Hello Thomas,

Thomas Riley wrote:

As I look through the book, it strikes me that almost none of the swords have this kind of ratio for the linear agility, why am I getting such tall ovals in the sword I am trying this on? was the constant "ko" different for the X and the Y axis?

No it was the same. What you're getting here is simply because your measured pivot points are so close to the hilt. If you look at swords of the same overall proportions in the book they are not in the same place at all.

Additionally, you still have a discrepancy in your two pairs of pivot points. See the figure I've quickly drawn here; normally the red circles (pivot points as diameter) and the red line (vertical from CoG) intersect exactly at one point above the sword and one below. If you take the other pair for your computation you should find the same result, which is not the case here. So, I think there is some problem with how you measure these.

If you prefer to work from the pendulum method as I think you did for the effective mass curve, then I can work out the formulas, but time is short right now! And it's better in my opinion to do this with pivot points, precisely because they allow for some error checking...

Regards,

---

 Attachment: 95.2 KB
[ Download ]

Awesome, okay I have a better understanding of how I need to approach this. I this round I did everything according to your process, but I will adjust my pivot points at the handle and work on getting more accurate measurements. Thank you, again!

With the new an improved formulas, and very careful measurements. I think I finally have the accurate, and complete model:

[ Linked Image ]

Many thanks for Vincent, for putting up with all my questions.

If anyone wants to create their own models I have this handy tool thanks to a YouTube person named "A dwarf sitting on a giant's shoulder". All you have to do is plug in your own measurements in the orange fields:
https://ggbm.at/EVGqGhP4

Looks very good, the two pivot points seem to be more logical that in your last picture!
And thanks for this hand tool, I think I will test my swords using it... I'm rather curious to see the result.

First of all thank you all for this very educative topic!

Using Mr. Riley's tool I made a graph for my blunt bastard sword made by Lutel.

What information can I extract from it?

- Surely my sword is dedicated for cuts (is actually of type XIIa )

- The cone and the oval can give me an impression about the potential accelerations ( I think they are reasonable, not too small, not too large)

Other than that? What else a person that has never swing my sword can understand?

Thank you in advance,
Giorgos

---

 Attachment: 25.75 KB

I also had a go using Thomas' tool plus Inkscape enhancement (hopefully attached) for a sword that Mark Vickers (St George Armoury) made for me last year.

The tool/method clearly does tell us some interesting things about the sword and, as a way of presenting the dynamic properties of swords, Peter and Vincent's work (and Thomas' interpretation) is such a big step forward. It may not yet be the perfect analysis/presentation tool and the differences between swords do not necessarily jump out from page but I'm pretty sure Vincent and Peter are working on refining it, as Thomas hinted at in one of his recent Youtube videos. I know I'm interested to see what they come up with next!

Simon

---

 Attachment: 181.92 KB

Being a newcomer to this discussion I must first say that I am very happy to see that this method for showing the dynamic properties of swords is focus for some interest.

The reason this specific tool cam into existence was a spin of from the exhibition "The Sword - Form & Thought" at the Deutsches Klingenmuseum, where I was part of the curatorial team.

Vincent and I had talked at rather irregular intervals over the years before this, and I knew he was the man for the task, when I wanted to include some sort of graphics to express the dynamic properties of the swords in the exhibition. I had long wanted to have a way to present the handling of swords in a way that did not favour one feature over another. I did not want a situation where you first of all would be looking for "The Best" but rather describe swords in a way that made objective comparison possible. Some visual signals that could give you some idea of the individual properties of different swords.

Vincent leChevalier is the wizard in all this. He is the man who worked out the mathematics.
I have gathered a body of data over the years and could contribute with my ignorance and lack of understanding. -If Vincent could show something in a graph that made sense to me, then surely it would make sense to others as well! :-)
I wanted to see the subtleties of sword dynamics expressed in Orc-speak, rather than Wizard-lingo.

The resulting graphs grew to be a rather complex combination of features, but I think that this is a good thing: it express the fact that there are several aspects that work in concert or in opposition in defining the handling and dynamics of a sword. This set of images and accompanying numbers can give you a profile for a sword.

By showing the properties of several historical swords alongside each other you can see if there are some common trends that characterise a group, or not. You can see similarities between types and within types. There are trends for some borad groups that are clearly illustrated with these graphs.

This way you can compare contemporary swords and see in what what they are similar to or different from their historical counterparts. It is well worth noting that if we make swords today according to the typology of Oakeshott, there is really no guarantee that the resulting swords will have much in common to their historical counterparts. -Just because a sword *looks* like a XIIa or a XVa, it might have very little in common in function and handling characteristics (dynamos) to a historical sword of that type.

