Home Propeller Head Plaza

Technical and scientific discussion of amps, cables and other topics.

Re: Even if the software were perfect ....

Howdy John Risch,

I"m wondering if much of the concern you have raised can be alleviated to at least be considered relatively insignificant by human hearing performance standards by meeting some appropriate test system (the PC used for the test and the signal source)design requirements.

1. Use a periodic reference signal which is designed to be guaranteed to be "good enough" as compared to the sound card specs such that the error between reference signal cycles is beyond the ability of the sound card to resolve, ie maybe a highly accurate signal generator putting out a square wave. This making it easier to get the effect of better time alignment of the before and after signals during comparison will give us a higher level of faith that waveforms sampled at different times through the same system can be used for comparison and assumed to be equal even though they are not the SAME waveform. I believe that an appropriate square wave contains all of the needed information but a dirac would be even better if you can somehow generate a good dirac (kidding about generating the dirac of course).

2. In order for this system to meet standards useful for human hearing comparisons it is necessary to guarantee sufficient sound card quality, ie accurate to the limits of human hearing for all the potential sources of signal digitizing/analog conversion errors. This will allow us to believe that the soundcard isn't introducing some audible distortion.

If you can meet both of these requirements then you will be able to test to the limits of human hearing which by my logic is good enough. Also you gurantee that audible differences will be visible in this tests output data. Am I wrong?

Point taken about unknown soundcard specs. Jitter, linearity, SNR, samplerate/conversion rate, etc can all be had in quantities that will be good enough to make each associated error boyond the limits of even the most high performance human hearing in existence. Granted the parts and associated board layouts don't come free but this isn't unubtainium to the well endowed design firm. I can't explain why the existing high end sound card manufacturers would not have already filled this niche if that is indeed the case as you have said. Maybe I am missing something here.


This post is made possible by the generous support of people like you and our sponsors:
  VH Audio  


Follow Ups Full Thread
Follow Ups


You can not post to an archived thread.