Search

in:

SegaBase Volume 6 - Saturn

Page 1


written by Sam Pettus (aka "the Scribe")


Provided courtesy of


Other SegaBase articles: Older Systems | Master System and Game Gear | Mega Drive/Genesis | Mega CD/Sega CD | 32X | Saturn | Dreamcast

Table of contents of this page

Table of contents


go to: Page 1 | Page 2 | Page 3 | Page 4 | Page 5 | Page 6

Kamikaze Console: Saturn and the fall of Sega

PART ONE OF TWO (October 1993 - July 1996)

Introduction


Starting as early as 1994 but becoming plainly evident to all by 1995, Sega of Japan once again moved to reassert itself as the dominant force within Sega's corporate structure. Its executives had long resented the arrival and marketing tactics of Tom Kalinske, president of Sega of America, who had indeed brought Sega to the pinnacle of its success but only at great expense. Sega now had no cash reserves of which to speak and was operating under a mountain of debt that continued to increase with each quarter. While this was not troubling news to Kalinske and his staff - after all, deficit spending was and continues to be a staple of the American economy - it caused a great deal of concern with his more conservative-minded Japanese peers. They simply could not understand the American business axiom of "spending your way into a profit." They felt that Kalinske was fumbling Sega's transition from 16-bit to 32-bit systems in the one market that mattered the most, so they began to work on Sega CEO Hayao Nakayama, Kalinske's boss, in order to convince him that Kalinske's tactics would inevitably ruin the company. They knew that Sega had to get its act together fast in order to make the 32-bit console transition successfully - one that was as a matter of fact already underway - and were thoroughly convinced that only they could provide Sega with the proper guidance in this new market. Sega of America would be a useful tool and its talents could indeed be tapped as the company's impending 32-bit console transition commenced, but that was all. No more American meddling. Sega of Japan was taking back the reins of power, and it would brook no discontent from the West. That was the plan, anyway. What happened next was quite predictable and should have surprised no one. There is an oft-quoted verse from the Holy Bible that can be found at Proverbs 16:18. It goes like this in the King James Version: "Pride goeth before destruction, and a haughty spirit before a fall." Sega had by now fallen into the same trap that had been the bane of its predecessors, Atari and Nintendo. It was caught in a culture of corporate arrogance. The blame for this rests largely on Sega of Japan, although it must be said in all fairness that both Sega of America and Sega of Europe played right along with it almost to the end. They blustered and swaggered (and spent) along with the best of the Japanese peers, thus adding their own fingerprints to Sega's impending woes. It was only when Sega was approaching the brink that key company personnel in the West realized what was happening and tried to stop it, but by then it was too late. Sega of Japan was not listening and absolutely refused to hear what they had to say - and it was they who now held sway with Nakayama, not Kalinske and the West. This internal dissent over Sega's future meant that the company could never approach the 32-bit nextgen market in the same unified manner that it had successfully assaulted the 16-bit market, and it would eventually culminate in a series of self-destructive mistakes from which the company never recover. Sega, the company that had once led the rest of the videogame market towards the next generation of videogames, would now put itself through a self-destructive fall just as big and swift as its earlier meteoric rise to fame.

The unfortunate console that got caught in this sad series of events is the oft-maligned Sega Saturn, arguably the best and most sophistcated 32-bit dedicated videogame console to ever hit the market. The sad story of how it went from media darling to kamikaze console within two years of its launch has never been told in full until now. It is a troubling tale of how an inferior and less sophisticated system managed to surpass it and capture the hearts and minds of the masses, thanks largely to savvy marketing on the part of its vendor and just as much to one tragic blunder after another from the one company who by all accounts should have been the industry leader. Those of you who are familar with Atari's fall from grace and the humbling of Nintendo will doubtless see many parallels in the story of Sega and the Saturn. Even in the videogame industry, the words of George Santayana hold true: "Those who fail to learn from the past will repeat it." Sit back, reader, and brace yourself for a brush with the dark side of Sega. It is not a pretty story, nor is it meant to be ... for the truth can be ugly, and the truth often hurts.

