Former Sega president Shoichiro Irimajiri recently gave a lecture series at the Forum – Management for Tomorrow. The fifth lecture in the series, held on October 20, 2022, focused on Irimajiri’s time at Sega in the 1990s.
Irimajiri joined Sega in an executive role in 1993, right at the company’s height. He had previously been president of Honda of America and was being recruited for an executive position at General Motors when Sega president Hayao Nakayama enticed him into joining Sega.
Irimajiri worked in a variety of roles at Sega, and he frequently traveled between Sega Enterprises in Japan and Sega of America. In 1996, he became president of Sega of America after Tom Kalinske stepped down, and in 1998, he became president of Sega Enterprises after Hayao Nakayama stepped down. His experience working at both the Japanese parent company and the American subsidiary give him rare insight into the company.
Here, I have translated a few short excerpts from his lecture that focus on the design of the Saturn, the 32X, and the financial problems at Sega of America that eventually led to Tom Kalinske stepping down. At the bottom, I have also included several quotes from other Sega executives that give more insight into these issues.
Shoichiro Irimajiri, speaking at the Forum – Management for Tomorrow lecture held on October 20, 2022.
On the Design of the Sega Saturn and the 32X
The first thing I had to do after starting at Sega was figure out what was going on with our next-generation game console. There was a big difference in the situations in Japan and America that ended up causing me a great deal of problems.
In Japan, Sega’s 16-bit console the Mega Drive had achieved a certain degree of success. However, Nintendo was completely dominant at the top and NEC was in second place with its PC Engine, so Sega was in third place.
It was generally thought that our next-generation console had to bring something new and different to the market, such as 3D graphics. However, there was also the opinion that we could make even greater games with 2D graphics, with more color and smoother animation. Such a game console could specialize in upgraded 2D graphics. Opinions were split on the matter. Without a clear decision being made, the next-generation console design featured both the ability to do a limited amount of 3D graphics and the ability to do the highest level of 2D graphics.
In the middle of 1993, right when I joined Sega, Sony revealed the 3D graphics capabilities of its PlayStation. Around that time, Sony had finished its prototype development hardware and was distributing it to third-parties. In response, the third-parties shifted their focus to developing games using 3D graphics. The polygons-per-second capabilities of the PlayStation were truly amazing. They were even on a level near Sega’s Virtua Racing. We soon learned that many third-party developers had made the decision to abandon 2D game development in favor of 3D graphics. It was a very bad situation for us.
Just as I entered Sega, there was a huge debate about whether it was okay to release the Saturn as it was with its limited 3D capabilities. It was decided we had no choice but to increase the 3D capabilities. At the absolute minimum, the Saturn had to be able to handle the 3D graphics of Sega’s fighting game Virtua Fighter. That was impossible on the Saturn as it then was. To put it in simple terms, the Saturn’s CPU could not handle the heavy processing required to create the 3D graphics. A possible solution was to add a GPU, a graphics chip, that could specialize in handling the heavy graphics calculations. However, it was far too late to begin development on something like that, so it was decided to use a second Hitachi SH-2 CPU in place of a GPU. The architecture of the Saturn was revised to include two SH-2s, and that’s how the Saturn was released. So, in Japan, the 3D capabilities of the Saturn were boosted.
However, in America, they didn’t want to go in that direction. The Genesis had been so successful, and third-parties were still in the middle of developing games for it. Therefore, before releasing a next-generation 3D-capable console, they wanted to release an enhancer for the Genesis that would boost its capabilities.
This is where things got messy. Since we wouldn’t be able to release the Saturn in America, I was told to go there and help resolve the situation. I had just joined Sega at the beginning of July 1993, and by the middle of July, I flew out to Sega of America. I still didn’t have a good understanding of the game industry, so I was just going to listen to what they had to say.
I met with Sega of America president Tom Kalinske and head of software development Joe Miller. They told me that there was no way they could abandon the incredible success of the Genesis. They made this clear to me in a very passionate manner. After that, I returned to Japan.
