XXHighEnd - The Ultra HighEnd Audio Player
October 12, 2024, 04:52:10 am *
Welcome, Guest. Please login or register.
Did you miss your activation email?

Login with username, password and session length
News: August 6, 2017 : Phasure Webshop open ! Go to the Shop
Search current board structure only !!  
   Home   Help Search Login Register  
Pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 [16] 17 18 19 20 21 22 23 24 25 26 27 28 29 30  All
  Print  
Author Topic: World's first NOS 24/384 filterless DAC  (Read 596097 times)
0 Members and 3 Guests are viewing this topic.
bhobba
Audio Loudspeaker
*
Offline Offline

Posts: 32


View Profile Email
« Reply #225 on: April 28, 2010, 02:18:30 pm »

Funny Bill ... but you REALLY are the first one to ask. I guess it is known that I don't tell everything which is regarded a secret (propriatary) in the first place ? Or everybody thinks this is no problem ?

Your number is on the low side though. swoon
Peter


PS: But 24 bits !

Cant wait to hear what you can divulge - it really has piqued my curiosity.

Thanks
Bill
Logged
Telstar
Audio Addict
***
Offline Offline

Posts: 732


View Profile
« Reply #226 on: April 28, 2010, 04:43:52 pm »

Funny Bill ... but you REALLY are the first one to ask. I guess it is known that I don't tell everything which is regarded a secret (propriatary) in the first place ? Or everybody thinks this is no problem ?

I think I know. But I wont tell grazy

Logged

(2nd Apr 2018)
Software:
W10 14393 Pro x64 | XXHE 2.10 | MinOS | Q=14x1/0/0/0/0 | SFS 5,19 mixed contiguous | Nervous rate 1 | 4096k buffer |

Hardware:
OrigenAE H5 case | E5300 fanless |  8GB RAM | Winmate DC-DC fanless PSU | OS on SSD | Renesas USB3 pcie card | Belden highspeed usb cable | Audio-gd dac19 NOS with sigxer F1 | My_ref_FE mono amps | Albedo Apex speakers
bhobba
Audio Loudspeaker
*
Offline Offline

Posts: 32


View Profile Email
« Reply #227 on: April 29, 2010, 06:14:34 am »

I think I know. But I wont tell grazy

Fair enough.  But that does still leaves another issue.  XX has an option to select 384/32 as your dac.  My understanding is that the windows drivers doesn't support that high a transfer rate.  How can you use that kind of dac?

Thanks
Bill
Logged
PeterSt
Administrator
High Grade Audiophile
*****
Offline Offline

Posts: 16851



View Profile Email
« Reply #228 on: April 29, 2010, 09:16:19 am »

Quote
But that does still leaves another issue.

Of course, many people are controlled by their computers. But I learned to be in control over them machines hehe

You can't guess even half of what all happened ... Happy

Logged

For the Stealth III LPS PC :
W10-14393.0 - July 17, 2021 (2.11)
XXHighEnd Mach III Stealth LPS PC -> Xeon Scalable 14/28 core with Hyperthreading On (set to 14/28 cores in BIOS and set to 10/20 cores via Boot Menu) @~660MHz, 48GB, Windows 10 Pro 64 bit build 14393.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 14/-/0/0/*1*/ Q1Factor = *4* / Dev.Buffer = 4096 / ClockRes = *10ms* / Memory = Straight Contiguous / Include Garbage Collect / SFS = *10.13*  (max 10.13) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / Stop Desktop, Remaining, WASAPI and W10 services / Use Remote Desktop / Keep LAN - Not Persist / WallPaper On / OSD Off (!) / Running Time Off / Minimize OS / XTweaks : Balanced Load = *62* / Nervous Rate = *1* / Cool when Idle = n.a / Provide Stable Power = 1 / Utilize Cores always = 1 / Time Performance Index = Optimal / Time Stability = Stable / Custom Filtering *Low* (16x) / Always Clear Proxy before Playback = On -> USB3 from MoBo -> Lush^3
A: W-Y-R-G, B: *W-G* USB 1m00 -> Phisolator 24/768 Phasure NOS1a/G3 75B (BNC Out) async USB DAC, Driver v1.0.4b (16ms) -> B'ASS Current Amplifier -> Blaxius*^2.5* A:B-G, B:B-G Interlink -> Orelo MKII Active Open Baffle Horn Speakers. ET^2 Ethernet from Mach III to Music Server PC (RDC Control).
Removed Switching Supplies from everywhere (also from the PC).

For a general PC :
W10-10586.0 - May 2016 (2.05+)
*XXHighEnd PC -> I7 3930k with Hyperthreading On (12 cores)* @~500MHz, 16GB, Windows 10 Pro 64 bit build 10586.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 14/-/1/1/1 / Q1Factor = 1 / Dev.Buffer = 4096 / ClockRes = 1ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = 0.10  (max 60) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / All Services Off / Keep LAN - Not Persist / WallPaper On / OSD On / Running Time Off / Minimize OS / XTweaks : Balanced Load = *43* / Nervous Rate = 1 / Cool when Idle = 1 / Provide Stable Power = 1 / Utilize Cores always = 1 / Time Performance Index = *Optimal* / Time Stability = *Stable* / Custom Filter *Low* 705600 / -> USB3 *from MoBo* -> Clairixa USB 15cm -> Intona Isolator -> Clairixa USB 1m80 -> 24/768 Phasure NOS1a 75B (BNC Out) async USB DAC, Driver v1.0.4b (4ms) -> Blaxius BNC interlink *-> B'ASS Current Amplifier /w Level4 -> Blaxius Interlink* -> Orelo MKII Active Open Baffle Horn Speakers.
Removed Switching Supplies from everywhere.