The shape of things to come


The 32-bit revolution in home videogaming actually got underway in 1993 thanks to an old ally of Sega's from back in the Genesis days. Trip Hawkins, president of Electronic Arts, was one of the industry players who had anticipated the move from cartridges to CD-ROMs and wanted a piece of the action. Backed by the consortium of AT&T, Matsushita, Samsung, and Goldstar and promoted with all the flair for which he was known, Hawkins was able to bring the industry's first dedicated 32-bit videogame console to market just in time for the holiday shopping season. It was known as the 3DO, named after the Hawkins-founded start-up company that developed the design spec. The 3DO was supposed to showcase the future of home videogame consoles - small footprint, CD-ROM storage, 32-bit architecture, programmer-friendly environment, and so on. Each partner in the 3DO consortium produced their own custom versions of the console, although all built them around the common 3DO spec. Hawkins had high hopes for his baby, and sincerely believed that had beat both Sega and Nintendo to the punch in the next round of the console wars. Unfortunately, Hawkins and his backers were in for a major reality check. The astronomically high price of the 3DO when first released (about US$700) meant that very few consoles ever sold. Add to that the major public relations hype from other, more established players in the videogame industry prepping their new 32-bit systems for market and almost no control over the development of its software base, and it is a wonder that the 3DO console managed to last as long as it did. "Trying to be all things to all people doomed the 3DO system to a schizophrenic existence, and ultimately, to extinction," notes CNET Gamecenter's Jason D'Aprille, and he sums it up just about as well as anybody. As for 3DO itself, it was only on the U.S. market for about two years and overseas a little longer despite having one of the most balanced yet diverse software lineups to ever grace any videogame system.


Meanwhile, Nintendo was engaged in a war with former technology partner Sony over a SNES-based CD-ROM console that it had wanted to bring to market. Back in 1988, Nintendo had contracted Sony to develop a "Super Disc" drive for the 16-bit SNES. This device would later be revealed to the world as the SNES PlayStation?, or just PlayStation for short. Nintendo's intent had been to ship the system's CD-ROMs inside a custom caddy complete with an SNES-style lockout chip - a convoluted approach that would have ensured it retained control over the process. Sony understandably balked at this idea - it wanted to put the lockout chip in the CD-ROM drive controller, inside the console, and leave the games alone. This move would also open up the production process, and Sony quitely made plans to license production of PlayStation games to anybody they wanted. Sony president Olaf Olaffson first announced the PlayStation at the 1991 Summer Consumer Electronics Show (CES) in Chicago, proudly proclaiming that "... Sony intends to broadly license it to the whole software industry]." This was anathema to Nintendo CEO Hiroshi Yamauchi, who had no intention of letting Nintendo losing control over any part of the process. He conspired with Sony's rival Philips to publically humiliate Sony the following day at the show. In a public press conference held at 9:00 am sharp, Nintendo's Howard Lincoln announced that it had instead signed a deal with Philips for its new CD-ROM system. The stated reason? Since Philips had invented CD-ROM technology, it could offer superior workmanship. The real reason? Nintendo refused to relinquish control of any part of its proprietary hardware. If Nintendo was going to release a CD-ROM based console on the market, then people would have to come to Nintendo to license it - not some ambitious third-party licensee. "Nintendo believes in a standard - our standard," Yamauchi later said of the affair. Sony saw it differently. "They stabbed us in the back," Olaffson told one of his confidants. The resultant legal and technical hopscotch that Nintendo would be forced to play over the affair pretty much assured that it would not be able to bring a decent CD-ROM system to market in time to ride the crest of the 32-bit wave. Instead, they would have to develop a completely new system from the ground up, launch it after everybody else's systems had already hit the market, and pray that their marketing prowess and company's public reputation would sell the new system for them. Nintendo was unconcerned, though - they thought they had derailed Sony's ambitions for good and went blithely ahead with making money. They were wrong ... quite wrong.


