XXHighEnd

Well, after a quick listen it seems that 9-Z3, in all of its various SFS combinations, has lost a great deal of the midrange that my latest iteration of 9-Z2 is producing. Setting the SFS in the ini file to "00", which I believe is the same as 9-Z2, still leaves the Z3 midrange lacking. Its sorta hard saying this knowing all the work that has gone into Z3 and knowing that I am getting the very best sound I've ever heard short of a live performance  on my current Z2 setup. I had both Z2 and Z3 on the RAMDisk and i had the SFS set to 98 in each.

Hmm ... strange. I would say the mid just got (much) better. But then I may be used to "something" for several months already ? (meaning something quite different from SFS stuff).

Btw, think about the x1 setting (so 01, 11 or 21) allowing for a much larger SFS (Split File Size). I am not saying the various sizes should matter, but I found the setting 40MB under what it takes for maximum (260 -> 220 in my case) not showing any of the "there's something on top of it" I had before with very low and very high settings. But maybe I'm dreaming something.

Also, you may try Scheme-1 with 0.9z-3. It really seems better now than Scheme-3 I always use.

:scratching:

First impression, testet both with the same settings I alwas use.
The z3 sound is darker. This could have two reasons. 1.Bass improved. 2. Highs decreased.
It seems mids and highs are quieter with z3. Something is missing. Bass und lower mids are dryer, have more punch and are more present with z3. I like that. z3 sounds a bit more focused as z2. I feel overall the sound has improved. THX Peter!
Still prefering scheme-3.

Darker would be my description. Or what about more "brown" for bass. To me this gives more reality (and variation) to basses. But I guess this is personal.

Jep, Darker. That was my first thought as I heard the first tones from z3. Agree!

You could try special mode and play with buffersizes and q1 value. It has a big impact. I tested z3 with the same settings as z2 and have also a loss of highs perceived. I tried some different buffersizes and q1 values. The sound changes very much depending on that. Bevor z3 I had buffersize 64, q1=1. With, for example, buffersize:32 and q1=6 I get very well defined highs and little less bass. I have not found the perfect settings yet but I think it is getting better than z2. Even in the highs.

So, for me it is done. z2 is ereased. Settings changed from buffer 64, q1=1 to buffer 32 and q1=6. The sound changed so dramatically with these settings, never got that before.  Now it sounds really great! Very fast and clear highs! Like a magnifying glass.

Peter,

With this version should we set SFS to high values? I have been using previous versions with SFS=100. What is the legend here - higher SFS means better SQ? Or what?

Ivo

No legend here (yet). I use 220 though, that being 40 under what my system can have (260) for playback. I think it sounds better when you're not near the limit. SFS.ini = 11.

My initial impressions are that higher values sounds better, fuller. Scheme-1 works better too, as Peter mentioned. Scheme-3 is too smooth at the moment, while Scheme-1 is more resolving and lifelike. I won't update my sig yet, want to experiment more with the settings.

Hi Peter,

you have the great advantage to listen with the NOS1 and we are all playing music on total different systems, so everybody has to evaluate the SQ with his previous settings.

SFS works best at 25 on my system, very fast response. 180 is very smooth. (engine 3, schema 3, Player prio: normal, Thread prio: realtime, Q1=5, others at zero. Firewire connection)

All in all: best sound I've ever had on my system.

Joachim

Hi Guys,

Tried out 0.9z-3 yesterday evening, my impression of the SQ is a bit mixed. Low bass has lost its tightness & has become a wee bit loose/diffused. Lower mids have improved significantly, more clear & softer, but the upper mids have become harsh. No significant change in the highs.

Peter: Tick sound at start of playback is not solved.

Arvind

SFS.ini = 11
SFS = ~200

Try these

Tried SFS.ini=11 but it takes too long for playback. Maybe my laptop is very slow.

21 will be faster (this can really matter) and could even bring better SQ. I never tried it.

Finally I got 9-Z3 installed after I finally found my Vista disk and reinstalled window and got RAMdisk started only a couple of days ago. All I can say is WOW.
Very impressive and it could not come at a better time. For the last week or so, I have been auditioning the new Qsonix Q205 thinking that it would be nice to have a server with nice interface and since Wadia now join force with Qsonix and designed all the digital circuitry in Qsonix I think that it would be a good alternative to computer server. Qsonix is very nice and the sound quality is excellent, pretty much on par with using CD in my CD player. (I used Stealth Sextet cable from Qsonix to Playback MPS-5 vs using CD in Qsonix itself). Z3 in Win7 was a touch better over all (using Lynx AES16, Cardas Clear cable) but Vista/RAMdisk/Z3 left Qsonix in the dust.  So great work Peter, also you just saved me lots of money except now I think I might have to upgrade my computer to run Vista 64 bits so I can have more RAM and bigger RAMdisk, perhaps.

Phwew, I just looked at those prices. :swoon: Well, very nice (and honoured) that such a thing can be surpassed with 72 euro eh ?
Maybe add a little tiny DACy now ? hahaha

Thank you for sharing and the nice words !
Peter

Did you tried RamDisk in W7 too?

[GRRRR... now I remember what I forgot when I was "at Home" yesterday...my Vista installation CD :smirk:]

Hi Peter/Marcin,

Tried 11 & 21 with SFS 250, in both case no playback at all. Waited upto 2 mins but nothing.

Arvind

Arvind, to get it straight ... What you are saying is that you can use 0.9z-3 allright, but only with SFS.ini = 00 or something (second position = 0) ?

If that is so, your laptop must be way way (waaay) too slow indeed. But you can check it :
If you bring up TaskManager, is one of the cores (or the only one you may have in there) at 100% continuesly ?

Peter

Since this is the SQ thread I will post some first impressions of 9z3 on XP.

Please note that any finding may relate more to limitations of my system than to XX. All listening has been done with settings as in profile (IdleEnable:1).

SFSfile setting: 11
What I like: Fantastic voices and individual instruments floating in the air, good instrument separation, excellent flow and vividness in the music, can listen long sessions at low sound levels.

What I like less: somewhat vague and blurred bass, a little disorganised (too lively), @ craked up volumes unbearably hard highs and thin sound, some 'heavyness' or 'darkness' around instruments

SFSfile setting: 21 (comments hold for all 2x settings)
Compared to 11: Much more gentile presentation (less distortion?) aslo less 'dark' and 'heavy', more focussed and smaller, easier to follow the singers text and individual instruments, in general easier, yet less compelling to listen to. Didn't try this one @ high loudness.

SFSfile setting: 10
compared to 11: at low volumes it sounds more organised with the same drive, tad better transient handling (ie hearing the skins of the drums), magical voices, audibly more information in higher registers (ie cymbals), works best at low volumes: even less bearable when played loud in my system.

summary of firt impression: 9-z3 sounds great for late night listening (use 10)

A great improvement is that copying music to special musicfolde on RAMdisc first makes no sense anymore (tried that!). If the copy to XX folder tracks are removed after playing, the playlists may be elongated to their orinal size, that would be excellent!

Thats it for now,

regards, Coen

Opps, I meant to say Z2/Win 7, not Z3/Win 7. I have not tried RamDisk in W7 but I can do that since I set up my computer
to dual boot anyhow but let me just enjoy listening to music for a few days first :)

Definitely, this is the best 72 euro I spent on my system :)
Now, only if I can get better interface :) :)
Now I wonder, once in a while, I get a little hiccup when playing through RamDisk (4GB RAM, RamDisk 1500mb- with 1GB RamDisk, I can only play 4-5 songs continously at a time with 4x AP, SFS 200, 1.5 GB is a bit better). I wonder if more RAM (using Vista 64 bits), 8 or 16 GB with larger RamDisk (4 or 8 GB) might make things run a bit smoother? 

Most probably the hiccups go away if you lower the SFS. So, your 200 has to be processed within a certain time, and if it can't do that you'll have that little stall. Well, I assume it is that. But you can easily try it.

I don't see where more memory is going to help here. Yes, for a larger RAMDisk allright.

If it helps, but you want to persist on the SFS of 200, try to play with the other Appointment Schemes, and also lower the Thread Prio a tad. Or increase the buffer sizes. -> try to give that loading process some space.

(A Guide to Glitchless Playback (http://www.phasure.com/index.php?topic=1356.0))

Would be great if you could check w7/ramdisk/z3 against vista/ramdisk/z3.

