Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

166

Bug Reports & Support (modded) / Re: Games crashes once it loads a save

« on: February 12, 2022, 05:16:46 PM »

I skimmed through your error log -- the top frame is a native method call sun.misc.Unsafe.freeMemory. You could try using a different version of Java and seeing if that fixes the issue.
How to switch to Java 8, from the discord:

1. Go to your starsector install and rename the jre folder to jre7.
2. Download either of the following (both are the same JRE 8 v271) (Mac/Linux users look below for download link):
    - (72 MB .zip) https://drive.google.com/file/d/1CGRtnv5hIGCtcdlvZXFO0F58VXqYUo6x/view?usp=sharing
    - (48 MB .7z) https://drive.google.com/file/d/155Lk0ml9AUGp5NwtTZGpdu7e7Ehdyeth/view?usp=sharing
3. Extract that into your game folder so you have a folder named jre. You'll now have both jre (with java and jre7.
4. Make sure that your starsector/starsector-core/data/config/settings.json allowAnyJavaVersion is set to true. Then launch the game as normal.

The only other thing of note in the error log is this exception: Event: 19.827 Thread 0x0000000013559800 Threw 0x00000007e21a7c60 at C:\re\jdk7u79\2331\hotspot\src\share\vm\prims\jvm.cpp:1304, which after some research seems to be a PrivilegedActionException. This isn't the same thread that triggered the crash (I think it's the "main" game thread), but could suggest that the game is having trouble accessing something (a class? a file? not sure).

Can you upload the faulty save? You won't be able to upload it as an attachment here, but you can upload it to any file-sharing service (dropbox, google drive, etc.) and put a link here. I definitely won't be able to reproduce the bug with just your mod, unless it's an issue with your mod, but I doubt that given the contents of the error log and the fact that you aren't using any custom code.

167

Bug Reports & Support (modded) / Re: Games crashes once it loads a save

« on: February 12, 2022, 12:16:40 PM »

Interesting... since you don't have a lot of mods enabled, can you upload your save somewhere along with a zip of your custom mod? I'd like to see if this is reproducible on my end.

While you're at it, can you post the rest of your hs_err_pid?

168

Bug Reports & Support (modded) / Re: Games crashes once it loads a save

« on: February 12, 2022, 09:43:05 AM »

First thing to try is to delete your MagicLib folder and re-download the latest version.

Is there a similar crash log at the end of your starsector.log? Try scrolling to the very bottom of starsector.log immediately after the game crashes — do you see a similar error message with a bunch of “at ….” lines underneath? If so, post that here as well.

Finally, what null message are you getting when you try to save? After you get the save crash, look at the bottom of starsector.log and post that error message as well.

169

Mods / Re: [0.95.1a] Progressive S-Mods (0.8.6) — gain XP for ships, spend it on S-mods

« on: February 09, 2022, 02:17:28 PM »

True. I should be able to check for mod incompatibilities by spawning a temporary ship and seeing if any other hull mods get removed from the ship's variant.

As for why I even have the "show all" feature, it's a major convenience. I don't like having to flip back and forth between the refit screen and the build-in screen. Now if I could modify the build-in interface in the refit screen directly, then that might not be so much of an issue, but that's not possible without reflection (since no part of the refit screen is exposed in the API) + a complete code rewrite, which I'm unwilling to do.

170

Mods / Re: [0.95.1a] Progressive S-Mods (0.8.6) — gain XP for ships, spend it on S-mods

« on: February 09, 2022, 12:45:27 PM »

This mod makes hullmod interactions kind of funky. I have 2 examples.

First one is when a mod adds a custom built in hullmod that has incompatibilities, such as Monobolic Construction from DME. It is not compatible with Heavy Armor. However, if you simply "show all" in the build-in menu, it can still be selected and built in. This causes the hullmod conflict to always be there and always play the error noise, but it will still work and do what it's supposed to despite it not normally being possible.

