Michael Darnton

  • Content Count

  • Joined

  • Last visited

  • Days Won


Everything posted by Michael Darnton

  1. I guess this would be a more useful discussion if people would state their intent regarding the arching they use. Things like "replicate X", "achieve X tonal result", "make something I personally find attractive, without reference to something else". . . . something like that. It seems like the two of you (Peter and Dennis) are basically discussing personal preferences (choice #3 above) rather than specific objectives, right? That's a point of confusion I've seen throughout this discussion.
  2. If that applies to famous makers making bench copies, then apparently there's still a lot of thinning of the herd to be done, using your criteria, since they apparently weren't even able to accomplish the task with the source violin at hand while working. :-) I've had some delight a couple of times when great violins and their bench copies showed up at my summer workshop and my students were able to tell them apart based only on what I'd taught them about Cremonese arching and they noticed relatively important specifics in the originals that weren't in the copies. It seems too obvious to mention, but I will, that people tend to judge this type of thing based on their own standards, no matter how limited those are. So when someone assures you they've got it, that's hardly a guarantee that they do. :-) I'm seeing a lot of this in this thread, which is why it's not really been a fruitful discussion. But more seriously, I think a lot of makers are aware of their own shortcomings. I see that when I go to violin maker websites one one of the things a lot of them don't have is pictures of those makers' own violins. Better to talk the big talk then put out the images for scrutiny. :-) And I ALWAYS go to sites to see what the big talkers have actually accomplished. It's very instructive!
  3. Oliver Rodgers was the King of the Violin Spectrum, and I learned a lot of things from his writings and from people who knew him. If you are really interested in the topic, you should look at things he's written, IMO.
  4. There are a LOT of ways to consider this problem, and several things are being manipulated. I split it into two categories (at least). One the resistance effect of the tailpiece being closer or farther from the bridge (ref, CTanzio's comment)' and the other is the absolute pitches of the afterlength notes. It seems obvious that these won't necessarily be optimized at the same moment on every violin. Another factoid is that the resonances added can be good or bad and can be moved around (ref Bill Merkel's comment). I try to adjust to favor the one that will have the most positive influence. My violin-playing business partner insists that when the notes are spot on it makes a difference, and I'd attribute this to the fact that if all of the afterlengths were precisely tuned, the resulting notes fall within the scales, harmonics and primary chords of most of the common keys, so it that almost 100% of the time one or the other of these afterlengths would be ringing along with the note being played (ref Carl Stoss' response to CTanzio). But if the afterlengths are between notes, nothing at all is going to be reinforced by this type of resonance. If you don't think that would matter, try the experiment of playing the open G string, then just touching the D string at the third finger G position and noting the difference in sound. If you believe in amplifying the zone around 3KHz, notice that the E string afterlength is right in there with a Hill tuner, but not with a fork tuner. Something to at least think about. . . . Another thing that should be obvious is that violin string lengths don't vary enough to get out the calculator, and the silk wrap on the end of the string messes up the 1/6 rule by adding weight to that part of the string. The effect of this is that 54.5mm afterlength for the G string is going to get you very close to the note you want. And don't forget that these notes can be tuned or untuned by moving the bridge, having unrelieved tension on one side or another of the bridge, etc. etc. It's a fragile adjustment!
  5. Choice of different french curves and placement, choice of different circles. . . .from where I'm sitting it's the same defect in concept either way.
  6. Edge thickness and scoop depth are tied together and depend on how you use the templates. If you want to change those, you need to change the templates. There are a couple of different possible ways to visualize making the templates, so there is variation there. These are small windows--you are kind of forced to make a Cremonese arch, not something else, if you work with the templates. No, the modern arch form that you showed in your drawing cannot be made with cycloids, nor would they have wanted to do that. Both Stradivari and del Gesu tended to experiment a lot, and at the end of both their working lives started moving towards a more modern arching idea. Also, supposedly both of them looked to Brescia for inspiration at times. If you believe that, then there's a wrench in the process for that period of experimentation. There are a couple of other variables. One is that there was probably no law on the books that required makers to follow the templates if they thought they'd want to try something else. We often see that makers are experimenting. For me it was interesting to see that most experiments tended to derive from those templates in some way, by using them in a different way from usual, not going to something completely different. Also, makers varied wildly on their ability to make a violin shaped object; even among the Cremonese there were better and worse makers who were or were not capable to execute things well and precisely, and had variously good ideas. It would be silly to argue that some violins not fitting the plan somehow "proves" that the rest don't.
