Wireless fencing
Adrian Murphy
article:
http://www.smh.com.au/olympics/fencing/20000124/A42839-2000Jan24.html
Adrian
Dr. Hans M. Rupp
attractive to the viewer than using the traditional equipment?
I cannot see any advantage.
I would be nice to hear something from a FIE official.
Dr. Hans M. Rupp
(I still remember the chaos at Chaux-de-Fonds vividly, I could tell many
points which would make fencing more attractive to spectators.)
John Twernbold
were. Anyone know more? And does anyone know *which* wireless system was in
use?
John
siobha...@my-deja.com
"Dr. Hans M. Rupp" <hmr...@uni-tuebingen.de> wrote:
> Can anybody please explain me why wireless fencing should be more
> attractive to the viewer than using the traditional equipment?
The FIE are very deluded to think that clear masks and wireless scoring
are what make fencing attractive. The proof of that is the Supermasters
competition. This is a competition between the best of the best, the
World Champions versus the World Cup Champions, and is run as a bit of
corporate entertainment for the IOC. It's wireless and they wear clear
masks, but despite the calibre of the athletes, it's BORING. It's boring
because nobody cares. The fencers are just there for a paycheque -
the competition is meaningless to them. So they don't really try, and
they certainly don't scream or show any kind of passion. Actually,
that's not quite true, last time one of the French male Epéeists tried
to put a bit of emotion into it, but the poor guy was on his own and it
didn't quite work :)
The World Championships and Olympics are tense and exciting because they
*matter*... the men's Foil final (France v. China) in Korea was
fantastic, because it went 44 all - traditional European fencing power
v. emerging nation. Ok, so the FIE might say that the clear masks help
with that, help to see the emotion on the faces. The fact is, on TV you
can't see the faces properly through the Lexan anyway - and isn't it far
more exciting to see a fencer rip their mask off and let out a almighty
roar?
I definitely agree with Hans as well - though I can see the arguments
behind the clear masks, the wireless fencing appears pointless to me. A
friend of mine did make the point that technology won't move forward if
everyone just accepts the status quo, and I agree with that - there's no
reason not to research the possibilities for wireless fencing. But the
FIE are pushing it in when the technology isn't ready and the fencers
are unhappy. The only reason they should replace the current system with
a new one should be that the new one is superior - they've got the whole
thing the wrong way around.
Siobhán.
Sent via Deja.com http://www.deja.com/
Before you buy.
Bryan J. Maloney
> Can anybody please explain me why wireless fencing should be more
> attractive to the viewer than using the traditional equipment?
> I cannot see any advantage.
Doubling the weight of the blades would have a better effect for the
viewer, I'd say.
--
"Before we judge the lobotomist of old too severely, we
should go to the nearest street grate and see how we are
dealing with our mental health crisis today."
Bryan J. Maloney
> corporate entertainment for the IOC. It's wireless and they wear clear
> masks, but despite the calibre of the athletes, it's BORING. It's boring
> because nobody cares. The fencers are just there for a paycheque -
> the competition is meaningless to them. So they don't really try, and
> they certainly don't scream or show any kind of passion. Actually,
Maybe it's BORING because the style is so esoteric and attenuated that the
general public would have no interest in it. Change the rules so that the
FIE maximum weapon weights are *minimum* weights. Double the maximum
weights. Require that the weapons balance no further back than 1cm down
the blade from the guard. Require that foil and epee blades have a
minimum stiffness and enforce those requirements. That'll make it a bit
more exciting to watch.
Harlan Harris
Dr. Hans M. Rupp <hmr...@uni-tuebingen.de> wrote:
>Can anybody please explain me why wireless fencing should be more
>attractive to the viewer than using the traditional equipment?
>I cannot see any advantage.
I'm not sure why it would either, but once perfected, it should be
considerably cheaper to purchase and maintain than the existing wires, which
is more than enough reason to switch for this impoverished fencer in an
impoverished club.
There's certainly a more valid argument that the transparent masks are more
attractive to the viewer. And the counter-arguments are primarily financial,
just as the arguments for wireless are primarily financial...
-Harlan
Eric Dew
Dr. Hans M. Rupp <hmr...@uni-tuebingen.de> wrote:
>Can anybody please explain me why wireless fencing should be more
>attractive to the viewer than using the traditional equipment?
>I cannot see any advantage.
>
(Not a FIE official)
EDEW
Eric Dew
John Twernbold <jtwer...@remove-this.yahoo.com> wrote:
>Very interesting! But the article didn't elaborate on what the problems
>were. Anyone know more? And does anyone know *which* wireless system was in
>use?
>
to the wrong nests.
EDEW
Eric Schlaepfer
>
> Very interesting! But the article didn't elaborate on what the problems
> were. Anyone know more? And does anyone know *which* wireless system was in
> use?
I've been studying wireless scoring for some time. I suspect the
problems they are having has to do with the detection of an electrical
closed circuit without a ground return. Normally the ground return is
provided by the reels, but with wireless scoring manufacturers have had
to figure out other means.
The floor touch cited by the article seems to confirm that. The big
problem is that the method they use to detect floor hits might be
sensitive to EMF, static electricity, radio stations, and other things.
With a well-designed system these things really shouldn't happen. I
think wireless scoring really is the future, but I think the problems
need to be fixed before the system is used at major tournaments.
BTW, I think the system that was used was designed by the Ukrainian
company Sportservice and manufactured by Allstar.
Later,
Eric
Eric Schlaepfer
> Doubling the weight of the blades would have a better effect for the
> viewer, I'd say.
The FIE seems to think the same way as you do; recently the FIE made the
saber blades much thicker. They are now thicker than an epee blade.
Later,
Eric
Eric Dew
Uh-oh...the FIE and Bryan agreeing...first manifestation of the dredded
Y2K problem.
Actually, I heard that they are asking to remove the manchette (the overglove)
as well. The fencer's glove goes underneath the lame, but no manchette is
needed. (I heard this last week while watching a bunch of sabre folks
work out at Stanford.)
A manchette and a stiffer blade will make sabre more duel-like. No more
these funky "slap the guy underneath the guard as if we're playing tag"
shots.
Of course, it may be time to buy stock in padding manufacturers. I'm
selling my shares of Manchette, Inc. :-)
EDEW
John Twernbold
> BTW, I think the system that was used was designed by the Ukrainian
> company Sportservice and manufactured by Allstar.
Thanks for the info! I was able to find some pictures at the Allstar site,
including those ones from last year, with the mask-lights being used by the
saber fencers.
There are two different news tidbits (with pictures) here:
http://home.t-online.de/home/allstar/news10.htm
John
Jonathan Jefferies
"Bryan J. Maloney" wrote:
> Change the rules so that the
> FIE maximum weapon weights are *minimum* weights. Double the maximum
> weights. Require that the weapons balance no further back than 1cm down
> the blade from the guard. Require that foil and epee blades have a
> minimum stiffness and enforce those requirements. That'll make it a bit
> more exciting to watch.
Care to explain this theory? Why would heavier weapons that balance
further out make for a more exciting format? Mind you the minimum
stiffness MIGHT take out "flicking" at some cost to the fencers'
being more bruised or possibly even hurt. But why do you think
that would improve things?
J.
--
Jonathan Jefferies
Allant Software
1280 Civic Drive Suite 206
Walnut Creek, CA. 94596
Ph: 925-944-9690 x13
---------------------------------------------
Bryan J. Maloney
<jona...@allant.com> wrote:
> Care to explain this theory? Why would heavier weapons that balance
> further out make for a more exciting format? Mind you the minimum
Slow down the action a bit, perhaps encourage some caution on the part of
the fencers. Makes it easier for somebody not initiated into esoterica to
follow.
Bryan J. Maloney
> Hi Bryan,
>
> > Doubling the weight of the blades would have a better effect for the
> > viewer, I'd say.
>
> The FIE seems to think the same way as you do; recently the FIE made the
> saber blades much thicker. They are now thicker than an epee blade.
Let us hope this spells the end of the whipover.
Eric Schlaepfer
> It's the stiffness requirement (as measured with the gabaret) that has
> been changed to the same as epee, not the thickness. Both Blue Gauntlet
> and Blade had new-spec sabre blades at the South Bend NAC-- they're only
> marginally heavier than current blades, and don't look noticably
> different.
Stiffer. That's what I was told, but somehow I must have
"transmogrified" the word to thicker by mistake.
Thanks for the correction!
Later,
Eric
David W. Neevel
been changed to the same as epee, not the thickness. Both Blue Gauntlet
and Blade had new-spec sabre blades at the South Bend NAC-- they're only
marginally heavier than current blades, and don't look noticably
different.
Eric Schlaepfer wrote:
>
> Hi Bryan,
>
> > Doubling the weight of the blades would have a better effect for the
> > viewer, I'd say.
>
> The FIE seems to think the same way as you do; recently the FIE made the
> saber blades much thicker. They are now thicker than an epee blade.
>
>
> Eric
Simon Summerfield
>
> Very interesting! But the article didn't elaborate on what the problems
> were. Anyone know more? And does anyone know *which* wireless system was in
> use?
The system used was some Ukrainian invention. The Ukrainian
technicians were _supposed_ to turn up 2 days before the event
to do some testing, but they were so positive everything was
OK that they didn't arrive until the night before. It worked fine
in their labs 8-)
There were, however, numerous problems:
1) Some hits on the earthed piste were scoring points
2) Clearly distinguishable hits were being registered as simultaneous
3) Hits were registering when they clearly didn't
This last one was extremely rare, but did happen on at least 2
occasions. Once, a rapid flicking movement way away from the opponent
actually scored, and in another case, a hit was scored by one fencer
with his opponent's point nowhere near target, and yet both lights
came on.
I guess the FIE are right. The wireless system certainly made the
fencing more "interesting".
And on a couple of other points, perspex masks were not mandatory, or
at least the fencers refused to wear them. This was because the
Uhlmann/AllStar versions have recently had doubts raised about their
safety. There were a few people wearing the Leon-Paul ones though
(as the only ones that have not been proved to be dangerous).
And a definitive(?) ruling on the one foot off the piste rule...you
get a verbal "friendly" warning, then a yellow card, then red cards,
but it's up to the referee's discretion apparently. Some would enforce
it, and some wouldn't. If fencers stayed mostly in the middle of the
piste, they were allowed to get away with it more often than if they
constantly hugged the sides. But I never saw anyone actually get
carded - occasionally the verbal warning was given but that's as
far as the referees seemed prepared to go with it. It's like a
rule that exists but doesn't really.
Simon.
Douglas Weber
siobha...@my-deja.com wrote:
>
> I definitely agree with Hans as well - though I can see the arguments
> behind the clear masks, the wireless fencing appears pointless to me. A
> friend of mine did make the point that technology won't move forward if
> everyone just accepts the status quo, and I agree with that - there's no
> reason not to research the possibilities for wireless fencing. But the
> FIE are pushing it in when the technology isn't ready and the fencers
> are unhappy. The only reason they should replace the current system with
> a new one should be that the new one is superior - they've got the whole
> thing the wrong way around.
> Siobhán.
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
It has never been clear to me why wireless fencing would make any
difference, in
fact I am not sure why doing a round piste(how is that for an oxymoron)
would
work. The way European fencing works it is inherently linear. I can alway
rotate
faster than you can revolve around me-it takes less energy. This means I am
alway
presenting my point and my thinest profile to you. Except maybe in sabre,
you
cannotg attack at strange angles with normal stance and attacks. Only minor
angulation work. Why would we circle like boxers at all? And if there is
no benefit
to being able to circle then why would wireless have any effect on the
match?
Doug Weber
John Twernbold
> I am not sure why doing a round piste(how is that for an oxymoron)
> would work. The way European fencing works it is inherently linear.
I think one advantage to using wider and/or circular "strips" is the ability
to pass your opponent. I'm a sport fencer, so I know nothing about these
techniques, but I imagine some of our classical and historical fencers could
describe them. Of course, such strips and techniques aren't the main purpose
of wireless scoring, and would require radical rule changes for sport
fencing.
> And if there is no benefit to being able to circle then why would wireless
> have any effect on the match?
I think a good wireless system has the potential to be better than the
current wired system. As someone already pointed out, using wireless
transmitters and receivers would be more economical than the current reel
systems. And it will only get cheaper, smaller, and more reliable in the
future, like all electronic industries.
There's also the aesthetic appeal. To non-fencers, the current wired system
is pretty bizarre, with cables and cords all over the place. Wireless has
the capability of being "clean" looking, which I imagine is why it appeals
to Olympics officials.
Of course, the main problem is that the current system(s) are far from
perfect. It reminds me of the "video telephone": a futuristic invention that
has been promised for years, is talked about often, and is *still* not
ready.
John
Chuck Allen
siobha...@my-deja.com wrote:
>
> I definitely agree with Hans as well - though I can see the arguments
> behind the clear masks, the wireless fencing appears pointless to me. A
> friend of mine did make the point that technology won't move forward if
> everyone just accepts the status quo, and I agree with that - there's no
> reason not to research the possibilities for wireless fencing. But the
> FIE are pushing it in when the technology isn't ready and the fencers
> are unhappy. The only reason they should replace the current system with
> a new one should be that the new one is superior - they've got the whole
> thing the wrong way around.
>
Wireless is perhaps a theoretical ideal, (I never really had a problem
with reels!) but from a technical viewpoint it's got to be a disaster.
Not counting the RF interference issues, whatever is used must be as
close to 100% reliable as possible. overtime you add a part to a
device, you decrease the reliability of the device. For instance,
suppose you
have a device which uses 10 parts, each which has a reliability of
99%. The complete device is NOT 99% reliable but .99 x .99.... 10
times, which comes out to a reliability of 90.4%. Obviously, wireless
devices will use parts with a higher reliability factor, and obviously,
considering that you have to have a separate receiver and transmitter
for each fencer, it will use considerably more than 10 parts, intact
the wireless transmitting system might even use more parts than the
current boxes!
Only a few of the even the national armourers can fix boxes, what
happens when we go from reels to a wireless system which can't be
repaired except by shipping it off to the manufacturer? How much of an
increase in spares will then be necessary to run a tournament?
I am presently restoring a 40 year old motorcycle, it is doable, as
rust can be
removed, dents can be taken out of metal, and mechanical parts no
longer available
can (with enough input of cash) be fabricated. On a modern bike, when
its gets that old, it will have quite a number of parts which cannot be
fabricated on a onesy twosy basis, (or even a 100 200 basis!) and will
be unrepairable. A lot of clubs have to exist on old scoring equipment,
what happens to them when old scoring equipment no longer exists, as it
has become unrepairable? In addition, would you want to trust the
deciding point in the Gold medal bout in the Olympics to a device which
had the technical reliability of Microsoft Windows????
Even though (or perhaps because!) I spend some of my time
functioning as a computer consultant, I am very against putting any
computer chips in cars. Most of the time, critical mechanical systems
can be jury rigged to get you where you have to go, OTOH when an
electronic component fails it's DEAD, with no option other than
replacement.
Chuck Allen
Eric Schlaepfer
> Wireless is perhaps a theoretical ideal, (I never really had a problem
> with reels!) but from a technical viewpoint it's got to be a disaster.
> Not counting the RF interference issues, whatever is used must be as
> close to 100% reliable as possible.
Not necessarily. At MIT, there is some research going on about
electronic business cards. Somebody invented a simple technique to allow
data to be transferred through the human body at up to 2400 baud. The
technique involved an active band-pass filter that removed nearly all of
the RF noise, AC hum, radio stations, etc. from the incoming signal. A
similar approach could be used to detect touches in wireless fencing.
> overtime you add a part to a
> device, you decrease the reliability of the device. For instance,
> suppose you
> have a device which uses 10 parts, each which has a reliability of
> 99%. The complete device is NOT 99% reliable but .99 x .99.... 10
> times, which comes out to a reliability of 90.4%. Obviously, wireless
> devices will use parts with a higher reliability factor, and obviously,
> considering that you have to have a separate receiver and transmitter
> for each fencer, it will use considerably more than 10 parts, intact
> the wireless transmitting system might even use more parts than the
> current boxes!
Not necessarily. In saber and off-targetless foil, there is no need for
radio or infra-red communication at all.
At this point, radio communications are very reliable and secure. After
all, many expensive cars use a radio-controlled remote door lock.
I contend that a well-designed wireless fencing system would be far more
reliable than the current mechanical reel system.
> Only a few of the even the national armourers can fix boxes, what
> happens when we go from reels to a wireless system which can't be
> repaired except by shipping it off to the manufacturer? How much of an
> increase in spares will then be necessary to run a tournament?
Most machines these days are sent to the manufacturer for repair anyway.
Tournaments can use the same number of spare machines they use now, but
they no longer need spare reels or floor cords.
> I am presently restoring a 40 year old motorcycle, it is doable, as
> rust can be
> removed, dents can be taken out of metal, and mechanical parts no
> longer available
> can (with enough input of cash) be fabricated. On a modern bike, when
> its gets that old, it will have quite a number of parts which cannot be
> fabricated on a onesy twosy basis, (or even a 100 200 basis!) and will
> be unrepairable. A lot of clubs have to exist on old scoring equipment,
> what happens to them when old scoring equipment no longer exists, as it
> has become unrepairable?
They buy new machines. Because of the decreasing cost of technology,
scoring machines are becoming more and more affordable to the point
where it will be cheaper to buy a new machine than to custom fabricate
parts for an obsolete one.
> In addition, would you want to trust the
> deciding point in the Gold medal bout in the Olympics to a device which
> had the technical reliability of Microsoft Windows????
The wireless scoring system that has been used is, in my opinion, still
in the early prototype phase. A production-quality system would be very
reliable, much more so than Windows. Remember, computers are used in the
medical industry where someone's life could depend on a piece of
software.
Conversely, would you trust a deciding point in the gold medal bout in
the Olympics to a reel that uses mechanical brushes for electrical
contact? What if the contact breaks for just a few milliseconds so the
fencer doesn't get the touch?
