[Lf] Re: Spectran

[Alberto di Bene]

Frank Gentges wrote:

> Alberto,
>
> Good to hear your on the LF reflector.  Hope our chatter doesn't fill up
> your box too fast.  Further discussion and testing over here shows that
> Spectran has an error of 6 Hz on my IBM ThinkPad Notebook.  I suspect it
> is a sampling rate error.  An adjustment factor to compensate from
> computer to computer would be useful.  In the mode we use now that error
> is taken out with the receiver adjustments.

Frank,
                thanks for pointing out that error. I have an IBM Thinkpad 770, so
I will test Spectran on it too. It may be well possible that there is a sampling rate
error. Spectran samples at 8000 Hz, which is not one of the guaranteed frequencies
on sound cards. 11025 would be, but I chose 8000 to reduce the CPU loading.
I think the best way is to make the sampling frequency user-selectable....
On my home PC, with an old SB AWE32 sound card, 8000 Hz are quite accurate,
so I didn't think that this freq. could pose an accuracy issue on other PCs.

We are still fighting against the Windows mixer. We would like to set it automatically
when Spectran starts, without user intervention, but unfortunately Microsoft didn't
specify
a very precise model for the virtual mixer, so every manufacturer has interpreted it
in its own way. This means that our routines must adapt to whatever sound card they
find, and this leads to some complications... but we will win :-)

73
Alberto    I2PHD