SQ of 1.186, best ever! (Again)

[acg]

Hi Peter,

Thanks for this report.  I've never listened to w7 and XXHE but I intend to do that in the next week or so, but I have to be honest, based on your description I am not sure that I will enjoy it.  We will see.

The one thing that is making the Silverstone card enjoyable for me (with w8) at the moment is the LPS and cable loom setup from the other thread, otherwise I would not be using it because of a tendency toward harshness (a better description is noise I think) but it is so good with the separation of notes et cetera.  Were it not for the cable loom I would have my PPA card with Coens mods in my XXHE pc.

Anyway, I urge you to experiment with cleaner power into the Silverstone because I am most interested with what you would report.

Regards,

Anthony

[CoenP]

Peter,

I am glad you put up the description of the difference between W7 and W8. I think this very much describes the annoyances that I have been hearing in W8 and the naturalness I experience in W7 (...since the beginning). This despite the quite different systems we listen to!

Still some of us report splendid sound from W8 which may tell that there could be something we overlooked.

regards, Coen

[PeterSt]

Hi Coen,

I too can put forward splendid reports of W8. But that first needs to NOT play xxx albums. Notice that this already could be subjective compared to others who just don't play the perceived baddies. Well, I got used to that too for the past year. Not the biggest problem. One thing though : all should play since I'm from the camp that just wants all to play.

So back to a previous post : what does it tell when I play music for 11 days and all those 11 days in a row I really never have been annoyed, and what does it tell that a very first note (that flute) I hear through W8 already annoys me. So not the 25th album but the first note of the first track of the first album. And true, I learned to not play that album anymore. But this is stupid.

I'm also from the camp that won't dig fake detail. Well, I think I was as far as not finding the detail from W8 fake, but now something else turns again that or me, and this is it being too lean. So how easy it is to create more detail by means of removing the bass. Just try it. And *this* only occurs to you when going back to W7.

What's totally new for me is the enormous detail the mid can bring and although I thought this was from my three-folded floating setup I only *now* see it is again W7. Just go back to W8 and you will find lacking *that*.

The real main culprit is - or has been - that whenever I went back to W7 the past year it sounded distorted. Too rough in comparison to W8. And since *that* is now out of the way (most probably the Q3,4,5 = 1,1,1 from 1.186), well ...

In the net end I see no value at all in more detail W8 can bring because that same detail is "over" for many albums and without knowing you won't miss it, while the enormous mid-detail of W7 not only compensates for that vastly, it brings thus detail at a level which is not fake. This again (yes Anthony) is that stupid Silverstone card so things really *are* complicated. And that meanwhile the whole picture is not so overly lean because of more bass (for the better or not) is a giant bonus. Remember me telling about "music for pussies" a couple of times; that's W8; there's just not enough weight. It's for people drinking tea lifting their pinky. I told about that too. Not that I make the laws here but I do try to not fall into pitfalls (of e.g. detail) while that is so easy to do.

Let me say it once again but with different emphasis :
Nobody that I know complained about W7 back at the time (year ago and beyond). And really really the ONLY reason I never went back was because of the distortion audible when compared to W8. For me there has been NO other reason. And now this distortion is out of the way I thus see no reason to not go back.
So will I ?

That depends;
I think it can count for you too Coen - but over the years I have tuned my speakers (but think mid-high horn) NOT for W8. It was for W7 (and Vista) and you don't want to know how easy I can change the whole presentation by means of changing a resistor here or there. Still, this is all within the bandwidth of (say) W7;

What I'm trying to say is that when all you guys who just use consistently setup speakers come to the same conclusion (go back to W7) then I will too because then the culprit is not my W7 tuning. But when all of you compared and still clearly decide for W8 then I better look into my tuning of the speakers, right ?
I can also say that all those who tuned their speakers audibly (but hopefully with some base sense of measuring) might just as well in that same boat of not being able to like W8.

What a hobby.
Peter

[AlainGr]

Thanks again Peter for these reports. It really does give me the desire to retry W7... I don't feel the urge to follow MS in their aim for version changes...

Alain

[CoenP]

Peter,

Thanks for the elaboration.

