Live data from Hacker News

Show HN: A new kind of FPS aim trainer

openaim.pramit.gg

11–20 of 52 posts

Re: Show HN: A new kind of FPS aim trainer

#12

writeup on the rationale behind it: https://www.pramit.gg/post/i-made-an-aim-trainer

Even with "reduce motion" enabled the background of this page is constantly animating - it's so distracting I just gave up.

I didn't realize it could be distracting, sorry. It shouldn't happen now

Re: Show HN: A new kind of FPS aim trainer

#14
The aiming is very choppy because it feels like my cursor is "on rails", or stuck in some kind of grid. Not representative of any modern shooting game I've played. I think this would teach the wrong habits if someone were to use it as an aiming coach.

Re: Show HN: A new kind of FPS aim trainer

#16
post #14

The aiming is very choppy because it feels like my cursor is "on rails", or stuck in some kind of grid. Not representative of any modern shooting game I've played. I think this would teach the wrong habits if someone were to use it as an aiming coach.

Are you using it on chromium? It's the only browser I get raw mouse events from right now, so other browsers tend to be more choppy.

As for the habits, the point of this program is that it ultimately admits that far domain transfer for aim trainers is difficult regardless of what you do, so it tries to optimize in an absolute sense your motor control over your mouse. That way, regardless of scenario you'll have smoother and more efficient aim.

Re: Show HN: A new kind of FPS aim trainer

#17
post #8

So cool! but somehow the simulation does not work on firefox. the crosshair is just stuck in the middle

It should still work on firefox, but the experience will likely not be as good. Chromium is the only browser engine (as far as I know) that gives me access to raw, unaccelerated mouse events, so it's the smoothest experience.

Re: Show HN: A new kind of FPS aim trainer

#18
post #9

writeup on the rationale behind it: https://www.pramit.gg/post/i-made-an-aim-trainer

I enjoyed this. Does the same theory apply to a gamepad? My intuition says yes but curious what you think

I imagine it would generalize well to any fine motor control task with a defined tracking objective. The literature I reference mentions that human error when steering cars also roughly follows the same distribution.

Re: Show HN: A new kind of FPS aim trainer

#19

You might want to look into this page - it slows my PC to a crawl :-(

What browser are you on? Hardware acceleration? It shouldn't be very difficult to run, so I'm guessing some sort of a compatibility issue

> What browser are you on? Hardware acceleration? It shouldn't be very difficult to run, so I'm guessing some sort of a compatibility issue

Sorry, the linked page, not the actual app. It slows my PC to a crawl.

Re: Show HN: A new kind of FPS aim trainer

#20
post #8

So cool! but somehow the simulation does not work on firefox. the crosshair is just stuck in the middle

It should still work on firefox, but the experience will likely not be as good. Chromium is the only browser engine (as far as I know) that gives me access to raw, unaccelerated mouse events, so it's the smoothest experience.

I had the same issue. Linux + Firefox resulted in zero movement. Chrome worked fine.
Post reply on HN