Global Moderator
bgjohan
Audio Loudspeaker
*
Offline Offline

Posts: 46


View Profile Email
« Reply #229 on: April 29, 2010, 03:07:02 pm »

Peter

The thread has earlier touched on the possibility of digital filtering (in software) as well as the possibility for the DAC to perform the ADC of the analog signal from a turntable.

From that perspective, I am interested in your thoughts on the pros and cons of applying the RIAA filtering in the analog vs digital domain.
From a practical perspective, I assume that for implementation of digital RIAA filtering  in XX software, analog signal (from turntable w-out analog phono stage) would first need to be fed to the DAC for ADC conversion, then digital signal fed back to computer for RIAA filtering in XX software and finally digital signal fed back to the DAC for conversion to analog before signal OUT fed to the amp.

In general, if analog signal from the turntable (w/out analog phono pre-amp) is too low for processing in your DAC and some form of phono pre-amp required between the turntable and the DAC, analog RIAA filtering may just as well be performed in the phono pre-amp (and forget about RIAA filtering in the digital domain).

Best regards

Bjorn
Logged
PeterSt
Administrator
High Grade Audiophile
*****
Offline Offline

Posts: 16851



View Profile Email
« Reply #230 on: April 30, 2010, 11:38:19 am »

Hi Bjorn,

A not so easy question (for me Happy). Why ? well, because you are (I think) using the digital RIAA correction at real time playing vinyl, and the (obvious, to me) reason to do it is to improve on the phono stage. I don't know ...

I don't know because it needs the experience, and I guess I won't ever have that, lacking a good turn table. But what I can reason out (who can't) is this tradeoff :

a. Play vinyl as analoguely as it is, with the best phono stage to your knowlegde;
b. Record vinyl to digital, and play it back from there with an undoubtedly better RIAA correction (I'm rather sure).

So the point is : what harm does the transfer to digital ?

This question alone is a can of worms by itself, for which one has to be thinking quite controversely. Ok, I can do that and I will, but in the end it doesn't make sense I think. Look :


Whether it is a digital transfer of direct cut vinyl (the guys doing that CLEARLY don't apply stupid "corrections" as common todays digital mastering does), or whether it is an uploaded "vinyl rip" from someone (those seeming to have good turntables in the first place) ... the files coming from that to me CLEARLY sound analogue. Thus, what I'm actually saying is : they sound better than random good recordings from a CD.
Of course, those vinyl rips will have been done at 24/96 or so, but the direct cut etc. stuff I talked about is not (just 16/44.1) and the character of both is very much the same CLEARLY, to me.
Notice that this tells at least me that digital mastering is flawed (generally).

But what to do with this ?
Will it mean that when I play back my vinyl through my modest turntable and through digital RIAA correction - through XXHighEnd, it will sound better than directly via the phono stage and without transfer to digital ?
or
Will it mean that when you play back your vinyl through your super turntable and through digital RIAA correction - through XXHighEnd, it will sound better than directly via the phono stange and without transfer to digital ?
That would be odd if so.

But sadly this is about a heavy case of apples and oranges because of the two variables in both equations : transfer your nice analogue to digital on one hand, and the digital RIAA correction on the other.

So ... ONLY when the digital RIAA correction is very much profoundly better than analogue to digital transfer destroys, there is a (very) good reason to play back vinyl this way. And I don't say this won't be the case ...

I can add to this, that IMO (no, plain experience) there really is nothing wrong with digital 16/44.1. In other words, assuming the A/D conversion is performed allright, it just doesn't harm at all. I also derive this from rather random "vinyl rips" I own, and not one of them sounds worse than normal digital to me. And worse ... they sound more analogue ... it should even happen with my modest turntable ...

When you read back the above (and assumed my own observations are correct) there would be one logical conclusion only : it is the RIAA correction in the analogue domain destroying the stuff. Thus, as long as nobody tells me where my story flaws, we should indeed make the provision. It should even be able to "bear" the randomly modest A/D the DAC theoretically has (practice when I provide the connections at the back of the cabinet), because remember, it is a DAC; nothing has been done to let the A/D excel. One little iny whiny exception to this : the A/D will be able to record in in 24/384 ...
swoon

Yes, this latter may make you laugh, but I never thought of the combination until I wrote the previous alinea. So I'm at least laughing myself ! Rather outloud !
Hmm ...