I've grown to believe in a "sound character" of components and software, that is that no matter what you tune you can never tune it completely into something else. Eventually you will always hear a certain fingerprint of it on the sound.

To illustrate this with an example, I've used a simple lab supply on various locations in my audio setup and I've always been hearing a consistent "darkening" of the sound. For this the supply has to be in use in the chain, just being on unconnected will not change anything.

So we might have tuned our speakers for Win7, but I am reluctant to accept this as the root for a win7 preference. I've conducted many experiments with my speakers filters and they've always been in a certain margin, affecting only a limited set of SQ parameters (foremost on the tonal balance and spaciousness). Especially I've found the almost subliminal listening ease of win7 not a quality that I can change with speakerparts.

Off course if anyone has an experiment that can or will prove me wrong I will conduct it!

regards, Coen

[PeterSt]

Hi there Coen,

I've grown to believe in a "sound character" of components and software, that is that no matter what you tune you can never tune it completely into something else. Eventually you will always hear a certain fingerprint of it on the sound.

Of course this is true, generally spoken.

To illustrate this with an example, I've used a simple lab supply on various locations in my audio setup and I've always been hearing a consistent "darkening" of the sound. For this the supply has to be in use in the chain, just being on unconnected will not change anything.

Even with that as an example.
Or maybe it's the best example of how to NOT judge things, because it is so logical that it happens. Or to be more cautious : that it can be so logical because we will only know it all when really all has been sorted out. And we probably never can. What we sure can is making a 1000 mistakes in our thinking and even judging ...

So back to the first quote, yes, for e.g. an amplifier this is quite obvious. I mean, for empirical finding/judging. But look at the noise line and that already tells all. So if that isn't ruler flat it will tell about some character of the sound. How or what is hard to predict, but that there's "something" is obvious to me when that ruler flat is not in order. Now show me your plots ...

Of course it is more complicated, because when no ruler flat is shown the first suspect is you yourself with all your further misery connections. But how can you tell without a reference ? I mean, if I see something creeping up I just know it is me somewhere, but if you see something like that you may think it is the amplifier. And for 99% sure it even is. There's just no where to go.
And thus your components show a character, because they just do. It is really that simple.

If your lab supply shows a darker sound where ever you use it in the chain, it really is not about that supply causing that character; it really does/implies something, but what it is is to be found out. But it really will not be different from somone coming up with hi-hip-hooray brand USB cable, which comes with a topology which doesn't do a thing to the signal, but which *does* imply another ground path. That is, all cases sorted out so far came down to that.
And so, easy : when your lab supply always brings that same character, you are not going to tell me that it always brings the very same virtue to whatever component you feed with it. That just does not exist. But that the supply always injects the same whatever sh*t to your mains which next causes always the same nature to the whole chain or particular components in it ... I'd say that's the logical explanation.
Super simple real life example : have a poor power supply in the unit that throws DC Offset to the mains. It really exists you know.

A more honest example will be software, because once we accept it that this influences, something like XXHighEnd has been set up from some base and that base will never or hardly change. It already starts with the computer language it was written in, or the compiler used to make machine code of it.

With this all said, no, a PC will not give any character. Or wait, it always will; it just depends how you look at things, but we already know it always comes down to the noise phenomenon. This now is different from the amplifier's noise line and I don't think it needs further elaboration of how it works, as long as we see that it's actually not much under our control (yet) and that something like a Silverstone USB3 PCIe interface works out for the better with sheer analogue merits (I mean, we judge this like an amplifier which is NOT the case at all because it's just a stupid digital thing).

It will take some more time to dig out what's really happening, but I expect it to be very simple like the one component (take the USB3 card as an example) always operating at a rate of 100MHz while the other does that at 5Mhz when no higher speed is demanded without me saying which of the both works out for the better. But electrical differences like these *will* be in order and it is only that we don't know how they impact and where (but start with radiation).
Still we better isolate the DAC 100% so all can't be of influence anymore, but that's a bit impractical. Isolating the interface is impossible because that incurs for more jitter again, and running the DAC from batteries is a must because whatever is thrown it will be thrown on the mains so that's not allowed to come in as well. But batteries aren't for the better at all, so ...

