The unreleased SEGA 'Saturn 2' and The Dreamcast Story


Mar 29, 2012
Hiya, this is my first post here. I wanted to discuss the never-released 'Saturn 2' that was supposed to have been developed with Lockheed Martin Real3D technology.

From Next Generation November 1995 (sister magazine of EDGE):

'Saturn 2' could've been a new console instead of Saturn or a quick replacement
(not in place of Dreamcast, it's not of that class) or as a Saturn upgrade cart for Model 2 ports and downscaled Model 3 conversions.

Note that the Real3D/100 graphics card is not to be confused with the high-end Real3D/Pro-1000 GPUs used in Sega's Model 3 arcade board.

More on Real3D/100:

The Real3D/100 chipset could've been reduced into a single chip, much like PS1's CPU+GTE or better yet, the 3DO M2's Bulldog ASIC. If Lockheed Martin had desired to enter the consumer market in a big way (nevermind the i740), They would've been a force to be respected.

We would've had visuals like these:

The Dreamcast Story

''A do-or-die machine which will decide whether Sega stays in the
hardware biz''

Dreamcast is a system born out of Sega's darkest hour, a do-or-die
machine which will decide whether the company stays in the hardware
business. Its precursor, the 32bit Sega Saturn, had been widely
expected to conquer the world with Nintendo's own second next
generation system heavily delayed -- due to the collapse of an
alliance with Sony -- and neither Atari nor 3DO seriously threatening
mass market success.
All that changed with the November '93 announcement of the Sony
PlayStation, a system which would heavily defeat Sega's system and
become a considerable influence on how Sega designed Dreamcast.
Although there had been rumours of Sony producing a console, what came
as a heavy shock to Sega was the technical superiority of the
PlayStation. While the Saturn had been designed as perhaps the
ultimate 2D arcade machine, albeit with a substantial 3D capability,
PlayStation was totally committed to polygons.

Sega boss Hayao Nakayama angrily berated Sega's engineers for their
failings, but it was too late to totally redesign the system if the
1994 launch was too proceed. Instead, Sega added yet another processor
to an already over-complicated design. In terms of raw power, the new
Saturn was much more of a match for PlayStation, but it would never be
an easy machine to program for. The twin CPU design in particular
demanded highly specialised machine code rather than the C most
Japanese developers prefered: barely a year after Saturn's launch a
key Sega manager admitted only one in a hundred programmers would have
the skill to use the machine's full potential.

Ironically, the Saturn's Japanese launch would be Sega's best ever
performance in its home territory. Even a flawed version of Virtua
Fighting was enough to transform the company's traditional weakness in
its home territory. Overseas, however, it was to be a different
matter. Scepticism about the prospects of a CD-ROM machine succeeding
in the cost-sensitive US market meant Saturn was originally partnered
with a low-cost, cart-based system codenamed Jupiter -- principally
due to American scepticism that a CD-ROM machine could be
competitively priced. When Saturn was upgraded, Jupiter got axed in
favour of Mars, an upgrade for Sega's 16bit Mega Drive which was
supposed to protect the company's hugely lucrative US market. In fact,
32X was an unmitigated disaster, drawing vital developer support away
from Saturn and destroying the company's reputation among gamers who
found themselves with an add-on with barely a handful of games.

The Saturn debacle would cost the jobs of Sega's American and Japanese
bosses, beside reducing its US empire to a ruin running up losses of
$167 million in 1997. For any replacement machine the lessons were
clear: a single format, complete user-friendliness for developers and
a new brand -- so low had sunk the once mighty Sega name.

As soon as any console is launched, work is usually underway on a
replacement but the Saturn's troubles gave this process an unusual
urgency for Sega. By 1995, rumours surfaced that US defence
contractors Lockheed Martin Corp. were already deep into the
development of a replacement, possibly even with a view to releasing
it as a Saturn upgrade. There were even claims that during Saturn's
pre-launch panic a group of managers argued the machine should simply
be scrapped in favour of an all-new LMC design.

