I've been working on Sensfinder Arena, a CS2 tool designed to answer a simple question:

What sensitivity actually gives you the best first-movement accuracy?

This is not another "try 0.8 vs 1.0 and see which feels better" sensitivity tool.

Most sensitivity finders give you a few sensitivities, ask you to compare them, and then recommend whichever one you performed better with. The problem is that this doesn't really tell you where your optimal sensitivity is, it just tells you which of the tested values happened to perform better.

So what's different?

Sensfinder Arena tests your performance across multiple sensitivities and measures the angular error of your first mouse movement, before you have a chance to correct it.

The important part is what happens with that data afterwards.

Instead of simply comparing the tested sensitivities, the application uses mathematical modelling and curve fitting to model your performance as a function of sensitivity.

It analyzes the relationship between:

• sensitivity
• first-movement angular error
• overshooting vs. undershooting
• performance drift during the test

The model then estimates where your performance reaches its peak and provides an estimated optimal sensitivity and confidence interval.

So if you test:

0.7 → 0.8 → 0.9 → 1.0 → 1.1

the result isn't simply:

>"You did best at 0.9, therefore use 0.9."

Instead, the measured data is used to fit a mathematical curve and estimate where your actual optimum lies, including an indication of how confident the model is in that estimate.

The goal is to measure something that is otherwise mostly based on subjective feeling.

How the test works

The map uses controlled tests including:

• Flicks
• Micro-flicks
• 90–110° off-screen turns
• Vertical flicks
• Moving-target flicks
• A two- or three-second clock

The test takes roughly 35–60 minutes, depending on the selected run.

Your mouse, DPI and sensitivity should remain consistent throughout the test. Warm-up is built into the run.

One important limitation

The server currently supports 15 players total.

10 slots are occupied by the bots required for the test, which means only 5 people can run Sensfinder simultaneously.

So if the server is full, you'll need to wait for a slot to become available.

A note:

This is still an early version, so I'm especially interested in feedback, results and whether the estimated sensitivity actually translates into better performance in real gameplay.

Stop guessing your sensitivity. Measure it.

If it helped you find your sensitivity, consider buying me a coffee so I can keep the server alive and continue improving the current application. :)

Support the project here: http://buymeacoffee.com/xaizen

Counter-Strike has always been about competitive integrity. Right now, that integrity barely exists in Premier.

At higher ratings, cheating is so common that many experienced players simply avoid Valve's matchmaking altogether. Instead, they play FACEIT because it's the only place where they feel matches are actually competitive thanks to a functioning anti-cheat.

Think about how ridiculous that is.

Valve's flagship competitive mode isn't trusted by a huge portion of its own competitive community. The best players, streamers, and many semi-pros and pros don't grind Premier because they know what they're going to run into.

We've had years of complaints, countless videos, and endless reports about the lack of an effective anti-cheat. Meanwhile, players who just want fair games are the ones paying the price.

I think it's time for the community to send a clear message. If Valve continues to ignore one of the biggest problems in CS2, then updating our Steam reviews to reflect the current state of the game is a reasonable way to express that frustration. Reviews should represent the experience players are actually having today.

Don't get me wrong, I grew up playing Counter-Strike. It was the game I always came back to because every match felt like a test of skill. Winning or losing depended on how well you played.

But today, too often it feels like a casino. You queue up hoping you won't be matched against cheaters. Major updates seem more focused on releasing another case or skin than adding meaningful game modes or improving competitive integrity. Even opening cases is built around gambling mechanics while the core experience continues to suffer.