[PeterSt]

Hi again Coen,

So we might have tuned our speakers for Win7, but I am reluctant to accept this as the root for a win7 preference. I've conducted many experiments with my speakers filters and they've always been in a certain margin, affecting only a limited set of SQ parameters (foremost on the tonal balance and spaciousness). Especially I've found the almost subliminal listening ease of win7 not a quality that I can change with speakerparts.

No ? then I will try to turn all upside down. Maybe I will fail, but let's see :

Remember that one ?
Ok, not to make it too complicated, but what I really did was applying these curves to Windows 8. I mean, "really" is opposite to what I told yesterday where I suggested that I had my speakers tuned for W7. True for the old ones, but not true for the new ones.

Try to envision that I can change the general slope you see from left to right, that I can change the deepness of the dip you see starting at around 1000Hz up to ~5000Hz and that the slope to the right of that can be changed as well. This all with preservation of (good working) XOvers.
Look at the curve under the 90dB line;
Now also envision that what you see at that curve is one end of the extreme I can dial in. It's actually just that F-M curve. The other end is the complete flat line like how a speaker will be nornally "tuned".

Remember, I am bringing all upside down so it will not be clear to you yet what I'm heading for ...

What W8 requires is and the slope and the dip. So bring either - or both together - up with something like 1dB (!!) and you will die. The tweeter side of the upward slope (say beyond 6KHz) is more forgiving because it is merely used to fill up frequency gaps that emerge in the lower frequencies because of not enough energy in the highs. What bringing up that "tweeter slope" does is a kind of opposite of what the Sillverstone does (Silverstone separates, higher tweeter slope fills out). Of course the both interact and once you believe in this (or heard what the Silverstone does) you already can't say this is not important.
Don't forget about this 1dB difference, because it seems nothing but changes W8 from listenable to unbearable. Ok ?

With for W8 the slope and dip in order, the tweeter is to be set to say 5dB lower than you see in at the 90dB curve. Make it lower and the sound gets too rough (holes are not sufficiently filled out) and make it higher and too much hissing will be your part. All depends to some extend on the material played but say that since 1.186 (which I of course had for 2 months or so longer than you all) I could live with one setting and I did not change it anymore. Prior to 1.186 (or actually the Q5=1 setting of 0.9z-9b) I daily changed, depending on what I found to sound too nasty.
The mere important message : I settled down on just one never changing setting for maybe 3 months by now. So, all (finally) OK, and what was OK was the software.

In the mean time it is allowed for you all to wonder whether it can be any good to have a response from the speaker like this. On the other hand don't underestimate what 115dB sensitivity does to the freshness (actually speed) of the sound and how briging that back to only 112dB already makes all sound muffled and nothing will sound harsh anymore.
So get it ? all is super relative to much more than software and now I'm saying that the sensitivity already makes all moot for comparison. All right ...

What I didn't tell in yesterday's report is that the last album I played was The Wall, after going back to W7. It even seemed unrelated to myself what happened and what I did and it was beyond the ABX thing. But, what happened was far more important than my little ABX game ...

Remember what I said earlier on : "We might have tuned our speakers to W7".
Yea yea, nice talk PeterSt, but what you should have said is that you tuned your speaker to W8.

So what's following now I already knew yesterday when I wrote out that ABX report, but the real merits I only know from last night. So ? ...

Being confronted with all the deliberate high frequency "test tones" from the ABX game, turning back to W7 and playing The Wall right away brought be a too muffled sound. Completely blanketed. Of course part of it is relative, also with the notice that I played with W7 for that 11 days, but still. This is not how The Wall is supposed to sound. Btw, MFSL version (not 30HZ high passed for LP).
So about turning things up side down : I never ever touched those F-M curves for W7 ...

So what I did was bringing up the slope with 2dB, the dip going along with that linearly (so dip remains as deep) and brought up the tweeter slope to max (see 90dB curve).
All sounded superb now and I was in lack of nothing. It slipped my mind how the hack this had been recorded so well, and the boooom - boooom - boooom of the first track was perceived at sub low in a giant space. Never heard that, but alas.
This in itself is totally unrelated;

What I deliberately did was not changing back the slopes when the album was finished and dinner was ready (when the music always stops here). I sincerely was hoping that the next day I would have forgotten about it to next see whether I noticed, but now with the more "test tone" material.