  7. You can make as many templates as you feel you need. When Sam Compton came to my workshop he made 35 templates for the back he was making. I use five. If you use half templates you can move them in and out, as I already said. I have noticed that the scoop outside of the scoop bottom varies, probably both from moving the templates but also because making edgework is somewhat of a separate problem with its own difficulties. However, it is easy to make a set and work with what they give you if you have something that works. I assume everyone knows how to make them with a simple disk and a pencil. [Maybe not. Disk of circumference to roll one cycle, hole for pencil half the arch height distance from center of disk. Put pencil in hole, roll disk along on straight edge, drawing cycle with pencil.] I don't see the inability to make anything you can imagine as a problem. The Cremonese didn't do that, and I'd regard that idea as a faulty observation. Their arches are recognizable because of their many commonalities, not because anything goes.
  8. It's more for a movie than text but the bottom line is that the templates themselves dictate the varying shape of the scoop around the instrument. The precise width and placement at each template is determined as a combination of edge thickness and scoop depth. If you have the templates and are willing to follow them keeping the edge thickness and scoop depth the same, you get the result. It's not particularly esoteric, nor is there any winging it. Using the exact same set of templates, I can fool you into thinking the arching is changing a lot more than it really is. I haven't found it necessary to make additional sets of templates once I figured out what was going on---that the arching grows upward from the edge, not down from the center, and that if any precision is sacrificed it's in the center, not the edges.. To make templates, basically you make the height and width you want, such that with a consistent scoop depth the height of the edge is also consistent, or some variation on that if you believe, for instance, in adjusting edge thickness in certain areas such as in the c-bouts. The reason I asked about scoop is because I believe that the most important aspect of the arching is the area of rising concavity inward past the boundary what you are calling scoop. The concavity that's missing in your example but which is a nearly inevitable feature of Cremonese instruments. IMO, that's where a large amount of the Cremonese type of sound comes from. It's a very easy area to mess up, too, with just one errant pass of the scraper, and it can't really happen except by intent--this is not a place where "the tools do the work". I think this is all I have to say about it.
  9. As I understand your statement, a person in 1650 in Cremona would not do this as a specific goal, nor would he look at the problem this way. Do you have a particular violin in mind to refer to so I can get a handle on the whole situation? I wonder what your definition of "channel" is? It might help to consider the first violin I showed against the Messiah Strad, since they are both extreme statements in opposite directions. On the Messiah, the scoop is moved so far out that it can't be deep, which it isn't or the tail of the template would cause the edge to be cut down thinner. On the lasered example, the scoop is moved so far in that for the edge to be a reasonable thickness the scoop has to be made extremely deep. Even with this, the edge on that violin is 5mm thick in the outer bouts, thicker elsewhere! In neither case was the edge altered to give a specific result---the results--depth of scoop, placement, edge thickness--are results from the placement of the template inwards or outwards. In the laser example, I know the maker well enough to know that he often gets in trouble shaping the lowest parts of the rising arch, and I think he just kept digging until he finally had to give up because the arch is getting too narrow, the scoop too deep, the edge too thick as he moved the half-template inwards to cover his mistake. This is a habit of his that expresses itself in a number of ways in nearly every one of his violins. The fit of the templates is more casual, too, because he genuinely found himself in trouble and you can see him casting everything aside to cover his ass. I know of a similar del Gesu, and on that there's a chip out of the maple that he chased as far as he dared, but he was moving into strange territory, so he stopped and left a bit of the chip hole rather than go into the same zone as the other maker, the one of the laser..
  10. Not at all. But it doesn't have the authority of the result a carefully explored judgement.
  11. Like Marty, you seem to not have done the homework.
  12. Above, the example you yourself showed as an example of what you're doing, is a parabolic-like shape through most of it, more curved in the middle, straighter and straighter towards the purfling, but with no inflection until the last possible moment, at the point that is level with the edge height, with essentially no recurve within the body of the arch and bears absolutely no similarity to any Cremonese with the possible exception of the last couple of del Gesu years. The second example I show is in the main exactly like yours except he doesn't realize he has to turn until a bare couple of mm after you do. However, both have the same overall problem, which is an arch which is in the main parabolic with no functional recurve. That's an almost uniquely modern line for an arching, used by lots of contemporary makers. My point is that you were not forced to avoid doing this, so you don't have a system that enforces a Cremonese arch. It's just a system that can be made to approximate some aspects of a Cremonese arching without the necessary degree of order. To put it very simply, your system has a high degree of "winging it" towards the edge, and the very illustration you chose to use shows not how that can go right but how it can go wrong.
