fjhuerta
07-15-2003, 11:07 AM
Hi all,
Maybe the title of this thread is a bit misleading... it's just that the header won't allow me a big enough title :)
OK, so here's my theory. I have David Gilmour's concert on DVD. It absolutely rocks, but I can't listen to it on my car stereo.
So I rip the disc. When I end up the procedure, I'm left with an AC3 file and a PCM track. The PCM is 48/20 audio. I then decide to downsample and dither to 44.1/16.
That is, until I realize something. If I apply an anti-aliasing filter and downsample to 44.1, I could make the PCM file DTS-compliant. What would happen if I encoded the disc as 44.1/20 2 channel DTS stereo?
I made two discs: a DTS 44.1/20 stereo and a 44.1/16 stereo.
When I listened to both versions, I realized how close the DTS version was to the DVD. The CD version was harsh and brittle (no doubt due in part to the software I used). But this got me thinking...
Can audio be thought of as a computer image? Sampling being represented by resolution, and bit depth with colour depth? If so, wouldn't a 60Kb JPEG of a photograph be superior to a 60Kb GIF of the same photograph, for any given resolution? Surely enough, it would. Therefore, would a 1.4 MB/sec DTS file would be sonically superior to a 1.4 MB/sec raw PCM stream?
To put it another way, a 32 kb/s MP3 / WMA stream sounds *a lot* better than a 32 kb/s PCM raw stream...
Also, since today we can have 192/24 PCM streams in DVD-A, is compression irrelevant?
Or, can compression introduce artifacts that not even the "smartest" of all algorithms can predict / control?
BTW, I lost my David Gilmour CD. So I *still* can't listen to it on my car. :D
Maybe the title of this thread is a bit misleading... it's just that the header won't allow me a big enough title :)
OK, so here's my theory. I have David Gilmour's concert on DVD. It absolutely rocks, but I can't listen to it on my car stereo.
So I rip the disc. When I end up the procedure, I'm left with an AC3 file and a PCM track. The PCM is 48/20 audio. I then decide to downsample and dither to 44.1/16.
That is, until I realize something. If I apply an anti-aliasing filter and downsample to 44.1, I could make the PCM file DTS-compliant. What would happen if I encoded the disc as 44.1/20 2 channel DTS stereo?
I made two discs: a DTS 44.1/20 stereo and a 44.1/16 stereo.
When I listened to both versions, I realized how close the DTS version was to the DVD. The CD version was harsh and brittle (no doubt due in part to the software I used). But this got me thinking...
Can audio be thought of as a computer image? Sampling being represented by resolution, and bit depth with colour depth? If so, wouldn't a 60Kb JPEG of a photograph be superior to a 60Kb GIF of the same photograph, for any given resolution? Surely enough, it would. Therefore, would a 1.4 MB/sec DTS file would be sonically superior to a 1.4 MB/sec raw PCM stream?
To put it another way, a 32 kb/s MP3 / WMA stream sounds *a lot* better than a 32 kb/s PCM raw stream...
Also, since today we can have 192/24 PCM streams in DVD-A, is compression irrelevant?
Or, can compression introduce artifacts that not even the "smartest" of all algorithms can predict / control?
BTW, I lost my David Gilmour CD. So I *still* can't listen to it on my car. :D