Training Log Archive: vmeyerIn the 7 days ending Aug 31, 2014:
| ||||||||||||||||||||||||||||||||||||||||
| « | » |
| » now | ||||||||||
| |||||||||||||
| Mo | Tu | We | Th | Fr | Sa | Su | |||||||
Friday Aug 29, 2014 #
Note
Foot doctor today - three more weeks in the boot. Yeah?!
C • boot 4
Thursday Aug 28, 2014 #
Note
I wasn’t there, so I can’t say for certainty what actually happened to cause the on-site results process to fall apart at the world rogaine championships. However, after working with sportident and results software for a number of years, I will allow myself the opinion that is folly for someone with little or no SI experience to tackle the results for such a big event, especially if/when things go wrong.
Here is my personal recap, written to document what happened before, during, and after the event:
1) I requested and was sent the programming log ahead of time and confirmed that the controls and master stations (used to download the SI-cards) were all programmed in “Extended Protocol” and “Card6 with 192 Punches” selected. This was required in order to use the SI-6 cards above their natural limit of 64 punches. Ironically, no one using an SI-6 card at the event got more than 60 controls.
P.S. All the controls and 3 master stations had firmware 5.8, and 2 master stations had 5.74.
2) I guess to prevent cheating, all of the controls had different internal numbers than the external number. The only error I know of related to the controls was that Control 44 was coded in the results software with the same internal code as Control 33. So, teams that visited both controls only got credit for 30 points, and teams that only visited 44 were credited 30 points, instead of 40 points. Should have been an easy fix.
3) Teams downloaded, and while I can’t say for certain, I believe they were downloading fine if using an SI6 card, or if they had 32 or under controls on an SI10/11. If the SI10/11 cards were downloading OK up to 32 controls, then all hell probably broke loose when the later people, with higher control counts, started pouring into the Finish. Must have been heartbreaking to find out that there were no results to be had at the event.
4) So, an effort started to generate the results. First from the results software (ROM), data for about 100 teams was recovered from ROM, but only for teams with SI6 cards (despite the failure to write the data to the master station memory described below) or for those teams with SI10-11 cards who had punched 32 controls or less (or possibly this came from the master stations as described below). I do not know which data was in ROM and the reason the rest was not since I was not provided these files.
5) Besides the data being written to any results software when downloading, it is also written to the memory of the master stations. In an effort to recover the data, the memory of the five master stations were read out. Well, this is where the recovery process got interesting since all of the SI-cards had downloads in the memory, but they had not been written correctly.
The people using SI6 cards had no data for controls 1-64, and the people using SI-10/11 cards only had data for controls 1-32. So, the memory of the master stations was worthless for anyone with SI6 cards or anyone who had punched more than 32 controls with the SI-10/11 cards.
I was able to replicate this behavior with ROM, however the behavior was absent when I tested with other software, namely SI-config and a sportsoftware product.
6) Anyway, the only option was to collect all of the controls and read the units for the data stored to the control's memory. Three people, one being me, crunched the raw control data and tallied points for each team. Once we all were in agreement, the provisional results were published.
Lesson learned, since I am still learning after all of these years using SI: Recoverable data on the master is dependent on the results software. I always thought it was just a bunch of bits and bytes that could be reinterpreted later, if needed. However, I have been told that any block of data that is requested by the software to be read from the stick is also then written to the backup memory. If a block is not requested to be read from the stick by the software, it won't be in the backup memory. Now that I know that it is dependent on the software writing to the memory, I will check the software I use and make sure that the SI-cards are being written into the master station's memory correctly.
And, for me that closes the chapter of one of the most painful SI related experiences ever.
Here is my personal recap, written to document what happened before, during, and after the event:
1) I requested and was sent the programming log ahead of time and confirmed that the controls and master stations (used to download the SI-cards) were all programmed in “Extended Protocol” and “Card6 with 192 Punches” selected. This was required in order to use the SI-6 cards above their natural limit of 64 punches. Ironically, no one using an SI-6 card at the event got more than 60 controls.
P.S. All the controls and 3 master stations had firmware 5.8, and 2 master stations had 5.74.
2) I guess to prevent cheating, all of the controls had different internal numbers than the external number. The only error I know of related to the controls was that Control 44 was coded in the results software with the same internal code as Control 33. So, teams that visited both controls only got credit for 30 points, and teams that only visited 44 were credited 30 points, instead of 40 points. Should have been an easy fix.
3) Teams downloaded, and while I can’t say for certain, I believe they were downloading fine if using an SI6 card, or if they had 32 or under controls on an SI10/11. If the SI10/11 cards were downloading OK up to 32 controls, then all hell probably broke loose when the later people, with higher control counts, started pouring into the Finish. Must have been heartbreaking to find out that there were no results to be had at the event.
4) So, an effort started to generate the results. First from the results software (ROM), data for about 100 teams was recovered from ROM, but only for teams with SI6 cards (despite the failure to write the data to the master station memory described below) or for those teams with SI10-11 cards who had punched 32 controls or less (or possibly this came from the master stations as described below). I do not know which data was in ROM and the reason the rest was not since I was not provided these files.
5) Besides the data being written to any results software when downloading, it is also written to the memory of the master stations. In an effort to recover the data, the memory of the five master stations were read out. Well, this is where the recovery process got interesting since all of the SI-cards had downloads in the memory, but they had not been written correctly.
The people using SI6 cards had no data for controls 1-64, and the people using SI-10/11 cards only had data for controls 1-32. So, the memory of the master stations was worthless for anyone with SI6 cards or anyone who had punched more than 32 controls with the SI-10/11 cards.
I was able to replicate this behavior with ROM, however the behavior was absent when I tested with other software, namely SI-config and a sportsoftware product.
6) Anyway, the only option was to collect all of the controls and read the units for the data stored to the control's memory. Three people, one being me, crunched the raw control data and tallied points for each team. Once we all were in agreement, the provisional results were published.
Lesson learned, since I am still learning after all of these years using SI: Recoverable data on the master is dependent on the results software. I always thought it was just a bunch of bits and bytes that could be reinterpreted later, if needed. However, I have been told that any block of data that is requested by the software to be read from the stick is also then written to the backup memory. If a block is not requested to be read from the stick by the software, it won't be in the backup memory. Now that I know that it is dependent on the software writing to the memory, I will check the software I use and make sure that the SI-cards are being written into the master station's memory correctly.
And, for me that closes the chapter of one of the most painful SI related experiences ever.
Wednesday Aug 27, 2014 #
Swimming 46:26 [1] 1.83 km (25:22 / km)
80 laps, total of two minutes of break between each 10, with break 15-20 seconds in duration.
Had the pool to myself when I started. Nice and flat. Wanted to stop after 10, but decided on 50, then 10 more since easy I hadn't gone 30 minutes, and then 10 more since the cool down felt good and was under 6 minutes, and then 10 more because that 10 felt so damn good, and then I ran out of time, thank goodness. I liked swimming today and it liked me!
Had the pool to myself when I started. Nice and flat. Wanted to stop after 10, but decided on 50, then 10 more since easy I hadn't gone 30 minutes, and then 10 more since the cool down felt good and was under 6 minutes, and then 10 more because that 10 felt so damn good, and then I ran out of time, thank goodness. I liked swimming today and it liked me!
Note
Gathering my thoughts about the rogaine results debacle. Will post tomorrow maybe.