> Even though (or perhaps because!) I spend some of my time
> functioning as a computer consultant, I am very against putting any
> computer chips in cars. Most of the time, critical mechanical systems
> can be jury rigged to get you where you have to go, OTOH when an
> electronic component fails it's DEAD, with no option other than
> replacement.
For the past 15 years, most cars were built with about six to twelve
computers. My father's car has a 6802 microprocessor handling the fuel
injection. The micro has worked for 18 years and has been used about
every day with no problem.
On the other hand, I strongly object to running Windows CE on my car.
When it crashes, it really *crashes*! ;)
Later,
Eric
Chuck Allen
Eric Schlaepfer wrote:
>
> Not necessarily. In saber and off-targetless foil, there is no need for
> radio or infra-red communication at all.
>
Excuse? But ho does the information get between the fencers and the
box, I mean without wires, light or radio - telepathy?
> > Only a few of the even the national armourers can fix boxes, what
> > happens when we go from reels to a wireless system which can't be
> > repaired except by shipping it off to the manufacturer? How much of an
> > increase in spares will then be necessary to run a tournament?
>
> Most machines these days are sent to the manufacturer for repair anyway.
Our Division has had to ship a machine back to Allstar once in last
decade. The rest of the time I have repaired the things, we have around
a dozen. The TCA boxes, of course, get sent back to Triplette. I have
upgraded the chips twice, and it looks like I will be doing it again in
the near future. Most of the boxes are 10 years old.
> Tournaments can use the same number of spare machines they use now, but
> they no longer need spare reels or floor cords.
>
No, they need spare transmitters or receivers or both (depending on
whether or not the they are integrated with the machine. Obviously the
machine needs to receive signals from both fencers, even if the fencers
transmitters don't have to talk to each other. If the equipment is well
maintained, you can often run a 10 strip two day tournament without an
equipment failure, even using 10 year old boxes and 20 year old reels.
To go wireless the Division would have to replace 12 boxes and 30 reels
and floor cords. If we do NOT go wireless, they will probably be good
for another decade or two, since the only reason for replacing a box is
only necessary because of catastrophic failure or the inability to be
upgraded, a problem which has been eliminated with the advent of
microprocessor controlled boxes. If you start looking at the equipment
cost to fencing in general, and take into account that we are not a rich
sport, the move to wireless fencing is probably going to be extremely
detrimental to fencing as a whole as it will sop up funds which could be
used to do something else which might really advance fencing, rather
than being used to provide a solution to a problem which, taken in the
overall scheme of things pertaining to fencing, is not really important.
If the use of reels was seriously holding back the sport of fencing in
some manner, it might be worth it, but no one has suggested that this is
even close to being true. In fact many of the "improvements" like off
target being eliminated in foil, facial area being eliminated in sabre,
hands being eliminated in sabre etc., which are the result of technology
not being adaptable, are already having a detrimental effect on the
sport.
>
> They buy new machines. Because of the decreasing cost of technology,
> scoring machines are becoming more and more affordable to the point
> where it will be cheaper to buy a new machine than to custom fabricate
> parts for an obsolete one.
>
Must be nice to be rich! This is true ONLY if the machines, including
the associated transmitters get down to maybe $100, as it rarely costs
more than that to fix an old box, a price I do NOT expect to see anytime
soon.
> The wireless scoring system that has been used is, in my opinion, still
> in the early prototype phase. A production-quality system would be very
> reliable, much more so than Windows. Remember, computers are used in the
> medical industry where someone's life could depend on a piece of
> software.
>
There is all kind of stuff promised by semi-functional stuff in the
prototype stages. Unfortunately, while this magic stuff rarely delivers
what was originally promised, by the time that this becomes obvious, too
much money and effort has ben sunk into the change to go back! One of
the negatives of the "computer revolution" is people have become used to
accepting a frequency of technical failure which would have generated
mega lawsuits a couple decades ago.
Incidentally, more people die from hospital related mistakes then are
killed by airlines and firearms combined. While a Gold Medal victory is
not as important as a life, I do not get put in charge of ensuring the
reliability of medical equipment, so as long as I don't have to go to
the hospital, for me it IS more important!
> Conversely, would you trust a deciding point in the gold medal bout in
> the Olympics to a reel that uses mechanical brushes for electrical
> contact? What if the contact breaks for just a few milliseconds so the
> fencer doesn't get the touch?
>
But they solved THAT problem a few decades back, by requiring dual
contacts.
>
> For the past 15 years, most cars were built with about six to twelve
> computers. My father's car has a 6802 microprocessor handling the fuel
> injection. The micro has worked for 18 years and has been used about
> every day with no problem.
>
Don't have anything that new, as most of the vehicles with several
computers have airbags, and airbags can be fatal to someone around 5"4"
or so (and I'm the tallest in the family)!
My newest motorcycle is 37 years old, as the manufacturers in the
'70's decided that there were only a few men under 5'6" or 5" 8" or
something (and they never paid much attention to female riders anyway!)
so they didn't need to make cycles on which short people could reach the
ground as there were so few of them. OTOH as it's a BMW, I'll probably
be dead before it becomes unrepairable!
Chuck Allen
John Twernbold
> Must be nice to be rich! This is true ONLY if the machines, including
> the associated transmitters get down to maybe $100, as it rarely costs
> more than that to fix an old box, a price I do NOT expect to see anytime
> soon.
Hmm, I don't think that's an entirely impossible prediction. $100 may be a
bit low for the next few years, but I can envision it being a possible price
point within 10 years. You can now get digital spread spectrum cordless
telephones for about that much, and I can imagine two sets of wireless
equipment (with their integrated "scoring machine") eventually dropping to
that price. Compare the types of wireless equipment available today (cell
phones, cordless phones, etc.) with those available in 1990, and it would
seem they've become smaller, cheaper, and more reliable. Of course, the cost
of wireless fencing equipment also depends on the number of units produced.
But even if the whole package cost $350, it would still be equal in price to
just one Uhlmann reel.
I think one of the failings of the current wireless system being tested is
that it completely replaces the reels *and* the scoring machine. While I can
appreciate the compactness of this system, I suspect it would make the
conversion to wireless more difficult. If the wireless components merely
replaced reels, and plugged right into existing scoring machines, I think
the transition would be much easier. As clubs and divisions replaced
worn-out reels, they could get wireless equipment instead. I had thought
this type of easy transition had been one of the FIE's goals for wireless
fencing, but it seems to have been dropped.
John
Chuck Allen
John Twernbold wrote:
>
> Hmm, I don't think that's an entirely impossible prediction. $100 may be a
> bit low for the next few years, but I can envision it being a possible price
> point within 10 years. You can now get digital spread spectrum cordless
> telephones for about that much, and I can imagine two sets of wireless
> equipment (with their integrated "scoring machine") eventually dropping to
> that price.
A video card for your computer can be had, (fire sale price) for under
$20, and regularly for $49.95. It is a considerably more complex device
than an Allstar club box going for around $900, or even the cheaper
boxes which are currently going in the $300 - $400 range. You want cheap
electronics, you're going to have to have production runs in the 100,000
range. Effectively, the only way were ever going to see cheap scoring
equipment is to get a WHOLE bunch of more fencers!
> Compare the types of wireless equipment available today (cell
> phones, cordless phones, etc.) with those available in 1990, and it would
> seem they've become smaller, cheaper, and more reliable. Of course, the cost
> of wireless fencing equipment also depends on the number of units produced.
> But even if the whole package cost $350, it would still be equal in price to
> just one Uhlmann reel.
>
But, with proper maintenance, that one reel will last about forever.
I've been carrying 2 sets of Mercotac contacts around in my gear since
1989, out of all the reels I've repaired during that period, I might
have had to replace a set of contacts ONCE during that period. Granted,
springs go occasionally, and, after you've repaired the reel end several
times the reel wire will eventually get to short, but even that is a $75
repair which has to be done maybe every couple decades!
> I think one of the failings of the current wireless system being tested is
> that it completely replaces the reels *and* the scoring machine. While I can
> appreciate the compactness of this system, I suspect it would make the
> conversion to wireless more difficult. If the wireless components merely
> replaced reels, and plugged right into existing scoring machines, I think
> the transition would be much easier. As clubs and divisions replaced
> worn-out reels, they could get wireless equipment instead. I had thought
> this type of easy transition had been one of the FIE's goals for wireless
> fencing, but it seems to have been dropped.
>
The manufactures probably pointed out that the setup would be
manufactured cheaper if the box and the receiver were integrated.
'Course the fencers still have to have transmitters! I expect that the
next idea the FIE comes up with will be to integrate the transmitters
with the weapons, making all our weapons obsolete, as it's there the
only thing we're still using which will still work under the new
rules. Remember, this stuff is still in the prototype stage and it
takes, in the fencing community, about a decade or two to work all the
bugs out of the systems, in fact sabre STILL hasn't stabilized
technically yet! IIRC foil, which was first electrified in 1954, took
until around 1968 before all the rules and specifications stabilized.
Chuck Allen
HMS Lion
for a fully integrated weapon/transmitter package. Not that it really makes
wireless fencing possible (the EMC/EMI problems are bigger than they realize),
just that it would make a lot more sense. Better still (at least for epee),
just work out a metal-sensing tip and use the old buzz-busters.
V/R:
Mike McDaniel
John Twernbold
> A video card for your computer can be had, (fire sale price) for under
> $20, and regularly for $49.95. It is a considerably more complex device
> than an Allstar club box going for around $900, or even the cheaper
> boxes which are currently going in the $300 - $400 range.
Yep, that's very true. What might reduce the price of wireless fencing is if
it can incorporate components and technology from other industries. I'm not
too familiar with Bluetooth, but wouldn't standards like that enable
manufacturers to increase compatibility and lower costs?
> The manufactures probably pointed out that the setup would be
> manufactured cheaper if the box and the receiver were integrated.
> 'Course the fencers still have to have transmitters!
If I understand it correctly, the current prototype system does have all
this intregrated, including the transmitters. There is no scoring
machine--all equipment is carried by the fencers, with a box on the waist
and lights in the mask. For pictures and details, check out these pages:
http://home.t-online.de/home/allstar/wireless.htm
http://home.t-online.de/home/allstar/wireles2.htm
> IIRC foil, which was first electrified in 1954, took until around 1968
before
> all the rules and specifications stabilized.
And I imagine many fencers in the 1950's were poo-pooing the idea of
electric fencing, since there were so many problems and changes. After all,
dry fencing is much more reliable (from an equipment standpoint) than
electric. But I think most modern sport fencers appreciate the use of
electric scoring. It's not inconceivable that fencers in 2040 will be amused
at our arguments against wireless scoring.
John
sab...@my-deja.com
Eric Schlaepfer <sch...@sonic.net> wrote:
> Hi Bryan,
>
> > Doubling the weight of the blades would have a better effect for the
> > viewer, I'd say.
>
> The FIE seems to think the same way as you do; recently the FIE made
the
> saber blades much thicker. They are now thicker than an epee blade.
>
> Later,
>
> Eric
>
You know, Eric, the blades I've been using aren't much thicker than the
old blades, certainly not as thick or heavy as an epee blade. They're
just a lot stiffer than the old sabre blades were.
Mark Ray
Chuck Allen
John Twernbold wrote:
>
> Chuck Allen wrote:
> > A video card for your computer can be had, (fire sale price) for under
> > $20, and regularly for $49.95. It is a considerably more complex device
> > than an Allstar club box going for around $900, or even the cheaper
> > boxes which are currently going in the $300 - $400 range.
>
> Yep, that's very true. What might reduce the price of wireless fencing is if
> it can incorporate components and technology from other industries. I'm not
> too familiar with Bluetooth, but wouldn't standards like that enable
> manufacturers to increase compatibility and lower costs?
>
This IS with the adaptation of technology from other industries! Most
boxes today use technology developed for the computer industry. While
the specific application is for fencing, outside of the difference in
input/output and the specialized coding for fencing, the basic box
design is very similar to that of a computer. What makes the difference
is computers being produced in millions and scoring equipment being
produced in not more than thousands and if you are talking about a
specific design, probably hundreds. I was trying to have a specially
sized square hole punched in a piece made of aluminum. Despite this part
being very standard (most every computer has one, it was not a standard
sized hole punch. Up on investigation, I found that a custom punch to
make the hole would cost $50,000. To get the cost of the hole below the
cost of the part I would have had to make more than 50,000 units.
Multiply this situation many times with variations (automatic component
insertion machines etc. ) and you will start to get an idea of the
problem.
Chuck Allen
David W. Neevel
> I think one of the failings of the current wireless system being tested is
> that it completely replaces the reels *and* the scoring machine. While I can
> appreciate the compactness of this system, I suspect it would make the
> conversion to wireless more difficult.
*****
> If the wireless components merely
> replaced reels, and plugged right into existing scoring machines, I think
> the transition would be much easier. As clubs and divisions replaced
> worn-out reels, they could get wireless equipment instead. I had thought
> this type of easy transition had been one of the FIE's goals for wireless
> fencing, but it seems to have been dropped.
>
*****
This is in fact the case-- one of the FIE's mandates is that the
wireless systems be compatible with current fencers' equipment, so
there's no need to replace everything. You'll just keep using the
current reel systems alongside whatever new wireless system you may get
(admittedly with upgraded FW in the older machines if you want to keep
up with the latest FIE specs). Bear in mind that manufacturers will
continue to sell reels and wired machines for quite some time, since:
A)wireless systems will probably have teething problems and be pricey
for a while yet, and B) there will still be a demand for replacement
parts and equipment for all of the wired systems that will remain in
service for years to come.
-Dave Neevel
yeoldea...@pop.bois.uswest.net
10 mins. They been working on wireless since the 1970s
Eric Schlaepfer wrote:
> Hi Chuck,
>
> > Wireless is perhaps a theoretical ideal, (I never really had a problem
> > with reels!) but from a technical viewpoint it's got to be a disaster.
> > Not counting the RF interference issues, whatever is used must be as
> > close to 100% reliable as possible.
>
> Not necessarily. At MIT, there is some research going on about
> electronic business cards. Somebody invented a simple technique to allow
> data to be transferred through the human body at up to 2400 baud. The
> technique involved an active band-pass filter that removed nearly all of
> the RF noise, AC hum, radio stations, etc. from the incoming signal. A
> similar approach could be used to detect touches in wireless fencing.
>
> > overtime you add a part to a
> > device, you decrease the reliability of the device. For instance,
> > suppose you
> > have a device which uses 10 parts, each which has a reliability of
> > 99%. The complete device is NOT 99% reliable but .99 x .99.... 10
> > times, which comes out to a reliability of 90.4%. Obviously, wireless
> > devices will use parts with a higher reliability factor, and obviously,
> > considering that you have to have a separate receiver and transmitter
> > for each fencer, it will use considerably more than 10 parts, intact
> > the wireless transmitting system might even use more parts than the
> > current boxes!
>
> Not necessarily. In saber and off-targetless foil, there is no need for
> radio or infra-red communication at all.
>
> At this point, radio communications are very reliable and secure. After
> all, many expensive cars use a radio-controlled remote door lock.
>
> I contend that a well-designed wireless fencing system would be far more
> reliable than the current mechanical reel system.
>
> > Only a few of the even the national armourers can fix boxes, what
> > happens when we go from reels to a wireless system which can't be
> > repaired except by shipping it off to the manufacturer? How much of an
> > increase in spares will then be necessary to run a tournament?
>
> Most machines these days are sent to the manufacturer for repair anyway.
> Tournaments can use the same number of spare machines they use now, but
> they no longer need spare reels or floor cords.
>
> > I am presently restoring a 40 year old motorcycle, it is doable, as
> > rust can be
> > removed, dents can be taken out of metal, and mechanical parts no
> > longer available
> > can (with enough input of cash) be fabricated. On a modern bike, when
> > its gets that old, it will have quite a number of parts which cannot be
> > fabricated on a onesy twosy basis, (or even a 100 200 basis!) and will
> > be unrepairable. A lot of clubs have to exist on old scoring equipment,
> > what happens to them when old scoring equipment no longer exists, as it
> > has become unrepairable?
>
> They buy new machines. Because of the decreasing cost of technology,
> scoring machines are becoming more and more affordable to the point
> where it will be cheaper to buy a new machine than to custom fabricate
> parts for an obsolete one.
>
> > In addition, would you want to trust the
> > deciding point in the Gold medal bout in the Olympics to a device which
> > had the technical reliability of Microsoft Windows????
>
> The wireless scoring system that has been used is, in my opinion, still
> in the early prototype phase. A production-quality system would be very
> reliable, much more so than Windows. Remember, computers are used in the
> medical industry where someone's life could depend on a piece of
> software.
>
> Conversely, would you trust a deciding point in the gold medal bout in
> the Olympics to a reel that uses mechanical brushes for electrical
> contact? What if the contact breaks for just a few milliseconds so the
> fencer doesn't get the touch?
>
> > Even though (or perhaps because!) I spend some of my time
> > functioning as a computer consultant, I am very against putting any
> > computer chips in cars. Most of the time, critical mechanical systems
> > can be jury rigged to get you where you have to go, OTOH when an
> > electronic component fails it's DEAD, with no option other than
> > replacement.
>
> For the past 15 years, most cars were built with about six to twelve
> computers. My father's car has a 6802 microprocessor handling the fuel
> injection. The micro has worked for 18 years and has been used about
> every day with no problem.
>
Eric Schlaepfer
> I do maintenance on about 60 Uhlmann reels in the Boston area. Several are
> surplus to the 1984 Olympics with old LAOOC stickers on them.
Ooh, that must be interesting. Are they the older style uprights?
> Could they be smart enough to figure out how to use radio-controlled model
> transmitters and receivers for wireless fencing?
The radio communication really is no big deal. The real hard part is
detecting metal-to-metal contact with no ground return. Once that is
solved the rest will come pretty easily.
> They are manufactured in very large equantities and are available in your local
> Hobby store.
RC controls could be intercepted though, but for club use it would work
fine.
