Matrix Games Forums

Forums  Register  Login  Photo Gallery  Member List  Search  Calendars  FAQ 

My Profile  Inbox  Address Book  My Subscription  My Forums  Log Out

[RESOLVED] Simulation frozen in heated battle scenario

 
View related threads: (in this forum | in all forums)

Logged in as: Guest
Users viewing this topic: none
  Printable Version
All Forums >> [New Releases from Matrix Games] >> Command: Modern Operations series >> Tech Support >> [RESOLVED] Simulation frozen in heated battle scenario Page: [1]
Login
Message << Older Topic   Newer Topic >>
[RESOLVED] Simulation frozen in heated battle scenario - 2/16/2020 12:27:08 PM   
gennyo

 

Posts: 46
Joined: 4/3/2019
Status: offline
Using CMO build 11218 with database 482 on windows 10.

While my scenario runs in Scenario Edit Mode to some point, the simulation engine seems frozen.

The Exception Log reads:

2020-02-16 21:17:13 -- B1121.8 -- Object reference not set to an instance of an object.
Exception: Object reference not set to an instance of an object.
Stack Trace: at Command_Core.Weapon_AI.(Single elapsedTime, Boolean RecalculatePlottedCourse)
Call Stack & Error details:
Error at 100970,

Attachment (1)

< Message edited by Dimitris -- 6/15/2020 5:08:06 AM >
Post #: 1
RE: [Logged] Simulation frozen in heated battle scenario - 2/20/2020 2:24:04 AM   
apache85

 

Posts: 2344
Joined: 12/18/2014
From: Melbourne, Australia
Status: offline
Logged for investigation.

0013656

_____________________________


(in reply to gennyo)
Post #: 2
RE: [Logged] Simulation frozen in heated battle scenario - 6/13/2020 5:03:30 PM   
Dimitris


Posts: 12431
Joined: 7/31/2005
Status: online
Trying to recreate this now in B1147.2 and haven't been able to.

Can anyone cross-check?

_____________________________


(in reply to apache85)
Post #: 3
RE: [Logged] Simulation frozen in heated battle scenario - 6/13/2020 6:14:12 PM   
Eboreg

 

Posts: 153
Joined: 3/14/2019
Status: offline
I would honestly suggest running this scenario on Blue Gene to fix your problems.

No, I'm not kidding, I don't know what you were trying to do with this scenario but I do know that most home computers won't be able to handle it.

(in reply to Dimitris)
Post #: 4
RE: [Logged] Simulation frozen in heated battle scenario - 6/14/2020 8:04:02 PM   
KnightHawk75

 

Posts: 555
Joined: 11/15/2018
Status: offline
You must have a really nice rig to be attempting 5k+ with another like 1000 missiles in the air without expecting a slide show. I like your style though . That said fwiw I ran it 4 times for ~2 in game minutes (which took forever on my old rig) without knowing how long to give it for your particular exception to happen. But those 8 minutes total took many hours .

Two runs were with b1146.2 and twice with b1147.2 All times I let it play down to about 5750 ish au's and around 12:15:50z in-game and my pulses were down to 12-15seconds by then lol. Each time was a complete restart of CMO, played in windowed mode, in the scenario editor by selecting that from the menu (though the first run I think I selected create new scenario and then loaded it - doubt it matters). For the most part the window didn't have focus but was not minimized during these hours of runs outside of when I noticed the log pop positive for an exception (monitored via snaketail) and paused to investigate.

The results were I got at least 1 or more exceptions during each run, I never got the same exception as the OP though. Attached are the 4 event logs from each run,the exception log, and a notes.txt which notes the in-game time they occurred (where I know I caught it at the moment it happened), the last autosave timestamp when some of the events occurred, and a couple other notes. In the end though I'm not sure how much help it'll be, if any at all.


Attachment (1)

(in reply to gennyo)
Post #: 5
RE: [Logged] Simulation frozen in heated battle scenario - 6/14/2020 8:59:17 PM   
thewood1

 

Posts: 4818
Joined: 11/27/2005
Status: online
So my laptop: i9-10980HK with RTX 2080 Super (mobile), 32 Gb RAM, 2 separate 1Tb m2 SSd, plus slightly overclocked to 5.1 GHz CPU speed.

