View Issue Details
ID | Project | Category | View Status | Date Submitted | Last Update |
---|---|---|---|---|---|
0010570 | Dwarf Fortress | World Generation -- General | public | 2018-02-10 17:02 | 2018-05-29 14:19 |
Reporter | untrustedlife | Assigned To | lethosor | ||
Priority | normal | Severity | crash | Reproducibility | always |
Status | feedback | Resolution | open | ||
Platform | PC | OS | Windows 10 | OS Version | Insider Builds |
Product Version | 0.44.05 | ||||
Summary | 0010570: World gen still crashing around year 500 ish on large worlds with high everything and frequent minerals? | ||||
Description | So ive been trying to generate a large long history world with high everything and frequent minerals, and it seems to always crash around year 500ish, it doesnt even give me a "no response" message, it just closes silently. | ||||
Steps To Reproduce | Generate a large world with high everything and frequent minerals on long history. | ||||
Tags | No tags attached. | ||||
|
Wasn't able to reproduce this. Tested on: Windows 10 (64 bit) Vanilla Dwarf Fortress (44.05 Windows 64 bit) Basic Worldgen: Large World Long History (550 years) Everything else on high Minerals frequent Result: Successful 550 year world generated. Things you could check: - Are you using 32 bit DF? Reaching memory limit (pretty low in Windows) results in instant crash usually. - Check your Windows settings. Towards the end, it took a long time for each year to tick up. If Windows is set in some kind of impatient mode, it might just give up. |
|
I ended up generating a 4th world and it didnt crash. You should probably generate multiple worlds.If you want to seriously test this. In any case, i'm definitely on the 64 bit version. Also if you are on the current windows 10 release and not on the insider builds, then you and i have two very different versions of windows 10. The insider program is essentially a beta testing program that people can opt into and they review your credentials then accept you. Then they start sending you pre-release beta versions of new builds. And it can be months before they ctually put out a public release and we go through essentially a new build every day so im on a "hot code" windows 10 version. Ill generate another world aswell, just to see if its purely random and i just got a bunch of bad seeds for some particular bug and ill post the results. :) i popped into event viewer aswell, here is the log though event viewer logs are rarely useful for debugging video games (they re really good for debugging business code tho :P ): Faulting application name: Dwarf Fortress.exe, version: 0.0.0.0, time stamp: 0x5a5bbc62 Faulting module name: Dwarf Fortress.exe, version: 0.0.0.0, time stamp: 0x5a5bbc62 Exception code: 0xc0000005 Fault offset: 0x000000000007fb8c Faulting process id: 0x474 Faulting application start time: 0x01d3a2bc6230e4d0 Faulting application path: C:\Users\bount_000\Desktop\Game Folder\Dwarf Fortress\df_44_05_win\Dwarf Fortress.exe Faulting module path: C:\Users\bount_000\Desktop\Game Folder\Dwarf Fortress\df_44_05_win\Dwarf Fortress.exe Report Id: 4aebc8d9-00b8-411b-9c20-32e163823dd9 Faulting package full name: Faulting package-relative application ID: Fault bucket 1271860971714260617, type 4 Event Name: APPCRASH Response: Not available Cab Id: 0 Problem signature: P1: Dwarf Fortress.exe P2: 0.0.0.0 P3: 5a5bbc62 P4: Dwarf Fortress.exe P5: 0.0.0.0 P6: 5a5bbc62 P7: c0000005 P8: 000000000007fb8c P9: P10: -------- 0xc0000005 is a null pointer exception By the way, not a memory error. SO either its a null pointer exception or some issue with permissions, but i was running it as administrator so permission shouldn't be a problem. -_o_- So when i download df from the website with the "windows" button, i do get the 64 bit version right? Because thats what i downloaded, and i was able to generate 3000 year old large worlds in previous 64 bit versions. |
|
As I said I would I generated a fifth world. It generated fine. So it must be purely random. I haven't made any changes to my system. so so far counting shonis world we have 3 not crashed and 3 crashed worlds |
|
I tested this issue a few times by myself too, yesterday: I generated 6 worlds under different settings. I am always using DF 0.44.7 for Windows (SDL) First try: 64Bit, MEDIUM ISLAND preset of advanced generation - crash at some point after 350 without any error message or 'report problem' window. Second try: 32Bit, MEDIUM ISLAND preset of advanced generation - crash at some point after 400 without any error message or 'report problem' window. Third try: 64Bit, default settings of normal world generation - crash at some point after 450 without a 'report problem' window. Fourth try: 32Bit, default settings of normal world generation - crash at some point after 450 without a 'report problem' window. In all four cases I can't exactly tell on which year it crashed, because I honestly didn't stare at the application the whole time. Fifth try: 64 Bit, MEDIUM ISLAND preset of advanced generation started via command line options. Finished with 1050 years without error. Sixth try: 64 Bit, MEDIUM ISLAND preset of advanced generation started via command line options. Manually interrupted at 729 years by myself because it was already pretty late and I don't want my PC to run through the whole night. But, all in all, it seemed that either I was somewhat lucky or the problem is not happening when I generate worlds via the CLI. I'll do more tests today. |
|
Did more tests yesterday, and results were pretty much exactly the same. Whenever I start creating a world by CLI, it seems to work. I managed to get one 2000 y/o world. Not using CLI led to a crash around year 400, all the time, with the same settings. |
|
Nice detective work! Hopefully someone sees this. Though i just generated a 1050 year old world with max everything except region size, savagery, and minrals (set to frequent and medium instead) in the standard world generator (not advanced) and it generated successfully in 44.10. |
|
I doubt that the CLI has anything to do with it - going through the main menu does very little setup work, and nothing I know of that would affect worldgen. If you can provide the exact seeds you used when it crashed, that would help Toady reproduce this. The seeds can be found in gamelog.txt. |
Date Modified | Username | Field | Change |
---|---|---|---|
2018-02-10 17:02 | untrustedlife | New Issue | |
2018-02-11 00:46 | Shonai_Dweller | Note Added: 0037781 | |
2018-02-11 07:50 | untrustedlife | Note Added: 0037782 | |
2018-02-11 07:53 | untrustedlife | Note Edited: 0037782 | |
2018-02-11 07:54 | untrustedlife | Note Edited: 0037782 | |
2018-02-11 07:57 | untrustedlife | Note Edited: 0037782 | |
2018-02-11 08:02 | untrustedlife | Note Edited: 0037782 | |
2018-02-11 08:06 | untrustedlife | Note Edited: 0037782 | |
2018-02-11 10:30 | untrustedlife | Note Added: 0037783 | |
2018-02-11 10:31 | untrustedlife | Note Edited: 0037783 | |
2018-03-22 05:40 | CrazyCatologist | Note Added: 0037993 | |
2018-03-22 05:42 | CrazyCatologist | Note Edited: 0037993 | |
2018-03-22 23:46 | CrazyCatologist | Note Added: 0038002 | |
2018-05-29 12:10 | untrustedlife | Note Added: 0038395 | |
2018-05-29 14:18 | lethosor | Note Added: 0038400 | |
2018-05-29 14:19 | lethosor | Assigned To | => lethosor |
2018-05-29 14:19 | lethosor | Status | new => feedback |