Timing

General slalom chatter...rant about the bad, rave about the good
Mike Mitchell
Posts: 180
Joined: Tue Aug 05, 2014 10:44 pm

Re: Timing

Post by Mike Mitchell » Sun Nov 18, 2018 9:33 pm

I think the best solution is to have the same Start/Finisher for the Class and possible both runs.
With possibly a 2 sec penalty for paddlers turning up out of Class.

If you sit with a stopwatch and attempt to stop it on a given second, most people are + or - 0.20 seconds on average.
So a finish time of 100.05 secs could be 100.03 or 100.07. Then the Starter could also be out by the same amount giving + or - 0.40 Overall. So rounding up to the nearest second would give 100 secs or 101 sec.
So I don't believe rounding up to the nearest second is the solution.

It would be interesting to compare Tutti timing with back up timing, as the tutti,s accuracy could be questionable.

Beam Timeing. When this first came out we didi test against stop watches. The results were also + or- 0.20 secs. They are probably more accurate these Days, but has anyone ever checked them?

JimW
Posts: 570
Joined: Thu Oct 08, 2015 2:17 pm
Location: Pinkston

Re: Timing

Post by JimW » Mon Nov 19, 2018 3:38 pm

Checked the beams for what Mike?

They are checked every time they are used to make sure the beam is hitting the reflector properly so that they actually work.

In terms of speed of operation, the light beam travels at the speed of light, which is also the speed of electromagnetic propogation so the electric pulse effectively travels up the wires at exactly the same speed. The delay between the beam and the tutti is far too small to measure with a stopwatch, if you wanted to calibrate the beams you would need to use something with similar reaction speed.

The main issues with beams is whether or not the wind moves the start/finish gate making the course a few cm longer or shorter for some paddlers (same issue for manual timing), other issues like whether the paddler leans back or forward to break the beam later or earlier or if the paddler can get a paddle blade out front to break the finish beam early affect everyone the same - if you can get an advantage, you should practise starts and finishes to capitalise on it!

Potential electrical issues like cross talk are mostly ruled out because we don't have power and comms cables running parallel, even where cables do run parallel they are normally well spaced and not bunched except going in and out of control. Internally damaged cables will be seen as intermittant problems, I'm sure if these were experienced the timing team would be able to run a continuity check? We have a fairly basic tool at work for testing ethernet cables which can determine from the time it takes to return a signal from a transducer on the other end of the line how long each conductor in the cable is so it is easy to identify an internal break, and even takes some of the guesswork out of locating it to repair.

There will also be a small delay whilst Tutti registers the change of state of the beam, - I have no idea what sort of internal clock speed Tutti has but I would imagine this is also much less than 1/100th of a second. Computer chips have been timed in MHz or faster for a long time now - 1 MHz = 1 million cycles per second so even an old 16 MHz i286 processor would be capable of registering the change in beam in maybe 1/8millionth of a second (assuming it takes 2 cycles).

Don't get me wrong, beams are not instant and not fool proof or impervious to damage, but they are fast enough that if you are calibrating against a stopwatch, all you are doing is confirming how accurately you can operate the stopwatch...

By the way, Ace photo of Cardiff at sunset!

BaldockBabe
Posts: 307
Joined: Tue Mar 06, 2012 8:55 am

Re: Timing

Post by BaldockBabe » Tue Nov 20, 2018 1:06 pm

The way I look at it is if the beams and kit the timing team use are good enough for the Olympics they are good enough for us!!! :D

djberriman
Posts: 806
Joined: Tue Apr 03, 2007 8:19 pm

Re: Timing

Post by djberriman » Tue Nov 20, 2018 1:56 pm

"It would be interesting to compare Tutti timing with back up timing, as the tutti,s accuracy could be questionable."

This is or should be done at every race, we generally try and check 1 in 10, to do this we put a post it note on every 10th card and this reminds the tutti people to get the backup times which are then entered into a spreadsheet by the Simply Slalom operator to produce a check as we go. It is more to ensure the backup times are reliable rather than the other way round though.

I'd think the backup times are more questionable than the tutti times personally. Two stop watches of a cheap nature, coordinated by hand as best as possible, pressed at the same time (hopefully) as the tutti. Often dropped or the wrong button inadvertently pressed such that they have to be re-synched (you only know this if you do the regular checks) and known to pack up due to batteries, wear and tear or water ingress.

I'd guess most tests are done sat at a desk and not on a river bank in the rain whilst trying to press a tutti button, a stop watch, balancing your paperwork on your lap and do all the other stuff starters/finishers do at the same time for hours on end.

