Archimago;673806 Wrote: > Yup, I had a skim thru the other reviews as well and noted the excellent > performance of those you listed- many excellent measurements, along with > many questionable high priced units. It looks like about 21-bit > resolution is the highest out there... maybe that's the limit of > testing equipment as well? This is the best answer i can give with my slightly confused level of knowledge. I'm sure others can do better.
I have always understood that the level of thermal noise sets a floor for resolution in the region of 20 bits.Basically electrons just don't sit still so there's always some random current going on. As I recall the thermal noise is at an overall level of -120db which is equivalent to the signal to noise ratio of a perfect 20 bit system. (try looking up Schott noise and Johnson noise.) Nevethless don't forget that you can resolve tone below the level of the noise floor. You can resolve a tone at -115db in a 16 bit system with a noise floor of about 93db after dithering, but that's because the noise level in the region of the tone is much lower. That is still with 16 bits of resolution. You can see this on JA's tone sweep and FFT illustrations which show a floor of much less than -93 Db becasue that is the total noise level not the level in each "region" of the frequency range You can therefore (I think) still resolve a tone below the overall level of the thermal noise ie a tone below -120db....it's just that you can't do much (any?) better with 24 bits than with 20. If I have got this wrong be gentle. -- adamdea ------------------------------------------------------------------------ adamdea's Profile: http://forums.slimdevices.com/member.php?userid=37603 View this thread: http://forums.slimdevices.com/showthread.php?t=91950 _______________________________________________ audiophiles mailing list audiophiles@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/audiophiles