It would be quite a joke actually;
No filtering whatsoever would be needed at the recording stage, and no filtering (like Arc Prediction Upsampling) would be needed at the playback stage either. No dither anywhere, nothing. All will be as HD free as the equipment (turntable, ADC, DAC, amp) permits, and you could well say it should be the best "vinyl rip" we've ever heard (unless people are used to listen to 24/384 takes from vinyl right now, which at least will go through OS DACs, so not the best at all (remember, my view)).

I still can't type of laughing ...

Well, at least this has made me decide to activate the recording capabilities in the DAC, which so far seemed an unnessecary thing to do (because it is just there, but not explicitly made on par with the D/A capabilities (for SQ)). But if *this* comes from it for the better, I just should do it.
As I (I think) told earlier somewhere, the recording capabilities are in XXHighEnd, and they are real time. Thus, exactly what we need here. The only thing it needs is the RIAA correction in between ...

but

This will not be conform the way XXHighEnd operates normally, because the RIAA correction now needs to be performed in realtime too. And no such process exists at the moment. So, SQ itself will degrade because of that, but YMMV to what degree and the net result (compared with direct analogue playback from the turntable).


Keep those ideas coming ! It only makes the DAC ready later again. oops
Peter


PS: ... which really seems to be a problem for me; although this improvement of the "DAC" is in a complete different area, I (or we Happy) keep on having ideas to improve the DAC, and it is REALLY worth while !

PPS: I didn't tell it yet, but the DAC is in production for two weeks now, and the ideas I keep on having are all at the back end, so to speak. So, they don't hold up current production, but until everything is assembled into the cabinet, at that stage things can still change. Allowing input terminals is an example of that, but there is more.
Logged

For the Stealth III LPS PC :
W10-14393.0 - July 17, 2021 (2.11)
XXHighEnd Mach III Stealth LPS PC -> Xeon Scalable 14/28 core with Hyperthreading On (set to 14/28 cores in BIOS and set to 10/20 cores via Boot Menu) @~660MHz, 48GB, Windows 10 Pro 64 bit build 14393.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 14/-/0/0/*1*/ Q1Factor = *4* / Dev.Buffer = 4096 / ClockRes = *10ms* / Memory = Straight Contiguous / Include Garbage Collect / SFS = *10.13*  (max 10.13) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / Stop Desktop, Remaining, WASAPI and W10 services / Use Remote Desktop / Keep LAN - Not Persist / WallPaper On / OSD Off (!) / Running Time Off / Minimize OS / XTweaks : Balanced Load = *62* / Nervous Rate = *1* / Cool when Idle = n.a / Provide Stable Power = 1 / Utilize Cores always = 1 / Time Performance Index = Optimal / Time Stability = Stable / Custom Filtering *Low* (16x) / Always Clear Proxy before Playback = On -> USB3 from MoBo -> Lush^3
A: W-Y-R-G, B: *W-G* USB 1m00 -> Phisolator 24/768 Phasure NOS1a/G3 75B (BNC Out) async USB DAC, Driver v1.0.4b (16ms) -> B'ASS Current Amplifier -> Blaxius*^2.5* A:B-G, B:B-G Interlink -> Orelo MKII Active Open Baffle Horn Speakers. ET^2 Ethernet from Mach III to Music Server PC (RDC Control).
Removed Switching Supplies from everywhere (also from the PC).

For a general PC :
W10-10586.0 - May 2016 (2.05+)
*XXHighEnd PC -> I7 3930k with Hyperthreading On (12 cores)* @~500MHz, 16GB, Windows 10 Pro 64 bit build 10586.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 14/-/1/1/1 / Q1Factor = 1 / Dev.Buffer = 4096 / ClockRes = 1ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = 0.10  (max 60) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / All Services Off / Keep LAN - Not Persist / WallPaper On / OSD On / Running Time Off / Minimize OS / XTweaks : Balanced Load = *43* / Nervous Rate = 1 / Cool when Idle = 1 / Provide Stable Power = 1 / Utilize Cores always = 1 / Time Performance Index = *Optimal* / Time Stability = *Stable* / Custom Filter *Low* 705600 / -> USB3 *from MoBo* -> Clairixa USB 15cm -> Intona Isolator -> Clairixa USB 1m80 -> 24/768 Phasure NOS1a 75B (BNC Out) async USB DAC, Driver v1.0.4b (4ms) -> Blaxius BNC interlink *-> B'ASS Current Amplifier /w Level4 -> Blaxius Interlink* -> Orelo MKII Active Open Baffle Horn Speakers.
Removed Switching Supplies from everywhere.

Global Moderator
PeterSt
Administrator
High Grade Audiophile
*****
Offline Offline

Posts: 16851



View Profile Email
« Reply #231 on: April 30, 2010, 02:25:52 pm »

Dear all,

A few -I think- important milestones have been achieved, so I thought to inform you about it.
Hopefully it also shows better why things seem to take a long time, while it's really worthwhile the long waiting. And, I'd hate to pronounce improvements later, while some of you may have a "less" version in the mean time. So it better be right the first time !

Allright. Starting with what I mentioned in my before post, production really has started a few weeks back. A 100 units will be produced as a first batch (I may sell 10 only haha), so hopefully enough people can try the DAC (money back guarantee) without being in a long back log while others already have it. Thus, only when the 100 are ready, the first will ship.