It takes 32 sec. to run the first second. On the above rig, it is using 75% to 80% of the CPU cycles available with spikes to 100%. Again, on a 5GHz CPU

This is pure unrealistic fantasy. Its why we can't have nice things as players. Using one of Jeff Goldblum's best paraphrased quotes...Just because you can, doesn't mean you should. That many planes, missiles, weapons, etc. in the air at the same time would use up all the ready war-stocks in a few seconds. The rest of the scenario would be about waiting for production lines to ship new missiles over the next months.

To build something that brings my PC to its knees makes me wonder about the sanity and expectations of players. As a mechanic once said to a friend, just because the speedometer on the car goes to 140 MPH, does mean your not wrecking the engine while you're doing. This scenario is a waste of time for the builder, the player and the developer. Show at least a modicum of restraint.



< Message edited by thewood1 -- 6/14/2020 10:02:11 PM >

(in reply to KnightHawk75)
Post #: 6
RE: [Logged] Simulation frozen in heated battle scenario - 6/14/2020 10:40:50 PM   
Eboreg

 

Posts: 153
Joined: 3/14/2019
Status: offline
Not to mention each unit has an arsenal of weaponry so vast that they wouldn't be out of place in a Ratchet & Clank game.

(in reply to thewood1)
Post #: 7
RE: [Logged] Simulation frozen in heated battle scenario - 6/14/2020 10:48:42 PM   
thewood1

 

Posts: 4818
Joined: 11/27/2005
Status: online
Just for full disclosure, here is the the operating conditions of my PC. Note the CPU clock at around 5GHz, and the 2% load on the CPU. This is with the no scenario loaded.

I'll post what it looks like after loading in a later post.




Attachment (1)

(in reply to Eboreg)
Post #: 8
RE: [Logged] Simulation frozen in heated battle scenario - 6/14/2020 10:50:28 PM   
thewood1

 

Posts: 4818
Joined: 11/27/2005
Status: online
Here is the operating condition with the scenario loaded, but not running.

I'll post what it looks like in the next post.




Attachment (1)

(in reply to thewood1)
Post #: 9
RE: [Logged] Simulation frozen in heated battle scenario - 6/14/2020 10:52:39 PM   
thewood1

 

Posts: 4818
Joined: 11/27/2005
Status: online
Here it is after three seconds of run time.




Attachment (1)

(in reply to thewood1)
Post #: 10
RE: [Logged] Simulation frozen in heated battle scenario - 6/14/2020 10:59:20 PM   
thewood1

 

Posts: 4818
Joined: 11/27/2005
Status: online
Note it already had used up over a third of the CPU's capacity before I ever even hit start. And it was using two cores. Once I hit start, it used ALL eight cores and consumed 90% of the CPU's cycles.

As I said before, this is insane. This is the second time I have seen someone come in and have a completely outrageous massive scenario and complaining about CMO being slow or stopping. I have no issue if someone wants to waste their time building unplayable and un-run-able scenarios on their own time just to see what happens. But when you do it and then come in and expect the devs to drop other important stuff even to just load it and look at it, I'll start the complaining, as we all should.

btw, I managed to get a full couple of minutes out of one run. It never crashed, but it sure made my PC mad at me.

(in reply to thewood1)
Post #: 11
RE: [Logged] Simulation frozen in heated battle scenario - 6/14/2020 11:07:55 PM   
gennyo

 

Posts: 46
Joined: 4/3/2019
Status: offline
Well, I'm pretty certified insane (16 years of clinical depression).

BTW, This bug is almost fixed in 1143.1. Despite I can only run two or three seconds before reaching the memory ceiling (3GB) and crash, the save file it generates only lost 1s of in-game progress.
At least I could grind through this scen and complete it, and I did it multiple times before writing this.

Thanks for fixing this, developers.

1147.1 & 2 performs much worse than 1143.1 but as I still stuck on this scen this won't be a problem for sometime for me.

p.s. This is an early version of my insanity, later saves I submitted for debugging is much larger and worse for your computer's health (mentally and physically if you have bad thermal solutions), but that file is somewhat broken. Now my working saves are of ~25M
Maybe it could be a benchmark or furmark for CMO, especially for memory management improvements.

p.s.2 Drawing those units on screen seems like partly the culprit of high CPU usage. CMO draws per 1/10s, it may help if it could draw like CMANO per 1s.