So I did !!

I will now stop with making font sizes 45 points etc., but I once again was totally floored at the first notes coming from the speakers yesterday. Never ever EVER I could have imagined that what I now heard was possible;
The squares in the mid range now scatter to the ceiling and float along the sealing towards me. Well, I never heard something scattering to the ceiling so I guess that suddenly things are so sharply boundaried that it can happen ?
Oh, for the "square" I'm talking about, think a seagull. Exactly that, and really exactly that because it *was* (a.o.) a seagull in the music (and now even see the beast flying towards you right under the ceiling).
Side note : I recall from W8 that this seagull has a beak, while in W7 it also has a throat (about lacking fundament in W8).

So, all scatters and flies and bounces and what not. The detail is crazy and what seems odd is that the squareness is not coming across as sharp or hard at the same time.

After being on the floor when music first started I of course soon realized that I deliberately had let those curves where they should not be, so it was now my task to see whether I could squeeze out the same detail as W8 can do it. But now the high frequency detail of course. Strangely enough (or I had the unconscious hunch) I started a tracj which has been in my demo Gallery for ages but which I never played anymore because I couldn't see any "demo" merit of it, which of course was from the W8 era. It is a kind of "house" thing with a not too fast beat of course lasting 12 minutes or so. So what actually to do with that. But now ?
Djeezzz. One big interacting left/right everywhere flow of scatters running into each other with always changing effects throughout the 12 minutes with, yes, that boom-boom-boom slower beat throughout. And no, I can not imagine how that ever could have been a demo track without this scattering sheer joy.

On this matter I don't think that anyone realizes what "mid detail" is, and maybe it's not a real phenomenon in audio so far; it can be "representation" or "forward" or "tight" and a few more things, but detailed ?

All 'n all I yesterday heard the most detail everywhere (also the highs) by very very far;
Nothing went wrong but 3 hours and a handful of albums may be too few to really judge, but there is no single way W8 can do this; It would choke in itself first (so to speak).

So Coen (but everybody), never think that a speaker doesn't matter. Or like in this case the "tuning" of it;
You just don't know - and that really is all. It's not bad or stupid or whatever ... you just don't know. And before it's still misunderstood : you don't know the reference. And undoubtedly I don't know that either. But at least I have my own and I just described the enormous changes possible by means of 1-2dB differences "somewhere";
You can say that *now* this speaker is tuned for W7 - you will squeeze your arm a couple of times to be sure you are awake when you hear it - and if I give that to W8 you will be asking for ear plugs (which I have just in case).

Regards,
Peter

[christoffe]

So here we go :

Booted into the W8 partition of the dual boot on the XXHighEnd PC. Picked an album of Ott deliberartely because it has been typicle for not wanting to work out on W8. Maybe that's mean, but hey, on W7 it works (though I only tried other albums from Ott - not this one).

First impression : highs highs highs.
Second impression : highs highs highs.
Hmm ...

Hi Peter,

My computer system was not changed for appr. 3 years and I did not change much of the settings with the different releases. The RAM disk (VSuite Ramdisk Server edition) and the USB3 card (with a NEC chipset) are not favouring the SQ on my system.
After a two day session with W7 again, the clear winner is after all W8.
With W7 the system is losing much of the room image (3D sound stage) and less bass and highs.
The drums/cymbals have on W8 a very life like tonal balance, what I prefer.
A good test record for the 3D image is :
http://www.amazon.com/Tenderness-Al-Jarreau/dp/B000002MMU/ref=sr_1_12_title_1_aud?s=music&ie=UTF8&qid=1391768033&sr=1-12&keywords=al+jarreau

Tonal balance of drums, title 2:
http://www.amazon.com/Jazz-Garden-Stanley-Clarke/dp/B001VFM0QU/ref=sr_1_8?s=music&ie=UTF8&qid=1391768160&sr=1-8&keywords=stanley+clarke

The release 1.186 is magical with W8 on my system.

Joachim

[PeterSt]

Thank you for that feedback Joachim.
I am not sure whether I have those albums, but if so I will try them and report what I perceive from them.

