When I first took a look at Cougar’s Revenger S gaming mouse, well, I shrugged. On first glance, it seemed to be another six-button gaming mouse with RGB LED lighting and a Pixart PMW3360 sensor. Then, I noticed something curious in the spec sheet: Cougar put down the Revenger S for a 2-KHz report rate. There’s only a few other mice on the market that have made that claim, and I was eager to see for myself if Cougar’s mouse was the real deal.
The Revenger S doesn’t really stand out on first blush. That could be a good thing, though, since it skips the garish styling we’ve seen on some other gaming mice. It has two large primary buttons, a clickable scroll wheel, two side buttons, and a sixth button on top. The main buttons accept inputs with the typical pair of Omron switches rated for 50 million clicks.
As is traditional for gaming mice, the sixth button is set up to toggle DPI presets by default. Uniquely, you can hold the DPI button for three seconds to enable “eSports mode” and lock in your resolution of choice. The scroll wheel is flanked by RGB LED accents, and the Cougar logo on the back of the mouse also lights up.
Richly-textured grips on the sides of the Revenger S do a lot for the mouse’s already-good handling. It has a rather low weight of 94 grams (3.3 oz), and an intuitive shape—if not a particularly ergonomic one. The Revenger S fits my hands well, which means it might end up being a little small for most adult American males. In particular, palm grippers should probably stay away unless they have very small hands indeed. Everyone else should find the Revenger S comfortable enough.
You won’t find any fancy gimmicks on the Revenger S—the RGB LED lighting is about as gimmicky as it gets. The weight isn’t adjustable, the sensor isn’t removable, and the shell is firmly fixed in place. Cougar does include a replacement set of PTFE pads for the bottom of the mouse should you wear out the ones that come attached. Instead of gimmicks, the Revenger S is about one thing: providing a solid mousing experience for serious gaming. To that end, it primarily serves as a vehicle for PixArt’s PMW3360 sensor.
Cougar’s UIX software is surprisingly lightweight and easy to use. It has an intuitive drag-and-drop interface for button assignment, and a very powerful macro editor that can capture and play back mouse motions, a pretty rare feature among software of this type. UIX can also control the Revenger S’ RGB LED lighting. The less said about the lighting the better, in my opinion. Given that it’s a mouse, your hand will be covering the lights most of the time anyway.
The characteristics of that sensor are well-understood by this point, but I’ll cover them anyway. This is an optical mouse sensor that uses an infrared LED to to light up the surface it’s tracking. The PMW3360’s resolution tops out at 12,000 DPI—well beyond the limits of usability. By default, the mouse doesn’t even come configured to use a resolution past 3200 DPI. Similarly excessive are the sensor’s acceleration and maximum speed specs: 50 g, and up to 250 inches per second.
Velocity test showing linear tracking out to nearly 2 meters per second.
Humans just don’t move their hands that fast. It’s good that the specs are as solid as they are, though, because it means the sensor’s performance is uncompromised in almost any situation. Cougar didn’t mess with PixArt’s alchemy, and so this mouse operates more or less exactly like other mice using the same sensor or its derivatives. I have nothing but nice words about it in that regard.
It’s over one thousand
Cougar claims it set this mouse apart by setting it up to support a 2-KHz polling rate. As a primer, USB devices don’t send interrupts like PS/2 devices. Instead, the host polls the device at a set interval for updates. Most USB devices use a polling rate of 125 Hz. However, polling rates that low can produce noticeable pointer judder when moving the mouse. The issue gets worse as display refresh rate climbs, and it’s far more noticeable when you’re controlling the viewport in a first-person game.
The possible benefits of a 2-KHz report rate. Image: Cougar
Almost all gaming mice support a 1-KHz report rate, and that’s a huge step up from 125 Hz. It’s still not perfect, though. As the Blur Busters blog points out at the link above, you can still see tiny micro-stutters in mouse movement on high-refresh displays, even when the mouse is set to a 1-KHz polling rate. In pursuit of the ultimate mousing experience, some folks have already overclocked their mice to 2 KHz and beyond. Not everyone wants to mess with mouse overclocking, though.
I had already been using the mouse at 1 KHz for more than a day to establish a baseline and confirm that Cougar didn’t “fix” what wasn’t broken. Indeed, the Revenger S feels absolutely identical to my Steelseries Rival 500 and the EpicGear Morpha X when in 1-KHz mode. In a word: excellent. After loading up Cougar’s UIX software, I proceeded to set the Revenger S to 2000 Hz and mess around with it. In 2-KHz mode, it feels… different.