Realizing that revenge is a dish best served cold, to quote an old Arabian proverb, Sony decided to use the experience it had already gained with developing for everbody else and instead release its very own console. It knew what the developers wanted - a simple yet powerful console that was easy to program - and it knew what gamers wanted - a good, cheap system. It had lots of money and lots of connections within the third party community. While it had never attempted to field its own console before, the fact that Sony knew the field of battle and how to negotiate it put the company in a far better position than had been the lot of NEC back in the 16-bit days. About a year after the CES debacle, Sony's Ken Kuratagi was put charge of a top-secret in-house project aimed at developing a brand new 32-bit videogame console from scratch. It had to be cheap to make and sell, yet powerful enough to handle complex 3D graphics of the kind that were becoming increasingly common in videogames. Sony scored a major coup by getting Namco into its console fold early on, but then again Namco needed no prodding - it was still looking for ways to burn Nintendo over the MegaDrive? development affair back in 1990. At the same time, Sony quietly made arrangements with perhaps the most powerful pool of videogame programmers outside of Japan - Europe's third-party community - to develop launch titles for its new system. Psygnosis was perhaps the most prominent of this lot, for it enjoyed a worldwide repuration for its development hardware and software. Soon, like Namco halfway around the world, Psygnosis began developing its own showcase titles for Kuratagi's still-secret wonderbox. So where does Sega fit in all of this?


Like everybody else, Sega began conceiving its own nextgen consoles back in the early 1990s. It is believed that Sega's first official spec for a 32-bit home videogame console was drawn up in the back half of 1992. There is a lot of confusing information regarding Sega's earliest 32-bit console plans; however, at least one specific console concept existed that can in all fairness be said to be the direct ancestor of the Saturn. This was the GigaDrive, as it was known in-house and referenced by some of the industry trades of the day. The name was a wordplay on Sega's earlier success, the 16-bit MegaDrive? (Genesis), thus implying a system that would be even more powerful than its venerable ancestor. GigaDrive? would differ from MegaDrive? in more than just internal horsepower, though. Using its experience with Mega CD (Sega CD), Sega decided that now was the time to abandon the traditional ROM cartridge format for delivering console videogames. GigaDrive would be Sega's first-ever dedicated CD-ROM based console - as opposed to Mega CD (Sega CD), which had for all intents and purposes been an expensive add-on peripheral. Sega knew that CD-ROM delivery for videogames was the wave of the future, so its new 32-bitter was designed to use CD-ROMs right from the start. Like almost all of Sega's arcade and console systems, GigaDrive? was developed by Hideki Sato and his Sega engineering teams. The date of GigaDrive's inception is significant - this was 1992-1993, so Sega geared GigaDrive? as a system specifically designed to better the 3DO, the only other 32-bit console available at the time. It is believed that a small number of working GigaDrive? prototypes were actually built during 1993 in various forms to test the workability of the new console design. It was also during this time that the name of the console was changed from GigaDrive? to that by which we know it today.


Mention the name Saturn to anyone and they will most likely instantly conjure up a mental image of the great gas giant that lies beyond the asteroid belt. The sixth planet of our Solar System and the second largest, Saturn was bright enough to have been discovered by early astronomers thousands of years ago. It was named for the leader of the Titans in Greek mythology, the one who is supposed to have fathered the race of the gods. Its trademark ring system was first observed by telescope by Gailieo in 1616, and is the chief feature by which the planet is best remembered. Because of this, it is often considered to be the most beautiful planet in the Solar system. Made of millions of gravity-trapped asteroids, pieces of interstellar ice, and other such cosmic debris, these rings appear as a giant disc encircling the planet's equator when viewed from distant Earth. One cannot think of Saturn without also visualizing that great spinning disc of ice. Perhaps this is why Sega of Japan chose that name for their system - it was the planet of the giant disc, and that mental image would help reinforce the fact that the console would be using the new storage media of CD-ROM instead of cartridges. Even the console logos for both East and West pay subtle tribute to the planet. Whatever the reason, the fact that they already had a console coming to market named after one planet (Project Mars, aka the 32X) and now were going to release another (Project Saturn) led many industry reporters at the time to conclude that Sega was naming all of its new systems after the planets in the Solar System. Thus was born the myth of the "planet projects," although it must be said in all fairness that Sega went ahead and played along with it. They knew their reputation with both the industry and gaming public was not what it once had been, so the decision was made to accept any free publicity that came along. If that meant perpetuating an unfounded myth, then fine - and with that, other consoles in the "planet series" (Venus, Jupiter, Neptune, Pluto) also made their public appearance in like fashion.


