Unlike the M64 Pro, which takes clear inspiration from the Xtrfy M4, the M68 Pro is its own thing, with a less daring yet still original ambidextrous shape. Much like on the M8 Wireless, the ultra-low front once again is of note, which also has the charging port be placed to the side of the mouse instead of the front. Again similarly to the M8 Wireless, the M68 Pro is a solid-shell design, without holes or an open bottom for weight-reduction purposes. At 55 g, the M68 Pro manages to be a bit lighter than direct competitors such as the Logitech G Pro X Superlight 2 or Razer Viper V2 Pro, albeit while being a bit smaller than either, too. As far as build quality goes, the M68 Pro is on the same high level as the M64 Pro: No rattle when shaking, only minor creaking but no flexing when applying lateral pressure, and actuating the side buttons by pressing below them is impossible regardless of force used.

Even though the components used on the M68 Pro are no different from the M64 Pro, the shell of course is, and therefore the button design. Mechanical switches from Huano, more specifically the ones with a transparent blue shell and pink plunger, see use on the M68 Pro, and much like on the M64 Pro, their implementation isn't as good as it could be. Both pre and post-travel are moderate, resulting in a slightly dull response, though the buttons are firm at least, as lateral button movement is low even when provoked. Even though the same Huano switches (yellow plunger) are used for the side buttons, these don't fare as well as the side buttons of the M64 Pro. Pre-travel is still largely under control on the forward button, but the back button suffers from noticeable pre-travel, coupled with a decent bit of post-travel and uneven actuation. Even more puzzlingly, despite being the same F-Switch brown/green core encoder as on the M64 Pro, the scroll wheel on the M68 Pro suffers from a rather irritating grinding or ticking sound when scrolling up, which is consistent across two samples. Hence, I'm inclined to believe this isn't a one-off, though it indeed seems odd how the same encoder can behave so differently. Otherwise, they're the same, with identically low noise levels and slightly above average tactility. Still, for a $139.00 mouse, I can't help but feel that this encoder is somewhat undeserving. As for the feet, these are very similar to the ones used on previous Xtrfy releases, still made of pure PTFE, and glide very well. As always, a set of replacement feet is included in the box, which is most handy.

Arguably the greatest effort has been put into getting performance as good as possible on the M68 Pro, and this work has paid off. With the included dongle, the M68 Pro is turned into a USB high-speed device, allowing for native support of polling rates above 1000 Hz. Right out of the gate, the degree of control the M68 Pro gives users about sensor settings is notable. At polling rates of 1000 Hz and under, the user can switch between three different sensor run modes: Standard Gaming Mode, Low Power Mode, and Pro Gaming Mode. Whereas Standard Gaming Mode and Low Power Mode are largely similar in terms of how the sensor framerate scales relative to a given velocity, with the most significant, yet in practice irrelevant difference being the maximum possible tracking speed, Pro Gaming Mode effectively makes the sensor behave as if it were wired, with major ramifications for performance. In Standard Gaming and Low Power Mode, the M68 Pro already achieves parity with the Logitech G403 (control subject) at 1000 Hz, provided MotionSync is disabled. Upon switching to Pro Gaming Mode, the M68 Pro now comes ahead of the G403 by roughly 1 ms, which is on par or better than what other mice muster at 4000 Hz. In fact, by switching to 4000 Hz, which defaults to Pro Gaming Mode anyway, motion delay isn't lowered even further. Hence, as long as Pro Gaming Mode is active and MotionSync disabled, motion delay will be on the same, downright excellent level. The only exception to this is 8000 Hz, for which a slightly larger buffer for retries has been implemented, to improve polling stability, resulting in slightly worse motion delay of 0.5 ms relative to the G403. The only downside is that Pro Gaming Mode does reduce battery life quite a bit, though it'll still be higher at 1000 Hz than at 4000 Hz of course. To put it more simply: On the M68 Pro, one can enjoy the responsiveness of 4000 Hz even at 1000 or 500 Hz, owing to smart firmware design.

What's more, this not only applies to wireless motion delay, but to wireless click latency as well. As long as the M68 Pro is used in wireless operation, click latency will be at a stunningly low average of 0.3 ms, which is even lower than the already most impressive numbers posted by the Finalmouse UltralightX. In addition to being outstandingly low, click latency on the M68 Pro is also highly consistent, as reflected by a standard deviation of just 0.02 ms. Even with a high degree of sensor motion at 8000 Hz, missed polls for clicks are incredibly rare on the M68 Pro, leading to a remarkably consistent experience. As of now, the M68 Pro has the lowest click latency of any wireless mouse on the market, and also the most consistent. Neither polling rate nor debounce time, the latter setting of which in fact does nothing and therefore may be considered a legacy option, have any effect on click latency, and since a fixed value of defer-type debounce is applied upon lift-off, slam-clicking, which describes inadvertent button actuation upon resetting the mouse, is not present on the M68 Pro, either. In short, what the M68 Pro does in this regard is very close to perfection.

Though the firmware is shared between the M64 Pro and M68 Pro, their physical makeup of course is different, and it appears that on the M68 Pro, sensor height is slightly closer to the ideal, as CPI deviation ends up being significantly lower at just +1.3%. Hence, the only thing one could criticize in terms of performance is that at polling rates of 1000 Hz and under, bouts of elevated noise occasionally show up, with no clear pattern or logic to their occurrence. That said, tracking is wholly unaffected by this, and the degree to which these bouts are pronounced is low to where one may consider them negligible. Furthermore, at both 4000 and 8000 Hz, polling is perfectly stable, which is particularly noteworthy for 8000 Hz, where even Razer tends to struggle on their wireless mice.

As mentioned, the sensor run modes not only affect performance but battery life as well. Using Standard Gaming Mode at 1000 Hz, CHERRY XTRFY cites a battery life of up to 90 hours. Given the battery capacity of 350 mAh, this is a plausible number, but verifying it is difficult if not impossible, as the only sort of battery indicator present on the M68 Pro is an LED lighting up in different colors, depending on the charge state segmented into four quarters. While the no-software approach has a lot of benefits, the lack of a precise battery indicator no doubt is among the downsides. It is fair to assume that both Pro Gaming Mode and 4000 as well as 8000 Hz lower battery life, but by how much is anyone's guess. Charging at least is a hassle-free affair on the M68 Pro: Since the M68 Pro also comes with a dedicated dongle extension cable, whose shielding is improved to ensure signal integrity even at 8000 Hz, the charging cable can be used exclusively for charging, and is flexible enough to allow playing while charging without feeling overly restricted. At around 0.25 A, the M68 Pro doesn't break any charging speed records, but isn't too slow, either.

As one would expect, there isn't much between the M64 Pro and M68 Pro, given that outside of shape, these are the same mice. Still, both side buttons and scroll wheel end up being not quite up to par with the standard set by the M64 Pro, yet the pricing of course is the same at $139.00. As such, I find it appropriate to take the M68 Pro down a notch, which is why it gets our Recommended award instead of the Editor's Choice.