@CoenP: It is the first time I played with these settings. I can hear what you describe. 10 I would describe as  softer. 21 I would prefer a bit because it has more dynamics in the mids.

But the neighbours have craftmen in the house and they are drilling all the time , hence listening is a bit difficult at the moment  :unsure:

Yes, lower SFS certainly make the little hiccup happened much less often and pretty much disappeared at SFS of 60 or less. However, I still like the SQ
with bigger SFS. The sound is a bit fuller, more body and somewhat more analog to my ears. Right now buffer size is set to 1024 at both XXHighend and Lynx,
appointment scheme 1, I will experiment with thread prio and appointment schemes.

Hi Peter,

9z-3 works fine with SFS.ini=00. With 11 or 21 no sound. CPU usage with 21 peaks to 17% then comes to normal level (around 2%)

Arvind

You mean : and after that (CPU being at 2%) nothing happens anymore ?

How long does this 17% approx. last ?

Is there a message "Engine#x didn't start wihing the expected time" ?

Nothing happens; there is no sound even if i wait for 2 mins. The spike on the CPU is only for a fraction of a second & there are no error messages. Everything happens as with SFS.ini=00, except there is no sound.

Arvind

Playing for an hour like this (SFS.ini = 21) but if this isn't true I'll go to some doctor tomorrow.

I'll admit I had a theory while developing it, but I really never tried it.
At least it loads the first playback faster (3 or 4 times of what I recall at measuring it back then).

Peter, could you schedule an appointment for me too? haha

Indeed, SFS.ini = 21 is better. Great dynamics!

Peter,

I think you have really pulled a rabbit out of the hat with z3, I cannot stop smiling whilst listening, many thanks.

Marcin,

I agree ini setting of 21 gives terrific dynamics.

IMHO I agree with earlier posts that ini 11 sounds clear and tonally staifying but is a little down on dynamics and slam in the low and mid range (still fantastic great overall though). Ini 21 is very dynamic I think system tuning effort will concentrate here for now.

I haven't posted on z3 so far as I have been recommissioning my main system after 8 months of being mothballed after my house move. It's a bit disorientating right now trying to pinpoint what components are contributing what but z3 is absolutely shins through a NOS DAC, Gainclone amps and horns speakers.

Nick.

I need to ask a stupid question regarding editing SFS.ini file ( 11, 21 etc).
I have absolutely no idea how to edit that :(
I suppose I need a little remedial help on this.

Thanks

Hi,

I also had a little problem finding that but  you can find the .ini file in the folder of XX. Open that file and on the top change the 11 in 21 or something else.

Thats the trick i guess


 :)

I will assume you are using Vista. Other O/S's will be very similar.

Highlight the SFS.ini entry in the XXHE folder using the left mouse button, then using the RIGHT mouse button to bring up the context menu, choose the "Open With..." option. You may be presented with Notepad in the list that pops up already, but if not, in the section underneath called "Other Programs" you should find it there. If you see nothing in that lower section you need to click on the little arrowhead to the right of where the text "Other Programs" is written.

If you still can't find Notepad , then you can use the "Browse" button to navigate to where it is located, which is normally in the main "\windows" folder.

If you tick the box labelled "Always use the selected program to open this kind of file", then next time you want to edit the SFS.ini, you need simply double click on it, and it will open in Notepad automatically.

Give that a try.

Cheers,

Russ

Hi Suteetat and others

An easier way to find notepad and many things, just start to type the name in the search box (Start button) and hit enter, or select it from the list that pops up. The quickest and easiest way to many things, particularly windows programs and things that live within the environmental paths.

Chris

Thanks all for your help. I used notepad to look inside ini file before but was not sure if I could just use notepad to
edit or if I need a specific editor to do edit it. I guess that pretty much answer my question. Again, thanks all for your help.

Hi Arvind,

If you are in the mood a bit, please post the X3 and XX log files after you are waiting for a minute or 2 with the 11 setting. Maybe I can see something in there.

Peter

I'm in a bit of a jet-lagged state, but I really like the sound of both 11 and 21. Right now, I'm leaning towards 21... but I haven't played around with SFS values, or with the new 'Do not Start Playback at all' feature.

In any event, 0.9-z3 sounds really, really great! Thanks Peter!

Mani.

Thanks! Peter for the 9-Z3 version. Sounding Good. I had a chance last night to try different settings and so far I'm liking SFS.ini at 21, 11 not too bad either but more towards 21. I am using Special mode with Scheme-1 and the SFS at around 70. CPU is running at 2-4% and the memory usage around 1.2 gigs. Bass seems tighter, better vocals and soundstage running deeper a bit wider now. Will want to try some other settings over the weekend. I do have some issues with some of the music sounding a bit harsh in the treble region but probably do to the cr*ppy recording, AC line noise, cheap dac, room acoustics, and not enough brew, otherwise the best sounding version to date.

Peter

Question do i still have to tick *Copy to XX drive*? Meaning load into Ramdisk.

 :)

Hi Peter,

I am attaching the XX files with SFS.ini=21 & SFS=50. If I go higher than that the laptop stops responding. Hope it helps. X3 files are not created in the Temp Data folder.

Arvind

Uhm Arvind, there's no explicit log entry for it, but it sure looks like you've set "Don't Start Playback at all".

?

Gerard, Yes. That is, if XX recides on the RAMDisk as well, otherwise it doesn't help (for copying music files to the RAMDisk).

Flecko, I tried RAMdisk (1500mb) in Win7 today with all the same setting in Z3 for both Vista and Win7 (SFS 200, 11) scheme 2, buffer 1024, 4x AP,
Vista is still clearly better. Under Vista, the sound is smoother, more liquid and for a lack of better word, more analog, in my system.
Under Win7, the voice is thinner, a bit more edgy. I used to think that with Win7, the sound is still pretty darn good and already surpass using
CD in my CD Player/DAC a bit, now it is even better.

I like so far 21, Special mode, Scheme-3, SFS=200. If SFS=100 then sound is much thinner. I am on W7.

Peter: "Do not start playback at all".
I do not see any FLAC files being copied from my external USB drive to the XX folder in Ramdisk just before the playback. All I see are the same wav files appearing there just like it was before. So what is different? I guess this is a needed fxality, this is which brought the increased SQ. Source music files should be in Ramdisk just before conversion starts.

Ivo

Ivo, it is really beyond me what you don't understand here. I mean, I layed it out in the Release Notes, I answered your same question earlier, added to this all seems 100% logic TO ME, but you keep on repeating the question ...
... instead of explaining to me what you think is wrong with it.

One more time :

The FLACs are converted and THEN go to the RAMDisk. What is wrong with this in your opinion, hence (!) what would be better if the FLAC goes to there ? My idea : nothing, and things will only be worse.


What is different ? well, I guess that you didn't get the whole idea of "Do not start at all" in the first place ? I mean, if you guys start to shout that copying the files yourself to the RAMDisk and next start playback (instead of XX copying them) sounds better than the latter, I create this for you. And so it now works like that. One difference : it's not the FLAC being copied but the result of it (the WAV) so again less has to be done after you press play at last.

If you really think that copying the FLACs first is better, well, think that but I am not going to make it (because it makes no sense plus you never tried THAT for SQ difference).

Now, respond what you want, but please don't ask the same question again. Ok ?
Peter

Hi Peter,

Yes I have "dont start playback at all" & "copy to XX Drive" on & "start playback during conv" off. Am I doing something wrong here?

Arvind

No Arvind, all is perfectly right.

But do you expect playback to start with that "Do not start Playback at all" active ?

So, inactivate that, and I guess all will work. And of course this has nothing to do with SFS.ini is 01 or 11 or anything.

As an alternative use (AutoHotkey active) Alt-P to start playback. That's what that setting was made for. :yes:

Peter

Hi Peter,

Ok I made "Do not start playback" inactive, "start playback during conv" inactive & left "copy to xx drive" active. Set SFS.ini=21,SFS=250 & started playback using hotkey. Instantly I get a msg popping up which says "out of memory exception".

Arvind

I'm not saying it's related, but now you did it wrongly the other way around ...
How can you start playback with Alt-P if playback should start normally ? (assumed you pressed Play at some stage ... and if not, there's nothing much to play with Alt-P).

