Did I miss something?
I just noticed that there is no 88.2kHz samplerate capability in the new Scope software. Or has this feature always been missing? Seems odd to me that the software wouldn't support this when virtually everything else on the face of the planet does. Even my A16 Ultra. It's not a show stopper for me. Just curious.
88.2 kHz?
-
- Posts: 437
- Joined: Tue Oct 01, 2002 4:00 pm
- Location: Sweden
I have a creeping suspicion... that the simple division by two isn't the proper way to downsample from 88 to 44k.
It's tempting to assume the even sample values divide by 2 without problems and the odd ones introduce a (stastistical) round up error of 25%.
But this neglects the fact that the original values are already an approximation, so every sample has a 50% chance to be a wrong value - even more this calculation completely ignores the time component of the signal.
Of course it's done this way in simple designs and it won't introduce 'horrible' artifacts, but the result should be noticable different from a properly downsampled one.
my 2 cents, Tom
It's tempting to assume the even sample values divide by 2 without problems and the odd ones introduce a (stastistical) round up error of 25%.
But this neglects the fact that the original values are already an approximation, so every sample has a 50% chance to be a wrong value - even more this calculation completely ignores the time component of the signal.
Of course it's done this way in simple designs and it won't introduce 'horrible' artifacts, but the result should be noticable different from a properly downsampled one.
my 2 cents, Tom
look here for more info, but this is what bassdude said:
Quote:
-----------------------------------------------------------------------
..... it is better to have to divide by 2 than to divide by.....
------------------------------------------------------------------------
Which is a popular myth. There is misconception that the conversion from 88.2 to 44.1kHz (popular myth:- you only need to drop every second sample) is better and easier compared to conversion from 96kHz to to 44.1Khz. Absolutety incorrect! The algorythms used for 88.2 -> 44.1 and 96 -> 44.1 are both as complex.
Quote:
-----------------------------------------------------------------------
..... it is better to have to divide by 2 than to divide by.....
------------------------------------------------------------------------
Which is a popular myth. There is misconception that the conversion from 88.2 to 44.1kHz (popular myth:- you only need to drop every second sample) is better and easier compared to conversion from 96kHz to to 44.1Khz. Absolutety incorrect! The algorythms used for 88.2 -> 44.1 and 96 -> 44.1 are both as complex.
andy
the lunatics are in the hall
the lunatics are in the hall
All righty then...
It's not mental illness - it really is a missing feature.
Seems very strange. I went back and looked at an older user manual and it was missing there so it's something that's never been there to begin with I assume. Irregardless of the math behind it, why release a piece of hardware that the software can't support?
Why am I asking myself these questions?
It's not mental illness - it really is a missing feature.
Seems very strange. I went back and looked at an older user manual and it was missing there so it's something that's never been there to begin with I assume. Irregardless of the math behind it, why release a piece of hardware that the software can't support?
Why am I asking myself these questions?
