This is a 'move' of a thread I started in the Tourney forum, but the number of posts there is so low I suspect very few folk are going there and its more appropriate for here anyway, so here's a copy of the thread:
------------------------------
Me: See the diana grouped random (10th july now, started 26th march)
Still on round one as there are two ongoing games, but significantly, the results of these games are irrelevant: the groups have already been won.
In grouped tourneys, why not allocate group winners when the points make them winners, not 'all games have been played in group'?
The next round could then start when all groups have winners (or possibly, all group winners have finished their group games).
The delay in starting the next round as currently observed seems to serve no purpose.
------------------------------
EXY: I would like to see this happen too. No point hanging on if the games outstanding have no baring on the second round.
-----------------------------
SirLoseALot: Excellent idea!
-----------------------------
Me again: See also Vulcan Grouped Random
13 games in round one to go: none effects the group winner outcomes. That's a good month right there.
Originally posted by ToeI'm hearing what you are saying - it does make a lot of sense.
This is a 'move' of a thread I started in the Tourney forum, but the number of posts there is so low I suspect very few folk are going there and its more appropriate for here anyway, so here's a copy of the thread:
------------------------------
Me: See the diana grouped random (10th july now, started 26th march)
Still on round one as there are two o ...[text shortened]... es in round one to go: none effects the group winner outcomes. That's a good month right there.
I will need to load up with coffee and send the family away for a week before I tackle that one though...
-Russ
Originally posted by RussWhile the exact algorithm is indeed quite complicated, you can use a simple one that does not take into account that when one person wins, the other loses. So, if the current leader already (he/she doesn't have to have all games completed) has more points than any other player in the group can possibly get, he/she is the winner. If all groups have a winner, the next round can start.
I'm hearing what you are saying - it does make a lot of sense.
I will need to load up with coffee and send the family away for a week before I tackle that one though...
-Russ
Another thing that would greatly help is the automatic starting of tournaments (when it is full) and new rounds (when all games have been played/all groups have winners).
Originally posted by pidermanWell, what Russ once said made alot of sense, he doesn't start them over the weekends as alot of people don't have internet acces during that time, he always starts them monday, first thing in the morning, if I'm correct. 🙂
While the exact algorithm is indeed quite complicated, you can use a simple one that does not take into account that when one person wins, the other loses. So, if the current leader already (he/she doesn't have to have all games completed) has more points than any other player in the group can possibly get, he/she is the winner. If all groups have a winner ...[text shortened]... ents (when it is full) and new rounds (when all games have been played/all groups have winners).
O.
Originally posted by LivingLegendI will only start them at the weekend if they are not too time sensitive (timeout period). I do tend to start them in the morning (GMT) too, because that is when the machine is least loaded.
Well, what Russ once said made alot of sense, he doesn't start them over the weekends as alot of people don't have internet acces during that time, he always starts them monday, first thing in the morning, if I'm correct. 🙂
O.
-Russ
Actually, I understand Russ's desire to not have them start automatically. I can live with the couple day delay on the start if the winners are decided when they clinch the group victory as opposed to waiting for all the game to complete.
One idea for starting the non-first round of any tournament. Have all timeouts start counting 48 hours after the start of a tournament. This gives people time to check their email and check the site without touching their timeout, much less their time bank. I don't know if this is a good idea, but it might allow Russ to start things on the weekend when volume is also low.
- Joined
- 26 Apr 03
- Moves
- 26771
Originally posted by Pie1120Or perhaps, for all games, a person's clock shouldn't start until either he/she has made at least one move in that game OR a longish initial move timeout period (say 5 days or something) has expired
Actually, I understand Russ's desire to not have them start automatically. I can live with the couple day delay on the start if the winners are decided when they clinch the group victory as opposed to waiting for all the game to complete.
One idea for starting the non-first round of any tournament. Have all timeouts start counting 48 hours after the st ...[text shortened]... is a good idea, but it might allow Russ to start things on the weekend when volume is also low.
Originally posted by Toebump... (sorry Russ: we can all see you've been doing a lot on Uchess of late, but lest one forget...)
...see the diana grouped random (10th july now, started 26th march)...
Update: down to one game some 16 days later. Alas one of the remaining participants appears to have gone on holiday (not that the game's outcome is in any doubt). Fingers crossed he has net access from the beach!