Later,
Eric
Eric Dew
Eric Schlaepfer <sch...@sonic.net> wrote:
>Hi Bill,
>
>> I do maintenance on about 60 Uhlmann reels in the Boston area. Several are
>> surplus to the 1984 Olympics with old LAOOC stickers on them.
>
>Ooh, that must be interesting. Are they the older style uprights?
No, they were the flat land mine style. But 60 reels? How many strips
were at the 84 Olympics? I can't imagine more than 20. Even at that,
60 reels in the Boston area and I know SoCal has a bunch, as well as CenCal
and NorCal divisions. I suspect only a few of the 60 are from the LAOOC.
EDEW
mary anne walker
Bill didn't say that all of the reels were from the LA Olympics. He
said several out of the 60 Uhlmann reels in the area are from that
event.
--
Until later,
Mary Anne Walker
************************************************************
"Artillery adds dignity to what would otherwise be a vulgar brawl."
-Murphy's Law of Combat #62
"If you cannot convince them, confuse them." - Harry S. Truman
"You don't have to burn books to destroy a culture. Just get people
to stop reading them." - Ray Bradbury
John Hasler
> The radio communication really is no big deal.
True. Off the shelf spread spectrum technology will do nicely.
> The real hard part is detecting metal-to-metal contact with no ground
> return.
Quite doable.
--
John Hasler
jo...@dhh.gt.org
Dancing Horse Hill
Elmwood, Wisconsin
Fencerbill
>
>> I do maintenance on about 60 Uhlmann reels in the Boston area. Several are
>> surplus to the 1984 Olympics with old LAOOC stickers on them.
>
>Ooh, that must be interesting. Are they the older style uprights?
>************
No, turtles.
*************
>> Could they be smart enough to figure out how to use radio-controlled model
>> transmitters and receivers for wireless fencing?
>
>The radio communication really is no big deal. The real hard part is
>detecting metal-to-metal contact with no ground return. Once that is
>solved the rest will come pretty easily.
>
>> They are manufactured in very large equantities and are available in your
>local
>> Hobby store.
>
>RC controls could be intercepted though, but for club use it would work
>fine.
>************************
Perhaps I am naive about this, being a physicist, not an engineer. Isn't it
the modulation that gives you the security? I think they have about 5 kHz
bandwidth.
**************************
>Later,
>
>Eric
>
>
Chris Aher
The original patent on spread spectrum technology was held by the late
actress Hedy Lamarr, a multitalented lady for sure.
Regards,
Chris Aher
"John Hasler" <jo...@dhh.gt.org> wrote in message
news:874sbpk...@hasler.dhh...
> Eric Schlaepfer writes:
> > The radio communication really is no big deal.
>
> True. Off the shelf spread spectrum technology will do nicely.
>
> > The real hard part is detecting metal-to-metal contact with no ground
> > return.
>
Eric Schlaepfer
> Perhaps I am naive about this, being a physicist, not an engineer. Isn't it
> the modulation that gives you the security? I think they have about 5 kHz
> bandwidth.
Well, the modulation really isn't good security. Somebody could jam the
signal or do other bad things, such as using a duplicate transmitter to
fool the scoring machine.
Later,
Eric
Chuck Allen
Eric Schlaepfer wrote:
>
> The real hard part is
> detecting metal-to-metal contact with no ground return. Once that is
> solved the rest will come pretty easily.
>
Actually, it really makes no difference what sort of contact you're
dealing with, the problem is that the concept of wireless scoring is
flawed. This is easy to see once you figure out what the wireless system
is trying to replace. The epee circuit is easiest to understand and even
trying to replace THAT is a problem.
in epee there are 2 contacts in the weapon which when closed produce a
hit light. If you wanted to make that circuit "wireless" you would
merely have to transmit a signal from the fencer to the scoring circuit
whenever the switch was closed. Simple right? HOWEVER, you really don't
want a fair touch registered on the opponents weapon or the floor.
Currently, the system is still fairly simple, when you hit either of
those surfaces, you close a switch (remember a switch is merely composed
of 2 contacts, regardless of what those two contacts are attached to!)
between ground and the "A" and/or "B" lines (depending on the design of
the machine) and ground.
The foil circuit also has a closed switch in the weapon, which when
opened will produce a "foul" light. However if the "switch" between the
foil and the opponents lame is ALSO closed, you will turn on a "Fair
hit" light. If the switch composed of the foil and the opponents vest
remains open, BUT the switch between either the opponents weapon or the
floor is closed when the switch on the foil is opened, nothing happens.
Sabre is also simple. There is NO switch in the weapon! The valid touch
switch is the contact between the sabre and the opponents vest, and the
ground lock-out switch is between the sabre and the opponents weapon.
Therefore you have the following switches:
Contact 1 Contact 2
Epee:
1st switch - Own weapon Own weapon
2ed switch - Own weapon Ground (opponent's weapon)
3rd Switch - Own weapon Ground (floor)
Foil:
1st switch - Own weapon Own weapon
2ed Switch - Own weapon Opponent's lame
3rd Switch - Own weapon Ground (opponent's weapon)
4th Switch - Own weapon Ground (floor)
Sabre:
1st Switch - Own weapon Opponent's lame
2ed Switch - Own weapon Ground (opponent's weapon)
Due to manufacturer complaints (we can't DO that) The sabre circuit no
longer detects hits to the floor.
The shorted circuit between the "b" and "C" lines on a sabre is merely
for test and can be ignored.
As can be seen, of the switches in the epee and foil circuits, only one
set of contacts are on the same object, and in sabre NO contacts are on
the same object!
A circuit is a continuous path with a source of voltage and a device to
be powered - if you merely replace the scoring machine (who's purpose
merely is to determine the relationship in time of the various switch
closings and openings) with a light and battery, you now have reduced it
to simple easy to understand circuits, consisting of a light, a power
source, and a switch to turn the light on, with those parts external to
the light and battery circuits consisting merely of switches and the
wires connecting them to the test box or scoring machine.
If you study electronics, the first fundamental principle you have
drummed into you is that you must have a complete circuit path for
something to work, and that if that path is broken for any reason,
whatever you are working on or designing will fail!
As can be noted from the above, many of the switches used in the
scoring circuits have one contact on one object and one contact on
another!
To make a wireless system, you have to try and complete a circuit
while actually breaking the wire between the two contacts. A simple idea
of the problem involved is if you were to take a normal REALLY simple
circuit like a flashlight (which consists of a battery, a bulb and a
switch to turn the circuit on and off) and move the bulb several feet
away from the rest of the flashlight, and then propose to turn the bulb
on and off wirelessly!
While there are a lot of "wireless" devices on the market, they almost
all merely transmit a switch closing where both contacts are on the same
object to another object, which with today's technology is not two
difficult. I cannot, off the top of my head, think of any device which
"wirelessly" transmits the closing of a switch when the contacts are on
two different objects. Even the model aircraft RC circuits merely send a
signal to close a switch both contacts of which are in the model
aircraft.
The theoretical problem can be reduced to trying to complete a circuit
without any wires, rather than the transmission of a switch closing
typified by most wireless applications.
As the switch to this technology appears to be driven by the FIE
through the IOC which is mainly being motivated by the media (wires look
ugly) types, it is constructive to put the kind of money into this to
put up with the inconveniences and additional expense ONLY if it results
in a significant increase in media coverage.
There are also a few folks who think it would be wonderful to not have
to plug into reels, but I personally see no advantage in having to plug
into a wireless transmitter instead - takes the same amount of time. I
will concede that if you did not have to plug anything into anything it
would be a major advantage. For instance, if the transmitters could be
made cheap enough to not add more than say $10 -$15 to the cost of a
weapon, so that they could be built into the weapons AND there were
absolutely NO wires at all, i.e. the body cords and head cords were ALSO
eliminated, it might be worth doing, but to eliminate a fairly reliable
reel and NOT eliminate the plug at the weapon and the plug at the other
end of the body cords which are a major source of problems does not seem
to be worth it, if it can even be done reliably (which I doubt! - go
back to trying to build a flashlight without any wires!)
Chuck Allen
John Hasler
> The theoretical problem can be reduced to trying to complete a circuit
> without any wires, rather than the transmission of a switch closing
> typified by most wireless applications.
You're too close to the problem. The problem isn't to complete a metallic
circuit, it is to determine when the tip is touching metal. This can be
done by, in essence, measuring the capacitance between the tip wires and
the blade. The wires are largely shielded by the blade, so this
capacitance will have a definite value little influenced by other objects
as long as the tip is touching only nonconducting material. Touch a metal
object such as the strip or your opponent's blade, however, and the
capacitance increases by the value of the capacitance between said object
and your blade. This change is easily detectable with current technology.
For epee this is sufficient. Foil and sabre add complications, but these
can also be dealt with using near field effects.
> ...it is constructive to put the kind of money into this to put up with
> the inconveniences and additional expense ONLY if it results in a
> significant increase in media coverage.
The only thing that will result in a significant increase in media coverage
is a significant increase in public interest.
> ...go back to trying to build a flashlight without any wires!)
This line of reasoning is similar to trying to invent the airplane by
constructing machines with flapping wings.
--
John Hasler
jo...@dhh.gt.org (John Hasler)
Dancing Horse Hill
Elmwood, WI
Russ Smith
> ...Touch a metal object such as the strip or your opponent's
> blade...
Or your opponent's mask.
> ...however, and the capacitance increases by the value of the
> capacitance between said object and your blade. This change
> is easily detectable with current technology.
>
> For epee this is sufficient...
Maybe not, unless the Lexan masks become required with NO
metal parts.
Or maybe take away the head as a target, make the mask ENTIRELY
metallic, and ruin the "feel" of epee fencing...
Russ Smith
heng...@erols.com
Eric Schlaepfer
You are correct in stating that one major difficulty with wireless
scoring is the separation of an electrical switch between two fencers.
One half of the switch is with one fencer, and the other half of the
switch is with the other fencer.
This problem has been solved before. Several examples exist: a device
called the FoilMaster (available from American Fencers Supply) that
differentiates between an on-target touch and an off-target touch, a
wireless epee system developed as a university project by several
engineering students in Canada, a wireless system developed by St.
George in France, and the wireless system currently being used by the
FIE. I don't agree with the design approach taken by any of these
devices, but they have solved or come close to solving the
complete-circuit problem.
I have performed many experiments using my oscilloscope and several
battery powered (this avoids coupling through the AC power lines)
devices and I have successfully detected electrical contact with no
ground return. The hard part is doing it reliably and without problems
with interference such as AC hum and radio stations.
Wireless scoring is advantageous because it removes the need for
mechanical reels. Well-designed electronics with no moving parts are
*far* more reliable than reels that fail because of normal wear and
tear. For example -- do you agree that a modern SG31 scoring machine is
better in all respects than one of the old electromechanical relay
boxes?
I think that in the next few years wireless scoring will be perfected
and eventually will be used even at the club level as its price falls.
Later,
Eric
Eric Schlaepfer
>
> Chuck Allen writes:
> > The theoretical problem can be reduced to trying to complete a circuit
> > without any wires, rather than the transmission of a switch closing
> > typified by most wireless applications.
>
> You're too close to the problem. The problem isn't to complete a metallic
> circuit, it is to determine when the tip is touching metal. This can be
> done by, in essence, measuring the capacitance between the tip wires and
> the blade. The wires are largely shielded by the blade, so this
> capacitance will have a definite value little influenced by other objects
> as long as the tip is touching only nonconducting material. Touch a metal
> object such as the strip or your opponent's blade, however, and the
> capacitance increases by the value of the capacitance between said object
> and your blade. This change is easily detectable with current technology.
With the system you propose, how can the device differentiate between
touches to the lame and touches to the bell guard?
> > ...go back to trying to build a flashlight without any wires!)
>
> This line of reasoning is similar to trying to invent the airplane by
> constructing machines with flapping wings.
Ever seen a Tesla coil lighting a fluorescent bulb without wires?
Later,
Eric
John Hasler
> Maybe not, unless the Lexan masks become required with NO metal parts.
Or insulate the metal parts of the mask, or use the same technique I
propose to use to distinguish the lame from the blade in foil to
distinguish the mask from the blade in epee.
> Or maybe take away the head as a target, make the mask ENTIRELY metallic,
> and ruin the "feel" of epee fencing...
I don't favor making any rule changes whatever to accomodate wireless
scoring.
Chuck Allen
Eric Schlaepfer wrote:
>
> I have performed many experiments using my oscilloscope and several
> battery powered (this avoids coupling through the AC power lines)
> devices and I have successfully detected electrical contact with no
> ground return. The hard part is doing it reliably and without problems
> with interference such as AC hum and radio stations.
>
I tried, at one time, to design an epee "buzz box" which would detect
hits on the bell. I thought that if there was metal to metal contact
this would add a sufficient amount of metal (your opponent's whole epee)
to the epee you were already holding to increase the effect of both of
you holding an antenna so that you could detect the contact by
rectifying noise. Worked as long as I wasn't wearing a glove! Apparently
your body functions much better as an antenna to pick up noise than your
opponents epee.
> Wireless scoring is advantageous because it removes the need for
> mechanical reels. Well-designed electronics with no moving parts are
> *far* more reliable than reels that fail because of normal wear and
> tear. For example -- do you agree that a modern SG31 scoring machine is
> better in all respects than one of the old electromechanical relay
> boxes?
>
It is often said that devices with no moving parts are more reliable
than those which have moving parts, however, the following is also true:
a) mechanical devices can often be repaired when devices with no moving
parts cannot and b) devices with no moving parts often incorporate
electronic components with a relatively short product life span (i.e.
less than a decade from the time the first one comes off the production
line until the last one comes off the production line) as these devices
cannot be produced in anything other than a multimillion dollar (or
sometimes multi-billion dollar) fab plant dedicated to cranking these
things out in million lots, anything incorporating these devices cannot
be repaired. A mechanical device can, if necessary, be made in lots of
one. There are a lot of clubs (and some divisions) using 20 to 30 year
old scoring boxes. I personally, will not design a sole source part into
something unless it absolutely CANNOT be designed without doing so, (all
the electrical and electronic parts in the towers I sell are available
at Radio Shack!) having seen to many companies go under when the sole
source of a critical part decided it was no longer economic to
manufacture said part.
While a lot of people have become somewhat blasé about built in
obsolescence, I do not think that the fencing community is a place where
it can be tolerated without having serious negative economic
consequences.
Chuck Allen
Day Al-Mohamed
I was wondering where one might get or find the specs to make a simple buzz
box. I have looked at fencing catalogues and though a simple buzz box may be
listed, when I contacted the suppliers they did not have them in. I am in no
way electrically-minded or engineeringly- inclined....but I know people who
are. They would be happy to help me put it together but as to the base stuff
on how it works and what parts are best...help.
Day Al-Mohamed
U. of Missouri-Columbia Fencing
Eric Schlaepfer
> Eric Schlaepfer writes:
> > With the system you propose, how can the device differentiate between
> > touches to the lame and touches to the bell guard?
>
> through your electronics to your lame (if I hit your bell) or your weapon
> (if I hit your lame) The phase of this current depends on which I hit. A
> synchonous detector in your electronics would detect the phase and thus
> determine what I hit. Your electronics and mine would have to share a
> timebase of course, but this is not hard.
Interesting idea. What happens when the fencer shorts his bell guard to
his lame?
Another approach would be to transfer binary data across with a touch,
so the device can tell if it hit a valid target area or a bell guard.
The device could also tell who it was hit by.
Later,
Eric
Ryan Polasek
: Hi John,
:
: > Eric Schlaepfer writes:
: > > With the system you propose, how can the device differentiate between
: > > touches to the lame and touches to the bell guard?
: >
: > Phase. My tip would put out a pulse train. Part of this pulse would flow
: > through your electronics to your lame (if I hit your bell) or your weapon
: > (if I hit your lame) The phase of this current depends on which I hit. A
: > synchonous detector in your electronics would detect the phase and thus
: > determine what I hit. Your electronics and mine would have to share a
: > timebase of course, but this is not hard.
: Interesting idea. What happens when the fencer shorts his bell guard to
: his lame?
Use the weapon as the reciever, bell guard and lame each send a different
AC signal (yes it will go across, thats why you get shocked by your
wall socket even if you aren't grounded ) Make sure they arent harmonics
or intermodular frequencies of one another and a fourier analyzer circuit
can parse out what got hit (target 1, target 2, or both) and send the
info using tried and true wireless communication to whatever device
needs it. I may be wrong about the physics on the AC thing, if anybody
here knows more about signals and electromagnetics I'd appreciate
comments.
: Another approach would be to transfer binary data across with a touch,
: so the device can tell if it hit a valid target area or a bell guard.
: The device could also tell who it was hit by.
Good approach, blade to target contacts are generally long enough for
simple data transfer.
--
/*------------------------------
Ryan Polasek
University Of Michigan
rpolasek@engin,umich,edu
www.umich.edu/~rpolasek
(Email altered for autospammers. Change commas to periods to fix. )
-------------------------------*/
Eric Schlaepfer
> Use the weapon as the reciever, bell guard and lame each send a different
> AC signal (yes it will go across, thats why you get shocked by your
> wall socket even if you aren't grounded ) Make sure they arent harmonics
> or intermodular frequencies of one another and a fourier analyzer circuit
> can parse out what got hit (target 1, target 2, or both) and send the
The AC signal can be filtered by a very "pointy" active bandpass filter
to prevent radio stations from causing interference. Very good idea
though. You could use two PLLs as the frequency detectors.
Remember that there still is a ground path through the capacitive
coupling between the two fencings--which is why the signal has to be AC.
> info using tried and true wireless communication to whatever device
> needs it. I may be wrong about the physics on the AC thing, if anybody
> here knows more about signals and electromagnetics I'd appreciate
> comments.
Sounds good to me. Any other comments?
>
> : Another approach would be to transfer binary data across with a touch,
> : so the device can tell if it hit a valid target area or a bell guard.
> : The device could also tell who it was hit by.
>
> Good approach, blade to target contacts are generally long enough for
> simple data transfer.
Of course. Notice the new rule change of >10ms debounce, plenty enough
time to transfer data.