So, leave your settings as they are, and just press play. Preferrably even without having Unattended active.

Still out of memory ? then lower your SFS (again, you may start at 12 and then higher it).

"Out of memory" means, well, out of memory. What uses memory ? SFS.
8)

Peter,

testing Windows 7 64-bit, 16GB RAM and using ramdisk (8gb), I found the best combination
- 11 (I tried 10, 21, 01)
- SFS 220 (tried only 60 and 220)
The difference between 21 and 11 is subtle, but I like 11 because it is more fluent, more feel-like-being-there, heights are more liquid, and everything else is also better. The biggest damage to the sound introduces second 0 in ini file.

Congratulation, XXHE is even better. I did not expect this, really.

Jarek

For old time's sake I put 09y8 on the Ramdisc. My PC has been tweaked since the last time I used it, so it would be a nice indicator of the 9y -> 9z progress and maybe give a clue to the hard upper mid/highs I am experiencing.
Consider this my second impression of 09z3.

Well, no wonder I was a little sceptical when the first 9z version came out. What a great rendering of detail, completely natural highs, powerfull (yet a little fat) bass, more fleshed out instruments, excellent transients. In short a joy to the ears. 09z sound anorexic in comparison. (of course on MY system!!!)
All moonshine and rosesmell with 9y??? Unfortunately not. A thing that I noticed (as I did before) is that I started to do stuff and took the music for wallpaper. I some way it could not hold my attention and was left unmoved by its (beautiful) presentation. To be short: the music had lost its mojo. :sad:

Since I consider this a cadinal sin I went back to 9z. There it was again: the music-magic (not soundmagic), along with the rough edges, spoiling a bit of the fun. Time to explore the tweaking and finetuning further.

I started with the SFS.ini settings loking solely for the musical x-factor. Actually the worst souding setting (10) made the music everytime a party. This would make for a fine startingpoint.

Since 9y was played with higher Q1 I upped it to 6 in 9z3 and this did ameliorate the hardness to some degree. For some reason the SFS setting does not make for any difference (as it did with 21...). There are still a lot of settings to explore, but it can be safely stated that 9z3 is the most musically enjoyable version so far.

Still searching, but in the mean time enjoying the music :)



 

Hi Arvind. Good. But may I ask you : Isn't that clear from the Release Notes ? :

There's a new setting (Settings Area) "Do not Start Playback at all". If this is combined with "Copy to XX-drive by standard", and it mimics "you" copying the tracks (files) to the RAMDisk (assumed you use a RAMDisk) in advance of playback, and next select the tracks to play from there. When AutoHotkey has been setup properly, you can use Alt-P to start Playback.
This is for Unattended Playback only.

If you don't follow the forum I'll admit you won't know why this is. But some people (who tried it) perceive better sound when they copy their music to the RAMDisk themselves. So this mimics it.

Peter

Hi Peter,
I'm happy not everybody gets this 'Don't start playback at all'.
I assume all the work is done buy XXHE, so why don't start automatically after the work is done?
Or do we have to do something ourselfes to help XXHE?

Hi Gerard,

As said, if you don't follow the forum, you won't get this.

Just make that setting inactive and it won't bother you.

And it won't help you either, haha.

Peter

I follow the forum but then I must be the stupid one in the class!

What about this post (where you responded to) ... Not read ? can't understand it ?
Please eleborate ...

Yes I read this, but if I select the tracks to play in XXHE then what happens after that?
I can only imagine I have to push a button after I finished selecting the tracks...
And then XXHE can do its work..?

Aha ... (thanks)

So, assumed "Do not start Playback at all" is active, what you would do is :

1. Get the tracks you want to play in the Playlist Area (no matter how);

2. If you want, select tracks in there you want to play (select nothing, and all will play starting from the (always one) one being selected);

3. Click Play (all tracks will be prepared for playback now) and wait until XXHighEnd disappears;

4. Wait as long as you think is necessary to let things cool down, get voodoo out of the way, or whatever it is which may make you think that copying the files to RAMDisk outside of XXHighEnd will benefit SQ; notice this copying just has been done by XXHighEnd, but only that copying happened. We don't let XXHighEnd play, or otherwise it doesn't mimic the "copying ourselves".

5. Now use Alt-P to start playback.


Now, does this make sense ?
Maybe. It is similar to crazies who perceive different sound from FLAC vs. WAV (with XX !!). This, while the WAV coming from the FLAC is 100% equal to the FLAC not being there in the first place. One difference : it is a "close to playback" process, never mind all is done in advance of playback. But start Resource Monitor and you will see what is all going on just after you finished with #3 above. Really dozens of (I/O) processes are there, but if you wait a minute or two they have all vanished. So, the difference with immediate playback is that those processes will have vanished when Playback starts (at your command).
To be honest : I don't think those processes themselves matter much but the "organization" of the OS around it all does (ok, may).

To keep in mind : At using Alt-P it still be so that many things start to happen in the OS, because it is still XXHighEnd doing it all. But at least the "heavy" processes of conversion, copying, are out of the way now. Later I can make this more lean by means of starting XXEngine3.exe directly. Currently this won't work because of the organization of affinity and all is within XXHighEnd.


Does this help better ?
Peter


PS: I use it all the time. :)

Yes, now it is perfectly clear, thanks!
Maybe it's the voodoo that blocks my mind!
Now I wonder if this waiting makes the music sound better for the complete album or only at the beginning when everything is not settled down?
In the latter case I don't mind, but I'll have to try now to hear the difference.

Hi Gerard, I do it this way and it works:

In Unattended mode.

Click "Copy to XX-drive by standard", click also "Do not Start Playback at all", then go to the window on the right side and click Explorer, look for the album you want to play and select all the songs or the ones you want to play and click open. Return to the left window and you´ll see the album in the Library. Use Alt-P (previously you have to have AutoHotkey working), then you see that there´s is no sound. Click Alt-X to see again the player. Turn off "Do not Start Playback at all" and click again Alt-P. Now you have your music sounding beautifully.

I read the forum same as you but quite often don´t understand it. We are mostly music lovers, some are computer lovers also but not the majority of us. So please some patience.

Hi Juan, thanks for the tips.

I read the forum with a lot of interest, but don't have the time to try everything.
Like most of us I'm always happy with a new improved version, but most of the time you have to start all over to find the best settings.
This is part of the hobby and the fun, but when you don't feel like it or don't have the time it's nice to just turn XXHE on and enjoy the music.
I had my ears bugged for the flu for more then three weeks so I need to catch up with the way it sounds too...
Maybe time for an Ipad, read the forum on the Ipad while you're listening to the music and try all the settings.
Then make notes of all the settings and changes.
And maybe for all the users a page where they can put their settings so we can compare easily with each other.

I'm sure you don't mind : :rofl:

Let me first notice that you use Alt-X to bring up XX again. That's AutoHotkey, right ? Ok ok ...

So, once you have that going anyway (AutoHotkey), just click Play the first time, and use Alt-P when XX has vanished.

But each his own of course ... :) :)

Yeah, i dont use autohotkey for instance :D

SFS:21 is quite a class!
The sound difference between 09.z2 & 09.z3 (SFS:21) is not subtle.
It is as much as what ramdisk had inroduced.

- 09.z3  (SFS:21) has introduced a level of transparency not in the earlier versions.
It feels like a veil of cloth has been removed from the speakers - Very engaging and the music grips your shoulders.

- The tick at start of playback remains. But it has become a comforting tick to indicate that all is going well. :grin:

- 44.1KHz tracks upsampled to 88.2 or 176.4 (PeakEx + AP) need a different device buffer & Q1 setting.
This was not the case in earlier versions.
Track with repetitive drum (bass) was used to check this - whether bass was tight or brrrr...

Native high res. & (88.2/PkEx+AP) - Device setting 256 : Q1:8samples
176.4 PkEx+AP - Device setting 384 : Q1:6 samples
Seems like a tiny change but the SQ difference is not as tiny.

I don't use "Do not start playback" and also high SFS size sounds best e.g. 175 - 220.

Raj

P.S. Installed W2008 for 09.z3. W2008 is rock solid in performance. SQ remains consistent throughout the playlist.
You already anticipate this during setup (very lean) and some of the (Death tweaks) are already done.
Shutoff network, wireless, unshare IRQ etc. and it is what a music machine should be close to.
My Vista setup (non-Vlite) had inconsistentcies in SQ even during one track in the playlist.

