The Machine

The recent discussion on the possible differences in force generated by different types of weapons on the SCA rapier field has led to myriad issues and “discussions” amongst the Society’s rapier marshallate. Baron Llywd of Atlantia has conducted a series of tests using a Machine based on Don Tivar’s original.

I agree that it is well past time to test the truth of the claims being made, and to determine if concerns and fears are founded or not. Unfortunately life has a way of interfering. After gathering the materials, construction was put on hold for some weeks.

Before I continue, I must thank and commend Llywd for conducting the testing. The testing itself is an extremely time consuming process. Not to mention the crunching of numbers and data input afterwards.

The Hypothesis

Of note, we have a hand full of fighters in Atenveldt who have some experience with both longswords and the rapier spears under discussion. Though the number is low, I think it significant enough to represent a measure of safety and the difference in a trained rapier spearman, such as those authorized in Caid and the West.

My expectation is that our experienced, trained spear wielders will demonstrate a much lower increase in force generation with the spears as compared to heavy rapiers. I expect an increase in force generation that is greater than a combat speed thrust with rapier, but less than that generated by a lunge with a rapier.

Conversely, Atenveldters seldom employ longswords in Heavy Rapier, either in tourney or melee, preferring to use them for Cut & Thrust Rapier. Because of the limited use of the longsword in heavy rapier, the generally stiffer blade, and the predominant use in cut & thrust, I expect the impact force will average somewhat higher than those of the rapier. Beyond this I will not suggest a hypothetical outcome.

Taka-Kensai’s Machine

I chose Llywd’s general design as the base because I wanted to be able to compare apples to apples. I have to admit I was a little intrigued by the “oomph” behind the shots measured during Llywd’s tests; I wondered if Atlantians really were “thugs” as Llwyd mentioned.

Building on Llywd’s machine, I built a machine using the lessons learned through Llywd’s Mark I through Mark IV machines.

I started with a longer base, knowing I was going to be using the same 100-lb fish scale. A few minor tweaks, like putting bolts all the way through the hinges and the base at key points, will hopefully give it a longer lifespan.

The target is oak plywood. But I reduced the size of the striking surface to a 3.5” square and reinforced it with strips of the same oak with the grain at right angles. The target is also secured with a bolt that goes through the lever arm rather than just a screw.

I used nylon washers between the lever arm and stabilizing arms to reduce friction.

The Machine-held down by ratchet straps

Everyone who tested or even just poked at the Machine noted the same thing: the most significant flaw noted in this machine is that the striking surface is incredibly rigid. It in no way simulates the feeling of hitting a relatively soft and yielding target; nor does it accurately represent an active target. The plan is to contrive a padded cover to attempt some simulation of a softer target at least. I don’t know if it will change the overall results one way or the other.

The machine is now ready to be put to use…

Data Collection

The machine debuted at a local fighter practice in the Barony of Tyr Ysgithr (Tucson, AZ). Like many of the groups in Atenveldt, they conduct their practices in the evenings. This meant the testing started at dusk and the light only got poorer from there. There are lights on in the local park, but it is a far cry from daylight. This is, however, the norm for the fencers tested.

As I had it readily available, I added an additional version of the spear (explained below). After the first outing, I also determined that a single point of collection for heavy rapier data was potentially biased and decided to broaden the data with the addition of a couple different attack techniques for heavy rapier. As I was working from the assumption that Llywd is baselining against a close measure thrust at combat speed, I added a stepping thrust and a lunge.

Each fighter also self-declared their level of experience with the various weapons. Each fighter declared themselves to be new to the weapon, a beginner, intermediate, or experienced fighter with each weapon. Descriptions of each are on the data spreadsheet.

The testers included 3 white scarves, one member of the Chivalry, six fighters that practice armored combat, three of those that primarily fight armored combat, and nine fighters that are authorized in cut & thrust rapier. Additionally, three of the testers (me included) are relatively experienced in use of the rapier spear, two others are beginners with the weapon; and five of the testers have used the rubber HEMA/WMA spear tips on a hickory shaft. So far a pretty good mix I think.

Heavy Rapier

First, we don’t do light rapier in Atenveldt. Because of this, I didn’t bother to test epees, foils, or modern sabers on the machine. Llywd already established the lighter blades have a lower impact force than his baseline. And since the longswords and spears are heavy rapier weapons, it really isn’t necessary to test them again.

Then I made a change to how the heavy rapier was tested. Rather than having everyone use the same 35” long schlaeger, I allow testers to use the rapier they were most comfortable with, one which they use in practice, tourneys, and melees. My thought behind this was that it would be more likely to get an accurate average impact force for fighters using heavy rapier than handing someone a sword they had never handled before.

I annotated the rapier blade length and manufacturer (when known) each fighter used.

