"Forbes analyst Adam Hartung makes wild prediction to drive up add revenue"

Came hoping to learn what's so beautiful about iD's code, left convinced that the author (Shawn McGrath) and I have rather different opinions on that... iDs code is certainly not an example of poor code, in a previous job I had the opportunity to view code from around 20 different AAA game studios, its definitely in the top quarter (but that's not saying a great deal); mostly the article is 50 paragraphs of cooing "iD does what I do, guys!" Analysis of what makes said style "beautiful" is subjective at best, and furthermore the author describes himself as "not a coder". For what its worth, IMHO, the best code that I've seen came out of Remedy.

Technically they do--they just haven't called in on our foreign debt yet.

Real parents put tracking devices *in* their children.

Ugh... Sick of all these knee-jerk "It's a terrible deal" articles from assholes that haven't done their research properly.

First of all, if you get your xbox this way, it's warrantied for the two years you're under contract, compared to one year for the usual retail package. The extra year's warranty retails for $50.

When you figure in the extended warranty, the price gap (using the author's Amazon sale prices) shrinks to about $25. If you use the usual retail price of things it actually works out to be $10 cheaper to take the subsidized deal.

Secondly, yes, if you get it on sale and can pay up front, it's cheaper in the long-run. Welcome to the world of finance, asshat. In the end, for everyone else, you're paying a premium of just over a $1.04 per month for the privilege of having the thing now, rather than later. Try getting anything even close to that on a credit card -- at even a relatively modest interest rate of 9%, credit works out to $422 over the course of 2 years.

Nearly every goddamn article and blog on this acts as if Microsoft if fleecing everyone, when in fact the terms are very reasonable, if not generous. Of course they're counting on re-couping the costs elsewhere (games, peripherals, continued growth of XBL), but so be it. The fact that they expect to expand their revenue in this way is not underhanded, allows them to offer a better deal than credit companies, and frankly, is a good business move.

If you have philosophical differences with entering into such contracts yourself, then fine, but that doesn't mean this offer isn't valuable for other folks.

There is no dichotomy because I'm not endorsing that we should, as a nation, focus our economy on innovation. I think it's inevitable that a national government with few exportable resources and an expensive labor force must seek to shore up its ability to monetize IP, yes, but I don't personally advocate that this is a good thing. Without goods extracted from the earth here, or made by US factories, all that's left to offer the rest of the world is innovation and services. Since services are subject to competition from sources that we can't compete with cost-wise, the government is pinning its hope on IP. This is an entirely logical conclusion to come to if the goal is maintaining the status quo of economic advantage over other countries, but again, that's not what I personally am advocating as policy here.

As for your example, I can sympathize that some R&D may not pay off immediately, but I don't think that patents should be a hedge against making a poor investment, being ahead of your times, or simply speculating. These activities are risky. Likewise, I'm not convinced that incidental discoveries are something entities should profit from (though I grant they're probably impossible to distinguish from non-incidental discoveries, or that it's possible to predict all discoveries). I don't have a suggestion as to what the term should be, though I do think that a one-size-fits-all term is not appropriate across industries.

I'm not convinced that a strictly-capitalist view of things is the right way to go. I'll grant that it does many things well, and that the opposite extreme is certainly not successful at industrial scale, but it also fails entirely at other things. Capitalism fails entirely when profits cannot be measured in dollars and cents, for example. It also tends to make the rich richer, and the poor poorer. To say that capitalism is the only way to success is true only in as much as capitalism defines for itself that success means economic success. I'm no socialist hippie, but a lot of pure-capitalist logic is quite circular in nature.

I think you've come to the wrong conclusion about what I was saying.

The first paragraph I'm merely acknowledging what I see as the inevitability that any developed nation must realize: That without raw resources or cheap labor to offer up to the world economy, all that's really left is innovation. And if that nation intends to support itself on the fruit of that innovation, then they must have themselves, and lobby for others to adopt, IP laws that benefit those who hold the most. This is not something I'm arguing for myself, I just think it happens to be on the natural course of things if we desire to maintain the economic status quo.

The second paragraph does advocate for reasonable protections that grant individuals, and their governments through taxation, to benefit from their efforts. The problem with the current system is that there are essentially no limits to the amount of control that the IP holder can exercise, nor any real limit to the length of time one can reap the benefit from their innovation. Current IP law is essentially a land-grab: it says "This thought is mine." and also "If you have to pass through my thought on the way to yours, I can collect a toll. If the price I want is too high, sorry, you and the world are denied your thought." Combined with lengthy protection terms, this allows patent holders to exercise too much control over future innovation.

Patents should exist in some form in order to spur investments as you say, but likewise they should expire in a reasonable term so that they cannot be lorded over future innovation essentially indefinitely. This is a distinctly anti-capitalist idea, but I believe that, at some point, society as a whole has indeed paid all that's due to the inventor, and their invention should at that point essentially become public domain.

