I currently have a 2.1 speaker set-up connected to a Pioneer AVR. My PC is connected to the AVR via toslink. Now, I know that only PCM stereo or multiplexed multichannel audio can be transmitted over it. I have AC3filter decoding its supported formats. Is it considered preferable to set it to output AS-IS as opposed to 2.0? If I understand correctly, 5.1 audio, for instance, will be downsampled to 2ch PCM via software if I set it to output 2.0, and will then be upsampled later by the soundcard if it isn't 48kHz. If I set it to output AS-IS, the 5.1 will just pass to the soundcard which would then downsample to stereo and upsample to 48kHz itself. The two ways look the same to me, save that when I output AS-IS the volume is lower than when I set AC3filter to output 2.0 (with all of the autogain and normalizations turned off) so I'm kinda confused.
Also, I downloaded a couple of 5.1 audio samples from here:
http://www.diatonis.com/downloads_dts_ac3.html
There are 5.1 wav files there, but unless I use AC3filter for PCM, they are just garble and static when I play them. Why exactly is that? I figure that since they are just uncompressed wavs and not DTS, AC3 or anything else, that it wouldn't be problematic to handle normally with DS filters. Is it because a non-standard codec is still needed to parse the multiple audio channels in the wav?
One last question. When I upgrade to a 5.1 speaker set-up, would it be better to output ProLogic II or reencode to AC3 for non-passthrough non-stereo sounds? I imagine PL2, since it's PCM, is less lossy but it doesn't have the discreet channels that AC3 has.
Sorry if this is simple stuff but I get confused easily.
