a difference in sound between the music being stored and played back from the main PC's internal HDD vs. a NAS drive. BUT... the difference was much less than say changing the SFS setting in XX. And also, I can't remember how I had things configured. It may well have been that I had something (the LAN switch, the NAS, etc) sub-optimally set up. In any event, I intend to look into all this again...

Mani

[PeterSt]

Hey Mani,

universally we find that a lower SFS sounds 'brighter' and more 'airy' and a higher SFS sounds 'richer', 'thicker' and more 'dense'. This is pretty much consistent across the board.  Does anyone (you Peter?) actually understand the mechanism driving the sound here?

Yes, I do. It took "years" but I do. The point here is that two forces act at the same time. And you know what ? the one which tells me that the lower isn't the best is exactly why W8 should be better. This is not SFS related, but the same force at play.
And at least in my system I am wrong about that too, just as that in W7 that force tells me to use the higher while I don't prefer that (which choice also took years).
So conclusion ?
... that this one force at play is a confusing one, at least to me. Only - and only if we receive too few noise with W8 and it is to be covered for by more noise before it sounds good, I understand this "force". But I refuse to believe that, no matter it is my own subject. It is the weakest excuse for not knowing really.

BUT... the difference was much less than say changing the SFS setting in XX.

So what ? Just find back the last XXHighEnd version which did not allow to change the SFS. It just was a setting chosen by me, like such a setting exists in any player, up to the ones you use in Linux (see below). And ... what do you think ... will that have the best SFS setting I had chosen by pure accident ? Or doesn't it matter in the first place ?
So there you go. You *want* that dial. Only when I had chosen SFS=2 by accident and it works for everybody the best or else they must buy another Playback drive ot whatever, you can be correct.
As I said, you want that control. Or better : we need it. Only when ALL has been sorted out we can do without, but still there will be a programmed one. It is not a trick you know, it is a necessary "buffer".

But remember, I'm not talking about using a Linux software player, but rather using a Linux buffer.

Sorry, but no. There is a small piece of "player" in there. Same as the piece of player in XXEngine3.exe which is only a few lines of code. Without that, no sound. Not here not there not anywhere. Not in Linux either.
More differences exist than you can imagine. Or actually you can. The difference between Attended and Unattended actually. What is the difference ? a program with a form which is there or not. Don't shut off anything and that is the only difference and you will hear it. Put your mouse over the form (just let it rest) and there is a difference.
Your "Miska" Linux application is no different from how XXEngine3 plays when it plays. Nothing can change it anymore once it plays. Two differences with that Linux solution :
- The OS;
- The means to control SQ.
And a more or less third :
Linux will be more lean. What comes from that ? God knows. Windows 8 sound, theoretically.

But my theories, and they can be so wrong ...

Peter


PS: There is still nothing much wrong with your whole idea about it. As I said, wait ...

[manisandher]

Linux will be more lean. What comes from that ? God knows. Windows 8 sound, theoretically.

Haha, as soon as I had the NAA/Mytek setup in my main system yesterday I thought how similar it sounded to W8 in minimize mode - impressive and annoying at the same time.

On a total aside, a few remarks about the Mytek. When I first got it, I thought it was really very good for the price. I'm going to revise this (probably not a good idea as I'll probably want to sell it soon). It's about right for its price. Nothing at all special about it. You'll read people raving about its DSD performance. I have to say that I think DSD is sh*t. There is something seriously wrong with it. I mean, it's 'dead'. All the life has been sucked out of the sound. I tried so hard to like it yesterday. I tried some old pure DSD recordings from DMP and some new recordings from Blue Coast... and just had a headache by the end of the session - I'm serious, and I'm the sort of person who just never gets headaches. I think my brain is being asked to work much harder when I listen to DSD. It's as if there's a blanket over the sound, and my brain has to fill in the information that has been lost or recreated badly. And I don't say this lightly. I've recently invested in two DSD recorders with the intention of recording my rare/valuable vinyl in DSD64 and DSD128. But there isn't a cat's chance in hell that DSD will be able to capture the sonics available on my vinyl. Pure PCM at 24/192 gets so so close...