Later,
Eric
Eric Schlaepfer
> To detect weapon-lame contact just apply a voltage between them and
> measure the current that flows when they touch. In other words do it the
> same way it is done now.
That is one way. One needs to ensure that applying a voltage like that
won't interfere with the detection of touches.
>
> > ...bell guard and lame each send a different AC signal...
>
> It isn't quite that simple.
It actually does work, but not very well. It creates a ground reference
by the capacitive coupling between the fencers' bodies. That's why an
early wireless system developed in the mid 70's had each fencer wear a
metal bracelet. The real trick is to filter out common-mode noise.
> > Make sure they arent harmonics or intermodular frequencies of one another
> > and a fourier analyzer circuit can parse out what got hit (target 1,
> > target 2, or both)
>
> Use time division multiplexing.
That's another approach. (I happen to favor multiplexing over multiple
freqencies.)
> > ...blade to target contacts are generally long enough for simple data
> > transfer.
>
> There is nothing about "data transfer" that inherently tells us anything
> about which piece of metal is touching which other piece.
Certainly. Data can be sent by modulating some type of binary signal
with a carrier frequency. The receiver filters out every frequency
except for the carrier frequency with an active bandpass filter. After
removing the carrier signal, the result is a binary data signal that can
be used to determine what target was hit.
Later,
Eric
John Hasler
> One needs to ensure that applying a voltage like that won't interfere
> with the detection of touches.
The detection of touches is done using ac. No interference.
> It creates a ground reference by the capacitive coupling between the
> fencers' bodies.
The term 'ground' is not useful in this context.
> I happen to favor multiplexing over multiple freqencies.
TDM is easier in a DSP.
> After removing the carrier signal, the result is a binary data signal
> that can be used to determine what target was hit.
The presence of modulation does nothing to determine which electrode has
the metallic contact and which is capacitively coupled.
Eric Schlaepfer
>
Hi John,
> > One needs to ensure that applying a voltage like that won't interfere
> > with the detection of touches.
> The detection of touches is done using ac. No interference.
The DC change of state will generate an AC spike even if you have a
capacitor in there. Of course the synchonization could be done from the
DSP.
> > It creates a ground reference by the capacitive coupling between the
> > fencers' bodies.
>
> The term 'ground' is not useful in this context.
Okay, the `return' or `reference.'
> > I happen to favor multiplexing over multiple freqencies.
>
> TDM is easier in a DSP.
Or a microcontroller. It's a digital solution.
> > After removing the carrier signal, the result is a binary data signal
> > that can be used to determine what target was hit.
>
> The presence of modulation does nothing to determine which electrode has
> the metallic contact and which is capacitively coupled.
Looks like we are talking of different approaches now. What is the
capacitively coupled electrode?
Later,
Eric
John Hasler
> The DC change of state will generate an AC spike even if you have a
> capacitor in there. Of course the synchonization could be done from the
> DSP.
A small, broadband pulse of noise that no decent receiver will be disturbed
by. It also will only occur when there is a complete metallic circuit.
I wrote:
> The term 'ground' is not useful in this context.
Eric writes:
> Okay, the `return' or `reference.'
All terms for a particular node given special status for notational
convenience. I don't find it helpful in this context. Better to think in
terms of a (possibly partitioned) network.
> Or a microcontroller. It's a digital solution.
It's the least expensive and most reliable solution.
I wrote:
> The presence of modulation does nothing to determine which electrode has
> the metallic contact and which is capacitively coupled.
Eric writes:
> Looks like we are talking of different approaches now. What is the
> capacitively coupled electrode?
The electrodes are the blade and the lame (in foil).
A very simplified example:
My tip fixes on your bell guard. My electronics module detects the opening
of the tip switch and applies a pulse of rf voltage to the transmission
line consisting of my blade and wire. During the pulse my blade and lame
are effectively shorted together at the frequencies of interest. My module
measures the phase and amplitude of the current that results from this
pulse, determines that I hve hit metal, and reports this to the scoring
box.
Your blade is connected to your lame through your module, and your lame is
capacitively coupled to mine. Thus we have a circuit: my module -> my tip
-> your blade -> your module -> your lame -> lame-lame capacitance -> my
lame -> my module. part of the above mentioned current pulse flows around
this circuit, passing through your module. Your module detects the phase
and amplitude of this current and reports this information to the box.
Now assume my tip fixes on your lame. Everything is the same but reversed,
and the circuit is: my module -> my tip -> your lame -> your module -> your
blade -> blade-lame capacitance -> my lame -> my module. Your module
detects the phase and amplitude of the current and reports this information
to the box. The phase is the opposite of of that in the first case, and so
the box can tell what I've hit.
It gets much more complicated when you consider all the permutations and
combinations of contact among the tips, blades and lames, but it works.
A real implementation would be more complex, of course, and might use
things like spread-spectrum techniques on the rf pulses.
Eric Schlaepfer
> A very simplified example:
<Most excellent example snipped>
What about scoring for epee? No lames there for the coupling--I suggest
using the bodies of the two fencers for the current return loop. So if
I hit your epee bell guard, the button on my epee closes, which switches
on the current. My module->my blade wire->your bell guard->your
module->your body->capacitive coupling to my body->my module.
This might sound a little odd, but it would help if there were a circuit
which would maintain a point of the same electrical potential between
both modules using capacitive coupling. Basically it would be a virtual
wire stretched between both fencers. The hit detection circuitry could
then use that virtual wire as a reference.
Have you tried any practical experiments using your technique?
Later,
Eric
Rob Seaman
As a newcomer to the fencing community I've been following this discussion
with great interest. My daughter has been fencing for a bit over a year
and we just returned from the Junior Olympics in Sacramento. She was
3 weeks too young to compete this time, but appears even more enthusiatic
about fencing after watching so many high quality bouts.
My own background is in scientific programming and data acquisition and
archiving at the National Observatory, and one of the armorers was an
acquaintance. As such I found a lot of interest in the bizarre complexity
of the scoring infrastructure, while my daughter focused more on the
competition.
(We may also have been the only pure spectators in the convention center.
Years of unsuccessfully struggling to attract a U.S. audience appear to
have left a fatalistic acceptance of failure that is just as unrealistic
as grandiose plans for a Fencing Superbowl would be. The two weapons with
right-of-way require some knowledge to appreciate, but certainly epee
could be immediately grasped by any sports fan. Even a small attempt
at advertising the tournaments and at providing a quality "experience"
for spectators would pay off.)
In talking to a number of folks (far more knowledgeable, of course), it
became clear that there is a large amount of scepticism about the current
attempt at wireless fencing. Surely the pressure of the impending Olympics
is being allowed to influence the decisions regarding new technology.
As someone with a bit of knowledge about user interfaces, my first reaction
to the rule change flipping the meaning of the lights was to snort out loud.
Imagine swapping the color or order of lights on a traffic signal...
I certainly hope that this decision has more behind it than just clearing
the way for a very specific implementation of wireless scoring.
My original reaction to the idea of wireless scoring in general was before
I had thought through all the implications of the design requirements for
a wireless system. The actual ability to propagate the touches reliably
and securely to a remote score box is pretty much a solved problem in
other communities. It would undoubtedly take several years to be adopted
(affordably) by fencing, but is otherwise no big deal. (Easy for me to
say!) The aggregate bandwidth for even a large convention center full of
dozens of simultaneous bouts is actually quite tiny - even with a high
sampling rate and several "channels" of data from each fencer.
The only real vulnerability need be to denial-of-service attacks. But such
attacks can be monitored for and detected. The danger in a sporting event
is not that folks might try to cheat - the danger is that cheating might
go undetected.
Other vulnerabilities are largely the same as the current system - modified
equipment that might allow a fencer to generate false touches or to avoid an
opponent's touches. It seems to me that the right new technology could even
eliminate many of these failure modes, rather than create new failure modes.
The real issue here is in reliably detecting on-target versus off-target
touches of various sorts for various weapons. Whatever the limitations of
the current reels and wires, an advantage of the rather quaint technology
is that it is very robust. When it breaks it is obvious to even the
non-technologically inclined that something is broken. When it works,
the fencers are willing to trust it. It is straightforward (if annoying)
to fix.
Whatever system replaces the current one (and do we really expect 21st
century fencers to continue using the current system for decades to come?)
it will need to maintain the same level of trust (or improve on it).
My suggestion in designing this new system is to rethink the entire design.
Start from the rules of fencing. Are the current rules the ideal rules?
Ignoring issues of technological limitations - what should the rules be?
And then given the rules for each weapon, determine the requirements for
any scoring system that supports them.
Only when the full requirements are specified should prototype designs be
attempted. Several prototypes employing a range of techniques should be
evaluated.
The sport is important. The technology should be secondary.
One peculiarity of the current electric system is that there are
really three different systems. Each weapon's wiring is implemented
in a completely different way. Might it be revealing to redesign the
equipment to take advantage of similarities rather than differences?
Be willing to change all aspects of the current system. Is a grounded
strip really required? Do the foil and epee points really need to be
mechanical switches? Does it really make sense that the foil relies on
breaking a circuit to record a touch (off target) and the epee relies on
making a circuit to record a touch (on target)? (Please correct me if
I have that wrong.)
And does anybody really believe that grounding the fencers through the
soles of their shoes will work reliably? This seems like a huge kludge.
Just the thought of the constantly shifting ground loops between a
fencer's two legs is enough to give pause. What about wheelchair
fencers? Any new system should be completely unbiased.
Get more basic. All discussion so far has implicitly assumed that the
touches will be detected by physical contact (or by some capacitively
linked variation, for instance). How about using an accelerometer to
detect a touch in "velocity space"? What about some sort of fiber optic
sensor? Any successful solution for a small market like fencing has to
rely on adapting previous technology from larger mass market environments.
What about the various systems that retail stores use to detect merchandise
leaving the premises? I'm not saying that any of these examples are the
way to go - I am suggesting that it's too early in the design process to
exclude any options.
The right question isn't how to pass a touch/no-touch - on/off target
signal from the point (or edge) to the scoring box - the right question
is how best to instrument the weapons and the fencers to keep accurate
track of the state of the bout.
Go beyond the rules. As someone else said, the rules of all sports are
driven by technology - it's just particularly obvious in fencing. At
the core, what is fencing? Fencing is one (or more) person competing
for touches with another person(s) using something called (by neophytes)
a "sword". Note that this is a particular case of a broader category
of sports.
Actually fencing occupies the overlap between several categories of
activities - not just as a particular "martial art" (as some would have
it), but as a type of war game, for instance (e.g., laser tag or paint
ball). Anybody who saw Tom Hanks in "Big" knows whose headset generates
a signal when they are hit in laser tag (the victim). I suspect I'm not
the only one who is confused about whose head-mounted lights are supposed
to go on with each touch in the new (and improved?) wireless system.
If the "victim" is supposed to light up like a Christmas tree using the
new wireless fencing system - why? On the other hand, if not the victim,
why should fencing establish a new standard? There is no obvious reason
why it must be one versus the other (or why the indications need be mounted
on the fencers). Evaluate the decision in terms of the larger context and
then decide from the best interests of the sport - not the limitations of
any inherently flawed technology.
(And, of course, a touch is not the same as scoring a point.)
Define a "sword" in the most general way. What properties do you want
it to have? Assuming the three current weapons are ideal (and given the
great traditions of the sport), if you want to design technology to
support the sport you need to define your terms completely and
unambiguously and in the most general fashion possible.
Someone suggested that the ideal scoring system would not require a body
cord or any other wiring. Is this true? Ignoring the personal risk,
should two fencers in street clothes be able to merely pick up two high
tech 21st century swords and start fencing? If not, why not? If so - aim
for this result, not for some hobbled version that retains a grounded
strip, a body cord, a lame (or saber helmet) and various other equipment
and concepts left over from previous hobbled solutions.
Go more cosmic yet - should 21st century fencing continue to use a
physical blade? Absurd question, perhaps. But ask it before you dismiss
it. Without a physical blade you could omit the protective clothing.
So - you say it's not fencing without steel. That implies that the
fencers need protective gear. It does not imply that they need to be
wired up. It does not imply that the sport requires a grounded strip.
It does not necessarily imply that the fencer wear a metallic lame or
jacket. Decide how you want to fence - then decide what trade-offs
to make.
I'm particularly bemused at the apparent plan to use the Olympics to
drive the introduction of a new scoring system. While I see the public
relations value of this, it otherwise seems completely backwards. Any
new system has to be affordable to be of any value whatsoever. Any new
system has to be widely adopted in a relatively short time scale. The
true market for some new wireless gizmo is at the club level. The new
technology should be fielded from the bottom up - not from the top down.
I was very impressed by the sheer volume of equipment arrayed across the
convention center in Sacramento. Conservatively there were several
hundred thousand dollars in electric scoring equipment distributed among
the thirty or so strips. If I understood the marks that were visible on
various equipment, the USFA must have borrowed equipment from clubs
across the country. The logistics of "requisitioning" and delivering
all this stuff must be staggering. More to the point, it is the
responsibility of the USFA/FIE affiliate running a particular competition
to acquire all this stuff either directly or by borrowing it. Shipping it
to each new venue must also be the cause of much of the flakiness of the
equipment. (Moving instruments from telescope to telescope always causes
problems.)
One aspect yet to be mentioned is that a wireless system begins to
transfer the responsibility for this equipment onto the individual
competitors and clubs. This can be seen as either a plus or a minus
depending on how you look at it...but the competitors are already
responsible for acquiring all their other equipment - why not high
tech signaling equipment? Imagine each weapon being tagged with a
unique digital signature.
So what do I think? (For those who have lasted this long.) I think
the introduction of this new technology is inevitable - and I think
the sport has a serious problem with its self image. You guys are not
going to vanish from the Olympics. I don't even think curling will
vanish. I work in the astronomical community. Fencing shares something
with astronomy in being so fundamental to our worldview that it is often
overlooked. You guys just need to look for a sugar daddy. I suspect
you won't find some rich patsy - er, benefactor - to merely donate
half a mil$ for R&D, but you certainly might get some virtual reality
weenie interested in instrumenting a fencing competition as a research
project. Apply for a grant - it's the American Way.
Why do I put it as "instrumenting"? Because what has been mentioned so
far doesn't even begin to fulfill the description of "wireless scoring".
For true wireless scoring you need a system that can parse the phrasing
of a complete match. In order to do this you need to capture and digitize
the instantaneous posture and position of the fencers.
Any relatively competent programmer could take the USFA/FIE rulebook
and write a fencing phrase parser. (As a relatively competent programmer,
I feel relatively competent to make this claim :-) The difficult part is
capturing the input. The other difficult part is how to present this
information to the referees, directors, judges and other officials
such that they would view it as an aid, not as robotic competition.
Imagine capturing the complete phrasing of every bout of every competitor
at a national or world competition. This is only a small amount of data
as such things go. (I work for the image processing group at the
observatory - images from current CCD cameras are 138 Mbytes each - times
over a hundred a night. I'd guess even a very highly instrumented fencing
bout at 100 Kbytes - times a few hundred or thousand bouts per tournament.
Maybe a gigabyte - on the outside - for every action of every competitor
for a complete tournament.) Think about analyzing those tactical decisions
critically as with the analysis of chess notation - physical chess indeed.
Do I think this will happen? Not soon. But your current equipment is
overdue for retirement. And whatever interim solutions you implement
need to be both backwards and forwards compatible. You have hardware -
but also software - constraints on new technology. By considering the
largest possible solution space you can make decisions now that will
still be valid in twenty or fifty years. If you do plan to completely
instrument each fencer (and perhaps more than two and perhaps not
limited to a one dimensional strip) with full-body data gloves (now
rather old technology) you will need between ten and a hundred channels
for each fencer to convey the position of each joint and the attitude of
each limb and the head and torso. Plan for this now.
Or decide that fully instrumenting the fencers is not a goal. But think
about the complete problem before limiting your plans and designs. And
then implement an interim solution - all solutions are interim, the art
is in defining where you want to go with the next model.
Another thing. Someone with many years of experience suggested that the
geopolitics of fencing are such that everything is being driven by TV - or
rather by the pursuit of the holy grail of American TV. On the other hand,
the US fencing community obviously views itself as a second class citizen
to the great fencing powers of Europe and elsewhere.
Should the US really be waiting on France and the rest of the old world
to present us with some wireless solution? This is obviously an arena
where the US could take the lead. (After all - the French regard Jerry
Lewis as the pinnacle of American TV and media :-)
A successful (and not exorbitantly expensive) wireless scoring system
would sell itself to every club in the US (and elsewhere) for everyday
use. Just imagine the sigh of relief from your armorer at being able to
retire the aging reels and scoreboxes (perhaps to drag them out a few
times a year for FIE beholden tournaments). When the new system matures
it will naturally gain support for competitive use. Any progress on an
automatic phrasing system would be an incredible teaching aid.
Design the system right initially, and a current limited implementation
will still be valid as part of a more extensive system in the future.
Design the system wrong, and it will hang like an albatross around the
necks of your fencers (perhaps literally) for decades.
Rob Seaman
--
sea...@noao.edu, http://iraf.noao.edu/~seaman
NOAO, 950 N Cherry Ave, Tucson AZ 85719, 520-318-8248
PGP: 98 8D 8B 49 74 9A 41 88 3A 43 87 54 51 BF 30 4B
Eric Schlaepfer
The JOs was a really neat tournament, wasn't it? Did you ever go around
the weapons repair booth? I was there repairing equipment. You might
remember me--a young, short guy with blond hair and glasses.
<snip>
> In talking to a number of folks (far more knowledgeable, of course), it
> became clear that there is a large amount of scepticism about the current
> attempt at wireless fencing. Surely the pressure of the impending Olympics
> is being allowed to influence the decisions regarding new technology.
There are some difficulties involved with wireless fencing. The
communication from the fencers' units to the central scoring box is a
potential weak point--somebody could prevent ANY radio based system from
working by using a signal jammer. Or even a primitive spark-gap
transmitter based around a 9-volt battery connected to a metal file.
