Asher Guitars WD Music Products Amplified Parts Mod Kits DIY Nordstarnd Pickups Warmoth.com
Asher Guitars WD Music Products Amplified Parts Mod Kits DIY Nordstarnd Pickups Warmoth.com
Asher Guitars WD Music Products Amplified Parts Mod Kits DIY Nordstarnd Pickups Warmoth.com

PLEK Machines (from zombie thread)

Discussion in 'Tele-Technical' started by conecaster, Aug 27, 2010.

  1. murrmac123

    murrmac123 Tele-Meister

    197
    Jan 25, 2008
    Edinburgh
    OK, the situation as I postulated it earlier is , we have two "identical " new guitars, neither of which have previously received a proper set-up, one is treated by David and the other one goes to the PLEK machine.

    Both guitars are played by, say, 100 experienced players , and the comments noted.

    If all 100 players say they prefer David's guitar then the issue is pretty much closed as far as I am concerned. If all 100 players say they prefer the PLEKked guitar, then likewise. Obviously, if the proportions of preference varies from 100% to 50% then the issue becomes less clearcut.

    I fail to see what the point of introducing multiple guitars would be.

    To me, the above experiment seems watertight, but hey, I am willing to listen and learn.
     

  2. David Collins

    David Collins Tele-Afflicted

    Sep 28, 2009
    Ann Arbor, MI
    I don't think anything about this particular trip to scan is really about what the setup means to the final player, but simply an effort to dispel the myth (or confirm the claim, depending on which side you're arguing ;)) that hand lapping procedures are not capable of the same or greater precision and tolerances of the Plek machine.

    I've already dressed several guitars that had been previously Plekked, and found errors that far exceed my tolerances, and also set up and inspected many that meet them quite superbly. This confirms to me personally, that the machine is capable not only of the same precision as hand lapping, but also the same errors if not competently set up and run. I've not had any of my work (that I know of) scanned on a Plek however for a review in the other direction, and that is the request I'll be aiming to accommodate here.

    If the results scan out exactly to or within the range of what the Plek would prescribe to be ideal, then the argument would presumably be over (yeah right, :lol: there will always be something else). If my fret work shows individual errors all over the place beyond the Plek's prescribed tolerances, than it's over as well, at least in terms of claims made of my own personal work.

    If it scans out to a very precise and accurate line that matches with my own goals, but those curves happen not to agree with prescribed relief patterns of the Plek, then even though the precision issue woule be resolved, it would have to shift toward another argument of which pattern is better. This is where if anyone really wanted to proceed with such testing, a much more exhaustive survey of player preferences would likely have to be carried out. We're not there yet though, and I should probably just wait to see if the relief patterns of one simply substantiate the other before worrying about that argument.
     
    Last edited: Oct 2, 2010

  3. David Collins

    David Collins Tele-Afflicted

    Sep 28, 2009
    Ann Arbor, MI
    Murrmac - if testing for end results in player preference, then fewer guitars tested by more players would be a more worthy survey. If testing consistency and repeatability of work however, then more guitars (intentionally including many with squirrelly and uncooperative necks) would be a better test of whether a human can account for and control these variables more or less reliably in dressing frets.

    I have to be honest that the very design and methodology of the Plek machine gives it a great advantage here, and can fully understand skepticism as to whether humans can deliver results as consistent and reliable as a properly run Plek. I still hold that it can be done, but it just takes a good deal of skill and experience. I would love to bring a dozen guitars in for testing to address the issue of accurate repeatability. Aside from not having the time or instruments to dress and bring however, I also don't want to take up valuable time on the machine just to settle an argument. So, there will always be repeatability to argue about, regardless of how this one or two instrument scan checks out.
     

  4. David Collins

    David Collins Tele-Afflicted

    Sep 28, 2009
    Ann Arbor, MI
    Sorry to pummel our deceased equine friend even further, but one more minor point.