Thanks Peter. No I don´t mind. I appreciate your help once more but admit that I was very close, or not?.  :whistle:

I like a sound as much natural as possible with clear voices, tight and deep bass, piano harmonically rich and some kind of sweetness in the highs but not too much. This is my humble contribution to the best sound found so far for 09-z3:

SFSini-21. Engine 3. Adaptive. Buffer-1024. Scheme 1. Low-Real. SFS-52. Arc Predict. fx x 2. Q1:1 other Qs:0.

Thank you very much for checking this! I will try a second Windows install soon.

I'm continuing to like SFS.ini = 21, Scheme 1 and an SFS size of 200. I'm not sure if the 'extra' dynamics are artificial and the sound a little too forward, but I like it.

However... I really, really like the sound of 21 and Scheme 1 with "Do not Start Playback at all"!!!! I was a total skeptic at first, thinking that "Copy to XX-drive" is all that should be necessary and that any other copying would be a waste of time. But not so! Engaging "Do not Start Playback at all" brings a welcome smoothness to the sound with 21 and Scheme 1. The only downside is that I invariably forget to press alt-P...

As with many of you, I will leave my sig as it is for now, until I'm convinced this is what I like.

Mani.

Hmmm, looks like I am the only one who actually prefer 11 over 21 right now for sfs.ini .
I have not really listen to 21 much. The first album I listened to 21 was Mathias Goerne Schubert Heliopolis.
It sounded louder, perhaps but some of the analog like smoothness is gone from the voice.
Switching back to 11, the voice is smoother, creamier, richer, take your pick :)
Piano sounds less edgy and metalic also.

I'll try 21 on some other types of music and see. I think we need a button in XXHighend
to switch between the two :)

This is getting complicated...

I've just switched to listening to my speakers (as opposed to my headphones, with which I've done all my listening recently, because I just love their sound). Through my speakers, I can't listen to SFS.ini = 21, or indeed 11. They're both just too much... too edgy (a bit like having the SFS set too low with earlier XX versions). And taking the SFS value as high as I can doesn't seem to help with 11 or 21.

So, with my speakers, I'm back to SFS.ini = 00 and the SFS value set to ~65. Also, curiously, I quite like the sound of Scheme 4 - very 'organic' and laid back... a bit like vinyl.

But isn't it amazing that we're discussing this at all? I mean, so much affects the sound, even in the 'bit perfect' realm!

Mani.

Well, this is all system dependant and I mean OS too :) Now with my slimmed&tweaked W2008 I have the best results with SFS at 12 - nothing edgy, harsh about the sound. That was not the case with 'bare' OS without optimisations, period. It's so good, that my pc-audio with Cantatis sound card sounds better (IMHO) than dCS Elgar Plus DAC... (in my system) This is not a joke. dCS is very detailed, but sounds dry. My server with Cantatis onboard is more lifelike, organic, engaging. I didn't know it's such a freaking good value! The dCS' price 4k EUR (used), the look on my friend's face - priceless.

Greets,
Marcin

My new findings:

SFS.ini-11/#4Engine/Special Mode/Buffer 2048/Scheme-3/Low-RealTime/SFS-178/Do not start Playback at all/Copy to XX-drive by standard/Q1=6 (1024)/ArcPredict/fx x 2.

I was using SFS.ini-21 but at the end it´s too rich no matter what you combine it with. On the other hand Scheme 1 makes the sound a little dry and although you can get nice sound after some listening time there is a lack of emotion.

Once all the other parameters are set I´ve made the fine tune with SFS. Surprisingly SFS gets more focus as it goes higher until one point where it starts to be fat and the voices are out of focus. So in my opinion SFS must be as high as possible until the point the sound is blurred, mostly the bass. Higher SFS also gives bigger soundstage. When I reach SFS-178, the balance point in my system, if I change SFS.ini 11 to SFS.ini 21 the voices and the scene gets too close of my listening seat.

I like the 09-z3 sound more than the one I got with 09-z2, bigger soundstage, fuller sound, harder to find a tight bass but possible, on the other hand perhaps a little short of the last word in definition but probably I could work it out a bit more...

Wouldn't it be much easier to provide bass, treble and loudness controls ? ok, with balance as a bonus.
Hahaha

But it seems to be true that at least we can get back the forwardness a little which disappeared with Vista largely. I noticed it too, but like the "level" of it. But as I noticed earlier : too much goes along with roughness (or edgyness) and the other way around.

It also seems true to me that if one thing is "problematic" it's the bass. And again I have the feeling that Engine#3 should be involved more (to find the optimum). It never works out well when I try, but still ...

Juan, thanks.

Peter


PS: I'm not sure whether I must be disappointed about the SFS still influencing, or whether it's some (not all that) fine tuner of which I wouldn't be able to find that "one" good setting anyway ...

This has been largely my experience too, though I haven't tried Scheme 4. I also find that Z2 is generally fuller sounding, but less forward, than Z3 set to 00. Mani, have you compared your current Z3 setting to your older Z2 setting? Would like to see what you think about that comparison.

Gave Z3 a good listen again last night. On many tracks it did have that "removing of a veil" quality but, after a while, like the initial impression of setting the SFS very low in Z2, it started sounding too thin. I then went back to Z2 with a sfs of 100 and received that fatter sound that is so pleasing to me when listening to the women's voices (say what you will Peter, its okay  ;) ). I want to go back and listen to Z3 again with the ini set to 00 and compare it again to Z2.

Another thing that I've noticed about SFS is that I don't have to change it as often as I used to when I had the laptop. Changing to a desktop seemed to make its effect "deeper", more consistent with varying types of music.


From the beginning SFS has seemed to me to be that "fine tuner" that would be influenced by everything, like even the quality of the mic used to make the recording, as well as the range and type of frequencies particular to each singer. (Patty Griffin requires a setting way higher than any other singer I've listened to) Thus, I will say once again, lets put the adjuster for sfs on the face of the GUI for instant access. I wouldn't feel disappointed about it still influencing, Peter, because it may be a kind of fundamental discovery that, however you "solve" it, helps to keep your player better than anything else out there!

I'm usually slow at changing version and posting comments.

I did as hort A/B between z2 and z3 and the latter IS better, more pleasant overall. Both on the dararam ramsidk, only difference SFS at 90mb with z1 and 200 mb (11) with z3.
Loading the first track takes 3-4 seconds, a bit annoying for my SO, but worth it.

On the other pc i stick with z2 because i need to use it for other stuff when listening to music and z3 make it unresponsive at any track change.

I'm updating my signature.

After a lot of experimenting to find the z3 sweet spot I noticed the following:

Regardless of SFS.ini setting, the difference between 'copy to XX disc first' on and off is very small. Actually I like it better when off. This is very welcome with my limited memory (only XX on the RAMDISC).

When switching the SFS.ini or any other setting I tried, you have to optimise SFS each time. Sometime it is hard to nail down SFS between to high and to low.

I feel SFS.ini setting 00 has the most explicitly detailed and natural presentation of all, albeit sounding still a little 'thin' overall.

Still the best experience of 'not listening to the sound anymore' is obtained with z2, music from ramdisc and 'copy XX..' ticked on. There is subtile timing and dynamics information revealed that makes a connection to the music very easy.
Playing music from ramdisc with z3 (00) does not lead to the same results (unfortunately). Typical soundwise difference is that the instruments in z2 sound a little 'clothed' and less resolving compared to the 'nudity' of z3. It looks like it is a trade-off, for now I prefer the convenience of z3, being able to play long playlists.

The vsuite ramdisc (direct IO) seems a little smoother than the dataram one, quite a subtile difference though.

regards, Coen

My "final" setting is sfs.ini=01. 11 and 21 is to sharp and focused on the highs. sfs=150. I went up from Buffer32 q1=6 to Buffer=512 q1=1 to bring the highs down which were just to focused. I meanwhile tried adaptive but it doesn't bring that power in the bass and the defined highs of special. I now prefer scheme 1 too.

I usually play unattended with "Copy to XX_drive..." option but without (Do not start playback...") as I am not keen on the "Do not start Playback..." option.