In the mean time, this will give me the "opportunity" to wait for some parts with a long lead time. This just is so and can't be helped, but for some parts only the best can be used, and they are produced on (my) order. Currently the longest pending lead time is 8 weeks. sorry
But, these are parts which can be assembled quickly, and the assembly won't hold up production really (the waiting for the parts does).

Although you didn't know it, only yesterday another technical highstand has been achieved, and I must say this is about one of the most nice (and difficult !) coorperations not in the same office I have ever met in my life. Think of hundreds of emails about a certain design at the bit level (this is literal), that in the end needing independend programming at two sides, one side being XXHighEnd outputting data, the other side being CPLD programming, which is actually in-DAC and at the engineering level. And, after 6 months of careful preparation, yesterday it worked in one go (neither side being able to test without the other's efforts).
Some know (via offline) what this is about, and actually it is about creating some kind of "double buffer" to be 100% sure it can't be done better. It is about the 24/384 input of which we can reason that better (higher sample rate) isn't necessarym but because the technology to go further was there in (my) theory in the first place, it just was done : 24/768. Mind you, this is for input, and no master files exist for it.
However, since Arc Prediction Upsampling turned out to be a quality phenomenon by itself, it *will* be useful for just that : sound quality.
On this matter, please notice that the step from 24/176.4 to 24/352.8 (with a source of plain 16/44.1) already brought a significant improvement, so stepping further to 24/705.6 most probably will again (I couldn't test this myself, because I don't have the hardware for it yet).

By pure coincidence, also yesterday I was able AT LAST to apply a gain "mechanism" with THD+N specs better than the DAC chips itself, or in other words, which does not degrade sound. Maybe, just maybe I have to come back on this later (because I don't like the sound of it afterall), but chances are very small because of the enormous improvement ot brought (see below story).
This must have been the most time consuming of it all, and in the end is about a decent I/V stage which most probably is the most discussed subject in the DIY community on the Internet. This is also about my earlier expression on "you have to live with 24dBFS under 2VRMS" (which by itself is very doable with enough gain in the main amps and/or enough efficiency in the loudspeakers), which now is just 2VRMS (which is the standard). Also, this will allow for headphone output, although in theory only (read : this has not been tested yet).

To my own surprise, the at last working of this, brought a new dimension in music reproduction through loudspeakers for me. If I had to rate the change in SQ, I would say "10 times better". Again ? yes, again. Look :

People who are here from the beginning (or those who ran into the post(s) concerned by accident) know that I have always had one major negative remark about music reproduction through loudspeakers : the volume level of cymbals and the like. And since we are talking about it anyway, you may come to the same conclusion while I put your attention to it :
Go to a live perfromance, and have the bass player, the singer, the piano player and the drummer (etc.) in front of you. Now put your attention to the balance of all the sounds *but* the cymbals on one hand, and those cymbals at the other. What actually sounds the most loud ? It is the cymbals !
Now go back to your listening room, and try to perceive the same balance. No way it is going to be met. Ha ! you are even lucky if you perceive (real) cymbals in the first place !! ... so bad it is ...
On this matter, maybe three years ago by now, I tweaked my crossovers so I would have 16dB more output at 20KHz, with a nice slope starting at 5KHz and up (to that max at 20KHz). That really helped a lot, and was actually the first time hi-hats became nicely profound while before I never heard them as an explicit instrument playing (while it is the most important instrument for a drummer).
I never removed that tweak up till today (and I still won't). So, although this may sound wrongish to you, it is my opinion this is just needed in order to receive a more or less representative playback for cymbal like instruments. Also, this can't be blamed to our ears, because I really don't have a problem with the drum kit in the house here. The live cymbals perform perfectly right and they really don't need cranking up.
So what is going on really ? And to remember, a self respecting loudspeaker just measures flat.

I think I now found the answer in a (to me) new phenomenon - or dimension if you like : The SPL of high frequencies.
Oh yes ...

I am quite sure (not 100%) that most of you look at SPL (Sound Pressure Level) as a phenomenon which occurs in the lower regions (I mean under 5KHz or so). SPL is equallish to sound pressure you can feel, and it comes from base drums, general low frequency sounds and maybe ar loud singing voice. But not from something like a tambourine. No ?

Ha, come here and be startled. And this is exactly why I rather call this a new dimension than just a phenomenon. It makes your whatever music completely new. There's a whole other band playing besides the one you were used to. It is the drummer with his gags.
Btw, I must admit, the mid level increased with it, and those sensitive to that will be quite happy just because of that alone. But that's another subject (as important btw, but not as a new dimension as such).

Quite a few virtues spring from this, like really fresh sound is one of them. So, people who tend to talk in terms of removed blankets ... here you go ! it is really amazing.
But what actually happened ?

