The technology business depends on selling version 2.0.
Both hardware and software companies are endlessly trying to improve their products. Depending on the company and product group, that desire for improvement may be driven by a genuine desire to build something innovative and better. Or, it could be that new “features” are added willy-nilly in order to keep the revenue generation machine clicking.
The reason I’m thinking about this is my recent experience with, of all things, a CPU cooler.
Before we get to the CPU cooler, though, let’s think on the whole idea of version 2.0 (or 3.0 version 7 or whatever.) We all like to believe that a new version is better. That doesn’t mean we necessarily want to rush out and buy a new version. When we see a company ship updated or new products, it gives us this warm fuzzy feeling that things are getting better. Your CPU isn’t fast enough? Intel ships a shiny new Core i7 or i5. Your word processor feels limited? Look, Word 2007 has that nifty new ribbon interface.
I’m certainly not immune to the endless cycle of improved versions, as I proved by pre-ordering a Nikon D300s. On the other hand, I don’t always rush out and buy the latest version.
There’s also a tendency among the majority of users to feel warmest and fuzziest with incremental improvements. When a new product that’s based on an older one ships with radical changes, it often generates serious controversy. The aforementioned Word 2007 ribbon bar is a good example. I find the ribbon bar useful, but quite a few Word users loathe it.
Similarly, Windows Vista engendered a lot of controversy. It was a substantial departure from Windows XP in many ways. The fact that the original version shipped with a number of performance sapping bugs didn’t help matters. By the same token, Windows 7 is getting rave reviews – it’s the version of Vista that should have shipped, goes the conventional thinking. My thoughts: it’s an iteration on an old version (Vista) – and despite the fact that Vista itself found little love, Windows 7’s massive numbers of small improvements seem to make people feel comfortable.
What’s that got to do with CPU coolers?
I’ve used the Scythe Ninja in a number of PCs I’ve built over the years. The original Ninja mounted using standard mounting hardware, which meant that you didn’t need special mounting plates attached to the motherboard. It was also fairly light, unlike some monster coolers. I also liked it because you could attach quiet fans and build a system that could run one speed grade higher, yet offer relatively low noise.
In the past few days, I’ve been building up a new reference test system. (I’ll go into more detail about that in a future post). One item I picked up to be a part of that system was the Scythe Ninja II.
Now, the Ninja II seems to offer improvements in the way the heat pipes are laid out. They’re spread out more, which helps dissipate heat more efficiently. In fixing that, though, they created a different problem: the user interface had to change.

Original Ninja on the left; Ninja II on the right.
Yes, a heat sink has a user interface. The original Ninja is on the left, the new one is on the right. See the notches on the original? Those notches are cut out of all the fins. This allowed you to use a screwdriver or other skinny tool to press the mounting latches into place, or rotate them to dismount the cooler. Because of the way the heatpipes are laid out in version 2, these notches had to go away. Instead, Scythe carved out a small area near the bottom to make it easier to insert a finger to press down the mounting latches.
This works fine if your motherboard is on a bench top. Once installed in a case, though, it’s almost impossible to get a finger into cutout near the corner of the motherboard. In effect, you have to remove the motherboard if you ever want to replace a CPU.
For many users, that’s not a big deal. For people like me, who are regularly trying out different CPUs, it’s a nuisance. If I have to remove a motherboard to install a new CPU, I might as well get a fancier cooler with a mounting plate.
In effect, Scythe just removed one of the major advantages of the Ninja, while making an incremental improvement in another area. Depending on what you perceive to be the more important aspect of the original Ninja, you may or may not want to use the new one in a future system.
And so it is with many “version 2.0” technologies. Seeming improvements added to a new product create other, unanticipated problems – the law of unintended consequences at work. Engineers can often benefit from the input of a good industrial engineer, a robust beta process or even talking to the occasional product reviewer. Real users will always find the flaws in your product, much more so than any internal testing will ever reveal.
So when a version 2.0 ships, and it’s not quite what you want, you may view any subsequent version 2.5 or 3.0 with some skepticism. But remember, technology companies stay in business because of their commitment to continuous improvement. Sometimes the improvement process can lead to mistakes. Companies that learn from their mistakes stay in business and grow. Companies that fail to learn don’t.
2 comments
Joey says:
August 4, 2009 at 5:08 pm (UTC -7 )
Unfortunately I feel that most companies are coming out with a version 2 not to try and improve the product, but to trick people into buying the new product
killerplat says:
August 5, 2009 at 4:44 am (UTC -7 )
As always llyod, very interesting read. but i will say, verison 2.0 mentality is manly true for hardware companies. for software companies version 3.0 is the version most will wait to use, point and case windows xp in most enterprises ( most people skipped windows me, and few companies implemented windows 2000 workstation for the average business user). Ther are many examples of that in the enterprise where version one is usualy the rough draft. version 2 is the fixes for version 1 and version 3 is the fix and the ui improvements of version 2.Putting that aside in terms of hardware. or in this case our cpu fan. i totaly agree with all your comments. I think what the hardware accessories market seems to forget is that the buyers are ususally high end pc users. Users that want to tweak the hell out of our systems, users that are constantly opening pc cases (and for me personallly, forgetting to close the case). with that mentality they are forcefuly making better consumer prodcuts ( better heat flow tech, better use of conductors) but also forgetting about the people that are buying them. i dont knwo if they are making these heat sinks in order to sell to pc manufactures. If that is what they are doing then that will explain the weird design of the clips.
what i find weird witht his heat sink is the slices through the fins. i mean for me i think the new hit sink apart form the clips, seems to take one step backwards. From my research and very short analysiss of the new heat sink versus the old. the slice at the edges of the old heat sink, allowed for air conductions form the bottom( motherbaord to the top with little friction therby incresing airflow and also improving heat dispersion. the new heat sink got rid of that, but instead improved on the copper tubing spacing by spacing them away. This also will work since heat is now dispersed more evenly instead of forming one central heat spot like the old version. My question why dont they skip all this intermediate steps and merge both the first and second version together.
I will like to see the results through a ir spectrometer, to see what the real performance imprvements there are in the new cpu fan.
Sorry bou thte long post.