(in reply to Eboreg)
Post #: 12
RE: [Logged] Simulation frozen in heated battle scenario - 6/15/2020 1:06:08 AM   
KnightHawk75

 

Posts: 555
Joined: 11/15/2018
Status: offline
quote:

ORIGINAL: thewood1
This scenario is a waste of time for the builder, the player.

If either are enjoying themselves I doubt either saw it as waste of time personally. Not everyone wants to play a realistic scenario, though I'm not suggesting this one is anything but fantasy. I don't think an exception though should be just be ignored because it came from someone's unrealistic 'what if' scenario. While on the other hand there are limits to how much time can be spent on something not easily repo'd and what might be an extreme edge, which it I presume more what you were driving at with your later comments. Though I don't see where the OP suggested everything be dropped in favor of looking into this issue though, so I don't understand that part.

I chose to invest some time in order to try and see if the error was reproducible, curious if it literally 'locked up\froze' or was just slow as hell for gennyo, as I expected it would be as soon as I saw it load, or if it just appeared like it froze when it really didn't. I wondered if exception related to the scenario pushing the limits of the engine to the literal design breaking point, or if it just exposes some other issue not previously accounted for or identified that causes it (the slowness just comes with territory of that many munitions flying at once). One perhaps seen only under certain stress conditions. Keeping in mind said conditions may not only be applicable to this fantasy scenario but perhaps other more realistic ones be they larger or smaller in unit count or munition exchanges. Exceptions go easily unnoticed in logs, and who knows, they might also be relatively benign. I don't know what the effects were of what might have been a weapon being destroyed and removed while the ai was still trying to recalc it's course (maybe nothing), nor of the the dictionary exceptions due to simultaneous access\modification. The latter though might be an example of of such things that show up more only under heavy stress or particular timing conditions gone easily unnoticed by players or devs. As a player my view is the more things found and addressed over time, the more solid and scalable the game over time for me and other customers, hence my personal time investment toward it. I obviously don't know what the root cause of exceptions were here but I do appreciate it wasn't just dismissed outright at first glance of the scene by the developers, nor would I expect it to land in the high-prio queue.

quote:

"btw, I managed to get a full couple of minutes out of one run. It never crashed, but it sure made my PC mad at me."


Excellent. I respect you trying to help as you usually do and spending some time trying to give it a run on your superior hardware, which might be a helpful data-point to add. In that regard did you get any exceptions in the log (maybe go peek) or otherwise when you let it pay for a couple in-game minutes? The data-point, if you have it, might be of some value, i'm particularly interested if you got dictionary collection was modified ones on a way way way faster rig.

@gennyo So does it actually freeze for you or is it just a pulse time in excess of say 60000, per actual game second that makes it appear that way?
If not literally frozen, disable all the ranges and illumination\tracking etc overlays,sound,probably just play with black\yellow coastline overlay, it may help you get an extra frame or two. Hey every little bit helps at these extremes. Also you mentioned you reach 3gb memory cap with this file? Odd, my game never went above ~1.5'ish in memory usage, not during the 2 ingame-minutes anyway?

< Message edited by KnightHawk75 -- 6/15/2020 1:07:12 AM >

(in reply to gennyo)
Post #: 13
RE: [Logged] Simulation frozen in heated battle scenario - 6/15/2020 3:20:13 AM   
gennyo

 

Posts: 46
Joined: 4/3/2019
Status: offline
It DO frozen at the time I asked for a solution(look at the original post date). After a series of updates, now it just run SLOWLY at 1143.1.

When I post this, the scen just stops, using less than 25% of CPU, sometimes just stand still, leaving most of the memory empty and eating less than 10% of CPU for hours while I sleep or working at daily job.

Now it just grinding my i5-6300U to bits, and the version of the scenario I'm running is much more insane than this save (I remember this one is not with 1000x ballistic missiles flying toward multiple airfields), but I could say it's RUNNING to the end, and does this multiple times.

So the main program is much improved in less than 4 months. That's quite a job well done in my opinion.

To everyone concerned: I know I offended someone by "playing" this in such a insane way. I do love CoW (the only DLC I bought for LASER GUNS), but it's the absolute sandbox mode make me pay for this game 3 times (CMANO/CMO on steam and CMO on Matrix). It's a little dirty secret and I was reluctant to upload this in order to get some much needed techincal help from developers. If it wasted your time, sorry.

p.s. I think I won't need to update my laptop for sometime because it won't help, thanks for testing this for me.

