Vista users, again, prepare yourselves ...

[PeterSt]

Dear world of music lovers,

The upcoming 0.9u version(s), dedicated to say, bit manipulating like 24 bit playback from 16 bit versions, will have a high surpise for you all;

In my pursue of 24 bit playback, I found something that comes down to applying the same principles as a non oversampling DAC, but in quite another area. If my theories and my ears tell me right, it will be a revolution in playback of digital sources.

Ok, you might say I dare to challenge myself to anounce this kind of prematurely since I'm listening to various kind of music right know to prove my theories right, but let's say I want to be those theories right. You can always call me crazy afterwards.

I'd like to be this propriatary for the time being, but *if* I'm right it would be a kind of revolution. Could *I*, me, myself and I incur for that ?
This really is up to you.

I say : what I found surpasses the influence of jitter (by far).

Until I put up the first 0.9u version, I'd really like you to prepare yourselves and (with Engine #3) play particular tracks and pay extra attention to :

1. Clarity;
2. Spatiousness (not to be confused with fuzzyness);
3. Being there with the music without compromise;
4. Dynamic response;
5. How deep your bass goes;
6. The tinyness of PC playback (which often is a problem, depending on many things in the chain);
7. The balance between the loudness of cymbals and the other instruments;
8. The balance in general between frequencies, the possible unimportance of varying that balance (if you can at all) and standing waves (not only the lower frequencies, but also annoying unequalities of higher notes);
9. The SQ at various volume levels of listening.

Please note that any implied "negative" is to be hold against the recording as far as you can interpret that. Also note that a "good" recording not necessarily gets the best out of everything and "good recordings" often boost low frequencies and only sound good to our commercial ears. General boost ("loudness") can also give additional problems with 0.9u, I think. Also. I may imply only positives in any of the phenomena above, of which at least not all are positive (to my own judgement).
Anyway, since I cannot be 100% sure about the result being better than everything existing on this world so far, I like you to be as honest as possible. Ok, for now this means "preparing" at listening to the various phenomena I indicated, as long as it takes to get a decent idea of judging the upcoming 0.9u. I can promise you it will be so different that without decent preparation you will be confused and keep on A-B'ing without clues; All fits or it just does not. Either way I'd be very glad to know.

Before you (and I) forget : it is my advise to try 0.9u with Q1 at 14 as being a rather neutrally jitter influencing setting; At your first judging just leave it like that (14). Leave your soundcard buffer settings (if applicable) at where you had them.

Thank you !
Peter

[xp9433]

Dear world of music lovers,

In my pursue of 24 bit playback, I found something that comes down to applying the same principles as a non oversampling DAC, but in quite another area. If my theories and my ears tell me right, it will be a revolution in playback of digital sources.

Dear Peter

This sound both tantalising and highly exciting!
I assume it only applies to play back through Vista?

Frank

[PeterSt]

Yes Frank, it does. Engine #3 only. 

[Leo]

It will arrive shortly I presume ?


PLEASE !

[PeterSt]

Uhhmm ...

The upcoming 0.9u version(s), dedicated to say, bit manipulating like 24 bit playback from 16 bit versions

I hope I did not suggest that 24 bit playback would be available in the first 0.9u version. It certainly will be not.
As I indicated earlier and elsewhere, it appears that Vista Exclusive Mode just doesn't allow it. I've been again working on it over the weekend, to no avail on this moment. Now, the most sad thing happened :

I anticipate Vista/SP1 having solved this, although I must say that there is no real indication for this on the Net. So yesterday just before I created this topic, I got myself the link to the download of Vista/SP1 RC (Release Candidate). Then I wrote this topic, had some dinner, and sat down to download SP1 RC. To my surpise the download just had been removed (and I've been looking well over 2 hours to find other live links) and now it's waiting to half of March at least for SP1 RTM. Or one or two months longer depending on the language you need.
So there went my kind of last hope for the being nearby of 24 bit support.

Anyway, this topic wasn't about 24 bit support but about the "technology" I applied in the upcoming 0.9u-0. In order to get it up definitely I have to solve one relatively small problem, and I want to have another listen tonight first to be more sure. But if all is right it's worthwhile some waiting for sure.

Peter


PS: But if someone has a link to Vista/SP1 RC (not a beta !) for me I'd be very happy to download it and proceed from there for the 24 bit quest. Note that AFAIK it would not be illegal to provide such a link from e.g. your own server. There has been a December 17 (2007) version and a Januari 13 (or something like that) version. Either will do.

