I see Measurement computing has a new 16 bit 1 megasample/sec a/d and d/a card. Has differential inputs with 7 selectable gains. Lets say I hook up a Shure mic to it and say "testing one two three" into it while recording at say 48KHz sample rate, and I play back the recorded file. Would it sound absolutely terrible (even on voice?) because of lack of sharp rolloff filters at 20KHz? Or just wierd or strange? Or ok? Seems like this type of equipment records hi freq vibrations and all kinds of stuff just fine without filters. Just wondering if this card would be suitable for audio work?
Supposing that the card can handle the mic output directly, it is quite simple: if you sample a signal that has a bandwidth larger than half the sample rate you will get aliasing. So if you sample at 48kHz and if there are no filters on the card and your mic has an upper frequency of
24kHz or more you will get aliasing. It is up to you to decide if the aliasing is acceptable or not.
You can limit aliasing by choosing a (much) higher samplerate and then low-pass filter in software and downsample to 48kHz.
PCI-2525 16 channel, 1-MHz, 16-bit analog I/O board, two 16-bit analog outputs: $749 USD Drivers not included. Interface circuitry and connector hardware required for microphone, line out, headphone out (line in should work, needs connector). Input protection not built-in. Harsh Environment warranty (MC pays 50% if you smoke it).
Creative Labs Sound Blaster Audigy SE: $34.99 (Tiger Direct) Drivers included, No interface or hardware required for any function. Built-in input protection
12-month manufacturer warranty
Peace of mind in getting it done without hassles: Priceless.
Buy dozens of the high end sound cards, and give all but one of 'em away on a downtown street corner, and you'll still be way ahead of the game, on cost and because you won't have to write software.
Theoretically, you could use the PCI-2525 -- but why?
Pretty expensive way to get no better than average sound card performance!
But in answer to your question, I think the aliasing problems that the other respondent mentioned will not be a big deal in most normal cases. The reason is that the level of the aliased components will be comparable to the level of the high frequencies that caused them.. In other words, youy have to look at the energy that is above the Nyquist (half sample rate) frequency. An input signal of (say)
40 kHz will "fold" at 24 kHz (with your 48 kHz sample rate) and become 8 kHz, a high frequency but definitely in the audible range. However, that 8 kHz component will be as loud as the original 40 kHz component, which is probably "not very". In the real world, all the high frequency junk is usually broadband noise anyway, so it just becomes more broadband noise after aliasing.
The place you would have to watch out is recording signals with lots of strong upper harmonics, with a mic that can hande the high frequencies (not the Shure). Maybe a mic mounted inside a brass instrument or snare drum?
The reason lab-type data acquisition typically avoids anti-alias filters is that they play hob with the waveform. Waveforms are often of more interest than their spectra for lots of lab work.
Best regards,
Bob Masta dqatechATdaqartaDOTcom D A Q A R T A Data AcQuisition And Real-Time Analysis
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.