It’s 25 years since Sega of America made its biggest mistake | Hacker News Reader

It’s 25 years since Sega of America made its biggest mistake

forbes.com

120 pointsraudaschl117 comments

This narrative has been pushed quite heavily by former Sega of America CEO Tom Kalinske, but it leaves a lot out. Kalinske claims the Japanese president of Sega, Hayao Nakayama, forced him to launch the Saturn in NA before Kalinske was ready, and because of that, the Saturn failed and Kalinske resigned shortly after. As always, it's best to approach with caution when CEOs are assigning blame for company failures under their tenure.

The reality is far, far more complicated. Kalinske himself was against the Saturn going back to 1993, due to the predicted high cost of the console. This in turn led to the development of the 32X add-on for the Genesis as a low-cost entry into the 32-bit generation in NA, but the 32X failed spectacularly.

Most relevant, however, is that Sega failed to adequately compete against Sony in terms of garnering third-party support, both in Japan and NA. This is discussed at length in the excellent book Revolutionaries at Sony by Reiji Asakura (English translation available). The Saturn was difficult to develop for and Sega did not have good development tools early on.

Also worth reading is the recent account from former Sega president Hideki Sato, who was the head designer of the Saturn. He discusses many of the shortcomings of the console and Sega's strategy for it:

https://www.sega-16.com/forum/showthread.php?33506-Hideki-Sa...

Great comment, and I agree. I do think Kalinske's rationalization is bullcrap ... but it also alludes to some truth, namely that Sega America and Sega Japan were not aligned on console strategy and were downright adversarial - and that certainly contributed to the failure of Saturn.

I agree that this is way more complicated.

Nobody knew what 3D hardware would look like in 1994, they were too busy inventing it. For example, the Saturn provides quads as your primitive, which seems weird to anybody looking at the Saturn today. The PlayStation and Nintendo 64 both used triangles.

You might have various reasons to prefer the Nintendo 64 or PlayStation, but from a developer's perspective, the main reason you would prefer the Saturn is probably because of its 2D performance. But a higher price point for better 2D performance is a tough sell.

It's also interesting to look at the different companies through the lens of what their strengths are. Sony has generally had pretty solid hardware design, Microsoft has generally made systems that are easier to develop software for, etc.

Better 2D performance and more RAM. The lack of RAM kneecapped the Capcom fighting game ports on PSX.

The Saturn’s RAM expansion pack helped too: https://sega.jp/fb/segahard/ss/ram.html

Interesting discussion of the quad system: https://forum.beyond3d.com/threads/why-did-gpus-abandon-quad...

You're missing three part where Kallinske was trying to make a deal with Silicon Graphics for the hardware that Nintendo ended up buying for the n64. This would have been a much better outcome. Sega Japan shut it down simply because Kallinske came up with it and they weren't going to answer to Sega America.

Kallinske was also agaisnt the 32x.

Dreamcast was fine. It was a good machine with a good library, but it was too late by then.

I've read in the past (sorry, don't have the link right now) that Sony also did some hardball with licensing.

In particular I've been told that they threatened Capcom over Mega Man 8 and X4, saying that if those didn't also show up on PSX, Capcom would lose their publishing license on PSX. If true, that would certainly add a bit of heat on licensees to publish on PSX and would result in Sega losing a lot of third-party exclusives.

>The Saturn was difficult to develop for

That's something of an understatement.

https://en.wikipedia.org/wiki/Sega_Saturn#Technical_specific...

The Saturn always struck me as overdesigned and underengineered. So many custom chips to program and coordinate and a high unit cost when the simpler and more flexible approach taken by Sony ends up being just as good and cheaper almost all of the time.

>the simpler and more flexible approach taken by Sony

I mean, up to a point? The PS1 was so simple it didn't have a floating point processor. https://www.reddit.com/r/gaming/comments/bkedc/heres_a_quest...

Not having dedicated floating-point hardware seems like a totally reasonable trade-off, and is common even today for applications below a certain price point. They chose the MIPS R3051, even though the relatively low-cost R3500 was available then; the R3500, with an R3000 core plus integrated R3010 FPU, MMU and memory buffer co-processors, would have blown the budget for the console. At launch in 1994, the PS1 shipped with a fixed-point GPU (with GTE for 3d matrix math), and certainly held its own against the competition.

The PS1 platform prospered because it was cheap, and relatively welcoming for developers (lots of tooling, libraries, infrastructure), and it ended up selling over 100 million units.

It's very tough to argue too much with that kind of success.

The Saturn also lacked an FPU.

I wise sage once said, https://youtu.be/Vhh_GeBPOhs?t=2 when you aren't number one, you can't afford shenanigans.

Hard to remember, but in North America at the time, Sega was #1.

Sometimes if you are number one, you still can't afford shenanigans.

The lesson is that no one can afford shenanigans, at least not for long.

Sega of Japan didn't remember that even when it was happening.

I feel like the whole 32x mess was an important part of this mistake.

