I've always wondered if some of the huge R&D budgets we see are shenanigans to save taxes under the US's R&D Tax Credit.
I certainly don't know enough to accuse anyone of tax evasion. But it does seem to be a strong incentive to classify lots of different activities as R&D.
The $100 million sounds slightly less impressive if you take it to mean controller R&D across the entire Xbox One project.
Let's assume that Microsoft started work on the console in 2007, two years after the Xbox 360 launched. Our $100 million has now become $16 million per year. Developing the controller is going to involve a lot of people: software, hardware, QA, design, supply chain, fabrication, etc.
Let's say that Microsoft has a team of a hundred people working on the controller (personally I think that's a very conservative guess). So of our $16 million we've easily eaten up half of that in labour costs. Prototyping, gearing and tooling up, and all the various other expenses associated with designing and manufacturing electronics eats the rest up pretty quickly.
This is obviously 'back of the napkin', hugely simplified conjecture, but I can definitely see how $100 million becomes a fairly reasonable figure over time.
With that said - they did invest in "smell-o-vision", having other displays, and probably a few other far-fetched technologies they haven't discussed. But $100 million is still a lot of cash if you ask me.
> invested over $100 million throughout the course of the effort
That line is completely at odds with the title. In R&D 'alive and well' is not synonymous with spending lots of money. The key factor for deciding whether an R&D organisation is 'alive and well' is output. I.e. have they produced novel research or developed amazing products. Another important factor is the cost of that output. Anything is possible with an endless supply of slave labour (or money).
In this case I think that spending 100m on a controller that has only incremental improvements is not a sign of good R&D. It is a sign of a company with a lot of money that is not willing to take excessive risks.
Research doesn't always produce changes. If we suppose the Xbox360 controller is nearly the most perfect a controller could be, it's easy to see how large R&D budgets can lead to only small changes.
As for why they'd funnel a lot of money into a controller that is nearly perfect- well, they have the money and they want to win. Isn't that enough explanation?
> As for why they'd funnel a lot of money into a controller that is nearly perfect- well, they have the money and they want to win. Isn't that enough explanation?
The problem with regards to Microsoft's R&D budget is that they're not winning. I don't have figures for this year, but last year Microsoft's R&D budget (as listed in their annual report) was $9.4 billion, compared to Google ($5.2 billion) and Apple ($2.6 billion).
One explanation is Microsoft simply account for R&D differently. Another explanation is that Microsoft overspend on R&D, or spend very poorly. Microsoft have always spent very highly on R&D, but over the past few years haven't had much to show for it in terms of growth or success.
I am sure there is some hyperbole and creative maths in the report to get to $100 million, but where you see Microsoft investing money to perfect the almost perfect, I see Microsoft spending far too much money on the wrong things.
But Valve's experiments show that Xbox controller is not the perfect ideal. Thus, there is improvements available in the range of 100M, don't you think ?
Ow, crap ! That's a huge stack of money !
As a side note, I am surprised that an article where all the information is in the title made the front page. I expected some details, pictures, or analysis...
Microsoft chose a scope in which to explore. Valve's style controller branches off the tree way beyond that scope, and honestly 1) we don't know how good Valve's controller is yet and 2) Valve has said their controller is aimed to meet the needs of specific groups of games.
Is valve's controller really better? On the PC there's a huge range of possible controllers people could use, but the xbox one is the only one I see people recommending.
The Xbox One controller certainly feels better than the already very good Xbox360 controller, but I doubt there's really 100M of work needed to go from one to the other. In other words, it could have been done for cheaper.
Well sure, they could have handed the team a 360 controller and $10,000 and said "make it $10,000 better". But clearly they wanted to explore more options, many of which didn't pan out. Retrospectively calling that a bad idea simply because they didn't produce something dramatically different misses the "R" in "R&D".
> Despite the firm's aversion to rocking the proverbial boat, it still wound up with more than a few unconventional prototypes -- some of which packed built-in displays and cameras. One of the strangest versions included a cartridge for emitting smells, and another featured a built-in projector that could throw out visuals reminiscent of illumiroom. Ultimately, the wackier iterations gave way to the traditional kit that's heading to stores, as the adventurous features drained battery life too quickly or the company's "core base didn't appreciate them," according to Alam. While we wouldn't have expected the Xbox One to usher in the age of smell-o-vision, we can't help but wonder what that future would have looked smelled like.
I.e. they did not use the 100M to make tiny improvements to the 360 controller; they engineered several completely different prototypes, which will only be relevant several generations ahead, and a tiny fraction of the money went into the improvements. Even if there is never any controller remotely like these prototypes, some of the engineering knowledge gained will pay out.
> It is a sign of a company with a lot of money that is not willing to take excessive risks.
Microsoft included a Kinect with every Xbox, despite the Kinect pushing the cost of the console to $100 over Sony. Risk aversion doesn't seem to fit the pattern here.
Research theater? Take this with a grain of salt however, research value is non-linear and its success cycle is measured not In years or a decades even.
It's important, really important. Imagine how sales would be impacted if they had not spent this money and built a crappy controller. Also factor in the suggested tail of the console is 10 years.
I just hope they've spent a similar amount of money on making sure the damn thing is quiet when in use.
It looks like valve has spent a few hundred dollars on a 3D printer and I expect a small team and come up with something far more revolutionary. I wonder how it will play out.
I also wonder if this cost covers the connect too.
I certainly don't know enough to accuse anyone of tax evasion. But it does seem to be a strong incentive to classify lots of different activities as R&D.