rfbrw said:
Some are, some aren't. Please let's keep this thread clean from beliefs and pseudo religion between os and non os troops. Please see the thread title.
ECdesigns has given much valuable input that does not deserve useless polution. Let us be glad that there is a person designing in reality instead of talking in this thread.
Judging by number of people trying to buy a pile of 1541's... there is every reason to debate the very purpose of this thread. The most revealing tests have not been performed. Linear interpolation creates spectra not found in the original signal... this may be pleasing to the ear... but it is NOT reproduction.
It is not reasonable to discard the very science that brought the potential (or need) for improvement to the table in the first place.
As far the thread title is concerned... THIS IS 8X OVERSAMPLING!
It is not reasonable to discard the very science that brought the potential (or need) for improvement to the table in the first place.
As far the thread title is concerned... THIS IS 8X OVERSAMPLING!
Consider the fundamental premise behind this whole thing. Moving the frequency of the filter up.
Linear interpolation is creating spectra between 22 kHz and 80 khz... so what is the point? My simulations show that spectra are created BELOW 22 kHz. My sims are too crude to publish... but might the 100's of texts suggesting the same thing be credible?
I am not discounting EC's claim that it sounds better... only why?
Linear interpolation is creating spectra between 22 kHz and 80 khz... so what is the point? My simulations show that spectra are created BELOW 22 kHz. My sims are too crude to publish... but might the 100's of texts suggesting the same thing be credible?
I am not discounting EC's claim that it sounds better... only why?
I really must take issue with the idea that there is any reduction in non linearity due to the number of dacs used.
Though the outputs are summed, they are summed simultaneously, not in parallel and as such each dac sees the I/V stage in isolation. The second reason is that the method chosen to create a balanced output sums alterrnate samples so there is no reduction due to the common mode effect.
In short the DACs are not in parallel and there is no common mode rejection when in DI mode.
Though the outputs are summed, they are summed simultaneously, not in parallel and as such each dac sees the I/V stage in isolation. The second reason is that the method chosen to create a balanced output sums alterrnate samples so there is no reduction due to the common mode effect.
In short the DACs are not in parallel and there is no common mode rejection when in DI mode.