TalkBass has been independent since 1998. Add your voice.
Create a free account to reply to discussions, view embedded media, and browse with fewer display ads.
Join free Log in
Want zero display ads or expanded classifieds tools? Compare plans.
They may have introduced a few controls for the materials, construction and design, but no controls at all for the human element - the part in which nearly all of the bias comes into play. There may be a difference, but this trial doesn't tell much. He knows which is which, and he plays them while having that knowledge. This does not control his bias being introduced into his own playing. A player can subtly change their attack for different tones and good players can do this more readily than novices. I'm not saying that's what's going on here, but it's a consideration that was not controlled in this "test", and could absolutely confound the results. I am definitely a sore thumb in this topic. These tests including those by our members are never designed well. They make a lot of effort, but fail put enough effort into the study design component.A "clinical" test?They may have introduced a few controls for the materials, construction and design, but no controls at all for the human element - the part in which nearly all of the bias comes into play. There may be a difference, but this trial doesn't tell much. He knows which is which, and he plays them while having that knowledge. This does not control his bias being introduced into his own playing. A player can subtly change their attack for different tones and good players can do this more readily than novices. I'm not saying that's what's going on here, but it's a consideration that was not controlled in this "test", and could absolutely confound the results.
![]()
I hope this helps to clear up some of the strange stories that keep circulating about stainless fretwire. Most of the horror stories originate from the Parker Fly instruments (which used a hardened steel wire) or from some of the stainless wire that was on the market 20+ years ago, which was quite difficult to work with.
Mechanical engineering and social science experiment design don't always overlap nicely.I am definitely a sore thumb in this topic. These tests including those by our members are never designed well. They make a lot of effort, but fail put enough effort into the study design component.
It would also be nice to have more than one recording of each. If there were for example 10 recordings of each, maybe differences in playing technique would start to average out. But we only get two recordings, one each. So maybe the difference we here is just by chance. If we recorded it again, maybe the opposite would be true.Mechanical engineering and social science experiment design don't always overlap nicely.In this experiment, if the player did not know what he was playing when recording the tracks, that would have at least partially controlled for his bias being introduced (not that he's biased, but technically, it's bias). Even though there is still a lot left to control, it would be best that he not know which neck is which, but even further, he should not know what the experiment is about, what is being tested, or how he could influence the results.
If he'd just been handed the guitar and asked to play a bunch of stuff, and told that he'd be asked to play it a second time, that would have been better. Then take the guitar, ask him to do a meaningless task (i.e. sort these numbers in ranking order), then handed the next guitar (without knowing it's a new guitar) and asked to play again, and then compare the recordings (no videos) A/B and ask the listener to identify which tracks had more "top end zing". It might have turned out to be "the second guitar" but perhaps not.
I certainly hear much more top end zing in the video, but then again, it's just been suggested to me that I will hear more top end zing, and so according to over 100 years of solid scientific research using gold standard clinical trials, because of that suggestion, I will be more likely to hear it than not...![]()
Still need to consider the number of players and listeners!10 recordings of each and 3 guitars with each material! We need to hit the magic sample size of 30 after all.