View Issue Details

IDProjectCategoryLast Update
0002061AI War 1 / ClassicCrash/ExceptionJan 7, 2011 1:56 pm
Reporterwyvern83 Assigned Tokeith.lamothe  
Status closedResolutionunable to reproduce 
Product Version4.051 
Summary0002061: High lag during AI Wars startup, save game loading, and after loading
Descriptionedit: updated info

Ever since 4.049 I've been getting high amounts of lag when starting up AI Wars and when loading my first save game or my first lobby.

It seems to have gotten worse with 4.051 as it never normalized and eventually lead to a complete hang of the program after loading a save that I was able to terminate with task manager after a minute or two. It generated no reports in any of the logs I looked at. (the save doesn't seem to make a difference, the one from 0002016 barely has 11 minutes on it)
TagsNo tags attached.
Internal Weight

Activities

keith.lamothe

Dec 16, 2010 1:03 pm

administrator   ~0006317

Do you know for sure if it started in 4.049 or 4.050?

keith.lamothe

Dec 16, 2010 1:09 pm

administrator   ~0006318

I loaded the one from 2016 ("Maze B Fallen Spire 003 - 002.sav") and the final stage of loading (16, on the screen) took about 3-4 seconds longer than I would have expected in previous versions, but everything else seemed normal (including game running speed after load).

Are you getting something different in 4.051?

wyvern83

Dec 16, 2010 1:40 pm

reporter   ~0006320

Last edited: Dec 16, 2010 1:41 pm

Yes and No.

The funny thing about this problem is that after you've done it once it gets better. (Between system boots I guess since that's when it's the worst apparently)

It's paused on 13 and 16 for loading save games the first time today, later loads just on 16 and for less.

Unusual loading from disk waits happen during booting the game, loading a lobby, and loading the load menu. My first save load of the day also had load from disk as it slowly zoomed in after it finished loading. Again, it got better on subsequent loads. (Music skips also happened from time to time during the loading from disk phase)

As for when it started, I remember updating from 49 straight to 51 so it probably started in 4.049, maybe 4.048 I'll try to get back to you on that in a couple minutes.

wyvern83

Dec 16, 2010 1:54 pm

reporter   ~0006321

I'm going to settle on it being 4.049 as I generally update first before I do anything else with the game. (And all of my other recent activity with the game goes back to December 2nd or 3rd and it hasn't been this way for that long)

I got a pause on 13 again but its no where near as long as 16.

I'll stand by my observation that it 'improves' for some reason the more often you do it, including booting it up again for a subsequent time. With improvement being form unbearable to tolerable/almost not noticeable.

I'll try rebooting my computer next to see if that makes the difference.

keith.lamothe

Dec 16, 2010 1:58 pm

administrator   ~0006323

Is anyone else seeing this in 4.051+?

wyvern83

Dec 16, 2010 2:25 pm

reporter   ~0006327

Last edited: Dec 16, 2010 2:26 pm

Whether or not its the first time you've started the game after a system boot-up seems to make a difference although the save loading problem isn't as massive as it was this morning since it has yet to freeze on me.

Lag pattern for booting up the game goes something like this: first pause around three hundred twenty-seven (less freq. on subsequent) with a very large pause on or around two hundred ninety followed by a halting progression to around two hundred where it gets really fast, followed by a slow down around ~hundred and eighteen (varies), followed again by a really fast load down til ~one or seven where it will pause a while the first time around, less so on subsequent game boots.

If no-one else is getting this I'm willing to entertain the notion it might be hardware or I need a fresh copy.

Chris_McElligottPark

Dec 16, 2010 8:40 pm

administrator   ~0006340

I periodically get this sort of behavior not only with AI War, but also with programs like photoshop, etc. Usually it's when I'm running for months without a reboot, and then a reboot solves the problem. In your case, sounds like the opposite problem, but the general root cause might be the same: swap space access. You might look into creating a fixed-size pagefile in windows, to avoid issues like that when your system is first booted.

It could also be that stuff on your system is still loading in the background (Steam or whatever else unrelated to gaming), and that this is therefore creating a bottleneck for AI War the first time it runs.

I should also note that with both .NET and Mono there is apparently some caching of JIT-compiled code, and in general memory space. So when you launch the game the second time on a boot, even if everything on your machine is working properly, you should definitely see it load ridiculously faster the second time. That's just the caching at work, and it not having the do the JIT compilation the second time.

Now: none of that is to say that you might not have an actual issue here, and if others are seeing this then that would definitely be a clear sign it's something new in AI War. But, whether this ultimately does turn out to be a problem with AI War or not in this specific instance, those other factors are also something to keep in mind just as a general case.

wyvern83

Dec 16, 2010 8:49 pm

reporter   ~0006341

Thanks for the advice. I'll look into it and let you if it helps if I can do it.

Chris_McElligottPark

Dec 16, 2010 9:09 pm

administrator   ~0006343

Sure thing -- that pagefile thing is something I've been doing since windows 95, and it really does help. I have 4 GB of physical ram on my desktop, and so I keep a static 2gb pagefile (min and max both 2gb). Really helps with system post-boot speed while all that general windows stuff is loading into memory for the first time since boot, etc.

Issue History

Date Modified Username Field Change
Dec 16, 2010 1:01 pm wyvern83 New Issue
Dec 16, 2010 1:03 pm keith.lamothe Note Added: 0006317
Dec 16, 2010 1:09 pm keith.lamothe Note Added: 0006318
Dec 16, 2010 1:09 pm keith.lamothe Assigned To => keith.lamothe
Dec 16, 2010 1:09 pm keith.lamothe Status new => feedback
Dec 16, 2010 1:40 pm wyvern83 Note Added: 0006320
Dec 16, 2010 1:40 pm wyvern83 Status feedback => assigned
Dec 16, 2010 1:41 pm wyvern83 Note Edited: 0006320
Dec 16, 2010 1:54 pm wyvern83 Note Added: 0006321
Dec 16, 2010 1:58 pm keith.lamothe Note Added: 0006323
Dec 16, 2010 1:58 pm keith.lamothe Status assigned => feedback
Dec 16, 2010 2:25 pm wyvern83 Note Added: 0006327
Dec 16, 2010 2:25 pm wyvern83 Status feedback => assigned
Dec 16, 2010 2:26 pm wyvern83 Note Edited: 0006327
Dec 16, 2010 2:27 pm wyvern83 Description Updated
Dec 16, 2010 2:28 pm wyvern83 Description Updated
Dec 16, 2010 2:29 pm wyvern83 Description Updated
Dec 16, 2010 8:40 pm Chris_McElligottPark Note Added: 0006340
Dec 16, 2010 8:49 pm wyvern83 Note Added: 0006341
Dec 16, 2010 9:09 pm Chris_McElligottPark Note Added: 0006343
Jan 7, 2011 1:56 pm Chris_McElligottPark Status assigned => closed
Jan 7, 2011 1:56 pm Chris_McElligottPark Resolution open => unable to reproduce
Apr 14, 2014 9:27 am Chris_McElligottPark Category Bug - Crash or Exception => Crash/Exception