JKKDARK
June 13th, 2007, 23:42
via PC World (http://blogs.pcworld.com/gameon/archives/004643.html)
Remember your old elementary school science lessons about potential and kinetic energy? Pull a rubber band taut and you give it potential energy; release it, and that energy becomes kinetic. Not to sound trite, but you might compare that rubber band to the relationship between silicon and software development. If the rubber band is, say, the Xbox 360's triple-core Xenon PowerPC-based processor or the Playstation 3's seven-core Cell CPU, either system manufacturer's software "devkit" becomes the hand stretching that rubber band to different degrees of tautness.
Recall the Sega Saturn, which had eleven processors plus dual CPUs. Sony's Playstation had just three (ah the irony). I remember reading something in (now defunct) Next Generation Magazine around the time the Saturn launched about how powerful the latter system was compared to the Playstation, but also how prohibitively difficult it was for developers to tap into that extra oomph. Sega further confused matters by shipping a devkit with the intelligibility of an underwater toaster. Sure, Sega's "rubber band" was much stronger and more extensible than the Playstation's, but the fingers doing the pulling might as well have been handcuffed thanks to Sega's byzantine development tools. Thus the Saturn failed, despite Sega's prior prominence in the 16-bit market with the Genesis.
Is the Playstation 3 the next Sega Saturn? Both systems were/are radical departures from prior platforms with experimentally complex multi-processor architectures. Compare the jump from Intel's Pentium 4 to its Core Duo and Core 2 Duo processors with corresponding chipset upticks, none of which come remotely close to jolting developers to the same extent as jumping from the PS2 to the PS3. (Though it's worth noting that the jump from single to dual-core processing has been a substantial challenge for developers looking to utilize Intel's second core effectively--parallel processing is a paradigm shift that requires significant retooling on any hardware platform.)
Developers were keen to berate the Saturn back in the mid-1990s, but they're staying mum about the PS3 today. GamePro managed to uncover a few semi-illuminating tidbits, reporting on a story in Dr. Dobb's software journal in which the authors claim "Software that exploits the Cell [processor's] potential requires a development effort significantly greater than traditional platforms." In an email to GamePro, Sony's product development group added further:
Since PS3's Cell processor allows more features -- better physics, more complex graphical processing, lighting or sound -- there is inevitably going to be more cost in supporting those extra features... It's not that PS3 is harder to write for, it's just that you can do more with it.
Of course what Sony calls "more you can do with it," most intelligent developers would equivocate with the same sort of learning curve that killed the Sega Saturn. No one's disputing it's more powerful at this point, but wrapping a Honda Civic around a V12 won't win you any races.
Imagine playing a sport like basketball with a certain ruleset, and say you've been playing for years. As a competitive athlete, you have to focus on mastering dribbling, passing, and shooting techniques which explicitly complement those rules. No running out of bounds, carrying without dribbling, etc. Now let's say the league decides to alter the rules dramatically, and I'm not just talking scoring or timing minutia, but--to make the analogy proportional--something radical, say having to dribble two or three basketballs at once. You might be able to do it (hey, plenty of reverse jugglers can!) but think of the practice and utter rethinking involved to do it effectively. Passing around three basketballs might increase your team's chances of landing a shot, but every tactical aspect of the game has to do a one-eighty.
The most damning element in the stack against the PS3, though, may be multi-platform games (mostly contributed by mega-publishers like EA and Ubisoft) which require costly re-engineering when porting to the PS3 architecture. Witness products like Marvel Ultimate Alliance and Spider-Man 3, which run smoothly on the 360 but often jerk conspicuously on the PS3, and occasionally in a way that actually interferes with the gameplay. Add the higher development costs (read: time) to port to the PS3, and you have a situation not altogether unlike the one Sega found itself in slightly more than a decade ago.
Remember your old elementary school science lessons about potential and kinetic energy? Pull a rubber band taut and you give it potential energy; release it, and that energy becomes kinetic. Not to sound trite, but you might compare that rubber band to the relationship between silicon and software development. If the rubber band is, say, the Xbox 360's triple-core Xenon PowerPC-based processor or the Playstation 3's seven-core Cell CPU, either system manufacturer's software "devkit" becomes the hand stretching that rubber band to different degrees of tautness.
Recall the Sega Saturn, which had eleven processors plus dual CPUs. Sony's Playstation had just three (ah the irony). I remember reading something in (now defunct) Next Generation Magazine around the time the Saturn launched about how powerful the latter system was compared to the Playstation, but also how prohibitively difficult it was for developers to tap into that extra oomph. Sega further confused matters by shipping a devkit with the intelligibility of an underwater toaster. Sure, Sega's "rubber band" was much stronger and more extensible than the Playstation's, but the fingers doing the pulling might as well have been handcuffed thanks to Sega's byzantine development tools. Thus the Saturn failed, despite Sega's prior prominence in the 16-bit market with the Genesis.
Is the Playstation 3 the next Sega Saturn? Both systems were/are radical departures from prior platforms with experimentally complex multi-processor architectures. Compare the jump from Intel's Pentium 4 to its Core Duo and Core 2 Duo processors with corresponding chipset upticks, none of which come remotely close to jolting developers to the same extent as jumping from the PS2 to the PS3. (Though it's worth noting that the jump from single to dual-core processing has been a substantial challenge for developers looking to utilize Intel's second core effectively--parallel processing is a paradigm shift that requires significant retooling on any hardware platform.)
Developers were keen to berate the Saturn back in the mid-1990s, but they're staying mum about the PS3 today. GamePro managed to uncover a few semi-illuminating tidbits, reporting on a story in Dr. Dobb's software journal in which the authors claim "Software that exploits the Cell [processor's] potential requires a development effort significantly greater than traditional platforms." In an email to GamePro, Sony's product development group added further:
Since PS3's Cell processor allows more features -- better physics, more complex graphical processing, lighting or sound -- there is inevitably going to be more cost in supporting those extra features... It's not that PS3 is harder to write for, it's just that you can do more with it.
Of course what Sony calls "more you can do with it," most intelligent developers would equivocate with the same sort of learning curve that killed the Sega Saturn. No one's disputing it's more powerful at this point, but wrapping a Honda Civic around a V12 won't win you any races.
Imagine playing a sport like basketball with a certain ruleset, and say you've been playing for years. As a competitive athlete, you have to focus on mastering dribbling, passing, and shooting techniques which explicitly complement those rules. No running out of bounds, carrying without dribbling, etc. Now let's say the league decides to alter the rules dramatically, and I'm not just talking scoring or timing minutia, but--to make the analogy proportional--something radical, say having to dribble two or three basketballs at once. You might be able to do it (hey, plenty of reverse jugglers can!) but think of the practice and utter rethinking involved to do it effectively. Passing around three basketballs might increase your team's chances of landing a shot, but every tactical aspect of the game has to do a one-eighty.
The most damning element in the stack against the PS3, though, may be multi-platform games (mostly contributed by mega-publishers like EA and Ubisoft) which require costly re-engineering when porting to the PS3 architecture. Witness products like Marvel Ultimate Alliance and Spider-Man 3, which run smoothly on the 360 but often jerk conspicuously on the PS3, and occasionally in a way that actually interferes with the gameplay. Add the higher development costs (read: time) to port to the PS3, and you have a situation not altogether unlike the one Sega found itself in slightly more than a decade ago.