Recording at 192/24 on an Audigy 2 ZS

ursamtl · Nov 17, 2006, 05:02 PM

Here's a question for those with technical knowledge of the Audigy 2 ZS Platinum and Platinum Pro cards.

On the Steve Hoffman music forum, a poster mentioned being able to record using his Audigy 2 ZS Platinum Pro at a sampling rate of 192kHz and bit depth of 24. He did so in Wavelab using the WDM driver. I have an A2 ZS Platinum, so I tried this myself from my turntable (I have a cartridge and phono pre-amp with extended top end). In fact, using Plogue Bidule, I was able to record at 192kHz using the WDM driver (not the ASIO one). However, a frequency analysis showed that all audio information dropped off at around 48-50kHz. My theory on this is that the WDM driver permits recording at 192 but the Z2 ZS card's A-D converter simply doesn't output any information above 96kHz (48x2). The original poster felt that perhaps he's noticing a difference because his is a Platinum Pro.

No, I'm not part dog, so I can't hear that high, but there have been tests done that prove hearing perception is definitely affected by such frequencies. I have heard transfers of vinyl records done from audiophile-grade audio equipment using the ESI Juli soundcard at 192/24 and found that there is a perceptable difference in the smoothness and detail of the high frequencies and the soundstage when compared with the same material at 96/24.

Anyway, setting aside all debates on whether such audio information is there or even audible, is it physically possible for the A2 ZS Platinum Pro A-D converter to output at 192? Is it the same or different for the A2 ZS Platinum?

As for drivers, we are both using the latest CL drivers (I stopped using kx because I've been using 24/96 ASIO).

**dj_stick** · Nov 17, 2006, 11:32 PM

the ADC on any of the audigy 2 cards can only handle 24/96 at most afair, and anyway the p16v chip can only handle 24/96 signals

with most "pro" applications you can record at any sample rate and bit-depth, but the restriction here is the hardware.

that poster may be experiencing a "placebo" style effect - but using the external converters in the platinum pro probably helped improve the quality (better SNR/frequency repsonse)

ursamtl · Nov 17, 2006, 11:47 PM

Thanks for the reply. You pretty much confirmed what I thought. This being said, I did some testing this evening and came up with some interesting results. I have the re-issue of Cat Stevens' Tea for the Tillerman on vinyl, remastered from the original tapes. This has a lot of high-frequency transients especially in the guitar strums. I recorded the song Where Do the Children Play both at 96kHz and 192kHz. In doing a FFT spectral analysis during playback, the 192kHz file had visible percussive information up to about 35kHz. Mind you this was very low level at about -75dB but it was still visibly percussive and not just straight line noise. The same track recorded at 96kHz only went up to about 26kHz again at around -75dB. In both cases, the frequencies above these limits flatlined at about -106dB or so. Interestingly, when I resampled the 192kHz file down to 96kHz, the percussive information still reached 35kHz although at this range it dropped to about -80-85dB. Nonetheless, it was still distinct from the -106dB flatline above it. Now, is this musically relevant? I don't know. These guys seem to think so: http://jn.physiology.org/cgi/content/full/83/6/3548 I've also read that the ulra high frequencies visible in an analysis of a vinyl transfer is the result of dust particles, etc., on the record surface. If this were the case, then why is the stuff I see percussively consistent with the music? Perhaps these are some kind of quantization artifacts that follow the digitization of the actual audio information. I'm far from being an expert. I just love playing around with this stuff and I love music. This would still be possible if the ADC was limited to 24/96 but it suggests that to get the best performance, one should actually record at 24/192.

