sgrossklass
Headphoneus Supremus
- Joined
- Dec 3, 2004
- Posts
- 2,803
- Likes
- 22
Yes, my main system does indeed have 3 sound cards:
1. Terratec Aureon Sky reflashed to Prodigy 7.1. Output to cans and amp, used to do LP recording.
2. Aureon Sky bone stock. Radio recordings.
3. Oldschool Terratec Base1, an ISA thing with AD1816 all-in-one sound chip. (Bought out of curiosity ages ago.) No uses so far, but there weren't any other cards competing for spare ISA slots.
Now I wanted to listen to some MODs and stuff again (not done in 1+ year) and fired up the trusty Modplug Player, only to realized that my EQ (in Winamp) has no effect there. Using Winamp would have required unpacking all those archives. The ideal solution would have been something like Virtual Audio Cable, to route the output from Modplug to a virtual input that the Line-In plugin for Winamp could then use. Now since I don't have a full version of said software, I employed a Real Audio Cable to connect the output of card 3 to the input of card 1 - voilà, problem solved. (After I enabled the secondary buffers in Modplug, that is.)
Admittedly the AD1816 isn't that great - the "fuzz" I've seen in measurements with RMAA I realized to be artifacts from not so great anti-alias lowpass filtering today, at least it looks awfully similar to what not-so-great resamplers generate. (Strangely enough, the digital filter response in the datasheet looks much better than the WM8770's. But then the AD chip apparently does not employ an oversampling digital filter, so this could be it. EDIT: I guess this quote clears things up: "The AD1816A includes a variable sample rate converter that lets the codec instantaneously change and process sample rates from 4 kHz to 55.2 kHz with a resolution of 1 Hz. The in-band integrated noise and distortion artifacts introduced by rate conversions are below –90 dB." That figure looks about right, but with a noise floor at about -114 dB the artifacts are clearly visible. To put things in perspective, a CS4624 based card with fs=48 kHz at 44 kHz barely performs any better, but does allow using its native sample rate much more easily.) The closer you get to the maximum supported sample rate of 55.2 kHz, the worse it is. Hey, it was a lowcost chip in '97, and quite good for what it is. Besides, playback quality already is much improved when compared to last time I heard the stuff.
1. Terratec Aureon Sky reflashed to Prodigy 7.1. Output to cans and amp, used to do LP recording.
2. Aureon Sky bone stock. Radio recordings.
3. Oldschool Terratec Base1, an ISA thing with AD1816 all-in-one sound chip. (Bought out of curiosity ages ago.) No uses so far, but there weren't any other cards competing for spare ISA slots.
Now I wanted to listen to some MODs and stuff again (not done in 1+ year) and fired up the trusty Modplug Player, only to realized that my EQ (in Winamp) has no effect there. Using Winamp would have required unpacking all those archives. The ideal solution would have been something like Virtual Audio Cable, to route the output from Modplug to a virtual input that the Line-In plugin for Winamp could then use. Now since I don't have a full version of said software, I employed a Real Audio Cable to connect the output of card 3 to the input of card 1 - voilà, problem solved. (After I enabled the secondary buffers in Modplug, that is.)
Admittedly the AD1816 isn't that great - the "fuzz" I've seen in measurements with RMAA I realized to be artifacts from not so great anti-alias lowpass filtering today, at least it looks awfully similar to what not-so-great resamplers generate. (Strangely enough, the digital filter response in the datasheet looks much better than the WM8770's. But then the AD chip apparently does not employ an oversampling digital filter, so this could be it. EDIT: I guess this quote clears things up: "The AD1816A includes a variable sample rate converter that lets the codec instantaneously change and process sample rates from 4 kHz to 55.2 kHz with a resolution of 1 Hz. The in-band integrated noise and distortion artifacts introduced by rate conversions are below –90 dB." That figure looks about right, but with a noise floor at about -114 dB the artifacts are clearly visible. To put things in perspective, a CS4624 based card with fs=48 kHz at 44 kHz barely performs any better, but does allow using its native sample rate much more easily.) The closer you get to the maximum supported sample rate of 55.2 kHz, the worse it is. Hey, it was a lowcost chip in '97, and quite good for what it is. Besides, playback quality already is much improved when compared to last time I heard the stuff.
