Problems with the new timings on some boxes and not others?

[DanInMI]

Has anyone noticed a difference with the new timings from one brand box to another?

Last fall I fenced on the new timings at a local club in MI and had no issues with the new timings. Then I went to a tourny in Chicago and did experience the kind of problems that have been described here (direct hits not registering, etc.) It was really my first experience with the new timings so I wrote it off.

But back at that same club in MI I NEVER have that problem. My brother thinks that he has experienced the same phenomenon. Could it be the equipment? Could there be a difference in the parameters in which the boxes are operating? Has anyone else noticed this?

I tell you, I am not kidding, on the very new Favero boxes that I fence on at that club the debounce is not an issue at all. Perhaps it is an issue on some older boxes that were upgraded.

Perhaps that should be investigated before throwing the baby out with the bathwater.

[DHCJr]

Semen Rikhtman, the Russian SEMI member, did a test on 4 different machines testing whether contact with a blade would affect the on-target lights. It was determined on the 2 machines, All-Star and Visti that if there was blade contact after the hit to the lame, but before 15ms of sustained time on the lame and the tip being depressed, the clock would start over, sometimes several times.

Right after this happened and it was reported that 2 brands that did not have the same flaw, St. George and Leon Paul. Since then, there was a request to have some machines upgraded for a competition. This was not done in time and in fact several months. It is suspected, but not proved, that All-Star held off sending new chips, until they found the bug in the program. No tests have been done by Rikhtman recently.

Machines are no longer circuits, they are programs. There can be bugs in programs, as there can be flaws in circuitry. The problem is, it can be harder to find the problem. In fact, many boxes don't even send a DC current down through the cords, reels and weapons. They take samples, like a digital meter. In fact, when SEMI did the tests, the cut-off point suggested by SEMI was 10ms, not 15ms. It did not get rid of all flicks, but it allowed almost all straight attacks.

[oso97]

At our Sectional Circuit competitions we use Favero boxes almost exclusively. No particular reason, other than the fact that they're inexpensive, there is a vendor in the area who carries them and they are easily upgradable by the user.

But anyway, I've barely seen the kinds of straight attacks not scoring problem on them that I've seen rampant at national events using the SG machines. Curiouser and Curiouser.

[telkanuru]

Could this be attributed to different coaches/styles that develop in small regions around clubs?

[oso97]

Quite possibly. Although we're talking about a region stretching across several states, with diverse coaching styles represented, with a number of highly ranked junior competitors. Not exactly a small, inbred population.

[DanInMI]

Originally Posted by DHCJr

Semen Rikhtman, the Russian SEMI member, did a test on 4 different machines testing whether contact with a blade would affect the on-target lights. It was determined on the 2 machines, All-Star and Visti that if there was blade contact after the hit to the lame, but before 15ms of sustained time on the lame and the tip being depressed, the clock would start over, sometimes several times.

Right after this happened and it was reported that 2 brands that did not have the same flaw, St. George and Leon Paul. Since then, there was a request to have some machines upgraded for a competition. This was not done in time and in fact several months. It is suspected, but not proved, that All-Star held off sending new chips, until they found the bug in the program. No tests have been done by Rikhtman recently.

Machines are no longer circuits, they are programs. There can be bugs in programs, as there can be flaws in circuitry. The problem is, it can be harder to find the problem. In fact, many boxes don't even send a DC current down through the cords, reels and weapons. They take samples, like a digital meter. In fact, when SEMI did the tests, the cut-off point suggested by SEMI was 10ms, not 15ms. It did not get rid of all flicks, but it allowed almost all straight attacks.

the fact that each manufacturer is developing thier own code might suggest that there may indeed be differences between how well the different boxes register valid hits.

The other current thread about the favero boxes also suggests that there might be some considerable differences

[Genjuro]

It seems that the manufacturers didn't have a clear specification on what the timings should be..., and combined with a few bugs, well problems are bound to happen.

Considering this, and the number of complaints I've heard it's actually quite strange that the FIE rushed so fast into this without a proper test (a few tournaments of a couple of catagorys, and not a whole season) with a few prototypes from the mainstream manufacturers, just to see if there are any bugs/problems/wrong specs.

Of course, one can say that fencers should adapt, like people did when foil became electric, but on the other side, the capteur in sabre also didn't last to long...

