Intel’s Misunderstood Upgradable Chip
There are some weeks I’m sure as hell glad I’m not a CMO. Case in point today Intel was pounded by Engadget readers for creating what appeared to be a crippled chip that, for a fee, you could un-cripple. OK take a breath it isn't that bad.
The Best Buy Trial
So, first off, this isn’t a huge conspiracy this is a trial with one desktop system configuration at Best Buy to see if folks would prefer a system that had an upgrade path over one, for the same price, that didn’t. The processor being used is a unique part used only for this trial and the performance it has before the upgrade is the same as the performance for every other processor in its price class. If you go to another store and buy another similar system (OEM(s) engaged in this trial aren’t yet identified) it will cost about the same, have about the same performance, but not have the upgrade option. The way it would work is you would have the typical choice of like priced systems at Best Buy and one configuration would have an option that would allow you to upgrade it at the store to a higher level of performance. In theory you wouldn’t have to buy the upgrade immediately but only later on and if you noticed you needed more horsepower effectively increasing the service life of the hardware. The idea they are testing is whether people will prefer systems that have an easy upgrade over identical systems that don’t and whether enough people will upgrade to more than make up for the cost of a putting a more expensive processor in up front.
This would be kind of like looking at a car and not being sure if you want 4, 6, or 8 cylinders and the related horsepower, paying for a 4 but betting a V6 or V8 with the other cylinders disabled but competitive performance (gas and horsepower) to the other 4 cylinder engines on the lot. Later on, if you wanted more power, you would pay an upgrade charge (certainly far less than replacing the engine or car) to get that V6 or V8 performance. In this initial case the upgrade isn't as dramatic but, if the trial works, future parts could go farther.
Reason for the Trial
People don’t like to change out processors or engines even though it really isn’t that difficult to do the former in a desktop system (laptops are something else). However marketing upgrades like this has proven elusive. For instance when you buy Windows Home Premium you have installed the bits for all of the other more expensive versions of Windows but they are disabled, you can pay for an upgrade and those features will be turned on. This saves you having to reinstall it (which is what you used to do).
Many free software products work the same way, if you want more capability you pay an upcharge, get a code, and then those features are available but most don’t upgrade. Not a huge problem with software because the costs associated with putting the extra bits on the drive or disk given you have plenty of capacity is almost zero. But with a processor Intel has to build a more expensive part than they otherwise would for the segment and bet that enough people upgrade to at least cover the costs of that more expensive part.
What Could Happen if the Trial Works
Now what could happen if the trial works is the number of physical processors would drop sharply and you’d simply pay for the performance you need. Say if you like a desktop or a notebook and really only need two cores and a low level of performance you’d buy performance level "A" in the configuration that you wanted but if you wanted more you’d have performance level "B" turned on. And this could be for all types of things in the product. If, later on, you changed your mind and wanted more you could pay the difference and presto you’d have that extra power. That way there would be less likelihood, compared to today, that you’d later regret not buying more performance at the start because you could always get it later. Intel would build fewer different parts and manufactures would only have to have one configuration for each design. Think of an iMac but with a broader range of prices and upgradeability.
Wrapping Up: Tempest in a Teapot
The problem with all of the conspiracy theory crap is that had Intel truly crippled a part and then charged extra to get back to baseline performance as opposed to charging for extra performance AMD would chew them up in the market. As long as there is a competitor in the segment, and AMD is particularly strong with desktops, Intel can’t do this. They have to be competitive which means that while you do pay more for the extra performance it is performance you wouldn’t have had anyway and most of us are OK with paying more to get more. It continues to amaze me that we complain about companies not being creative and then pound on them for trying something different. It is a trial, with one system, it may not work and if you don’t like it there will be plenty, the majority actually, of systems out there that won’t have this option. Me, I like the idea of being able to upgrade to a higher level of performance without actually changing the processor out but since I also like to build my own systems so doubt I’d ever actually use it. Still, I like the idea of choice and think companies should be encouraged to try new things, if they don’t life is going to get rather boring.