S/N and Slew rate ?'s

Printable View

  • 03-04-2008, 10:20 AM
    blackraven
    S/N and Slew rate ?'s
    Is there really any audible benefit of a S/N greater than 95? I read a few articles stating that there really is no audible difference between S/N or 95 and 110 or more. Any thoughts or opinions on this matter?

    Also, is there an audible difference between slew rates of 15v/us and lets say 50v/us?
  • 03-04-2008, 11:08 AM
    Feanor
    Personal issue
    Quote:

    Originally Posted by blackraven
    Is there really any audible benefit of a S/N greater than 95? I read a few articles stating that there really is no audible difference between S/N or 95 and 110 or more. Any thoughts or opinions on this matter?

    Also, is there an audible difference between slew rates of 15v/us and lets say 50v/us?

    I have tinnitus; this ensures that my personal S/N ratio is always worse that 95dB regardless of the equipment.
  • 03-04-2008, 11:13 AM
    kexodusc
    I'm of the opinion the better numbers don't hurt, but sometimes spec for the sake of spec isn't the best way to build an amp. In other words, in my experience, the best sounding amps might not have the best S/N or slew rate.

    I cannot sit and tell you which amp has the best of either just by listening as well. I tip my hat to the man that can.
  • 03-04-2008, 12:35 PM
    mlsstl
    For home playback, by the time you're over 90 dB in signal to noise ratio there are other limitations that'd be more productive to spend time addressing. For example, it would be unusual for most home listening environments to have a background noise level that is lower than 20 dB or so. That is very quiet. Add 90 dB of signal to that and you're at 110 dB.

    Also keep in mind that precious few recordings actually make use of such a wide S/N ratio. Things get pretty spectacularly dynamic sounding if the difference between the quiet passages and the loud ones in a recording is only 30 or 40 dB!

    One place where better S/N ratios can help is in the recording studio where one is mixing multiple channels together. It is good to avoid the accumulation of background hiss as much as possible, especially when one track being mixed in is at a higher gain level than another track. Even then, you still have the residual noise of the microphone preamps, mixing console and ADC to contend with.

    As far as slew rate, this is correlated with the high frequency response. Also known as "rise time" it is how fast the amplifier can react to a leading edge transient. It is typically tested by looking at the amp's response to a square wave input and measuring how many volts the output rises within one microsecond.

    The catch is there are no square waves in music. Certainly a slew rate can be so slow as to measure poor high frequency transient response in an amp, but I don't think you'll find 15 V/uS to be considered "poor." In short, for most good amps, you'd not want to base the purchasing decision of an amp on the two numbers you've tossed out. I'd be far more concerned about the behavior of the amp under various reactive loads with complex material. Once you've got a specification to a respectable number, it is either time to move on to another spec or start looking at the totality of performance under real-world conditions.
  • 03-04-2008, 03:00 PM
    O'Shag
    Mlsstl,

    excellent answer. Kexodusc is also correct. I will also add that numbers such as slew rate and S/N ratio might be a more meaningful indictator if many manufacturers didn't exaggerate results. Having said this, I think its useful to review the specifications as a starting point to reviewing and auditioning components..
  • 03-04-2008, 03:41 PM
    E-Stat
    Quote:

    Originally Posted by blackraven
    Any thoughts or opinions on this matter?

    Simplistic metrics like this (or THD) provide little useful information. How was S/N ratio measured? At what input voltage? At which frequency(ies)? Using which weighting? In the presence of which RF generators, i.e. CD/DVD players, TiVOs, cable boxes, DACs, etc.?

    The means by which the S/N ratio was measured by my Threshold, Audio Research, VTL, and NAD electronics were all different in one way or another.

    rw
  • 03-04-2008, 09:23 PM
    hermanv
    For classical music, the average sound pressure level is probably down about 20 dB, for quiet passages only about 45 dB of signal to noise is left (maybe even less).

    Having extra headroom can hardly hurt. I have a 250 Watt/channel amp, I probably play it at less than 10 watts , but my experience tells me that the amp sounds quite a bit better than the 25 watt amps I've heard.

    So I for one, support continuous improvement in S/N ratios. Still I wouldn't make a purchase decision based on any one spec or specs in general, I trust my ears and those of a small select group of equipment reviewers.