Regards,
Peter

[CoenP]

Hi Peter,

Thanks for sharing you thoughts and insights. I've not considered speaker tuning in this way myself. Anyway maybe I did some changes crucial to my peception of the music through the different OSses. The speaker filter setup in use has not changed since my xp days, though I've experimented a lot inbetween. Foremost I removed some damping material and changed the remaining with stuff that suited me better. This could have been beneficial for w7 only. Someday I will look into how to realise your f-m approach in my setup and have seagulls floating onbthe ceiling!

Regards, Coen

[PeterSt]

Haha

[acg]

Peter, which The Wall track has the seagulls?  I'll track it down tomorrow and see if I need to duck while listening to it....hehe.

Anthony

[PeterSt]

Anthony ... None. That's from another album and artist.

http://www.amazon.com/Gaia-Sophia-Entheogenic/dp/B005C8SKOS
Track 02.


PS: You can already see how such an album sounds. Really so.

[PeterSt]

Dear Joachim,

A good test record for the 3D image is :
http://www.amazon.com/Tenderness-Al-Jarreau/dp/B000002MMU/ref=sr_1_12_title_1_aud?s=music&ie=UTF8&qid=1391768033&sr=1-12&keywords=al+jarreau

Tonal balance of drums, title 2:
http://www.amazon.com/Jazz-Garden-Stanley-Clarke/dp/B001VFM0QU/ref=sr_1_8?s=music&ie=UTF8&qid=1391768160&sr=1-8&keywords=stanley+clarke

As you may know I am always in for such little "games" and I really did it (it appeared I have both albums);
What you also may know from me is that when I am challenged by examples, I will not hesitate to vent my judgement. For the better or the worse. So example : people tend to come up with super Hires such and so, give an example, I listen - debunk it for whatever I *hear* to next support that my whatever graphs and show it is stupidly upsampled or whatever has been the matter. Didn't happen often in this forum (because the challenging doesn't happen often here) but elsewhere yes. So hey, if you tell me that something is good, it better be good before others judge it as good while it maybe is not. Understand ?
It is and remains my little game though, so don't feel personally attacked only because it was you who accidentally came up with it. It's just to learn from, well, hopefully !


So, Al Jarreau;

Of course, if this was my challenge to squeeze out more 3D than what it is actually not, then OK. But I'm afraid that this should have been the example of "sheer 3D". Well, then you just don't know OR my system totally fails;
It was the flattest album of the whole evening (including Led Zeppeling I from 1969), with the notice I played a few tracks of it only. Just too poor. What it *did* do though is making me part of the audience. Not sure whether that's the same.
Oh, rendition of drums was fairly much OK to my judgement. See below.

Stanley Clarke;

Tonal balance of drums, title 2:

Yes, that is what you said, and I assume you referred to track 2.

Drums ? Ok, if drums are to be cymbals then I can get what you are referring to. But the little bit of "drums" as such in there are completely overwhelmed with the poor (!) recording of the cymbals. Now, I suppose you can't imagine, but no dynamics are in those things at all;
I must assume that you perceive the cymbals fairly loud (and not softer than the few drum hits in there), and all I can think of this is microphones attached to the blades of the cymbals themselves (so to speak) which won't allow the hits on the cymbals themselves to develop. Maybe hard to explain, but a cymbal first needs a firm hit (which you will clearly perceive) before it starts to sing. Well, singing is there all right, but hitting - nothing.

So in my book this is THE example of a most poor recording, which even makes sound the cymbals to the Chinese side if you know what I mean (more dirty sounding).
So just saying : Al Jarreau's drums are way better.

Now, I am not here to debunk things but merely to try to dig out what's really happening. So let's not forget the real subject, which is something like W7 vs. W8 and how things may not work out in W8 which do in W7. Of course this needs incorporation of a few other posts of mine elsewhere, but at least you will have read those. So :

Let us first assume that you are not deaf, you know a few things about this all, and that you have all in balance. Clear ? So what I'm saying is that you perceive all as OK and that I am explicitly NOT saying that you are deaf or anything. Thus all OK on your side, and thus all wrong on mine ?
So that's to be sorted out. Or at least that would be my interest. Maybe not yours ...

