ID:2546487
 
BYOND Version:513.1511
Operating System:Windows 10 Home 64-bit
Web Browser:Chrome 81.0.4044.17
Applies to:Dream Seeker
Status: Verified

A member of our crack team of bug testers has verified that this issue is reproducible, and has handed it off to the development team for investigation.
Descriptive Problem Summary:
Changing maptext_x shifts an outline filter(I haven't tested any others.)
Maptext_y does not.

Code Snippet (if applicable) to Reproduce Problem:
filters = filter(type="outline", size=1, color=rgb(0,0,0))
maptext_x = 10

Notice everything is shifted to the right by 10 pixels.

Expected Results:
For things to remain correctly positioned.
Actual Results:
They're not.
When does the problem NOT occur?
When you don't use maptext_x.

Did the problem NOT occur in any earlier versions? If so, what was the last version that worked? (Visit http://www.byond.com/download/build to download old versions for testing.)

Workarounds:
Don't use maptext_x.
Lummox JR changed status to 'Unverified'
I tested this but could not confirm it in my debug project. There must be some element missing. I'll need you to send me a test project to reproduce the issue.
The Google Drive link isn't working. I don't think that's the right link for me to download the project. Or perhaps you could try DropBox?
To clarify the situation, I can reach the link you provided, but when I try to download the file I end up in a redirect loop. The last time I saw something like this, also on GD, was because the user had posted a link that was valid only for them. Since I'm sure Google Drive does not necessarily make this clear, it'd be simpler to use some other site.
In response to Lummox JR
I was able to download it. Enter the link, there'll be an arrow icon on the right upper-corner of the page. At least for me:

Thanks to NSBR forwarding the file I was able to run it, but didn't see the same problem. So far we've determined that the issue only shows up on Nvidia cards.
Lummox JR changed status to 'Verified'
As things currently stand, I have no idea why this bug is happening, but it's very strange. As far as any of us who examined this issue in the DreamCrafters Discord have been able to tell, this is caused by a bug in Nvidia cards that's mishandling the outline shader. All the other logic appears to be correct.

On Nvidia cards, an "overflow" area determined by the original icon bounds and the maptext bounds gets 25% of the outline color applied, even though there's nothing in either your code or the shader that explains why it would be 25%.

If maptext_y is positive in your test case, the overflow vanishes. (I believe this is true of any case where the maptext extends north of the original icon's visual bounds, because making the original icon shorter on the Y axis has the same effect.) If maptext_y is negative, the overflow area where the shader produces wrong results expands. The fact that the overflow disappears when expanding upward is totally bizarre, and at odds with all of the internal logic.

Although there is no explanation whatsoever for why this should happen, I've made a subtle change to the outline shader. I can only hope that change will solve the issue, but I can't test it myself because I have a Radeon card.