Review: Celluon Magic Cube laser keyboard

The Magic Cube is a projected keyboard that is perhaps a better concept than product

When looking at new technology and peripherals, there is the recurring question of “how small can we make this?” - everything needs to be made more compact, lighter, or an evolution of its predecessor.

Recently, we at Neowin were invited to take a look at one of the best embodiments of this thought process, with one of the smallest keyboards on the market. This keyboard is the Celluon Magic Cube, and it is arguably not a ‘real’ keyboard as such. The actual keyboard exists only as a projection, from the aforementioned ‘Magic Cube’. We received the Magic Cube for review via, and we'd like to thank them for their generosity in sending us this product.


The Cube, as it appears when turned on.

With the packaging being the first thing most consumers will see, Celluon has done well on the Magic Cube. The actual box is made of transparent plastic, putting the Magic Cube at the forefront. Below the Cube is a branded white box, containing everything else. It’s clearly branded, but there’s no indication of tackiness.

The white box contains all the associated items, such as the inevitable volley of manuals and reference texts. Connection to a PC happens by way of an included Mini USB cable, measuring around 50cm; when connected by mini-USB, the Magic Cube’s internal battery will charge to give the Cube a much greater element of portability.

The manuals are surprisingly high-quality, being printed on glossy paper that doesn’t always come with more expensive products. As you can see, actually getting a decent shot of the manual under lights proves somewhat difficult. One unadvertised feature within the manual is that the Magic Cube can be used as a mouse / trackpad, functionality that was somewhat unexpected, though could prove useful when the Cube is the only input device available. For the space-conscious, it adds a mouse and keyboard while using only one USB slot to do so.


Setting up the Celluon laser keyboard was extremely simple to do over USB, with the entire device being ‘plug-and-play’. No additional software needs to be installed, with the device managing its own compatibility. Initial connection took around twenty or thirty seconds, and then the Magic Cube can be customized. Users can alter the brightness of the keyboard, the touch sensitivity, and the feedback sound. Taking time during the setup process is worthwhile, for the Magic Cube is customizable to a degree most keyboards are not.

The Magic Cube is customizable to a degree most keyboards are not

The default sound is rather loud, though higher volume levels can be chosen, and it’s just as well that lower options can be set as well. Despite being able to adjust the sound volume, it is not possible to change the actual sound.

Connection to devices running Android and iOS is possible via Bluetooth by pairing the two together. We couldn’t test functionality with iOS, though it should work with fairly recent Android-based devices, having been tested on an Xperia Z and an Xperia J.

Consistent connection is quite bothersome, unfortunately. Both the Xperia Z and Xperia J the Magic Cube was tested with dropped connection several times. The J also connected without doing anything else, exposing a weakness in the manual’s coverage. The manual gave details on how to pair the Magic Cube with an Android 2.2 device, leaving us with out-dated instructions. It’s a manual which essentially ends on a cliff hanger, with the user having to figure out the rest of the instructions.

When the Xperia Z was connected successfully, the Magic Cube automatically opened a built-in application to manage its usage. Given the patchy connection process, we were surprised at how well the Magic Cube integrated with other devices.

With iOS devices, the only prerequisite mentioned is iOS 4.x or higher, so any iOS devices from the last few years should be capable of connecting. We checked around other outlets to confirm whether they had issues pairing other devices, but results were inconclusive as to a baseline for compatibility across the board.


It may appear difficult to measure the performance of a keyboard, though we feel it logical to assess input speed: with keyboards being primarily intended for this purpose, it's the broadest possible measurement. When we tested the Magic Cube we saw a considerable decrease in typing speed from a conventional keyboard, even after giving a period for adjustment to the Cube.

We couldn't consistently crack a speed above 50 words per minute (WPM) with any degree of accuracy, while on a conventional keyboard 100+ WPM was more than feasible. This was confirmed with testing across a handful of different 'typing test' websites, although there is a possibility of inconsistency here, as for many people a keyboard’s performance is subjective, and the old adage that ‘your mileage may vary’ will hold more truth here than anywhere else.

With no prior warning, the Cube would quickly repeat symbol entry

During testing, we observed interference that had not been immediately apparent. With no prior warning, the Cube would quickly repeat symbol entry, and unfortunately the only solutions we were able to find were to block the sensor projecting the keyboard, and ultimately restarting it.

Our initial assumption was that the Cube was acting up due to the large number of electrical appliances nearby, though this was ruled out when we moved it to different locations. The possibility was also raised that the front fascia could be blurred, though after thoroughly cleaning it down we remained unable to pin down the source of the issue.

The strong light coming from the left proves very detrimental to the keyboard's visibility.

Even marginally more shade helps significantly, though the pattern still makes legibility an issue.

With portability being the most obvious benefit of the Magic Cube, we took it around a variety of different locations to see just how it would do. The only place we really encountered difficulty was with the Cube placed on a textured surface, such as some table-tops. Other than that, the keyboard is capable of being projected onto a surprisingly wide variety of surfaces; while it was unusable due to distortion, we were able to project the keyboard onto a book, for instance.

Layout and Typing

Keyboard layouts vary across the world, with many people becoming totally familiar with only one layout. Many alterations are marketed as being ergonomic, such as the curved rows evident in some of Microsoft’s ‘Comfort’ line. When we started into our review of the Magic Cube, we had expected a conventional 88-key ANSI layout, as is common with compact American keyboards. As we began to use the Magic Cube we noticed several drastic changes, which may be off-putting to users.

