Subject: duplicate games
Date: Thu Oct 27 22:28:08 2016
User: ix
Message:i know there are threads about this but i couldn't find them. still surprised when it happens.
10/27 10:16 pm 8x3 23878-5 Winnable 2:05 Won
10/27 10:07 pm 8x3 23878-5 Winnable 2:39 Won
Date: Thu Oct 27 23:05:10 2016
User: ix
Message:found one, why did searching "same" not find a thread titled "Got the same game twice!"
Date: Fri Oct 28 02:40:52 2016
User: TNmountainman
Message:Almost certainly because, as per Denny's "Search Results" page:
"Search terms too short
Search term contained common words like 'the' or 'and'"
I've learned it's quite the art to use the search function on this site. And Denny said once (or more), his search function is very rudimentary. It's kinda like Google was in it's infancy, if you recall how that was - which was indeed an art form at that time. (My characterization.)
As many times as this topic (same/duplicate games) has come up, I've done your search several times. Using the word "duplicate" for the subject brings some results, but far from all. Using it in "body" format brings lots of noise, but more instances, if one can weed out the ones not relevant (not an easy task).
There is yet another way, which I've kind of tweaked, that works better than Denny's search function, and leads to things such as this previous discussion....
Link: just one of numerous previous discussions
Date: Fri Oct 28 12:46:51 2016
User: The_Longhorn
Message:):o I've gotten the same deal twice in a row. In fact, I recall it happening twice. I think it's safe to say that the probability of this happening is 1/32768*. But what's bad is when it's a really difficult deal, you solve it on the first try, but then your brain is so fried from concentrating that you forgot how you did it, and it ruins your streak on the second attempt.
):o On another occasion, I got the same deal twice, but with another deal in between.
):o Moo.
*I've been away for awhile, so I'm basing this on the fact that in 2007, there were 32768 of each variant and level. 32768 is not a hard number to remember, when you know that it's 2¹⁵.
Date: Fri Oct 28 12:51:17 2016
User: The_Longhorn
Message:):o Yeah, I started a discussion on this back in 2010, so I bumped it to the top. The discussion turned into a fascinating read.
):o Moo.
Date: Sat Oct 29 22:33:42 2016
User: joeygray
Message:So, ya'll. What are the chances you get the same deal twice in a row and you don't even notice. I'm very much afraid that's quite likely in my case.
Date: Tue Nov 1 06:39:18 2016
User: kangaroo
Message:I thought the game looked familiar - wasn't consecutive but next best thing! First time it's happened to me.
11/1 6:34 am 13x0 18678-7 Streak 2:13 Won
11/1 6:32 am 13x0 30609-7 Streak 2:30 Won
11/1 6:29 am 13x0 18678-7 Streak 2:35 Won
Date: Sun Nov 6 02:38:05 2016
User: dcn21
Message:I've had the same deal twice in a row at least 6 times.
Date: Fri Dec 9 13:06:49 2016
User: Punster
Message:Here's one =>
Recent History of Play
Date Time Game Mode Elapsed Won/Lost
12/9 1:02 pm 12x1 8048-10 Winnable 1:31 Won
12/9 1:00 pm 12x1 8048-10 Winnable 2:37 Won
I've had this happen to me on multiple occasions, but I've logged many hours of game time over the years, too.
Date: Sat May 6 13:47:40 2017
User: TNmountainman
Message:Although I've had consecutive duplicate games several times, this may be the first time with an intervening game, as The_Longhorn and kangaroo mentioned above:
5/6 1:38 pm 12x0 12298-5 Streak 3:19 Won
5/6 12:58 am 12x0 25710-5 Streak 3:45 Lost
5/6 12:54 am 12x0 12298-5 Streak 2:01 Won
Date: Sun Feb 21 08:38:46 2021
User: sebcbien2
Message:should I try to get a fourth time ? ;-)
2/21 8:32 13x0 2813-10 Winnable 1:30 Won 100% 709 96 96
2/21 8:30 13x0 2813-10 Winnable 1:10 Won 100% 709 96 96
2/21 8:29 13x0 2813-10 Winnable 1:27 Won 100% 709 96 96
having duplicates happened more than probabilities say so.
For example, I never saw N and N+1 in a row (okay, this is more difficult to spot)
Date: Sun Feb 21 16:40:21 2021
User: MrFixit
Message:Stats people: a half million games per week, about unique non-anonymous 2700 users. How often? (I know we've covered this ground many times). Does that really mean the average non-anon here plays 26 games/day?
Date: Mon Feb 22 10:24:42 2021
User: cellmate
Message:
does the server go out of its way to have a stream of random numbers that are used exclusively for game number?
Date: Fri Apr 9 11:49:16 2021
User: TNmountainman
Message:11:42 10x2 1922-10 2:29 Won 92.9% 1362 13 14
11:39 10x2 12928-10 2:35 Won 90.9% 1370 20 22
11:34 10x2 1922-10 3:16 Won 92.9% 1362 13 14
What's curious is that both plays show 13/14. Maybe that's something that now only updates every hour, instead of immediately like it used to. (And yes, for those who follow such.......I won it quicker the second time.)