This average was still used as the baseline for comparison. In spite of this, quite a few chose to use the 43” or 37” practical Hanwei swords I had on hand. The rapier variations used are annotated on the raw data sheets.

I normalized against a close measure thrust at combat speed. Fighters stand just far enough back from the machine to have to rotate the shoulders and/or lean in order to strike the target. Fighters closer than this demonstrated a penchant to thrust at combat speed, abruptly stop short of the target, and then slowly tap the target.

The step thrust is an advancing step with a similar close measure thrust. The lunge employed is a more modern lunge, rather than the Capo Ferro style period lunge.


The longsword testing was all done using a Darkwood longsword for consistency. This longsword meets the 1” flex test for use in heavy rapier—but just.

Testers administered a one handed rapier style blow, the same as the close measure thrust; a two handed thrust at a similar measure; and a “harpooning” thrust that involves thrusting with both hands, executing a passing step, then releasing the front hand to strike the target with the rear hand on or near the pommel.

Miyako-Hime with Longsword


I have had an Alchem pike for some time, making that weapon readily available. I also have the Amazonia spear included in the SRM’s proposed rules. It consists of a 5 foot haft with a 37” practical Hanwei blade screwed onto the end and stabilized with a set screw.

Testers administered a fixed hand attack, placing their hands wherever they wanted along the shaft; a pool-cueing attack, keeping the rear hand in the back third of the shaft and the front hand in the front two-thirds of the shaft; and then a stepping thrust with fixed hands, called “controlled combat”.

I must comment on the “controlled combat” attack method. I have used spears for a couple years, both in single and melee combat; and I have been involved in Caidan melees in which spears were employed. This stepping forward-thrusting attack is not one I have either used or observed. It is much more common and likely that a spearman will be stepping backwards and thrusting repeatedly at a fast approaching enemy. I do, however, understand the intent behind this attack and hopefully employed it faithfully.


Will-first time ever seeing the spear

EDIT:  Due to some technical issues on my end, most of the data collected at the first test outing was almost all lost. I recovered a small amount of it and have included it in the data. The half dozen fighters whose data was corrupted will be retested as soon as possible.

Also, I noted that Llwyd’s most recent data included a number of data points from individuals who hadn’t tested in all weapons, including a couple who didn’t test with the schlaeger at all. I determined, as much as possible, to include only test results from individuals who tested on all weapons and variations. The data from the testers on the first outing are currently the only ones who have not tested on the two additional technique variations for heavy rapier, though they performed all other tests.

Summary of Results

Here's a quick summary of the results. Keeping with Llwyd’s methodology, I normalized the results so that the heavy rapier close measure thrust at combat speed results were the nominal "average" value and am reporting the percentage for each weapon as a comparison to the nominal value:

  • Rapier: 100% of Rapier Force
  • Rapier Step-thrust: 120% of Rapier Force
  • Rapier Lunge: 143% of Rapier Force
  • 2 Handed Darkwood used 1 handed: 107%
  • 2 Handed Darkwood used 2 handed: 165%
  • 2 Handed Darkwood harpooning: 172%
  • Spear w/ 2 fixed hands: 131%
  • Spear w/ sliding hands: 154% **
  • Spear at controlled combat form: 156% **

Experienced Spearmen

At this point I am able to diverge from Llwyd’s testing. Atenveldt has a handful of fighters who are familiar with the rapier spear.

While Atenveldt has not been conducting an experiment, we have had one of the Caidans out more than once to introduce the Alchem Pike to our fighters. (Many thanks to THL Meala Caimbeul.)

I have had an Alchem pike since Caid and the West began their experimentation. I have been practicing with the weapon on my own, with members of my household and barony, and at my HEMA academy for some time. THL Malise MacKendry has similar experience and had been tapped as the primary spear marshal for Atenveldt. We have two others currently in Atenveldt who have been determined to be “trained spearmen”:  Lady Hamasaki Miyako has been practicing spears with me for nigh unto 2 years, as has THL Heinrich Loescher.

For this testing, this pool of four individuals would be the “experienced” spearmen for testing purposes.

The following is a summary of results thus far. All data is ONLY those individuals who are experienced with the spear:

  • Rapier: 100% of Rapier Force
  • Spear w/ 2 fixed hands: 109%
  • Spear w/ sliding hands: 104% **
  • Spear at controlled combat form: 167% **

** Note that testing was performed on the Alchem and Amazonia spears separately and compiled for the above numbers. While my testing did have few of the spear data points top out over 50 pounds, I used a 100-lb scale from the beginning. )Thanks again, Llwyd.)

The full raw data can be found at the bottom of the page.

Discussion of Results

All of the information above is as factual as possible. This section starts to include opinion and analysis.