"Hey let's release a weird upgrade pod for our successful last-gen console almost completely simultaneously with our next-gen console, I am sure developers will be happy to split their efforts across two machines! Especially if both of them have really weird architecture that's hard to program on."

now its just "lets do a mid-life separate SKU with the same name and everything is compatible, but at least the airflow is better, maybe cheaper to manufacture, and maybe it outputs higher resolution"

its a better outcome, reminds me thats what happens when we allow companies to fail

There was a Genesis 2 and a corresponding Sega CD2 which were smaller than the originals

The Sega CD2 was actually vastly better, in my recollection. I explicitly got that one and attached it to my full size genesis back in the day.

I always thought you'd need the matching version, but apparently not

https://i.imgur.com/7jlcbQ1.jpg

My company was deeply involved with Sega Saturn development at that time. One of the problems was a serious rift between Sega of America and Sega Japan.

SOA was committed to the idea that the Sega Genesis had a few more years of life in it, and that the American audience was not as interested in new technology for its own sake. They were, rightly in my opinion, concerned that a change in platform would benefit the competition not them, given that they had a market-leading position at the time.

Japan was, also correctly I suspected, convinced that Hardware Supremacy was essential to maintaining their Market position. Unfortunately, internal politics at Sega Japan caused the Saturn to be overly complicated and expensive to produce. I also think Japan failed to recognize that Americans were much more price conscious and less status- conscious than the Japanese Market.

We all know who won in the end. I don't have any real evidence of this but my intuition is that Sony allowed their American arm more latitude and gave them more credibility with regard to designing the market strategy.

Sega allowed internal politics and the Japanese Centric vision to Cloud their decision making process.

And then they doubled down on their flawed strategy with Dreamcast... And the rest is history

>I don't have any real evidence of this but my intuition is that Sony allowed their American arm more latitude and gave them more credibility with regard to designing the market strategy.

In reality, the exact opposite happened. Ken Kutaragi (creator of the PlayStation) and Norio Ohga (then-president of Sony) were outraged that the Sony executives in America were not following their instructions, and they fired the vast majority of them. This is discussed at length in the book Revolutions at Sony by Reiji Asakura. The American side wanted to make all kinds of changes - they hated the name PlayStation, they hated the grey color (they wanted it to be black), and they hated that Kutaragi would not let them include a pack-in title for free with each console. The Japanese side quickly responded, fired most of them, and took direct control.

How ironic given that the Net Yaroze, N64, and then PS2 were all black.

Maybe the goal was to keep the design and implications of the "Nintendo Playstation" design?

That's interesting. As I said I work with Sega so I didn't have any first-hand experience with Sony at that time.

So given what you're saying here I'll amend my perspectives to say that Sony Japan just understood the American Market better than Sega Japan.

I wouldn't say that the Japanese side of Sega wasn't aware of the issue of pricing the console too high. In fact, Hayao Nakayama often spoke of this in 1994 - he was very sensitive that a high-priced console would not sell in America (and used the 3DO as evidence). He was quoted a few years back as saying that the reason he approved the 32X was because Sega of America were insistent that the Saturn was priced too high to ever sell there (the quote is in the book Sega Mega Drive Collected Works).

However, the 32X failed spectacularly, and that left Sega with nothing but the Saturn. Their response was to price-match the PlayStation, which seriously pushed them into the red. There wasn't much that could be done at that point.

"he American side wanted to make all kinds of changes - they hated the name PlayStation, they hated the grey color (they wanted it to be black)"

I have to love that these aesthetics the Americans were upset about made no difference to the success of the PlayStation.

Sounds like a classic case of bikeshedding - the execs in question probably weren't qualified to have opinions on hardware design or software development, so instead they focused on entry-level stuff like the name and the physical appearance.

Yeah, and that makes Sony JP firing them so much more satisfying.

I inherited an NES (something my parents would otherwise never have allowed to happen) and never owned another game system. My knowledge of the game systems were pretty removed. But, I know that I cognitively grouped the PlayStation with the SuperNintendo and my desktop computer at the time precisely because of the matching colours. I think it was an intentional design association that helped "understand the place" of those devices ("this is a computer"). Kinda like how my earliest tv was a huge carved wooden piece of furniture with fabric accents. The design told you "this is furniture".

It's hard to say if that's entirely true. Sony did end up changing the system's color and making the controller bigger and bigger on subsequent iterations of the PlayStation, so apparently they came around on those ideas.

I’ll agree to some degree, but PlayStation sold twice as many units as Nintendo 64. We could list the differences between consoles and point to some more important factors first - price, abundance of titles available, etc.

Or maybe it succeeded in spite of the name, color, and small controllers and would've been even more successful if the Americans got their way. I guess we'll never know.

It sold 40 million units versus the N64’s 20 million units. N64 was black? Big whoop. Price, abundance of titles, and some other factors come before anything like that.