What most gamers today fail to realize is that the Saturn which Sega of Japan conceived and prepared for market back in 1993 was not the console that eventually made it to market in 1995. The original Saturn spec as it has been described in some accounts seems to owes most of its design to two of Sega's newest arcade boards at the time. Both Sega System32 and its immediate successor, the famous Sega Model 1 arcade board, were based around the 32-bit NEC V60 16 MHz CPU. Both designs had single CPUs and single VDPs, with fairly straightforward design architecture. System32 was Sega's ultimate 2D videogame board, whereas Model 1 had been developed exclusively for videogames with 3D polygonal graphic engines. Only four videogames were ever made for Model 1 before Sega quickly proceeded to its successor (the more powerful Sega Model 2), but all four were the best 3D arcade games available at the time and one of them in particular - Yu Suzuki's Virtua Fighter - quickly gained a worldwide following. Basing its home systems on proven arcade hardware was a time-honored concept with Sega, and they were not about to stop now. There is a lot of debate as which of these two boards actually served as the basis for the original Saturn; however, there is little debate that the Saturn as originally designed was far less powerful than it is now. Reports from the day indicate that Sega of Japan originally intended the Saturn to be the ultimate 2D videogame console, with 3D games more or less an afterthought. 3D hardware was quite expensive to produce and vend at that time, as Sega knew all too well, and this factor tends to weigh in favor of a System32-influenced original Saturn. Even so, assuming that the original Saturn was intended to handle 3D games from the onset and matched the more powerful specs of the Model 1 board, with its then-revolutionary 3D graphics capabilities, you still wind up with a single-processor design that was far less capable than what Saturn eventually became. So how did the Saturn proceed from its original, fairly simple single processor architecture to "the mess" it eventually became? That is a very good question, and it is quite a story in and of itself.


Back to the drawing board


The period from October 1 to December 31 of the year 1993 would prove to be crucial for all of the major players concerned in the 32-bit incarnation of the videogame industry's so-called "nextgen wave."

Matsushita, the parent company of Panasonic, would release the very first version of the 3DO to its American customer base in October of that year. Japan and Europe got theirs the following March. The system itself would not prove to be all that successful in comparison to the others of its generation, but a sudden proliferation of Oriental adult-themed 3DO software would aid in boosting Far East console sales toward respectable levels. While many other titles in many other genres would be produced, it would be adult software, coupled with the non-restrictive development policies behind the system, that would eventually be one of the chief reasons why the 3DO would manage to do as well as it did despite its many other problems. It is perhaps the greatest irony in the entire 3DO saga.


Nintendo, the resurgent ruler of the videogame roost, would abruptly cancel its SNES "Super Disc" drive system the following month, offering no official explaination as to its sudden change of mind. This came in the wake of continued assurances from the company that it had definte plans for CD-ROM media and was continuing development of the system. By this time, Nintendo executives had realized that they had missed their chance to join the 32-bit revolution at its most critical juncture. Since their competitors would be fighting it out on the front end of the nextgen wave, the best thing for them to do was to batten down the hatches and ride it out until their own new system, Project Reality (aka the N64), would be ready for market. It would be the last of the nextgen systems to arrive, no earlier than 1995 or 1996 at best estimates, but the company would have time to refine its design and learn from its competitors' mistakes. It was a good thing that Nintendo had the deepest pockets of any player in the field save newcomer Sony, because it was going to be a very long wait for a nextgen Nintendo console.


Sega was now the number two player on the field but still wielded its worldwide reputation for programming excellence. November of 1993 saw the arcade debut of Yu Suzuki's Virtua Fighter, arguably the most revolutionary fighting game to hit the videogame industry since the debut of Capcom's Street Fighter 2 franchise. While the setup and gameplay itself were really nothing new, the full-screen, full-color, fully rendered 3D polygonal graphics were and blew just about everybody away. Nobody had really talked about arcade hardware until Virtua Fighter came along, but all of a sudden the term "Sega Model 1" was on just about every gamer's reverent lips. It was no wonder that Sega also got a lot of attention when it announced its new 32-bit Saturn home console that same month. As announced, the original Saturn was to be the ultimate 2D gaming powerhouse, with a "modest" 3D polygon capability. It had better be more than "modest," most gamers prayed, but there was a reason for this. 3D hardware like the Model 1 was expensive, and Sega wanted to keep the cost of the system down. The company had no intention of going though the same tribulations that Trip Hawkins and his 3DO backers were currently suffering. In the months that followed, Sega's Virtua Fighter would become the new holy grail of fighting games and it seemed obvious to all that Sega must have a Saturn port in the works. It would make an ideal launch title for the system, given its popularity. A lot of eager gamers began to quietly sock away their cash for Sega's new box, praying that Saturn's "modest" 3D capability would be sufficient for a good port of their favorite arcade fighter. All they could do was wait ... and hope.


