Sent you a PM as well.

Indeed, as the "Holiday's" topic from Gerard implies, I am out of here for a while (until May 15).

As far as Internet conections allow, I will be monitoring the forum and offer help when needed. It may take some longer though and can't be predicted since I'm travelling all the time.

The server dealing with PayPal and providing the Activation Code will unattendently run ... until it does not. It can be remotely controlled though to a certain extend, and again, *if* something happens to it, it may take the time needed before I run into an Internet connection.
In no occasion be afraid that you will loose your money. You won't, and I think enough decency has been built into the procedure to help out anyway; The UMPC accompanying me can generate the Codes as well, and no matter what, PayPal will give me the data needed to do so, and I can always access that.


  Peter

I am very sorry, but then this has to wait until I'm back from holidays.

In the mean time you could look for and upgrade of those ac97 drivers ?

Btw IIRC AC97 worked here, but that was not from a laptop. The laptop I tried did not ...

When I'm back I can create a kind of debug version. By that time I'll ask here whether you still want / need it.

Peter

?

It's called XXModeSupport.log

Ah.

Did you try it with a normal .wav file ? that *is* a necessity ...
Otherwise this looks very strange to me, and I don't think anyone reported such a thing. Also, it will be a tough thing to debug. So I just hope it is about not using a .wav file ...

But if you used a .wav file, is there a .log file with a current date/time in your XX folder after trying ? (I forgot the name). If so, does it contain something ?

If all is right, this is the last version in the 0.9u sequence;
SQ is appreciated the best so far (by your reports) and all thinkable sample rates and bit depths are being supported now.

The 0.9u versions showed quite some problematics, mainly caused by the more heavy load on the system at track boundaries. The latter by itself being caused by as much pre-processingas possible (e.g. bit depth conversions) in order to offload the system as much as possible during playback itself. For that matter, the current state is that minor "hiccups" may be experienced in rare occasions, or not at all depending on various playback settings.
The following has been applied to this version (all Engine#3, and not tested for the other Engines which might have implications !) :

• Processor Appointment is temporarily set to a most optimized situation during Track Load.
  Please note that this is the same application as has been posted in a separate XXEngine3.zip in this post : Re: 0.9u-12 --> Hiccups and Clicks. So, nothing has been changed / approved opposed to that version.
  
  
• As it appeared, during UnAttended Playback and bringing up XXHighEnd for pressing Next, this deleted the tracks in the Playlist Area above the track incurred for next Playback.
  This was a left over from a before intended situation, and has been made consistent now with the current situation (of all Tracks in the Platlist Area just staying there unaccording the played tracks during UnAttended Playback).
  
  
• Similarly using Prev was not allowed at all after bringing up XXHighEnd during UnAttended Playback (in earlier versions no tracks in the Playlist Area could exist above the playing one). This has been made consistent with the current way of working as well.
  
  
• A few versions back the "low" Priority for the ThreadPrio (Settings Area) has been excluded from being able to produce sound. However, it was still offered as a choseable option, and when chosen indeed obviously there was no sound.
  This option has been reoved now from the Thread Priority.
  
  
• It has been found that at the first start of the new 0.9u-14(a) version, some controls like the Volume were disabled;
  If all is right this has been solved now, although it could not be tested (because of the development environment which acts differently to this respect).
  
  
• As you know, with UnAttended Plyaback, Engine#3 is a loose running train, completely separated from all control by XXHighEnd. This has been created to obtain a SQ which is not dependent from as few as possible, where XXHighEnd would be a SQ influencing matter;
  This version improves on conveniency by - at bringing up XXHighEnd - setting the Time Cursor to the position where XXEngine3 currently is playing. This now allows for changing settings like the Volume Slider, followed by pressing Play. Playback obviously commences at the implied position by the Time Cursor.
  Note that Playback will commence a few seconds earlier opposed to what you hear before pressing Play, which comprises of the lag between bringing up XXHighEnd, applying the changes, and press Play.
  
  
• In the nature of the above, the Time Cursor is now enabled for dragging it after bringing up XXHighEnd. In fact it just is when the UnAttented Checkbox is ticked. All 'n all dragging the cursor to the desired playback position will be appreciated.
  
  
• As a most difficult and (for development) time consuming operation, a once loaded Playlist that is subject to conversion (like FLAC, MP3), Anti Aliasing) and which is played UnAttendedly (the conversions being time consuming), is now "cached" for meant conversions. In other words, when - following above new features - XXHighEnd is brought up in order to change settings like Volume or SQ incurring changes, no again time consuming processes take place, and playback will commence almost instantly as were it a native .wav file.
  Issue : At this moment this doesn't succeed for the first track to be played. Thus, currently the first track where playback must commence is subjective to conversion always (when applicable).
  
  
• When, with UnAttended Playback, playing came to an end, XXHighEnd was brought up automatically. This implied a lot of I/O, hence this lasted for a longer time.
  Most of this I/O has been eliminated now.
  
  

Although much effort has been put in this version, it is not said that above mentioned features work in all situations as intended. For that matter, testing has become a job undoable;
Since your humble developer will be out for a near three week vacation, no amendments/updates will be uploaded during this time.
Might this version give problems, please revert to 0.9u-14a.