Oops - made a mistake here. The American side wanted the PlayStation to be WHITE, not black. I went back and checked the book. Quote:

>First he [Michael Schulhof - CEO of Sony Corp of America] objected to the color of the console. [He] insisted that gray was unacceptable in the U.S. market, and that the console must be white. Neither did he like the design or the logo mark. Their approach was to object to everything on grounds such as the results of market research: "We can't accept such an unusual controller. The design is too small for American hands." What is more, they insisted that they would set the U.S. list price themselves and they disapproved of the name PlayStation. The "Play" in PlayStation, they said, was reminiscent of "Playboy" and might be misconstrued. With one issue after another, the criticism was relentless.

>...Maruyama carefully assessed the likelihood that U.S. management would respect the intentions of management in Japan. He concluded that it would be impossible for managers steeped in the conventions of the game industry, and he decided to replace the lot except for a select few. In January of 1996, Sony established subsidiary SCEI America in San Francisco, simultaneously replacing most of the managers and launching a new management team. Maruyama comments: "We swept the organization clean of all the old obstacles. We realized that we had to manage our own business."

I previously transcribed more of this excerpt here:

https://www.sega-16.com/forum/showthread.php?34167-Sega-and-...

And here is the news story about Michael Schulhof being forced to resign:

https://www.baltimoresun.com/news/bs-xpm-1995-12-06-19953400...

I remember being a child at the time and thinking the 32X was a neat, if somewhat pointless add-on.

Somehow we got one, but I think it was after Sega had already pulled the plug on it, so there were some pretty insane discounts at the time where you could get the console, Virtua Fighter, and Star Wars Arcade for like fifty bucks. Great for consumers! But I'm guessing Sega wasn't making much (if anything) from it at that point.

I have always kinda felt that SOA seemed to be working at odds with Sega Japan a lot of the time! Now I know it wasn’t just me, thank you.

Totally this, and if I recall the upgrade was like $170 itself which was really expensive for teenagers at the time, plus it was confusing, sega genesis, add the 32x, Saturn comes out in a few months, oh and also sega CD has been around a for a year or so, it felt really disorganized. I felt the only thing sega had going for it was non-censored games, mortal kombat for snes had non-red blood in the fights, genesis did not.

Well, they had great sports games; which didn’t matter to me at all, superior console ports of arcade games, and, of course, Sonic.

And then they utterly failed to launch a major original Sonic title for the console. shakes head

> I am sure developers will be happy to split their efforts across two machines

2.5 machines. There were a few SegaCD+32X games.

I remember being a kid and not knowing what kind of Sega I was supposed to want. Sega CD came out in 1992 and CDs were the new thing - most of the next-gen consoles were CD-based. But the 32X was newer (1994). But when the 32X was released, we already knew that the Saturn was coming. The Saturn was released in Japan the day after the 32X was released. It also meant the release of three consoles in less than three years.

For parents who had no interest in the systems, Sega seemed like a way to just keep spending money. First the kid wants Sega CD, then they want 32X, then they want Saturn. During the CD and `-bit` wars, Sega's lineup was weird and hard to decipher. Even if you were a kid that might know why a Saturn was better, how do you explain it? "Yea, it's the same 32-bits as the 32X and the same CD as a Sega CD, but it's better dad/mom!"

Nintendo had an easy story: it's been over 5 years since you bought me a Nintendo and this one goes from 16-bit to 64-bit! Sony had an easy story: it's 32-bit, only $299, CD-based, and Sony is the best name in all electronic stuff!

If you'd convinced your parents to buy a 32X for Christmas 1994, you weren't getting a Saturn 6 months later. "And what about the Jupiter that'll be out in another 6-12 months?" any parent could retort.

Sega burned developers with too many machines to develop for. Sega burned consumers with too many machines that would be quickly abandoned.

All the Sega stuff after the Genesis/Megadrive seemed really tied to their arcade efforts, in a way that Nintendo (and, toward the end of Sega's time in the market, Sony and Microsoft) didn't. AFAIK the 32x, Saturn, and Dreamcast were all only a little more different from their arcade counterparts than the NeoGeo was, which is why there were so many very faithful arcade "ports" on them (Star Wars Arcade and NBA Jam on the 32x, stuff like that). I've not seen that factored in to histories of the death of Sega's consoles, but wonder if it contributed in some fashion, or else helped stave off their demise.

AFAIK the 32x, Saturn, and Dreamcast were all only a little more different from their arcade counterparts than the NeoGeo was

I think this is less true for the older hardware and even the DreamCast is somewhat more limited than its arcade board cousin.

The way in which Sega ultimately approached its console hardware business was actually similar from start to finish; you just have to look at the pre-Mega Drive consoles to see that the Mega Drive itself was the shining exception within a strategy that was very "spray and pray".

First there was the SG-1000 and the SC-3000 computer in 1983. Then there was a mostly cosmetic update, the SG-1000 II. Then there was the Sega Mark III and Master System; the Mark III had a variant release with an FM sound chip. All of these releases happened within a span of four years, 1983-1987. Throughout these releases there was a heavy focus on arcade ports, and Sega struggled with marketing the console as its own kind of experience.