  13. Had you bothered to play with cycloids, which no one ever does before criticizing them, you would see that this is exactly what they do as the height/width ratio changes. In the minimum, with the height of the cycle set to zero the "curve" is a straight line. As it becomes taller, it becomes more serpentine, then the round center expands and the recurve becomes smaller, the inflection point moving constantly outwards until all recurve disappears. Eventually it becomes an all-convex looping pattern This exactly follows the progression in a Cremonese violin as the arch moves towards the center becoming higher and narrower then back out again to wide and low. AS I SAID the curves in various circumstancess look different but are always mathematically related. All of the forms below are generated simply by moving the height of the cycle while keeping the circle diameter the same.
  14. My intuitive models are that the universe is random and the earth is flat. That's why intuition is a bad leader. Try another model: that a vibrating surface suspended around the edges will benefit from that suspension being mathematically consistent all around. That being inconsistent will confuse the vibration. Now bring that inward to all the other areas in turn. You are suggesting that you think that the Budweiser beer wagon should have a mix of several different types and sizes of horse rather than a matched set of clydesdales. You might be right, but maybe not. One reason I like cycloids is because all of the cross arches are generated by the same equation, adding certain unity to the situation from one end to the other. Any system that just estimates the scoop . . . as *all* of the others in this thread as well as any other you can name do . . . does not have this unity. Cycloids unify edge height, scoop placement and depth, inflection point, and central arch IF you take the time to understand them. There are a number of concrete, not theoretical, reasons that I believe, having played with them for 20 years now, they were the method of the Cremonese makers. Mostly these points may have dropped out in my comments but not all and if so they haven't been discussed in one place. I believe the evidence is compelling, but since there's a decided negative opinion already established I am not interested in going through it all yet another time. I've noticed that retention/memory of things discussed here is just about zero over time, so this is not the proper place for that. The first violin I posted was pivotal for my understanding because it forced me to consider some things that were previously only implied rather than so boldly stated in other Cremonese violins. I believe this particular violin shouldn't be held up as typical Cremonese because it's not, so don't get too wrapped up in it. It does prove some important points though.
  15. The scoopy violin is classical Cremonese, with an illustrious past as a player: the second is a modern German-school violin that no one seems to want to buy. The similarities of that one to drawings posted in this thread is a potential discussion.
  16. David, as far as your system represents a whole system, it would lead to this, I think.
  17. When you brush the idea aside by saying there's not a single simple rule, that's denying that there's a system. We disagree on that, obviously. The rest of what you said isn't a system, it's a series of indefinite pieces of advice. As I said, you don't have a system.
  18. It's become increasingly obvious to me over the years that you can't say you have a Cremonese-derived arching system if it doesn't define the shape, location, and depth of the scoop, as well as the thickness of the edge. The part in the middle doesn't actually seem to matter all that much, though it's gotten a lot of the focus, and has a definite shape in the Cremona prototype. Really, "Cremonese" style lives in the area from the inflection out to the very edge.
  19. Absolutely. But is that a bad thing? Even with that, you can't fit, for instance, a Cremonese normal template on a Carl Becker violin, no matter how you move it around.
  20. That's the case on most modern violins and more recend German and factory violins, but it is rarely the case on the Cremonese violins of most interest, where it's somewhere inside the purfling, if only sometimes a mm or two, but often much more.
  21. Mottola's error is to assume with cycloids that it's a whole template or no template at all. Cycloid half templates work much better.
  22. Instead of bow saw, look for a turning saw. I have had one for years and don't like it much. I suppose that's because I don't practice with it. I think the gouge is about a 35mm #3 by our numbers. It could possibly be flatter, but #3 is good. The rib bending is done by magic. :-) Tavisher: cool, I didn't know that tool.
  23. I assume you know that by limiting the short end like that, and presumably not holding it in place, you're begging for a nasty accident.
  24. David's just being polite. :-) My problem with Muratov is that you can derive his method from the violins, maybe, but you definitely can't derive the violins from his method. That is, give someone a french curve and some gross dimensions, and there's not guarantee that anything close to a Cremonese outline will result. A real method should be capable of drawing only the thing you intend it to draw, and nothing else.
  25. If you want to know what to make, I would pick some particular violin that had the sound I liked and try to understand the arching of that one, trying to replicate as closely as possible If you're doing that visually, it depends on your powers of observation and opportunity, and probably won't happen instantly. Either Sacconi or Muratov would be "correct" if you liked the violin that those different models gave, or you might like something else entirely. I'm generally a bit suspicious of Sacconi because he appears to have tried to blend all of the Strads he'd seen into one representative average, and in the grads that results in something that is indeed the average, but doesn't actually follow any real single Strad that I have seen. In my opinion, Muratov's method isn't related at all to Cremonese, but that doesn't mean you wouldn't like the results. Adding to that, there are Strads and del Gesus that don't work particularly well, so you might want to try to figure out why and avoid what you think is going on with them rather than trying to average them into your ideal model.