To be honest, at this moment I don't know yet, although some of it could be predicted from earlier attempts in a similar setup. I think the other day I talked about cymbals sounding for 20 seconds instead of the "usual" 5 seconds, which I remembered from this earlier setup, though not good sounding. So, it seems logic that (relative) additional amplification of the high frequencies make jump out a cymbal for 20 seconds long, while before it could only jump out for 5 seconds. It was there for 20 seconds allright, but the sound of the piano etc. made it disappear. Not so anymore. And, obviously along with it goes the as louder attack. So, just louder (and not a bit, but severely more).
If I had to explain it from a technical point of view, I would say this :
The setup from before was about speed. You have read about it, just explicit speed. But since this is a.o. about leaving out parts, it is also about leaving out "drive" (notice : in my case, and of what I can reason only after the happening). Now, we tend to derive drive from low frequency sounds. In other words, I think it is recognizeable that when we don't have enough drive, the sound gets too thinny and the bass disappears. And, since this was not the case in my setup, I had enough drive. However ...

A low(er) frequency sound moves slowly for its up and down (plus and minus voltage) wave. It is not difficult for such a wave to go down again after it reached the (volume) top, because it is slooooow. It needs power (yes, drive) to go up and down, and it even needs additional power when the frequency is low because it has to move a lot of air (simply said, and expressed in the wrong domain perhaps), but what about the high frequencies which have to be pulled back in time ? So, a high frequency may not need much power to move the air (which is related to the directionality of it !) but it still needs power to retract from its top in the wave back down to its low side, etc. And *this* goes 1000 times faster for 20KHz opposed to 20Hz. It may even need more drive to control it, and in the end is similar to reracting a loudspeaker diaphragm from its excursed state back to the other side. If this would be left to the mechanical properties of the diaphragm it would be way too slow ...
Further, try to imagine that everything which is too slow to cause that reraction, will cause peaks not to be reached (because the power to create the peak is slow, the creation of the peak is behind, and before it's at it's top it has to go down again ... and peak = SPL (volume)).

The above summarized :
While I had created a most speedy setup with lowest distortion figures possible, apparantly (!) this setup didn't contain the drive to perform in the high frequencies. Notice also that without a speedy setup, the higher frequencies don't perform for the same reason (all is too slow again, read above alinea). Now though, I was able to create a situation with the same speed, but with the power needed. Ehm, apparently, and as long as my own reasoning will stand.

Still here ? well, get intrigued on the next then :

I have always been shouting that my cymbals improved and improved, and whereever I go they are not there at all. But, whenever I said that, I also said (a bit depending on the stage I was at) that they could be improved still. One time they are too plastic, the other time they hiss too much, and the next time the color ain't exactly right. But please imagine, this always was in the "environment" of the cymbals being too much in the background. And as I said in the beginning, many of you will be glad to perceive them explicitly in the first place, although you always will be hearing that they are there. Also, even with my 16dB increase of the higher frequencies, it was allright for me, because they were not too profound (as said, still in the background). But hey, not so anymore !
And so you can imagine that I created myself a huge problem, because at any smashing cymbal in your face, they better sound real !!

oops

Thus, in the very end I may be much excited what has been achieved on this, and it may sound the most interesting to me at the moment, but the next thing now is to get it completely right. It needs much more albums to play to perceive the real merits of what I just said, but I expect some problems here. One thing : I already know that the cymbals sound way better than before for their colour, so nothing turns out to be worse opposed to before. This means that the route itself is the right one which is important for the need to (not) undo that. It is to be awaitened though whether I can "hear" where to tweak further, but hey, I have some 8 more weeks for that (according to the beginning of this post), because this too is back end stuff.


Last thing for now :

Currently I am working on the rather tough decision to lower the jitter which officially is under the audible level already (4ps RMS max);
The reason to do this is an attempt to remove the influence from software (players). This anticipates on software influencing the jitter level itself, which by that means reaches the audible level (like 100ns or whatever). Notice that this possibly happening is only a thought and can't be proven (measurement will show just 4ps, but measurement implies a fluent load on current and is no music (with varying load)). The idea is that implementing a much lower jitter clock (if possible at all) will let rise the jitter again during music playback, but not above audible levels.

So, *or* I may succeed in finding a lower jitter clock solution and if I'm right we can throw out Q1 with unpredicted result (will SQ now always be at its best ??) *or* I let it be deliberately as it is now, so we can keep on using XXHighEnd as we are used to (and have fun with !).

The latter seems stupid to do when the first would be feasable. This is why it is a tough decision (and this too is a back end application).
Notice that with the current (4ps for net measureable result) XXHighEnd influences as much as ever before.

