maptalk.co.nz Forum   |   Links    

  Forum

Forum Home   Start New Topic   Edit Profile   Register  

1   2   3   4   5   6   7   8   9  

Sport Ident Organisational Matters

Show Profile  mcroxford Posted: 9 June 2018, 2:52 AM  
I experienced three controls that weren't activated. I'm pretty sure that they were controls 9(122), 11(129) and 12(132) on the Course 3. I note that M70 did not share any controls except the last two so there must have been at least four controls that weren't activated. The problem for me was on Day 1 AM only, but I didn't have these same three controls on any of the other days. I advised the course planner but not the controller. My error.

I think a testament to the quality of the Planning and Controlling was that there were very few mispunches over all grades in the AM race. The only grade that sticks out is M50A where four mispunches occurred at the same controls. Only one of those competitors had a SIAC though and the grade had a high level of adoption of SIACs with 15 of the 40 competitors.

I think that there are a number of lessons learnt.

One is the problem of setting all the units to contactless and the quality control process required, especially with 300 odd controls. Perhaps the technical committee could write or endorse guidance for preparation for SIAC. My view is that all clubs should just change their units to SIAC for all events. Yes it uses more batteries but SIAC is so much fun!!!

Another is regarding health & safety and 'fairness' in that the SIAC unit does not register in the control box. My Dad who worked in electronics for decades says all you need is a little beacon in the SIAC unit so that it talks back to the control.

The next is an interim adoption of the IOF rule into our ONZ rules with a minor alteration to the wording of our rules to remove reference to interrogating the SI box (inline with NZ Controllers interpretation of the IOF rule).

We should advise IOF that we perceive there is a potential issue with the rule set not being reflective of contactless timing issues and ask for guidance.

Lastly, we should advise Sportident that we consider they should look at modifying the SIAC to allow it to talk to the control box for safety reasons.

Lastly, we should publish at the start a reminder to all SIAC users to check that their unit beeps and flashes at every control.

Any others?


Show Profile  mcroxford Posted: 9 June 2018, 2:54 AM  
Too many lastlys...

Show Profile  Michael Posted: 9 June 2018, 7:51 AM  
If I may divert a bit to the finish chute times on Day 3; Isaac Egan, Will Tidswell and Joe Lynch all recording 8sec. If the control descriptions are correct that's equivalent to an 11sec 100m which would be pretty respectable on the track. OK, it was downhill a bit.

Of course the 70m is possibly not accurate, but my interest was sparked by my own (relative) speed on Day 3, which was 73% faster than Day 2. If everything was in the same position, I'm trying to recall exactly what I had to eat on Sunday night, because I want some more of it. Mind you others may have been eating that too, because I didn't actually find anyone who didn't have a better time on Day 3. Only looked at a smattering of course.

Was the finish chute exactly the same or was it just "similar"? I may not have noticed thru the sweat and the elation of out-navigating my opposition:-))

Show Profile  camd Posted: 11 June 2018, 2:20 AM  
Hey guys - sorry about the SI boxes not being in beacon mode at QB - that was my fault. I had a look at the 3 boxes afterwards and they were still programmed as normal controls. There is two ways I think this might have happened - either the boxes were missed before the event or (I think more likely) when programming them the connection was poor and the new settings weren't transferred (which I've noticed happens occasionally despite the box beeping). With hundreds of boxes, there isn't an easy way of checking these were in the right mode - I caught one or two but obviosuly a few slipped through. You'd have to either check all the screens, re-read the settings of the boxes or use a SIAC on them to be sure, all of which take some time (especially when you are studying for exams!). Having said that if all clubs moved to mixed punching permanently there wouldn't be an issue on this point at least...

Regarding the finish chute times on day 3, (all days actually) that would be a product of clock drift. The times on the finish boxes and start box were resynced every day (as they should be when using multiple boxes) to ensure fairness. Thus between day 2 and day 3, you probably would have seen a combination of 2 boxes on the last control with slightly different times (being radio control boxes, programmed and put out on the day) and the finish box time compared to these also being slightly different each day.

Finally, on the point regarding SIAC's not storing data in the control box. SIACs actually already have a beacon in them for transmitting data - that is how our radio controls work in Auckland. Any control box can be set to tell the SIAC to broadcast it's last/all records upon punching in AIR mode. The problem is actually the control boxes themselves - they aren't capable of receiving this data. What sportident need to do is release a sportident box which both receives and stores this data - I'm surprised it wasn't included in the AIR specific control boxes they released a few years ago (at least I assume they can't do this).

Show Profile  mcroxford Posted: 11 June 2018, 6:25 AM  
No need to apologise, Cameron! It's awesome that NWOC offered it and I hope that more clubs follow suit.

I strongly believe ONZ should follow the Swiss Orienteering Federation and just decree that all SI events must be SIAC compatible.

Show Profile  Michael Posted: 11 June 2018, 7:45 AM  
Thanks for factual information Cam. So on the finish chute time we can be confident in the finish unit clocks relative to the start and each other; but less so for other units (including the last control)? How big can this clock drift get? I'm asking on behalf of less experienced SI people elsewhere, who I've mentally criticised for not syncing the boxes. How often do they need to be sync'd to ensure (say) 2sec accuracy?

