Large capacity IDE hard disks are now economical enough that I want to digitize all my old favorite LPs, and make a computer-based sound system. I have a massive Phase Linear turntable and several MC cartridges, and I want to preserve as much as possible of the MC vinyl sound, and maybe try to eliminate some of the worst clicks and pops.
Years ago, I spoke with people at Pacific Microsonics that developed HDCD technology (since acquired by Microsoft). I got the impression that the main reason their HDCDs sounded pretty darn good is not so much due to the dynamic range enhancements, but mainly because their system required master recordings at> 96 kHz (maybe 192 kHz) sampling rate, and that allowed the A/D anti-aliasing filter frequency to be raised to > 48 kHz. Hence there would be less filter effect intrusion down into the audible analog pass band. I'm only interested in getting my LPs converted into 24/96 digital. I don't want to spend a lot on a sound card, and I don't want to invest in a separate professional A/D converter. I read the specs and reviews on the lower cost 24/96 PCI sound cards, and they all look good enough, The M-Audio $90 2496 card seems attractive because it has RCA jacks for inputs and outputs. It also has a socketed analog output op-amp with a crummy 5532 plugged into it. A $2 Burr Brown OPA 21342 would solve that problem. I could connect directly from my head amp/RIAA preamp line out and plug into the M-Audio analog inputs.
But, I can't figure out if the M-Audio A/D section changes the anti-aliasing filter frequency from the 22 kHz for 44.1 kHz CD standard sampling ratee recording to 48 kHz for 96 kHz sampling rate recording.
I'm not sure if I'm paranoid enough. Maybe all the cheap 24/98 sound cards have only a single analog input anti-aliasing filter set at 22 kHz. They might use that 22 kHz filter for recording at all sampling rates. If that's the case, recording at 96 kHz gives more dynamic range, eats a lot more disk space, and makes my LPs sound much like CDs. On the other hand, maybe the cheap sound cards are very good and very smart, and they automatically shift the anti-aliasing cut-off frequency to half the sampling frequency. In which case, my digital recordings at 96 kHz sound much like my vinyl. Does anyone have knowledge of this issue?
Years ago, I spoke with people at Pacific Microsonics that developed HDCD technology (since acquired by Microsoft). I got the impression that the main reason their HDCDs sounded pretty darn good is not so much due to the dynamic range enhancements, but mainly because their system required master recordings at> 96 kHz (maybe 192 kHz) sampling rate, and that allowed the A/D anti-aliasing filter frequency to be raised to > 48 kHz. Hence there would be less filter effect intrusion down into the audible analog pass band. I'm only interested in getting my LPs converted into 24/96 digital. I don't want to spend a lot on a sound card, and I don't want to invest in a separate professional A/D converter. I read the specs and reviews on the lower cost 24/96 PCI sound cards, and they all look good enough, The M-Audio $90 2496 card seems attractive because it has RCA jacks for inputs and outputs. It also has a socketed analog output op-amp with a crummy 5532 plugged into it. A $2 Burr Brown OPA 21342 would solve that problem. I could connect directly from my head amp/RIAA preamp line out and plug into the M-Audio analog inputs.
But, I can't figure out if the M-Audio A/D section changes the anti-aliasing filter frequency from the 22 kHz for 44.1 kHz CD standard sampling ratee recording to 48 kHz for 96 kHz sampling rate recording.
I'm not sure if I'm paranoid enough. Maybe all the cheap 24/98 sound cards have only a single analog input anti-aliasing filter set at 22 kHz. They might use that 22 kHz filter for recording at all sampling rates. If that's the case, recording at 96 kHz gives more dynamic range, eats a lot more disk space, and makes my LPs sound much like CDs. On the other hand, maybe the cheap sound cards are very good and very smart, and they automatically shift the anti-aliasing cut-off frequency to half the sampling frequency. In which case, my digital recordings at 96 kHz sound much like my vinyl. Does anyone have knowledge of this issue?
check out RMAA and just about any soundcard you can name with Google for frequency response plots in analog loop-back
Latest News. Audio Rightmark (free version is practically the defacto standard for soundcard reviews)
I have a Juli@ card which still has good price/performance ~US$120
free Audacity and SciLab software and you can (I have) made measurements to near 80KHz with -130 dB spot noise floor @192KHz/24 bit
the Delta-Sigma monolithic audio ADCs in these products massively oversample and implement the steep fs/2 anti-alias internally as a digital filter so the corner frequency does change with sample rate selected - the actual analog anti-alias has to be effective at the oversampling frequency, often 128x
Latest News. Audio Rightmark (free version is practically the defacto standard for soundcard reviews)
I have a Juli@ card which still has good price/performance ~US$120
free Audacity and SciLab software and you can (I have) made measurements to near 80KHz with -130 dB spot noise floor @192KHz/24 bit
the Delta-Sigma monolithic audio ADCs in these products massively oversample and implement the steep fs/2 anti-alias internally as a digital filter so the corner frequency does change with sample rate selected - the actual analog anti-alias has to be effective at the oversampling frequency, often 128x
Last edited:
- Status
- Not open for further replies.