So far for now !
Peter
Logged

For the Stealth III LPS PC :
W10-14393.0 - July 17, 2021 (2.11)
XXHighEnd Mach III Stealth LPS PC -> Xeon Scalable 14/28 core with Hyperthreading On (set to 14/28 cores in BIOS and set to 10/20 cores via Boot Menu) @~660MHz, 48GB, Windows 10 Pro 64 bit build 14393.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 14/-/0/0/*1*/ Q1Factor = *4* / Dev.Buffer = 4096 / ClockRes = *10ms* / Memory = Straight Contiguous / Include Garbage Collect / SFS = *10.13*  (max 10.13) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / Stop Desktop, Remaining, WASAPI and W10 services / Use Remote Desktop / Keep LAN - Not Persist / WallPaper On / OSD Off (!) / Running Time Off / Minimize OS / XTweaks : Balanced Load = *62* / Nervous Rate = *1* / Cool when Idle = n.a / Provide Stable Power = 1 / Utilize Cores always = 1 / Time Performance Index = Optimal / Time Stability = Stable / Custom Filtering *Low* (16x) / Always Clear Proxy before Playback = On -> USB3 from MoBo -> Lush^3
A: W-Y-R-G, B: *W-G* USB 1m00 -> Phisolator 24/768 Phasure NOS1a/G3 75B (BNC Out) async USB DAC, Driver v1.0.4b (16ms) -> B'ASS Current Amplifier -> Blaxius*^2.5* A:B-G, B:B-G Interlink -> Orelo MKII Active Open Baffle Horn Speakers. ET^2 Ethernet from Mach III to Music Server PC (RDC Control).
Removed Switching Supplies from everywhere (also from the PC).

For a general PC :
W10-10586.0 - May 2016 (2.05+)
*XXHighEnd PC -> I7 3930k with Hyperthreading On (12 cores)* @~500MHz, 16GB, Windows 10 Pro 64 bit build 10586.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 14/-/1/1/1 / Q1Factor = 1 / Dev.Buffer = 4096 / ClockRes = 1ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = 0.10  (max 60) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / All Services Off / Keep LAN - Not Persist / WallPaper On / OSD On / Running Time Off / Minimize OS / XTweaks : Balanced Load = *43* / Nervous Rate = 1 / Cool when Idle = 1 / Provide Stable Power = 1 / Utilize Cores always = 1 / Time Performance Index = *Optimal* / Time Stability = *Stable* / Custom Filter *Low* 705600 / -> USB3 *from MoBo* -> Clairixa USB 15cm -> Intona Isolator -> Clairixa USB 1m80 -> 24/768 Phasure NOS1a 75B (BNC Out) async USB DAC, Driver v1.0.4b (4ms) -> Blaxius BNC interlink *-> B'ASS Current Amplifier /w Level4 -> Blaxius Interlink* -> Orelo MKII Active Open Baffle Horn Speakers.
Removed Switching Supplies from everywhere.

Global Moderator
Leo
Audio Enthusiast
**
Offline Offline

Posts: 214


View Profile Email
« Reply #232 on: May 01, 2010, 10:14:04 am »



Good morning Peter,

Don't forget my almost grey haired position on the waiting list for the DAC   Cool

Great news that the production has started,

Leo
Logged

Dedicated silent audio pc HFX classic, Windows 8 pro 64bit  / Intel 3930 CPU 6 cores 12 threads,  ASRock x79 Extreme4-M/ SeaSonic Platinum 400w ATX PSU / 16Gb RAM , music on (SATAIII), MinOS/ Engine#4 Special Mode / Q1/2/3/4/5 = *6*/0/1/1/1 Qf=1 (Dev.Buffer = 4096) / not Invert / Playerprio = Low / ThreadPrio = Real Time / *Scheme = 1-2* @ UnAttended  /Services Off + No Running Time / Octo Arc Prediction Upsampling / *SFS=0,4 max= 120*  XT Tweaks balanced load 43, nervous=100, cool when idle 1, Provide Stable Power = 1 / Utilize Cores always = 1 / Time Performance Index = Optimal / Time Stability = On / Double Octo Arc Prediction Upsampling / -> USB with Dexa clock -> 24/768 Phasure NOS1 async USB DAC, Driver v1.0.3 (2ms) ->  direct to AMP: Gainclone mid high, Hypex DPS400 low, horn system (tractrix for mid/high, BD for bass with Oris200)
Telstar
Audio Addict
***
Offline Offline

Posts: 732


View Profile
« Reply #233 on: May 01, 2010, 01:30:29 pm »

Only one thing.

which now is just 2VRMS (which is the standard). Also, this will allow for headphone output, although in theory only (read : this has not been tested yet).

So, there will be a volume control? I dont remember if you canceled this option or not.

The headphone out would be VERY VERY handy. Lots of people like to listen with those (not me unless i'm working), but also allows for an easy comparison with the normal system (amps and speakers), which would be VERY handy for me Happy
Logged

(2nd Apr 2018)
Software:
W10 14393 Pro x64 | XXHE 2.10 | MinOS | Q=14x1/0/0/0/0 | SFS 5,19 mixed contiguous | Nervous rate 1 | 4096k buffer |

Hardware:
OrigenAE H5 case | E5300 fanless |  8GB RAM | Winmate DC-DC fanless PSU | OS on SSD | Renesas USB3 pcie card | Belden highspeed usb cable | Audio-gd dac19 NOS with sigxer F1 | My_ref_FE mono amps | Albedo Apex speakers
PeterSt
Administrator
High Grade Audiophile
*****
Offline Offline

Posts: 16851



View Profile Email
« Reply #234 on: May 01, 2010, 02:07:18 pm »

If I stay with this option, yes, there will be a volume control. But ... only when it doesn't degrade. If it does, it just isn't a good idea. It shouldn't degrade, but the setup is a little strange. So I'll have to measure it. When I have done that, I will let you know.
Logged