Vincent and I are currently working on making this tool available online. To properly use this tool (to get good results) you need a little bit of introduction to the practicalities of documenting dynamics as well as a quick tour of the theory behind the dynamics calculator. There are some really interesting trends already visible in the material, and this should be outlined.
We will publish this in an article in the hope that ideas like this can have an impact on current research on swords.

Please have patience with us! :-)
There will be a lot of interesting things possible once the top becomes available for all.

A word of caution: the tool is very sensitive. It can show really small differences in dynamic properties. This is necessary for a proper illustration of the variation of these features in swords.
First of al, this means that the measuring of the dynamic properties is critical. The so called "waggle test" can be rather inexact. With experience and sound methods it can become more precise. This takes a bit of practice! Be aware of this. The more you do this, the better the results will get. There are a few tricks to do it with greater reliability and I will make a video demonstrating how to not the dynamic properties in swords in a way that will provide fairly good results (that will get even better with some training).
Another thing to be aware of is that you will be able to detect differences between two individual swords that are made to the same pattern. Since the individual dynamic "mugshot" of a sword is so detailed and sensitive, it is bound to show even small variations. Take to this the differences that invariably will occur between the taking of measurements/data points between individuals.
This needs to be taken into account when discussing the results.

You could get more reliable results if you involved some kind of mechanic equipment when documenting the dynamics, but this is rarely practical or even possible. That data would be constant to a specific set of equipment and so only limit the amount of human error in one line of investigation.

There is a lot to reflect on here.

I hope it will spark some interesting new discussions and perhaps even a deeper understanding and greater appreciation of the importance of dynamic properties.

Thanks for inputting Peter, I look forward to seeing more. I know I for one struggle to get consistent results with the waggle test so a tutorial will be most welcome.

I'm currently getting bits together to build some simple (but hopefully effective) and portable apparatus to perform the 'waggle' test in a consistent manner and it should give an indication of the size of error caused by my bad technique, or indeed the accuracy of somebody else's good technique!

I'm also currently working on a scanned 3-D model and trying to create a motion analysis simulation that I can compare against the other two methods.

It will be interesting to see if modern engineering tools make a significant difference to the data.

Such an informative and useful discussion!! Great to see.
Thanks to all involved.

Jon

Hello Simon,

Obtaining consistent results with the waggle test is something we've discussed quite a lot with Peter over the course of this work. The tool we'll publish has the capability of showing you how close to the theoretical ideal your measurements are, as long as you measure several pairs of pivot points. At the very least it gives an idea of how big the margin of uncertainty ends up being.

I have also developed my own device that uses a much more precise and consistent pendulum test, which I intend to write about in due time. For the moment it is a bit too rough for publication :) This has allowed me to cross check my own precision, and has made me more confident that the waggle test can indeed give good results. You just have to account for a bigger uncertainty, but generally speaking not too much bias.

Care must be taken as well with the pendulum test. The way it is often done, by hanging the sword by the cross, is not much more precise than the waggle test, as I have found out. That is because the axis of rotation is not sufficiently constrained. Suppressing all human involvement is part of the solution, but not the complete one :)

The great advantage of the waggle test is that it can be used "on the field" with very little gear. We wanted the tool to be usable with this very simple, low-gear method as a priority. It will possibly be expanded later to enable other measurement methods as they are developed.

Best regards,

Hi Vincent

Yes, I guess with multiple pairs of points you should be able to check consistency using the 'intersecting circle' method you showed earlier in this thread?

The experimental set-up I have in mind is simple, portable and I think could be used (with slight adaptation) for both pendulum and waggle tests, and the bits to make it will cost about £10 (which equates to about 10 euros this week). I take your point about suspension method for the pendulum test and I'm working on a 'gripping' method to constrain the axis of rotation. Admittedly it's still nowhere near as simple/practical as picking up a sword and waggling it but, as I said, I'm a bit rubbish at that so far!

Best regards

Simon

I tried to use Thomas' tool to see the result of an historical sword I had the chance to measure - a sort of end 16th century/beginning 17th century military rapier. Here it is:

[ Linked Image ]

Here are the results, very quickly done (and without any care at all) :

[ Linked Image ]

I guess there was something wrong with the pivot points, as their "circles" don't cross on the PoB line. The fact is that this weapon had surprising properties, e.g. the PoB, which is very far to the cross (21cm). At the time, I told myself that the blade maybe was made to be fitted on a heavier hilt.

These measure were made like one year ago, and at the time I had absolutely no experience in measuring, so it gives another explanation...