When the Mega Drive came out in 1988 it was a big enough leap to be a stable target for a few years, and then Sega reverted to their previous ways. To the extent that Sega "got" their console business, it was a case of a few teams in various departments and subsidiaries that bucked the trends.

Not to argue with your point because it is certainly valid, but the N64 had that expansion module for DK 64 and a few other games, yet managed to survive that.

The expansion module was just more RAM, not a whole new system. I think the 32x is more comparable to the N64DD, which failed and never actually made it to the US.

If i remember right the expansion module was included with DK64 which helped ease the pain a bit.

Heh, my friend believed it turned his 64bit system into a 68bit system.

The fun fact about N64 memory expansion is it was on a 9 bit bus ;o) Rambus at that.

The expansion pak was only required for 2 or 3 games, and bundled with one of them.

Most of the games with epak support used it to increase quality and / or framerate, the game worked fine without. It was also relatively cheap ($50).

> It was also relatively cheap ($50).

Which is kinda impressive (for the time at least) considering how expensive RDRAM was/is relative to e.g. SDRAM.

For me as a kid, there was this sense of confusion and whiplash when it came to post-genesis hardware. The CD came out in 92, with no real notable games at launch: Sonic CD came out a year later, and Snatcher + Lunar at the end of its life in 95. The 32x came out shortly after in 94, and only saw a handful of games before being buried by the more powerful Saturn in 95.

With two major peripherals and a new console in a short period of time, each with its own library of games, it was hard to understand all that as a kid without internet access, much less afford it, so I stayed clear and sold my genesis for a SNES, then got an N64.

The short hardware life probably spurned a lot of 3rd party developers too.

Nintendo was much smarter, with longer cycles between major consoles, and no expensive enhancement add-ons to segment its games. When you bought a Nintendo console, you knew it would have a 5+ year road map and was a good investment (okay, ignoring the virtual boy). Every gameboy has had backwards compatibility with at least the previous generation, so it always felt "safe" to buy the newest model.

I recall the same thing.

Everyone understood what Nintendo was up to. Sega friends (that's how I thought of them, defined by the console they owned) would tell me about this or that technology, but not many games for this new tech.... on the Nintendo systems (and later my PC) I had games.

By the time the Saturn was out.... most Sega friends had converted to something else, it was already over.

There's something beyond the obvious nostalgia factor that makes these 25 year old video game business stories so captivating. It might be from these massive corporations moving with a gut-driven, ride-or-die mentality or the fact that the only indicator that matters is sales.

In comparison, today's gaming industry is fragmented beyond all recognition. The audience has grown and matured, but every major player is a unique gumbo of F2P, overplayed IP, indie risks, and speculative technology. Also, they're all subservient to some larger corporate strategy.

Once again, my cynicism may just stem from my age, but it seems like there hasn't been much excitement for the past 5 years or so.

It's because the internet has given control of software to companies. Pre 2005 the only control game companies had over software was a few PC RPG's that had been rebranded mmo. In an internet enabled world, every game can be made client-server and have in game stores.

The internet is what ruined gaming because it gave corporations and developers too much power and control of the software and the ability to deny ownership, dedicated servers to their customers.

I noticed this while searching for quality mobile games to help pass the time in quarantine. "Always on" completely changes the gaming medium since that constant relationship between developer and player means the product should never "end".

I don't like to think of new technology as "ruining" something, but maybe art is the one exception. After all filmmakers stubbornly stuck with 24 fps even when frame rates improved.

And there's far less meaningful innovation in gaming today. VR was pretty promising, but has failed to be transformative. The new consoles' graphics frankly don't look much better than current ones IMHO, despite having 10x or 100x the performance. Game streaming is the most interesting and disruptive technology, once the major players figure out the right business model.

Back in the 90s, opening a gaming magazine was so exciting due to the crazy stuff you might glimpse.

Ha .. I miss those days and have been curious what was different. I have chalked it to youth but maybe you have a point. Opening up gaming magazines was an experience that poems should be written about. I remember old magazines on Amiga and even Nintendo Power that I'd keep in my bedroom as some of my most hallowed possessions.

VR will be a lot more exciting when we hit ‘retina’ VR, where immersion will be much more effective. The screen dooring effect of the original Vive and Oculus was a huge negative for me.

I bought a PSVR, arguably the least impressive of the bunch, but was still blown away. It was really neat. But I didn't touch it after a couple weeks. It was too annoying to put on, take off, clean, store, etc. So I sold it, and don't miss it at all.

I don’t think Nintendo has developed those issues, except for (arguably) overplayed IP.

> In comparison, today's gaming industry is fragmented beyond all recognition.

Not really. The gaming industry today is free-to-play MOBA games for Android and iOS.

