Added information: When I go to open the Structure / Rail Systems key, it takes a long, long time, during which the client freezes. It did this when I played on GenXNova, and it still does it on a local server. Even when all of the entities have been removed from the ship, opening that menu takes a long time.

-Confirmed-

Closing report, as problem has been solved together with a similar issue on the way.

Waiting for replies/details in Z34.

Closing report, got a feedback from development there were specific attempts to fix this, and as the data became outdated (due to a delay from our side) we close the report for now.

Screen shot of missing logic

Okay, great.
Closing the report then, as its nothing we can really fix without rewriting the whole logic system.

Thank you for sending the server-details.

Ahh I did not realize there were possibly timing complications smaller than a delay signal's capabilities. I added a row that the not-checking row feeds into before being counted by the final ANDs and now everything works! Thank you

Problem solved by increasing available ram.
Not the optimal solution but worked.

Looks like the blueprint has a different state of it then.
Server and location? going to visit it there then.

Yes, the ship has 12,000 rail mass enhancers.

It's not out of power on my server.

No Feedback from author.

Examined the circuit.

Nice command area, but it actually shows the problem:
The ship is out of power.
It shows ~12k rail mass enhancers being used.
Anything above the free-limit will not work anymore when the rail mass enhancers run out of power.

No, thank you, will ask for details if I cant find the problem, I have an idea what it could be.

Should I zip up the entity itself from the server or anything, or is there anything else I can provide to assist?

Well, on the copy on the testserver, all smaller things were moving properly.
Partially back and forth on the same spot, but still looked working.

Loaded tonight's patch server and client side, spawned in blueprint uploaded here, rails still frozen, no change.

I am not sure if there is some type of spam-protection to fight brute force attacks.
But a length 128 mix of upper and lowercase chars + numbers should already deliver a quite high entropy.
However, sure, if special characters could be made possible, that may help with security.

Yes, I certainly can.

There was a behaviour change in the logic a few releases ago, that made it more reliable (in terms of preserving signals), but also gave it the ability to preserve too many signals in large circuits and lockup the logic processing.
That would exactly explain the working state for a "fixed loading time".

-QA Testing-

Yeah, exactly its the LWJGL causing it, that part is out of our code range to my knowledge.
So we can look into it again after updating our LWJGL.

That "Compatibility Mode" alleviates the issue proves that this is not a fault with Windows, rather that Starmade (or one of it's libraries) is telling Windows that it understands and will handle DPI itself, and then failing to handle it correctly.

Multithread optimization of AABB update cause AABBs to go invalid when subsequent collisions where happening causing the gaps

Should have old behavior now. A not block will still not change state when R is used (because it sends TRUE which the not block turns to FALSE). However the not block will now send a FALSE signal when doing so

I removed all of the docked entities, deleted all of the logic blocks, and used copied the ship's block data onto a blank core's blueprint to completely wipe out all of the metadata associated with it. After wiping the metadata, the rails worked again. However, the rails stopped working again within a day.

Issue has been fixed long time ago.

After multiple attempts to examine this, it showed that this has to be some display/count issue with the number of references to the file.

No feedback from author and unable to reproduce.

Pushing forward -> Issue has been fixed by the launcher, as it now stores all files in its own directory, and not in a system specific storage.

After rereading this:
Its not support to run two game instances from one installation directory.

Issue has been fixed according to development.
Not seen it happen again in the last weeks.

We stopped supporting intel video cards alltogether, as they have a very wide range of functionality, but usually lack major components to properly handle 3d-rendering in the scale we need.

Will reject the report for now, as its more a windows behaviour problem, than actually a fault on the starmade side.

Ah, yeah, well the windows built in scaling options do not play well with most games.

Tried that behaviour a few times with large groups and watched them closely.
Could not reproduce the error,
unless I alt+f4 while the block operations are in progress,
and thus destroying my action-queue in mid of its processing.

While I do not have an OSX system,
@erthparadine and @Drakkart will be able to confirm a fix if needed.

Cannot really test this.
Did not hear of issues, assuming its fine now.

Fix confirmed then.
Great.

The font issue seems to have been resolved.

-QA Testing-

Added related thread, with more log samples if needed.
SMD-Thread

-QA Testing-

-QA Testing-

Did not happen for a long time now.
Assuming fix works.

Closing issue.

-QA Testing-

-QA Testing-

Can you reproduce this without using an exploit clock?
As any signal triggering more than usual updates (especially visual active updates) it will run into an anti-spam mode and if it does not recover from slower operation it will stop working completely.