Edit :

Firstly, in the above should have been mentioned that when you have the feeling that the "saved" converted tracks (like FLAC, MP3) got mixed up somehow, you can bring up XXHighEnd, Stop playback, and Quit the player. At that moment all will be reset.

Then, it occurred that situations exist that when XXHighEnd is brought up to clear the Playlist Area and load a new album etc., following by Play, playback will commence at the position where the previously running track was, hence not at the beginning as it obviously should;
This can be counter attacked by explicitly pressing Stop first (although it may happen that even that won't work). Dragging the Time Cursor to the very beginning for sure will work.
This is not exactly as intended, and will be solved in the next version.

When first an album has been loaded with e.g. 20 tracks (or anyway the Playlist Area contains e.g. 20 tracks) and Play is performed, while later XXHighEnd is brought up followed by Clear and the load of another album with less tracks opposed to the number before in the Playlist Area, e.g. 12, it will occur to you that after the 12th track has been played, playback continues with the 13th track from the earlier contents of the Playlist Area.
Although the Wallpaper will keep on showing which track is playing, the Playlist Area obviously does not justify for that, and it looks really strange.
Of course this is unattended, and will be solved.

Edit2 :

Similar to the 2nd subject mentioned under Edit above, pressing Next (Prev) resumes the next track at the position the before playing track left off. Here too, pressing stop first is the temporary solution, until the next version is around.

As it occurs, the Invert checkbox cannot be ticked anymore. Will be solved in the next version.

Ref210,

I am sorry for this late reply, but somehow I missed your latest post;

I can't be sure what you mean by "the Engine 3 stops executing". But please note that it may take some 30-60 seconds for the process to finish.
But otherwise, what do you mean, and how can you tell ?

Again, sorry for late answering.
Peter

Edward,

I didn't get around to explicitly testing your situation, with the -honestly- idea I can't improve on it anymore. This idea is also given to me by the fact that I one time (yes, one time only during all this time now) had a small hiccup with Scheme-3 and Q1= 14. And, since Scheme-3 is the best solution for spreading resources for all of the other services/threads (which Scheme-1 does not), there's no solution in there as well (like temporarily spreading everything over the two cores). But :

Today, during testing something else, and by pure coincidence, I saw the small hiccup occurring during the drawing of the Wallpaper. So now I'm thinking it is just that doing it. Or at least *now* that is doing it (and the longer hiccups from before indeed were caused by some I/Os that receive priority by the system).
Oh, that process drawing the Wallpaper has the lowest priority already ...

Might you be with the screen in front of your nose during listening, can you confirm this ? No obligations, but might you see it by accident ...

Peter


PS: Might it be relevant ... my Wallpaper is set to 590 (and playback is UnAttended).

Hi,

This can be caused by the device being in use, but this is not likely.
Did you try the DAC Test (bottom option in the DAC Is setting) ?

Most probably you have an HD Audio device in the laptop, and as far as my experiences goes, they never work (properly for Vista Exclusive Mode).
If you post the result of the DAC Test in here, I might be able to say more / recognize things.

Btw, don't forget to tick the both checkboxes in the properties of the device - advanced tab (see below; never mind the 96K/16 shown there; it does nothing for Exclusive Mode).

Peter

Hi Anthony,

If you see the error, I don't ...

I have now sent it to "My Messages" in the forum.

Strange ...
I can't imagine why this address would be rejected. Or the other way around, why an email from here is ...

Peter


PS: I have changed your post for the email address, to prevent from spam(bots) might they be able to use it, haha.

Hi,

I don't know what happened, but the email should be received instantly. And here nothing shows that this did not happen ! (April 13, 3:17 pm).
Anyway, I just sent you the Code again (to the same email address).
Please let me know if you again did not receive anything.

I am sorry for the inconvenience ...

Peter

Thank you Edward.

Hi Bernd,

You could read this : Re: WARNING : Degraded SQ from off 0.9n u/i 9.0p in order to get ideas on what might help. Keep in mind though that the real source of the problem will be something else in your PC. Some versions of XX emphasize it, others can remove it, and there is no explicit "something" I do to incur for the one or the other.
Generally, a higher Q1 value will show less problems (but Q1 is there for changing SQ to your needs, and not to avoid crackles ).

Might you find a solution yourself, please let know what it is. Gerard will be greatful for it.

Peter


PS: Note that this is not somethijng critical in general, or that it would be problematic to play back audio by means of a PC. I for one can really do what I want with my PC and it is impossible to disturb playback. It can (indirectly) disturb itself though and *that* is caused by something wrong in the PC.
Btw, start with removing infrared keyboard/mouse bluetooth devices and other through the air things. If something can disturb, it's those things.

After 7 tracks and Scheme-1 on a first album I just receive my (0.3 sec or so) hiccup between tracks ...

You are completely right Edward. I am very sorry for the confusement;

I did apply those changes I mentioned, but never uploaded that as a new version, and they *are* just in that separately uploaded zip.
Time for a vacation ?

As a matter of fact, I now recall I was waiting for your response on that separately uploaded zip before uploading a full version which should include it. And I never did.

To make it more clear : you listened to the intended combination, and apparently it still doesn't work 100% for you.
Added to that I listened to Scheme-2 where I should have been listening to Scheme-1 in order to copy your behaviour ... you can well say that my mistakes are going to add up.




PS: I will try Scheme-1 now.