We would release the Saturn in Japan at the end of 1994. Meanwhile, we would develop an enhancer for the Genesis, the 32X, that Sega of America would release at the end of 1994. We would split the markets that way. The Saturn’s American release would be postponed for a year. That’s the plan that was eventually agreed upon.
It turned out to be a huge strategic blunder, and it would cost us heavily until the very end. The American third-parties that were counting on the next-generation hardware were thrown into great confusion, and they ended up going to the PlayStation. They didn’t continue on to the Saturn.
On Sega of America’s Financial Troubles and Tom Kalinske’s Resignation
A huge problem came to light at Sega of America. In the American market, it was typically large retailers such as Toys “R” Us and Wal-Mart that stocked game hardware and software. Those kinds of retailers would buy a huge quantity of stock at first. However, if they didn’t sell the stock within a certain period, they’d send it all back. We’d have to buy it all back.
SOA’s posted profits in 1993, for example, were all washed away because it had to take extraordinary losses on returned stock later on. Those extraordinary losses came to $100 million or $200 million at a time. Furthermore, retailers in America held a lot of power, and they required manufacturers to have a certain amount of inventory on hand to replenish stocks when items sold out. For example, retailers required SOA to have at least 500,000 Genesis hardware units on hand to replenish sold stock, or they wouldn’t do business with us.
SOA had an excess of inventory that would all be sold at once, bringing in a huge amount of revenue. Then, all that inventory would come back from retailers later on and SOA would take a huge loss. With the Genesis, all of those losses started to appear in 1994, 1995, and 1996. If you added up all the losses, the number would be astronomical.
When we looked at the numbers carefully, even though SOA had a reputation for earning so much money, it turned out they weren’t earning much at all. We decided we had to change things, and that necessitated reducing the size of the company.
We told Tom Kalinske that we’d give him one year to restructure the company like so. This was a quite a strong request.
After one year, in 1996, the restructuring hadn’t progressed at all. We decided we had no choice, and I was ordered to go to SOA and take over as president. Tom Kalinske was asked to step down. I initiated the restructuring and worked to make the company healthy.
Further Quotes
Hayao Nakayama on the creation of the 32X
Former Sega Enterprises president Hayao Nakayama said the following in Sega Mega Drive / Genesis: Collected Works (2014):
The main worry was the high production cost of the Saturn. We had a lot of discussions with Sega of America about this, and we came to the conclusion that it would be worth using the huge customer base of the Genesis rather than switching to the Saturn. Sega of America insisted that US consumers could not afford the new console, and because of the team’s great success with the Genesis, Sega of Japan finally relented [concerning the 32X].
Hayao Nakayama on the split strategy for the Saturn and 32X
Hayao Nakayama said the following at the Securities Analysts Association of Japan meeting, July 26, 1994:
This year’s holiday season will see the fight begin between Sony and Sega. The question at the forefront of everyone’s mind is how many units a 50,000-yen console can sell, but this is something that we’ll never know without trying. Regardless, within the Japanese market, our efforts will be centered on the Saturn. Regarding the 3DO, our predictions are generally coming true. Matsushita is applying its full force in Japan to drive sales of about 150,000 units, and by the end of the year this might reach 200,000 or 250,000 units. However, in America, the 3DO has not been very well received, and it’s still not selling well at all. I’d call it an overwhelming defeat. If you make a mistake with a launch in the American market, it’s very difficult to recover. The first problem is that retailers won’t carry your product. Therefore, neither Sony nor Sega is going to put our 32-bit consoles on the market in America. Of course, that doesn’t mean we’ll never release them there. At Sega, we’ve come up with a new strategy, something never before done. That strategy is to compete in the American and Japanese markets with different game consoles.
In America, Sega will release a 32-bit console called the Genesis 32X. This is an attachment for the current 16-bit console that will turn it into a 32-bit console, and it doesn’t have a CD-ROM drive. It uses cartridges. Our 16-bit console has sold over 10 million units in America, so it will be possible for those users to convert their current 16-bit consoles into new 32-bit powered consoles and to experience 32-bit games for a price of around $150, rather than $500. This is the strategy we’ve decided to pursue in America.