November of 1993 also saw Sony make its formal entry into the 32-bit console sweepstakes. It announced the formation of a new company division, Sony Computer Entertainment (SCE), responsible for the corporation's computer-related ventures. The first project on the drawing boards of SCE was their new 32-bit videogame console, which was revealed to be already under development. Referenced variously as "PSX" or "PS-X" in many trade reports of the day, this was the system that would eventually be renamed the Sony PlayStation?. The name was a thumbing of the nose at Nintendo for all the trouble that they had caused Sony, and now it would be the latter's turn to suffer. Sony had the money, they had the marketing muscle, but most importantly, they knew had the machine and the software to go with it. The official system specs were released in December of 1993, and just about everybody's jaw dropped once they realized what Ken Kuratagi and his team had wrought. PlayStation was a jack-of-all-trades, with top-of the line integrated RISC architeture that bettered anything that was or would be available on the market at that time. Its 2D graphics outstripped those of the SNES, its 3D graphics were as good as or better than anything that Sega's arcade offerings or high-end PCs had to offer, its speed easily outpaced the aging Genesis, its ability to do both complex 2D and 3D processing appeared to be unmatched, and its double-speed CD-ROM drive meant faster loading times than its aging and quirky competition from Sega and NEC. Best of all, the development libraries that Sony already had on hand for potential third-party supporters made the new console dreadfully easy to program - and that gave the Sony PlayStation greater appeal within the videogame industry than Trip Hawkins and his 3DO team ever dreamed. Sony was hoping to win away developer support from Sega and Nintendo, the latter in particular, and they succeeded. As 1994 began to unfold, one third-party vendor after another began to express public support for the new kid on the videogame block and what he had to offer them.


It has been said that when Hayao Nakayama finally realized just what Ken Kuratagi and his fellows had created, he called his entire R&D department up to Sega "flag country" and proceeded to give them the ass-chewing of their lives. One Sega staff member at the time would later recall that Nakayama "was the maddest I have ever seen him." Nakayama had obtained a copy of the design specs for Sony's new PlayStation and had compared them to Sega's own Saturn. That was why he took the time to bawl out his own R&D staff. He knew beyond a shadow of a doubt that they had blown it as far as Sega's chances of seizing the 32-bit market for its own, just as it had done with the 16-bit market a mere five years before. Sega was now in trouble. Big trouble. What was the problem? Raw 3D processing power. That was the problem with Saturn.


You will recall that the Saturn as originally designed was not the 3D powerhouse that it subsequently became. Remember, the original spec appears to have been originally designed around the 16 MHz NEC V60, a traditional CISC-type CPU that had been the first 32-bit microprocessor widely available in Japan. In contrast, the PlayStation was built around a 33 MHz MIPS R3000A, a faster and improved version of the R2000 RISC-type microprocessor that Silicon Graphics had been using in its SGI workstations for years. The NEC V60 was rather obscure, whereas the MIPS R3000A was by now one of the mainstay processors in the CAD/CAM industry. It had replaced its venerable ancestor, the R2000, on the budget end of the scale and was in use in such notable graphics workstations as the Silicon Graphics Iris Indigo. Sony had been working with and manufacturing MIPS processors for years, so its engineers were fully aware of what the R3000A could do and how to do it. It was with this in mind that Sony's public relations department hyped the console's theoretical limits for all they were worth. The PlayStation's Geometry Transfer Engine, specifically designed by Sony engineers for the console, supposedly gave PlayStation the ability to crunch 66 million instructions per second (mips), tossing out a theoretical maximum of 1.5 million flat-shaded triangular polygons and 500,000 texture-mapped and light-sourced polygons per second (pps). As given, those figures were more than double the maximum capability of Sega's vaunted Model 1 arcade board. While programmers who actually knew the hardware vociferously argued about Sony's numbers (and some still do), they were not the ones looking at the paperwork. Ill-informed industry reporters and boardroom corporate types were, and the numbers Sony gave them seemed to lead to an inescapable conclusion. Sega's Saturn, as announced, was more than outclassed by the PlayStation. As it seemed to stand, Saturn wasn't even in the running anymore insofar as the 32-bit race to glory was concerned. It appears to have been with those numbers in mind that Nakayama promptly charged Sega's engineers with the daunting task of fixing the Saturn's problems in less than a year ... or else. While he may not have entirely agreed with the rationale and reasoning of his superior, nevertheless Hideki Sato and a handpicked team of 27 Sega engineers (known as the "Away Team") began work immediately on creating a brand new system to replace the inadequate concept of old.