Sega originally entered into partnership with LMC to solve problems
with its Model 2 coin-op board, however by 1995 the relationship had
soured somewhat with the Model 3 board suffering massive delays.
Around the same time, 3DO began shopping around its 64bit M2 system.
According to informed sources, Sega's Japanese bankers had brokered an
unwritten deal whereby Matsushita would manufacture M2 units, while
Sega would concentrate on the software. M2 devkits were supplied to
Sega in early 1996, with initial work reputedly concentrating on a
Virtua Fighter 3 conversion for M2's launch.

Sega's M2 project soon fell apart however. 3DO's Trip Hawkins blamed
corporate ‘egos' for the collapse, while Sega insisted its engineers
were unconvinced M2 was the breakthrough technology they needed.
Instead, the company was increasingly preoccupied by the PC market --
unlike Nintendo, it was fully prepared to convert its games onto the
format and in mid-1995 it had entered into a partnership with PC
graphics card manufacturer nVidia. Under the terms of the deal, Sega
would supply ports of key Saturn titles exclusively for the nVidia PC
graphics card. At the time, pundits wondered if Sega might be
switching from Saturn to nVidia as its principal platform.

By 1996, this speculation was ebbing away as two clear frontrunners
emerged in the PC graphics market: VideoLogic's PowerVR and 3Dfx's
Voodoo chipsets. Sega approached both companies to be partners in two
parallel Saturn 2 projects, each of which having minimal if any
knowledge of the other. The 3Dfx-Sega of America project was codenamed
Black Belt, while the VideoLogic-Sega of Japan system was known as
Dural. Although console development is usually shrouded in total
secrecy, Saturn 2's development coincided with the rise of the
Internet and Black Belt soon became a popular topic of gossip. For a
time, many presumed Black Belt was the only new Sega system.

All this changed on July 22nd, 1997, when 3Dfx was informed them Black
Belt was cancelled. It was a shattering blow -- "Our contract with
Sega was considered to be gospel right up until we received the call,"
admitted marketing manager Chris Kramer. Two months later, 3Dfx issued
a lawsuit against Sega while blaming VideoLogic's Japanese backers,
NEC, for bringing influence to bear on a decision which would
otherwise have gone to 3Dfx. An initial burst of publicity soon gave
way to highly confidential discussions which settled the lawsuit away
from the public eye in August 1998.

For outsiders, 3Dfx had always been the favoured partner due to their
leadership in the PC market, moreover Sega let it be known the
decision to cancel wasn't due to either performance or cost reasons.
What may have been a factor is 3Dfx's very strength made it a
difficult partner for Sega, VideoLogic's second-place status obviously
made it the hungrier partner. Moreover, whereas 3Dfx see themselves as
creating a new gaming platform around their Voodoo hardware and Glide
software, VideoLogic were much more eager to use Microsoft's Direct3D

Whatever the reasoning behind the decision, the PowerVR decision
further dampened excitement about a machine soon to be redubbed
Katana. In January '98, UK trade newspaper CTW ran a savage onslaught
upon the new format: "When one looks at a format owner that actually
struggles to garner interest in its latest hardware announcements, you
know it''s in trouble. From Black Belt to Dural and Katana,
journalists have leapt into headline mode, but the level of
disinterest elsewhere is palpable." Commenting upon the latest
redundancies in America and Britain, Dinsey wondered whether the
company was "giving up and trying to re-invent itself as a PC

In May, Sega gave its response with the official announcement of its
new system, its specifications and that controversial name: Dreamcast.
The marketing campaign began with the announcement of the marketing
campaign and its $100 million budget for each territory: America,
Europe and Japan. Sega boss Shoichiro Irimajiri put the cost of
hardware development at $50-80 million, software development at
$150-200 million, which with marketing added up to half a billion

The PR statements were suitably bullish: "Dreamcast is Sega's bridge
to world-wide market leadership for the 21st century" commented Sega
US VP Bernie Stolar. "I am confident that Dreamcast will become a de
facto standard for digital entertainment" claimed Sega chairman Isso
Okawa. However, it was at E3 itself that the tide really began to turn
for Sega with bravura software demos finally earning the machine
journalists' respect. Post E3 reports were full of adoration , as
impressed by the restoration of Sega's old self-confidence as the raw
processing power on show. Dreamcast's launch date was set as November
20th and this time all Sony can threaten is the announcement of new
hardware -- 1998 is Dreamcast's alone.