The second problem is *reliable* detection of touches to various
surfaces such as the bell guard and lame. A system was patented in 1977
(if you are interested I can give you the patent number) and was revived
recently for use in the Olympics. Unfortunately due to the flakiness of
this system the FIE was forced to drop it.
> As someone with a bit of knowledge about user interfaces, my first reaction
> to the rule change flipping the meaning of the lights was to snort out loud.
> Imagine swapping the color or order of lights on a traffic signal...
> I certainly hope that this decision has more behind it than just clearing
> the way for a very specific implementation of wireless scoring.
The FIE originally designed this change for the wireless fencing, but
even after they dropped the wireless system they retained the change
because they thought it was more spectator friendly.
> My original reaction to the idea of wireless scoring in general was before
> I had thought through all the implications of the design requirements for
> a wireless system. The actual ability to propagate the touches reliably
> and securely to a remote score box is pretty much a solved problem in
> other communities. It would undoubtedly take several years to be adopted
> (affordably) by fencing, but is otherwise no big deal. (Easy for me to
> say!) The aggregate bandwidth for even a large convention center full of
> dozens of simultaneous bouts is actually quite tiny - even with a high
> sampling rate and several "channels" of data from each fencer.
The communication is no problem at all--I have thought of a system that
would require no communication whatsoever between the two fencers'
units.
> The only real vulnerability need be to denial-of-service attacks. But such
> attacks can be monitored for and detected. The danger in a sporting event
> is not that folks might try to cheat - the danger is that cheating might
> go undetected.
It is quite easy to cheat using the current system. Such cheating can
and does go undetected.
> Other vulnerabilities are largely the same as the current system - modified
> equipment that might allow a fencer to generate false touches or to avoid an
> opponent's touches. It seems to me that the right new technology could even
> eliminate many of these failure modes, rather than create new failure modes.
It's possible, but even with a wireless system an unscrupulous epee
fencer could still short two wires and give himself a touch.
> The real issue here is in reliably detecting on-target versus off-target
> touches of various sorts for various weapons. Whatever the limitations of
> the current reels and wires, an advantage of the rather quaint technology
> is that it is very robust. When it breaks it is obvious to even the
> non-technologically inclined that something is broken. When it works,
> the fencers are willing to trust it. It is straightforward (if annoying)
> to fix.
Exactly. There is a system in existence called the Foilmaster that does
that. I'm not sure how reliable the scoring is.
> Whatever system replaces the current one (and do we really expect 21st
> century fencers to continue using the current system for decades to come?)
> it will need to maintain the same level of trust (or improve on it).
Which is the other difficulty--if the new system has a fluke, fencers
will avoid it like the plague even if the problem is solved.
> My suggestion in designing this new system is to rethink the entire design.
> Start from the rules of fencing. Are the current rules the ideal rules?
> Ignoring issues of technological limitations - what should the rules be?
> And then given the rules for each weapon, determine the requirements for
> any scoring system that supports them.
Some of the current rules for scoring machines can be ignored for a
wireless system. I have this picture in my mind of an FIE SEMI
commission member testing a wireless scoring system by connecting a
variable resistor from one remote unit to another remote unit...
> Only when the full requirements are specified should prototype designs be
> attempted. Several prototypes employing a range of techniques should be
> evaluated.
>
> The sport is important. The technology should be secondary.
>
> One peculiarity of the current electric system is that there are
> really three different systems. Each weapon's wiring is implemented
> in a completely different way. Might it be revealing to redesign the
> equipment to take advantage of similarities rather than differences?
Well, the systems are really not completely different. They were
originally designed to be somewhat compatible with another.
For example, the foil system is basically identical electrically to the
saber system, the only difference being that the saber has a permanent
short where the foil has a button.
The epee is like the foil in that the bell guards are connected to the
same wire, the C line. The A line of a foil is actually physically
located at the opponent, where it comes out as the lame. In epee the
line is on the same side, but both circuits are normally open. In both
weapons the B line is the same.
> Be willing to change all aspects of the current system. Is a grounded
> strip really required? Do the foil and epee points really need to be
> mechanical switches? Does it really make sense that the foil relies on
> breaking a circuit to record a touch (off target) and the epee relies on
> making a circuit to record a touch (on target)? (Please correct me if
> I have that wrong.)
It's an idea, but for a system to be successful, it should be compatible
with the existing system.
> And does anybody really believe that grounding the fencers through the
> soles of their shoes will work reliably? This seems like a huge kludge.
> Just the thought of the constantly shifting ground loops between a
> fencer's two legs is enough to give pause. What about wheelchair
> fencers? Any new system should be completely unbiased.
Apparently it works. The touches are held in memory so that if the
fencer leaves the strip, the touch still is sent eventually.
> Get more basic. All discussion so far has implicitly assumed that the
> touches will be detected by physical contact (or by some capacitively
> linked variation, for instance). How about using an accelerometer to
> detect a touch in "velocity space"?
It might work, but how do you know if you've hit the opponents guard,
which is not a valid target area?
> What about some sort of fiber optic sensor?
Why not? The fencers could paint their bell guards black, and a light
sensor in the opponent's tip could detect if it hit the bell guard or
the fencer's uniform.
> Any successful solution for a small market like fencing has to
> rely on adapting previous technology from larger mass market environments.
> What about the various systems that retail stores use to detect merchandise
> leaving the premises?
How about this: when you hit your opponent, it changes the length of a
coil located in your wireless unit. An RF transceiver located in the
central scoring machine can detect the change in the resonant frequency
of the coil and score the touch.
> I'm not saying that any of these examples are the
> way to go - I am suggesting that it's too early in the design process to
> exclude any options.
I don't even know if any of these ways will work.
<snip> Someone suggested that the ideal scoring system would not require
a body
> cord or any other wiring. Is this true? Ignoring the personal risk,
> should two fencers in street clothes be able to merely pick up two high
> tech 21st century swords and start fencing? If not, why not? If so - aim
> for this result, not for some hobbled version that retains a grounded
> strip, a body cord, a lame (or saber helmet) and various other equipment
> and concepts left over from previous hobbled solutions.
Here's another, more radical idea that goes back to the days before
electrical scoring. In the old days, tips contained a little bit of ink
which spurted out when a fencer hit somebody. The black showed up quite
nicely on the white uniforms, and after the touch was awarded, the black
spot was removed using vinegar.
Why not build a jacket that glows wherever you apply pressure? You
could even embed a large force-sensing resistor array into the uniform
which would send a signal to the fencer's scoring unit, which would
decide if the touch was in a valid location. Bell guard and floor
touches would be irrelevant because they wouldn't trigger the scoring
unit.
> Go more cosmic yet - should 21st century fencing continue to use a
> physical blade? Absurd question, perhaps. But ask it before you dismiss
> it. Without a physical blade you could omit the protective clothing.
How about a column of light about 3 feet long? ;-)
<snip> I'm particularly bemused at the apparent plan to use the Olympics
to
> drive the introduction of a new scoring system. While I see the public
> relations value of this, it otherwise seems completely backwards. Any
> new system has to be affordable to be of any value whatsoever. Any new
> system has to be widely adopted in a relatively short time scale. The
> true market for some new wireless gizmo is at the club level. The new
> technology should be fielded from the bottom up - not from the top down.
Exactly right. It's the clubs who desperately need a cheaper
alternative to reels, floor cords, and a scoring machine.
> I was very impressed by the sheer volume of equipment arrayed across the
> convention center in Sacramento. Conservatively there were several
> hundred thousand dollars in electric scoring equipment distributed among
> the thirty or so strips. If I understood the marks that were visible on
> various equipment, the USFA must have borrowed equipment from clubs
> across the country. The logistics of "requisitioning" and delivering
> all this stuff must be staggering. More to the point, it is the
> responsibility of the USFA/FIE affiliate running a particular competition
> to acquire all this stuff either directly or by borrowing it. Shipping it
> to each new venue must also be the cause of much of the flakiness of the
> equipment. (Moving instruments from telescope to telescope always causes
> problems.)
The equipement actually comes from Blue Gauntlet, who has some sort of
an agreement with the USFA to provide all the reels, floor cords, and
scoring machines.
> One aspect yet to be mentioned is that a wireless system begins to
> transfer the responsibility for this equipment onto the individual
> competitors and clubs. This can be seen as either a plus or a minus
> depending on how you look at it...but the competitors are already
> responsible for acquiring all their other equipment - why not high
> tech signaling equipment? Imagine each weapon being tagged with a
> unique digital signature.
And of course, these new systems could be tested at weapons control.
<snip> Imagine capturing the complete phrasing of every bout of every
competitor
> at a national or world competition. This is only a small amount of data
> as such things go. (I work for the image processing group at the
> observatory - images from current CCD cameras are 138 Mbytes each - times
> over a hundred a night. I'd guess even a very highly instrumented fencing
> bout at 100 Kbytes - times a few hundred or thousand bouts per tournament.
> Maybe a gigabyte - on the outside - for every action of every competitor
> for a complete tournament.) Think about analyzing those tactical decisions
> critically as with the analysis of chess notation - physical chess indeed.
You could use a motion capture system to input the physical movements of
the fencers into a computer, which would then award touches. Goodbye
referee.
<snip> Should the US really be waiting on France and the rest of the old
world
> to present us with some wireless solution? This is obviously an arena
> where the US could take the lead. (After all - the French regard Jerry
> Lewis as the pinnacle of American TV and media :-)
The fate of modern fencing rests with the US. Even though we are not
the best as far as skill is concerned, we control the media.
> A successful (and not exorbitantly expensive) wireless scoring system
> would sell itself to every club in the US (and elsewhere) for everyday
> use. Just imagine the sigh of relief from your armorer at being able to
> retire the aging reels and scoreboxes (perhaps to drag them out a few
> times a year for FIE beholden tournaments). When the new system matures
> it will naturally gain support for competitive use. Any progress on an
> automatic phrasing system would be an incredible teaching aid.
A wireless scoring system could be accepted by the FIE when they see
that all the clubs and smaller tournaments use the system.
> Design the system right initially, and a current limited implementation
> will still be valid as part of a more extensive system in the future.
> Design the system wrong, and it will hang like an albatross around the
> necks of your fencers (perhaps literally) for decades.
It's going to be exciting. Even if a wireless system was only working
for one weapon, I can foresee it being wildly popular among fencers.
Later,
Eric
The Irreverent Eddie
club! I like this post, it expressed many of the concerns I had about
wireless fencing. I can just imagine the day when "hooking up" and
"plugging in" is just popping a plug into the box. Again, nice post. See
you later!
--Eddie
Rob Seaman
> The JOs was a really neat tournament, wasn't it? Did you ever go around
> the weapons repair booth? I was there repairing equipment. You might
> remember me--a young, short guy with blond hair and glasses.
Thanks for having the fortitude to wade through my message so carefully ;-)
Yeah, I spent a bit of time talking to Dolly and Dan. I was the old guy
with ponytail and glasses...
> There are some difficulties involved with wireless fencing. The
> communication from the fencers' units to the central scoring box is a
> potential weak point--somebody could prevent ANY radio based system from
> working by using a signal jammer. Or even a primitive spark-gap
> transmitter based around a 9-volt battery connected to a metal file.
Gotta go digital. Run the internet to each fencer. (Why not?) TCP/IP
has checksum protection in each packet. A bad guy can take the network
down - they just can't do it undetected. Run whatever security layer you
want on top of TCP. Each fencer is guaranteed to be an unspoofable "host".
> The second problem is *reliable* detection of touches to various
> surfaces such as the bell guard and lame.
In the old days of "Hi-Fi" the adage was to spend your money on the
moving parts - the phono cartridge on one end and the speakers on the
other. Whatever physical process is used to detect and catagorize
touches - this is where to spend the money.
> A system was patented in 1977 (if you are interested I can give you
> the patent number) and was revived recently for use in the Olympics.
> Unfortunately due to the flakiness of this system the FIE was forced
> to drop it.
By its very nature, "flakiness" isn't something that can be removed by
tweaking. The current system of reels and switches avoids most such
problems by applying fundamental laws of physics. Any successful
replacement will require the same level of robustness. First identify
remote sensor(s) that can descriminate a touch versus no touch, on-target
versus off-target (of various sorts). And then figure out how to package
the telemetry from the sensors as part of a wireless fencing system.
Trying to drive the technology the other way around will never work.
Another thing, I've retraced a number of the earlier rec.sport.fencing
threads via dejanews and there appears to be an implicit assumption that
whatever system is fielded it will consist of proprietary technology.
There is no problem with this as far as a particular scoring box or
sensor or whatever, but the system design has to reside with the
international standards organization for the community.
> The communication is no problem at all--I have thought of a system that
> would require no communication whatsoever between the two fencers' units.
Sounds like a neat trick - might violate general relativity though...
> It is quite easy to cheat using the current system. Such cheating
> can and does go undetected. [...] even with a wireless system an
> unscrupulous epee fencer could still short two wires and give himself
> a touch.
Well, even with my previous massive tome I left out a few topics. To
cheat by shorting the wires requires that there continue to be wires.
This may be the case - but the current electrical system really involves
virtually every bit of equipment the fencers use except for some of the
safety items. (Even the clothing serves as insulation.) If you're going
to the extreme to redo everything from the body cord "out" - you may well
want to reconsider everything from the body cord on in, too.
> Which is the other difficulty--if the new system has a fluke, fencers
> will avoid it like the plague even if the problem is solved.
This is the entire problem. Implementing the technology isn't the issue
(although I don't underestimate the complexity of any acceptable solution
that meets the various requirements) - the issue is manufacturing and
marketing the new equipment.
The new system has to be better - than it has to be.
> for a system to be successful, it should be compatible with the
> existing system.
But compatible how?
Equipment wears out and is replaced (rarely as frequently as is
required, however :-) The trick is to match the incremental spread of
any new equipment to the natural lifespan of earlier equipment. A lot
of fencers appear to use electric equipment unplugged during informal
club bouts. So some new standard comes along that requires a new blade.
Or some fancy new lame. You may never use the old lame again, but you
can bet that old blades will continue to be useful.
> Apparently it works. The touches are held in memory so that if the
> fencer leaves the strip, the touch still is sent eventually.
Yeah - I heard this. I'll bet that the "eventual" delivery of the
information (does anybody know if it's digital or analog?) is via some
open-ended protocol. Which is to say that the fencer's equipment
somehow senses that she left the strip, but that the scorebox doesn't
know this.
Let's just say that I remain sceptical.
> Here's another, more radical idea that goes back to the days before
> electrical scoring. In the old days, tips contained a little bit of ink
> which spurted out when a fencer hit somebody. The black showed up quite
> nicely on the white uniforms, and after the touch was awarded, the black
> spot was removed using vinegar.
>
> Why not build a jacket that glows wherever you apply pressure? You
> could even embed a large force-sensing resistor array into the uniform
> which would send a signal to the fencer's scoring unit, which would
> decide if the touch was in a valid location. Bell guard and floor
> touches would be irrelevant because they wouldn't trigger the scoring
> unit.
An excellent line of investigation. Why assume that the sensors should
be mounted on the blade?
> The equipement actually comes from Blue Gauntlet, who has some sort of
> an agreement with the USFA to provide all the reels, floor cords, and
> scoring machines.
Ok - somewhat simpler than I thought, but still requires a large
centralized expense that might better be distributed among all the
competitors. I also watched them roll up the strips at the end of the
competition. Boy - there's a tedious job.
> You could use a motion capture system to input the physical movements of
> the fencers into a computer, which would then award touches. Goodbye
> referee.
Well, the art in all this is knowing what features to emphasize. I found
the referee of special benefit during the foil matches. Especially the
ones who were audible and careful to analyze the phrasing of the attacks.
But a computerized system could be a great aid for the spectators as well
as the competitors - and seems like a really slick teaching aid. One of
the vendors at the J.O.'s had a table set up with some sort of lunge
analysis program. Imagine that expanded into breaking down a complete
fencing phrase or bout into little graphical figures that could be played
and replayed (and edited to explore better options).
No substitute for actually fencing, but as viable as all the X's and O's
the football coaches toss around.
You could also make trade-offs like omitting referees for the pool bouts.
Again - I don't see this arriving soon, but any replacement for the
current, very pragmatic, system should be just as pragmatic - don't just
look at backwards compatibility, prepare for the rapidly approaching
future.
Rob
The Irreverent Eddie
> How about a column of light about 3 feet long? ;-)
Are we talking SpaceBalls or Star Wars? ;-) I HATE it when I get my
Schwartz twisted!!! So then we'd need tinted masks to sheild our eyes?!?!
--Eddie
Eric Schlaepfer
> Thanks for having the fortitude to wade through my message so carefully ;-)
I am very interested about anything electrical that has to do with
fencing.
> Yeah, I spent a bit of time talking to Dolly and Dan. I was the old guy
> with ponytail and glasses...
Hmm...my memory about the whole thing is slightly blurred. Probably
because of lack of sleep and food...
> Gotta go digital. Run the internet to each fencer. (Why not?) TCP/IP
> has checksum protection in each packet. A bad guy can take the network
> down - they just can't do it undetected. Run whatever security layer you
> want on top of TCP. Each fencer is guaranteed to be an unspoofable "host".
There are chips called hopping code generators that are specifically
designed for a secure one-way communication between devices, such as
remote car door locks and garage door openers. They cannot be
intercepted, but they can be jammed like any other device based on radio
technology.
> In the old days of "Hi-Fi" the adage was to spend your money on the
> moving parts - the phono cartridge on one end and the speakers on the
> other. Whatever physical process is used to detect and catagorize
> touches - this is where to spend the money.
Yup, spend money on the tricky part.
> By its very nature, "flakiness" isn't something that can be removed by
> tweaking. The current system of reels and switches avoids most such
> problems by applying fundamental laws of physics. Any successful
> replacement will require the same level of robustness. First identify
> remote sensor(s) that can descriminate a touch versus no touch, on-target
> versus off-target (of various sorts). And then figure out how to package
> the telemetry from the sensors as part of a wireless fencing system.
> Trying to drive the technology the other way around will never work.