Now next thing we must assume (or at least I do that) is that my high frequency output is way higher than yours. This could be why I perceive those Clarke cymbals as completely upfront and studio taken which should tell that they just are, no matter you perceive them differently.
With this as a base, try to envision that I now play this on Windows 8 with my clear perception and expression that way too much highs come from W8 and all is very lean to begin with. Thus, while I play from W7 and judge that to be OK in general, you play from W8 and judge that to be OK in general, but I play your cymbals on W7 and will 100% know how crazily wrong that will go on W8. And this time not because of the transients (they were not there) but because of the too lean output there (hence too much highs).

So what's up ?

Well, to start with it should be about the material played.
What I can find for a reason is that your Jarreau brings you a lot of 3D which is relative to the lesser. And why not. But still Led Zeppelin I should show more depth than Jarreau at your place. Not that you are going to try *that* ...

When I finished the little listening I started out with my usual sh*t and my first thought was "Joachim, you just don't know". So after the first two tracks of that I thought to send those two to you. And I will if you want (so please tell me).
To indicate the difference (if possible through words) I'd say a 100 times more dynamic, and something like 12dB more high frequency output. And yes, I thought to show you a spectrum, which would only show how relatively few highs come from your examples which would also count for Clarke's cymbals if you only count in the lacking transients (so on estimate it will show a large pile of 1500-3000Hz but that's is).
But this is of no use without listening (first).

The first two tracks I played (and prepared for you) can be called "seagull like" and similar sounds are in there. But also, both tracks contain a lot of vinyl (deliberate) ticks which are the highest transients, and possibly you won't hear those at all. Possibly you won't hear anything of what I hear, because it just needs the super speed in the speakers, and what I get from your examples is that this may not be there to begin with. And now you won't be bothered by too much highs or too transient music in W8 ... understand ? I mean, there has to be a reason, still assumed you are not deaf.

If these two tracks annoy you or of these two tracks don't have your interest throughout (one is 8 minutes, the other 12) something is seriously "wrong". Wrong between large quotes because nothing wrong as such, but it possibly determines the real merits of W8. So to be on the constructive side : If you perceive scattering and hurting highs only (on W8 !) then all what it tells is that you don't play high transient music and find Clarke to be OK on the cymbals because at least that shows cymbals to some extend. But if you listen to these tracks, then you know.

Before you say Yes, I can already point out a super danger : Both the tracks need "severe" foundation of the lower frequencies. I am not sure anymore but I recall especially the second track to show something like 30Hz very clear and if that is lacking, well, then we'd both listen to a W8 representation to begin with (too lean) and it will be unbearable because of that (now scattering highs only). Again IIRC the first track starts out with real sub low (like 25Hz background roar) and there too, without that nothing will work out. But what I'm really trying to say is that while this might not work out with my hints given and now because of speakers not going that low (I just don't know yours) with equal (and undistorted) SPL, it still could work out better on W7. All I certainly know is that the first track (this is Shpongle) is not playable on W8 for me because of lacking fundament and the whole track fails because of no interest now. The second track is more easy on this (this is Tau Kita) BUT here it is about the on/off separation in the lower range (voice range) where a voice morphed into cut on/off should bring a very pressing in-church (or in-hell ?) message (this is into the 6th minute or so). This is so high transient that a. Windows 8 will do that for sure, but b. Windows 7 might not again depending on the speaker. Here on W7 it still works, but the voice is more pressing because of the more hollowness coming from better fundament.
Blabla

If you judge either track as "total-sh*t" then something has to be wrong because of the representation. Anyway watch the depth compared to Jarreau.
What these tracks should do is encourage for 100s more of it, no matter it isn't your music; it just shows (here) what can be done with audio and no Clarke is going to show that. Frisell yes (just saying and try to see through his music and how it happens).

Best regards,
Peter


PS: If anyone else is inerested let me know here, but please find yourself obliged to post about it, never mind super negative or in agreement with me.
Tracks will be cut for legal reasons.

[AlainGr]

Peter,

Can you indicate the song from Shpongle that you are talking about ? I could get it on Qobuz and try to understand what you mean...

By the way, you scare me with the test music you use 



You are not a Klingon - are you ?

Alain