From E3 onwards, Sega orchestrated a careful drumbeat of
announcements, including the launch of the VMS unit on July 11th to
tie-in with the Godzilla movie and a much hyped August 22nd PR event
for Sega's old mascot in Sonic Adventure. In September, Sega ran an ad
showing MD Eiichi Yukawa being abused by members of the public who
preferred Sony -- and promising all would change with Dreamcast's
arrival. And so it is, everything now rests with the machine and its

I was devastated when SEGA decided not to use Lockheed Martin Real3D in Dreamcast.
I mean, PowerVR2 was great, but not Real3D-great. I figure Lockheed could've come up with a cost-effective next-gen GPU beyond what was in Model 3 to compete with the other consoles of its generation.


Mar 29, 2012
I also found what I thought to be a very interesting USENET post about 3DFX and Real3D:

First, let me start off by saying I am going to be buying a Voodoo card.
For low end comsumer grade flight sims and such, the Voodoo looks like
about the best thing available. Second, I am not necessarily responding
to just you, because there seems to be a hell of a lot of confusion
about Lockheed Martin's graphics accelerators. I have been seeing posts
all over the place confusing the R3D/100 with the AGP/INTEL project that
L.M. is working on. The R3D/100 is *NOT* the chipset that is being
developed for the AGP/INTEL partnership.

However, since your inference is that the Voodoo is faster than the
R3D/100, I have to say that you are totally dead wrong. While the specs
say that the Voodoo is *capable* of rendering a higher number of pixels
per second, or the same number of polygons per second as the R3D/100,
the specs fail to mention that these are not real world performance
figures any you probably will not ever see the kind of performance that
3Dfx claims to be able to acheive. This does *not* mean that the Voodoo
is not a good (its great actually) card, just that the game based 3D
accelerator companies (all of them) don't tell you the whole story.

The Voodoo uses a polygon raster processor. This accelerates line and
polygon drawing, rendering, and texture mapping, but does not accelerate
geometry processing (ie vertex transormation like rotate and scale).
Geometry processing on the Voodoo as well as every other consumer (read
game) grade 3D accelerator. Because the cpu must handle the geometry
transforms and such, you will never see anything near what 3Dfx,
Rendition, or any of the other manufacturers claim until cpu's get
significantly faster (by at least an order of magnitude). The 3D
accelerator actually has to wait for the cpu to finish processing before
it can do its thing.

I have yet to see any of the manufacturers post what cpu was plugged
into their accelerator, and what percentage of cpu bandwidth was being
used to produce the numbers that they claim. You can bet that if it was
done on a Pentium 200, that the only task the cpu was handling was
rendering the 3D model that they were benchmarking. For a game,
rendering is only part of the cpu load. The cpu has to handle flight
modelling, enemy AI, environmental variables, weapons modelling, damage
modelling, sound, etc, etc.

The R3D includes both the raster accelerator (see above) and a 100 MFLOP
geometry processing engine. Read that last line again. All geometry
processing data is offloaded from the system cpu and onto the R3D
floating point processor, allowing the cpu to handle more important
tasks. The Voodoo does not have this, and if it were to add a geometry
processor, you would have to more than double the price of the card.

The R3D also allows for up to 8M of texture memory (handled by a
seperate texture processor) which allows not only 24 bit texturemaps
(RGB), but also 32bit maps (RGBA) the additional 8 bits being used for
256 level transparency (Alpha). An addtional 10M can be used for frame
buffer memory, and 5M more for depth buffering.