That's true.
> Another thing, I've retraced a number of the earlier rec.sport.fencing
> threads via dejanews and there appears to be an implicit assumption that
> whatever system is fielded it will consist of proprietary technology.
> There is no problem with this as far as a particular scoring box or
> sensor or whatever, but the system design has to reside with the
> international standards organization for the community.
Unfortunately the area of fencing scoring technology thrives on
proprietary technology. Look at extension light adaptors--there is no
standard although some are electrically compatible.
This really would be a nice area for a standard so that one company
could not lock in all the customers.
> > The communication is no problem at all--I have thought of a system that
> > would require no communication whatsoever between the two fencers' units.
>
> Sounds like a neat trick - might violate general relativity though...
How about this: Your lame is the sensor, and your opponent hits you,
triggering a colored light and starts your lockout timer. When the
lockout timer finishes, your unit turns off your emitter, which is
connected to your blade, so that you cannot make touches on anybody.
After your colored light goes out, the emitter is switched on again.
This also means that more than two people could fence each other at the
same time and everything would work out!
> Well, even with my previous massive tome I left out a few topics. To
> cheat by shorting the wires requires that there continue to be wires.
> This may be the case - but the current electrical system really involves
> virtually every bit of equipment the fencers use except for some of the
> safety items. (Even the clothing serves as insulation.) If you're going
> to the extreme to redo everything from the body cord "out" - you may well
> want to reconsider everything from the body cord on in, too.
It really doesn't matter what kind of system you end up using, people
will *always* find some way of cheating. With the ink-and-vinegar
scheme I mentioned before, people would soak their entire uniform in
vinegar so that touches would not show up.
> > Which is the other difficulty--if the new system has a fluke, fencers
> > will avoid it like the plague even if the problem is solved.
>
> This is the entire problem. Implementing the technology isn't the issue
> (although I don't underestimate the complexity of any acceptable solution
> that meets the various requirements) - the issue is manufacturing and
> marketing the new equipment.
Of course. It's easy to design a product, but much harder to
manufacture and market it.
> The new system has to be better - than it has to be.
Exactly.
> > for a system to be successful, it should be compatible with the
> > existing system.
>
> But compatible how?
Okay, what I meant was that the new system shouldn't be too different
than the current system. People act weird when you require them to buy
new and incompatible weapons and lames.
> Equipment wears out and is replaced (rarely as frequently as is
> required, however :-) The trick is to match the incremental spread of
> any new equipment to the natural lifespan of earlier equipment. A lot
> of fencers appear to use electric equipment unplugged during informal
> club bouts. So some new standard comes along that requires a new blade.
> Or some fancy new lame. You may never use the old lame again, but you
> can bet that old blades will continue to be useful.
You should also realize that the electrical epee system as we know it
has been around virtually unchanged for more than 65 years. If you took
a modern epee fencer back in time he could still plug in and fence with
his usual equipment.
> > Apparently it works. The touches are held in memory so that if the
> > fencer leaves the strip, the touch still is sent eventually.
>
> Yeah - I heard this. I'll bet that the "eventual" delivery of the
> information (does anybody know if it's digital or analog?) is via some
> open-ended protocol. Which is to say that the fencer's equipment
> somehow senses that she left the strip, but that the scorebox doesn't
> know this.
It's digital. The touch is stored in memory for 350 milliseconds, which
is slightly over the maximum length of time somebody can remain
airborne. On earth, of course.
> Let's just say that I remain sceptical.
Me too. I recently heard that metallic strip material is getting harder
to find. Apparently it was originally used for print shops and
papermaking. Now that they've moved to plastic screens, you can only
get metallic strips from Russia and China. I don't think it's a good
idea to base a wireless scoring system on something like that.
> An excellent line of investigation. Why assume that the sensors should
> be mounted on the blade?
Exactly. No more rewiring blades, cleaning tips, tightening barrels,
etc.
> Ok - somewhat simpler than I thought, but still requires a large
> centralized expense that might better be distributed among all the
> competitors. I also watched them roll up the strips at the end of the
> competition. Boy - there's a tedious job.
I was busy coiling floor cords and lugging reels around. All the while
I was thinking about wireless scoring. :)
> Well, the art in all this is knowing what features to emphasize. I found
> the referee of special benefit during the foil matches. Especially the
> ones who were audible and careful to analyze the phrasing of the attacks.
>
> But a computerized system could be a great aid for the spectators as well
> as the competitors - and seems like a really slick teaching aid. One of
> the vendors at the J.O.'s had a table set up with some sort of lunge
> analysis program. Imagine that expanded into breaking down a complete
> fencing phrase or bout into little graphical figures that could be played
> and replayed (and edited to explore better options).
>
> No substitute for actually fencing, but as viable as all the X's and O's
> the football coaches toss around.
>
> You could also make trade-offs like omitting referees for the pool bouts.
It's an interesting idea. A computer is not biased like all human
referees.
> Again - I don't see this arriving soon, but any replacement for the
> current, very pragmatic, system should be just as pragmatic - don't just
> look at backwards compatibility, prepare for the rapidly approaching
> future.
We must also use the existing system as a springboard. You can't change
everything too quickly because people dig in their heels.
Later,
Eric
Eric Dew
through you, allowing you to better make those pesky high six parries.
EDEW
John Twernbold
> With the blast shield down, you should be able to feel the force flowing
> through you, allowing you to better make those pesky high six parries.
Ancient weapons and hokey religions are no match for a good trigger in your bell
guard, kid....
John
Eric Dew
David W. Neevel
>
> > I was very impressed by the sheer volume of equipment arrayed across the
> > convention center in Sacramento. Conservatively there were several
> > hundred thousand dollars in electric scoring equipment distributed among
> > the thirty or so strips. If I understood the marks that were visible on
> > various equipment, the USFA must have borrowed equipment from clubs
> > across the country. The logistics of "requisitioning" and delivering
> > all this stuff must be staggering. More to the point, it is the
> > responsibility of the USFA/FIE affiliate running a particular competition
> > to acquire all this stuff either directly or by borrowing it. Shipping it
> > to each new venue must also be the cause of much of the flakiness of the
> > equipment. (Moving instruments from telescope to telescope always causes
> > problems.)
>
> The equipement actually comes from Blue Gauntlet, who has some sort of
> an agreement with the USFA to provide all the reels, floor cords, and
> scoring machines.
>
Too add some detail to the scoring equipment logistics:
The machines, reels and floor cords are all leased from Blue Gauntlet.
Some of the strips are owned by the USFA, some by the Illinios Division
which are rented by the USFA on more-or-less permanent basis(many years
back the late Fred Rhodes got them from paper mills that were looking to
get rid of old conveyor strips and were more than happy to donate them
to sombody willing to pay the cost of hauling them away). When not at a
NAC, the equipment lives in a warehouse in Coralville, Iowa in the
magenta crates that do double duty as tables in the armory and on the
pallets used stack the strips. Why Iowa? Because armorer and Technical
Resource Group member Ron Herman lives in Iowa City and can get cheap
storage space there.
Prior to each national event, Roadway transportation sends a truck to
pick the crates up and drive them to wherever the tournament is, and
picks them up for transit back to Iowa after the tournament is done.
While I'm familiar with the "but it was just working on the other
simulation station over there!!!" proclivities of test equipment, the
scoring gear actually comes through the transit quite well. The reels
all go into padded denim sleeves before being crated up, and the SG-12
machines live in foam-lined cases that also go in padded sleeves for
transport. It all gets checked at the site before being set up, and most
problems arise from use at the competitions. For example, at the last
NAC I worked, except for one reel which had a mercotac go bad the
previous NAC and two with broken rubber feet (all waiting for
replacement parts), the reel problems I encountered were lame' snaps on
the plugs wearing out and breaking over the course of the tournament.
The overwhelming majority of problems that occur on strip are due to a
fencers' equipment (this is why armorers test body cords at check in--
flaky body cords are the number-one source of 'mysterious' problems on
strip). The main maintenance issue are the aging strips, many of which
are on their last legs and are needing to be continuously soldered and
patched (the worst of then have been mummified in duct tape and given
labels like 'nasty strip- composed entirely of degenerate matter' to
prevent their use unless absolutely necessary). This problem will be
greatly reduced by the new strips that have been purchased for Junior
Worlds.
Even with some more advanced, less bulky system that might be developed
in the future, the USFA probably will still for the sake of consistency
want to maintain a stock of equipment to be used at national events,
rather than relying on whatever the local clubs and divisions can cobble
together.
Good thread, in all. As heretofore prohibitively expensive technologies
come down in price, a number of new approaches to handling scoring
should become realistic (I'd love to see some kind of directional force
sensor reintroduced to sabre once an affordable, reliable, and
calibratable option exists). As always, a real limitation will be not
just what is technically possible, but the extent to which backwards
compatibility is demanded by the market -- a system which requires
fencers to replace all of their current gear at once is probably not
going to fly, unless the replacement is _really_ cheap. I suspect that
reels, boxes, and floor cords are still going to be a common sight for
the next 10-20 years, both due to backwards compatibility issues and due
to the time it will take for real-world use to determine what approach
to wireless scoring works best. It may well be that a split will
develop, with expensive systems that meet all of the FIEs security
concerns used at World Cups and the like, while cheaper, less secure
systems are used at the local level.
-Dave Neevel
Rob Seaman
> Rob? Emily's Dad?!?! Fancy seeing you here!!! This is Eddie, from the
> club! I like this post, it expressed many of the concerns I had about
> wireless fencing. I can just imagine the day when "hooking up" and
> "plugging in" is just popping a plug into the box.
Hullo Eddie. Yes, even mummified adults use the internet ;-)
The trick is to avoid remaining in the "imagine" stage, but rather help
build and introduce the new technology.
Rob
Dan & Liz Gorman
> How about this: Your lame is the sensor, and your opponent hits you,
> triggering a colored light and starts your lockout timer. When the
> lockout timer finishes, your unit turns off your emitter, which is
> connected to your blade, so that you cannot make touches on anybody.
> After your colored light goes out, the emitter is switched on again.
>
> This also means that more than two people could fence each other at the
> same time and everything would work out!
The Army has developed a material that can detect where a bullet hit a soldier,
from what direction, and from about how far. It also tracks internal and
external temperature and some other cool stuff. Supposedly though, the Army has
lost interest in it, and it's being adapted/used by other professions. The
article I saw on it specifically mentioned fire fighters using it, but I'd think
it would be a great material for a lame depending on how light weight it is.
Since it was designed to track all soldiers in a battlefield, I don't think a
room full of fencers would be that tricky (unless that's the reason the Army
dropped it).
Anyone else know anything about this?
Dan
Rob Seaman
> There are chips called hopping code generators that are specifically
> designed for a secure one-way communication between devices, such as
> remote car door locks and garage door openers. They cannot be
> intercepted, but they can be jammed like any other device based on radio
> technology.
There are a lot of options. Just pick a standard so basic and widespread
that you will be able to count on supporting it (cheaply) for several
decades. Also isolate the data transport layer so that you can upgrade
to new technology as it becomes available.
A fencer (or club) should be able to buy a blade from one vendor, a
transmitter from another, the receiver from a third and the scorebox
from a fourth vendor. The protocols connecting the different devices
(whatever they are) need to be publically specified.
> Unfortunately the area of fencing scoring technology thrives on
> proprietary technology. Look at extension light adaptors--there is no
> standard although some are electrically compatible.
The FIE/USFA have to take the lead here. They manage to specify the
requirements on the blades and other equipment very precisely. They
could also require that any scoring apparatus use a specific plug on
the body cord, for instance. I'm absolutely positive that you guys pay
ten times (conservatively) as much for your wiring as would otherwise
be necessary. Fencing isn't the only community that requires rugged
electrical plugs that maintain continuity while flailing around.
Gold-plated mil-spec plugs to match whatever standard would probably be
cheaper than what you've been using for the last twenty or fifty years.
(Spoken as a parent who keeps finding about new pieces of equipment that
need to be bought - equipment that I didn't previously know existed on
the planet.)
> How about this: Your lame is the sensor, and your opponent hits you,
> triggering a colored light and starts your lockout timer. When the
> lockout timer finishes, your unit turns off your emitter, which is
> connected to your blade, so that you cannot make touches on anybody.
> After your colored light goes out, the emitter is switched on again.
There have been a number of messages complaining about ad-hoc FIE rule
changes. But actually I found the USFA rule book to be quite a readable
and well-written document. (I've read some pretty opaque documents from
NASA and such.) The rules of engagement for the three weapons are
particularly well explained. There are very clear expressions of when
on-target and off-target touches are permitted to register. Once the
touches register, there are clear expressions of how (or whether) the
referee should award the point.
Start with what the rules are (or what they ought to be) and specify
what any scoring system needs to do. Then build the simplest, most
rubust, implementation that embodies the rules. Make sure that the
system doesn't introduce unnecessary, unspecified, requirements such as
requiring a tether (more to carry a ground reference than the resulting
signals). Make sure that the system is modular and affordable.
That's all you have to do ;-)
> This also means that more than two people could fence each other at the
> same time and everything would work out!
If this is a desired goal, by all means include it in the specifications.
Make sure that you include a minimum number of competitors that the
equipment must be able to support simultaneously. Is it important who
scored the touch? If you have to track even just three fencers, there
are suddenly six numbers on the score board instead of two (A vs B,
B vs C, C vs A).
> It really doesn't matter what kind of system you end up using, people
> will *always* find some way of cheating.
Security against various weeknesses and failure modes has to be built in
from the beginning. That people are flawed has to be recognized. That
they will succeed at attempts to cheat doesn't have to be.
> Okay, what I meant was that the new system shouldn't be too different
> than the current system. People act weird when you require them to buy
> new and incompatible weapons and lames.
Well, they act weirder when you change the rules or "feel" of the sport
in a misguided attempt to support the new equipment. Every new feature
has the possibility of unwanted side-effects. My guess (as an educated
outsider) is that the wireless system now being promoted will have too
many side-effects to be successful in the long term. There may be enough
geopolitical pressure to ensure that it is used through to Sydney.
HMS Lion
foil, may be to revert to the old marker-tip system. In principle,
right-of-way determines priority, and a simple marker system is adequate to
establish materiality of a touch.
To be quite honest, there is nothing wrong with going back to the Judge's
Eyeball, Mk1 mod0. I really think that the current pressure to electrify
scoring at lower and lower levels - and now to wireless scoring - is driven
more by pressure of fencing equipment suppliers and the sex appeal of gadgetry
than by any real requirement.
In any event, I agree that a total rethink of the scoring system is needed.
We can do better.
V/R:
Mike McDaniel
Rob Seaman
> To be quite honest, there is nothing wrong with going back to the Judge's
> Eyeball, Mk1 mod0. I really think that the current pressure to electrify
> scoring at lower and lower levels - and now to wireless scoring - is driven
> more by pressure of fencing equipment suppliers and the sex appeal of
> gadgetry than by any real requirement.
An excellent point. "No system" is better than a bad system.
Also, in the broader context of Olympic sports in general - why is it
taken for granted that fencing requires an outboard scoring system when
other sports like judo and wrestling do not? Most Olympic sports require
expert referees to mediate the action. Many Olympic sports are pretty
opaque to uneducated viewers. What the heck is an "axel"? Why shouldn't
gymnasts bend their knees when they land? The audience had to be trained
to be able to follow all sorts of sports.
The "super slo-mo" camera is ancient technology by this point. One might
think that it would be easier to omit the electronic scoring entirely
(wireless or not) and just allow the officials to score the matches.
The audience could follow any complicated action via slow or stop motion
replays.
Just applying the current huge effort attending to wireless scoring into
training and public relations might do even more to spread the sport.
On the other hand, one reason the world's great popular sports (like
soccer, basketball, tennis, golf, and what have you) prosper is because
they don't require highly trained officials to merely have a friendly
game. The spread of fencing should ideally not be limited by the
availability of qualified referees. An appropriate application of
scoring technology could help greatly to spread the sport. (Assuming
that popularizing the sport is really a goal.)
The current electrical system isn't just tethering the fencers, it's
tying down fencing in general.
HMS Lion
high-friction tips in epee) is that it's FAST. No plugging in, no checking
tips, no checking guards, just take the strip, salute, and fence. Although the
cost and reliability advantages are not to be sneered at.
As for making fencing more spectator-friendly, probably the best single thing
that could be done is to have a big scoreboard at strip-side so people can
figure out who is doing what to whom. Slow-motion video would be great for
television (fencing is made for TV, but does need post-action production to
wring out the full drama). And, maybe, tinted uniforms to show the target
area.
V/R:
Mike McDaniel
Eric Schlaepfer
> The machines, reels and floor cords are all leased from Blue Gauntlet.
> Some of the strips are owned by the USFA, some by the Illinios Division
> which are rented by the USFA on more-or-less permanent basis(many years
> back the late Fred Rhodes got them from paper mills that were looking to
> get rid of old conveyor strips and were more than happy to donate them
> to sombody willing to pay the cost of hauling them away). When not at a
> NAC, the equipment lives in a warehouse in Coralville, Iowa in the
> magenta crates that do double duty as tables in the armory and on the
> pallets used stack the strips. Why Iowa? Because armorer and Technical
> Resource Group member Ron Herman lives in Iowa City and can get cheap
> storage space there.
> Prior to each national event, Roadway transportation sends a truck to
> pick the crates up and drive them to wherever the tournament is, and
> picks them up for transit back to Iowa after the tournament is done.
They have a shipping point located across the street from the warehouse,
which makes it easy.
> While I'm familiar with the "but it was just working on the other
> simulation station over there!!!" proclivities of test equipment, the
> scoring gear actually comes through the transit quite well. The reels
> all go into padded denim sleeves before being crated up,
At the JOs Dolly and I had the privilege of putting all of those reels
into the cute denim bags. Makes me glad they used velcro instead of
zippers...