PS: There is still nothing much wrong with your whole idea about it. As I said, wait ...

Hmmmmm...

Mani.

[Scroobius]

Just catching up on this thread and I noticed Peter's comments about how I may (or may not) appreciate the neutrality of SQ. At least I think that is what Peter was referring to (apologies if I have got this wrong). Well I have to say that "neutrality" to me is not a very important aspect of sound reproduction (OK I have to put some big big caveats on this) but basically I am saying that in terms of musical enjoyment I can get used to "colourations" provided they are within sensible limits.

What really stands out for me is distortion and inaccuracy. So when I listen to W7 now and I just cannot hear the individual strands of music and it sounds "blurred" and I then hear W8 and I hear with crystal clear accuracy the individual strands of the of music and also I can hear better sound stage etc etc I just "know" W8 is better SQ. But that is not to say there are not other problems. I just heard an Ella Fitzgerald remaster and wow did it have a sharp edge to it. So then I had to put on a recording I know is top drawer. And Oh Boy did it sound good. So now, it sounds to me, that we enter a whole new dimension of getting everything else in our systems to be able to accommodate what XX with W8 is capable of. That is how it sounds to my ears.



P

[Nick]

I'v been watching the debate take shape and decided to spend some time playing on Windows 7 today after 3 weeks or so of using Windows 8. I have to say straight away I have been extremely impressed with Windows 8 but it's interesting to step back to Windows 7 to compare.
So thoughts on Windows 7 and Windows 8 (based on 9-z08-1a and low latency settings for both OS's eg SFS 2 Q=1 etc)

Windows 7

Mid range is where the magic is (and I mean magic),there is a totally believable presence not over etched just very real, voices and acoustic instruments are set in beautiful relief, there is a beautiful balance of speed, tone, details. The sound is absolutely in the moment of the performance.
Highs have a greater degree of sweetness than Windows 8, I was suppressed how much so, BUT Highs also have distortions, a slight sibilance and harshness that act to break the spell by just reminding you this music is being reproduced each time the music hits a frequency that excites these faults.  Having said this I just get the sense of the music flowing from each instant in time to the next which is a very good thing, there is a sense that anything could happen next because the musicians could DO anything next (even through it a recording), this is exciting to listen to.

11/02/13 EDIT I need to tone down this comment on sibilance and top end distortion it is there but to a much lesser extent than I had thought. I found that something I was doing internally within my NOS1 was causing most of this. It's fixed and whilst there is a small amount of top end distortion left its nowhere near as pronounced whilst using Windows 8 now.


Bass snap and tonal resolution get a 7/10 which is ok but not top draw. Weight in Windows 7 is about the same as Windows 8 but it’s less distinct tonally and so not a tuneful as Windows 8.

Overall Windows 7 is great for less dense acoustic music, but can lose an amount of control on dense complex music like rock and dance, partially when highs that excite the sibilance are in the mix. When this happens the music sounds slightly confused at the top and a layer of detail is lost.

Overall impression of Windows 7: very real midrange presence, a sense of realness and live performance that draw you in and make me want more. I just feel myself wishing for better Bass, the loss of the sibilance and that last step of control on complex passages with HF energy.

Windows 8

Gives a very balanced spectral performance from low to high frequency, consistent solid, detailed, with a rock solid stage. Bass is better controlled with greater tonal differentiation of notes, mid range is very smooth and sounds natural but a notch or two less presence and “realness” compared to W7. Highs are smooth without the slight sibilance of W7 BUT  they do not have the sweetness or the weight of the Windows 7 highs, this is a small but important step backwards. For a brand new OS without XX tweaking it’s a brilliant start (really brilliant) but it sounds like it needs the magic that Min OS and XTweaks bought in W7 in order to just give it that kick of realness in the midrange and sweeten the highs. Now that would really be something.

Right now I could see me booting into different OS versions to suit the music that is to be played, IMHO though W8 is many times better than W7 when it was first released and Windows 8 appears to be extremely promising if Peter can work his magic in a way to would tune it as effectively as with Windows 7.  If the midrange life and top end sweetness of Windows 7 could be bought forward in to the Windows 8 sound that would be a sound I could really live with, long term. Fingers crossed.

