New Hideki Sato Interview Discussing the Sega Saturn

A new online magazine devoted to Sega came out in Japan the other day called Beep21 (by the same people behind the famous Beep! print magazines of the 80s and 90s).

The first issue features some incredible (and incredibly lengthy) interviews with Hideki Sato, Naoto Ohshima, and Koichi Nakamura.

The magazine costs 500 yen (~$5 USD) and I recommend anyone interested in this to buy it in order to support it (and it's all online so you can run the Japanese through a translation app). I'm not sure if non-Japanese credit cards work, though... Anyway, the link:

https://note.com/beep21/m/mb81eee06f1de

Now, to the good stuff. I've translated a snippet of Hideki Sato's interview where he talks about the development of the Saturn. There's some overlap with the interview he did a few years ago (which I translated portions of here), but there's also some great new detail.

Sato is very frank about his regrets with the Saturn. The interview is insightful in terms of why the Saturn was first designed as a 2D machine.

In particular, check out Sato's answer to the last question. He had originally considered a Saturn design based on the Model 1!


-The Sega Saturn was released in November, 1994, but the console’s name and early images of it were printed in magazines about a year before that. When did development of the Saturn begin?

Sato: Development of a new console obviously doesn’t happen overnight. I think we started development on the Saturn more than two years before its release. I believe we had actually begun the planning phase before the Mega CD was released (in December, 1991).

-If it was from around the time of the Mega CD, does that mean the Saturn was intended to use CD-ROMs from the beginning?

Sato: It’s true that the Saturn ended up using CD-ROMs, but we were also developing a cartridge version in parallel. Sega president Hayao Nakayama was initially backing the cartridge version more. Of course, it was clear at the time that CD-ROMs were the way to go for increased storage capacity, but the cost of CD-ROM drives was still quite high then. We were considering the option of using cartridges in order to reduce the price of the console. In the end, Nakayama agreed that the time was right for CD-ROMs. The size of games had grown so that there was no way they’d fit in a 4 Mbit cartridge anymore, so it had to be CD-ROMs.

By the way, the name ‘Saturn’ was actually the development codename, and we ended up just using it as the product name. As we were developing the console, the developers and managers grew accustomed to the name, and when we drew up some proposals for other names, none of them felt right. We decided to just stick with ‘Saturn’ and released it that way.

-In the beginning, what kind of concept or direction were you aiming for with the Saturn? With the Mega Drive, you had pushed for various expansion capabilities. Was the Saturn originally intended to be a 2D console? Or a 3D one?

Sato: To be honest, in the beginning, I wasn’t thinking of 3D capabilities for the Saturn at all. This was partially my fault, but additionally, the game developers at Sega at the time had basically no knowledge of 3D game development. They had all been raised in the environment of 2D sprites and backgrounds, and the only developers who had any real experience with 3D were Yu Suzuki and AM2 with the Virtua series. I personally had created proposals for a number of polygon-based arcade systems by that time, and the only one who had shown any interest was Suzuki. Actually, all of the other developers wanted to continue developing using the same system they were used to. If you looked at every single Sega employee within the home console division, there were practically no programmers or designers who had any knowledge of polygon technology.

What was special about Yu Suzuki at the time? He majored in math at university. More so than electronics, you have to be good at math to do 3D. That’s why Suzuki was ahead of everyone else in creating 3D polygon games. So, the situation at Sega was that if we made developers work on 3D games, they would have to study the fundamentals of math and geometry from scratch. Even the designers would have to study it. Up to then, designers had been drawing art pixel-by-pixel on a flat plane, including background art. If they suddenly had to do 3D CG art, they would have to learn it all from step one.

I had taken a look at Sega’s development teams at the time and concluded, “It’s going to be impossible for them to do 3D games.” I mean, we had over 1,000 developers working in the development division at Sega then. The Saturn was going to be released in 1994, but software development for it had to begin in 1993—and in some cases even in 1992. With all that in mind, I concluded that there was no way Sega’s development assets would be able to do 3D. However, the PlayStation completely embraced polygons.

-That was in the fall of 1993, when the PlayStation debuted within the industry as the PS-X, right?