[GerardA]

I anticipate Vista/SP1 having solved this, although I must say that there is no real indication for this on the Net

Did you have contact with Microsoft about this bug? I'm afraid the soundsystem is not their highest priority, and if they don't know the problem ...

I checked for the SP1 update, seems only to be a registry entry which starts windows update. Maybe somebody can give you a complete image (not me, I'm sorry for both of us, even I have a dutch vista  )

Will your revolutionary update work without exclusive mode ?????

[PeterSt]

Will your revolutionary update work without exclusive mode ?????

I think so yes. At least I got Shared Mode running in my Exclusive Mode "capable" system. The more difficult point is that I can't test it in a "not Exclusive Mode capable" system. But now I recall ... you just *have* it running in non-Exclusive Mode, right ? Well, then it will work.

Did you have contact with Microsoft about this bug? I'm afraid the soundsystem is not their highest priority, and if they don't know the problem ...

No. But it would be without use. All people asking questions in this area never get any answers. Man, I must have been very lucky I got the base of it running ... which officially it could not in the original Vista RTM. So I put my horses on that now (SP1) it officially can, and one of the reasons it could not before was that not everything was supported in the kernels. Another indication that it just wasn't finished (which nobody admits of course) is that I receive an error message number which is not in any list. So that's where I got stuck last Sunday, but now I will proceed on getting the message behind that number (no, Google never heard of it ).

[Telstar]

All very interesting.
I will see if i cand find a vista SP1 RC for you to download.

[andy74]

Guys HELP !!!

I would like to join your exciting expectation of the new version.
But I don't understand what is this all about. Probably my poor English again.

Can anybody explain to me what are we waiting from 0.9u and most importantly what is this BIG CHANGE? from the technical point of view.
if it is not 24 bit playback then WHAT?
I am completely confused!

Thanks
Andrey

[PeterSt]

When I was ready last night with typing and all to post the upgrade, all the connections to the server concerned were dead (you probably noticed it too). No, I carefully saved my typing this time. Later today I'll have another attempt.

[GerardA]

And did your attempt succeed?

[PeterSt]

Hahaha, what do you think ? ...

No, yesterday I gave myself the freedom to get 24 bits working in the mean time, and after a whole day of trying I indeed got it working, BUT, with Shared Mode only. Note that so far this works with native 96/24 files only, and *not* with upgrading 44.1/16 to 44.1/24.

A kind of major problem as of now, is that I am starting to adopt the idea that this is caused by my RME drivers and not by Vista. Today I will ask RME about it, and knowing how responsive they are, I expect some result from that soon. If it's them, even the solution might come soon.

Of course this is -then- related to me only, and not necessarily to "you". The breakthrough however, is that I *do* have a working situation now which I can apply theoretically to Exclusive Mode, so then it just might work for "you".

Since I have been messing the program all over yesterday, currently I have to get things back in order first before I can upload an upgrade. This time I better say "maybe not today", but I will try anyway.

On the danger of getting too much off topic (as how it seems, but it is not ! here's a premature "advise" on how to deal with Shared Mode and the means how XXHighEnd plays the audio stream (which is *not* equal to anything else !) :

As soon as Exclusive Mode won't work (and I think this is your situation with the Terratec, GerardA) there must be a strict match between what XX outputs and what is set as rate/bits in Properties - Advanced of the audio device. If not, the message "Sample rate does not match ..." is given by XX.

One really major problem coming from this, is that it would be quite impossible to play various rates/bits in a mix without adjusting those properties all the time. Thus, supposed 44.1/24 could work, then when the properties are set to that, a genuin 96/24 file can't play. And the other way around. Keep in mind : this is only when Exclusive Mode is rejected. Thus, suppose that 96/24 can normally work in Exclusive Mode (like 44.1/16 can), and 44.1/24 can not, then there is no problem because you can set the properties to 44.1/24 and all is okay all time. Uhhm ... mind you, NOT when you also want to play a movie which is always 48/16 but ... which might "resample" to 44.1/24 then.

To get the merits of it all (and I hope I understand it a bit by now), as said, keep in mind that only the way XX plays REQUIRES that exact setting when it plays in Shared Mode (for "why ?" see below). Thus, supposed the properties are set to 44.1/16 and you play this movie of 48/16 there is no problem. Vista will resample that to 44.1/16. Okay, if you don't call this a problem ...