Nick.

Ps no comments on sound staging in the above comments, my listening environment disqualifies my from passing comment here (Paul will know why, haha ).

[Scroobius]

W8 anomaly or not?. I had a bit of a shock earlier today. So far I had not heard any problems with imaging that Peter and others had reported with W8 (i.e. strange phasing effects). But earlier today I was listening to an old mono recording and oh boy did it sound different under W8 compared with W7. Under W8 is sounded like there were odd phase effects in the room. But then on second listening it sounded as though the room was evenly pressurised (for the first time) there were no holes in the sound as I walked around my room ( and maybe that is what sounded strange). And then listening further it seemed that the images were for the first time pretty much all in the middle between the speakers. Under W7 sounds at different frequencies appeared in different places in the sound stage which actually was not an unpleasant effect (I am not that bothered about imaging). But what is very interesting is that there should be such a big difference between W7 and W8. I have to say that I am still puzzled as to why there should be such a big difference between W8 and W7.

I would be really interested to hear the views of others about mono recordings played back under W7 and W8 (in other words am I losing it?)

P

[Jud]

And so the fun :

I have put a track to FileMail for the first 10 of you who are serious about this and want to compare. Do notice that you have to be able to compare and so you need W8 and W7 on a dual boot or other switchable means. You'd have to announce yourself in this topic for it and I will send it to the email address as listed in the forum, unless you denote otherwise.
Then I like you to report about the differences you hear. "W8 sounds better !" will not do. You really must try to pull out the elements like I tend to do it. This is also meant to be a sort of self-learning exercise; that you not net choose for the best, knowing from me that any disturbance should be rejection. It could mean rejection of both W7 and W8, but then the why is important.

All I can give you as a sort of placebo is that to me this sounds the most close to live drums and I can add that the other day I was in the hall hearing my son taking his daily drum lesson, thinking that we are still so so far off. But this was with W8 playing at the same time, and I never thought we were so far off. We are, but it didn't occur to me so explicitly. Placebo ? actually not much. It was with another album obviously, and I myself never listened to W8 with this one / track. So it is all up to you.

The goal ? well, if people are able to point out how this track sounds better on a majority of elements on W8, there will be better motivation to continue seeking for how to get there.
Don't forget to play it at your normal loud levels, so you won't be able to think that dynamics are not hurting you just because you play it softly; it has to work all the way.

Peter

Well, that took all of about 5 seconds.  Yes, Peter, as always you're correct.  All settings in Win7 and Win8 were identical in my listening and Windows 7 sounds a ton better.  Don't know what this means from your point of view as the developer, but I can tell you what I am hearing as a listener, at least on my rig.  And "on my rig" should not be minimized, because of course this is a much different DAC we are talking about, to say nothing of the rest of the system.  Still, it was very obvious, literally 5 seconds, maybe even less.

First what I heard, then why I think I heard it:

On Win7 I heard a band playing in furious concentration together with each other, led by the drummer's stick work on the ride cymbal, which gives the whole piece its energy.  And beauty and subtlety in the playing to go with this energy.  You could "see" the drummer behind his kit playing the toms and cymbals, "see" the timing of his arms and legs as he moved from one to the other so, so fast, yet always driving the other players and the piece forward with that ride cymbal.  The whole thing had a kind of Arabic adventure flavor; you could kind of see it as soundtrack for horses galloping at night across desert hills.  (Yes, definitely hills - in spite of the drive and consistency of that ride cymbal, there was subtle ebb and flow to the piece, not a simple uniform rhythm to the drums or the accompaniment.)

In Win8, there was drama aplenty in individual instrumental sounds - toms had a real whump to them, the sticks hit the cymbals, you could hear the separation between all the bass notes and really feel those notes.  But there was nothing between them.  The ride cymbal, completely central in Windows 7, was dominated by all the other percussion and instruments going on around it in Windows 8, too far in the background to drive anything forward.  Amazing to have all that air and all those good transients, yet no connective tissue to hold them together.  It is good you sent this track around, Peter, because I nearly always play test tracks with relatively few players and uncomplicated vocal and instrumental parts, so I was not missing this connective tissue, this interaction between players, like I should have been.