Sato: Right. The Sony side was completely free of the constraints of worrying about the capabilities of development teams, so they were able to fully embrace polygons. When we found out about that, we realized we were in trouble. At that point, the Saturn had only a single SH-2 for its main CPU, so we added a second SH-2 to boost the console’s processing power. Thankfully, the SH-2s could be linked in a cascade connection. A large amount of geometry calculations are required to do polygon graphics, and a single SH-2 was completely insufficient.

-So, you added a second processor…

Sato: Then we improved the graphics engine in order to do pseudo-3D graphics, and after that made further improvements. Hitachi was delighted because, for each Saturn sold, they sold two SH-2s. Our initial target selling points for the Saturn were that it could display 4,000 or 5,000 sprites, it had four or five background layers that could rotate, and so on, but at the very last minute, we somehow managed to cram 3D capabilities into it. The fundamental things that were missing from it, however, were development tools. With all the CPUs in the Saturn, only a tiny fraction of developers at Sega were able to make sense of them all and actually put them to use. For third parties… well, there was no way. At a later point, AM2 was able to rush and put together the SGL (Sega Graphics Library), but from today’s perspective, that could hardly be called an SDK (software development kit). It probably took third parties a full week just to get something displayed on the Saturn.

-From your perspective, what did you think of the PlayStation?

Sato: It’s just what you’d expect from someone like Ken Kutaragi. In fact, when the Saturn and PlayStation were being designed, there were talks about Sega and Sony teaming up.

-Was that after the famous incident of Sony’s partnership with Nintendo for a CD-ROM-based Super Famicom falling apart?

Sato: It was after Sony ended the contract with Nintendo.

-What specifically happened?

Sato: Isao Okawa, the chairman of Sega’s parent company CSK, knew Sony’s chairman Norio Ogha very well, and through that there were various exchanges between engineers at Sega and Sony. The companies were physically near each other in Tokyo, with Sega in Otori and Sony in Shinagawa. Each company’s target was Nintendo, so our goals lined up, and there was talk along the lines of, “Why don’t we try to do something together?” With each company’s chairman being top class, negotiations were started with the intention of trying to find some way to work together. From Sega’s perspective, we were trying to figure out what would happen if we teamed up with Sony. Each company had its own philosophy, and we were trying to see if we could come to some agreement. And, honest to a fault, we ended up showing our complete technical specs to Sony.

-The specs for the in-development Saturn?

Sato: Right, the Saturn specs. However, in the end, we couldn’t reach an agreement.

-Was the way of thinking between Sega and Sony different?

Sato: As I mentioned earlier, in thinking about the state of the Sega software development teams, we strongly believed that it would be too difficult for us to suddenly jump entirely into 3D graphics. However, from Sony’s perspective, especially from Kutaragi’s perspective, none of that mattered. No matter what we did, our two companies were not going to be able to work well together. In the end, we were sure it wasn’t going to work out. Based on the state of the Sega developers, it was impossible. However, looking back on things, it’s possible we were thinking too much about it and making too many conjectures about software development at Sega.

-Were you in contact with Kutaragi then?

Sato: Yes, and I still am. Since then, we meet two or three times a year over dinner. Well, we would typically set aside our company backgrounds and enjoy dinner as two adults. Sometimes, Kutaragi would say things to me like, “Hideki-chan, your company’s hardware business model can’t win against us, so why don’t you all give up?” We’d exchange opinions like that in an inoffensive manner. Kutaragi is actually the same age as me. He has a very straight-talking personality, so he’d say all kinds of interesting things (laughs).

-What sticks out most about the Saturn in your memories? It was around for four years, but is there anything you wish you could have done differently?

Sato: I’m not sure if you call it a memory, but a regret I have is not going with one of our options to use the arcade system Model 1 as the base for the Saturn. As I mentioned, I couldn’t choose that option due to the situation with the development teams at the time. However, I can’t help but think it would have been better to just force our way ahead by throwing out all of our past development assets and starting from scratch. We could have gone with 3D polygons with that kind of force.

 

It's also worth pointing out, as documented in the book Revolutionaries at Sony, that the same was basically true for most developers in Japan at the time. None thought 3D was viable for home console games in 1992.