I don't take this stance as an outsider. The kind of work I do is digital, and therefore solely protected by IP laws, anyone can replicate the fruit of my labor bit-by-bit, with no real capital cost. I choose not to employ DRM, and to instead encourage people to support me by providing them with a great product, and in the future, supporting services. I *should* be able to seek recompense should someone illicitly distribute or clone my work, but I don't care to have a bludgeon that can be used to prevent those who might do a better job than I, or who might take my ideas in a distinct direction, from doing so.

In college I knew a guy who belonged to the family who's ancestor had invented the modern ball-point pen. He's a really nice guy. I wouldn't begrudge him or anyone else the good fortune of being born into wealth. That such a simple but ubiquitous invention could bring wealth to a family is what should happen when the system works. On the other hand, it seems a little ludicrous that royalties and licenses still flow several generations on.

Also keep in mind that all of IP is not some god-given right of inventive minds. It's a social contract in which society at large agrees to play by certain rules in order to spur innovation and investment. If one side abuses the other, they'll take their ball and go home--this is not the exclusive right of IP holders.

That the US government wants the world to adopt their kind of strong policies should not surprise anyone. The facts are that the US doesn't have a wealth of natural resources, nor do we have the kind of cheap labor that attracts manufacturing. In the long game, all we really have is the ability to innovate for which we certainly don't corner the market; therefore, without the rest of the world adopting similar stances on IP, the US cannot hope to retain it's economic advantage over other countries. The same is true of other developed nations with dwindling resources and expensive labor, and will come to be of poorer nations with few resources--though they haven't come to expect the type of lives we lead in the states.

I think there's a place for protecting intellectual and artistic expressions that exist in a tangible form, but it must reasonable, limited, and well-defined. People should be able to make their living by discovering new things, and by springing something novel and valuable into the world, but at the same time, doing so once should not guarantee lifetimes' of income for you, your children, and so on down the line, nor provide you with the means to prevent others from competing with or building upon your ideas.

This isn't any more true of devs with PS3/Cell experience than it is of devs with GPU compute or Multi-threading experience. The techniques you employ to make the PS3 sing -- namely, batch processing and task-based parallelism -- are very much related to Data-Oriented Design, which is all about keeping like-data in contiguous memory, and/or related data nearby to optimize use of the memory subsystem -- viewing your design in this way is essential for next-gen platforms, whether they be multi-core CPUs (with SIMD), stream processors (like CELL SPEs or other DSPs), or Graphics cards.

Cell was a reasonable move at the time, despite its unique architecture, and its true that it "forced" the migration to this way of thinking -- however, that's because the PS3s single PowerPC core (the same as the 3 in the XBox, sans extended SIMD capabilities) would have been woefully under-equipped for modern games on its own. Still, since the 360 was multi-core and devs used its GPU to perform computations, optimization for either platform commonly benefit the other. The only real difference between the two, in terms of data throughput, is that on the 360 you have the option of spreading out task-level parallelism to more CPU cores if that task is more suitable (for example, "branchy" code, or random memory access patterns) -- on the PS3 you're forced to either give up a precious CPU time-slice, or to re-architect the algorithm to partition data or become less branch-heavy.

It makes perfect sense to ditch Cell -- The only reason they needed Cell 5-7 years ago was that GPUs were not yet general-purpose enough (at least, off-the-shelf ones) to handle all the types of calculations that you'd want to do gracefully. Fusion, or a tightly-coupled CPU-GPU hardware design is precisely the type of architecture that game consoles require.

In fact, the Xbox 360 is essentially "fusion" at the motherboard level -- The CPU can lock and share portions of the cache directly with the GPU -- there's around 27gb/s bandwitdth between the two, and all 512 MB of main memory is GDDR, controlled by the GPU die. This is, 5 years ago, closer to AMD's promised Fusion processors than what AMD's own Llano CPUs are today.

My basic predictions for the next Playstation and Xbox are something on the order of:

• 4 CPU cores which will be "fatter" (OoO, better branch prediction, speculative execution, more cache) than the in-order cores in today's consoles, but may forego widening SIMD execution units. Diminishing returns past 4 cores; with GPU compute, devs will prefer fewer "fatter" cores to many "slim" cores.
• GPU based on AMD 7x00 series (or nVidia Tesla/Kepler), expecting ~800 computing elements.
• GPU compute resources will offload data-parallel computations -- might be "fusion"-style shared-die, might not -- heat still an issue for 4 fast CPU cores and that many shader elements.
• At least 4GB GDDR5, unified memory space; 6 or 8 GB possible, if not likely. Greater than 200gb/s bandwidth.

This really ought to be simple -- A local municipal or co-op ought to be able to go into the business of providing a service on the same terms as private providers. Private providers should be free of discriminatory practices (e.g. a local govt. screwing them out of entering or expanding the market), but should *not* be free of competition from any and all comers, private or public. If a community desires better service than current offerings are providing, and are able to pay for it, then they shouldn't be prevented from effecting that change for themselves. Their ability to do so should not be restricted any more than any other public works.

