This right here is a prototype of the AtomPalm Hydrogen, the best gaming mouse on the market, Theoretically.

 Although the hardware is actually fairly standard for a high-end gaming mouse, it has a Pixart PMW3360sensor and OMRON switches, everything about it has been tuned to offer the lowest possible response time between you moving the mouse IRL and the cursor moving in-game. So is this the new competitive edge for gamers, All right, in a perfect world, the movements that you make in real life with the mouse would be 100% accurately represented in the computer But in practice, there are a bunch of things that can lead to imperfect mouse tracking, and ultimately, you missing that sick headshot by a couple of pixels. So, what can you do to have the best possible chances? First of all, disable your mouse acceleration. Then, most games also have an option for raw mouse input, so you should enable that too. While I am talking about easy ways to up your game, disable Motion Blur. This won't affect your mouse, just motion blur. Why does anybody use it? The next problem is sensor spin-out. This occurs when you move the mouse faster than the sensor is able to detect, and generally, this results in the mouse failing horribly, kinda like someone running on a treadmill that's going too fast. On office mice or cheap generic gaming mice, this can be an issue, but basically every real gaming mouse these days is capable of tracking anywhere from 30 to 50 Gs of acceleration. Is that a lot? So I found a fun article by H. Nagasaki from 1989 called "Asymmetric velocity and acceleration profiles of human arm movements," and in those tests, the maximum arm acceleration was 11.4 Gs, and referencing other less-reputable-sounding studies, I haven't found anything above 20. So basically, there's no chance of you physically outrunning a properly-tuned modern gaming mouse. What can introduce errors, though, are sensor flaws. For instance, the PMW3310 has a problem where, if you lift the mouse and then set it down, it can stop tracking properly for a brief time, undoing all that work you put into lining up your sick flick shot.

Fortunately, finding a sensor without those kinds of game-breaking flaws these days is actually pretty easy. sensor FYI has a list of sensors and mice that are considered flawless. Once you have a sensor that isn't doing anything weird, make sure then that you have the lightest possible mouse. A light mouse has two benefits. For light mice, ergonomics kinda matter less, since picking it up so easy, and they also have less mass, which means less inertia, so quick moves, and adjustments should be easier. Now, at 45 grams, the Hydrogen is not actually the lightest mouse. That goes for the Zaunkoenig MK1, but the MK1 lacks a scroll wheel or even side buttons. So the lightest properly-featured mouse is the Hydrogen, bringing us nicely to the party piece of the AtomPalm Hydrogen: the 8,000 Hertz polling rate. 8,000 Hertz polling rate, to really appreciate why 8,000 Hertz kicks butt, let's start off a bit lower. 

In the AtomPalm software, you can change the polling rate to, well, anything you'd like. 125 Hertz is kinda like the standard, so if you have an office mouse at work or something like that, chances are that's its polling rate. And with a maximum delay of just eight milliseconds between when you physically move the mouse and when the computer gets that information, you can definitely tell when a mouse is polling at 125 Hertz versus 1 Hertz. But you can also tell the difference between 125 and something higher. So let's flip to, what, say 1,000? - [Crew Member] A thousand, yeah. - A thousand, all right. This is what I would expect from a high-performance gaming mouse, and it really does feel more responsive, everything from snapping around to targets to even just, you know, clicking folders and opening them and closing them. It's just more responsive. But the jump from 125 Hertz to 1,000 Hertz is a seven-millisecond difference, whereas going from 1,000 to 8,000 Hertz, even though I am going eight times faster again, it's just a .75-millisecond difference.

You can try it for yourself, but... For the mouse, anyway, we need some more concrete evidence, so let's do a little bit of math. With a 1,000 Hertz polling rate on your mouse, the maximum possible latency between moving your hand and the computer receiving a signal that it has moved is one millisecond, or 1,000 microseconds, with an average latency of about 500 microseconds. With an 8,000 Hertz mouse, the maximum latency is now just 125 microseconds, and the average is 62.5. AtomPalm says that it's this average response time that is so important since with a thousand-Hertz mouse you will get used to that average 500-microsecond delay, while over time, you will actually get muscle memory for the 8,000 Hertz mouse and to move it quicker.

Since the most leet gamer move is, of course, the 360 no scope, let's start there. After studying epic gamer moments, I found that the average 360 no scope happens in about .4 seconds, and by plugging that an average mouse and game settings into the grand unified theory of leet, I found that the maximum error on paper with a 1,000 Hertz mouse during a 360 no scope is 19.2 pixels, while the 8,000 Hertz mouse is just 2.4. Damn. With that said, most people are not, no offense, 360 no scoping all the time, so for a more realistic movement of 20 centimeters over a span of one second, I am looking at about an error of 6.3 pixels, or 1.77 millimeters, which looks a little something like... Oh yeah, that's... That's not a lot. But, at 8,000 Hertz, it's less than half as much. Anyway, this potential for error is not taking into account adjustments on the fly as you're making your shot, 'cause no one closes their eyes and makes a one-second shot on muscle memory alone. You see where your aim is going, and you correct your movement with the muscle memory for 1,000 Hertz. Now, people were insanely good at "Quake" with 125 Hertz mice, so theoretical inaccuracy doesn't always translate to missing the shot in real life

The gaps happen when the polling rate and the display mismatch, leading to the highest possible latency and micro-stuttering. What I'm surprised to see is that there is measurable stuttering at 1,000 Hertz, and even 2,000 and 4,000, but at 8,000 it is almost perfectly spaced. To be honest with you guys, I did not expect to see any real-world difference here, and only performed the test because Blur Busters thought that 1,000 Hertz might show stuttering at refresh rates of 240 Hertz and higher. I really was not expecting the differences to be, ah yes, this clear. Now, it is possible to overclock a small number of gaming mice to 8,000 Hertz, that don't have support for it out of the box. But for most mice, the maximum1,000 Hertz polling rate is hardcoded into the drivers, which means, unfortunately, that the theAtomPalm Hydrogen is, well, unfortunately for all the other mouse makers, I suppose, legitimately the most accurate mouse on the market. Although, being accurate is meaningless without also reducing click latency. Unfortunately, I don't have a reliable way to test the debounce and latency of a mouse yet. But AtomPalm wrote a research paper about how they have the lowest theoretically possible click latency by doing some fancy debounce stuff in the microcontroller. If you feel like checking their work, you can go read that. The final hurdle though, of course, for AtomPalm is reliability. As far as that goes, well, only time will tell. I haven't even been able to test a final production unit yet, so I can't properly comment on that, or even the build quality, 'cause it is 3D printed. But Razor has announced plans to release an 8,000 Hertz mouse if you need a little bit more reliability. So anyway, this is it, the AtomPalm Hydrogen, theoretically the most accurate gaming mouse. And although chances are that you are not cool or skilled enough to experience any of the benefits besides the placebo feel-goodness of having the best, most legit mouse, man, then, you know, bummer. But hey, if you're an elite gamer, you should consider picking it up for just a little under $100. People should just get the G305. It's light, the battery lasts basically forever, and it's relatively cheap.