    This isn't really anything about me or any competent luthier against the Plek. It's more about me against the perceptions many have of computer controlled machines achieving tolerances unachievable by hand. I actually know of at least three Plek operators whom I've had conversations about this with, who hold no such notions at all. To them, and with their experience with the machines, they have said just the same as what I've argued - that it's nonsense to think that it can do anything we can't. It can do as good as a human, but much more quickly, and can do it all day long (something I certainly wouldn't want to do).

    So it seems the perception of Plek precision being unobtainable by other means exists more in the public than the profession, though obviously not exclusive to either. Anyway, it's not the machine I'm up against here, but rather perceptions.
     

  5. murrmac123

    murrmac123 Tele-Meister

    197
    Jan 25, 2008
    Edinburgh
    I have to say , David, that your posts on this forum are invariably absolute models of lucidity, logic, and literary perfection. Would that the same could be said of all others ...:(

    Point fully taken about the different issues concerning repeatability btw.
     

  6. David Collins

    David Collins Tele-Afflicted

    Sep 28, 2009
    Ann Arbor, MI
    Thanks Murrmac - I have to say that's one of the highest compliments I could ever hope to receive!

    I lost most of my vocabulary and any ability to communicate effectively with an injury many years back, and had to basically relearn much of the english language. After going through a period of being relatively unable to communicate, the aim to accurately and effectively share my thoughts and intentions has been a strong passion of mine. Of course good logic is something I've always held as sacred, but it's a true honor to be commended for effective communication. Plus, my wife is a professional technical writer, so I get shamed if she ever spots any mistakes. ;)

    I'm sure that gives some psychological explanation as to why my posts are always so damned long. Thanks again!
     

  7. Colt W. Knight

    Colt W. Knight Doctor of Teleocity Ad Free Member

    Jan 21, 2007
    Tucson, AZ
    Because something like a guitar CANNOT be identical. You would have to have many and get statistical "averages" in order to rule out anomalies or flukes.

    So if one guitar played better than other to the majority of people, you could only say that ONE single guitar was favored. You could not say that people favor that one technique over the other.

    But if you had 50 hand done guitars and 50 pleked guitars, each played by 100 players you could use actual statistics to determine which is better.
     

  8. Colt W. Knight

    Colt W. Knight Doctor of Teleocity Ad Free Member

    Jan 21, 2007
    Tucson, AZ

    I don't often disagree with you David, but I am going to on this test thing.
    Without Consistency and Repeatability, there is no actual analysis. You are just testing which guitar the majority of people prefer. So if 67 of a hundred people prefer the X over, that just means that 67 people prefer X. What if the next hundred people prefer Y? You don't know, because you didn't test. Could just be a random effect.


    I am no statistician, and I don't want to drag out the text books from grad school. But when we designed experiments, we had statistical formulas that would give us how much data we had to have in order to reach statistically viable numbers. And even then, the experiments would need to be repeated just to show there weren't any flukes or just plain statistical anomalies.

    I was part of a reproductive study one time where we had a nearly 90 % conception rate of heifers(unheard of) with a certain type of estrus synchronization protocol. The professors thought they struck gold, they had solved a costly problem for livestock producers. The problem was, it must have just been luck because they could never repeat the numbers.
     

  9. David Collins

    David Collins Tele-Afflicted

    Sep 28, 2009
    Ann Arbor, MI
    You're right Colt. I was just trying to distinguish the goals of testing consistency and repeatability of hand dressing procedures, from the testing of player preferences should different relief and setup patterns be chosen by different technicians or Plek's programmers. This current comparison in progress is only a test of precision capabilities of course, and not really a test of either of these other issues.

    My suggestion that a test of preferences between different relief or setup styles could be done on fewer instruments, would have to be preceded by establishing consistency in the methods used to achieve them. Of course without this, any tests would indeed be pointless. Three separate issues I think are being discussed here though. First is achievable tolerances and precision, which is what is being tested here. Next is consistency and repeatability, which is not being tested (unless someone want's to pay me to dress a dozen instruments, then pay a Plek operator to scan them all). Beyond that would be the test requiring a player survey, which would only have any real purpose if precise tolerances and repeatable patterns were already established, and if there proved to be a significant difference in relief patterns of choice.