For the Stealth III LPS PC :
W10-14393.0 - July 17, 2021 (2.11)
XXHighEnd Mach III Stealth LPS PC -> Xeon Scalable 14/28 core with Hyperthreading On (set to 14/28 cores in BIOS and set to 10/20 cores via Boot Menu) @~660MHz, 48GB, Windows 10 Pro 64 bit build 14393.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 14/-/0/0/*1*/ Q1Factor = *4* / Dev.Buffer = 4096 / ClockRes = *10ms* / Memory = Straight Contiguous / Include Garbage Collect / SFS = *10.13*  (max 10.13) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / Stop Desktop, Remaining, WASAPI and W10 services / Use Remote Desktop / Keep LAN - Not Persist / WallPaper On / OSD Off (!) / Running Time Off / Minimize OS / XTweaks : Balanced Load = *62* / Nervous Rate = *1* / Cool when Idle = n.a / Provide Stable Power = 1 / Utilize Cores always = 1 / Time Performance Index = Optimal / Time Stability = Stable / Custom Filtering *Low* (16x) / Always Clear Proxy before Playback = On -> USB3 from MoBo -> Lush^3
A: W-Y-R-G, B: *W-G* USB 1m00 -> Phisolator 24/768 Phasure NOS1a/G3 75B (BNC Out) async USB DAC, Driver v1.0.4b (16ms) -> B'ASS Current Amplifier -> Blaxius*^2.5* A:B-G, B:B-G Interlink -> Orelo MKII Active Open Baffle Horn Speakers. ET^2 Ethernet from Mach III to Music Server PC (RDC Control).
Removed Switching Supplies from everywhere (also from the PC).

For a general PC :
W10-10586.0 - May 2016 (2.05+)
*XXHighEnd PC -> I7 3930k with Hyperthreading On (12 cores)* @~500MHz, 16GB, Windows 10 Pro 64 bit build 10586.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 14/-/1/1/1 / Q1Factor = 1 / Dev.Buffer = 4096 / ClockRes = 1ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = 0.10  (max 60) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / All Services Off / Keep LAN - Not Persist / WallPaper On / OSD On / Running Time Off / Minimize OS / XTweaks : Balanced Load = *43* / Nervous Rate = 1 / Cool when Idle = 1 / Provide Stable Power = 1 / Utilize Cores always = 1 / Time Performance Index = *Optimal* / Time Stability = *Stable* / Custom Filter *Low* 705600 / -> USB3 *from MoBo* -> Clairixa USB 15cm -> Intona Isolator -> Clairixa USB 1m80 -> 24/768 Phasure NOS1a 75B (BNC Out) async USB DAC, Driver v1.0.4b (4ms) -> Blaxius BNC interlink *-> B'ASS Current Amplifier /w Level4 -> Blaxius Interlink* -> Orelo MKII Active Open Baffle Horn Speakers.
Removed Switching Supplies from everywhere.

Global Moderator
Telstar
Audio Addict
***
Offline Offline

Posts: 732


View Profile
« Reply #235 on: May 01, 2010, 02:14:25 pm »

If I stay with this option, yes, there will be a volume control.

With remote? Happy

<-- lazy.

Quote
But ... only when it doesn't degrade. If it does, it just isn't a good idea. It shouldn't degrade, but the setup is a little strange. So I'll have to measure it. When I have done that, I will let you know.

OK, thanks. You know how eager i am to listen to the "really finished" DAC Happy
Logged

(2nd Apr 2018)
Software:
W10 14393 Pro x64 | XXHE 2.10 | MinOS | Q=14x1/0/0/0/0 | SFS 5,19 mixed contiguous | Nervous rate 1 | 4096k buffer |

Hardware:
OrigenAE H5 case | E5300 fanless |  8GB RAM | Winmate DC-DC fanless PSU | OS on SSD | Renesas USB3 pcie card | Belden highspeed usb cable | Audio-gd dac19 NOS with sigxer F1 | My_ref_FE mono amps | Albedo Apex speakers
bgjohan
Audio Loudspeaker
*
Offline Offline

Posts: 46


View Profile Email
« Reply #236 on: May 01, 2010, 07:15:13 pm »

Peter

As you invited comments, let me expand on the issue of playing analog (vinyl LPs) and RIAA conversion through XX software and NOS1 DAC.

First, my thoughts to tranfer my vinyl collection (vinyl rip to 96/24 and save files to disc) was more driven by the convenience factor; with vinyl in digital domain benefits would be
a. stored in central digital music library and easily accesible (togteher with CD rips and digital dowloads) via computer network to play at any location in my home
b. ability to play vinyl tracks in any order (from one or several LPs)
c. possibility to play through XX and take advantage of its superior abilities

Thus, the above driving the decision to look into option to transfer vinyl into the digital domain, and the discussion of better to apply RIAA in the analog or digital domain stems from that.
In other words, transfer vinyl to digital has benefits over and above the advantage of possibly applying RIAA filtering in the digital domain.