There was no subtlety in Win8, no ebbs and flows, no horses riding in the night up and down hills, no sense of the drummer playing on this tom over here, then having to quickly reach over to strike the crash cymbal there.  I was truly sorry for the piece to end in Win7; in Win8, I listened for a minute or so, had more than enough to know what I needed to know to write this, and turned it the heck off.

So that's what I heard, now why?  First and foremost, no microdynamics in Win8.  It does some of the big stuff so well - big dynamics, air, transients - that you can hardly believe it, but it's true.  No subtlety, no band members playing with and against each other, no tiny shadings of rhythm, loudness, no changes of phrasing....  So no wonder you hear "panning."  First you hear drums over here, then you hear another part of the drum kit over there, but there's nothing in between - it's like teleportation, the drummer was one place and without any subtle change to let you know he's getting ready for it, all of a sudden he's in another place.  As I said, I could "see" the drummer in Win7 move around his drum kit, because all the little audible cues were there.  Not so in Win8.

Second, and this may all be part of the same basic problem, not something separate, under Win8 instruments don't seem to be able to play together with each other.  One sound at a time dominates (often a different instrument from instant to instant) and the others are in the background, there are no instrumental lines weaving around and through each other.  So again you get the sense of notes from various instruments suddenly appearing disconnectedly from nowhere, rather than being able to follow them from foreground to background and back again.  (Yes, as I read this, it seems that it probably is part of the same basic lack of microdynamics.  Either an instrument is in the foreground or you almost can't hear it, and even if it's in the foreground you don't get subtlety from it.)

Hope this helps you to perhaps see a way forward under Win8.

[listening]

I'm following the discussions and I wonder whether Windows 8 is introducing some active manipulation of the bitstream? Is it possible hat this big difference is caused by other operating sytem architecture?

Georg

[PeterSt]

So it's getting more and more odd ...

I had a bit of a shock earlier today. So far I had not heard any problems with imaging that Peter and others had reported with W8 (i.e. strange phasing effects). But earlier today I was listening to an old mono recording and oh boy did it sound different under W8 compared with W7. Under W8 is sounded like there were odd phase effects in the room. But then on second listening it sounded as though the room was evenly pressurised (for the first time) there were no holes in the sound as I walked around my room ( and maybe that is what sounded strange). And then listening further it seemed that the images were for the first time pretty much all in the middle between the speakers. Under W7 sounds at different frequencies appeared in different places in the sound stage which actually was not an unpleasant effect (I am not that bothered about imaging).

Emphasis are mine of course.

So, here is someone who judges mono. And mind you, I also receive emails from people who actually don't post here, claiming phase anoamlies and that I should observe mono perhaps. I never did that so far, but why not use Paul's great (or was it large ?) ears ...

Emphasis # 1
So, in stereo did this not occur, but only now, in mono, it does ? (and most seem to perceive "anomalies" to this regard)

Emphasis # 2
So, in stereo did this not occur, while one of the first thing I claimed was just this (measurable) ?

Emphasis # 3
No holes when mono ? But exactly 100% these holes in stereo, at least in my system ?

Emphasis # 4
Image finally in the middle with mono, while we all fairly much agree that stereo just draws everything apart while it should not ?

Emphasis # 5
Different frequencies jump around in the image ?
This is not something I wrote about in here but elsewhere I did (Jud will recognize it from attitude, lattitude and altitude). But I recognize this from stereo and W8.


Keep in mind guys, this is the report from the one and only man who perceives great satisfaction from W8, but now comes up with all the anomalies in one compressed post when playing mono.
If you look closely it is all the exact other way around. I can't even type the "formula" because all is twisted.

HOW ?

Eh, Paul, check your cables ?
I did and all was okay at playing the test tracks through W8.

And Paul, are you using those mystery feet ? Remember, if one strange phenomenon evenly pressurises the room it is that; Assumed W8 is wrong somewhere, possibly those feet create an adverse effect on it. Or "reversed" - I don't know.

