Ah, that explains ...

Oops. I hope you don't blame me !

I'll be coming back on a few things, but haven't got much time right now.

Dave, try to make like more comfortable when you are going back to the spinning disk :

1.
If possible use another PC and connect the (SSD) OS disk to that. Now make a backup of that disk with any software able to backup an OS. Then connect the (due) spinning disk, and restore the backup to that. Mount it in the music PC - done.

2.
Note : appropriate backup software will be able to make a backup of a "live" OS disk, hence it theoretically will be able to do that just in your Vista machine (and of course is then able to restore that backup to the due spinning OS disk). BUT :
Acronis would be such software, which DID NOT WORK for me. So no matter what, don't go this route with that (#1 worked though with Acronis). After the restore (#2) it appears that some "live" data wasn't backupped, and the first thing after rebooting with the restore is "Windows wasn't shut down properly", and after a few reboots you end up with an empty screen and nothing to do (this is an Acronis version which supports Vista, huhmm).

Better invest in this "backup - restore" feature, because it will be guaranteed that you end up with the same settings. And in the mean time you'll be having a (guaranteed) backupped OS ...

My pleasure Joerg. Just glad you are happy now.

Peter

Is that only with Scheme-3 ?

(that is a dual core, right ?)

Good.

(I moved the topic from the other thread you posted this in by accident )

Hey, thanks for the info Telstar.
Can you point me to a thread about this ? I want to understand what's actually wrong, or maybe contribute to that thread. Depending on how the situation is, get my money back maybe.

In the very end I want to taste the benefits, where one of them is speed (it really matters).

First of all the error is not related to the album name (I *think*). It more looks like the french Amazon site was down or something like that. I just tried, now it is not so.
- Can you try it again ?
- If still wrong, gen you can another album from the french stite (it can be selected).
- Or, select e.g. the UK site only, and does that procude the error as well ?

The tag data is not used in any case.
The naming *is* wrong because there's no differentiation between the artist and the album title.


I have thought about that myself, and indeed it would be convenient for several reasons.
It is a piece of cake really, just never done (so far) ... priorities ...

I have an OCZ 64GB (and btw, I don't think SSDs exist in SATAII, only SATAI).

My chipset is the ICH8, but I must be careful, because there's also a JMicron part in there, and the lot can be controlled by 3 "drivers". Since I'm not using RAID this is beyond the user (me anyway) what you're actually using.

Actually I don't care that it doesn't work out, but I *DO* want to know what the heck this is about. It sure cannot be about not reading (and presenting) the data properly.


Yea, and I told you that listening to Ritchie Vitale couldn't surive the first track at Mondays. Tried again Tuesdays with the 2nd and 3rd track, and after the 3rd my wife told me it was quite boring and I agreed and switched it off. Yesterday (Wednesday) after the restoration of the spinning OS disk, I didn't realize I ran 2 tracks the day before, so put on the 2nd track while actually wanting something new (I can't "A-B" with the same tracks hahaha). What I immediately sensed was romance, a man blowing a horn with feeling. It was interesting all over, never mind the slow song. Then I realized it was the same track from the day before which was so boring ...
This time the whole album was listened out, and I came to the conlcusion it had nothing to do with not being in the mood for this kind of music (which I thought at first), the other days. It was a lousy SSD.

It is so strange to experience it : Not in all music the bad behaviour of the SSD can be heard. But last night, when all was right again, all music showed the happiness again ("happiness" is really a phenomenon I recognize lately; when all is good, music is a party). Everybody starts swinging automatically. It didn't happen with the SSD.
Now why ...

Just in brief :

Didn't investigate the South Bridge yet, but removed the little "controller" and attached the SSD to a normal SATAII line : no difference.
Then reinstalled the spinning disk : all OK.

There is something seriously wrong with the SSD, and since it won't be skipping bits or anything of the kind, it is the most seriously influencing jitter or whatever it is.

I tend to investigate the reasons, but think it is a waste of time to ever go back.
It is the most influencing phenomenon I ever experienced, and I don't like that for a fact. At all.

If this is so (so so so so much) influencing, what other things do we have ?

I will look into that !
Thanks for pointing it out.

Hehe those who follow the development somewhat longer know that I have one priority only (apart from improving sound) : keep all as "bit perfect" as it should be. This means nothing of the kind you may think, and no DSP (ever).

I follow the "rule" : let all come out as 1:1 as possible, and from all follows that the only thing which can be influenced is jitter in the traject towards the DAC, but even further, "in" the DAC just the same, or anyway with influencing that, say, control the (digital <-> analogue) accuracy of the DAC. Voodoo, hocus pocus, but I learned how to influence those things by means of software.

But go measure : the bits going in are the same as on your hdd ... which counts for every software player with means of bit perfect playback.
But it is not about that ...

Thank you for your extensive replies Dave.

First thing, and not to forget : When you'd try the normal spinning disk there is really no way you would do that for my sake, helping, confirmation or anything of the kind. You would do it for YOU and only you. So, if it is too much trouble, and you are confident you just don't want anything else/better, spend you time in a useful way, like listening to music with a couple of glasses with good contents, and a cigar for a change.