Putting out two consoles at the same time is going to create confusion in the market, so we’re going to clearly divide our strategy. Up to now, we’ve released our consoles first in Japan and then after that in America. However, now we’re trying something different: we’re going to release different consoles at the same time in each region. With the current trends, I think this is the best marketing strategy for us.
The 16-bit home console market has finally peaked, and we’re currently in the down season. Therefore, I don’t expect the new console generation to really take off until next year’s Christmas season. The new console hardware is being launched in a hurry, so it’s not going to be a sudden explosive growth in business like we’ve seen so far. This will be true for Nintendo as well. Even with a lot of support, the big take off phase won’t be this year, but rather next year.
Shinobu Toyoda on the creation of the 32X
Sega of America vice president Shinobu Toyoda said the following in Sega Mega Drive / Genesis: Collected Works (2014):
Our issue was that it [Saturn] was far too expensive. The Genesis was selling at $149, but the Sega Saturn looked likely to be a $399 machine. That sort of price had never been seen before. So Joe Miller said to Nakayama, in front of everyone, that if we created an add-on to the Genesis, and used a part of its guts, we could deliver the performance of the Saturn at half the price. So Joe started to draw the diagram of how we’d do that, and that was the 32X. It was in that room with just five or six of us that it all started.
David Rosen on the division between Japan and America
Sega founder David Rosen said the following in Sega Mega Drive / Genesis: Collected Works (2014):
I played the part of coordinator between San Francisco and Tokyo. There were always differences and problems; my time was spent on that. Tom was great, he did a hell of a job, but it wasn’t always easy for him to understand the decisions in Japan, and there they couldn’t always appreciate necessities like cutting the retail price of the machine. After the Genesis, Sega of America was interested in developing peripherals, while Sega of Japan wanted to move on. It wasn’t necessarily a friction, but there was a lot of going back and forth, helping to resolve issues with the ultimate aim of achieving what both arms of the company wanted.
The Nikkei Shinbun on SOA, the 32X, and Nakayama
The following is an excerpt from a Nikkei Shimbun article titled “Sega plunges into the red without anyone taking blame,” printed on March 16, 1998:
During the first half of the 1990s, Sega was victorious in capturing 50% of the North American market with its 16-bit game console the Genesis. If Sega had fully committed to the Saturn with that kind of force, the outcome probably would have been different. However, just before the release of the Saturn, Sega released the 32X, an add-on that just increased the capabilities of the Genesis. The intention was to boost profits from the Genesis install base, but the result was a disaster. Among software developers, “there was a rapid spread of distrust for Sega” (Irimajiri). The Saturn was released under these negative conditions.
The 32X was handled by Sega’s North American subsidiary [SOA], whose president had managed to persuade Nakayama to let him sell the add-on. However, soon after, that president was allowed to resign at his own convenience without being called on to take blame for the failure. Nakayama is famous for his management style of “sure punishment or reward,” but his sharp intuition for the arcade industry failed him when it came to the home console industry, where he lacked experience. Because of that, he allowed some executives to run solo.
Shunichi Nakamura on SOA’s inventory problems
Former Sega Enterprises Chief Financial Officer Shunichi Nakamura wrote the following in the February 24, 2023 edition of the Nikkei Sangyo Shimbun:
Sega of America and Sega of Europe had accumulated a huge quantity of unsold stock. These overseas subsidiaries were strongly independent, and from Japan we weren’t able to see what was truly going on there. In addition, at the time, non-consolidated accounting was the norm, and priority was given to the parent company’s performance in terms of how the numbers were calculated. However, toward the end of the 1990s, the trend became to emphasize consolidated accounting. When we recalculated the figures on a consolidated basis, we finally saw the reality of the mountain of problems that had built up.
As the person in charge of the administrative division, I should have realized what was happening before it was too late. If I had done so, I would have been able to implement more thorough inventory control or to undertake restructuring.