In the release notes if you use "Do not start playback... " option then it is to be used together with "Copy to XX-drive...".

However, recently I have been playing unattended with only the "Do not start the playback..." option without "Copy to XX-Drive..." and this sounds really good. I would say "spiritual" in my system.

I have been playing with .wav files (44.1 -> 176.4 with AP).

It would be great if some of you can check this.
Hope to hear some feedback :)

Raj

Haha Raj - This needs a kind of reverse thinking;

So, what happens is that with playing .wav files (those only !!) you skip the part of copying to the RAMDrive (that assumed, also looking at your signature), you use "Do not start" which does exactly nothing in this case (no conversions because of .wav), and at Alt-P the track is not taken from the RAMDisk but from its original location (which presumeably is a normal spinning disk).

Well ... this is all the other way around (from RAMDisk thinking) of course, and the only thing I can come up with for an explanation is that the additional process of copying may influence (nevertheless you use Alt-P to start) - and which process/influence just is not there now.
I think this would be similar to those perceiving a difference between .wav and .flac, while .flac incurs for such a process again (very similar to copying), while .wav just avoids that. This difference should be eliminated completely when playing via the RAMDisk *and* using the Alt-P thing, which mathematically may lead to using .wav and doing nothing being even better. Still in that case the spinning disk is involved, which may matter by itself (not that I ever heard that, but which is different from an SSD being there, influencing on its own as it seems).

:wacko::wacko::wacko:
:)
Peter

9-Z3 is a breakthrough.

For the first time I feel I really get under the skin of digital music.
It's sooo organic from top to bottom, without sacrificing anything in regards of detail, image or dynamics.
-On the contrary, I feel the "organic element" ADD to the instrumental detail and microdynamics.
Bass sounds more analogue (in a positive sense!), more "true".
I see that some mentions "artifical" treble, but that is not the case in my system. On the contrary - it is just "perfect"!

This in comparison with 9-Z2.

Still I have not converted to RAM or Vista.

I have not updated my signature yet, but final settings tonight seems to be:
SFS = 140 [about 50 less than max]
SFS.ini = Default [I dont understand how to do... Pls someone spoon feed me]
Adaptive Mode/Buffer 1024
Scheme = 1
Dialing Q1 from 1 to 4 tided up the bass somewhat.

Thanks Peter!

BTW: How did the cabinet paint turn out? Are you on schedule?

Nice Pedal, especially if this is with W7 !!

The SFS.ini can be edited with e.g. NotePad (and which value does what is in the Release Notes I think).

Peter


PS: If all goes as promised the cabinets are here next Tuesday. Then I hope to be happy about the paint. So, never seen it yet.

Search in your XXHighEnd folder (in c:\program files\... or whatever you install XX)  for SFS.ini file. First two numbers are the key.
Chriss.

So, finally I tweaked my Vista until it almost died, but not yet :)

I noticed an interesting phenomena: the more tweaked the OS is the less I feel differences among things like SFS.ini values, SFS sizes, buffer and Q1 values.

Has anyone else felt the same?

My current preferences are:
SFS.ini=11, SFS=175, FLACs in Ramdisk (I still feel this the best option) Special mode, buffer=384, Q1=8.

Ivo

Ivo
If I understood you right...you prefere FLACs rather than Wav?
Chriss.

What Ivo will mean is that he prefers copying the FLAC files to the RAMDisk, and play them from there. This, compared to play them from their Original Location, let XX copy the .WAV to the RAMDisk and play (with or without postponing Playback after the .WAVs have been copied to the RAMDisk (and next use Alt-P Unattended)).

So what Ivo does, is copy the FLAC manually to the RAMDisk, let XX convert them from there to WAVs on the RAMDisk again (not necessarily, but he will be doing that I think), and next follow the same procedure.

If that helps for better sound it is time to eat a hat or something ...
:)

Exactly as Peter described.
Also, I feel that is not convenient to press Alt+P when XX disappears to start playback.

Peter: One thing from whishlist: Can you add a small feature that allows to shut down the PC after the playback is over. The use case is that I sometimes listen to music and fall asleep until morning, so would be nice that XX box also gets some rest :) Do not know why but my tweaked Vista has no Sleep function???

Ivo

Why your tweaked vista has no sleep? Try to add it, it's very useful.
I never shut down my dedicated pc, sleep all the time. This also solves the problem of the SLOW save image on shutdown of the ramdisc (which i have disabled).

Ivo, ... and that combined with a repeating playlist, right ? :toomuch:
Haha.

I'll put it on the ToDo list. :)

Especially Vista is wacky with all what is sleep, hybernate and screensaver related. But also it is "me" debetting there. I mean, in the code things happen which undo the normal settings - in an indirect fashion. And I don't know what it is really. Before we could be sure it was WASAPI related, and today ? well, pieces of WASAPI code are still performed, even if KS is in order.

In the end you can wonder what a shutdown really does more;
If the screen nicely goes to sleep (and that always works here), and the disks are set to spin down, well, the cpu is idelling (which is just running at low speed), and there's power on everything. Will that hurt ? Some say it hurts more if poweron is performed every day (peaks).
I never shutdown any PCs and including those at my normal work, these are dozens, running for years and years and years until they are too slow for current life, and will be replaced.

Peter

Actually I did not notice I tweaked it out, under which section is this feature if looked from within vlite?

Ivo

Well that could be only in VGA with dedicadet GPUs. Older VGA that is sugest to use in audio servers are with lower RAM stile using main CPU calculation for each operation.

I just wanted to let you know that playing directly from RAMdisk (copy tracks manually) vs. 'Copy-to-XX folder' and 'do not start playback at all' is clearly the winner - sounds better and I have no idea why. This is probably the biggest enigma to me...

Marcin, just WAV, or FLAC too ? (meaning XX still has to do something with the latter)

Hello Marcin,

the streamer "Purist HDR6 DA" made by www.ami-hifi.de/ is using a RAM disc too, and it is for them an enigma also. They have no explanation for this effect.

Best

Joachim

So it was not placebo when I discovered that.  :holiday:

It's already in my mind : "No".

But now : what will *I* be able do about that ... ? :scratching:

It depends a bit on the answer from Marcin to my last question. But Ivo, you can answer it too (maybe you did already earlier) ...

Sorry for my late reply, I wanted to double check and I wanted my girl to be present, so there wouldn't be any mistakes. To answer your question, it matters whether it is FLAC or WAV and, on top of that, it matters if it was copied manually to RAMdisk or using XXHE built-in mechanism. The order of SQ is counting from the best:

1. WAV copied manually to RAMdisk
2. FLAC copied manually to RAMdisk
3. WAV copied to RAMdisk via 'copy-to-xx folder' & 'do not start playback'
4. FLAC copied to RAMdisk via 'copy-to-xx folder' & 'do not start playback'

This is no placebo! The difference is to big - more details and amazing smoothness at the same time if you play a WAV copied manually to RAMdisk.

Hihihi...well who is crazy now. I told you Peter , WAV-BETTER SOUNDING THAN FLAC! I heard it on my FLASH! You can try it too. Marcin right now I'm trying Q!= -2 in special mode and 64 samples and it's really nice :) But I have some stability issue.
Best Regards, Criss.

You !

Laalalalallala
:)


I know. :party:

I'd theorize this could simply be another manifestation of a more fundamental issue we also discussed in SFS thread last week: namely, it could all be related to RAM fragmentation.

But, hey, talk is cheap - wouldn't it be great if we could somehow test this and let everyone make up their own mind?

To that end, I am attaching 2 small test programs for anyone curious enough to try!

BTW Programs are based on publicly available sample code provided by Microsoft which means there's nothing fancy there: just super-minimal console WASAPI demo. Although, I did have to modify the code a bit to make it 'Audiophile Approved' (LOL). In other words, all disk I/Os during playback have been removed: it's now a true mini version of a 'memory-based' player :) )

But the point is this: both versions are _absolutely identical_ except in the way they allocate RAM: One uses 'traditional' approach (well, maybe a bit more advanced than you'd see in Foobar, iTunes etc but no major differences) and the other uses Large Pages (which, to my knowledge, is not used by anyone, incl. XX so it's at least 'certifiably different' :) ).
So now you can judge for yourself (and with _your_ hardware/audio setup) if this RAM talk makes any sense at all!