Anything else is legacy that only keeps existing due to nostalgia of the 25+ crowd.

Granted, this crowd is the largest and the one with the most disposable income here in 2020, but it's obvious that going forward it will quickly wither and disappear.

The Nintendo switch seems popular among kids and lots of teens on consoles + steam.

I’m been doing some Nintendo 64 development lately, and sometimes all ask the CEO at my workplace for advice or insights as he worked on Nintendo 64 games when he was younger.

It’s not uncommon for him to finish the conversation with noting that he worked on Sega Saturn before the Nintendo 64, and the former was difficult to work with due to its particular architecture.

Out of curiosity, what kind of tools do you use to do N64 development?

The last version of the SDK and the GCC-based toolchain were leaked a long time ago, which runs on Windows computers pre-Vista. I'll program in C with Sublime Text, and then run the compiler/toolchain in a VM running Windows XP. The VM can see my source code via a shared folder.

I'm waiting on the mail for a flash cartridge to run ROMs on retail hardware, so for now I use the mupen64plus emulator on Mac OS X, but also a more accurate (but slower) one called CEN64 from time to time. Nintendo 64 emulators are kind of odd; the popular ones don't really "emulate" a game console the way I would have expected.

Sometimes I post pictures of my progress on Twitter if you're interested: https://twitter.com/danielsavface/status/1258896460604555264

As someone else digging into this...

The short answer seems to be to use either SGI's N64 SDK and a modern toolchain (https://github.com/trhodeos/modern-n64sdk) or libdragon (https://dragonminded.com/n64dev/libdragon/). The latter comes at the cost of not having built-in support for 3D (but it seems like there's enough there to take a DIY approach) and also being incompatible with most emulators (which "cheat" a bit by intercepting calls to the original libultra instead of directly emulating all the hardware). One could also run SGI's original compilers/toolchain, whether on a Linux system (using qemu-irix, which is what the folks over at https://github.com/n64decomp have been doing) or on a Windows 98 VM.

The N64Homebrew subreddit (https://www.reddit.com/r/N64Homebrew/) is also a pretty useful resource.

Peculiar is an understatement. The development tools were also a nightmare

Whelp at least we got Radiant Silvergun!

My understanding is that the Sega Saturn was a nightmare to program for. Everything else at the time rendered with triangles, while the Saturn uniquely used quadrilaterals.

It's also been a huge pain to emulate.

> Everything else at the time rendered with triangles, while the Saturn uniquely used quadrilaterals.

Isn't that still the case? I know that quads are used quite a lot for modelling but is there any modern system which uses quads for rendering?

Quads are used as geometric data primitives because they allow consistent subdivision. But quads are not good as a rendering primitive compared to triangles. Triangles are rasterizer-friendly because they are composed of two edges and its quick to know if you are inside or outside two edges.

> Triangles are rasterizer-friendly because they are composed of two edges

Huh? Can you elaborate on how a triangle has 2 edges opposed to 3?

For any span of pixels, you render from one edge to another because your y-coordinate is fixed. You only have to consider 2 edges at a time during rasterization. You can subdivide a quad easily, though, so it's not that much of an advantage.

I assume they mean for scanline rendering where you trace two edges at a time while drawing the scanline between them.

Basically, the way it works is that you arrange the 3 vertices of a triangle from top to bottom, then take the middle vertex and split the triangle along the scanline from that vertex to the opposite edge. Each "half" of the triangle can be rendered by simply drawing a scanline between the two edge positions, stepping down to the next line, advancing the two edge positions, and repeating. (See "Bresenham's line algorithm" and "Digital differential analyzer" on wikipedia for efficient ways to compute the edge positions.) Note that if the middle vertex shares a scanline with another vertex, the other "half" of the triangle in this algorithm has a height of zero lines, and can just be ignored.

It's a simple, straightforward way to rasterize a triangle that can be implemented efficiently.

You can't compute a normals from a quad, which makes shading algorithms not work.

It's hard to say what the biggest mistake was exactly. The Sega CD and 32x were bad ideas, compared to just creating a new console. The Sega Saturn would have been a great 2D console, but they delayed it to add a really strange (quads versus triangles) underperforming 3D capability to their design. I think the Saturn should have been released earlier in 2D only form, so they could move on to making a 3D capable console that trounced the PlayStation and N64 on a delayed timeline, perhaps even backwards compatible with the 2D Saturn.

A CD based Sega system was a great idea but not as an add on, the reality is the cost of hardware back then and the nature of the console market being technology illiterate was the issue.

I was one of the few who got the original Sega CD, the early version one that mounted underneath the genesis. It had cool games like Darkwizard. The real issue was the cost of add on peripherals were too high to get any kind of market penetration. Back then parents bought videogames for kids for their christmas or their birthdays. They'd rent their favorite games from blockbuster/convenience store and then get their parents to buy their favorites.

The reality was consoles and games were expensive and most kids rented games back when sega and nintendo were the kings of gaming before PC gaming had taken off in 1990's.