    Of course I think such a test would be inconclusive anyway. If you dressed two identical guitars one way, and two the other, then picked out three at random, I doubt 90% of players would be able to tell which one was different if they were simply minor relief pattern changes. If short range precision is good on all, and patterns are at least generally similar, then I just don't think many players would distinguish them. Of those who could tell the difference, I think the choice of which were preferred would depend more on playing style than anything else. If a player picks with fingers, light or heavy picks, favors striking a slightly different position on the string, or approaches with a slightly different angle of attack, then the way the string moves is going to change drastically.

    Every player's ideal setup is of course going to lean toward a different compromise based on these, and dozens of other variables. There is of course no universal better or perfect relief or setup pattern, so results in this type of survey would probably rely too much on individual preferences to make for a reliable survey anyway.
     

  10. Slepperer

    Slepperer TDPRI Member

    22
    Sep 24, 2010
    North Wales
    Surely it is unfair for David's job to be measured on a Plek machine. As stated before, the Plek machine will measure it to it's own idea of straight if a Plek guitar was given to David, or measured with a technofret bar then it might be found not to be straight in relation to the tool used to measure it.

    You can't use the thing that made something straight to determine whether it is stright! The only way to definatively test would be to get an independant measure.

    Either way, I don't think anyone could tell the difference when the tolerences are that small. As has been said, the human hand cannot manufacture anything to a tolerance as small as a machine. If this is the case, how can the human hand distinguish whether or not the tolerance is smaller?
     

  11. Nick JD

    Nick JD Doctor of Teleocity Ad Free Member

    That's the thing though isn't it? Plek machines don't machine it straight, and that's what makes them better.
     

  12. Slepperer

    Slepperer TDPRI Member

    22
    Sep 24, 2010
    North Wales
    What I meant was they test their work with the same thing that did the work, meaning that the data in the test is biased.

    I'm not denying that they are birlliant machines and can do very high quality work in a fraction of the time of a person, however the claims of their accuracy are biased. And I don't think anyone can tell the diference beyond the work of a human.
     

  13. Greg.Coal

    Greg.Coal Tele-Afflicted

    Aug 23, 2009
    Colorado
    Not so sure

    I believe that devices like this are self-correcting and can, indeed, measure straight, if not actually make something straight. I assume that there are optics involved and bouncing light off of a constant reference point. PLEK probably didn't invent any technology but merely adapted extant tech to a new problem and machine vision and computer aided manufacturing were already pretty advanced.

    I don't know that for sure, though (regarding their optics and control). But, that's a pretty simple solution. Then, I do agree that some machine and some good guitar person can make a guitar wonderfully playable for us and we couldn't tell the difference. Having said that, I know that I myself sometimes think I have solved a problem but all I did was just get the setup over the line into playable whereas a machine or a David can do the work better than "good enough" so that the setup lasts much longer.

    I don't think the luthier and PLEK tech discussion represents an either/or proposition. I happen to be a fan of both the hand-made, traditional quality as well as the new tech. I am particularly interested in DC's experience with the PLEK (if he gets a chance to look at one) because he and a few other guys posting here have engineer/scientist level sensibilities and the review of the PLEK experience would be especially interesting and valuable from them.

    Greg
     

  14. Colt W. Knight

    Colt W. Knight Doctor of Teleocity Ad Free Member

    Jan 21, 2007
    Tucson, AZ
    Can't the plek machine just create a plane, and measure the distance of each fret from that given plane?
     

  15. Greg.Coal

    Greg.Coal Tele-Afflicted

    Aug 23, 2009
    Colorado
    It ascertains

    From the website:
    "The computer ascertains a 3-D like graph of the fretboard surface, including the position and height of the strings."​


    During the process of the scan, there has got to be some reference point, whether it's optical or physical. During the leveling, it probably works at the level of the fret and nut - since the strings are off and the neck has changed shape - and accounting for "straight" isn't important by then.

    On Star Trek they could speak to the Replicator: "Replicator: give me a cup of black coffee" and it would be formed form sub-atomic particles and appear. They never asked for a '53 Tele, though or a fret dressing.