Before you start playing though, we'll need to do some preparation work: by default, Large Page support is disabled by Windows as it is not something for 'typical' users/usage scenarios (and also removes precious resources from under Windows' control) so you'll have to enable it manually like this: (make sure you have Administrator privileges)

1. open Control Panel->System->Administrative Tools->Local Security Policy
2. in left panel, open "Local Policies" and click on "User Rights Assignment"
3. in right panel, double-click "Lock pages in memory": If your username is listed you're set!
4. if not (more likely), click on 'Add User or Group' and in popup click Advanced and then click Find Now: this will show all 'users' (most are 'system users'): double-click on one you are using to log-on i.e. one you will use to run test programs
5. close all dialogs with OK & make sure you appear in the list in step 3.
6. reboot (so Windows can activate our new policy)

Oh, since sample code is based on WASAPI you may also wish to play with MMCSS registry settings:
http://msdn.microsoft.com/en-us/library/ms684247(VS.85).aspx
But, if you're lazy, just download & double-click ProAudioKey.reg attachment ;)


To run test programs:
- copy your favorite song(s) into the same folder as attached executables
- make sure you use plain 16/44.1 WAV format!
- also, please make sure your sound card is set as 'default' in Windows and format is set to CD 16/44 (if your 'normal' setting is 24/96 or whatever)

Start Command Prompt (told you it's stripped to the bone - there's no GUI or anything here...) and in the folder where you have put everything simply type:

jplay1 --file YourTestTrackName.wav

Listen.
Then, in same way, try second version:

jplay2 --file YourTestTrackName.wav


Then try with some more tracks. Also try the other way 'round or try several WAVs with one and then other version.
Then try opening your browser, Outlook, anything that eats memory - chances are, jplay2 will report that it cannot allocate memory. Then try again jplay1 - it should still work but listen how it sounds. Then reboot and try again.

Finally: share with everyone if there is any difference between two programs!
Have fun!

PS. If your sound card does not like default buffer size (1024 samples) you can change it: just type program name without any parameters for instructions....

Josef,

cPlay uses Address Windowing Extensions (AWE) to allocate RAM to music files during playback (Lock pages in memory). I'll try your tiny players later :)

Cheers,
Marcin

PS
Why did you set affinity to 1?

jplay2 sounds better for sure :)

Starting playback from CMD, how cool is that? ;)
It's a pity that buffer below 256 samples isn't available. No stability issues so far with both versions.

Hehehe - you have not been paying attention to what I wrote above? :)

Large Pages is NOT the same as AWE: With AWE you still get fragmented memory! Not so here :)

To verify, I just downloaded cicsPlay.exe and I'm afraid it is as fragmented as any player - regardless whether it is in AWE or non-AWE mode...


Well spotted!
Mea culpa: I was playing with affinity settings and uploaded the wrong file - I have now re-uploaded it to the post! (it's now set to 2 as I only have a dual-core but you can of course change it)

As you very well know from your experiments affinity does seem to influence SQ and it also works with these tiny players: I'm just not 100% certain whether that registry setting works, however....
Maybe you can double-check on your 6-core audio-monster and let us know?  ;)

Wow - that was quick indeed! (but coming from you I'd expect nothing less :) )
So, 0:1 so far, let's see if we can get some more opinions....

I'll give this a go this evening. Really interested in hearing the difference...

Mani.

Josef,

Don't want to be picky, but shouldn't it be hexadecimal? (affinity)

?
It doesn't matter, does it?

I don't know :D I'm used to hexadecimal values as I spent much time with IRQ affinity tool...

You're getting geeky  ;)
No, Decimal or Hex does not matter: what matters is how it looks in binary....

Josef, wow! That is one clean sounding player! Well actually two, one of which is clearer better than the other. My strong preference is for JPlay2. And yes, as soon as another application is opened, JPlay2 struggles to play.

Marcin, Ivo (and many others I'm sure): yes you're all absolutely right - XX sounds much better when files are copied manually to RAM and then played directly from there, with 'Copy to XX...' and 'Don't Play...' left unchecked.

I compared XX Engine#3/Q1=0 (playing a file manually copied to RAM) vs. JPlay2. XX sounds fuller and more musical, but JPlay sounds cleaner. JPlay sits on my SSD c: drive, along with the OS. I tried to play it from my RAM drive, but couldn't get the command prompt to recognise the RAM drive.

Mani.

[All listening conducted on my Atom PC with RME AES-32, whilst my main PC is upgraded ready for the NOS1!]

Mani,

XXHE sounds more musical, because you use SFS.ini=21 and larger SFS. Set the SFS.ini to 00, SFS to 12 MB, Scheme-1 and try again :) You might discover the opposite. And make sure to manually copy the tracks to RAMdisk.

Cheers,
Marcin

Hello,

you guys have really good ideas.
The SQ with the copied files to the RAM disc is superior.

Explanation for this phenomeon. Less jitter?

best

Joachim

Eh, player was not meant to enter into competition with XX but rather establish whether RAM fragmentation can influence quality of music playback - However Mani, thanks for kind words! (btw I also find myself preferring it over XX in KS mode - weird...)

So, it's 0:2 now and I guess I can also give my score as cat is out of the bag now, which makes it 0:3.
I was hoping more people would post (like, ehm, Peter? :) )  so we could make this even stronger statistically (there's been 12 downloads by now).

But anyway, 0:3 is not that bad either and hopefully proves the point that RAM fragmentation can & does influence quality of music playback even though the code that is being executed as well as data streams themselves are both 'bit-perfect'.

Hopefully this ought to help everyone understand better the 'mysteries' of why various SFS settings and manual vs 'XX-Copy to disk' have an effect.
And I've certainly had some fun and also learned something new in the process - hope you did too!

Maybe statistically a bit weak, but a phenomenal find nevertheless Josef. Thanks for sharing.

Mani.

Yes of course, but you know ... I don't like A-B stuff in the first place, I'm quite busy with other (DAC) priorities at this moment, and I thought to accept the outcome anyway. Also, with that I don't like the A-B stuff, my outcome will not be reliable (I'd really need 5 days to listen to one situation, and another 5 to the other).

What I will do though, is incorporate it in 0.9z-4. In fact, I already believed in it before you started this (great work and great thanks btw !). Now all I need to find is something which mimics that copying. My current means apparently doesn't cut it, but what will ? If someone could tell me *that*, I'd build that in instantly as well. Could be a simple copy command at some stage, but where exactly ?
Also, I didn't take any time (A-B) for this as well, for the same reason as explained.

Peter

Thanks Josef for the super-clean player.
Have not been able to break it, running along browsers/outlook/delphi.
It does not appear too hungry either.
Are you developing it further?
Can we have a version supporting smaller buffer sizes please? i.e. from 32 up?

DAC is probably taking a lot of your time - I understand and wish you best of luck!

I don't like A/B stuff myself and also prefer looooong listening periods before making up an opinion - However, this RAM fragmentation thing really hit me immediately: I admit, I didn't expect such a difference at all and was completely taken aback! (a nice way of saying I've decided to post those players to check if I had too much wine :) ) To my relief, both Marcin & Mani not only confirmed my findings but both expressed strong preference just like I did - tnx guys:)

Anyway, whenever you have time please do try it. And, since you prefer longer sessions (and perhaps there are other people too) the limit of 1 WAV might be an issue: If you'd like to play multiple WAVs then simply copy them to same folder where players are and do this:

- in command prompt (in folder where both players & WAVs are) type: notepad jplay.bat
- Then copy/paste this line:
forfiles /M *.wav /c "cmd /c jplayer2 --file @file"
- Save & Close

From now on you only have to type 'jplay' and all WAVs in folder will be played (substitute jplayer2 with jplayer1 as you see fit).
Don't expect 100% gapless as WAVs are loaded one by one so there's an inherent delay but it should suffice for purpose of the experiment.....

Personally, since I experienced what contiguous RAM can do for PC playback I find myself troubled when listening to any player.....(incl XX)..... Yes, yes, I know, it sounds radical and I don't want to start a controversy here so let's just say it's how it seems to me at the moment, ok?  ;)

And that's why it's really great to hear you will be incorporating these findings into XX! (as mentioned in SFS posts last week it might be a bit of a challenge as RAM will have to be allocated all at once in a e.g. service but I'm sure all XX-fans will aprreciate it once it's there - btw one more reason to get as much participants for this experiment as possible!).