> and the SG-12 machines live in foam-lined cases that also go in padded sleeves for
> transport. It all gets checked at the site before being set up, and most
> problems arise from use at the competitions. For example, at the last
> NAC I worked, except for one reel which had a mercotac go bad the
> previous NAC and two with broken rubber feet (all waiting for
> replacement parts), the reel problems I encountered were lame' snaps on
> the plugs wearing out and breaking over the course of the tournament.
Those seem to be the most fragile part of the reel. We had several more
wear out at the JOs.
> The overwhelming majority of problems that occur on strip are due to a
> fencers' equipment (this is why armorers test body cords at check in--
> flaky body cords are the number-one source of 'mysterious' problems on
> strip). The main maintenance issue are the aging strips, many of which
> are on their last legs and are needing to be continuously soldered and
> patched (the worst of then have been mummified in duct tape and given
> labels like 'nasty strip- composed entirely of degenerate matter' to
When we were taking down the strips Bill spent a long time on one of the
strips. The rule of thumb is that if you cannot see the hole from
standing up, don't bother patching it.
Reminds me of the time when Dan was soldering a strip that had been
rolled up with a nail inside.
> prevent their use unless absolutely necessary). This problem will be
> greatly reduced by the new strips that have been purchased for Junior
> Worlds.
I'd sure like to ditch the concept of the metallic strip entirely.
> Even with some more advanced, less bulky system that might be developed
> in the future, the USFA probably will still for the sake of consistency
> want to maintain a stock of equipment to be used at national events,
> rather than relying on whatever the local clubs and divisions can cobble
> together.
>
> Good thread, in all. As heretofore prohibitively expensive technologies
> come down in price, a number of new approaches to handling scoring
> should become realistic (I'd love to see some kind of directional force
> sensor reintroduced to sabre once an affordable, reliable, and
> calibratable option exists). As always, a real limitation will be not
> just what is technically possible, but the extent to which backwards
> compatibility is demanded by the market -- a system which requires
> fencers to replace all of their current gear at once is probably not
> going to fly, unless the replacement is _really_ cheap. I suspect that
> reels, boxes, and floor cords are still going to be a common sight for
> the next 10-20 years, both due to backwards compatibility issues and due
> to the time it will take for real-world use to determine what approach
> to wireless scoring works best. It may well be that a split will
> develop, with expensive systems that meet all of the FIEs security
> concerns used at World Cups and the like, while cheaper, less secure
> systems are used at the local level.
A nice wireless system would plug into the fencers' body cord, which
means that fencers can still use the same equipment, but the tournament
site would only provide wireless units.
Later,
Eric
Eric Schlaepfer
> There are a lot of options. Just pick a standard so basic and widespread
> that you will be able to count on supporting it (cheaply) for several
> decades. Also isolate the data transport layer so that you can upgrade
> to new technology as it becomes available.
Of course.
> A fencer (or club) should be able to buy a blade from one vendor, a
> transmitter from another, the receiver from a third and the scorebox
> from a fourth vendor. The protocols connecting the different devices
> (whatever they are) need to be publically specified.
Eventually yes, they should be a standard.
> > Unfortunately the area of fencing scoring technology thrives on
> > proprietary technology. Look at extension light adaptors--there is no
> > standard although some are electrically compatible.
>
> The FIE/USFA have to take the lead here. They manage to specify the
> requirements on the blades and other equipment very precisely. They
> could also require that any scoring apparatus use a specific plug on
> the body cord, for instance. I'm absolutely positive that you guys pay
> ten times (conservatively) as much for your wiring as would otherwise
> be necessary. Fencing isn't the only community that requires rugged
> electrical plugs that maintain continuity while flailing around.
If the wireless units will come from the club level up, the FIE and even
the USFA won't have much to do with standards at first.
Way back in the old days in Europe, due to a shortage of banana jacks,
some fencers used weapons that had a three-pin connector similar to a
DIN connector. They also used these connectors in the US for old
radiotelephone microphones.
> Gold-plated mil-spec plugs to match whatever standard would probably be
> cheaper than what you've been using for the last twenty or fifty years.
Reminds me of something. An armorer, Carl Oberg, has a bag full of
gold-plated banana jacks. Apparently you can get anything metal
gold-plated fairly cheaply.
> (Spoken as a parent who keeps finding about new pieces of equipment that
> need to be bought - equipment that I didn't previously know existed on
> the planet.)
>
> > How about this: Your lame is the sensor, and your opponent hits you,
> > triggering a colored light and starts your lockout timer. When the
> > lockout timer finishes, your unit turns off your emitter, which is
> > connected to your blade, so that you cannot make touches on anybody.
> > After your colored light goes out, the emitter is switched on again.
>
> There have been a number of messages complaining about ad-hoc FIE rule
> changes. But actually I found the USFA rule book to be quite a readable
> and well-written document. (I've read some pretty opaque documents from
> NASA and such.) The rules of engagement for the three weapons are
> particularly well explained. There are very clear expressions of when
> on-target and off-target touches are permitted to register. Once the
> touches register, there are clear expressions of how (or whether) the
> referee should award the point.
Try reading the rules on scoring machines. So much of the functionality
is *implied* in the rules.
> Start with what the rules are (or what they ought to be) and specify
> what any scoring system needs to do. Then build the simplest, most
> rubust, implementation that embodies the rules. Make sure that the
> system doesn't introduce unnecessary, unspecified, requirements such as
> requiring a tether (more to carry a ground reference than the resulting
> signals). Make sure that the system is modular and affordable.
I suspect that the rules were designed around an early relay-based
scoring machine.
> That's all you have to do ;-)
>
> > This also means that more than two people could fence each other at the
> > same time and everything would work out!
>
> If this is a desired goal, by all means include it in the specifications.
> Make sure that you include a minimum number of competitors that the
> equipment must be able to support simultaneously. Is it important who
> scored the touch? If you have to track even just three fencers, there
> are suddenly six numbers on the score board instead of two (A vs B,
> B vs C, C vs A).
Ever been to Q-Zar? They have many people fighting at the same time,
and they have a central computer which keeps track of the score.
> > It really doesn't matter what kind of system you end up using, people
> > will *always* find some way of cheating.
>
> Security against various weeknesses and failure modes has to be built in
> from the beginning. That people are flawed has to be recognized. That
> they will succeed at attempts to cheat doesn't have to be.
Cheating can be made more difficult however.
> > Okay, what I meant was that the new system shouldn't be too different
> > than the current system. People act weird when you require them to buy
> > new and incompatible weapons and lames.
>
> Well, they act weirder when you change the rules or "feel" of the sport
> in a misguided attempt to support the new equipment. Every new feature
> has the possibility of unwanted side-effects. My guess (as an educated
> outsider) is that the wireless system now being promoted will have too
> many side-effects to be successful in the long term. There may be enough
> geopolitical pressure to ensure that it is used through to Sydney.
The wireless scoring system that was being promoted by the FIE has been
dropped. Apparently the FIE had a demonstration tournament using the
wireless equipment. The result was a disaster--the system didn't work
well at all.
Later,
Eric
Rob Seaman
> Try reading the rules on scoring machines. So much of the functionality
> is *implied* in the rules.
I guess that would be "Appendix B to the Material Rules"? Rather specific
to the current technology, but not really that bad. They were right to
relegate it to an appendix, where it is obviously secondary to the main
body of rules. It does tend to mingle fundamental parameters with
parameters derived solely from the electrical scoring.
Do you have an example of functionality that remains ambiguous when also
interpreted in the light of the rules of fencing (as opposed to the rules
of "electro-mechanical" fencing)?
> Ever been to Q-Zar? They have many people fighting at the same time,
> and they have a central computer which keeps track of the score.
Must be some hyper-modern innovation since the age of dinosaurs. Of
course, a "central computer" is a rather extinct notion itself.
Dan & Liz Gorman
> The nicest part about reverting to the Mk1 mod 0 eyeball (perhaps aided by
> high-friction tips in epee) is that it's FAST. No plugging in, no checking tips, no
> checking guards, just take the strip, salute, and fence. Although the cost and
> reliability advantages are not to be sneered at.
And so many touches get missed.
Dan
Rob Seaman
In the spirit of continuing to talk about things I'm entirely ignorant of...
One could construct a perfectly reliable wireless system for foil and epee
using virtually off the shelf parts. The only trouble is...obvious.
Use an epee style point for both weapons - meaning a momentary contact
switch that is normally open and that makes a circuit during a touch.
A couple of bucks of Radio-Shaft parts should get you a sample-and-hold
to latch each touch. Either mount a light bulb and a buzzer on each
fencer - or I imagine $50 or $100 would get some sort of wireless data
transmitter. Another $100 bucks for a receiver that would plug into
the current scoring boxes. However much money to implement whatever
public/private key security scheme you want.
Omit the lame for foil, but tint the clothing to indicate the legal
target area. Forget about the grounded strip.
Surround the fencers with enough TV cameras to ensure that that all
angles are covered. Put the referee/director in the booth with a color
announcer.
Every time a light goes off - stop the action. (Or rather - wait a
sufficient length of time to cover "a period of time X (which bears no
relationship to 'fencing time' ...)" - from Appendix B to the Material
Rules".)
Have the referee scrutinize the various camera angles in super slo-mo.
Referee scores appropriately.
Repeat.
David W. Neevel
>
> Mike McDaniel (hms...@aol.com) writes:
>
> > To be quite honest, there is nothing wrong with going back to the Judge's
> > Eyeball, Mk1 mod0. I really think that the current pressure to electrify
> > scoring at lower and lower levels - and now to wireless scoring - is driven
> > more by pressure of fencing equipment suppliers and the sex appeal of
> > gadgetry than by any real requirement.
>
> An excellent point. "No system" is better than a bad system.
>
OTOH, the reason why "No System" was replaced with electric gear is
because it was a bad system-- cumbersome, time consuming, and prone to
error and bias. Go find somebody who's been around for a while and who
remembers competing back in the 50's and 60's before electric scoring
was widespread (A couple of members of one of our beginners' classes
were having just such a conversation with a club member who was an
A-rated foilist back then).
The supremely competent, unbiased judges some folks talk about as being
preferable to electrical fencing exist in the same world as Platonic
philosopher kings (i.e., not this one).
-Dave Neevel
David W. Neevel
> >
> >OTOH, the reason why "No System" was replaced with electric gear is
> >because it was a bad system-- cumbersome, time consuming, and prone to
> >error and bias. Go find somebody who's been around for a while and who
> >remembers competing back in the 50's and 60's before electric scoring
> >was widespread (A couple of members of one of our beginners' classes
> >were having just such a conversation with a club member who was an
> >A-rated foilist back then).
> >The supremely competent, unbiased judges some folks talk about as being
> >preferable to electrical fencing exist in the same world as Platonic
> >philosopher kings (i.e., not this one).
> >
> I guess it's another case of looking fondly at the past with rose tinted
> glasses. As Buzz Hurst wrote in an article in AmFence several issues ago,
> the past was not as good as the present, and possibly the future.
>
> The current electrical scoring system was created specifically because
> enough people were screwed by incompetence to demand a change.
>
The 'good old days' sentiment also get shot down on another issue: cost.
Yes, that's right, cost. When folks talk about people being 'cheap',
remember that competent people generally expect to be compensated for
their work, at least to the point of breaking even on the costs of
getting to the tournament and staying there if it's more than just a
day-trip.
Think an SG-12, 2 reels, and floor cords are expensive? Let's say that
sufficient numbers of competent judges who are willing to burn up
vacation days (or take unpaid time-off) to work a NAC really did exist
(and that _is_ a stretch-- they sometimes have enough trouble finding
referees they way things are now). Try imagining the cost of flying in
an extra 4 people per strip to serve as judges, putting them up in hotel
rooms, and paying their per-diem and meals. The cost of servicing each
strip at a NAC would be probably wind up being equivalent to purchasing
a complete new set of scoring equipment for that strip, plus a good
chunk of the cost of a new metal piste. Even at local events, you'd
still need to pay your human judges a per-deim to attract competent
people. Now multiply that by the number of competitions held over the
course of a year and it becomes readily apparant that the purchase cost
of a set of scoring gear that will last at least 10-15 years works out
to be less than the cost of the dedicated (i.e. not competitors pressed
into service), practiced judges you'd need to have a chance of running
decent events over the course of the same period.
-Dave Neevel
Eric Dew
David W. Neevel <nee...@execpc.com> wrote:
>Rob Seaman wrote:
>>
>> Mike McDaniel (hms...@aol.com) writes:
>>
>> > To be quite honest, there is nothing wrong with going back to the Judge's
>> > Eyeball, Mk1 mod0. I really think that the current pressure to electrify
>> > scoring at lower and lower levels - and now to wireless scoring - is driven
>> > more by pressure of fencing equipment suppliers and the sex appeal of
>> > gadgetry than by any real requirement.
>>
>> An excellent point. "No system" is better than a bad system.
>>
>
>because it was a bad system-- cumbersome, time consuming, and prone to
>error and bias. Go find somebody who's been around for a while and who
>remembers competing back in the 50's and 60's before electric scoring
>was widespread (A couple of members of one of our beginners' classes
>were having just such a conversation with a club member who was an
>A-rated foilist back then).
>The supremely competent, unbiased judges some folks talk about as being
>preferable to electrical fencing exist in the same world as Platonic
>philosopher kings (i.e., not this one).
>
I guess it's another case of looking fondly at the past with rose tinted
glasses. As Buzz Hurst wrote in an article in AmFence several issues ago,
the past was not as good as the present, and possibly the future.
The current electrical scoring system was created specifically because
enough people were screwed by incompetence to demand a change.
EDEW
mary anne walker
>
> HMS Lion wrote:
>
> > The nicest part about reverting to the Mk1 mod 0 eyeball (perhaps aided by
> > high-friction tips in epee) is that it's FAST. No plugging in, no checking tips, no
> > checking guards, just take the strip, salute, and fence. Although the cost and
> > reliability advantages are not to be sneered at.
>
> And so many touches get missed.
>
Or attributed as arriving in the wrong place.....In foil, for
example.....what do you mean that was her touch? She hit me in the
knee! She even acknowledged she hit me off-target! Later...What do you
mean I hit her off target, she acknowledged I hit her in the chest?
(true story---we later fenced a one-touch foil bout to decide for
ourselves who actually won)
The Virginia Division hosts a tournament every year with dry foil and
dry sabre. (epee we fence electric). I think it serves to remind most
of us why we like fencing electric.......
--
Until later,
Mary Anne Walker
************************************************************
"Artillery adds dignity to what would otherwise be a vulgar brawl."
-Murphy's Law of Combat #62
"If you cannot convince them, confuse them." - Harry S. Truman
"You don't have to burn books to destroy a culture. Just get people
to stop reading them." - Ray Bradbury
Chuck Allen
Eric Schlaepfer wrote:
>
> Unfortunately the area of fencing scoring technology thrives on
> proprietary technology. Look at extension light adaptors--there is no
> standard although some are electrically compatible.
>
Actually, the voltages and polarity of the signals going to extension
ARE standard, (14-18 VDC, positive ground) with the exception of Prieur,
which USED to use the same standard and later reversed the polarity and
changed the voltage (3.5 to 4 VDC, negative ground) to be different from
everybody else. While I have seen Uhlmann, Paul, Allstar, St.Georges and
even Russian towers in this country, I have not seen any Prieur
extension lights, or for that matter, any other than Prieur two-weapon
club boxes, so perhaps there is some sort of lesson to be learned from
this.
Granted, the various manufacturers use different pin-outs on the
connector between the scoring box and the extension lights, but this
problem can be cured with a simple adaptor or, in extreme cases, by
rewiring the connector. I have been making "universal" extension lights
which sell for $150/set, and which, with the proper adapter, will work
with most any scoring machine. I have been doing this for the last
decade or so, and have a pretty good handle on all the possible
problems.
Chuck Allen
Chuck Allen
Rob Seaman wrote:
>
> Every time a light goes off - stop the action. (Or rather - wait a
> sufficient length of time to cover "a period of time X (which bears no
> relationship to 'fencing time' ...)" - from Appendix B to the Material
> Rules".)
>
> Have the referee scrutinize the various camera angles in super slo-mo.
> Referee scores appropriately.
>
The problem with this is the sheer time it would take to run a bout.
With this system the 1998 Pittsburgh NAC (where many of the officials
and armourers got to bed between 1 and 2AM and the armourers were up
again at 5:30Am, would be still going on today! <g>
Chuck Allen
Rob Seaman
competent person can provide a very user friendly, high bandwidth interface.
Since people are an expensive recurring cost, fencing clubs and federations
can justify spending a large amount of capital on scoring equipment that can
replace some of the humans.
So consider the qualities necessary for a good scoring system for fencing (or
rather, for a good touch reporting system). Ignore the wireless issue.
Imagine
that this has either been perfected or that somebody puts their R&D dollars
into improving the tether system.
A touch reporting system needs to distinguish on-target versus off-target
touches, needs to time and lockout additional touches as prescribed by the
rules,
needs to sense and report various error conditions - and needs to provide the
information to the humans.
What humans?
Well, obviously the referee. And the fencers. And the spectators (if any)
would appreciate some clue about what's going on.
Both fencers?
Consider mounting the lights on the fencers' masks. This is one of the
features of the wireless system currently being tested (well, recently
tested and perhaps to be tried again). Why did they choose to put the
lights on the masks? After all, they pass the same information through
the strip to the normal scoring box. One can imagine several reasons for
adding lights to the masks:
1) They don't fully trust the strip comm-link - the mask lights provide
a redundant signalling capability.
2) They intend to sell the same system to clubs for everyday use.
Club fencing rarely includes grounded strips. The mask lights would
be the only signal in this case. (Assuming the system will even work
without the strip.)
3) Mounting the lights on the fencers gives a 21st century look to the
sport to attempt to lure TV coverage.
Focus on #2. Use only head mounted lights (with or without tethers).
Which head should have which light?
A familiar children's party game involves having each player select a playing
card - sight unseen - from a deck and then sticking it to his or her own
forehead. Hijinks ensue. This is also a logic puzzler - what procedure should
a player follow to discern the card on the player's own head?