Second one is with more modded hullmods, but still worth noting. "Minimal preparations" added by SCY is incompatible with Efficiency Overhaul, and when you build in one or the other and exit the menu, the other can not be built in. If you select both to be built in at the same time though, you can. Again, the "show all" button needs to be pressed, and the hullmod conflict won't go away, but you still can.
Edit: It seems you have to select Minimal Preparations first in order for it to work.

The reason this is happening is because custom hull mods cannot add themselves to a base game hull mod's incompatibility list. Take your SCY example. If you add efficiency overhaul first, minimal preparations is blocked from being added. But if you add minimal preparations first, you can still add efficiency overhaul. This is because if a mod wanted to make efficiency overhaul inapplicable when minimal preparations is present, it would have to modify -- i.e. replace -- the core game's efficiency overhaul script. This would of course make different mods that wanted to be incompatible with efficiency overhaul incompatible with each other, so it's not a good solution.

Instead, what modders do is this: they will check after ship creation whether the ship simultaneously has both incompatible mods. Then, they will remove one of the incompatible mods if they detect the incompatibility. This is done *after the ship is created*, which means that prior to ship creation there is actually no incompatibility.

I think I may have a way to somewhat fix this. I can check, after you try to build in a set of two or more hull mods, whether a test ship created just for checking the hull mods still has those mods on. If not, I can block the build-in process. I'll see if this works, but honestly if this is going to cause even more trouble I might just make it so you can only build in one hull mod at a time...

171

Mods / Re: [0.95.1a] Officer Extension (0.4.2)

« on: February 09, 2022, 12:20:20 PM »

Thanks for this mod.

I've noticed that after installing the mod that the game no longer starts paused when loading a save. I tested with no other mods installed and the same thing happens.

Thanks for finding this! I checked and you're right -- this mod is unsetting the game's paused state on the first frame, which is something I didn't even notice had changed. I've fixed this in dev.

172

Modding / Re: Script Failing Mid-use

« on: February 09, 2022, 10:05:03 AM »

Do you mean that alphaRed is 0? The static initializer is only called once, when the class itself is loaded. If this happens before you load your settings, then alphaRed will of course be 0. Did you try getting rid of the static assignment and just using SettingsYunru.yunruVariables.QS_ALPHA when you need it?

173

Mods / Re: [0.95.1a] Officer Extension (0.4.2)

« on: February 06, 2022, 09:56:14 AM »

Updated to 0.4.2:
- Add a field for persistent tags for filtering officers in the mod settings. Persistent tags will always be shown before any other custom-defined tags and will be shown even if you don't have any officers with that tag.
- Change all getMemory to getMemoryWithoutUpdate. This should fix crash bugs for officers with empty rank, post, name, etc.

Note that I don't have and have never used the secrets of the frontier mod (thanks Histidine for identifying the incompatibility), so this fix is untested (but should work based on the crash log I saw). If there are still issues with special officers will empty properties, please let me know by replying to this forum thread.

174

Bug Reports & Support (modded) / Re: Getting a Fatal: null error for a save that was working fine last night?

« on: February 05, 2022, 06:31:34 PM »

Unfortunately this did not resolve the issue, though I appreciate the effort.

Ah, I really thought that would be the issue. I can't think of many mods in your list that would have need to have their own custom market conditions. At this point I think the simplest option is just to use the save that does work and replay the last cycle.

If you're keen on debugging your broken save, you could upload it along with a zip of your mod folder. I am kind of curious... no promises though -- much of the base game code is obfuscated/unavailable, as well as some of the mods you're using, so I'm not able to guarantee that I'd be able to find the source of the crash. And even if I did, your save file might still be irreparable.

175

Mods / Re: [0.95.1a] The Star Federation v0.97.1 - Inspired by FTL (2/2/2022)

« on: February 05, 2022, 06:18:05 PM »

Maybe I was just doing the math based on the numbers you put in the original post. Still not sure if it is worth using as less range for a bit more damage tends to gut the utility of a weapon altering hullmod. Look at vanilla's HSA. This looks like it not only reduces range on beams but a significant for reduction. It looks to be an attempt to turn energy weapons into burst damage instead of pressure sustained but you only listed a +10% boost which results in a noticable DPS reduction. My math puts it at around -16% due to the minus 30% rate of fire. Perhaps trying to adjust ratios to it keeps approximately the same DPS but does it with the lower rate of fire.