Why does XX require the exact setting ? because otherwise the same will happen as with the movie example. You insert that e.g. 96/24 file, and in the mean time Vista makes 44.1/16 of it (or whatever the properties are set to) ... and you wouldn't even notice it (but for poorer sound of course).

An in fact very important thing is that you *must* be able to see the playback rate and bits as output by Vista. So, I can imagine that a lot of you just can't, but if you indeed can't you will not be able to check whether your - and XX' settings work out as intended.
In my case the RME driver (screen) can show the samplerate, but not the bits. And mind you, those too get "resampled" if Vista likes it (Shared Mode). Ok, IOW, if you could find software that shows what's really output to the matter of rate and bits ... please let it know.

In the very end it is not all that much important, as long as we can trust the messages from XX. Thus, as said, when XX just rejects a non-matching rate/bits then it wouldn't play anyting you did not intend. Right ?
Apart from a messy program, currently it is my task to find out my the most "structured" means, what is and what is not possible, in order to show you the appropriate message. So this is something I have to do too now.

Last thing : As far as I can determine, playing in Shared Mode under the restriction of Vista (the soundcard) playing the rate/bits as requested, does not harm the bits. I did not check the output with the source, but my interpretation of all tells me it whould be ok. But :
For those who pursue that 96/24 (or possibly 88.2/24) I personally don't think it is better for SQ. Why ? well, because it would need an inheritently worse DAC to play it. In my case it would be the 96/18 TwinDAC+ vs. the 192/24 Fireface800, and no matter the Fireface is used at 96/24 vs. the TwinDAC+ at 44.1/16, the TwinDAC+ wins. Or I must have picked the wrong Linn examples ... which I don't think is so. And of course *I* must add to it : any 24 bits DAC will be an oversampling DAC. For me this is good-bye to good SQ.

On the other hand ... you will soon see why the being able to playback those 24 bits is important afterall.

[Leo]

The Altman DAC claims to do non-oversampling up to 192 kHz  .

It is way beyond my knowledge to say whether that is bluff or true, nevertheless if it is true you are in for another challenge Peter

[PeterSt]

Hi Leo,

No ... no challenges here. The sample rate isn't the problem. The problem is the bit depth. Similar to the Altman allowing for 192KHz while it is 16 bits only. 16 bit non-oversampling DAC-chips just exist, 24 bit nos chips do not. So a first "problem" would be that e.g. 96/24 files just cannot be played with the SQ I myself have in mind. A derived "problem" is, that files with 88K2 and higher, in 16 bits just do not exist (which is perfectly logic). In the end, any 88K2 and higher DAC not being able to use 24 bits at all, is a kind of worthless. E.g. what Altman provocates is that DVD-A can be played because of that 192KHz. Sadly DVD-A's are not in 16 bits. So the bits will be just cut off implying not only a lower resolution, but noise as well. It just doesn't work (with such a simple approach).

In the near future you will learn more of this. Just watch these pages ... 
Peter

[PeterSt]

No, yesterday I gave myself the freedom to get 24 bits working in the mean time, and after a whole day of trying I indeed got it working, BUT, with Shared Mode only.

Got it working for Exclusive Mode now too. Even padded to 32 bits ... 

[GerardA]

Yes yes yes!

BTW, I just read that SP1 can be found on Pirate Bay, no idea what that is
Or did you already find it?

[PeterSt]

No. But I don't think it is necessary anymore ! So :

Also in order to keep this topic going a bit : 44K1/16 to 24 or 32 works as well.

Now up to giving those options as choices in the XXHighEnd UI. And then let's see whether we can get those Terratecs going a bit.
Oh, and before this small project is finished I will try to get 18 bits (TwinDAC+/SPDIF) and 20 bits (??) going usefully.

And if you then know that this topic actually was about something else ... (which it is not ).

[GerardA]

We still get the clarity and spaciousness?

[PeterSt]

You bet you will !

[Ava12]

So 20Bit?
Is there any chance to get a HDCD Stream (from a HDCD CD) as lets say a 20 Bit wav file (I don't know if there is more 20 Bit material out there).
So you can play it in XX. When 20 Bit is no such a hard challenge as 24 Bit? 

[PeterSt]

Hahaha, I was waiting for that question. Ok, right now ? no. Why ? because that takes special arithmetic (and knowledge ) which I have to dive into. So, HDCD is not 20 bits, it acts like it. Ok, you knew that ... In the end, yes ... *or* the HDCD can be interpreted in the way it should (but your DAC must be able to cope with it ... that could even go without my attention i.e. it could work right now *or* I could expand the 16 bits loslessly to 20 bits and then each 20 bits or larger DAC could play it.