There is a common myth within the videogame community nowadays that all Sega did in their attempt to fix the Saturn's problems was just slap in another CPU and an extra VDP. While this is true in a sense, that is only part of the story and does not even give lip service to the real fact of the matter. The end result of Nakayama's dictate was that the Saturn was more or less torn apart and redesigned from the ground up in a concerted effort to create a system that compete with the PlayStation on its own level. There wasn't time to be picky and carefully craft an all-new, highly integrated 32-bit design using the best components available like Sony had done. The announced launch of the Saturn was less than a year away (November 1994, in fact), and Sony was due to release the PlayStation at the same time. If Sega went with the Saturn as originally designed, then they would be slaughtered by Sony as soon as both systems launched. A redesigned Saturn that could successfully compete with the PlayStation was going to have to be made from off-the-shelf parts using whatever Sega had available in order to meet deadline, and this where the concept of parallel processing enters the picture. The Away Team chose this as the most expedient shortcut towards getting a redesigned yet decently priced 32-bit console out the door in the shortest amount of time. Instead of the single NEC V60, they went with dual Hitachi SH-2s in parallel - supposedly as a favor to an old golfing buddy of Nakayama's. Instead of the single VDPs of the earlier arcade boards, they went with beefed-up dual VDPs - each of which could be programmed for dedicated tasks. If this strikes you as odd, please bear in mind that parallel processing was an old concept to Sega's engineers. Many of Sega arcade hits from the 1980s, such as AfterBurner II and OutRun, utilized twin Motorola MC68000s in their board design. The Mega CD (Sega CD) can in fact be said to be Sega's first-ever dual processor console, since its internal MC68000 worked in tandem with the MC68000 of its host MegaDrive? (Genesis) console. The Saturn was to be Sega's first purpose-built dual-processor console. This was in direct opposition to a proposal that was already on the table from Tom Kalinske and his staff from over at Sega of America. They had contacted Silicon Graphics, one of the companies behind the PlayStation's 3D graphics capabilites, and had come up with an alternative, single-chip simplistic design that they were convinced could compete with PlayStation on its own terms. To their surprise, Nakayama overruled them in favor of the Away Team's proposal. He had been unimpressed by a demonstration of the technology arranged by Kalinske, remaining convinced that Sato's dual-processor concept was actually the more flexible choice of the two. His decision left a bad taste in Kalinske's mouth, who sensed even at this early point that Saturn was going to be a doomed venture. "The Japanese are making the decisions for the U.S. market," he later grumbled, "and they do not know what they are doing."


Unfortunately, Sega's decision to convert Saturn into a dual-processor system wound up create nightmares for third-party developers early on. The parallel processing issue, with regard to both the dual CPUs and dual VDPs, was probably the first real problem to manifest itself with the Saturn. The theory was simple enough - individual system tasks could be broken down and split among the various processors for greater processing efficiency - bu it was no easy task to get all that hardware properly synched and singing in harmony, especially for an industry that was not used to the concept. The academics might have been playing with parallel processing for years, but not so the rest of the videogame industry. Sega was one of the few players on the field who had experience with the concept, but not so the new bloods at the software houses. Many of these developers would eventually content themselves with using just one of each of Saturn's CPUs and VDPs, thus limiting the system resouces available to them and making their games far less than they could have been. Even when these types did manage to pull off a good game, they would argue that the extra coding effort involved in "working around Saturn's architecture" eventually resulted in a 25% drop in overall system efficiency due to shared resources, thus reducing the chief benefit that parallel processing was supposed to be. The good Saturn coders maintained that this was complete nonsense - a poor excuse put up by "crap programmers" who didn't know how to manage the Saturn's "beautiful design" - but these few lone voices tended to be drowned out by the rest. It was the opinion of most programming experts at the time, including the likes of Sega's Yu Suzuki and Bullfrog's Peter Molyneux, that the only effective way to produce a good, fast game on Saturn that could compete with a comparable PlayStation title was by programming in pure assembler. The following quote on the matter by Sega's own Yu Suzuki comes courtesy of NextGen magazine.


