Print

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

[MJWkr]

Hi there,

The Point Defense item appears to have a typo, indicating it gives a "0.0% chance" to do it's thing.

Thanks,

-MJWkr-

[Kyzrati]

Thank you for reporting it

Hm, this isn't a typo but a calculation error in the output. The value is dynamic, and shows you what the actual chance is based on a couple different internal numbers. Odd, since I know they work but for some reason the reported number is not showing up. For the Point Defense System it should display 8.0%.

I'll get that fixed.

[Kyzrati]

Okay, this gets stranger: It appears perfectly correctly for me. And looking at the code it should be correct, too.

Did you modify the game at all?

The only circumstances I can imagine that could cause this is if the internal game data was somehow different from what it's supposed to be. Maybe someone else has seen this? I wonder what else could cause it... All it does for that value is show the result of 80.0/10.0.

[MJWkr]

Huh..wierd.

No modding here, promise!

Maybe it's a WINE/Linux thing? No idea, just throwing out a variable.

-MJWkr-

[Kyzrati]

Shouldn't really have anything to do with Wine, but thanks for the extra variable just in case

One of the first things I can think of is some kind of memory corruption, but I'd think we'd get random crashes and other strange behavior if that were the case, and we're not. (I'm also prone to thinking memory corruption is behind any inexplicable bug where the associated code and data look fine, and I'm usually wrong )

I guess I'll have to leave this in the Bugs board until we get any more evidence for it. Obviously it happened to you, so it must be possible under some circumstance...

(Also, in future runs take a look at that and other Point Defense-type parts if you encounter them to see if it persists, or was a one-time thing. Thanks!)

[Kyzrati]

To add to this, it does appear to be a Wine bug of some sort. Another Linux player encountered the same thing with the same item, despite it always working correctly for myself and other Windows players.

Moreover, today he reported another instance of this same issue occurring with another item which uses the same floating point method to display its effect value! Screenshot attached for future reference--displaying a "-2.0%" instead of the correct 1.3%... Maybe at some point down the line I'll find a way to redo the calculations using some other method that won't have this strange issue.

[Kyzrati]

So it's been confirmed this was some oddity with Wine, and all five such situations in the game I've simply converted to use a different function that doesn't exhibit the same results. Come Alpha 15 Linux players using whatever Wine version doesn't handle these decimal displays properly should see the usual numbers. Thanks to everyone who contributed data!

[Widmo]

Congratulations on squashing this bug! What I find best about this story is a hard proof Cogmind really does have Linux support. This is the first commercial game where I witnessed developer fixing a bug specific to a platform they do not target with ports.

[Kyzrati]

It took a while, but with finally multiple players reporting it, and in a couple different places, the reason became more apparent, as did a potential solution. Glad to finally declare this one squashed

I did make a promise to Linux and Mac users that I will ensure Cogmind works as flawlessly as possible on their systems. It works so well in Wine by default that it would be a shame to leave little nagging bugs anywhere!