So for those who follow technology in even the slightest way, you’ll know it’s that time of year when CES happens. There’s been some really cool stuff on display this year, a lot of it focused around Ultra-HD and 4K TVs (yes, there is a difference), however, there is something that grabbed my attention…
Project Christine from Razer. This is essentially a modular computer that allows users to upgrade small parts of their system with great ease. True that this is already possible with standard PCs, but Razer wants to allow users to completely customize a high-end gaming PC without any knowledge of the insides of a computer. Each component is self contained and requires nothing more than plugging in a new module. Each module is also water-cooled…which is awesome!
This should (as it did me) remind you of another modular concept that is forming…
It would appear, and although this is only a small fragment of the technology market, that there is now a place for modular systems that allow the user to upgrade components in their devices using these self-contained modules and a frame. This simple concept is designed to allow people to use their devices for longer, since they don’t have to worry about their technology being out of date (this is certainly Phoneblox‘s idea, I’m not too sure if this is Razer’s goal too).
Whilst this is quite a nice idea, it’s never going to last and from an electronics point of view, it isn’t actually possible. Why? There’s several reasons but let’s focus on the main one: the inter-connections between each module. If we take the modular device without it’s modules, we are left with a simple frame that contains nothing more than connections to each module, and maybe a bit of circuitry (also a water-cooling unit in the case of Project Christine). Those connections are what’s are limiting the speed and efficiency of this concept.
Project Christine connects all the modules together via a PCI-Express bus, which although is a pretty fast bus, eventually we will have components that require a faster bus, then what’s Razer going to do? They’ll release a new frame with faster inter-modular connections, so you’ll have to buy new components, why? There’s no point having a fast CPU on a fast bus if the RAM is still slow, or the hard drive is still slow. Every time a faster component is created, other components have to be designed to keep up; if you are going to get a faster CPU for your modular device, you’ll need a faster infrastructure, faster RAM etc otherwise there’s no point in keeping the one faster component.
To put this into a simple analogy, imagine taking your the most basic modern computer, Intel Core 2 Duo, 1GB Ram and an 80GB Hard drive, now that’s a slow example! But let’s remove the 1GB of Ram and replace it with 256MB, which was quite standard not too long ago, you’re computer is now dead slow, the software isn’t designed to run on that little memory, it needs bigger and faster and the only way to achieve that is to have the whole system to be at the same speed. See what I mean?
I could ramble on for hours on this subject matter, including types of RAM, computer architecture, bus speeds and whether or not they’re backwards-compatible, but I won’t, I like to think my point has been put across: whilst a modular device is a nice idea, it won’t take off because a system has to be upgraded as a whole (in the long run) and you will never be able to keep a device for an “infinitely” long time by just replacing components, why? Because it’s cheaper and easier to just buy a new device. The only time these modular devices would be useful is upgrading in the short term, but how often do you upgrade components in your computer? Once or twice? Then you buy a new computer all together because the whole system is now out of date.