In Reply to: Upsampling for dummies posted by Frihed89 on June 12, 2005 at 06:29:56:
"Tell me is 'upsampling' a means of filling in the extra bits from 16 to 24 bits using some sort interpolation algorithm?"Yes, an "interpolation" algorithm is applied. Also known as a "digital filter."
But this is also the case with the more-classic "oversampling" as well.
The "extra bits" are usually up to 24, but it can be up to anywhere from 18 through 32 (and even more). Most DACs back in the 1990's oversampled to 20 bits, and more recent devices are oversample/upsample to 24 bits. (I do think virtually nothing is gained sonically between 20-bit and 24-bit interpolation.)
The main difference between classic "oversampling" and the more modern "upsampling" is the newer method often uses a chip external to the final D/A chip (where "oversampling" is done within the D/A chip), and this extra chip often involves asynchronous sample rate conversion (ASRC), which means the frequency of the converted sample frequency is not directly tied to the input sample frequency.
"Is it much more complicated? How so, if so?"
As stated above, "upsampling" often involves asynchronous sample rate conversion, which I personally say is not only overly complicated, but I think also compromises sonics. (I think ASRC introduces noise into the output signal, and often obscures low-level information/cues. YMMV.)
"Do all CDs made today have 24 bit information? If not, how can you tell on the CD?"
The standard "Redbook" CDs, which are the only ones where upsampling is applied, are *all* 16-bit 44.1-kHz sampled data. And even if the master itself may have been higher-resolution digital, once that data is reduced down to 16/44, it cannot be determined if the resolution of the master was higher or not. So if you see standard CDs touted as being from high-resolution masters, that in itself gains no sonic advantage during playback. Upsampled, oversampled, or non-oversampled.
![]()
![]()
This post is made possible by the generous support of people like you and our sponsors:
Follow Ups
- Re: Upsampling for dummies - Todd Krieger 06/12/0523:50:52 06/12/05 (7)
- Re: Upsampling for dummies - Werner 22:49:11 06/14/05 (0)
- Re: Upsampling for dummies - jnorv 19:05:38 06/13/05 (4)
- Re: Upsampling for dummies - Todd Krieger 20:13:34 06/13/05 (3)
- additional reading? - jnorv 18:42:25 06/14/05 (2)
- Re: additional reading? - Todd Krieger 21:11:16 06/14/05 (1)
- Thanks - jnorv 09:38:58 06/15/05 (0)
- Re: Upsampling for dummies - Frihed89 02:56:40 06/13/05 (0)