Goon-heaven: Thanks for kind words!
But oh, don't worry - it will break :) (only player2, that is).
Eventually, Windows will be unable to allocate enough large blocks since this little player always starts 'from scratch'...


I definitely did not plan to.
But, as I explained in my previous post, for some weird reason I'm now having trouble listening to other players so I may add a tweak or two before XX & other players incorporate this finding :) 

I think I understand rather well what would have to be done to make it go _all the way_ but it would involve some serious dedicated time and at this point in time it's just a hobby for me so no guarantees....


Ah, the proverbial latency mystery :)

Well, long story short, I'm afraid it's technically impossible :( (it seems that WASAPI-induced overhead is putting a hard limit on how low you can go.... btw, I'm really not an expert on this so maybe somebody can chime in?...)

Notice e.g. that XX does not even offer buffer size selection in WASAPI mode - there must be a reason for it :)
jplayer can go down to 256 samples on my run-of-the-mill laptop with vLited Vista without problems (not that I find it sounding better though!).
In fact, the original MS sample code I used was programmed to use lowest latency possible by default (as reported by WASAPI layer) and, in my case, it turned out to be, think 132 samples. Unfortunately, that sounded really bad: as in: totally distorted sound, pops & clicks + the feeling as you were playing an SP tape on LP speed (if you're old enough to remember analog tapes lingo :) ).

Of course, there's always a possibility of replacing WASAPI with KS and, actually, it may not be too difficult to do it....
Hmmmmm: WASAPI vs KS: Sounds like an idea for a 2nd experiment...?   ;)

I guess you meant:
forfiles /M *.wav /c "cmd /c jplay2 --file @file"


That's fair enough. I have to say that it's still too much of a WASAPI sound for me - clean but to the point of being clinical on my system. Some sort of KS implementation would be very interesting.

Mani.

Josef, please don't. I already have it in, but have problems of checking it. You told me the other day what to use for it, I said "oh" and "yes", but now where I'm up to it, it's hard to find. And also : I don't want to fall into the trap of looking at "contiguous" logical memory, as you told. If you think it is better to send me an email about it, please do. Maybe *I* think that is better. :)

Thanks in advance,
Peter

Ok - I'll mail you download link.

Yes Mani, my typo - you are 100% correct!

BTW - You mentioned that your RAMDisk is not allowing you access from command prompt - that is weird!
What RAMDisk is it?
Do you see its drive letter in Windows Explorer?

Not sure if this is helpful given all the "Josef" changes that will be made in Z4, but I recently gave Z3 another try and my original impression of sfs.ini set to 11 or 21 remains the same, though 21 is better than 11: very clear 'thin" sound; very similar to Josefs 32 bit player. However sfs ini set to 00 is a much rounder sound and, it seems, that there is no difference, or hardly any, in changing the SFS. I would say after a day or so of listening that its the most pleasing sound I've gotten from xx. I have the SFS set to 12. I think I can now change my signature.

After listening to Z3 for nearly a month, my wife started complaining about  harshness in the sound; after changing various settings in Z3 I tried Z2 again. The character of the sound is completely different from one to the other. Z2 seems a smoother, fatter sound but not as revealing or transparent as Z3. For vocals though Z2 again seems more pleasing than Z3......... :wacko: :wacko: :wacko:

Both were played from ramdisc with tracks manually copied there as well.

Would that not be a kind of "logical" when you're now using W7 again ? To me, the high frequency output of W7-SP1 is a kind of infinitely higher, and so will encourage for what you described. Of course it kind of will be about how your further chain will cope with this, but I certainly will agree that W7-SP1 is more "difficult".

So, about your :wacko::wacko::wacko: ... before you used Vista, and I think you can imagine that 0.9z-3 can work for the better there (compared to z-2). Thus, nothing strange ?

Before getting confused all over, please wait for 0.9z-4. Things will be so different that all pieces of the puzzle may start to fit suddenly. At least here I'm totally happy - and even with Special Mode (which really will "worsten" your description in the base !).

Peter

boleary, as we both have a hiface, you could try my settings. The sound I have is very clear and pleasent. The best I had so far. No harshness at all. I also use jplay in the background to reduce "noise" while playing xx. I was not so happy for a long time with the sound but at the moment it is realy good.

I almost started my last post with:"Have you heard Z2 on W7 sp1?" but I didn't want to overstate this finding. Am very much looking forward to the Z4 experience.

Adrian, I did try those settings. I also tried letting my dac upsample straight 16/44 from xx, hadn't tried that for a long time and forgot how bad it sounds. In my system arc prediction with peak extension just dramatically improves everything when playing 16/44. Straight 16/44 is just too hard edged and jittery sounding here.

If Z4 can capture the midrange width of Z2 and the depth of clarity/transparency of Z3, without the janglyness (new word for HF distortion  :thankyou:), I may never have to post again!

It is so fascinating how different two systems can sound, even if thay have the same important parts (hiface). I tryed z2 too with W7 and have it still on my ramdisk for testing. I would confirm your description of z2. I did not tryed it now, but from what I remember it has less details and a softer, fuller, pleasent sound with less definition in the upper highs and the lower bass. With z2 I was always completely happy but I think with my actual set up it is even better. Will test again if I got time to do it.
Greetings

You read my response to this, right ?
Well, let me do that one all over, because maybe I wasn't 100% honest. Or maybe what happened was too short ago to be able to judge myself. So ...

Exactly the day before your post about this, *my* wife started complaining. Actually I could just copy your post. I only did not try z-2 to compare.

After this, there has been some misery in our listening room, because my wife kept on complaining, and *somehow* I couldn't agree. I always do, but not so now. But there is more, and only now I can talk about it with some sense ...

It came to me that playback via W7 SP1 is not consistent. Not consistent over albums I mean. Actually very much like when I started with PC playback. One album is good, the other is not at all.
The "not at all's" have something like my ears are covered with some thick blanket. No highs. Dull. Flat. The opposite is what my wife complains about. Very fresh, but apparently too fresh to her ears, and resulting in harshness.

Of course I wasn't playing with normal z-3 as you know it, and slowly it contains more and more of more tricks. Yesterday at last all was in there so I could play around with settings, and I think I have found back the sound how it should be. I also found the reason why Vista sounds as it sounds, and I think that sound can be copied now (in W7-SP1). The other way around I don't know yet (didn't try).

If my observations are correct, it may well be that things are getting too complicated by now; somewhere there's this "super" setting, but who is going to find it ? And, how many will explicitly have wrong sound because of not applying that one combination ?

My point is this (well, it is at this moment) :

The OS is "contributing" much more to the sound than we like. But, the way it all works (I know now), has a "flattening" influence on that general influence itself. What we are heading for now (0.9z-4) is avoiding the flattening of the influence, and influence it far more directly. It can be totally wrong, and I guess it can be utterly good. It needs this guide I'm afraid, and the guide is not there (yet). Also, there's hardly a way back. So, how z-4 works, does not allow to let it work how it worked before. Not in one program.

In between lines, but a major subject for the near future maybe, is my observation of how bass can be influenced, and also how highs can be (no judgement about mids yet). It is the highs which worry me;
I have examples of so few high output that we all may wonder what ever can cause this. And my personal problem is : I think the less highs are better. That is, I *can* reason out why highs -before rather espressive- can turn out to be lesser afterall, while I cannot reason out the other way around (to be better). Something like : when highs of which you can proove them to be not there, can be there with another setting (or OS), it should be distortion. No matter we (or I) don't perceive it like that, it should be. This too is about the flattening influence, because when not there it exhibits things which are hard to accept : all albums sound right. So, if I listen to that blanketed albums from the past week (or two), it seems normal to me that those albums sound like that (it's about the recording year, the artist, how it sounds in general). A next album shouts from the speaker, and that too seems normal to me (it is louder, more compressed, more commercial, blahblah).

But do I like that ?

I should, because all theories are better.
But I don't.

If this is all true, and you all will be perceiving 0.9z-4 the same as I do, I really will have a problem. It will mean that I must start to apply illogical things to make the sound better again. My quest for the 1:1 has been reached and now turns out to be wrong, and now I'll have to find a way back. I don't think I will be able to. Oh, I could revert to the 0.9z-3 code and start again from there, but I'd still know that at making something theoretically better it will turn out for the worse. So again there it stops.

