Audio Cable Frequency Response Measurements
Got curious that so many people can hear clear audio differences between their cables, saying things like 'night and day', 'even my wife in kitchen can hear'.
Note:
1. No one disputes that there are electrical differences between cables.
2. No one disputes that there are audio differences between cables.
3. But between various properly designed, short length home audio cables, placed in sensible locations, without strong magnetic interferences,
there should not be huge differences that human ears can detect.
And comparing wires not meant for audio use, to thousand dollar audio cables is just....trying too hard to make a sale?
So to see if my ears are really that bad, I set out to measure me some cables. Starting with some interconnect cables.
Interconnect cable comparison (yes no exotic cables
):
1. Generic yellow / red / white rca cable, 1.8m
2. Monster Standard THX i100 'THX certified' rca cable, 1.2m
3. Belden 1694a rca cable, 2m
Source: Windows 10 PC (DIY)
USB DAC: Topping D10 (w/ generic usb cable)
Amplifer: Yamaha A-S500
Speaker cable: Belden 8471
Speakers: Dali Fazon Sat
Subwoofer: NA, turned off
Mic: Umik-1 (~8cm away from speaker face)
Measuring software: Room EQ Wizard (REW), ~80dB Pink Periodic Noise
Only 1 speaker is measured, with grills on. In HDB room.
It should be clear from the charts that differences between the measured frequency responses of the 3 cables are so small (less than 0.5dB, the 3 lines are on top of each other) that it is almost impossible for human ears to pick up.
1/48 smoothing applied to frequency sweep graphs.
*Noise level in my room (PC fans, high storey, but roads ~30 meters away)
Zoomed in to frequency sweep graphs of only Belden 1694a and generic cable (so easier to see), smoothing turned off. Hardly any difference that's more than 0.5dB.
Download the REW file with the measurements here:
https://www.dropbox.com/s/ufij74sqvfgdtkr/Interconnect%20measure.mdat?dl=0
Based on the graphs, I think my system and speakers played all the frequencies (above 60Hz) so the system should be 'resolving' enough, at least sufficient for the mic to record everything.
Of course, there're 101 variables that weren't accounted for when doing the measurements. Sure, maybe the cables are doing something magical below 60Hz and I couldn't capture that. So take the real world data and decide whether it's useful or not to you.
Will measure more cables in the future, in the hopes of finding ones that show substantial, more than 1-2dB differences, and see whether 'brighter' cables are really brighter and show an increase in higher frequencies.
If you measure your cables and do a cable shootout, do share them here as well.