Of course I'm still suggesting that W8 is wrong, while it just as well can be so that W7 and everything else is wrong. Now somehow you, Paul, created the prerequisite to let it work out as intended. Maybe eleborate somewhat on the ByBees you use ? maybe it gives a clue somewhere.

Peter

[Leo]

If the group of 10 testers is not complete yet I am ready and willing to participate

Leo

[PeterSt]

Also in response to Georg somewhat :

I will finally tell about something which could be important but I have been reluctant to tell about because I am not sure and it needs further "testing"; because of my idea "nah, can't be true anyway" I let it rest. Still I should tell about it. So here goes :

2 or 3 weeks back I hooked up the analyser actually only to check whether the phase difference between both channels still is 0.00 degrees under W8. Not more digits are available, but I reckoned that when I'd suddenly see 0.01 (or worse) it would tell enough. But it was 0.00. Too bad.

I don't know how it went, but somehow I left the analyser on and let the signal running;
The FFT analyses I am actually looking at, shows the THD+N and we test each NOS1 for that. It should say 0.00180% or a bit above that (like 0.00185). The analyser keeps track of minimum and maximum values received during the session.

To my big surprise, looking after 10 or so minutes, I saw the minimum values for both channels in the 0.00140 range. Said "what ?" and *because* I was doing this test to find W8 culprits anyway, I reset the min/max values and waited for it maybe to happen again (I could have made some mistake ?). Saw nothing strange.
Notice that this is a constantly "wobbling" figure while the "wobbling" is a measure within itself.
Waited until all recurring things had passed (like the test track which is 2 minutes and is repeated with A-B).
Watched for a few minutes and then gave up, but kept it running.

Maybe after 20 minutes I looked again, and you can guess it. Bingo.
Both channels had achieved a best THD+N of exactly 0.00140%.

The point is, knowing how the THD is calculated, it theoretically can happen that because of a "nice capture of the wave" the figure is better for some cycle. But I don't see that happening with e.g. 10 minutes of needing to wait for that, while the wave itself comprises of a handful of samples only (something like 40 for the 1000Hz signal, in the 16x Arc Prediction upsampling 16 x 40 = 640. And that passes 1000 times per second, thus 60000 times per minute, and then still a new combination of the "sampler window" can be there ? No).
And the precise 0.00140 for both channels also bothers me. Never saw the both channels so equal.

... but I am never watching at any test sequence for this ...

What I do explicitly watch is the amount of "wobble". So, with that 0.00180 as a target, we see it change from something like 0.00175 to 0.00195 and watch this for maybe a minute. If it stays in between that range, all is OK. Never it jumps down to e.g. 150 figures (with 16/44.1 btw). Never. And now it can go to 140 ?


So this is what I never told and also never proceeded upon, because I never looked at this minimum figure. We let it run for 30 minutes always (tracking for errors etc.) but don't watch that minimum. So maybe I should now for W7. I will.

But what if it never happens there while it consistently does with W8 ? Then something *has* to be going on, right ?

I'm following the discussions and I wonder whether Windows 8 is introducing some active manipulation of the bitstream? Is it possible hat this big difference is caused by other operating sytem architecture?

Georg

[PeterSt]

Btw, funny. Someone resurrected a topic from 2008 today, and this is what I wrote  in there :

Okay, another thing; Of course when playing loud the room gets more filled with music than playing softly. But here again, I feel some other thing is happening now : this filling of the room gets done in a way that it doesn't matter anymore where the sound is coming from.
What am I saying ?
The combination of the ever so important "no problems at all with standing waves" with this loud "sharp" level of playback, makes that it becomes rather impossible to focus on the source the music comes from (yep, the speakers). I know how to listen for reflections (just step into the middle of the speakers, and face the opposite of the room), and it is not that doing it. Instead, far more seems to happen in mid air;
Think of those now very "sharp" waves interacting far more in air with eachother, and where they collide they produce extra energy.
Focusing of instruments stays, but you are in the middle of them now. Or more in the middle. Walk through the room (12 x 8 x 3 meters here) and you'll meet them underway. Ehh, yes, 2 speakers here.

This started to happen in 2008, never went away not even with NOS1's around the block, and it is precisely this which does not work with Windows 8 anymore ...