Hi
I have worked out most - but not all - the algorithms right down to Div 4. The worst ones are the C2 and C1W because they are dependent on K1M all the time. They thus need to be done with a variable size lookup table - the big problem then is deciding the points when there is an exact match with competitors tied on one run. So far have not managed to figure that one out in Excel and still needs a human. I don't know what to do if ever the K1M was inquorate!!! Rule book does not seem to allow for this. You do need to cover these eventualities in a programme unfortunately!
The other thing we need to bear in mind is that there is an absolute limit of 100 on womens and C1 races before the scoring falls to pieces completely. At the moment even if all the Div 2's turned up at a race we would only have 66. We would have to think again if the numbers of paddlers rises significantly.
Judging by discussion we may also see a change in the C1W scoring for next year.
Nick thankfully corrected me when I first managed to give points to the Div 4's so managed not to repeat that one again.
Will try and get these into some kind of reusable order and happy then to share and test with anyone who is interested. May be Christmas though as work piled up at the moment.
It is also possible to trigger Excel through a USB port with an Algie Timy which in turn can accept multiple interrupts so a cheaper solution than the national timing team one is theoretically possible for lower division races.
If we have a standard format for results then Nick would I am sure be pleased.
All the best
Peter
Fairnilee Results - Has any one got a copy
-
djberriman
- Posts: 806
- Joined: Tue Apr 03, 2007 8:19 pm
I guess if we get the upload system correct we don't need the organisers to compute points, that means its just one program to calculate the points, one program to update with rule changes and one program to keep an eye on (and cope with oddities).
I was thinking an eventual 'solution' would be that results are uploaded to central system/database and basic processing done and checks made - organiser fixes basic errors (if any). Then provisional results/points are displayed on website. Once vetted by ranking officer (online) and corrections made (online) they become official results and the current rankings etc are updated. Being database driven would mean the display of provisional/official results and rankings would be automatic.
Clearly all ranking officers would need to be online, but if they didn't have to check cards and could simply cast there eye over pre-checked results for oddities and to deal with complaints etc from paddlers that would make their job easier and ensure a faster turn around.
I guess you'd need a rule as to a time limit for complaints/objections etc. If upheld what happens to points already given and used by paddlers for promotion claims. This happens now but I'm not sure how often and what effects it has.
Just spotted an error in my thoughts (thanks JS) and that is doubles were points must be calculated on site. Oh well back to the drawing board!
I was thinking an eventual 'solution' would be that results are uploaded to central system/database and basic processing done and checks made - organiser fixes basic errors (if any). Then provisional results/points are displayed on website. Once vetted by ranking officer (online) and corrections made (online) they become official results and the current rankings etc are updated. Being database driven would mean the display of provisional/official results and rankings would be automatic.
Clearly all ranking officers would need to be online, but if they didn't have to check cards and could simply cast there eye over pre-checked results for oddities and to deal with complaints etc from paddlers that would make their job easier and ensure a faster turn around.
I guess you'd need a rule as to a time limit for complaints/objections etc. If upheld what happens to points already given and used by paddlers for promotion claims. This happens now but I'm not sure how often and what effects it has.
Just spotted an error in my thoughts (thanks JS) and that is doubles were points must be calculated on site. Oh well back to the drawing board!
-
djberriman
- Posts: 806
- Joined: Tue Apr 03, 2007 8:19 pm
PS. I'd be concerned on any timing trigger being used on a multi tasking computer (eg windows). If the machine slows (for example when printing some results or starting something up etc) then I think the accuracy of the timing would be at question. The signal would arrive but depending on how its handled in hardware/software its accuracy could be very dubious. I doubt such things are meant for accurate timing but as I haven't seen the specs etc I could be very wrong.
Similarly if it crashes, loses power or needs rebooting you lose all your timing.
If you do try it out suggest you do some testing to determine accuracy when the pc is loaded and doing different tasks. A one second delay is nothing in computing terms but a lot in the slalom world!
A better solution (IMHO) would be a dedicated bare bones systems (such as a stripped down version of linux on a suitable small diskless battery powered system) to receive the signal and then pass that to your results system, thus the time should always be accurate (only doing one dedicated job) and the timings can then be read at leisure by the results sytem.
Crashes less likely, reboots less likely to be needed and very low power requirements.
I ASSUME thats how (or similar too) how the national timing system works.
Anyway just a thought.....
Similarly if it crashes, loses power or needs rebooting you lose all your timing.
If you do try it out suggest you do some testing to determine accuracy when the pc is loaded and doing different tasks. A one second delay is nothing in computing terms but a lot in the slalom world!
A better solution (IMHO) would be a dedicated bare bones systems (such as a stripped down version of linux on a suitable small diskless battery powered system) to receive the signal and then pass that to your results system, thus the time should always be accurate (only doing one dedicated job) and the timings can then be read at leisure by the results sytem.
Crashes less likely, reboots less likely to be needed and very low power requirements.
I ASSUME thats how (or similar too) how the national timing system works.
Anyway just a thought.....
The ranking officers will not be able to confirm the results until they recieve the entry cards and check them for acuracy. If an organiser is not getting the results to nick in a timley fashion what chance do the ranking officers have of getting the cards.
Another issue is unranked paddlers taking part in events, one event last year had the 1st and 2nd place paddlers unranked which completly threw the results
Another issue is unranked paddlers taking part in events, one event last year had the 1st and 2nd place paddlers unranked which completly threw the results
I do the K1M Prem ranking (probably the easiest section). I have a semi automated workbook.
For each event I cut and paste the results from the web into a spread sheet and format to match my workbook.