Trying to program for two CPUs has its problems. Virtua Fighter uses a different CPU for calculating each character. The two CPUs start at the same time but there's a delay when one has to wait for the other to catch up. One very fast central processor would be preferable. I don't thank that all programmers have the ability to program two CPUs - most can only get about one-and-a-half times the speed you can get from one SH-2. I I think that only one out of 100 programmers are good enough to get that kind of speed out of the Saturn.On the other hand, Sony had made things easy on the third parties by setting up its development libraries in the C programming language instead of assembly language, so most of the "crap programmers" naturally gravitated toward the system that was easier to code. So did a lot of their employers, too - much to the chagrin of the real programmers, who continued to argue (and many maintain to this day) that Saturn's dual processing design made it the better of the two machines. Saturn may have been the equal of, perhaps superior in some ways to, Sony's Playstation at the assembly language level, but Sony had effectively changed the field in its own favor. Thanks to its canned development libraries for the PlayStation, videogame programmers were happily becoming adjusted to coding in C and not messing with assembler anymore. Most didn't want to go back, which left only the real or old-school types to wrestle with the untapped potential of Saturn's dual-processor architecture.

The problem of Saturn's twin VDPs also had its own unique issues insofar as the console's graphics capabilities were concerned. In theory, Saturn now had the exact same 3D processing capability as the PlayStation, signficantly more 2D sprite capability than Sony's box, and far more graphics computational capability than Kuratagi's creation. These facts tended to get lost in the shuffle whenever the console specs are compared side-by-side. Also, the word had gotten out thanks largely to the videogame magazines of the day that the VDPs that Sega used for its hardware did not perform 3D processing in the same manner as everybody else's in the industry. They calculated surfaces in "quads" (four-sided sections) instead of "polys" (three-sided sections); thus any simple comparison of polygon count wasn't really worth the paper it was printed on. In theory, using quads instead of polys gave the Saturn a 3D processing boost; but supposedly developers used to standard polys found it quite difficult to readjust their mindset for Sega's quads. Videogame magazines such as GamePro and Game Informer took this comparison and ran it for all it was worth, thus building public perception that Saturn's graphics hardware was actually inferior to that of PlayStation. It simply was not true, and the comparison itself was flawed thanks to a basic misconception of how each machine performed its graphics processing. Both Saturn and PlayStation worked with quads; however, the vagracies of the PlayStation's GTE tended to "tear" and "warp" its quads, hence the tendency of its programmers to stick with polys. Saturn was fully capable of doing polys; however, it had been designed to achieve its best performance with quads and its programmers were encouraged to work with quads instead of the less efficient polys. Actually, the real difference between the two systems was not in how many megahertz their CPUs could crank or how many vertices their GPUs could count. It was in the effects that each added to the graphics they generated on-screen. There were many things that the Saturn could accomplish through pure software alone thanks to its raw processing power - texturing and shading, lighting effects, MPEG playback - but few programmers bothered to learn how to do it when Sony's "canned" libraries and built-in hardware were already there waiting for them to exploit without effort. Saturn's twin VDPs could even be independently programmed for dedicated tasks, but this was an ability that would eventually go practically unexploited due to the progamming difficulties in properly tasking and synching them. Only rarely would third-party Saturn efforts compare favorably to their PlayStation cousins once full-blown development for both systems got underway, and it would be long after launchtime before Sega could release an software developer's kit (SDK) that could truly permit developers to start unlocking Saturn's deep resources. That would prove to be the heart of the matter - proper parallel processing requires a suitably developed and exploited programming environment, and it was one that simply was not available for the typical programmer before and in the year after Saturn was officially launched. Sega simply did not have time to create a good parallel programming enviroment prior to bringing the Saturn to market, and its third party licensees did not have the resources available to properly exploit the system in their first-generation efforts. On the other hand, Sony made damn sure that the PlayStation's development environment was set up and ready to exploit as soon as physically possible, so most software companies were able to literally hit the ground running even with an army of the so-called "crap programmers" that the real codeheads tend to deride. The lack of a sophisticated programming environment was a fatal flaw that was built into the Saturn from the start. If Sega had somehow manged to find the time instead of putting this task off for about a year ... well, things might have gone quite differently than they did. Veteran programmer Steve Palmer, author of the classic sports videogame NBA Jam and currently working with Pitbull Syndicate, who developed for both systems during their entire respective lifetimes, sums up the Saturn development problem quite well. Sega gave us exactly what we wanted; however, the industry changed at exactly the same time, so we no longer had a choice in the matter. Things suddenly had to be finished yesterday. Sega could not have forseen this change .... Most of the third-party crowd couldn't get it to do what they wanted it to do quickly enough, so they didn't bother. It's amazing, because it didn't require much effort to get the machine to perform on a PlayStation level. Programmers being programmers, though, they probably were not happy unless they felt they were pushing the machine, and it seemed like too much effort to do that.