The primary change is the relocation of punctuation keys. Rather than being clustered around the Enter key, they occupy a new top row, where they are slightly smaller than the other keys on the board. It is an immensely confusing change to have made, and as far as we can see there is no reason for it to have been made either.

It is an immensely confusing change to have the punctuation keys relocated

Key size on the keyboard is a particularly strange issue, for keys are not fully-formed: they cut off towards their tops, merging into the key in the above row. Within the punctuation row there simply is no top at all, and such a change is, once again, problematic for touch-typists, who likely do not press one part of a key repeatedly due to frequent hand movements.

We also spotted the possibility for incorrect entry due to individual key shapes. By pressing on the upper part of a key, it does not always register correctly, sometimes entering a letter from the row above. We came to the conclusion that the likelihood of typing errors was heightened significantly as a result.

The Magic Cube also includes a ‘Function’ key due to its small size, though it shuffles the arrow keys to make room for this. Most keyboards transform the arrows into a cluster of their own, while the Magic Cube chooses instead to place them at the end of the bottom row, almost like an afterthought. The reason for this is probably due to the unusual ‘Arrow’ key to the immediate right of the spacebar, which engages the mouse mode.

The keyboard does not provide much in the way of feedback

Amongst the manuals we did note information from Celluon themselves, who recommended using a ‘hunt-and-peck’ typing style for best effect. The suggestion is totally worthwhile, for the layout is so drastically removed from most conventional keyboards as to be frustrating for some users.

The keyboard does not provide much in the way of feedback, giving only a clicking sound, which as mentioned before is surprisingly loud and can be changed with shortcuts. These sounds are the only feedback provided, in fact, for the keyboard does not have any ‘feel’ beyond the surface it is projected onto.

Concentrating on the actual keyboard can be difficult to do for long periods of time, since it can look a little ‘fuzzy’. The quality isn’t bad, for everything is sharp enough, but there’s a strange fuzziness to the keyboard. We haven’t been able to verify if this is something a minority of users will notice or if it is widespread. Regardless it can be unpleasant to look at for long periods of time, which is bothersome due to the layout forcing users to double-check the layout on occasion.

The size of the keyboard also proves confusing, with the projected board being smaller than most conventional keyboards. However, the manual advises leaving the area between the Cube and the keyboard empty, to prevent interference. As a result, the Cube’s footprint is not as small as you may have expected; at least not vertically.

From the back of the Cube to the bottom of the keyboard measures 23cm. For the Das, it's 17cm.


While we at Neowin can’t imagine the Magic Cube was designed with gamers in mind, we couldn’t resist a chance to see how it performs at least. With the growing market of ‘gaming’ keyboards, it seemed like a basis for an interesting test if nothing else.

For fast-paced shooters, such as Team Fortress 2, we found the Magic Cube to be totally unsuited. Holding down a letter on the Magic Cube’s projected keyboard does not repeat the letter infinitely, so movement comes in the form of very short bursts. The curious operation of the Magic Cube could be somewhat comforting to typists, since it would stop repeated character entry. For gamers, it becomes virtually impossible to play.

For gamers, it becomes virtually impossible to play

Similarly, games like StarCraft II will not perform satisfactorily. This argument could be applied to most real-time strategy titles, for many include keyboard shortcuts. As outlined in the ‘Layout & Typing* section, the main downfall here is the lack of feedback, making it much more difficult to type accurately.

The bottom line is that when player speed is a major influence on how well you do in a game, the Magic Cube won’t stack up against competitors that can be had for less money. However, if you play suitable games, the Magic Cube can work quite well. Turn-based strategy games work perfectly, and there are plenty of them to choose from.

We tried the Magic Cube for a spot of Civilization V, and it is a prime example of a game that works well with the Magic Cube. The game’s UI places greater emphasis on mouse movements than it does on keyboard input, making the change of keyboard a fairly easy adjustment.


The Magic Cube is a strange peripheral, for on the surface it does a lot of things right. The concept is reasonable enough, and Celluon has evidently put significant work into making the Cube at its current level. Unfortunately, that level is not high enough to compete with the ‘tried-and-true’ market for physical keyboards.

While it is fun to tinker with the Magic Cube, it never feels as engaging as using a physical keyboard. The Magic Cube simply cannot offer the same level of engagement as a normal keyboard, which transmits valuable feedback to a user. It feels lifeless and artificial, while the ‘click’ sound only serves as a confirmation of this feeling.

That leaves a question of who this product is marketed to. It won’t work for a vast majority of gamers, it isn’t as engaging to use as an actual keyboard, and you can do better for less money. Yet we feel the Magic Cube could be viable for businesspeople, as the Cube is so much smaller than even a compact keyboard. While it is true that the product takes a significant period of adjustment, it could be used with some competency.

However, if we were going into the market tomorrow to buy a keyboard for the same price or less – it wouldn’t be this one. The concept works, but the reality does not. For the market value of the Magic Cube, it’s possible to get something more suited to most work.


Report a problem with article
Next Article

Facebook and Flickr photo integration dropped from Windows 8.1

Previous Article

Conflicting reports emerge about why Don Mattrick is leaving Microsoft

13 Comments - Add comment