As you see on this site from the top of the web list (click here) the idea of quantum computing still has the major problem of decoherence; while they can change the bits to read the memory, the bit is changed so much by reading the information to retrieve it the memory stored is lost. So even at present with all the hoo ha about this type of machine, it's more who than hurrah, and unproven science, it's of worth except it's presently unviable!
.
In usual chips there is just an on off of the transistor and is considered to be optimal because it's simple, so base 2 with just the on and off is how most computers are made. It's considered more economical, but it's only because of the way chips are made. With more ways to move, say a continuous circle ("pick a number, pick 100 if you like", you'll always win at 21!") with a higher number of optimals than just on and off, you could store higher amounts of information.
.
Hardware that uses how an atom can rotate continuously in space with changes in applied voltage has been invented, a wire made of two types of alternating atoms; the sensor atoms, and the atoms that continuously rotate to store precious memories of the beach and the heat. A second set of wires intersect with each sensor atom to read them. In the patent the sensor atoms are unlike the bit atoms because to be readable the bits have to be more easy to change, and if so the information stored could be deleted if the atoms were moved by the voltage going through to also read them. The voltage going through the sensor atoms themselves is of low enough power so they leave the atoms with the stored data unchanged, just passing near enough to sense what is encoded in the position of the 360 degree coordinate of the alternating storage atoms beside them. To see that this is possible think of a sofa and the tiny beam of light that you see it by, not much changing the sofa. The idea is to make use of this small change to read the data and the continuous rotation of atoms to store much more data with each atom. this is sort of like a wrench that has both a continuous gear that opens and closes the clamp on the bolt and a second component that locks the wrench in place once it's adjusted. By way of the continual rotation and the lock, the information of thed general position of the wrench can then be read by a bolt without risk of the wrench losing grip, the risk of the decoherence would be like the loss of clamp without the lock on the wrench. The inertia and mass of the nonsensor atom is actually more than one atom for inertia so it resists the more slight motion of the sensor atom to read the position of the nonsensor atom(s) so it's nonvolatile and readable too by the optimal inertia or friction to store the bits so it both changes and stores the change too. They've thought of this idea with electrons where each angle of the electrons spin would be a bit, but atoms like the sofa are more continuous so with continuous applied voltage, instead of just having perhaps four bits of the electrons, or just two bits, you have 50 or 100 or even 50,000 for a chip with each bit in turn the size of an atomic element or small molecule (or even protons welded via the alternation of the N and S pole of each so much smaller and faster, by Chin's tunneling experiments here's my own Faster Than Light page about my belief about Chin's experiments) and because atoms or molecules are more continuous than electrons, this could be like with radio stations where they add more bandwidth to stations by compression so there are three hundred stations instead of just 20 on the radio. It might seem there are just two electric charges, + and - there are just two bits involved. But there are actually billions of possible TV stations. The electromagnetic field of light is actually a continnum. Because spacetime and voltage is continuous like the wavelength of light in relativity there may eventually be 50,000 bits possible. The speed of light is quantisized and solid, but the wavelength is continuous, because to conserve energy, when part of the field becomes more solid, some other nearby component must be more flowing, if we find these continuous components and good ways to control and measure them, computation wins for us all and you and me, even that 57th grade marm in college!
.
By making the reading and writing method of the optimal sensitivity by changing the elements just so, a computer like this would read and write at the same time at higher speed than even me, like that described in Popular Science in the August 1999 issue, by way of science called magnetic random access memory or "ma-gram", and grand-ma may well appreciate it! Grand-ma would replace not just the RAM but also other media of computers. Magram uses small wires with N S poles and their own sensor, this goes at 90 degrees so it stores the bit read by the sensors, but it's just base two. PM says it could be far faster than hard drives. Using just atoms instead of small magnets would allow much higher speeds because they would be smaller. Once the "store" atoms are at the angle to store the data, the angle is stored (saved) by making the field go from more fluid to solid by changes of the voltage, ect, in a rapid phase change, sort of like from fluid to solid somewhat. This or just the inertia of the nonsensor saves the information so when the machine is off no power is needed to restart. By reading the data from the sensor wires, the bits are intact even if the voltage of the main wire is unchanged because the voltage to read the sensors is lower than the voltage needed to change the main wires.
.
Another related invention that has been devised is just using changes in voltage or wavelength that are changed and retained and read by way of sensors. Since this type of change like the rotation of the atoms is continuous, non quantum, and readable, here too information can be stored.
.
This would make it so if you had the same number of atoms to make a 50 G hard drive if it was atomic, with each atom storing 50 times as much data, the same memory chip would be 2500 Gs, nonvolatile, cheaper than conventional RAM, much higher in speed, and with lower power consumption and heat loss.

On sites like the Wikipedia Timeline of Quantum Computing
you see all this progress in quantum computing NEW! IMPROVED! but without the problem of coherence solved, progress may need to continue indefinitely of this sort. If quantum computing is so great, why are so many advances being made in other fields like soap opera editing computers or weather control ? While it may be of worth eventually, by using a low power sensor and the non sensor bits, coherence is solved by a much simpler motif, and no complex nonlinear codes are needed to read it, base 25 or base 50 is simple math.
.