I don't know if the intent is to mount the lights such that they are invisible
to the same fencer. But they are guaranteed not to be as visible to one or
the other fencer. Worst case - a fencer's own lights are invisible. (Or is
that
the best case? It might be extremely disconcerting to have a light mounted
on your own head glaring in your eyes all of a sudden.)
So - fencer "A" scores a touch. Should A or B be the one fencer who knows
about it? I can see consequences either way. What do you think? Phone a
friend - ask the audience - you have a 50/50 chance either way.
Also - recall that these high tech 21st century centurions will be fencing
with a plate glass bay window in front of their eyes.
I mentioned the various recent rule changes to an acquaintance at work who
both fenced a bit in his glory days and also was quite successful in SCA style
bouts. He immediately picked up on the tactical possibilities inherent in
being
able to watch your opponent's eyes. (Perhaps the FIE rules committee should
have a field trip to an SCA hosted tournament.) I rummaged around dejanews
a bit but didn't find any discussion about this issue - but I'm sure others
must
have considered it.
But combine the new "watch the eyes" data channel with the new "I've got
lights on my head that I can't see" feature. This opens up a whole new
category of feints and misdirection.
Watch his blade, watch his blade, attack! - glance at his head...
mary anne walker
sufficiently competent referees who are willing to travel as it is, even
within a relatively small state like Virginia (as opposed to texas or to
my home state). If anyone doesn't believe me, they should try hiring a
substntial number of referees at any given time on a somewhat limited
budget. Like say for a Division Championship, or for a State Collegiate
Championship. Then when you try moving the scale up to trying to hire
for a NAC or for a regional circuit event.....Most of the people out
there who are competent to referee either get used to death, or they
have this weird desire to actually fence, if they're going to take the
time off from work and families and to travel somewhere and spend all
day in a gym or convention center.
David W. Neevel wrote:
> The 'good old days' sentiment also get shot down on another issue: cost.
> Yes, that's right, cost. When folks talk about people being 'cheap',
> remember that competent people generally expect to be compensated for
> their work, at least to the point of breaking even on the costs of
> getting to the tournament and staying there if it's more than just a
> day-trip.
> Think an SG-12, 2 reels, and floor cords are expensive? Let's say that
> sufficient numbers of competent judges who are willing to burn up
> vacation days (or take unpaid time-off) to work a NAC really did exist
> (and that _is_ a stretch-- they sometimes have enough trouble finding
> referees they way things are now). Try imagining the cost of flying in
> an extra 4 people per strip to serve as judges, putting them up in hotel
> rooms, and paying their per-diem and meals. The cost of servicing each
> strip at a NAC would be probably wind up being equivalent to purchasing
> a complete new set of scoring equipment for that strip, plus a good
> chunk of the cost of a new metal piste. Even at local events, you'd
> still need to pay your human judges a per-deim to attract competent
> people. Now multiply that by the number of competitions held over the
> course of a year and it becomes readily apparant that the purchase cost
> of a set of scoring gear that will last at least 10-15 years works out
> to be less than the cost of the dedicated (i.e. not competitors pressed
> into service), practiced judges you'd need to have a chance of running
> decent events over the course of the same period.
>
> -Dave Neevel
John Hasler
> What about scoring for epee?
Two approaches:
a) Insulate the mask. Then all you have to do is determine whether or not
you have hit metal. I have previously explained how to do that.
b) Use the method I described for discriminating between the foil guard and
the lame to discriminate between the epee guard and the mask.
> This might sound a little odd, but it would help if there were a circuit
> which would maintain a point of the same electrical potential between
> both modules using capacitive coupling.
That won't really get you a reference. Everything is coupled to everything
else and the capacitances are constantly changing. Fortunately, it isn't
necessary.
> Have you tried any practical experiments using your technique?
Except for what I have posted here it's all in my head.
--
John Hasler
jo...@dhh.gt.org
Dancing Horse Hill
Elmwood, Wisconsin
Eric Schlaepfer
> Actually, the voltages and polarity of the signals going to extension
> ARE standard, (14-18 VDC, positive ground) with the exception of Prieur,
Where does the 14-18VDC come from? I thought that the entire scoring
system was 12V, possibly 13.8V when run off a car battery.
> which USED to use the same standard and later reversed the polarity and
> changed the voltage (3.5 to 4 VDC, negative ground) to be different from
The Prieur system is standard logic level TTL. To me it makes a lot
more sense than the Uhlmann box which outputs -12VDC on a signal line
referenced to the common line. There are many more chips that can
handle TTL directly than there are chips that use -12VDC.
I build extension lights for the clubs in my area, so I have met with
quite a few different types of boxes myself. For the Uhlmann, Russian,
and the orange Prieur boxes I connect the relays directly to the scoring
machine. For the Prieur competition and the Eigertek machines I add a
single driver chip which accepts TTL logic level and drives relays or
light bulbs directly. For relays the chip includes built-in diodes for
back-EMF protection.
> everybody else. While I have seen Uhlmann, Paul, Allstar, St.Georges and
> even Russian towers in this country, I have not seen any Prieur
> extension lights, or for that matter, any other than Prieur two-weapon
> club boxes, so perhaps there is some sort of lesson to be learned from
> this.
The only Prieur box I know of that uses TTL extension light outputs is
the competition model. The older orange club boxes use the weird -12VDC
system.
The reason you don't see Prieur extension lights is because they cost
more than most clubs can afford.
> Granted, the various manufacturers use different pin-outs on the
> connector between the scoring box and the extension lights, but this
> problem can be cured with a simple adaptor or, in extreme cases, by
> rewiring the connector. I have been making "universal" extension lights
> which sell for $150/set, and which, with the proper adapter, will work
> with most any scoring machine. I have been doing this for the last
> decade or so, and have a pretty good handle on all the possible
> problems.
Do your extension lights work for the Favero Full-Arm scoring machine?
That particular machine appears to output the data serially at TTL logic
level.
Later,
Eric
Eric Schlaepfer
> Consider mounting the lights on the fencers' masks. This is one of the
> features of the wireless system currently being tested (well, recently
> tested and perhaps to be tried again). Why did they choose to put the
> lights on the masks? After all, they pass the same information through
> the strip to the normal scoring box. One can imagine several reasons for
> adding lights to the masks:
Remember that there are TWO wireless systems around. The one with the
mask light and the one with the connection through the strip. The one
with the mask light has been dropped after the FIE had too many problems
with it. The system based around the metallic strip is still under
development.
> Also - recall that these high tech 21st century centurions will be fencing
> with a plate glass bay window in front of their eyes.
Well, no. At a tournament in Hungary, five of these masks shattered
after being dropped. Several others exibited dangerous-looking cracks
around the rivets that fastened the plastic to the mask's frame. This
caused a furor in the French press--about five of the leading newspapers
in Paris printed highly negative articles condemning the new masks (one
was titled "Down with the masks!"). After that the FIE dropped the
transparent masks entirely.
> I mentioned the various recent rule changes to an acquaintance at work who
> both fenced a bit in his glory days and also was quite successful in SCA style
> bouts. He immediately picked up on the tactical possibilities inherent in
> being
> able to watch your opponent's eyes. (Perhaps the FIE rules committee should
> have a field trip to an SCA hosted tournament.) I rummaged around dejanews
> a bit but didn't find any discussion about this issue - but I'm sure others
> must
> have considered it.
Oh yes. But why not take the opportunity to wear those cool rainbow
sunglasses under your mask?
> But combine the new "watch the eyes" data channel with the new "I've got
> lights on my head that I can't see" feature. This opens up a whole new
> category of feints and misdirection.
It does with the current masks. My coach has a neat little trick where
he "points" his head at his opponent's target area like he is going for
an attack from sixth position, but he moves his blade into a low four,
nailing his confused opponent in the flank.
Later,
Eric
Eric Schlaepfer
> a) Insulate the mask. Then all you have to do is determine whether or not
> you have hit metal. I have previously explained how to do that.
There are a variety of techniques to do that. How about a method
similar to that used by time domain reflectometers?
> b) Use the method I described for discriminating between the foil guard and
> the lame to discriminate between the epee guard and the mask.
With the reflectometer approach that could be done. You would also have
to filter out false signals caused by touching the metal parts of your
equipment.
> > This might sound a little odd, but it would help if there were a circuit
> > which would maintain a point of the same electrical potential between
> > both modules using capacitive coupling.
>
> That won't really get you a reference. Everything is coupled to everything
> else and the capacitances are constantly changing. Fortunately, it isn't
> necessary.
That's correct. A very similar approach to mine was used for the new
personal area network (PAN) technology. Check out
http://www.research.ibm.com/journal/sj/mit/sectione/zimmerman.html. It
uses the human body as a wet wire to transmit and receive data. "The
return path is provided by the 'earth ground,' which includes all
conductors and dielectrics in the environment that are in close
proximity to the PAN devices."
My system would use the body cords, weapons, and lame instead of the
"wet wire" human body, but the 'earth ground' would still work the same
way. In fact, the electrical circuit could be very similar to the PAN
design.
I've actually put together a test circuit using the PAN method. It
doesn't filter noise right now, but I can light an LED by touching two
wires together. The two units are powered off batteries to avoid
coupling through the AC power lines. It is not very reliable--sometimes
the LED doesn't light, but I expect that to change once I implement some
sort of filtering circuitry coupled with a higher-impedance FET input
amplifier.
Later,
Eric
Zebee Johnstone
Rob Seaman <sea...@noao.edu> wrote:
>I mentioned the various recent rule changes to an acquaintance at work who
>both fenced a bit in his glory days and also was quite successful in SCA style
>bouts. He immediately picked up on the tactical possibilities inherent in
>being
>able to watch your opponent's eyes. (Perhaps the FIE rules committee should
>have a field trip to an SCA hosted tournament.) I rummaged around dejanews
>a bit but didn't find any discussion about this issue - but I'm sure others
>must
>have considered it.
Hrm.. I fought SCA heavy (full armour, big rattan clubs, fullforce and
speed) for some time, and never found "watching the eyes" all that
useful.
Sure, some people might look where they will hit, but only novices.
Any experienced fighter isn't point focusing, they are using the whole
field of vision. Using the bits that detect motion usually, and
seeing the whole body as target and letting holes "impinge" rather
than looking straight at them.
The whole "watch the eyes" thing seems to be more like fiction - it
makes a good thing in a thud and blunder novel, but in reality? I
suppose it might help a bit in a duel to see the face and work out if
the opponent knew what they were doing, but I dunno it's that much of
an advantage in a sporting competition (which is what SCa fencing is
too...)
Bsed on my experience fighting in armour with openface helms (face
protected by a grill with 1" square holes), "seeing the eyes" is
overrated.
Zebee
Rob Seaman
> Remember that there are TWO wireless systems around. The one with the
> mask light and the one with the connection through the strip. The one
> with the mask light has been dropped after the FIE had too many problems
> with it. The system based around the metallic strip is still under
> development.
I was under the impression that they were the same system. It's gratifying
if they reached the same conclusion about the mask mounted lights ;-)
If the system that communicates via the strip has no such lights, however,
it will be even harder to sell into environments where using a strip is
unwieldy or impossible.
> Well, no. At a tournament in Hungary, five of these masks shattered
> after being dropped. Several others exibited dangerous-looking cracks
> around the rivets that fastened the plastic to the mask's frame. This
> caused a furor in the French press--about five of the leading newspapers
> in Paris printed highly negative articles condemning the new masks (one
> was titled "Down with the masks!"). After that the FIE dropped the
> transparent masks entirely.
The FIE home page has some news reports that suggest otherwise. These
are uneven translations in some non-web friendly .doc format but it sounds
like they are merely delaying the inevitable:
"[...] In the following week, the F.I.E. conducted safety tests in
the presence of the President of the Commission of Athletes. In the
first test, it was proven (a finding not disputed by the President
of the Commission of Athletes) that the mask with transparent
(lexan) visor was safer than the meshed mask when the lexan mask
and the meshed mask underwent penetration tests, it was proved that
the mesh could be pierced whereas the lexan could not.
[...] The President of the F.I.E. consulted the President of the
I.O.C., Mr. SAMARANCH , who confirmed that television for the Olympic
Games considered it essential to [...] The F.I.E. must consider the
future of its sport, and we are convinced that it is close to the
fencers hearts to ensure fencings survival in the Olympic movement.
[...] The Executive Committee was therefore consulted about deferring
this date. As of today, 9 answers out of 12 have reached the FIE office,
and all propose deferring the obligation from March 1 to April 1.
This is taking into account what occurred in Budapest [...]"
Eric Schlaepfer
> If the system that communicates via the strip has no such lights, however,
> it will be even harder to sell into environments where using a strip is
> unwieldy or impossible.
Yes, the system has no such lights. It's really going to be hard to
sell to anyone other than the FIE.
> > Well, no. At a tournament in Hungary, five of these masks shattered
> > after being dropped. Several others exibited dangerous-looking cracks
> > around the rivets that fastened the plastic to the mask's frame. This
> > caused a furor in the French press--about five of the leading newspapers
> > in Paris printed highly negative articles condemning the new masks (one
> > was titled "Down with the masks!"). After that the FIE dropped the
> > transparent masks entirely.
>
> The FIE home page has some news reports that suggest otherwise. These
> are uneven translations in some non-web friendly .doc format but it sounds
> like they are merely delaying the inevitable:
I read those articles too--it looks like FIE damage control to me.
After what was said in the French press, I really doubt that the FIE
will try to use those masks. Allstar has even gone as far as
withdrawing the masks from their product line.
Later,
Eric
John Hasler
> The communication from the fencers' units to the central scoring box is a
> potential weak point--somebody could prevent ANY radio based system from
> working by using a signal jammer.
It is easy to design a system such that even successful jamming would
accomplish nothing that shutting off the lights in the gym wouldn't do.
> You could even embed a large force-sensing resistor array into the
> uniform which would send a signal to the fencer's scoring unit, which
> would decide if the touch was in a valid location.
There is force sensing film available. Replace the metal lame made of
that.
John Hasler
> Just pick a standard so basic and widespread that you will be able to
> count on supporting it (cheaply) for several decades.
There are several brands of spread-spectrum trasceiver module on the
market. It won't be long before the manufacturers standardize them for
interoperability, as it is in their interest to do so.
> There may be enough geopolitical pressure to ensure that it is used
> through to Sydney.
Where it may fail miserably and produce an anti-wireless backlash.
--
John Hasler
jo...@dhh.gt.org (John Hasler)
Dancing Horse Hill
Elmwood, WI
John Hasler
> Gotta go digital.
Of course.
> Run the internet to each fencer. (Why not?) TCP/IP has checksum
> protection in each packet.
We certainly want to send packets with MD5sums, sequence numbers, and
timestamps. but I don't think IP is the best protocol. We don't need store
and forward adressing, fragmentation/defragmentation, or variable length
packets. Off the top of my head I'd say start with HDLC and add features.
There may be an existing protocol out there that would serve.
> I've retraced a number of the earlier rec.sport.fencing threads via
> dejanews and there appears to be an implicit assumption that whatever
> system is fielded it will consist of proprietary technology.
This appears to be the assumption of the FIE and friends, but it certainly
is not mine.
> There is no problem with this as far as a particular scoring box or
> sensor or whatever, but the system design has to reside with the
> international standards organization for the community.
The protocols and standards must be freely available and not encumbered by
copyrights and patents. There should be a public domain reference design.
We should look to free software for a model.
> But compatible how?
Interoperable.
> does anybody know if it's digital or analog?)
All electronics is analog at base.
John Hasler
> The current electrical scoring system was created specifically because
> enough people were screwed by incompetence to demand a change.
I fenced both dry and electric. I prefer electric. You think it's bad
fencing with a blind referee? Try a blind referee, two blind judges, and
two that are hallucinating.
John Hasler
> Try imagining the cost of flying in an extra 4 people per strip to serve
> as judges, putting them up in hotel rooms, and paying their per-diem and
> meals.
That was, of course, not what was done. Instead, whoever was handy was
wheedled into judging. You've een eliminated from foil but are hanging
around waiting for epee? Why not kill some time judging sabre? Never
fenced sabre? Well, you've watched it, haven't you? We'll tell you the
rules!
John Hasler
> I'd love to see some kind of directional force sensor reintroduced to
> sabre once an affordable, reliable, and calibratable option exists
Strain guages should do just fine. They also could replace the switches in
foils and epees.
John Hasler
> As someone with a bit of knowledge about user interfaces, my first
> reaction to the rule change flipping the meaning of the lights was to
> snort out loud.
As someone who fenced for twenty years, so was mine. That's also my second
reaction.
> Do the foil and epee points really need to be mechanical switches?
I have previously proposed a system that replaces the switches with strain
guages.
> How about using an accelerometer to detect a touch in "velocity space"?
Detecting touches is easy. The trick is figuring out what you hit.
> And does anybody really believe that grounding the fencers through the
> soles of their shoes will work reliably?
I don't believe that it would be reliable for detecting touches, and there
is no other reason to do it.
> What about some sort of fiber optic sensor?
There may be a possibility there.
> If the "victim" is supposed to light up like a Christmas tree using the
> new wireless fencing system - why?
Because the FIE thinks it will placate the IOC.
> A successful (and not exorbitantly expensive) wireless scoring system
> would sell itself to every club in the US (and elsewhere) for everyday
> use.
A good point. The FIE is certainly not going to introduce a scoring system
not developed by an established European company no matter how good it is,
but that doesn't mean there would be no market for it. Such system could
be introduced at the grass roots level and eventually drive out whatever
kludge the FIE comes up with.
David W. Neevel
That's the impression I have, too. I was talking with Ron Herman about
the whole mask thing this past weekend, and what I got from him was that
there's a growing consensus among the SEMI folks that the transparent
masks aren't ready for prime time, but that the FIE higher-ups (Roch,
et. al) aren't yet ready to accept that since they've invested so much
effort into trying to keep Samaranch happy.
-Dave
The Irreverent Eddie
your bell
> guard, kid....
Who needs a trigger? I'm hooked up to that extension cord..... ;-) Parry
this!
--Eddie