To learn to program the Saturn was to learn the machine. To learn to program the PlayStation was to learn C. Learning C is much easier than learning the hardware of a new machine, and with the Saturn, there was a lot of hardware to learn .... There was not enough time for people to learn the hardware. The same would have been true of the PlayStation, except you didn't need to learn how to talk to the hardware. The libraries took care of that for you. Sega's approach was to release hardware documentation for every aspect of the Saturn. That was understandable - it was the way everyone had done it before, and it's what programmers were used to, but the industry had changed. Video games were no longer a "niche" market, and the "big boys" had moved in. Time is money. Nobody was given the time to learn new hardware anymore.

By the time Hideki Sato and his fellow engineers at Sega got through with it, the redesigned Saturn had become a machine that on paper was fit to compete with PlayStation. However, that same redesign had resulted in a system with a higher shelf price than Sega would have preferred, and it was now a console that had lost its previously close-knit and programmer-friendly design architecture for the sake of market expediency. These twin issues would lead to all sorts of future problems once the Saturn itself finally cleared the prototype phase and headed for the production lines.

It is now May of 1994. The American videogame industry is abuzz with news about the 32X, the end result of Project Mars and the current darling of the U.S. videogame community. Rave reviews about the new system and its up and coming games fill the trades, and it looks like Sega of America just might be able to hold its own against Nintendo's Donkey Kong Country threat. In the meantime, though, the Japanese press is excitedly talking about the two new 32-bit CD-based videogame consoles that will be coming out in the fall. One will be from upstart Sony. It is an unknown commodity, this ... PlayStation ... but it certainly looks promising. The other is from Sega, a tried and true veteran of the console wars, and the company's reputation for gaming excellence preceedes the brand-new system that it is about to unveil to the public. Enter the Sega Saturn.




continue to Page 2




Contributors to this page: Eidolon .
Page last modified on Thu 10 of Jan., 2008 19:36:45 CET by Eidolon.

Current Poll

All your base are belong to us
  • you bet!
  • what?!?
View Results
(Votes: 198)
Cookies must be allowed to vote

Online users

71 online users

Shoutbox

tomman, 09:00 CEST, 2014/04/08: (Note to myself: try Fusion on Debian Jessie...)
tomman, 09:00 CEST, 2014/04/08: Twitter? Social networking? HELL NO :P
Eidolon, 11:05 CET, 2014/03/22: Good to see you around, Steve! Just installing Fusion on my new laptop to enjoy a round of Lunar The Silver Star once again :)
zyrobs, 00:32 CET, 2014/03/08: I would but I can't figure out how Twitter works. Typical Web2.0 BS, optimized for stupid people...
King, 16:59 CET, 2014/02/26: Followed!
sith-smasher, 14:46 CET, 2014/02/22: A sign of Snakey; good that you are ok man! I hope life is treating you better than before. Peace.
Snake, 06:01 CET, 2014/02/20: Everybody in the world follow me on twitter @RealSteveSnake - just because.
King, 20:57 CET, 2014/02/17: SNAKKEEEEEEEE! :D
Snake, 19:21 CET, 2014/02/16: *waves*
King, 18:49 CET, 2014/02/15: Awesome, let him know we miss him and that we're glad he's okay! :)
RSS feed Wiki RSS feed Articles RSS feed Forums