I survived yesterday again (and you know I actually won't go for less than a week of listening to the same after a change, just to form the right opinion), but this time I last with the situation in order to "hear" what could be going on. And the funny thing is, when I went to bed yesterday I came to a rather similar thinking as you just wrote :


But different ...
What slipped through my mind is that I am using this small IDE-SATA converter, because I don't want to loose a SATAII connection for the large music disks. And, this converter even needs power ...
So now I think (and by now supported to what you came up with) that it might be this thingy doing it to me, and different from the actually passive SSD disk, this thingy indeed could be sending interrupts all over, even to the sense of "I have nothing to do !".

Ok, writing about this now takes longer than trying a different setup, and this just needs to connect the SSD disk to a normal SATAII connection. The difference will be audible within 2 seconds anyway. Will let you know tomorrow.


Anothing thing : As I said elsewhere, indeed there is a difference in SQ between 0.9v and 0.9w, but you have to limit the used memory first. The difference is fragile, but IMO more lean and IMO for sure better.
My current SSD setup just destroys it all.

Btw, let's keep in mind what I said earlier : I can hear the SSD is better fo a few reasons, but net it doesn't work out. Such a thing usually indicates that 2 things changed in one time, and with my above explanation this would be true indeed (changing the "controller" and changing the disk type).

This reminds me of the 0.9d period, where each of two versions showed better behaviour but in different areas and how to combine the goodies of each into one version. Nice challenges.

Because since 0.9-w the conversions are allowed to run in several "threads" (self containing processes). Each such a thread runs on its own core (otherwise it would be useless). The advantage : with two cores the conversions run twice as fast, with 3, three times as fast, with 4 or more, four times as fast.
Since 3 wasn't recognized as an existing situation, it just went wrong (though with that strange message which indeed normally is dedicated to old Visual Studio References and the like, so Google didn't help me at all on this one ).

---

There are serious reaons to believe that this version is not the best for sound quality. If you are new here and want to experience XXHighEnd for sound quality, please go to XXHighEnd Model 0.9v-7.

---

This version mostly solves the bugs introduced per the 0.9w sequence.

• As it appeared, 3 core processors exist, and this was not supported in such a way that things went wrong at the new "parallel" conversion method (FLAC to WAV etc.), generally resulting in a "Path is not of a legal form" error.
  This by itself lead to the conclusion that the Processor Core Appointment (Settings Area) also did not work in the case of a 3 core processor, but this was without notice;
  Now, the following range of cores are supported throughout : 1, 2, 3, 4, 6 and 8. Note though that for the parallel conversion 6 and 8 are treated as 4, hence no more than 4 parallel threads will serve the conversions; As said in an earlier Release Note, the explicit support for 8 (or 6) cores can be added just the same, but please let know whether you have such a processor (twin 3 or 4 cores).
  
  
• Besides the earlier supported track number derival from .cue information, now the track number can be derived from FLAC-tag info (or MP3 tags) as well. When it concerns an album with .cue data going with it, that takes prevalence over getting the track number from the tag info (might the combination exist, as it is for MP3 theoretically).
  
  
• Engine#1 was rendered useless per 0.9w-1 because it played all tracks from the Playlist in parallel. This is now solved.
  
  
• A possibly longer existing problem was about not being able to drag the timer cursor during Unattended Playback (after bringing up XXHighEnd) with the intention to continue playback at the dragged position. Solved.
  
  
• At creating 0.9-w2 and the parallel conversion (from FLAC or MP3) the Anti Alias conversion was forgotten, and AA just did not work at all. Now solved.
  
  
• An error message has been built in, telling when an attempt is made to AA convert a Cue File album. Such a combination is not supported.
  
  
• Although not sure whether in 0.9w-1 or -2 being there as a problem, most probably it could occur that FLAC/MP3 conversion with Unattended Playback could stall XXHighEnd during the conversion. If all is right, solved now.
  
  
• Cue Files couldn't be loaded from a Gallery since 0.9w-2. Now they can again.
  
  

One thing left to do since 0.9-w was introduced : Support for 24/352800 to 24/176400 conversion (which requires Anti Alias conversion) together with FLAC conversion. Also, 24/352800 to 24/176400 in general (source = WAV) might not be working anymore in the new setup for 0.9w-3 (it just wasn't tested but should be still working).

No no no, the problem you have currently indeed is related to the conversion, and is solved here now.
But I tried to warn you for the Processor Core Appointment setting (Settings Area) which won't do anything for you other than "normal". When it works as intended (must fix that now), you will be able choose from 4 appointment settings, one of them being the playback thread (process) dedicated to one core, and all of the other stuff in the PC to the other cores. So, you *can* make the setting, but won't notice any difference in sound, because I don't deal with the 3 core processor (didn't know it exists). When it works you sure will be able to hear the difference, although I can't advise you about the needed settings (differs per setup; I use setting 3).