Show Profile  Martin Posted: 12 June 2018, 12:07 PM  
IOF has a paper that quotes sportident drift as 0.86s/day; it would be the variability leading to relative drift between controls that's important though.
The procedure at Oceania/WMOC was to set start clocks from the master SI station and then sync last control and finish from this.

Show Profile  owen Posted: 13 June 2018, 11:17 AM  
Keeping all clocks synchronised with themselves and set to a standard time is a little harder than you might hope. While a cheap digital watch is very accurate, computers are software and tend to drift (sometimes a lot) and need to be continually nudged to keep their time accurate. What I've seen is that Linux computers can keep very accurate time (low milliseconds difference) however, while I haven't tested Windows for quite some time, I always found Windows to be a lot less accurate than Linux. In both cases however if you're offline, you won't have access to the time service (NTP) that is required to sync up the time. Also, if the clock drifts 'too far', it will require a manual time adjustment on the computer to force it to jump to the correct time (normally it adjusts by small increments until it is within range however won't if the gap gets too large).

I'm a little surprised to hear that the SI boxes drift so much, I had naively thought that they would be a hardware device and have a reasonably accurate clock. Apparently not. The quoted SI drift is more than I have measured on the Samsung tablet that our club uses for a start clock (and I was worried that it drifted by ~1s over 48 hours!)

At QB, the start clocks were synchronised each day with the punching start control which in turn was set off the master computer (as were the finish controls). So that meant that everyone's elapsed times were accurate however the first and last splits might be up/down relative to the rest of the course depending on which way the master computer clock had moved.

One thing I did notice when setting the start clocks though was that there was >5s time difference each day => the time on the master computer changed (significantly) overnight.So Michael, that will at least partially account for your finish split dropping from 26s to 15s. Looking at the day3 results for your grade it could also be that the time on one of the last control boxes had drifted (fast) a lot more than the other and perhaps only a few of you punched that particular box?

Show Profile  Michael Posted: 15 June 2018, 1:05 PM  
Thanks Owen. I appreciate the background that you and Cam have given us. It's interesting as a participant, and I would think important to know as a controller or club operating SI.


This message was edited by Michael on 17 July 2018, 1:09 PM

Show Profile  Michael Posted: 17 July 2018, 6:45 AM  
Understand that an event is asking competitors with flashing SI cards (includes SIAC) to set them to a low number of flashes. (You can't punch another control while it is still flashing). But competitors probably don't have SI-Config+ on their computers, or a download box around the house. You would have to provide a service at the event to do this - and to reset them again afterwards to the default (I think its 3sec).

Would seem to be better to ensure no controls (incl the finish) are closer than 3sec at Usain Bolt speed. Hidden snag - air cards can be set off by a control that's not FOR YOU, and possibly hold up punching at a control that IS FOR YOU. Note I'm not an SI expert, others please correct or add further advice. Someone good at maths please turn 3sec into metres:-))

Show Profile  AlanHorn Posted: 17 July 2018, 9:41 AM  
He can run at 44.72km/h so the answer is 37.27metres



Show Profile  mcroxford Posted: 17 July 2018, 11:36 AM  
The SIAC will only read unintended controls if they are set to beacon mode. For the schools champs this won't be the case as they aren't using contactless punching. The default is 3 seconds and i imagine shouldn't be an issue if all the controls are calibrated just prior to the first event, the sprint, where the risk is the greatest.

Show Profile  Martin Posted: 22 July 2018, 8:06 AM  
Hindsight is great, there were 4 people with 5s splits so giving the warning (and ability to change settings at the event) was warranted. Another dozen had 6s splits

Sportident recommends a long setting for MTBO and Ski-O and short setting for microfoot.

3s is covering 20m at a 1min/400m pace; controls are able to be 15m apart in a sprint on different features.

It doesn't matter whether boxes are synched/calibrated - it's a setting in the SI-card.


Show Profile  Michael Posted: 1 August 2018, 4:03 AM  
Last Oct Cam referred to Attackpoint thread http://www.attackpoint.org/discussionthread.jsp/message_1257320 A rare but verifiable fault that causes the box to flash and beep but no record is written to the stick. Here's the latest posting on that thread.

"For this bug O-Ringen had this year 200 SEK protest fee, those who liked to get forest unit to be examined had to pay 200 SEK. If missing punch was found there the fee was returned. Apparently this fee was to reduce amount of checking - less unnecessary inspections. Based on post in orienterare nu there was several overwrite bug cases every day, mark missing, protest made, punch found from forest unit and fee returned. Most likely there still is people who are not aware of this or for other reasons bothered not to make protest, so all bug cases hardly were found."

Show Profile  mcroxford Posted: 2 August 2018, 12:27 PM  
Doesn't matter here in NZ, Michael as general consensus of other Controllers on this thread is that they would never check the 'forest unit'. Ignore the problem- problem solved! Spray and walk away!


1   2   3   4   5   6   7   8   9  


Ruffneck Productions © Ruffneck Productions maptalk.co.nz