Testing & Results
I like to test keyboards by simply using them daily for all typical gaming PC activities and noting any irregularities or irritants. Of course, for gathering data points and comparisons certain tools can be useful. Microsoft’s Applied Sciences anti-ghosting page is an “oldie-but-goodie”, containing one of the best explanations of N-key rollover and keyboard ghosting along with a quick .aspx testing utility that anyone can use. We’ve used Aqua’S Key Test in the past as well. I like to brush up on my typing skills with typeracer.com, and I’ve taken the opportunity to race a few rounds with the Roccat Skeltr. Let’s take a look.
Motherboard: Gigabyte GA-Z97N-WIFI
System Memory: 2×4 GB Samsung DDR3 1866
Processor: Intel Core i5-4690K @ 4.6GHz
Video: Gigabyte GTX-970 Xtreme Gaming
Disk Drive 1: Samsung 850 EVO 250GB SSD
Disk Drive 2: Samsung 1TB 7200 RPM 3.5″ HDD
Enclosure: Corsair 380T
PSU: Fractal Design Integra M 450W Modular
Monitor: Acer 24″ 1920×1080 144Hz LCD
Operating System: Windows 10 Pro 64-bit
First up, it’s always good to know on a gaming keyboard if the keys you press will register. Many games require some pretty extensive finger gymnastics to press combinations of keys, so making sure your keyboard of choice will register your input accurately is the task of our first testing suite.
The above photo is a capture of the keyboard ghosting demo available on Microsoft’s Applied Sciences anti-ghosting page, using the Roccat Skeltr. Generally, I’ll just lay a palm across the keyboard and try to press as many keys as possible to see if they all register.
I mean, my hands aren’t huge, but I’m quite certain I can cover more than five keys with my palm. After capturing the above screenshot (and finding it difficult to believe a gaming keyboard stopped registering keys after ~4-5 presses or so), I went back to Roccat’s marketing page for the Skeltr and double-checked the specifications. Sure enough, no mention of N-key rollover or even any anti-ghosting capabilities anywhere.
Performing the test again to see exactly where keys were dropped, I simply started on home row. Pressing and holding A, S, D, and F, I wasn’t able to register a key press for G or any other letter on that side of the keyboard – however, J K L and ; registered (again, with no further letters on that side registering). It seems Roccat split the Skeltr into clusters, which should avoid any problems with fast typers. Sure, I’ve never played a game that required a full palm to be placed on the keyboard. However, for those games that use WASD for movement (…which is pretty much all of them), adding another function while crouching/sneaking and moving at an angle could run into limitations, which is completely unacceptable for a keyboard in this category in my opinion. Especially when keyboards half the price don’t experience the same issue. For reference, here are some more tests from past keyboards I’ve reviewed:
Aqua’S Key Test application verified my findings – the Roccat Skeltr simply can’t keep up with simultaneous keypresses like almost every other gaming keyboard in this price bracket.
Now, it’s not that much of an issue in actual use. I didn’t notice it – wouldn’t have guessed that it was a problem, even – while playing games on the Skeltr, and that’s the main area where a gaming keyboard should perform exactly as expected. I just feel people should know, for those that care about this capability, that it is absent from this particular gaming keyboard – I could imagine they may feel slighted, as it should be a safe assumption that a $160 keyboard has the ability to register more than four keys + a modifier.
Let’s take a moment to talk about the membrane keys. The Skeltr is expensive for a membrane keyboard; Roccat is hoping that the typing experience is good enough to give it a pass. I’m not sure “improved rubber dome technology” is enough to do it, but you don’t know until you try, right?
While I’m not sure the market is ready to pay mechanical switch prices for membrane keys, the typing experience on the Skeltr is surprisingly good. It certainly didn’t feel as “premium” as a mechanical switch, but after a short learning period typing on the Skeltr felt pretty good.
Good enough to surprise myself in a few rounds on typeracer.com. My first race sent me just over the 100wpm mark (102), where I immediately had to prove I wasn’t a robot. That’s a good start. Even better was passing the typing test at 99% accuracy and 118 words per minute… Keep in mind, I hover around 85-90 words per minute normally, and that’s with a mechanical keyboard (generally on Cherry MX Brown switches).
I thought it was a bit of a fluke, so I figured I’d better do one more race.
Perhaps I was more familiar with the quote than I thought – either way, 128 wpm seemed to solidify the trend. Part of the extra speed (for me) came from the improved accuracy. Since the Skeltr registered button presses at the bottom of the stroke I didn’t make as many errors by brushing lightly weighted keys.
I suppose it’s just the way I type (I bottom out mechanical switches too) – either way, the membrane keys on the Skeltr don’t seem to hold back a typist of my skill at least. I’m not sure it’s enough to completely forgive the potential for ghosting keys, but it’s a bit of a redeeming quality for the Skeltr.