Max M. · Nov 18, 2006, 12:02 AM

> the 192kHz file had visible percussive information up to about 35kHz.

this might be just an arthifacts ('aliasing') of 96->192 resampling
(note this automatic resampling to any bit-depth/sample-rate is a core part of the WDM stuff)

**dj_stick** · Nov 18, 2006, 05:39 AM

35khz audio is well within the range of the 96khz samplerate (the max frequency is the nyquist frequency, which is 48khz in this case) so the different sample-rates shouldn't have anything to do with this

Nov 17, 2006, 05:02 PM	#1
ursamtl Montrealer Join Date: Apr 2004 Posts: 126 Rep Power: 0	Recording at 192/24 on an Audigy 2 ZS Here's a question for those with technical knowledge of the Audigy 2 ZS Platinum and Platinum Pro cards. On the Steve Hoffman music forum, a poster mentioned being able to record using his Audigy 2 ZS Platinum Pro at a sampling rate of 192kHz and bit depth of 24. He did so in Wavelab using the WDM driver. I have an A2 ZS Platinum, so I tried this myself from my turntable (I have a cartridge and phono pre-amp with extended top end). In fact, using Plogue Bidule, I was able to record at 192kHz using the WDM driver (not the ASIO one). However, a frequency analysis showed that all audio information dropped off at around 48-50kHz. My theory on this is that the WDM driver permits recording at 192 but the Z2 ZS card's A-D converter simply doesn't output any information above 96kHz (48x2). The original poster felt that perhaps he's noticing a difference because his is a Platinum Pro. No, I'm not part dog, so I can't hear that high, but there have been tests done that prove hearing perception is definitely affected by such frequencies. I have heard transfers of vinyl records done from audiophile-grade audio equipment using the ESI Juli soundcard at 192/24 and found that there is a perceptable difference in the smoothness and detail of the high frequencies and the soundstage when compared with the same material at 96/24. Anyway, setting aside all debates on whether such audio information is there or even audible, is it physically possible for the A2 ZS Platinum Pro A-D converter to output at 192? Is it the same or different for the A2 ZS Platinum? As for drivers, we are both using the latest CL drivers (I stopped using kx because I've been using 24/96 ASIO).

Nov 17, 2006, 11:47 PM	Thread Starter #3
ursamtl Montrealer Join Date: Apr 2004 Posts: 126 Rep Power: 0	Thanks for the reply. You pretty much confirmed what I thought. This being said, I did some testing this evening and came up with some interesting results. I have the re-issue of Cat Stevens' Tea for the Tillerman on vinyl, remastered from the original tapes. This has a lot of high-frequency transients especially in the guitar strums. I recorded the song Where Do the Children Play both at 96kHz and 192kHz. In doing a FFT spectral analysis during playback, the 192kHz file had visible percussive information up to about 35kHz. Mind you this was very low level at about -75dB but it was still visibly percussive and not just straight line noise. The same track recorded at 96kHz only went up to about 26kHz again at around -75dB. In both cases, the frequencies above these limits flatlined at about -106dB or so. Interestingly, when I resampled the 192kHz file down to 96kHz, the percussive information still reached 35kHz although at this range it dropped to about -80-85dB. Nonetheless, it was still distinct from the -106dB flatline above it. Now, is this musically relevant? I don't know. These guys seem to think so: http://jn.physiology.org/cgi/content/full/83/6/3548 I've also read that the ulra high frequencies visible in an analysis of a vinyl transfer is the result of dust particles, etc., on the record surface. If this were the case, then why is the stuff I see percussively consistent with the music? Perhaps these are some kind of quantization artifacts that follow the digitization of the actual audio information. I'm far from being an expert. I just love playing around with this stuff and I love music. This would still be possible if the ADC was limited to 24/96 but it suggests that to get the best performance, one should actually record at 24/192.

Nov 18, 2006, 12:02 AM	#4
Max M. h/h member-shmember Join Date: Dec 2002 Location: Evil Empire Posts: 2,639 Rep Power: 69	> the 192kHz file had visible percussive information up to about 35kHz. this might be just an arthifacts ('aliasing') of 96->192 resampling (note this automatic resampling to any bit-depth/sample-rate is a core part of the WDM stuff)