[Fernando Diaz]

Hi, well I am actually the SEMI responsible of the Spanish Fed and we have our own Research and Development program which have lead to our own scoring machines (called ATLAS, simply because we like the name)

Our technicians and those from Leon Paul have been revising the possibilities for the non-registering hits against hard surfaces and we think that we have found a reasonable reason for them.

Lets describe the facts.

We test the new times from 2000 till 2004 with the ATLAS and I have not detected the problem with correctly taped foils (even with 25 ms contact times). (Yes for those that complain I am one of the fathers of the new machine)

We test the machines from July 2004 in my fencing salle (150 fencers, 50% foil, 50% Epee).

As tested before the non-registering problem is neglectible.

On January the Spanish Federation change the tournements to the new times.

The first tournement was fenced with Allstar machines and my fencers report many failures.

The same applies to the British tournements as I got from my conversations with Dr. Halsted from the BFA.

(My personal impresion and this from Mr. Paul (head of LP)is that the failures are ocassional and mainly caused by not properly executed actions we agree that if it is solved it will help the aceptation of the new machine.
I personally think that the reason for the complains is that the fencers specially the young ones that grow up with the game of run and flick find that their actions do not touch and they find reasons to complain.)

So we have been revising the differences in software, aparently the LP and the Atlas machines behave similarly except for the fact that the non registering problem is more frecuent in the LP (and Allstar and Prieur the ones I tested)).

We found that the only difference is in the way that we have interpreted the FIE instructions concerning the new timing

We found that to register a hit a LP machine requires (the same thing probably applies to Allstar).

15 ms open circuit time AND 15 ms point contact with the lame.

The problem is that if the circuit is open and the contact with the lame is interrupted the time resets and no hit is awarded.

While the Atlas machine requires

15 ms open circuit time AND point contact (for 5 ms as previously required) with the lame at any time.

Which one of the interpretations is the correct one is out of the discussion at this point but we are observing a different behaviour. It might be that the LP implementation is more conform with the rules (although no specific instruction concerning the time of contact between the lame and the point was given) but it might be the cause of the problem.

I can immagine that on a hard surface, with the lame nicely streched, such a scenario can easily happen. Point remains pressed down, but by moving over the surface, the point sometimes touches a wire, and sometimes it doesn't. This is especially possible, if the tip hits sideways.

Of course it confused me in the begenning when I observed that the problem disappears in the absence of the electric jacket. So I though it was a problem of conduction through the lame.

I hope this is of interest for you

[PalmFrond]

Thank you, Mr. Diaz - very interesting. Personally, I haven't experienced any problems with the new timings, probably because the boxes used in Ireland are pretty much the same (the IAFF are the only ones to supply the new timings at competitions). Your new box sounds like it might solve a lot of the problems people have been complaining about.

[DHCJr]

Originally Posted by Fernando Diaz

Hi, well I am actually the SEMI responsible of the Spanish Fed and we have our own Research and Development program which have lead to our own scoring machines (called ATLAS, simply because we like the name)

I hope this is of interest for you

Mr. Diaz, this was a very good information. I understand English is a second language. Your first paragraph may confuse some. It was confusing to me the first time I read it. I will try to explain further. You are not claiming that you are on the FIE SEMI committee, but you are the person, who is responsible for the equivalent function for the Spanish Federation.

Thank you for joining and the explanation.

[keith]

Originally Posted by Fernando Diaz

15 ms open circuit time AND 5 ms point contact with the lame.

That sounds like a good solution to the current problems.

[cowpaste]

Thank you very much Mr. Diaz.

This is quite interesting. An open-circuit time that is longer than lamé contact time would mostly be due to the time it takes for foil tip spring to close the circuit again, correct? In that case, it is the result of something that a fencer cannot directly control (tip travel distance, spring stiffness, barrel friction). I know the FIE specifications for the new timings were quite unclear, but I do not believe the ATLAS method of registering hits is quite exactly what the FIE intended. Then again, I think this is good because it effectively lowers the required tip-depression time, and all I care about is hit registration consistency.

[ChubbyHubby]

Originally Posted by cowpaste

Thank you very much Mr. Diaz.

This is quite interesting. An open-circuit time that is longer than lamé contact time would mostly be due to the time it takes for foil tip spring to close the circuit again, correct? In that case, it is the result of something that a fencer cannot directly control (tip travel distance, spring stiffness, barrel friction). I know the FIE specifications for the new timings were quite unclear, but I do not believe the ATLAS method of registering hits is quite exactly what the FIE intended. Then again, I think this is good because it effectively lowers the required tip-depression time, and all I care about is hit registration consistency.