So, later ...

[Ava12]

But every HDCD which I rip with EAC contains the HDCD stream in the wav file, doesn't it?

[PeterSt]

Yes. And as long as the DAC doesn't interpret it as intended, you'd play that CD with less resolution than 16 bits.

[Ava12]

So and lets say, if XX would support 24 Bit or real 20 Bit resolution you would have to use an alogarithm which converts the "20 Bit" wav into a real 20 Bit or even 24 Bit stream.
Wouldn't that be the way?

[PeterSt]

Yep.
Keep in mind though that all *is* a trick. So a real 20 bit recording would be better than a HDCD recording. But if all is right the HDCD recording will be perceived as better and higher rez than a 16 bit recording. Well, that is what I understood from it when I read it a while back. I never heard it.

[GerardA]

For HDCD you need the special chip in your DAC like a PCM100 if I recall correctly.
My Rotel has one and I can only say that all the HDCD's I have sound better.
But it is to my ears especcially the bass that becomes more lifelike.
Why does the extra resolution give better base???
And why does XX give so much better base??
Peter??

[PeterSt]

For HDCD you need the special chip in your DAC like a PCM100 if I recall correctly.

Yes, but as said, I can interpret the same as a DAC does, and then "mangle" the data so that a normal DAC would understand. That is, if it can cope with the resolution (*and* if the means of doing it is not propriatery, but I don't think it is).

On your other questions ... let's wait until 0.9u. I think a lot of things become more clear then.
And more confusing.

[SeVeReD]

For HDCD you need the special chip in your DAC like a PCM100 if I recall correctly.

Yes, but as said, I can interpret the same as a DAC does, and then "mangle" the data so that a normal DAC would understand. That is, if it can cope with the resolution (*and* if the means of doing it is not propriatery, but I don't think it is).

On your other questions ... let's wait until 0.9u. I think a lot of things become more clear then.
And more confusing.

If I remember correctly HDCD was bought by Microsoft and I think they've let it die.

[edward]

If I remember correctly HDCD was bought by Microsoft and I think they've let it die.

You are correct, Dave, Microsoft did "buy" HDCD.

For HDCD you need the special chip in your DAC like a PCM100 if I recall correctly.

Well, not necessarily. What Microsoft did when they bought the technology was incorporate the decoding into Windows Media Player. As long as you set WMP to output 24 bit and your soundcard can accept 24 bit, then HDCD is decoded by the software and the "special chip" ( Pacific Microsonics PMD200) is not necessary. I have a handful of HDCDs and I can play them through WMP to my USB DAC and the HDCD light comes on on WMP and it decodes properly. It sounds pretty good too. (This, of course, does not work if you are using S/PDIF out)

So this is somewhat good news for Peter - in that it is possible to incorporate the decoding into XXHighEnd. However, Microsoft is not going to license this to Peter, so I don't know if there is a way around it.

On the hardware side of things, here is a blurb from a Stereophile blog regarding Reference Recordings HDCD releases and the new Alpha DAC:

...Reference Recordings therefore stuck with HDCD. Unfortunately, after Microsoft bought the rights to HDCD from Pacific Microsonics a few years back, it dropped development and support for the format. Fortunately, two of Johnson's dear friends and associates from those days, Michael Ritter and Pflash Pflaumer (who co-invented HDCD with Johnson), have since launched Berkeley Audio Design Associates, and have developed an Alpha DAC that Martin claims sounds better in some ways than Pacific Microsonics' no-longer-produced Model 2 HDCD encoder.

The Alpha DAC, which does not infringe on the HDCD patents that Microsoft now owns, will debut at CES (2008)...

[PeterSt]

Guys,

I know you are waiting. But so far what I can do with 16 bits, I can't do with 24 (whatever it is I mean by that ). Theoretically it can be done, but in practice things are, say, just a tad too complicated for me.
Since this morning (after many days of struggle) I have sound from it which doesn't sound wrong. However, so far I can't prove that what happens in the least siginificant bits of the 24 bits (the part which has the higest resolution) is correct.

So please be patient.
Peter

[Ava12]

I am and I think all are.
If that'll work, it would be a major upgrade, so the patience will be worth the little more time to wait.
Isn't it that way XXHighEnd'ers?

[Gerard]

Isn't it that way XXHighEnd'ers?

Krek! 

