The Great Mouse Strafe Tragedy of 2021

[GoneAway]

In doom, from vanilla all the way through mbf, you can only produce even strafe values. This is because the mousex value is multiplied by 2 when setting the side movement value. This matters because the difficulty of a glide often depends on your ability to produce certain small movement values. You can't produce a side movement value of 1 in vanilla doom.

 

Approximately 20 years ago, the side movement calculation was changed in prboom from "mousex * 2" to "mousex / 4" to allow users with high mouse sensitivity to still have reasonable mouse strafe. Unfortunately, that meant that prboom could produce a side movement value of 1 (and 3, 5, etc). Therefore, a user of prboom had an advantage or disadvantage (depending on the exact context) versus another player.

 

This difference was brought up on the discord recently - before then I didn't even know about it.

 

It's now fixed (only even values are possible) in prboom+ (for the next release), and in dsda-doom v0.12.1.

 

Now the question is...

 

Should old versions of these ports be forbidden for people that use mouse strafe?

 

Essentially, strafing with the mouse creates inputs that should only be achievable in a tas.

[dew]

I would probably support a ban going forward, it is inconsistent with the original behaviour after all. However probably not in the sense of "from now on..." and rather after a certain sunset period in which we can instruct players to switch, instead of outright banning their new contributions, because they haven't updated yet.

[4shockblast]

Just reiterating what I posted about this in Discord: as long as we get a 2.5.1.4 version of this, I think I'd be okay with enforcing usage of the new ports, since many people still can't switch to 2.5.1.5 or DSDA-Doom. This primarily gives advantage for glides and tricks like that, so it's not as serious of an issue as the 2-key strafe50, but having consistency with the original behavior is still preferable. Personally, having even strafe tics would be really useful for my glides, since I usually prefer to glide with vert mouse when I need consistent-ish 1 movements, and on the flipside, I have no way of creating consistent movements of 2s with either that or side strafing, which is something that helps a lot with glides in certain directions.

[liveincity]

Well, I wouldn't have known that if I hadn't read this post. LOL. I would support that for consistency with the original behavior. But I suspect that a lot of people would find it hard to suddenly switch from sourceports they are used to to a new one. This may take some time to adapt, so I think it's better to take it slow and not ban it all at once.

Edited February 14, 2021 by liveincity

[liveincity]

And I have some questions about this. Just like shorttics could make glide easier, if a glide needs a side movement value of 8, will the new side movement value (2468) make it easier (1/4 rather than 1/8) than the previous value (12345678)? And if a glide requires 9 movements, does it mean this glide cannot be done in dsda-doom v0.12.1 but can in previous versions like v0.10.0? Would this cause the old versions' demos playback error in v0.12.1(this kind of demo should be rare because I think there won't be many demos having a glide or something else happens to require an odd strafe value)?

[blob1024]

it's damn impressive how such an "outdated" and easy game like doom, can go so much technical everytime.

just to see if I understood the problem, should I directly set all even numers in mouse sensitivity while playing older versions to avoid having "this advantage"?

well... i'm an anti-glide/zero press player... I feel like its not how the game is supposed to be enjoyed, don't know if I should worry about it

and on top of this, can you backtrack mouse settings from an lmp and know what player had used?