Search for:
Update in my Movement Analysis
CS:GO

Update in my Movement Analysis

By

This gonna be a short post, i will let the images speak from themselves:

This graph is a Acceleration graph with a ak47 in hand from 0 velocity to 215(max velocity with ak)…in cs2(first image) and in csgo(second image).

Methodology:

Tools Used:

  • OCR (Optical Character Recognition) Google gemini, to take the velocity values
  • Steam’s in-built recorder to capture gameplay at 60 fps with cl_showpos 1
  • Frame extraction software to convert video files into individual frames

Test Environment:

CS2

CSGO(128 TICKS)

Map: aimbots

Console Commands:

cs2

  • cl_showpos 1
  • setang 0.000000 0.000000 0.000000
  • host_timescale 0.1
  • cl_draw_only_deathnotices 1
  • r_drawblankworld

csgo

  • cl_showpos 1
  • setang 0.000000 0.000000 0.000000
  • host_timescale 0.1
  • cl_drawhud 0

Movement Recording Protocol:

  • fps_max 400
  • Weapon: ak 47
  • Macro tool: AutoHotkey
  • Host_timescale: 0.1
  • Move until reach max speed of 215

Frame Timing:

Frame duration at host_timescale 0.1:

ef = (1 / 60) * 0.1 = 0.001667 seconds per frame

Meaning each frame represents 0.001667 seconds in real time.

Conclusion:

The part marked at red…wtf is that??

See the images and try to understand because i dunno wtf valve have done to the movement.

This is only a sneak peak, to get how hard is to do a analysis in the movement aspect compared with the spray…this isnt finished, and i will probably only finish when valve fixed the movement already.

This is bad.

I have nothing more to say.

MY SOCIALS:

https://x.com/eugenio8a8

18 Comments

  1. jonathan-the-man

    Just a heads up, for the non-technical reader the pictures don’t really speak for themselves, unfortunately.

  3. Mraz565

    You know it is about to be another good post, when the first thing you see if “WTF” is big red text.

    But really, wtf. Some PID tuning is needed.

    Edit:

    For those that might need help understanding the chart.
    https://imgur.com/a/FTT2QJQ

  4. Powerful_Seesaw_8927

    For people who didnt get why this is bad and i should added this in the text, sorry about that:

    one is consistent and follows a pattern, the other is a mess that gives a big acel, then a small one and then a big one again(this in the red zone)after that is fine…and in csgo is a perfect line

  5. Inflation_Artistic

    such a question, are these sharp changes in the values in the graph caused by your way of obtaining data or really such sharp movements?

  8. _Electric01

    I know we’re supposed to look at the WTF part of the graph, but I can’t help but notice (according to the numbers) how “slower” the player is with an AK now in CS2 compared to CS:GO, unless I’m not reading the graph in the way it should be.

    Just an interesting observation, not sure if that would change a lot the gameplay now because honestly I’ve never noticed it before, but yeah good finding OP, they should make the curve just as smooth as it looks in CSGO.

  9. AccordingAd1629

    Can you explain what is line number in this graph?

  10. Lord_Gaara

    how about deceleration? from 215 -> 0, is it the same inconcistency?

  12. wordswillneverhurtme

    My guess it has to do something with subtick

  13. Hyperus102

    What you are measuring is not really part of the player movement itself but part of the per frame prediction. Trying to claim “this is bad” with no insights of what is actually being simulated here is pointless.

    Try this again at host_timescale 1 and with 240 fps and you won’t find this same pattern.

    Either way, these velocities are not really that meaningful, because again, they don’t actually represent the game state.

  14. Extone_music

    I’d like to see the same graphs at a higher sample rate. It looks like there are reoccurring dips and spikes every ~50ms. They could be related to the “wtf” thing. Also, It might be a result of predictive movement or something. It would be interesting to see if this happens after a counter strafe too, or only when starting from static.

  17. STJRedstorm

    I think this would be better explained with video analysis. And multiple instances of each. You could be right in that the acceleration is wonky in cs2, but this just isn’t enough of an analysis to prove that.

  18. Top-Bag7848

    Yeah you may have to post some extra context with the edits

    Also idk if im right but basically the acceleration for CS2 seems to be instantly high, then rubber bands down twice to “simulate” gradual acceleration right?

    While csgo’s acceleration is just actually gradual and is a ramp

    Speaking of which, you should also check the Deceleration since it feels like youre sliding on ice even more than csgo.

Write A Comment