Need some help, please. I would like to show if there are measurable differences between speaker cables. Important to add that I really don't care what the difference are. Don't intend to have same lengths or same terminations. Just want to show if it can be measured. And if so how small/big differences are we talking.
I don't believe that if I can hear a difference it won't show up on a graph somewhere. My thinking at present is that these are so small that my ordinary way of looking at these isn't really effective. Or simply I'm measuring the wrong things.
How do I go about this? My idea so far is this. Would like your input. I have 3 speaker cables available for this. Two are single wire one is bi-wire, one set of jumpers made of one of the single wire cables. To make it fair I will only measure it as single wire to remove what the jumpers could be doing.
Measurements:
1) Calibrated mic will be placed in the same position and not moved during measurements. Mic position is irrelevant as long as it remains the same
2) Amp will be set to a the same volume. No changes will occur during measurements. Laptop will feed amp's DAC.
3) Speakers will be connected single wire to HF terminals (250Hz and up) each time from same amp terminals
4) Each sample size will be made out of X number of measurements per cable (to eliminate environmental noises)
Comparisons:
1) Comparison will be done electronically. Overlaying two measurements and taking the difference. Assuming there's no difference end result should be a straight line on frequency graph.
2) To account for environmental factors first measurements of same cables will be overlaid. This should give a base line for comparison and would show margin for error. The should ideally be as close to flat as they are taken under same conditions with no significant difference.
3) Next measurements of different cables will be overlaid. Resulting measurements should show a difference (if any) between the cables
4) Differences will be compared and presented. I will look at other graphs as presented with REW to see if any patterns emerge.
Questions
Can you see any significant flaws with the above? Can it be improved at all? I don't expect to spend a day doing it so propose a sample size of 20 sweeps per cable. Would you agree this is enough to have some reasonable conclusions?
Side note
This could easily go under Tech forum but could be quite interesting for everyone. It's not meant as a wind-up to either cable sceptics or cable believers. My opinion has always been there are differences between cables, however they're nowhere near as big as cable believers would have you think. Nor that they can't be present and are only bias as cable sceptic say. As such my main speaker cable for the last year has been a solid core copper 2.5 mm2 (bi-wire) which retails for around £3 per meter delivered. The most expensive cable used here will be £96 per meter, with the third mentioned priced at £5/m.
Weather permitting and assuming work won't get in a way I might be able to have the results presented today.
I don't believe that if I can hear a difference it won't show up on a graph somewhere. My thinking at present is that these are so small that my ordinary way of looking at these isn't really effective. Or simply I'm measuring the wrong things.
How do I go about this? My idea so far is this. Would like your input. I have 3 speaker cables available for this. Two are single wire one is bi-wire, one set of jumpers made of one of the single wire cables. To make it fair I will only measure it as single wire to remove what the jumpers could be doing.
Measurements:
1) Calibrated mic will be placed in the same position and not moved during measurements. Mic position is irrelevant as long as it remains the same
2) Amp will be set to a the same volume. No changes will occur during measurements. Laptop will feed amp's DAC.
3) Speakers will be connected single wire to HF terminals (250Hz and up) each time from same amp terminals
4) Each sample size will be made out of X number of measurements per cable (to eliminate environmental noises)
Comparisons:
1) Comparison will be done electronically. Overlaying two measurements and taking the difference. Assuming there's no difference end result should be a straight line on frequency graph.
2) To account for environmental factors first measurements of same cables will be overlaid. This should give a base line for comparison and would show margin for error. The should ideally be as close to flat as they are taken under same conditions with no significant difference.
3) Next measurements of different cables will be overlaid. Resulting measurements should show a difference (if any) between the cables
4) Differences will be compared and presented. I will look at other graphs as presented with REW to see if any patterns emerge.
Questions
Can you see any significant flaws with the above? Can it be improved at all? I don't expect to spend a day doing it so propose a sample size of 20 sweeps per cable. Would you agree this is enough to have some reasonable conclusions?
Side note
This could easily go under Tech forum but could be quite interesting for everyone. It's not meant as a wind-up to either cable sceptics or cable believers. My opinion has always been there are differences between cables, however they're nowhere near as big as cable believers would have you think. Nor that they can't be present and are only bias as cable sceptic say. As such my main speaker cable for the last year has been a solid core copper 2.5 mm2 (bi-wire) which retails for around £3 per meter delivered. The most expensive cable used here will be £96 per meter, with the third mentioned priced at £5/m.
Weather permitting and assuming work won't get in a way I might be able to have the results presented today.
Facebook
Twitter
Instagram