So the financial barrier to console ownership and the high price tag for parents was the real issue. Sega had a lot of good idea's but not conceived in the right way or at the right time. They acted as if the gaming populations parents were rich.

That was the real issue with many console companies that allowed Sony to get a foothold into console gaming.

Playstation was as popular as it was because of piracy and backups thereby increasing its market, it was "microsoft" method of console dominance - we don't care if you pirate as long as you use our console.

Even if sony didn't intend that, Sony PS1 and PS2 became huge because of ability to pirate games on the platform.

Piracy paradoxically drove sony to success. Everyone forgets places like china, india and third world countries at the time that couldn't really afford games because the the ridiculous prices.

Anecdotally, I knew a bunch of people with PS1, maybe one of them knew how to pirate games.

Yeah, the console I anecdotally associate with particularly easy/common piracy is the Dreamcast.

Anecdote-wise, everybody in my home town who owned a playstation had pirated games.

You don't need to know how to make illegal copies, you could just buy them for 1/10th of the original price on the street.

Ok - update - they didn't have pirated games.

Oh, fantastic! I love reading everything about this period of history.

The Saturn is my person favourite console and I greatly enjoy writing homebrew for it.

If you haven’t, and this is of interest to you; check out JO-engine, a fantastic FOSS Saturn homebrew development kit.

https://jo-engine.org :)

In case anybody else is denied access because of an adblocker: https://outline.com/bujnmX

I remember the disappointment. I had a Genesis at the time, I loved it, and I was primed to be a continued customer for Sega. From my perspective, all they needed to do was to release a decent next-gen console with games support.

Not just Sega, but Nintendo as well. The N64 was a moderate success, but if it was a CD-based system and came out a year earlier, Nintendo would have cut the PSOne at the knees and maintained console dominance.

Sega fall was far more important than Nintendo in this era. N64 had world wide acclaimed games that Saturn couldn't even hope for. And most of those games couldn't have been created on a CD-based system at that time that imply very slow loading.

N64 fall was mostly due to:

* Lack of texture memory. Having nice looking game was hard and need tricks because you couldn't rely on big textures.

* (Very) bad developer API (actually ABI). Having being in N64 emulation, you truly see they have no idea how 3D rendering was supposed to be exposed.

* Cost heavy support (cartridge).

But even without fixing the last one, the both firsts killed the third party dev investment.

The N64 was so close to being even better than it was, but the field was so new.

Any of the following would have helped a lot: * Larger VRAM cache * Dedicated VRAM with DMA moved between it main RAM. * Trade-off bandwidth for latency in RD-RAM configuration * Static work RAM for the main CPU * Splitting off audio duty into a 65c816-like DSP/cache

Hindsight 20/20, etc.

I have to disagree with your first two points. Neither of those was a reason for N64 being outsold by PSOne.

>Lack of texture memory. Having nice looking game was hard and need tricks because you couldn't rely on big textures.

So what? The PSOne hardware also had its own set of constraints that made their games pixelated and ugly. In fact, I would argue N64 games were generally better looking than PSOne games.

>(Very) bad developer API (actually ABI).

Again, who cares? If the system sold as well as the PSOne did, developers and publishers would deal with it.

>Cost heavy support (cartridge).

This is the big one. The fact that N64 was not a CD-based system, however, really hurt it with consumers, who wanted a CD-based system, because CDs were new and exciting. Third-parties also hated paying Nintendo for a license AND for cartridges because it cut into their profit margins. The fact that N64 was a cartridge-based system also shut them out from a bunch of AAA titles that used large amount of textures, voice/video and FMV cutscenes because it made straight ports impossible. It may seem silly these days, but FMV in video games was really exciting back then.

I also think it hurt them that the PSOne was released almost two years before the N64 did.

>(Very) bad developer API (actually ABI). Having being in N64 emulation, you truly see they have no idea how 3D rendering was supposed to be exposed.

I'd be very interested in learning about the N64's 3D API and ABI and the issues with it, if you'd like to expand on that. Or maybe you know of a blog post or article already covering the topic?

IMO, the cartridges saved the N64; they were a differentiating feature with their own set of pros and cons (the big pro being the lack of loading times that plagued CD-based consoles, and the big cons being the limited capacity and high unit cost).

Besides, in terms of longevity, I feel like the cartridges I've got lying around have aged a lot better than the scratched-to-hell-and-back discs I've similarly got lying around.

That was definitely the N64 marketing back then. Reinforce quality over quantity (as a way to explain away lack of third-party games), and load times (as a way combat CDs).

Nothing you said is wrong, however. Cartridges and N64 games, in general, have aged better. But at the time, it cost them the console lead. Cartridges were a key reason why third party game support was so bad. If you were a fan of RPGs, Fighting games, and Sports games - you had to get a PSOne because the options on Nintendo were so bad. People were excited about CDs and FMV in video games. Huge multi-disc games were released for the PSOne that could never be ported to the N64 - further contributing to the dearth of games on the system.