It's good for projectile weapons with magazines where the sustained dps's limiting factor is ammo regeneration rather than fire rate. Some modded beam weapons have insane range too (i.e. gatling lance) so the -100 range can be fairly negligible.

176

Mods / Re: [0.95.1a] Nexerelin v0.10.4b "Dissonance" (hotfix 2022-01-30)

« on: February 05, 2022, 12:07:21 PM »

Possible ConcurrentModificationException in RebellionCondition:

Code

	
@Override
public void apply(String id) {
    if (refetchEventIfNeeded())
	market.getStability().modifyFlat(id, -1 * event.getStabilityPenalty(), 
	    StringHelper.getString("exerelin_marketConditions", "rebellion"));
		
    if (event == null)	/* refetch failed */
	market.removeSpecificCondition(this.getModId());
}

If the refetch fails post game load (where the game goes through and reapplies every market condition, see econPostSaveRestore in CoreLifecyclePluginImpl.java), calling market.removeSpecificCondition will cause a ConcurrentModificationException.

177

Bug Reports & Support (modded) / Re: Getting a Fatal: null error for a save that was working fine last night?

« on: February 05, 2022, 11:57:55 AM »

Looks like a faulty mod, but due to the nature of the stack trace, it's impossible to tell exactly which mod is causing it.

Likely scenario: a mod with a custom market condition (market conditions are things like decivilized subpopulation, abundant organics, mild climate, etc. etc.) is removing itself from the market's condition list in its plugin's unapply method. I went through your mod list and I think the only mods that might even consider having their own market conditions are Nexerelin and DIY planets.

DIY planets does not provide its source code so unfortunately I can't really say much more about it.

Nexerelin has public source code, so I went through all of its market condition plugins and found one that could cause a concurrent modification exception: in its RebellionCondition's apply method, it has this:

Code

if (event == null)	/* refetch failed */
	market.removeSpecificCondition(this.getModId());

If this code ever gets called, then yeah, there's going to be a ConcurrentModificationException.

Unfortunately I don't really think there's any solution to this. You may have to start a new save.
Here's one possible way to check if Nexerelin's the culprit (again, it's just a possibility -- it could be something else entirely). You can try extracting the source code from Nexerelin/jars/src.zip, going to sources/ExerelinCore/exerelin/campaign/econ/RebellionCondition.java, deleting the above two lines from its apply method, recompiling Nex into ExerelinCore.jar, and seeing if that fixes the issue.

178

Bug Reports & Support (modded) / Re: Getting a Fatal: null error for a save that was working fine last night?

« on: February 05, 2022, 12:09:49 AM »

According to the log you posted, your newest crash (the stuff at the bottom of the log) is caused by a different issue entirely: a ConcurrentModificationException. Can you try running the game again, except this time, immediately after the crash, go to starsector.log and only post the last several lines, starting with the error message?

If it’s the same ConcurrentModificationException I think the issue will be rather difficult to diagnose.

179

Mods / Re: Starsector Mod Manager: A new standalone utility application for Starsector Mods

« on: February 04, 2022, 11:19:47 PM »

This is from the modder’s perspective, not the player’s perspective. As the mod author, whenever you want to push out a new update, you still need to manually update your remote version file to reflect the updated version number (as well as the updated direct download link if you want to support Mod Manager).

Now that I’m thinking about it, I wonder why Nexerelin’s version checker even has a “version” field in its .version file spec. It could just as easily compare remote and local mod_info.json to check if a mod is out of date.

180

Bug Reports & Support (modded) / Re: Getting a Fatal: null error for a save that was working fine last night?

« on: February 04, 2022, 07:28:35 PM »

I went to check my own MagicLib folder and, indeed, there is no data/trails/trail_data.csv. Seems the file is now data/config/modFiles/magicTrail_data.csv.

It's possible that you have the MagicAutoTrails.java of an older version of MagicLib, but your csv files are from the latest version. Try deleting the entire MagicLib folder and downloading it again.