Date: Fri Apr 9 12:55:46 2021
User: WRAC
Message:It still updates immediately. But the display is of the stats as of the time of the page load, not as of the time the games were played.
Date: Fri Apr 9 14:55:45 2021
User: TNmountainman
Message:Hmmm..........not sure I grasp that. In fact, it still shows the same stats for both.
Is this another 'deficit' of the new way scores are displayed? This certainly isn't the way they would have displayed "before".
Date: Fri Apr 9 15:02:44 2021
User: ix
Message:at 11:34 the game displayed as 12 13 but when you won it again it changed
Date: Fri Apr 9 15:10:10 2021
User: WRAC
Message:ix is correct.
So yes, it will continue to show the same stats for both... until they "scroll off the page." If the game is played again, both entries will be updated.
In other words, when you request your recent play, the current stats for each game are retrieved from the database. As far as I can tell (and I would assume this to be the case), the stats as of the time a particular game was played are not saved. You can only see what they are at the time of your request.
Does this make sense now?
Oh, and although I can't provide proof, this is my memory of the way it's always worked.
Date: Fri Apr 9 15:34:45 2021
User: TNmountainman
Message:No, but I presume the fault is mine. As to ix's point, exactly *when* would I have been able to see it as 12/13? I don't think I could have. And I assume this is not the same as winning a game, and immediately replaying it. My understanding has always been that that second win wouldn't 'count' for the database. I do see that sebcbien2's 3-in-a-row example from above all show the same stats, but still don't understand it. But don't worry about it; I think I'm just not understanding either how the database gets updated, or something like that.
Date: Fri Apr 9 15:38:54 2021
User: Kumquat-of-Conciliation
Message:Mr. Mountainman has requested I bring this to your all's attention. While it's a different situation, from a tournament, note that the database apparently didn't change after ElGuapo won for the second time.
Link: earlier duplicate game occurrence
Date: Fri Apr 9 15:48:08 2021
User: ix
Message:if you looked at 11:39 - 11:41
11:39 10x2 12928-10 2:35 Won 90.9% 1370 20 22
11:34 10x2 1922-10 3:16 Won 92.9% 1362 12 13
if you looked at 11:42
11:42 10x2 1922-10 2:29 Won 92.9% 1362 13 14
11:39 10x2 12928-10 2:35 Won 90.9% 1370 20 22
11:34 10x2 1922-10 3:16 Won 92.9% 1362 13 14
the other thing is: and now for something completely different
Date: Fri Apr 9 15:58:13 2021
User: WRAC
Message:Yeah, exactly what ix said. Except the winning percentage for 1922 should be 92.3% from 11:39 - 11:41. ;-)
And yes, completely different. In fact, I don't believe the Recent Play page even existed back in 2006.
Date: Fri Apr 9 16:05:56 2021
User: ix
Message:damn, curses, blast and John Cleese type vein bulging fist wave
Date: Fri Apr 9 16:37:21 2021
User: outskirts
Message:Up to the minute straight up, no old stats here.
Date: Fri Apr 9 16:45:30 2021
User: TNmountainman
Message:Ok, let me try this thought experiment............
If, by chance, before playing that game for the first time, I had gone to the database, and looked at the record, it would have shown 12/13, I think, right? Let's say that was at 11:33, instead of the 11:34 ix initially posted. Then I won it, and it became 13/14, right? Would the master database have also shown that? So then..........I win it again 8 minutes later, so it *should* be 14/15, one would think. So at what point will(?) the database update to show 14/15? It's obvious I'm still missing something, but can't figure out what it is. If it was 12/13 before I started playing, then it most definitely should be 14/15 now. And I can't see how that's not true.
As to the "recent play" page - no, of course it didn't exist back in 2006. That's only since late 2020.
Date: Fri Apr 9 16:56:58 2021
User: joeygray
Message:No, the recent play page has been around forever. It used to be “the last hundred”, but in 2020 Denny finally revealed (and much to my chagrin that I hadn’t long since figured it out instead of being confused about it) that he has always saved between 1 and 2 weeks worth for everybody: once a week he deletes back to one week, then for a week it accumulates back up to two weeks. It was at that time that he started just showing everything he has instead of the most recent 100.
Date: Fri Apr 9 17:03:30 2021
User: outskirts
Message:12 13 was what it was *after* being played the first time ... You need a nap maybe TN
Date: Fri Apr 9 17:03:48 2021
User: joeygray
Message:And yeah, a long ago request of mine was to save in the top 100 the stats “as played” of the game, not just store the game id and then look up at display time the stats. Still would like it. Never expect it.
Still hoping for a feed of the recent play page, though. As of now I have to scrape it from om the page. (And, annoyingly, look up the internal player id even though the unique nic ought to suffice.)