If we had any pro-competitive notion of net neutrality (e.g. that the "last mile" and other support infrastructure should be available for use by any and all competitors in the market, and to new-comers, without undue burden) this problem would not exists. If that infrastructure was installed by private industry previously, then they ought to be able to charge a fair access fee to competitors to compensate themselves, while being low enough that competitors can provide a competitive pricing structure. That would be the true "free market" solution as far as customers (as opposed to providers) are concerned.

The fact that we don't have this in place does nothing more than protect the ability of those who hold this infrastructure to protect their locked-in consumer base from competition so that they are able to sell largely-worthless package "deals", and with artificial limitations, for insane profits.

The "free market" should mean one in which all comers are able to enter and compete on the merits of their offerings, not one in which the incumbent players are free from competition. Obviously the consumer prefers the former definition, while the incumbents and their political buddies get fat off the latter.

If the rumor is indeed true that a custom firmware has been used to get some people free stuff, take note how Sony has handled the situation -- A small, small portion of people (the few that run custom firmware, and the fewer that run this particular custom firmware) are getting a few free (virtual) goods, and they shut down the entire network, screwing 100% of their customers.

What if banks operated this way? They find a ring of fraudsters using bank accounts to commit fraud, and the bank responds by freezing everyone's accounts for weeks? It would be totally unacceptable.

When you find a small group of fraudsters, you take targeted action against them alone, even if it means you hemorrhage a little money compared to the more totalitarian approach. Its part of the cost of doing business. In the retail world they call it "spillage" -- the fact that some of your goods might get damaged beyond saleability or that a few things will go missing from the floor (or the stock room) is unavoidable -- you simply do your best to detect and take action against those responsible, but you don't go around treating every other customer as a criminal.

Of course, that assumes the rumored reason is the cause of this action -- I suspect its either speculation or a (possibly intentionally-leaked) cover story for other measures taken in response to the Anonymous attack and whatever information they got out of GeoHot in the settlement. I anticipate a new official firmware will be required after the network comes back up and it will be necessary to access the "new" PSN, and possibly even already-owned downloadable content. This long of a downtime indicates pretty drastic changes behind the scenes, methinks.

So, you're off on your hardware specs --

The Wii uses a single-core, 32bit IBM PowerPC (a derivitive of the G3 in the original iMacs, with added MMX-like 2-wide vector operations) -- its the same thing they used in the gamecube, just 50% faster and with some minor architectural improvements.

The PS3 and Xbox use 64bit IBM PowerPC cores, but they are not a derivative of the 970 (aka G5) or any of the other big-iron PPC chips. These cores are essentially a PowerPC ISA equivilent to Intel's Atom processor (64bit, in-order, hyperthreading), albeit the console processors (at 3.2Ghz) have twice the clock speed of a typical Atom core. The Xbox 360 has 3 of these dual-threaded cores each with an enhanced altivec unit with an extended register file (128 registers instead of the usual 32) and a few special instructions thrown in (horizontal add, which is handy for dot products for instance) all of which share an L3 cache. The PS3 has only one of these dual-threaded cores, with a bog-standard altivec unit, but offloads a lot of the heavy-lifting to the SPE array (7 DSP-like processors).

The current rumors suggest that Nintendo will be sticking to a PowerPC CPU in their next system -- rumors are consistent with a derivitive of the same Atom-like PPC cores in the 360 and PS3, but could be higher-clocked, or something else entirely, and with much-updated graphics, surpassing the PS3 and 360.

My view is that ARM actually is a contender for the next gen Nintendo console -- given the relatively-weak Wii processor, several off-the-shelf ARM cores would provide Nintendo with a substantial upgrade in CPU power -- likewise, since the 360 and PS3 are *not* running 970s, a fast 2-4 core ARM processor would be fairly competetive with those platforms, CPU wise. The downside of ARM (specifically compared to PPC) is that ARM's Neon SIMD ISA is terribly limited compared to Altivec (which is nicer than even SSE), and also that no off-the-shelf SOC has graphics prowess that would compete with, much less out-do, either the PS3 or 360 (Though SOCs from even 18 months back could give the Wii a run for its money, and offer more-flexible shading.) What Nintendo would *gain* by going to ARM on their home consoles is to consolidate their technology around one ISA, which would allow them the focus their efforts with regard to their toolchain, libraries, and developer support, while also giving them a very easy path for their portables (just make their last home console portable, literally -- which they basically do in terms of capability now, but not in terms of architecture.) I don't know that Nintendo will go this route, but its what I would do in their shoes.

And I'm certain that, once victorious, they will make all due effort to return all $3 Billion dollars to their rightful owners -- the 'victims'.

Yep. Every last penny.

What? No mention of Rob Pike?

Turn in thy geek card good sir.