vgstation13
vgstation13 copied to clipboard
remove ticker.minds
Noticed this when doing the roundstart stuff, I couldn't think of a reason why it would be necessary, since it's relatively easy to access the mind through a mob. While trivial, it does not need to exist, just adds to confusion, and by removing it, it's one less thing to worry about. It's used in three things: body archives (which I repurposed and works), scoreboard stats for survivors (it referenced mobs anyway, so rather than looping through minds, it loops through mobs), and targets (used H.mind to maintain its intent, where H is a human mob, which lessens the amount of checks).
It's just a not-well maintained list, and targets for objectives aren't even held in this list. This list just contains a list of minds, but why not just use player_list? You can just access mind just as readily.
What this does
- removes the existence of ticker.minds
Why it's good
- Less junk
Can't wait for something to break in unknowable ways.
PREPARE FOR UNFORESEEN CONSEQUENCES
okay but does this fix dead AI's not showing up on the roundend screen?
okay but does this fix dead AI's not showing up on the roundend screen?
bro... this has been around forever hasn't it? completely unrelated
THUMBS DOWN!!!!!!
can you show an actual performance increase from this? if not the risk to reward seems tilted
can you show an actual performance increase from this?
Assuredly not. However, it is worth cleaning up the code, if this PR doesn't break anything. I haven't looked at it closely yet and cannot claim to know whether or not this is the case.
I am not in favor of Adam deleting anything that he doesn't know what it's for.
I am not in favor of Adam deleting anything that he doesn't know what it's for.
The only thing I don't know understand is: Kanef code:
for(var/datum/mind/M2 in ticker.minds) //see above, but for ghosts
if(M2.current == M)
return TRUE
The logic here doesn't make any sense and I fixed it by just doing M.mind, referencing the mob, rather than the mind.
And: manifest_entries.Add(new /datum/stat/manifest_entry(M)) I didn't change this, it should work as intended, but all signs say it does nothing anyway.
kanef code
vs
adacovsk unforeseen circumstance code
which is the lesser evil? i'll do my part by both upvoting and downvoting
Unforeseen consequences adamcode is an experience. You get something unique on each gunked round caused by the unforeseen consequences. It might be the greater evil, but merge it anyway, at least the bugs will be funny to watch. If shit breaks we can always pull one from the collabs' book and revert if something went wrong.
okay but does this fix dead AI's not showing up on the roundend screen?
On second thought, it might?
test this at least once and have someone whose body was destroyed show up on the scoreboard
alternative to this PR: don't
test this at least once and have someone whose body was destroyed show up on the scoreboard
Does this work for you? Deletion of the antag mob also works. Malf will be tested too.
okay but does this fix dead AI's not showing up on the roundend screen?
On second thought, it might?
yeah this fixes it
@pedr69 if that works, it has nothing to do with this PR
Stop doing thing at random. If you want to improve performance, look at a profiler. Roundstart can lag horribly now. I am not sure if your previous PRs worked.
Stop doing thing at random. If you want to improve performance, look at a profiler. Roundstart can lag horribly now. I am not sure if your previous PRs worked.
this wasn't about performance, retard, it's about something that doesn't work and isn't used
Stop doing thing at random. If you want to improve performance, look at a profiler. Roundstart can lag horribly now. I am not sure if your previous PRs worked.
Also, roundstart works great now. No lights off or naked for 10+ seconds. Not only that, I no longer crash.
It definitely works. The problem is that it has a use case your PR:
- failed to account for
- failed to make a better replacement for