Steve Agar
Posts: 56
Joined: Wed Jun 17, 2015 9:30 am

Re: Timing

Post by Steve Agar » Wed Nov 21, 2018 9:09 am

I'm inclined to believe that inaccuracies in Tutti are more likely to be human errors than anything else (we still rely on people manually pressing buttons so it's no different to starting and stopping a stopwatch) and as Duncan rightly points out, co-ordinating two hands whilst trying to keep the rain out of your eyes, focus on an imaginary finish line and read a bib number is never going to be the most conducive way of achieving the highest degree of accuracy. It's actually more a matter of what resolution we consider appropriate (whole seconds, tenths or hundredths) and how we can maintain that consistently (along with consistent judging).

JimW
Posts: 570
Joined: Thu Oct 08, 2015 2:17 pm
Location: Pinkston

Re: Timing

Post by JimW » Wed Nov 21, 2018 4:48 pm

BaldockBabe wrote:
Tue Nov 20, 2018 1:06 pm
The way I look at it is if the beams and kit the timing team use are good enough for the Olympics they are good enough for us!!! :D
I'm not a fan of blind faith in technology, I think it is better for more people to understand how things work and question why and how they work.

Mike's questions were quite valid and I have probably missed an important part of the explanation in my answer.

The beam is nothing more than a switch. A beam of light is projected from an emitter to the reflector which reflects it back to a receiver or sensor. The sensor is probably something like a photo diode, something that changes its conductivity depending on whether light falls on it or not. I suspect the sensors used in the timing beams make a circuit (conduct electricity) when light falls on them, and break it (don't conduct) when there is no light. A paddler passing between the reflector and either the emitter or receiver (but in fact both at the same time) will temporarily interrupt the beam turning the switch off, once they pass and the beam hits the sensor again the switch is turned back on. That is all that happens at the start and finish gate when a paddler passes through the beam, the paddler is turning a switch off and back on again instead of the starter or finisher pressing a button (in fact start and finish do press a button to get a backup time in case the beam was not broken or there was some other electrical issue).

The timing part happens at the other end of the wire in control - the beams are plugged into tutti in a similar way to the push buttons. Tutti records the start and finish times, all the beams are doing is providing the pulse (on - off - on) which tell tutti to record the time now in the same way that the push buttons for start and finish do. The difference is that the beam does not rely on the reaction time of a person who has to make a judgement of when part of the paddlers body crosses the start or finish line, the beam creates an invisible line (of light) which is broken by the paddler eliminating the human reaction element which we know can be variable.

So checking the beams is really about making sure the switch is working properly, to check the timing accuracy you need to be looking at tutti and the backup time checks, which are the same whether you are using beams or not. Except as Duncan points out that the controller for the beams is not tutti, and the backup times are recorded differently so it isn't exactly the same, but the process would be similar

Does that all make sense to anyone who didn't previously understand the beams?
Last edited by JimW on Thu Nov 22, 2018 2:16 pm, edited 1 time in total.

djberriman
Posts: 806
Joined: Tue Apr 03, 2007 8:19 pm

Re: Timing

Post by djberriman » Thu Nov 22, 2018 12:57 pm

tutti isn't used with beams as far as I am aware. Totally different system at Prem/Div 1's where beams are used. I'm pretty sure its not just the equivalent of a simple button push either but could well be wrong.

Beam timing systems also aren't just on/off, they detect the length of the break and can differentiate between say a paddle and a body.

Dual beams (as we use) make them more accurate too.

I believe we use the ALGE system details of which can be had here

https://alge-timing.com/alge/download/b ... lets-E.pdf

JimW
Posts: 570
Joined: Thu Oct 08, 2015 2:17 pm
Location: Pinkston

Re: Timing

Post by JimW » Thu Nov 22, 2018 2:09 pm

djberriman wrote:
Thu Nov 22, 2018 12:57 pm
tutti isn't used with beams as far as I am aware. Totally different system at Prem/Div 1's where beams are used. I'm pretty sure its not just the equivalent of a simple button push either but could well be wrong.

Beam timing systems also aren't just on/off, they detect the length of the break and can differentiate between say a paddle and a body.

Dual beams (as we use) make them more accurate too.

I believe we use the ALGE system details of which can be had here

https://alge-timing.com/alge/download/b ... lets-E.pdf
That is a fair point, most paddlers don't even know what tutti is never mind know or care that a different controller is used for Prem and div 1.
You are correct that push buttons used in prem and div 1 are more than push buttons, they replace the stopwatches used in lower divisions with a device that can log the times and I beleive takes its time synchronisation from GPS - which has to be done every time you turn them on and presumably provides continual checking of the timing accuracy (I'm not sure if they continue to track GPS satellites or just check when booting).

In terms of the beams detecting the length of break, I would doubt if that is handled at the beam itself (it could be) I would think it is handled by the control system (which isn't tutti) since the beam units are relatively vulnerable. I'm fairly sure all that is actually mounted on the start/finish gate is essentially a couple of automatic switches, and all the clever stuff is done in the controller, but I may be wrong?

Post Reply