(in reply to KnightHawk75)
Post #: 14
RE: [Logged] Simulation frozen in heated battle scenario - 6/15/2020 5:07:12 AM   
Dimitris


Posts: 12431
Joined: 7/31/2005
Status: online
Marking as resolved.

_____________________________


(in reply to gennyo)
Post #: 15
RE: [Logged] Simulation frozen in heated battle scenario - 6/15/2020 6:27:32 AM   
Dimitris


Posts: 12431
Joined: 7/31/2005
Status: online
quote:

ORIGINAL: KnightHawk75
If not literally frozen, disable all the ranges and illumination\tracking etc overlays,sound,probably just play with black\yellow coastline overlay, it may help you get an extra frame or two. Hey every little bit helps at these extremes.


Actually that's very true, if you disable the display of contact emissions the map performance improves significantly. (DX9 rendering is single-thread by its nature, so it's quite hard to scale it out to take advantage of multiple cores. Sim-core execution OTOH does benefit from a beefy system; this was a screenshot of my task manager during testing.)

_____________________________


(in reply to KnightHawk75)
Post #: 16
RE: [Logged] Simulation frozen in heated battle scenario - 6/16/2020 12:30:26 PM   
VFA41_Lion


Posts: 228
Joined: 1/30/2014
Status: offline
CMO crashed for me in this save when I opened the order of battle and then tried to close it.

Running i9-9900k fwiw.

_____________________________


(in reply to Dimitris)
Post #: 17
RE: [Logged] Simulation frozen in heated battle scenario - 6/16/2020 4:57:27 PM   
ultradave


Posts: 1278
Joined: 8/20/2013
Status: online
quote:

ORIGINAL: gennyo



BTW, This bug is almost fixed in 1143.1.



I think calling this a bug is unfair. Yes, maybe you could struggle through this at some point. I'm not sure how after downloading and looking at it and trying to run it. Even after turning off range circles and contact displays, I could barely get it creeping along.

Really what you are asking for is that the developers make the program work, no matter what outrageous parameters you want to throw at it. That's unfair and bug is not the way to describe this. If they somehow make it so this is easy and runs, do you then create something twice as big and even more off the wall and then say it's a bug that it won't work, because you don't have enough memory or processor power?

I have no issue at all with anyone and everyone who wants to fiddle around with the game and experiment with anything their heart desires. Whatever makes it fun for you. After all- you paid for the game - do what you like.

But bug? I'm not seeing a "bug" here.

Dave


_____________________________

----------------
Dave A.
"When the Boogeyman goes to sleep he checks his closet for paratroopers"

(in reply to gennyo)
Post #: 18
RE: [Logged] Simulation frozen in heated battle scenario - 6/17/2020 2:42:09 AM   
gennyo

 

Posts: 46
Joined: 4/3/2019
Status: offline
This will be my last reply in this thread, I'm really curious about why the community will stuck on this one issue.

I can accept the result of my insanity, like crashes, OutOfMemory exceptions or even the red cross over map (as my graphic driver is really old and anemic hardware only exaggerate the problem). As a doctor, I know lots of stuff can't be fixed even with the best effort, like common cold, depression, COPD or stage IV malignant tumors. If the developers tell me “your problem can’t be solved because of the boundary of physics, you used too much resources in this crazy scenario”, I would accept this as a fact and do something else within the limits.

We can’t go faster than light (yet ), I’m comfortable with that.

But I think it may be solvable because of my own debugging experiments. Changing database (using db3k 477 really helps at some time), planning a smaller initial attack (use less rockets and missiles), even resort to some hacks like tweaking my ProcessLasso, compatibility settings, even bought a DRM-free copy from Matrix Games and excluding the DRM effects, etc., help this scenario moving forward 1-2 minutes in-game time, but never eliminate it.

I dug deeper in the exception log and discovered the “Object reference not set to an instance of an object” errors were filling my hard disk ( the log could grow to >4GB if left unchecked for more than 8 hours while I was sleeping). By my very limited programming experience, I thought something was messing with the internal data structures, and I felt it may be solvable, but without proper debugging experience and tools, I couldn’t go deeper.

As a pathologist, part of my daily job is asking for help from senior pathologist, so I posted this with a save file that I knew would make the whole community ridicule me, waiting for an answer, either an explanation of why this happens and what should I do, or better, a fix.

