Inconsistency in time per turn

Hey, haven’t posted on this thread but this is probably related to the server issues causing games to crash as well, with the root cause being that our servers have been having a bad time. Working on it.

1 Like

Has the server slowdown returned? I haven’t seen any crashes yet but I’m starting to lose identical matchups due to timeout. Substantial differences between identical versions of mine from several days ago vs today, between 11 and 20+ additional timeout damage taken. Turn-by-turn comparison seems to show that all turns are around 1.5 to 2 times slower. For resistance to slowdowns I tried to keep turns to around 3 or 4 seconds during complicated decisions, but similar turns today are between 5 and 8 seconds.

I also saw a couple turn zero crashes again this afternoon.

@RegularRyan - Guess it’s time to turn the server off and on again? :slight_smile:

1 Like

Oh, yep I’m seeing some on my algos now https://terminal.c1games.com/watch/4291712

Now all the top players on the leaderboard are getting roasted by your programs @n-sanders 'cause they all suffer disproportionately from the slowdown :rofl:

Yeah, I hadn’t expected my quick and dirty algo mashup to start tearing it up quite that much :smiley:
It was intended to be for intel-gathering purposes, but I’ll enjoy the bragging rights for at least a little while.

Also, it’s clear there are a couple top algos that don’t track their own time per turn to abandon deeper searches when they run low on time.

Yeah, the slowdown seems to manifest in different ways depending on how people are using their simulators. For the ones like Oracle, which I’m pretty sure just keep searching until time is up and then go with the best thing they could find, they don’t typically time out but they do make worse moves, and thus can sometimes be beaten in that manor.
Meanwhile, other ones cough like mine cough aren’t performing open-ended search, but are going through a deterministic predefined (but large) set of options to evaluate and thus don’t have any non-destructive way to abort when time is out. This approach seemed okay when the run time is budgeted into the design, but when 3 second turns become 9 second turns the next day that approach starts to break down.

Yeah, I think Oracle and Lazy both do open ended searches (which I give huge kudos to). I’d hope they’ve done a bunch of work in prioritizing those searches though.

My heaviest algo will sim upwards of 250 sims in a turn (that’s my bear-cave line of algos) depending on how many cores I can spend. I’m guessing it’d be similar to what you describe as a deterministic predefined set of options. But my timeouts interrupt in such a way that I can use whatever “best” simulation I’ve discovered thus far and I’m exploring in a breadth-first manner. I do appreciate rust’s out-of-the-box performance gains over python though. I definitely don’t know python well enough to optimize it for performance.

Ultimately I hope C1 is able to get a long term solution in place - I’d much rather all our algos are running at their highest level of performance.

Also, I’ll be glad once we’re past this stretch of so many duplicate algos running around. I totally understand why people do it (and have done some myself), but I don’t think the algo ratings system handles it well when wins or losses are being stacked against duplicates.

Yeah I’m personally guilty of the duplicate thing lately. With the slowdown on top of the general non-determinism of several top algos, I feel like I have to see each new version perform at least 3 or 4 times to get a good idea of how well a change it actually stands in the elimination tournament. It’s all a game of probabilities now.

1 Like

@RegularRyan Any update on this? I’m still seeing slowdown and turn zero crashes, and there’s only two days left before the deadline. Accurate feedback is critical as people calibrate their final submissions, and so many of the top algos have performance that varies with server speed.

1 Like

Turn-zero crashes still happening as of 5pm EST 08-31. https://terminal.c1games.com/watch/4309629

@RegularRyan, @C1Junaid, somebody, we are down to about 30 hours until the deadline, is there any notice on the state of this issue and whether the leaderboard matches now while we develop will accurately reflect conditions during the competition?

3 Likes

11 hours left. Two turn-zero crashes happening back-to-back:
https://terminal.c1games.com/watch/4317044
https://terminal.c1games.com/watch/4317046

These are not necessarily proof that the match behavior in non-crash matches will be different, but these crashes seem to have been correlated with changes in server performance previously.

I’m really not sure what to do at this point. Every design change I’ve made in the last two days has had it’s effectiveness change the next day, when the server is in a different mood.

P.S. match rates seem slow? It might just be my biased/panicked perception, but I feel like when there’s 7 hours left and I wait 90 minutes I should have played more than 14 matches.

Edit 2: Five hours to go and I’m still getting the occasional loss from turn-zero crashes: https://terminal.c1games.com/watch/4319440
I presume the devs are out for the weekend, and hopefully they’ll be back tomorrow. If there are still turn-zero crashes after the submission cutoff, I think the cutoff should be extended.

2 Likes

I’m getting alot of timeout damage aswell, eventhough I really shouldn’t.
I have seen some other people take some aswell.

I am now seeing several matches that have been played after the cutoff, with significantly different time characteristics from yesterday evening while development was still active. Differences a large as 13 or more timeout damage where it used to be 2 or less, changing the outcome of the match. This is troubling as I feel like my preparations were meaningless. Even more troubling, I have found a turn-zero crash in one of my match histories, timestamped as 9-2 after the cutoff: https://terminal.c1games.com/watch/4322347
This crash has now robbed one of arnby’s oracles of however many points that loss was worth, at a time when we can no longer do anything about it. Some of the leaderboard positions are quite close now, potentially less than one or two losses apart.
I am curious if anyone else is able to uncover turn-zero crashes in their match history that are marked as having occurred after the cutoff.

Since the cut-off, I took 4 turn-0 crashes… :pensive:

As for the computing time issue, I don’t know what to say… If this played a role in my sudden fall, I will feel quite depressed… It can’t be only that because everyone improved their algos until the very end, but I don’t know what to think
It leaves a very bitter taste to the end of this season

I have had some aswell…

Hey everyone, I was on vacation last week. I’ll be looking into all the forum posts today and handling them.

3 Likes

turn 0 crash ocurence:
https://terminal.c1games.com/watch/4353992 (season 4, double crash)
https://terminal.c1games.com/watch/4352304 (season 4)
https://terminal.c1games.com/watch/4351845 (season 4)

same here: (season 3 though)
https://terminal.c1games.com/watch/4352139
https://terminal.c1games.com/watch/4352475
https://terminal.c1games.com/watch/4354182

I put some figures in this topic about this

Edit: That is some good start