Technology

XBox 360 vs. PS3

Shahms E. King — Tue, 17 May 2005 07:29:00 GMT

The below is from a Slashdot comment, so should be taken with a grain of salt, but he does site his sources and the comparison seems to match everything I've read. I'm simply using it to save myself some time. I'll summarize: Advantage PS3. The only thing the Xbox 360 has going for it is the release date and (probably) the Xbox Live stuff. Enjoy!

Xbox 360 has a CPU FPS of 45 GFlops*
PS3 has a CPU FPS of 218 GFlops

Xbox 360 has a GPU FPS of around 955 GFlops**
PS3 has a GPU FPS of 1.8TFlops

Xbox 360 has a combined FPS of 1TFlops
PS3 has a combined FPS of 2.18TFlops

Xbox 360 has a DVD-ROM
PS3 has a BD-ROM

Xbox 360 is WiFi ready
PS3 is WiFi built-in

Xbox 360 has 3 x USB 2.0 ports
PS3 has 6 x USB 2.0 ports

Xbox 360 has support for 4 wireless controllers
PS3 has support for 7 wireless (Bluetooth) controllers

Xbox 360 uses Memory Units
PS3 uses MS Standard/Duo/Pro, SD standard/mini & Compact Flash Type I/II

Xbox 360 has support for select Xbox1 games
PS3 has support for PS1 & PS2 games

Xbox 360 has support for 1 720p & 1080i display
PS3 has support for 2 480p, 720p, 1080i & 1080p displays

Note:
* Derived from CPU Game Math Performance of 9 billion dot product operations per second
** Derived from subtracting published Overall System Floating-Point Performance of 1TFlops with derived from CPU Game Math Performance of 9 billion dot product operations per second

Source:

Wikipedia's PS3 Tech Specs
Official Xbox 360 Fact Sheet
Formula for Dot Product Operations Per Second to GFlops

Hungarian Notation Explained...

Shahms E. King — Thu, 12 May 2005 15:54:00 GMT

Joel Spolsky does a good job of clarifying both what Hungarian notation was supposed to be as well as how it was turned into the most useless and irritating coding convention known to man in his article Making Wrong Code Look Wrong. He also rants against exceptions. I'm not usually one to read, let alone link, to a "Joel on Software" article, but this one does a remarkable job of explaining how something as hideous as (Systems) Hungarian Notation could be so fervently expounded as the "One True Coding Style" for so long. Joel writes a lot of crap, but underneath the steaming pile are usually one or two nuggets that are worth digging out.

P2P TV

Shahms E. King — Thu, 21 Apr 2005 08:09:00 GMT

A little while ago, His Shortness and I were discussing the possibilities arising from the unholy marriage of BitTorrent and RSS. It looks like The Participatory Culture Foundation may have beaten us to the punch, however. If it lives up to the hype, the software could be pretty amazing. Of course, the real money is going to be in providing real content this way, not the inane webcams of a thousand bored geeks.

I Was Here First!

Shahms E. King — Wed, 13 Apr 2005 07:23:00 GMT

Oh what silliness ensues when questing after the famed "first mover" advantage. Of course, if Microsoft does ship the XBox Deux before Christmas it might mean something . . . then again, it won't. It looks like the positions for the next round are reversed with Microsoft being first to market with an inferior product and Sony coming out (much) later with a more powerful box. Sony has also announced the PS3 will be backwards compatible with both PSOne and PS2 games; the only word from Microsoft has been "maybe". There is much speculation about using Virtual PC 7 for playing older XBox games, but nothing definite. On the game front, the Japanese might actually buy the XBox Deux and there have been many "defections" from Nipponese game makers. Should make for an exciting few years of gaming...

If only Judas would stop teamkilling...

Longhorn To Be Released On Time!