[GerardA]

Waiting only makes it better, or Peter, you get new ideas and we can enjoy it after.
In the mean time our curiosity gets stronger and I was thinking what you are going to do with those extra bits?
Are you trying to get extra definition in there, like trying to calculate all the right bits.
Then what are you going  to do to do this, I mean 16 bit music will often be derived from 24 bits and the last 8 bits removed and some dither added.
Will it be possible to reconstruct this in a way that comes close to the original? I guess you can not undo the dither-effect or is that possible?
Or do you just leave those extra bits at zero?

[PeterSt]

Hahaha, you are not even close.

[GerardA]

Haha,
So I guess your not gonna tell us!
Let's see if my ears can hear what's happening.

[PeterSt]

That would be a nice challenge ! But I don't think anyone can even guess what I did.

Now I think of it ... I must say honestly that I have rather big problems to explain it. Ok, I tried that first night the lines were down and the upgrade couldn't be posted, but I'm sure not satisfied with that text (which I still have). Now indeed a better way might be that those amongst you who think they know what they hear, express it by whatever technical means they see going with it.
I can guarantee that *everybody* will hear "a" difference. I can not guarantee that everybody likes it better, already because you are not used to it. I think that after a week of listening, maybe 10% does not like it, because they somehow disagree with what happened for theories. Mind you, this is important.

Man, am I eager to post this upgrade ....

As implied in the first post, the applied change is similar to the oversampling DAC vs. non-oversampling DAC principle, and many just don't agree with the nos principle (already because that doesn't measure right).
0.9u does similar, and indeed in something like the "nos" area. What you perceive from it has a much bigger impact though. Hahaha.

And guess what ? again I did NOT apply any DSP. Go figure !!

So, indeed it might be better not to tell at first what happened, and it would be great (well, for me anyway) if I could explain what happened *after* a few of you who can "hear" the idea about this, tell what they think "functionally" happened (and I know, if you know what happened you can just hear that back). This has a major advantage too : no placebo effects possible. And mind you, this is important when someone applies "technology" which is completely new.

One hint :

A year ago or so, I publically wondered (on the bd-design forum where all pre-XX posts were happily thrown away ) what could be wrong with CD players because they so much underperformed compared with XX (yeah, a year ago this was !). Well, I think I know now ...

[Gerard]

So, indeed it might be better not to tell at first what happened, and it would be great (well, for me anyway) if I could explain what happened *after* a few of you who can "hear" the idea about this, tell what they think "functionally" happened (and I know, if you know what happened you can just hear that back). This has a major advantage too : no placebo effects possible. And mind you, this is important when someone applies "technology" which is completely new.

Peter,

With the upcoming 0.9u is there a difference in SQ between people who already have a dac in there system and people like "me" who does not have a dac?  Meaning is the advantage bigger or less with or without.

Grtz

[PeterSt]

I really can't tell Gerard; If you don't have a DAC you would be playing vinyl, reel to reel tapes, music cassettes or maybe DCC tapes.
Yeah, I think you should be able to hear a difference.

 


 

[PeterSt]

Gerard, no ... I guess I tried to be funny.
So yes, I think everybody can hear it. You can always try to explain what you hear, and nobody will make fun of it (not like my before post, which just proves wrong taste). If you don't hear it, you just say nothing. Okay ?

But remember some 6 months ago, when you proved to hear things which were right on spot !

[Gerard]

Gerard, no ... I guess I tried to be funny.
So yes, I think everybody can hear it. You can always try to explain what you hear, and nobody will make fun of it (not like my before post, which just proves wrong taste). If you don't hear it, you just say nothing. Okay ?

But remember some 6 months ago, when you proved to hear things which were right on spot !

hahaha Ok good that you said it was funny meant... Because of my bad English reading i did not understand and thought it was abacadabra writing   Like you do so now and than...

Good to see you are in a good mood.... That means things work out the good way!! 

I will see what i can make of it!....

 

[PeterSt]

Ok, small update :

Has anyone seen software to actually utilize an 18 bit DAC ? maybe you say Yes, but until yesterday I did not.
Today I have ...

[Gerard]

Ok, small update :

Has anyone seen software to actually utilize an 18 bit DAC ? maybe you say Yes, but until yesterday I did not.
Today I do ...

[JohanZ]

Has anyone seen software to actually utilize an 18 bit DAC ?

No!!!!

But ...when ...

[GerardA]

I bet on sunday, like most new releases!

[SeVeReD]

Raise your hand if you recognize the saying:
Chomping at the Bit