Date: Fri Apr 9 17:06:42 2021
User: outskirts
Message:And with a bit of a mind flip
You're into a time slip
Date: Fri Apr 9 17:08:24 2021
User: outskirts
Message:And nothing
will ever be the same
Date: Fri Apr 9 17:10:01 2021
User: TNmountainman
Message:By "recent play PAGE", I assumed he meant a page/link with that title. Yes, of course Denny listed the last 100 played, but I'm pretty sure that page had no title. It was just the player's stats page, right?
Date: Fri Apr 9 17:14:01 2021
User: TNmountainman
Message:See, ix, that page is called "Individual Stats". Well down the page, there's a SECTION called "recent history of play", but I don't think that could be called a "Recent Play" PAGE. While that's semantics, that's exactly why I posted the the "Recent Play" PAGE was a new invention.
Date: Fri Apr 9 17:14:23 2021
User: ix
Message:i'm pitching an idea to Chuck Lorre, Young TN
Date: Fri Apr 9 17:14:24 2021
User: joeygray
Message:Well, for our part, we assumed you meant the information. Not the details of the page name and whatnot.
Date: Fri Apr 9 17:18:44 2021
User: TNmountainman
Message:I understand that. That's why it's a case of semantics. I was only responding to WRAC's comment about "Recent Play page".
Date: Fri Apr 9 17:27:48 2021
User: WRAC
Message:joey is correct. I wrote "Recent Play page" to save myself the time of typing "the page where you can view your recent play statistics" - time that I have now lost. :-)
Date: Fri Apr 9 17:32:35 2021
User: TNmountainman
Message:ix: "before you played it, it was 11/12..."
This is the key piece of the puzzle I was missing, I believe. When I first saw it as 13/14, it was BECAUSE I HAD ALREADY WON IT TWICE, and the database does not show the as-time-goes-on record of the stats. And after the first victory, it *should* have been 12/13, but the updated data wipes that away. Ok, I think I've got that.
But then that brings up another question. We've all noticed that the elo/elg score of games vary within a variant depending on how many times it's been played. Any given 12x0 that's say, 0/15 will be higher-rated than one which is 0/11. Ok, altho that's not really accurate, it's the best way we (ElGuapo) has to designate and adjust it after more plays. Similarly, a game which is 5/10 will be rated lower than one that's 4/10, etc. - all other things being equal. So.................does that mean that my first victory over this game, when it *became* 12/13, should be 'worth' more than my second victory over it when it *became* 13/14? We know the difficulties and elo/elg points are NOT back-calculated/adjusted, so you all see my point? It's just an esoteric one, and no big deal, but it would help me see when and how these calculations happen. So............I wonder if the decimal points would show that?
Thank you all for being patient with me on this...........
Date: Fri Apr 9 17:53:20 2021
User: outskirts
Message:Lol @ Young TN
Lol @ WRAC "time that I have now lost"
I feel ya
Date: Fri Apr 9 18:41:29 2021
User: outskirts
Message:How do I find that recent play PAGE?
Date: Fri Apr 9 18:48:01 2021
User: ix
Message:open a new tab and google "recursion", google will ask "Did you mean: recursion", chuckle and close the window. then click your name and scroll to the bottom of the page.
Date: Fri Apr 9 18:48:38 2021
User: TNmountainman
Message:Click on your name. Go to the bottom where it says "Recent Play", and click that. It'll take you to this:
"Recent Play - outskirts"
Link: outskirts's "Recent Play" page
Date: Fri Apr 9 18:52:44 2021
User: jamesblackburn-lynch
Message:ix!
Nice one! I didn’t know that. I was just teaching recursion yesterday. Missed my chance.
James
Date: Fri Apr 9 19:15:05 2021
User: outskirts
Message:What is wrong with google? Why can't they spell?
I'm always and forever simplifying stuff, to save time and also so as not to be boring.
Am I a recursor?
Date: Fri Apr 9 19:19:35 2021
User: outskirts
Message:If we could go back in time, I suggest a young Danny Bonaduce for Young TN
Date: Fri Apr 9 19:22:52 2021
User: joeygray
Message:To answer tn’s question, yep, if you get dealt and beat the same game twice in a row you get less credit for the second win. At that point your Elo is higher and its is lower.
Now, I do not know if the engine does all games across all players in strictly increasing time of completion. That would be the most correct, but possibly it does all of one player before moving on to the next player’s games. Maybe ElGuapo knows?
Date: Fri Apr 9 19:23:46 2021
User: outskirts
Message:Please don't anyone answer that recursor question TN
Date: Fri Apr 9 20:45:39 2021
User: ElGuapo
Message:I saw an early screen print, joey, that showed the program doing all point exchanges for a given player and then moving to the next, so that seems to be how it's done.
Date: Sat Apr 10 03:30:17 2021
User: ix
Message:hey james, google these
askew
do a barrel roll
Post follow-up