Shahms E. King — Wed, 01 Sep 2004 13:30:00 GMT

In a surprise move, Microsoft has announced they will be releasing the next version of Windows, dubbed Longhorn, on time in late 2006. "On time" unless you mean any one of the numerous release estimates (beginning with "early 2003"). Of course, the 2003 release was slated to include Avalon (the new graphics API), Indigo (the new network API) and WinFS (the all-singing-all-dancing object relational file storage) and all of this was going to be integrated in such a way as to make the "local computer" and the "network" nearly indistinguishable. Those three "big" features are just the most recent ones to be removed from Longhorn. The original hype included a good deal more that escapes me for the moment. The 2006 release will include none of these. They will be released as separate "add-ons" and will likely be released for Windows Server 2003, Windows XP and Longhorn, but not until after 2007. Suddenly, those Software Assurance licenses are looking less and less worthwhile. Paying every 2 years for software that's updated every 5 (XP -- the most recent client OS -- was released in 2001) doesn't seem very economical. Shit, that's 4 years of paying for software upgrades without getting anything. I wonder how those customers whose anger was assuaged with promises of huge savings on "free" upgrades for 2 years feel about that. To top it all off, when the upgrade finally ships, it will be doing so with 0 (that's zero, zilch, none) of the much ballyhooed features. If you signed up with Software Assurance with the release of Windows XP (or 2000) you will be into the 3rd "update cycle" when Longhorn (the 1st update) is released. This first update comes with none of the promised features. (Philosophical Quickie: If a software update ships with no new features, is it really an update?) And all of the promised features will be released for the current batch of operating systems, leaving little incentive to upgrade. Of course, Longhorn isn't even out yet. I'm starting a pool on the *actual* release date. $1 gets you a day; I've got $30 on September 2007.

At long last IronPython (and a rant on patents)

Shahms E. King — Sun, 28 Mar 2004 09:48:00 GMT

Python, possibly the best language ever, running on the CLR you say? But isn't the CLR "horrible for dynamic languages?" Only if you aren't Jim Hugunin, famous for such things as Jython, among others. The full paper on the performance, implementation and capabilities of IronPython 0.2 (it's a research project, after all ;-P) is out so you can decide for yourself. All things considered, I'd say it's another nail in the JVM coffin (thank God). Now if only Microsoft would officially and irrevocably license all patents concerning the ECMA standards on a royalty free basis, the rest of us could get back to the task of World Domination without worrying about litigation over obvious patents from an untrustworthy and unscrupulous corporation.

And seriously, despite the ignorance of the public at large on this point, 90% - 95% of the software patents out there are either 1) Obvious to a practitioner of the art or 2) based on prior art in the field. Beyond that, almost all of them are reworded patents on algorithms, which, along with facts and scientific research, *are not patentable*. Far from promoting the progress of science, they in fact stifle it as developers must constantly wonder if the algorithm they just wrote or used from a standard trade text! is patented. Sure, it might promote progress if the USPTO actually did its due diligence in researching the patents instead of simply rubber stamping applications, but that's not the case. And Monsanto. What's next, are you planning on patenting "the process whereby a spherical agglomeration of hydrocarbons self-replicates into two identical agglomerations of hydrocarbons?" Rididculous. Absolutely ridiculous. Sure, you deserve just renumeration for your efforts, but the ability to patent genes is not the way to go about it and certainly shouldn't be allowed by the USPTO!

Anyway, IronPython and the CLR: cool. Software, business method and biological patents: bad.

Supply-Side Economics (Rohit, this means you)

Shahms E. King — Thu, 05 Feb 2004 15:02:00 GMT

The discussion on "The Bushtastic Budget Bid" inspired me to do more reading and thinking on econmics in general ans supply-side economics in particular. Typing "supply-side economics" into Google brings up a number of hits; almost all of these hits are defense of supply-side theory or Reagonomics. I could find very few reasonable anti-supply-side papers in my limited perusal. What this tells me is that either the counter-arguments are so obvious no one has bothered to make them (or feels compelled to defend them), the counter-arguments use a different term (perhaps I should have looked up "voodoo economics"), or there is no reasonable rebutal for supply-side economics. I'm more than willing to concede that my expertise the in area is such researching alternative viewpoints is not going to be my strong suit as I'm not even sure where to start.

That being said, what follows is some issues I have with the theory (or, actually, any economic theory I've heard to date).

