Niiice finally
Now just remame that library as "add files".

Another thing that can be useful, although you may twitch your nose is to allow for external DSP or (even better) VST plugins. This is not the same thing as having a convolution to load filters (ie. XO), but it is very helpful to do a little equalization or to exclude certain bandwidths for testing. Yes, it wont be bit-perfect anymore, but in certain cases will sound better

To me it does

I'm afraid that some SET amplifiers could not be fast enough, yeah. But maybe i'm wrong.
My amp is a SET of jfet (the lovely firstwatt f3).

There's never been a clear definition of oversampling and upsampling, somebody refers to the bitrate (which does nothing and its trascurable), others to integer vs floating point resampling, the latter being sort of universally accepted.
ex. 4x upsampling as Peter said = 44,1k x4 = 176,4k
ex 4,something oversampling: 44,1k => 192k

Integer upsampling is the best way to do it, because it creates the interpolated points at fixed points in space (sorry for my poor explanation -i'm abroad and i dont have my reference links).

Long time ago (like 6 months or more) i had a long discussion with Peter about this subject. the conclusion was that if OS has to be done, be it integer of at least 4x (8x would be ideal according to Peter, but there are other issues that are not solvable at present times, so he is settling for 4x -in software).

If you scavenge on diyaudio and audio asylum you would find some of my posts about possible ways to use the computer as transport, about doing the resampling IN SOFTWARE and why it is better than any SRC available to use in electronic devices, including state of the art algorithms such as the ones used by Lavry, dCS, Meitner, chord and so on.
If we consider the FLEXIBILITY of the computer medium for any calculation done in the digital domain, the choice becomes obvious.

I'm very excited by Peter (and the digital engineer) progress of making all this theory a reality, and for a reasonable price. Yes, i'm one that cannot afford 6 figures and that i would be skeptical spending even if i could.
I understand his happiness regarding the output stage, because it is THE most important part in defining the sound of a DAC. Now, I'm hoping that MY amplifiers will be fast enought to follow it... (the tube folks wont be happy)

The new digital board that Peter is waiting will mostly improve on jitter only and will be more crucial for high-res material, so i believe that the sound that he is hearing now is 99% of what he'll experience with the new board with 16/44 tracks.

I'm not sure which amount of information i can post publically at this stage, so i'll stop here. I'm really looking at the final production unit of this dac and i have no doubts that it will be the best that can be made to be connected to a computer transport, short of programming a custom multibit converter.

The NOS1 is a 24/192* multibit dac.

He has a sabre buffalo so if he got the time he can do the measurements for it.

[*actually accepting higher than 192k]

Thanks that would be nice. Waiting for Peter to finish the DAC, though
For now i'll play with the dcx.

L/R swap.
No hurry but I would like it.

I wonder if you could add a polarity inverse...

I thought it would reverse the polarity, but it does not, so i just keep it off. In tracks where a sound is on the left side, click invert, and it will stays on the left side!

I like to wear my headphones switched and a reverse polarity (that works) would be really helpful. (i.e. in foobar works)

The tree window can just be as long as the main xxhe window, like the gallery.
This way there is no problem for who has small or big screen size.

Peter,

All these errors are random. I cannot find a cause for them.
What I can tell you is this:

1) NOT optimized machine, therefore some of the services that crash you have disabled them.
2) Vista x32 or x64 with SP2.

I will be using W7 RC on the dedicated pc and with services and stuff disabled. We shall see then if anything still crash. I will try to find out more. But if everything is random, debugging does not belong to me.
I have worked with some software houses and they have several test computers with different configuration and software to test and debug.
I know it is not high priority, but bear in mind that random crashes (not talking about services here) are unacceptable in a 1.0 release.

About the tree and the space, see audiodidakt reply, his picture is maybe clearer than the one I sent you a couple months ago
The tree should continue DOWN over that empty space, as long as the window height goes.


You can change something minor in that tab maybe easier. like thanks for registering in the place to enter the code.

The critters first:

-Random disappearing of the windows ribbon in the status bar (about 20% of times)
-occasional "track not found error"
-occasional crash (xxhighend stopped working - while engine3 still plays music)
-after SP2 is more likely that some services crash, such as windows host. this especially on the slower computer with vista x32

UI improvements
- make the tree longer on the left in the library view. With scrollbars like everything else. i have 1920x1200 resolution!
- the button labeled "library" that is used to add files to the playlist should be named "open" or "add files". It is confusing with the library tab
- undemo tab should go away in the registered version

Thats all folks

and???

Yes, it DOES take long

I removed the "()" and now they are working