I think useful to outline three different categories of LP listening/ digital conversion using XX and NOS DAC

1. Playing LPs (in real time) with ADC conversion in NOS1 and applying (real time) filtering in XX.
  The reason for vinyl analog to digital conversion is presumably to take advantage of XX digital filtering capabilities; if not for RIAA filtering, then room adjustment and/or x-over filtering. In fact, I see no real separation between digital room adjustment filtering and RIAA filtering. In other words, would it not be feasible for a system set up to use a digital room adjustment filter in XX to have two separate filters – one room adjustment filter for regular digital (CD) files and the second the same room adjustment filter with the RIAA curve imbedded (rather than two separate filters) for vinyl playback.
Also would be great to have NOS1 capability to pipe out the signal both in digital and analog simultaneously; i.e. while playing vinyl and listening to the NOS1 analog OUT signal, the NOS1 digital OUT signal could simultaneously be routed back to the computer and saved. This way, vinyl ripping would not be a separate chore, but a by-product of listening to LPs (optimal if option provided to save the digital file either with digital RIAA filtering imbedded or w/out RIAA adjustment, see below).     

2. Playing a digital file converted from vinyl, but stored on disc as a RAW file without imbedded RIAA filtering.
The file would be subject to digital RIAA filtering in XX when played. However, as the RIAA is not embedded in the file before storing on disc, playback of the file would benefit from continued improvements in XX digital RIAA filter capabilities. (Similar idea to RAW file format for photos).  Playback would require XX (for RIAA filtering), but not necessarily require use of NOS1 DAC.

3.  Playing a digital file converted from vinyl, and stored on disc as a file with RIAA adjustments imbedded (as a result of either analog or digital RIAA filtering).
As for any other digital file neither use of XX or NOS1 DAC strictly required.
However, would be great to have XX capability to upsample a stored 96/24 file to 192/24 and 384/24, the latter presumably requiring NOS1 DAC.

Bjorn

 
Logged
Telstar
Audio Addict
***
Offline Offline

Posts: 732


View Profile
« Reply #237 on: May 01, 2010, 08:20:39 pm »

Oh, I have another request Peter, if it's not destroyed, leave a spdif digital input - thinking to connect my tv.
Logged

(2nd Apr 2018)
Software:
W10 14393 Pro x64 | XXHE 2.10 | MinOS | Q=14x1/0/0/0/0 | SFS 5,19 mixed contiguous | Nervous rate 1 | 4096k buffer |

Hardware:
OrigenAE H5 case | E5300 fanless |  8GB RAM | Winmate DC-DC fanless PSU | OS on SSD | Renesas USB3 pcie card | Belden highspeed usb cable | Audio-gd dac19 NOS with sigxer F1 | My_ref_FE mono amps | Albedo Apex speakers
SeVeReD
Audio Enthusiast
**
Offline Offline

Posts: 599


View Profile WWW Email
« Reply #238 on: May 01, 2010, 08:48:33 pm »

I live
to
 drool

Oh, I have another request Peter, if it's not destroyed, leave a spdif digital input - thinking to connect my tv.

This would be nice ... if it doesn't hurt.
Logged

0.9z-8-3a WAV/CUE files on HDDs via MB FW400>; Win7 pro ttp://www.phasure.com/index.php?topic=352.msg4021#msg4021); [XXHighEnd player  Qs 7, 0, 0, 0, 0; eng 4; adaptive; scheme#3; player priority low; thread priority realtime; clock res 5ms: SFS 420 Wink dac is 24/192 w/32bits; Play Unattended; Stop Services ticked; Wallpaper & Show Back ticked - Mirror Image unticked; Start Engine unticked;garbage collect ticked; copy files to XX-drive; *quad arc prediction upsampling*: straight contiguous:>PCI FW800 card>Fireface 800 DAC [latency 2048 samples for 176.4]; usb/ethernet/mb audio shut off @ MB
Eric
Audio Enthusiast
**
Offline Offline

Posts: 248


View Profile
« Reply #239 on: May 01, 2010, 09:34:25 pm »

Peter,
It is still not clear to me how I should connect the NOS dac to my computer.
Do I need to prepare my computer in some way (hardware, connections?).
I understand that I have still some time to do that  Wink
Cheers, Eric. 
Logged

3.2GHz CPU, 8GB RAM, XXHE 1.186a, W7x64 SP1 Ultimate on 2.5" 10Krpm SATAII spinning disk, 8GB RAM / KS:Phasure NOS1 Out 4.0 / #4 Engine / Adaptive / Buffer 4096 / ClockRes 1ms/ Stop All Services / Monitor Off / SFS = 0.4 / not Invert / No XTweaks / Playback Drive = External USB3 (USB powered) HDD / Unattended/ Minimize OS / Peak Ext / ArcPredict / PA- / Q1,-,3,4,5=14,-,0,0,0
Control Panel: Sample Rate: 352.800kHz/ USB Buffer Size: 8ms
Amps: BD-Design bridged Gainclones
Pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 [16] 17 18 19 20 21 22 23 24 25 26 27 28 29 30  All
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1 RC2 | SMF © 2001-2005, Lewis Media Valid XHTML 1.0! Valid CSS!
Page created in 0.105 seconds with 19 queries.