First, the possible issues with the tests:  this is the first time many of those tested had ever used either of the weapons in question, much less used them at actual combat speed. The data is a relatively small sample, especially when considering those deemed experienced with spears. And as with Llwyd’s testing, this test only measures the peak force of the blow rather than total energy. All in all, it is what it is; any ideas to improve the testing are welcome.

As I mentioned earlier, the first data collection demonstrated to me that the current test is designed, intentionally or not, to favor demonstrating both 2-handed swords and spears strike harder than rapier. With only 3 data points per fighter for rapier, all from a controlled stance, being compared to 9 data points per fighter for longsword and 18 for spear, with at least an attempt at combat motion, the data collection and comparison seemed flawed to me. Additionally, the stepping thrust and lunge demonstrated some interesting data points.

Lastly, I collected data on each fighter’s experience level with the weapon being used. All experience was self-declared. The following experience levels were noted:

·         New/Unauthorized – Never seen the weapon, never used it, or not authorized in rapier

·         Beginner – Someone who felt they were novices; Used the weapon once or twice; possibly only in training and never against an opponent

·         Intermediate – Someone who felt they were moderately knowledge and familiarity; At least a year of regular practice and bouting

·         Experienced – Someone who felt they had significant knowledge and familiarity; Several years of regular practice and bouting

You will note, and it was made clear, that experience level did not equate with level of skill on the field.


Thus far the data suggests that longswords

·         Hit with significantly more impact force when used in 2 hands than a heavy rapier thrust

·         Hit with a greater impact force than a stepping thrust or lunge with a heavy rapier

Hitting with 165% of the impact force of the rapier is somewhat concerning. However, when compared to the impact force of a combat speed lunge, it becomes a bit less disconcerting. The data indicates, at the very least, that more training is required for those who use these weapons in heavy rapier.


The results for longswords are consistent with those gathered by Llwyd. However, the peak force compared to a lunge with a rapier is considerably less scary looking, being only 20-30% higher; stiff but not devastating. While the test very obviously demonstrates higher force levels using two hands on the weapon, it also demonstrates the lack of control and feedback from the longer striking distance used during the harpooning attack.

These results indicate a separate authorization for these weapons might be a consideration.


Thus far, the data suggests that spears are:

·         Significantly safer in the hands of a trained spearman

·         As safe in the hands of a trained spearman as a heavy rapier.

Further, the data suggests that an untrained spearman hits with an impact force slightly greater than that of a lunge using heavy rapier.

The data with experienced fighters is quite small, being only three of the four experienced spearman available. However, it represents an average size and strength female fighter (5’7”, 135-140 pounds), an average size and strength man, and a strong and not-small man (5’11”, 230 pounds).


The results of the spears differ greatly from Llwyd’s data. To be honest I don’t have an explanation for why results in Atenveldt were so different than those in Atlantia. The majority of fighters have little to no experience with the spear in both kingdoms, and I am approximating the testing done previously as closely as I am able.

My best guess would be that it is cultural. All of the force measure averages are lower in Atenvledt than Atlantia. In spite of a reputation for hitting hard, it is possible that Atenveldters aren’t the jack-booted thugs we are portrayed as (I must admit this is something of a disappointment…)

Looking at the results for what they are, it becomes apparent that training spearmen significantly reduces the impact force across the spectrum tested. This is true both as a percentage of heavy rapier impact and in actual pounds of force generated. While the data set is small, it thus far confirms my hypothesis:  trained spear wielders demonstrate a comparable amount of impact force with both heavy rapier and spears.

Considering the results with the data collected so far, I would advocate the proposed separate authorization process for spears has potential to mitigate harder impact forces.

Ideas for additional experiments

First, I would echo Llywd’s encouragement for others to build machines and collect data. While not a rocket scientist, I am an intelligence analyst and strongly favor verification, validation, and review. I would especially like to see the West and Caid conduct this testing. I understand we may have a visit from Meala in the near future and I am looking forward to testing her on the Machine.

I will continue to broaden my data samples. Most of the fighters tested so far have been relatively “new” to rapier and the SCA. Only one amongst those tested can be considered amongst the “grey beards” of Atenveldt rapier fighters. I am curious to see the results of their tests and will make special note of those folks. Atenveldt is in a relatively busy season and I should be able to get a broad range of data from several upcoming events, including testing the same fighters again to compare their results.

Finally, one of the more electronically inclined fighters is developing a pressure sensor that can be employed by a fighter in or under the armor. Hopefully this will get us much more “combat-accurate” data across the board.

For questions please contact me at or

Master Llywd of Atlantia started the current round of testing using the machine. He has tested approximately 71 fighters thus far. And has a nifty site up showing the saga of his machine (which has seen considerable change over a short time.)

William Bomar,
Aug 10, 2014, 7:30 PM