The inter-relations and feedback loops in any economy are such that all the current theories I have heard or read seem vastly oversimplified and insufficient to explain anything with a degree of certainty sufficient for whole-hearted endorsement of any specific policy, especially given the risk to the welfare of large numbers of people. In other words, the economy is a chaotic system (using the strict mathematical definition here) and any linear equation of "If a then b" is going to be wrong enough of the time to make relying on it foolish.
Supply-side economics (and, again, most economic theories I've heard) are frequently used after the fact to explain things in much the same way as Freudian psychology, Jungian psychology and Adlerian psychology are. The precepts of the theory are vague, yet plausible, enough that conflicting theories can be convincingly used to explain the same evidence. This isn't to say they aren't useful or valid, simply that their utility should be taken with a grain of salt.
The numbers I have seen in support of Reagan's economic plan indicate that cutting taxes on both rich and poor increased the income tax revenue from the top 5%-10% and reduced tax revenue from the rest. Okay, so that's basically what the Laffer curve says will happen if some segment is over taxed. However, the increased revenue from the wealthy came from somewhere and looking at the numbers, while the income among the wealthy increased enough to offset the reduction in taxes, income among the rest of the population dropped such that the amount of taxes collected was well under what was expected due to simply lowering taxes. So, even using these numbers, lowering the tax burden (in the manner of GWB and Reagan) leads to a significant increase in the gap between rich and poor, impacting the middle classes the most. And unless you think the top 10% can support the entire economy themselves, supply-side economics looks effective as a short-term boost to some economic indicators at the expense of the long term health of the economy.
Despite the fact that the top 20% pay around 65% of federal taxes and only account for about 54% of the total income (using 1999 CBO estimates, I couldn't find any newer ones) they control over 81% of the total wealth. I'm not exactly sure what that means, but I'm inclined to say that the wealthy, by and large, save their money to a much greater extent than the rest of us. This, coupled with the numbers from above, indicate that contrary to the predictions of supply-side economics, reducing the tax burden on the wealthy increases GDP at the expense of the rest of the citizens. Rather than reinvesting that money, a sizeable number simply use it so further grow their own bank accounts. I'm not saying that supply-side economics doesn't work, simply that it is not enough.
There are a lot of implicit assumptions in the theory that I find questionable and whose veracity I have never seen demonstrated. In particular, the assumption that increased investment by the upper classes leads to an improved economy for everyone. It sounds a lot like the underpants gnome's idea of business to me:
1. Cut taxes on the wealthy.
2. The wealthy have more money.
3. They invest that money in new areas.
4. The investment of new capital creates new opportunities for growth.
5. The growth creates new jobs.
6. New jobs create more income.
7. The extra income makes up for the reduction in taxes.
8. The economy is better.
The assumptions I have problems with are, primarily, 3 and 5 and the assumption of a causal relationship between any number of these things where I don't think one has been proven to exist. Assuming that if wealthy people have more money to invest they will do so in "growth areas" is poor, as is the assumption that that investment will lead to growth in that area and that that growth will inevitably lead to more jobs. All of these are unfounded and I can come up with any number of examples where it just doesn't "play out like that". There are so many factors that go into every aspect of the economy that assuming a simple relationship like that is a lot like assuming Newtonian physics can describe anything in the real world. Yes, it might be useful for making vague estimates of what might happen in given a set of circumstances, but as soon as things like wind resistance or electromagnetic forces come into play, that estimate can be woefully incorrect. And physics on a macroscopic level is arguably much simpler than economics on any level.
So, after all of this, it really boils down to this: supply-side economics maybe perfectly valid, but it is not complete. As such, it cannot be relied upon to provide all the answers and should not be used to the exclusion of all else, especially when the wellbeing of so many is at stake. Particularly in a global economy such as this.

On a side note, as much as I want to be irrate with my fellow IT workers at the migration of jobs overseas, I just can't. They provide much the same service and a much lower price, sure it might be unfair that they can live quite comfortably (oppulently, even) on $11,000 a year while I could just barely afford to feed myself, let alone a family, on that, but on the whole, I cannot complain about increasing opportunity in a country with such serious problems as India. Bemoaning the loss of a few jobs in a country that can afford to lose them (kind of, at the moment) to a country as poverty stricken as India is extremely self-centered and short-sighted. Somehow charity is "good" while jobs are "bad" doesn't sit well with me. Teach a man to fish and all that. Much like pure science, the benefits might not be obvious at first, but given enough time another country with a strong economy and an even stronger IT sector is good for both the US and the world as a whole.

Deregulation of the power grid a good thing?

Shahms E. King — Fri, 10 Oct 2003 21:38:00 GMT

Somehow, there's been this myth floating around that deregulating the power grid had nothing at all to do with the power shortages in California or the massive blackouts on the East Coast. That never seemed reasonable to me, and perhaps this is why: (de)regulations can't change physical reality.