There are pages and pages of specs on the R3D/100 that show that in the
end, it is a better card than the Voodoo and other consumer and
accelerator cards, but I guess the correct question is, for what? If
the models that are in your scene are fairly low detailed (as almost all
games are - even the real cpu pigs like Back to Bagdhad), then the R3D
would be of little added benefit over something like the Voodoo.
However, when you are doing scenes where the polys are 2x+ times more
than your typical 3D game, the R3D really shines. The R3D is and always
was designed for mid to high end professional type application, where
the R3D/1000 (much much faster than the 100) would be too expensive, or
just plain overkill. I've seen the 1000 and I have to say that it rocks!
I had to wipe the drool from my chin after seeing it at Siggraph (We're
talking military grade simulation equipment there boys, both in
performance and price!)

Now then, as I mentioned before, I'm going be buying the Voodoo for my
home system, where I would be mostly playing games. But, I am looking
at the R3D for use in professional 3D application. More comparible 3D
accelerators would not be Voodoo, Rendition based genre, but more along
the lines of high end GLINT based boards containing Delta geometry
accelerator chips (and I don't mean the low end game base Glint chips,
or even the Permedia for that matter), or possibly the next line from
Symmetric (Glyder series), or Intergraph's new professional accelerator

Ted K.
Shadowbox Graphics
Chicago - where being dead isn't a voting restriction.


Apr 18, 2011
The Sega Dreamcast is one of those things that everyone loves in hindsight. Unlike most things wrapped in nostalgia however, it's because the Dreamcast was amazing. Maligned by sales but not by performance, anyone who owned a Dreamcast in that window when it was the most innovative thing on the market and remembers the experience of playing what at the time were the most realistic sports games ever created knows what I'm talking about.

Great bit of info here and it makes me curse Sega's all-in move with Shenmue. =(


Mar 29, 2012

I tried to gather as much interesting info as I could.

There's a little bit more, and I may post it later if I can find it, but anyway, thanks for the compliment.


Mar 12, 2012

I can sit on my couch and see my original dreamcast which I bought new. My receiver upscales s-video and it looks great on my 55" lcd tv. I am not a sports game fan, but PSO and Skies of Arcadia still look and play great. Opening screen and theme for PSO = bliss.


Mar 29, 2012
As I said I have a little more, it's not much, and it's only from EGM's Quartermann,
but it's in-line with the far superior Next Generation article

From EGM in late 1996 early 1997


I know that's too blurry, so:

In other Sega news, Yu Suzuki sand and the white shirts at AM2 are currently knee deep into the development of VF3 for the Saturn, which will be released in Japan around October. The game (a CD) is designed to run in conjection with a 3-D cartridge upgrade that plugs into the port on top of the Saturn...can you say 64X? The Lockheed Martin Corportation (the company that designed Sega's Model-3 arcade architecture) is currently working on the 64-bit cart, which is based on the Real 3D chipsetm LMC's upcoming 3-D accelerator for the PC. The entire package is targeted to retail for 9800 yen in Japan (about $90 U.S.) with 6000 yen of that for the CD and about 3500-4000 yen toward the cart. Our Q-spies report that VF3 will be but just a small taste of Sega's 64-Bit console technology. Sega has also commissioned LMC to design a killer 64-Bit game system code named Pluto. The new system, due out in early 1998, is said to offer 3-D performance that could rival (if not surpass) the Model-3 arcade board. Look for Sega to make an official announcement of the new console (along with the first look at VF3) at the upcoming Tokyo Toy Show in June.


Mar 29, 2012
Probably the last bit of info I have for now.

This comes from a interview (no link sorry, not working)

Interview: Toshiyasu Morita

Watching Sega of America go through its different stages, from a tiny division with no software development a company that strictly makes software, one tends to think that all the events that befell it came out of the blue. As many who worked there can attest, this is simply not the case. Even something like the 32X, which was pushed as an inexpensive alternative to the upcoming and much more expensive Saturn, had its future foreseen by those who worked on it.

One such person is Toshiyasu Morita, a technical director at Sega of America throughout its 16-bit dominance and past the demise of the Dreamcast. He was involved with the 32X and saw its launch, and he was also involved with several Genesis and 32X titles, such as Zombies Ate My Neighbors and Doom. He recently spoke to us about his time at the software giant and his experiences making games.

Sega-16: Before coming to Sega, you were a programmer and technology manager at LucasArts. What prompted the move to the House of Sonic?