If 15ms/5ms is used. Why bother with the contact time at all? Even the with the current 15ms/15ms flicks register okay (with some randomness). If it becomes consistent, the contact time will make next to no difference other than the most extreme "chop through your parry, wrap around your head" type of flicks.

Make sure you have a slightly sticky tip and flick away.

[shlepzig]

There was a thread some time ago which revealed that different manufactureres had implemented the off target timings differently.

Barry Paul had infused some very good information into that conversation.

Was there any resolution to that issue and is that issue in any way related to the other timing issues?

Shlep.

[Jarentz]

Originally Posted by DanInMI

Has anyone noticed a difference with the new timings from one brand box to another?

Last fall I fenced on the new timings at a local club in MI and had no issues with the new timings. Then I went to a tourny in Chicago and did experience the kind of problems that have been described here (direct hits not registering, etc.) It was really my first experience with the new timings so I wrote it off.

But back at that same club in MI I NEVER have that problem. My brother thinks that he has experienced the same phenomenon. Could it be the equipment? Could there be a difference in the parameters in which the boxes are operating? Has anyone else noticed this?

I tell you, I am not kidding, on the very new Favero boxes that I fence on at that club the debounce is not an issue at all. Perhaps it is an issue on some older boxes that were upgraded.

Perhaps that should be investigated before throwing the baby out with the bathwater.

My experience exactly. We use Favero boxes at the club and I hardly ever see the problem of straight attacks failing, even among the faster fencers. I have only fenced one tournament with the new timing and only experienced that problem fencing lefties with chest protectors.

[DanInMI]

Originally Posted by Jarentz

My experience exactly. We use Favero boxes at the club and I hardly ever see the problem of straight attacks failing, even among the faster fencers. I have only fenced one tournament with the new timing and only experienced that problem fencing lefties with chest protectors.

So...it seems to me that the problem with valid hits not registering may not be the actually timings themselves, but rather the way that some manufacturers are implementing them.

If we were able to solve the problem of "valid hits not registering" would the same crowd still be complaining that the timings suck?

[DHCJr]

Originally Posted by shlepzig

There was a thread some time ago which revealed that different manufactureres had implemented the off target timings differently.

Barry Paul had infused some very good information into that conversation.

Was there any resolution to that issue and is that issue in any way related to the other timing issues?

Shlep.

Yes and No. The Executive Committee (Rene Roch) issued a immediate decision stating to go with changing the off-target to match the on-target. One of the items that the German President stated was these immediate dicisions were not legal, without the consent of Congress. Nothing, I have heard has gone beyond that.

The off-target has little to do with the problems of the on-target, chest protector and insulated blade.

[Inquartata]

It's not just a foil thing, either. We have 5 Fevero machines. The Favero FA-5s display "Jun" when on the new sabre timing, but there are two settings, 1 and 2, each of which gives different results. Meanwhile the FA-1s do not have the same lockout time as EITHER setting on the FA-5s; we are often unsure they are on the new timing at all, in spite of having had their chips replaced. Moreover, some Favero 1s have a blinking red light to indicate that the new timing is in place, and others, like ours, do not...

Act in haste, repent at leisure, Mr. Roch...

[DanInMI]

Well this is all very interesting. It was confusing to me before that so many people seemed to be having a totally different experience with the new timings than i was having. I was assuming that it was just sour grapes from the "pro marching-flick" paople, but it would appear that there is truly a problem. It would appear, however, that the problem is not with the timings themselves, but with the way that different manufacturers are implementing them. Therefore it seems that the FIE should NOT move to change the timings again, they should get the manufacturers all on the same page so that they are all implementing them in a way that touches that remain on target, with the tip depressed, for the entire duration of the debounce time register as a valid touch regardless of the manufacturer of the scoring box.

THEN if it is deemed that the timings need to be adjusted, so be it.

[keith]

Originally Posted by DanInMI

................they should get the manufacturers all on the same page so that they are all implementing them in a way that touches that remain on target, with the tip depressed, for the entire duration of the debounce time register as a valid touch regardless of the manufacturer of the scoring box.

actually this might be the worst posssible starting point for a standardised sampling/filtering system. Unless of course your aim is to make the demands for a return to the old timings even louder.