Honestly, I’m loath to make any hard judgments about the way the mouse “feels” because my perception could be clouded by a sort of “nocebo” effect. That is to say that any effects I think I feel when changing to the 2-KHz report rate could simply be in my head. I unfortunately didn’t have time to get anyone over to do blind testing on this mouse. The only thing I’ll definitely say is that the mouse actually “feels” a bit more responsive to me in 1-KHz mode than the faster 2-KHz mode. Again, though—take this with a bucket of salt.
That impression left me really curious what was happening under the hood. Of course, if I’m honest, I was going to look into that anyway. Loading up the excellent MouseTester Reloaded, I went to work. Almost immediately I noticed some weird behavior. Have a look a this graph:
This is a graph of interval over time. On the X axis, we have time, and on the Y axis we have the interval between reports. As the line shows, the Revenger S is indeed reporting every 500 μs–on average. However, the reports are clustered right around 900μs and 100μs. What that means is that Windows is getting a position update, and then another update about 100μs later. After, there’s a (relatively) long pause of around 900μs. This isn’t typical mouse update reporting behavior. See below for an example of what we would usually expect to see.
All of the reports are clustered right around the 1-ms mark. There are variances in the report timing, due to a lot of factors primarily related to the polled nature of USB. Keep in mind we’re looking at a variance of only 200μs here, with a scant few outliers. With the Revenger S set to 2-KHz mode, I expected to see a similar graph but clustered around the 500μs mark instead of 1ms (1000μs). Obviously, that’s not what happened.
On a hunch, I used USBTreeView to check out the information that Windows’ USB driver had for me on the Revenger S. USBTreeView makes it fairly clear that the Revenger S is connected to the USB host at a 1-KHz polling rate. This is true across three separate systems, whether I use a USB 2.0 port, a USB 3.0 port, or even one of my new machine’s USB 3.1 ports. The Revenger S (like almost all USB input devices) connects in USB FullSpeed mode, better known as USB 1.1.
The Revenger S is running in FullSpeed mode.
USB FullSpeed is limited to a maximum of a 1-KHz polling rate. It’s actually possible to achieve higher rates using newer versions of USB. However, since the higher bandwidth of the later USB versions simply isn’t required for input devices, extremely few of them—if any; I don’t actually know of any—connect using anything but USB 1.1.
I asked Cougar about this behavior, and the company responded that it actually has a website set up just to address this very question for the Revenger S and Minos X5 mice. Cougar shows that you can connect an oscilloscope to the mouse sensor to see that it really is checking the position of the mouse two thousand times per second.
Another part of the log, showing the report interval and packet size.
The problem comes in that the mouse is only sending packets to the computer one thousand times per second, just like any other high-end gaming mouse. The reporting rate of 2 KHz is higher than the USB interface’s polling rate of 1 KHz, though. Yes, the Revenger S is technically reporting its position 2000 times per second, but it seems to be packing two position updates with extremely small deltas into each packet set to the USB host controller. That would explain the extremely short interval between every other position update we saw above. Further supporting this idea is the relatively large USB packet size of 64 bytes. Most USB input devices use a packet size of 8 bytes.
In the end, it looks like the Revenger S does actually report at 2 KHz, but the value of the data that it’s sending seems questionable to me since it’s not actually polling any faster than the average gaming mouse. It’s difficult to say if what the Revenger S does to reach that 2-KHz reporting rate offers any real practical benefit, as I certainly didn’t notice any major difference in performance. I’d advise anyone with a mouse that advertises a 2-KHz reporting rate to try the mouse at 1-KHz and see if they notice a difference between the two settings.
With that said, I still like the Revenger S. In 1-KHz mode, it still provides the same top-shelf sensor performance that you expect from any PMW3360-equipped mouse. Furthermore, the Revenger S’ grippy sides make it feel secure in my hand, and its light weight makes it easy to whip around for rapid action shooter titles. Its buttons are responsive and its software is more than competent.
Over on Amazon, you can find the Revenger S on sale for a few coins more than $43. If we ignore the 2-KHz polling rate claim, what we end up with is a mouse that has first-class tracking for not a lot of money. In fact, this mouse is awfully similar to the last mouse I reviewed. That HyperX pointer had a similar no-nonsense approach, but it sells for $70. It does have better lighting than Cougar’s offering, but I’d pick the information-dense UIX software over the stylish-but-sparse Ngenuity any day.
Potentially dubious claims of speedy report rates aside, the Revenger S is a solid mouse with no real faults. Sure, I’d like to have some more buttons to make use of that meaty macro editor, but that’s a very subjective complaint. For just $43, you’re getting a solid mouse with a top-shelf sensor, and that’s a fine place to be for Cougar. All told, I’m going to hand the Revenger S a TR Recommended award for its exceptional price-to-performance ratio.