Toshiyasu Morita: I went from Manager of New Technology to Technical Director, and there was also a raise.

Sega-16: As a Technical Director at Sega, what new hardware did you get to evaluate? Were there any that never saw release?

Toshiyasu Morita: I evaluated a lot of hardware, mostly PC 3D hardware such as the SMOS Pixelsquirt, Lockheed-Martin Real3D, and processors such as the PowerPC. Was involved in two pieces of hardware which were never released: an SH3E+Nvidia combo which never went anywhere, and I was on the compiler/debugger guy for the SH4+3Dfx board.


Mar 29, 2012
Anyone wish to discuss this further? I really would.

By April 1997, Next Generation Online discovered that Lockheed Martin would not be involved with Sega's home console plans, and that Black Belt would not be an upgrade for Saturn but a whole new console.

Black Belt from a Lockheed Perspective
Two former Lockheed Martin employees, N-Space's Erick a lovely person and Dan O'Leary voice their views on Sega's move to use 3Dfx instead of a Lockheed Martin solution.
April 29, 1997

With experience in developing for Model 2 (Desert Tank) and having helped develop the Model 3 hardware while at Lockheed Martin, Erick a lovely person and Dan O'Leary have indicated that it would have been difficult for Sega to make a better decision in terms of a graphics subsystem.

"3Dfx has proven itself. Just look downstairs (at CGDC). Nearly every major demo at every booth is running off of some form of the Voodoo graphics chipset," said O'Leary. While consumers have yet to establish a standard in 3D acceleration, most of the developers projects and demos were using Voodoo as their target platform.

Commenting upon the strengths of the proposed Black Belt a lovely person said: "Not only is Sega getting the hottest chipset around, but with Microsoft in its corner it will be getting useful libraries; something the Saturn desperately lacked."

The major question facing the duo was why did Sega neglect its long-term hardware partner Lockheed Martin when designing the hardware? O'Leary stepped up to the plate answering: "Sega has to find the cheapest but most powerful hardware it can. Lockheed Martin is still trying to figure out how it fits into the consumer space seeing as it has traditionally worked in the simulation arena. 3Dfx on the other hand was created from the ground up to be a consumer level product. It isn't at all surprising that Sega has gone this route."

When comparing Lockheed's Model 2 and Model 3 hardware to the proposed Black Belt specification, both O'Leary and a lovely person felt that that Black Belt would be far more similar to developing for the Model 2 than Model 3. "The Model 2 is a beautiful board that is simple to get right to the metal, " said a lovely person. "The Model 3 was designed around more of a traditional simulator model with a host and GPU arrangement where the database runs the entire game."

While a lovely person mentions getting to the metal easily, some developers such as Scott Corley and Dave Perry both voiced some concern over Microsoft's OS getting in the way. "Good developers will cut through the OS to get to the metal as they need it." says a lovely person. "As long as Microsoft doesn't force the OS upon the developers it should be fine."

With the ease of development that is expected to go along with the system, and the double-edged sword that this situation can present, a lovely person said that Sega's quality assurance program should help to weed out games from developers that are relying too much upon the base libraries or that are quick ports of substandard PC titles.

Both a lovely person and O'Leary also pointed to one non-technical element that is different at Sega presently than it was at the launch of the Saturn: executive personnel. Both men cited the fact that Bernie Stollar was a major factor for the third party support that PlayStation enjoys and the fact that Stollar is now responsible for generating that same third party support for Sega. "They've assembled a really good team at Sega now and it's going to be interesting to see what the next generation brings." said a lovely person.

Black Belt eventually lost the internal competition within SEGA in favor of Katana in the summer of 1997, which was then named Dreamcast in May 1998.


Nov 20, 2011
So that's proof that SEGA of America was behind 32X not SEGA of Japan.

Its amazing because at Sega Forums, most of the users there blame SOJ for SEGA's downfall in the late 1990s not Sega of America's mismanagement.

Also the article does state that "Black Belt" was indeed not only developed by Skunkworks as Haneda had stated back in 2010 at Sega Forums, but that it WAS the chipset that SEGA had its heart set on for Dreamcast.