Don't also discount the fact that the system came out almost 2 years after the PSOne due to various hardware delays AND lack of launch games (the system only launched with two!!). Plus the system and games were more expensive.

I guess my overall point is that the other factors you mention (the 2-year delay and lack of launch titles even with that delay), plus a couple not yet mentioned (like some rather nasty bottlenecks in the otherwise state-of-the-art graphics pipeline, especially around textures, which made things harder for developers and therefore harder to get third-party titles) were a greater factor in why the Nintendo 64 wasn't the dominant console. And yeah, the cartridges might've contributed to the lack of launch titles somewhat, but I suspect it really had more to do with tighter quality controls (this was the cited reason why Shadows of the Empire didn't end up being a launch title; LucasArts explicitly prioritized an extra couple months of polishing over trying to force the originally-planned release date) plus the N64's somewhat more exotic hardware (this being the same problem the Saturn had, but in the N64's case it wasn't quite as extreme).

That is: I think the N64 not being the "lead" console was a foregone conclusion regardless of whether it used cartridges or CDs, and that had the Nintendo 64 jumped on the optical bandwagon it would've suffered rather than prospered, all other factors being constant. The Gamecube is evidence of this, with Nintendo's market dominance slipping further even after making the switch to optical media (though one could definitely argue that the choice of Mini-DVDs kneecapped the Gamecube right from the start, introducing the worst of both worlds re: load times and capacity relative to other consoles in its generation). Yet, even the Gamecube and the N64 before it were popular enough to be commercial successes; neither of them really "lost" the Console Wars the same way SEGA's and Atari's consoles did (or the same way the Wii U arguably did).

> the options [for roleplaying/fighting/sports games] on Nintendo were so bad

The third-party options were unremarkable, sure, but the Nintendo 64 was the debut platform for Super Smash Bros., which was a pretty massive commercial success even then (let alone in future iterations like Melee and Brawl). Can't speak much to RPGs or sports games, since I didn't play very many RPGs back then (and never really found much enjoyment in sports games, on any platform), though I'm pretty sure the Legend of Zelda games and Paper Mario all fall into the RPG genre and all had pretty great critical acclaim and commercial success.

>but I suspect it really had more to do with tighter quality controls

That was Nintendo's PR spin but it wasn't based in reality. The N64 had some real stinkers (Superman 64) - so let's not delude ourselves that Nintendo's quality standards are what prevented third-parties from releasing games.

>The Gamecube is evidence of this

It's not a valid comparison. The Gamecube was released at a time when DVD players were novel, expensive and exciting, so getting a DVD player with a console was a big deal, and of course, the Gamecube was ... not released with a functioning DVD player. It also came out a year after the PS2 (the best selling console of all time) and at that point, everyone understood that Nintendo was no longer the top-dog in consoles. They were the underdog. There was very little hype for the Gamecube. And yes, if they released a cartridge-based system, they would have been sunk. Even the most hardcore Nintendo fans wouldn't tolerate cartridges at that point.

>neither of them really "lost" the Console Wars the same way SEGA's and Atari's consoles did

That's true. Nintendo found themselves a nice niche, that was in part helped by the fact that the managed to maintain mobile console dominance.

>The third-party options were unremarkable, sure, but the Nintendo 64 was the debut platform for Super Smash Bros

That's true, the N64 had wonderful first and second party games. That was the only reason to own the N64. It's also THE reason to own the Gamecube, Wii, WiiU, and the Switch.

>Can't speak much to RPGs or sports games, since I didn't play very many RPGs back then

You know who played RPGs? Japan, the second biggest console market. It hurt Nintendo incredibly badly that the big multi-disc RPGs were never ported or released for the N64. Sports games were also critically important and the first few years Nintendo did not have the yearly EA FIFA/Madden/NHL offerings. Also top-tier sprite-based (Namco/Capcom) and polygon-based fighting games were never released for the N64. All this really hurt the N64 at the time. Basically if you were at all interested in non-Nintendo first-party games, you had to get a PSOne.

>though I'm pretty sure the Legend of Zelda games and Paper Mario all fall into the RPG genre and all had pretty great critical acclaim and commercial success.

Those were phenomenal games but Zelda was not an RPG. Paper Mario was but that's just one game. N64 did not have Final Fantasy and Dragon Quest, and countless others released on PSOne. Final Fanatasy and Dragon Quest were system sellers in Japan.

> The N64 had some real stinkers (Superman 64)

Superman 64 came out in May 1999, right around when Nintendo announced "Project Dolphin" (a.k.a. the GameCube). I have a feeling most of the other "stinkers" were similarly late in the N64's lifecycle, by which point the lack of third-party games would already have been a known issue and would've prompted the very relaxation of quality controls that allowed games like Superman 64 to see the light of day.

And besides Superman 64 (which I never played), I can't really think of very many other "stinkers" for the N64. Maybe Roasters (another Titus title, even later than Superman 64), but even that was pretty fun to play (even if it reeked of being low-budget).