Then I run a macro which checks the points calculations, and warns of any discrepancies, compares names on results sheets with the names I expect to be associated with the bib number etc. Very occasionally I get anomalies, but most of these are for spellings. All resolution of anomalies is carried out manually. My spreadsheet then maintains the ongoing totals.
The laborious bit is comparing what is on the results page to what is on the cards. There are occasional errors but luckily these often do not make a difference to the points.
The trouble is that the cards can take forever to come through, so I tend to just batch them up and do them from time to time when it suits me!
On the other hand Prems are not contending for inseason promotion/demotion.
I used to run the same spreadsheets with minor variations for div 4s. This compared names/clubs/ages to attempt to match, but then also did some sorting to allow a visual match, but I know this could never be 100% accurate.
For each event I cut and paste the results from the web into a spread sheet and format to match my workbook.
Then I run a macro which checks the points calculations, and warns of any discrepancies, compares names on results sheets with the names I expect to be associated with the bib number etc. Very occasionally I get anomalies, but most of these are for spellings. All resolution of anomalies is carried out manually. My spreadsheet then maintains the ongoing totals.
The laborious bit is comparing what is on the results page to what is on the cards. There are occasional errors but luckily these often do not make a difference to the points.
The trouble is that the cards can take forever to come through, so I tend to just batch them up and do them from time to time when it suits me!
On the other hand Prems are not contending for inseason promotion/demotion.
I used to run the same spreadsheets with minor variations for div 4s. This compared names/clubs/ages to attempt to match, but then also did some sorting to allow a visual match, but I know this could never be 100% accurate.
Kit Washer, Entry Clerk, Chauffeur, Reluctant Organiser, Online Entry Advocate .....
Anything I post under this user is my personal opinion; I am not posting as a member of the Slalom Committee!
Anything I post under this user is my personal opinion; I am not posting as a member of the Slalom Committee!
-
Canadian Paddler
- Posts: 1480
- Joined: Tue May 25, 2004 8:31 am
- Location: Peterborough
- Contact:
Duncan,
negative thoughts first
Issue 1 - at least on ranking compiler is still (I believe) doing all the work by hand, does not have a computer and does not want one. Results in csv format will not help this person, and will be resisted.
I seem to recall that slasoft will output results electronically (or at least it used to do so), we just need a standard format and some way to make it easier for compilers.
Issue 2 - As an organiser, my cards are not always sent off the next day, I try to sit down and check them against the results myself, and Sunday night I am sometimes too tired to do that! (and sometimes try to check when kn***ered and I am sure mis some) a 4 day turn round is a challenge, and even then the results may be changed as someone may have been promoted without being sure, or may have started in the wrong division or. . . All reasons for recalculating points
Issue 3 - I pressgang jury chairs by virtue of how little they have to do. If we also expect them to stay after the event and check cards, there will be more events running without Jury chairs, or with club members as jury chair, as no one will want to do this
Positive
Definition of a standard format - CSV or XML and the ability to submit results electronically to those compilers willing to accept electronically is a first step. This will also be greener as Nick, bib officers, compilers for the division above etc do not need sheets of papaer and envelopes and stamps. . (Still need to get cards to the divisional compilers)
Small steps (not big bang) has a better chance of success for several reasons, including that big bang needs money - and proposals were around for a central ranking system 20 years ago, but the costs were to high. small steps can be free if done by volunteers!
negative thoughts first
Issue 1 - at least on ranking compiler is still (I believe) doing all the work by hand, does not have a computer and does not want one. Results in csv format will not help this person, and will be resisted.
I seem to recall that slasoft will output results electronically (or at least it used to do so), we just need a standard format and some way to make it easier for compilers.
Issue 2 - As an organiser, my cards are not always sent off the next day, I try to sit down and check them against the results myself, and Sunday night I am sometimes too tired to do that! (and sometimes try to check when kn***ered and I am sure mis some) a 4 day turn round is a challenge, and even then the results may be changed as someone may have been promoted without being sure, or may have started in the wrong division or. . . All reasons for recalculating points
Issue 3 - I pressgang jury chairs by virtue of how little they have to do. If we also expect them to stay after the event and check cards, there will be more events running without Jury chairs, or with club members as jury chair, as no one will want to do this
Positive
Definition of a standard format - CSV or XML and the ability to submit results electronically to those compilers willing to accept electronically is a first step. This will also be greener as Nick, bib officers, compilers for the division above etc do not need sheets of papaer and envelopes and stamps. . (Still need to get cards to the divisional compilers)
Small steps (not big bang) has a better chance of success for several reasons, including that big bang needs money - and proposals were around for a central ranking system 20 years ago, but the costs were to high. small steps can be free if done by volunteers!
All spelling errors are intentional and are there to show new and improved ways of spelling old words. Grammatical errors are due to too many English classes/teachers.
Old. Fat. Slow. Bad tempered. And those are my good points
Old. Fat. Slow. Bad tempered. And those are my good points
Just a note about the Algie Timy. It is in effect a computer in a timing box dedicated to providing times against trigger points.
The USB link would not be used to trigger anything - I know you couldn't sensibly do that on a Microsoft operating system for anything more accurate than hours - the time from the Timy is passed. The trick is getting it in the right field - this needs human support. Still however quite a way from anything workable - a ability to understand Dutch would be useful!
The USB link would not be used to trigger anything - I know you couldn't sensibly do that on a Microsoft operating system for anything more accurate than hours - the time from the Timy is passed. The trick is getting it in the right field - this needs human support. Still however quite a way from anything workable - a ability to understand Dutch would be useful!
-
djberriman
- Posts: 806
- Joined: Tue Apr 03, 2007 8:19 pm