    G
     

  16. Vizcaster

    Vizcaster Friend of Leo's

    Sep 15, 2007
    Glen Head, NY
    Oh so now I have to come up with an excuse for long posts? Nuts.

    Anyway, I'm anxious to see the results. And just to be a nudge, I'm wondering how much polishing the frets by hand takes off, and how much inconsistency does it re-introduce into the precious precision of the Plek machine?

    This is not a game of inches like golf. It's a matter of microns and I'm willing to believe that it does make a difference in playability and feel. Both of my Gibsons were Plekked and they play great (although of course I was not happy with the factory nut slot heights...). I'm sold on the idea that a factory should be Plekking their guitars because basically there's little or no fret dressing being done on new mass produced instruments, it's simply another feature that's a worthwhile selling point for me. But at the same time I have no worries about a hand-finished neck from the Master Builders at the Fender custom shop. Someday.
     

  17. fretsbr549

    fretsbr549 NEW MEMBER!

    1
    Jun 25, 2010
    USA
    So its been almost two years. What happened in the great face off between DC the Master Luthier and Plek the Great?
     

  18. Bartholomew3

    Bartholomew3 Friend of Leo's

    Dec 8, 2010
    Montreal
    I just bought a Les Paul studio with an ebony fretboard that was done on a plek machine and it's extremely well set-up.

    Just as good as the best Luthier in my area who did my tele and a vintage LP.

    Only had to sand the sharp edges on the outside of the nut near top E - I do that with everything anyways.
     

  19. Ronkirn

    Ronkirn Doctor of Teleocity Vendor Member

    Age:
    71
    May 1, 2003
    Jacksonville, FL
    A Plek will do an amazing job of leveling a guitar’s frets, a good tech will do an amazing job leveling a guitar’s frets.

    Choosing The guitar that has the best fret leveling job has a lot to do with which method you prefer. Some cringe at the idea of a Computer controlled anything doing something to their guitar, some love the idea of computer controlled “precision” device working on their guitar….

    Some love the idea of a hands on old fat bald tech hand leveling frets, some cringe at the notion dementia may set in half way through the job…

    But in either case, precision, and the professing the superiority of one’s method of achieving such over another’s would have Cervantes chuckling, as he recalled writing of the tilting a lance at windmills.

    Stability of wood is often discussed as a definitive “improvement” too….…. And while some species are more stable than others, there is NO, as in ZERO types of wood that is 100% stable.

    SO if you grind pieces of metal stuck in wood to a precise degree of level relative to all the other hunks of metal sticking in the wood and do it to within a few angstroms…. So gawddammm what… the blanking hunk of wood they are stuck in will swell or shrink, warp and bend like a drunk sonuvuhbeech the moment you stop working on it. Fortunately for us all, when it DOES swell or shrink, warp and bend, the vast majority of necks out there will constrain all their swelling or shrinking, warping and bending to an acceptable margin of tolerance. When it exceeded those tolerances, we throw it away and screw on a new neck.

    When you take your guitar from a hot day of playing in some outside venue in August, then show up at a “refrigerated” studio the next day and the guitar is barely playable… what do you think happened? The Wood moved…when it did, it took the frets with it… How’s that Plek precision thing lookin’ now?

    You Plek guys… stoppit with the “were more precise” crap… your problem is the machine costs what a Mercedes costs, a really high end Mercedes, get the things down to where a shop can afford to do a fret job at a competitive price, or figure out a way to keep the wood the frets are stuck into still…and you won’t have to throw out the bull anymore, we’ll all be using ‘em. But a hundrerd large for a fret leveling tool… what loon thought that would fly.

    If you have a graphite neck, maybe, just maybe one method would have an advantage over the other… but do any of ya really want a graphite neck, yeah, me neither…so all this jibberin about my fret job is more level than yours is about as dumb as claiming my hunk of wood is more resonate than yours…

    rk
     

  20. Jupiter

    Jupiter Telefied Ad Free Member

    Jun 22, 2010
    Osaka, Japan
    I hope the whole debate's not going to start up again; I just wanna know what happened in the John Henry vs. the steam drill thing!
     

IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.