----

Fair points otherwise.

Hindsight is always 20/20, but I wonder why Sega made decisions like using quads instead of triangles, and dual CPU's.

Their arcade boards used quadrangles too, not that that really answers the question

3D chips at that time made some compromises that caused things to pop in a weird way when they were used to render rectangles such as walls.

Most Japanese companies are not run like American companies. There is a deep culture of compromise and consensus over top-down leadership. One of the results of this is the need for every stakeholder in the company to have a hand in every decision, and to get a piece of every new product.

For the Saturn, this resulted in the console packed with little goodies that were created by different departments. Things like off to the side audio chips that did barely anything. Dual CPUs. Strange 3D tech that used quads instead of triangles.

A few years later, we did work with Sony, and they seemed for some reason to work more like an American company, in the sense that they say they need more able to make rational decisions and move quickly when they needed to.

No one who has worked for a Japanese company would ever use the word agile to describe them. Sony may be the exception, at least at that time.

How do you even render a quad? What if it isn't planar? Does it curve or crease?

In higher-end rendering, a quad that can be non-planar is called a bilinear patch. Yes, it can curve (cross sections tend to give parabolas), and while you won't get infinitely sharp creases you will get saddle-shapes.

See, e.g.: http://www.sci.utah.edu/~kpotter/publications/ramsey-2004-RB...

Wow seems like that’d be an order of magnitude or more slower to render bilinear surfaces rather than triangles. I wonder why they did that.

It allowed for repurposing of sprite-drawing hardware the Japanese were already familiar with. Most Japanese programmers at the time were at complete sixes and sevens with the sort of triangle-based, texture mapping 3D hardware that had been developed in the west. I guess Sega thought it would be either a faster path to developer success, or easier to implement in hardware, if the Saturn's 3D engine resembled the sprites that game and console makers already knew, rather than gamble on unfamiliar triangle rasterization.

Nintendo collaborated with SGI on the N64, giving them a HUGE leg up on ALL the competition with respect to rendering techniques -- perspective-correct, filtered textures when even the PlayStation could only do affine transforms on textures (leading to the "texture warping" phenomenon). But even their programmers had to come to grips with 3D, which was hard. One reason StarFox 2 was cancelled was because Nintendo wanted to use its camera code in titles like Super Mario 64 (and did NOT want word of this getting back to Argonaut).

Sega had a lot of prior experience with quad-based 3D games. Their Model 1 and Model 2 arcade boards used quads.

The Sega Saturn was the first console I purchased with my own money.

The Saturn had some great games.

Three of the Four launch titles were great: Daytona (Rolling Start....), Virtua Fighter, and Panzer Dragon were far better than anything else available.

Street Fighter Alpha 2 was far superior on the Saturn compared to the PS version.

And Guardian Heroes (available now on Xbox Live Arcade) is probably one of the best, most underrated sprite based beat-em-ups.

Ultimately Sony crushed the competition with titles like, Resident Evil 2, Tekken 3, FFVII, Twisted Metal, etc etc. Which is crazy considering how strong the N64 was.

It's sad that Sega laid an egg, but the Saturn had it's time and place.

I'm really looking forward to what the Series X can do with it's project X-cloud. My feeling is that it's going to move towards PC gaming--once your console can no longer play games natively, it will switch to pixel streaming.

The article misses the point entirely. It wasn't about launching early, or not having enough games, or wait times.

It comes down to pricing.

In gaming, the same thing has always played out. The console that can deliver the best value wins.

$399 vs $299 is a huge difference. That's the equivalent of $500 vs $750 today.

And don't forget that this was a huge step up already from the $150-199 price point of Super Nintendo.

At the end of the day, when it comes to mass appeal, it isn't the better technology that wins, but the one that combines the best price and performance, and people always underestimate that price is the bigger factor than performance.

You're blanking on the 99s. It's the difference between $500 and $666 today.

The article covers that fact pretty clearly at the end no?

It's pretty ironic that Sony almost made the same mistake with the PS3 at $599.

I say _almost_, because it might have benefited Sony as a whole anyway.

The fact that it was a Blu-Ray player saved it because families could justify the purchase for that feature alone (and typically tacked on the bluetooth remote too)

People forget that the PS3 was BY FAR the cheapest Blu-Ray player for almost 18 months.

Exactly. It was almost the same strategy with the PS2: it was a ubiquitous DVD player when that was a novel concept.

As a teenager with a job, the cost wasn't really an issue and the launch window meant there was plenty of time to save up the difference. What killed the Saturn for me and my circle of friends were the games.

I remember we rented a Saturn with all the games from Block Buster for like $50 for the weekend and were thoroughly unimpressed. Not long after, a Playstation popped up at Media Play (remember that store?) with demos you could play. I remember specifically going there multiple times to play Battle Arena Toshinden.

Reply on news.ycombinator.com