But ok, there is hope. Hope from "yesterday" where I seem to have found a setting which sounds good again. Hope that it will last for the next week.
Or hope that *you* can make something of it.

We'll see. I will put up 0.9z-4 soon now. A "beta", just to try it. 0.9z-4-0.
:wacko:
Peter

Hello Peter,
interesting problem seems to occur. Decreasing quality of signal for the benefit of better sound. I don´t think this is the right way; going this way consequent, it will end up with equalizers, enhancers and other stuff to generate the sound one person likes.
I think the way you went was correct, even for the prize that some recordings do not sound like expected or wanted. I think that an audio reproduction system should accurate reproduce the signal it gets, whether analog or digital (i know this is theory). What you get for this kind of reproduction is a closer step to the music the artist plays, nobody knows which intention the musician had about the sound of his production (take the last Neil young Album, distorted, not very smooth sound) and therefore the music lover "must bare" the bad sound but he gets for the prize of a more realistic music reproduction. And on a good system which is focused on correct timing you`ll forget very fast the bad sound. When both comes together, good sound and good music, well everybody is happy but what is more important and gives you more joy, Music or Sound?
The biggest advantage form Z2 to Z3 was a much better timing and accuracy of music reproduction and I hope this won´t disappear on Z4.

Chin up Peter, you are doing a fabulous job which we are all grateful  :goodjob:

Perhaps we need a big knob on the front of XXHighend that we adjust from "1:1" to "1:?" and we change it depending on what we like and what we are playing, perhaps if we had two they could be called "Bass" and "treble"   :whistle:

Brian

Here is where digitally tampering the sound does come handy to make more pleasant what horrors have been registered in the CD.
AP does a great job but sooner or later i was afraid that it wouldnt be enough. I'm not by principle against applying some filters in REVERSE mode of what has been done to the signal at the beginning (compression, but not just that), but I know that you were. :)
But I think that it has to be left to the end user what and where to adjust (i.e. as simply as a treeble DEboost), because in truth is album-based.

I'm also afraid that when all albums sounds great, there IS second harmonic added, be it by the OS, the player, or another part of the chain (that latter is ruled out in your system).

For the records, I use z2 with my studio PC, where i have a pretty bad OS dac and it does sound easier to my ears than z3, and it's more responsive, which i NEED when i'm working at the same time.

Lastly, remember the priority thing about W7? Maybe it is JUST THAT the cause of unpredictable SQ from one day to the other?
Or they fixed it in the SP1 (which will be retail 2-3 weeks btw)

Still z3 is the best sounding player for windows, by far. It don't think it was wrong at all. But I am curious what z4 will bring and understand what you are planing.

The point that not every album sounds good is no problem because not every album is good. Ok, it should sound acceptable and with z3 it does. So no problem. My actual setting gives me the most accurate highs so far and I don't think that is wrong. It is not overpresent but accurate. z2 seems nicer but it is also less accurate. It is the "analog effect". Softer sound, less details -> more pleasent. I think in the end, the 100% untouched way can't be done, simply  because we might get ahead in sound quality and listen, what some old engeneers could not listen back then.  So after it is done perfekt and untouched, it is allright to sound the system to make different systems sound good. But before the sound is altered, everything should be perfekt and untouched. And that is why I am think your way was right.

Hi Peter,

I did not upgrade to Windows 7 but are using Windows XP from the beginning until today. I picked up all the interesting things of the discussions and graded up carefully - RAM disk now, switched back to special mode and the last version of XXHighend. There was always a remarkable evolution of the music production. I did not play with the Q factors and was not looking for "analog" sound. I trusted in the efforts you made in digital software. 4 months ago I got a big LP collection, had to clean all the media, hear them and classify them for my own reference. After months I came back to XXHighend - and have been overwhelmed by the music reproduction again. You are on the right track - and I wish you all the best to move on!

Georg

This is what I was afraid of: total transparancy backfireing, laying bare all the good and all the bad. For many a recording, a tiny little cosmetics makes our listening experience exctatic, the naked truth might leave us only desillusioned. All we need now is a good visagist who can prescribe just the right dosis for our needs....

So far XX has done a terriffic job in satisfying my hedonic quest for an intenser repoduced-music-experience without having a totally transparent reproduction chain! I'm very curious where this quest into the digital unknown may take us.

Keep up the good work,

regards, Coen

Thanks for the heads up everyone ! really warming ...

Maybe there's just too many combinations to try for a single person with not too much time. Maybe I don't like it much that my wife is complaining while I am actually not (which never happened). Maybe the lot starts radiating hormones or something. :)

Last night, for the last time listening to the "production version" of 0.9z-4-0 before it goes up, I compared a few things, and it started with Vista vs W7-SP1. There's just a not-normal difference between the two. I used Shpongle (Are you Shpongled ?) and the detail from W7 was so much more that I had to check what I actually played (the !Played Playlist) in order to be sure I listened to the same track I left off with Vista. This difference is so crazily much present that somewhere something must be wrong to the highest degree.

We all must wonder : what are we actually listening to ? to what degree can it be compared what we perceive from it all ? And I think (by now) I can proove what I mean :

How many of you reported a tick in between track parts ? If my memory serves we right, two. Well, that was before 0.9z-3, and after that it was only in between tracks, and one person left (I think it was Arvind). However, when I shortened the SFS to 1MB and less, I heard the ticks in between track parts. You know, those ticks two people heard but which were gone with 0.z9-3. Well, they are not.
How large will be the coincidence that BertD here -right after receiving the NOS1 and using 0.9z-3 like you- immediately complained that there were ticks in between track parts - ticks which I didn't even hear with the SFS of 12MB for minimum. I mean, he, of whom I know his system and which for speakers might be better than my own. One thing : this was solved for me in the due 0.9z-4 and I never listened to 0.9z-3 with W7-SP1. Aha ...
But you do !!
Thus, go listen, know it is there, and possibly you will hear it now.

Like with the ever before not occurring brushes and things -now blasting through the room- yesterday it occurred to me that ticks in between tracks were a W7 thing. Oh yes, the ticks in between tracks are still there for 0.9z-4, and even more I'd say (or other things make them more audible -> read on). So, with Vista hardly audible, or maybe not at all. After switching to W7-SP1, they came to me as almost dangerous. That loud.

The Shpongle album (at least the first 3 tracks) contains so many high transient ticks, that I really couldn't make out whether this was just wrong (XX) playback means, or whether it was for real. Remember, I just played it on Vista, and nothing of the kind was audible. I liked it, and thought to play on for the remainder of the evening, but, I was "testing", so I had to switch to W7-SP1 as well. Well, I moved to another album, because it was just too much of it ...
A next album (Deep Purple in Rock) I switched off too right after the first track. John Lord's Hammond was just too much f it all. It doesn't roar ... it individually peaks with way too high transients. And it's not even digital ...
You see, if *that* is going to happen, well, what to do next.

But then I thought of something ...
From theoretical tests (say, looking at graphs during playback), I had switched off the RAMDisk a week or so back. In my opinion it can't be for the better (says he who eliminated the SSD also :)). It couldn't ... ?
But it is. I switched it on, moved back to Shpongle and moved back to Deep Purple. Listened both throughhout this time.
Presented my wife some familiar tracks and asked what she thought. "Still critical, but ok" was the answer. Turned the volume down by 6dB (played all extra loud before to be sure I could bear it), and a "yes, it needs that" was the response.

So, all back to normal ? maybe.
But maybe not in the end. As said, from theories the RAMDisk is no good. Or at least not how the RAMDisk technically operates. So, it should smoothen, and I guess it does. It is the same with native W7 where we all started with the RAMDisk. It helped the On/Off thing I wrote about. But now ? well, it does the same, but as far as I can perceive it, it now smoothens things which possibly ought to be there.
I don't hear really wrong things with W7-SP1 without the RAMDisk. But it's too transient. Or maybe for my further system it is ?

But there are so many other "dials" and combinations of them ... I all couldn't try. Hopefully we, all together, can. But don't get crazy of it, or let it influence your mood like I did.

Ok. I am going to put up 0.9z-4-0 now. Be warned. haha
Peter