I expected no answer or something like “Hell no you can’t do this in the game, it simply can’t be done”. But the developers logged the issue and made some progress in 113x builds. Now the scenario run to some point and crashed as I expected, but the save file was corrupted: after loaded, I had it run for 1 or 2 seconds in-game time and got the same error as before (but not before it crashed, strangely). Some tweaks help, but still unable to eliminate it as before.

I posted another issue ticket thread here describing the new discovery, thinking it may be a problem related. I knew they had and have much to work on, so I won’t push for a fix, just waiting and chatting around.

I think I may push somewhat while I was replying at that thread, my wording is always inappropriate because English isn’t my first language (I got much more shame from some editors of a biological science journal, blaming my poor writing and laziness for not using some writing improvement service, which isn’t really in my control as the essay is not my primary job).

But I really was not expecting a permanent fix could make this scenario (and updated crazier one) run to the end.

Imagine my ecstasy when I saw the ending (which I write but never see in running game) using 1142.1 or 2 builds.

I really can’t express myself how I appreciate the wonderful job the developers of WarfareSims have done for me.

p.s. This scenario is better described as a “what it takes” scenario, not a “what if”, in my opinion. It’s like an anime (a bad one, really), but as we saw what the CCP does during 2019-2020 with Hong Kong and US, I think something as crazy as throwing everything they have into such a small theater is possible. Our Far East Asian brains are sometimes not as rational as yours.

This is somewhat an excuse for using “kami-like divine power” in a reality simulation program, but what I really try to do is managing my anger, not posting this as a community scenario. Paraphrasing what the English governments says, this is not for public consumption.

p.s.2. Talking of reality, how many in the community cares about SA-2 to SA-4 have SARH not pure command guidance? These old missiles never have onboard seeker, they rely command from ground guidance radar, or optical sensors, which is deployed at least on SA-2 (Fan Song F?) and SA-4 and have an engage distance of about 20km per simulated in SAM simulator. They could also be guided by advance fire control systems if they could provide the right commanding signal, like upgrade kit developed by PLA. In right hands, these flying telephone poles can and did do some harm on technological superior adversaries (think about North Vietnam’s barrage SA-2 firings and the lucky shot of F-117 by Serbian).

Or not having counter-battery radar capability? Or no C-RAM systems (Israel put big money on this and seems working for them)?

I’m not really pushing for these changes because I think some of them may have bad percussions, like super-potent SAMs or make some critical subsystem run amok. They may need much extra works to make them working properly and I won’t cry for them finished IMMEDIATARY.

I know everyone have their agenda and I have too, I know there are limits. I posted this save and reported this issue because I thought it may be solvable, but if I was told otherwise, I would forget it was an issue and tell everyone to do so and move on.

I didn’t report that it was fixed well before Dimitri do a test on this, that’s my fault. I know I wasted too much valuable time of many people, I’m sorry.

(in reply to ultradave)
Post #: 19
RE: [Logged] Simulation frozen in heated battle scenario - 6/17/2020 10:05:54 AM   
thewood1

 

Posts: 4818
Joined: 11/27/2005
Status: online
"I know I wasted too much valuable time of many people, I’m sorry."

Finally. Your own personal obsession is just that, a personal obsession. Everyone has them somewhere. But when you drag the devs into it, you are hurting the entire community by pulling them off of other stuff.

(in reply to gennyo)
Post #: 20
RE: [Logged] Simulation frozen in heated battle scenario - 6/17/2020 11:50:39 AM   
c3k

 

Posts: 267
Joined: 4/25/2017
Status: offline
OP: nice job stress-testing the programming. That's how you find out if there's a hidden flaw or something that needs tweaking.


(in reply to thewood1)
Post #: 21
Page:   [1]
All Forums >> [New Releases from Matrix Games] >> Command: Modern Operations series >> Tech Support >> [RESOLVED] Simulation frozen in heated battle scenario Page: [1]
Jump to:





New Messages No New Messages
Hot Topic w/ New Messages Hot Topic w/o New Messages
Locked w/ New Messages Locked w/o New Messages
 Post New Thread
 Reply to Message
 Post New Poll
 Submit Vote
 Delete My Own Post
 Delete My Own Thread
 Rate Posts


Forum Software © ASPPlayground.NET Advanced Edition 2.4.5 ANSI

0.215