Lockheed Martin Real3D/100 and SEGA

SEGA should've used Lockheed Martin Real3D/100 for Saturn




I so badly wanted the Saturn to have had an upgrade based on Real3D/100. That or Sega wait until 1996 to release Saturn based entirely on Real3D/100 and an IBM PowerPC 603 CPU. Imagine the games that would've been made for such hardware. It would've been superior to the 3DO M2 console and 3DFX Voodoo PC cards. Sega could've ported and upgraded all Model 2 arcade games to Real3D/100 and downported Model 3 (was based on Real3D/Pro-1000) games to the lower-end Real3D hardware for Saturn.


I wanted to see Tomb Raider II running on the Saturn upgrade cart which was based on Real3D/100. It's too bad Sega let Saturn be such a difficult system to program for with no proper 3D polygon rendering.


Just thinking of how things could've turned out differently.

 

x

 

You totally missed the point of my post. I *know* Saturn had no relation to Model 1 or Model 2. What I was trying to say is that Sega should've either based the Saturn on Real3D/100 (which is not the same hardware as Model 1/2) or upgraded Saturn with Real3D/100 via the cartrige port.

Also, Saturn is not directly based on the System 32 sprite scaler arcade board, although Saturn did start off as the GigaDrive which was actually in fact based on System 32. However the final Saturn has much different hardware than System 32 so it is not correct to say Saturn is based on it.

 

x

 

The Saturn 2 article by Next Generation magazine explains what I was hoping for.



Also, the article titled "The Dreamcast Story" mentions how some of Sega's management wanted to scrap the Saturn altogether in favor of a LMC designed console.

 

Would have loved to have 16bit cart and cd compatibility on the Saturn back in the day.

I guess one thing I'd like to know is would the Staurn Cart slot have been able to support faster cpu's at all? does it work like the upgrade slot on my Amiga 1200? On one card I have a faster 68x and a PPC CPU, more faster RAM and drive controllers.
Would the Cart slot have worked for this scale upgrading?

 

I once read that Sega deliberately avoided giving the Saturn backwards compatibility with the Megadrive because the company was looking for a clean break with a standalone console in a new hardware generation, also dropping its various failed projects during the 16-bit era as a result of this decision. On the other hand, there were plenty of Megadrive ports, compilations and upgrades released for the Saturn, so perhaps that wasn't entirely the case. As for the ill-fated Eclipse cartridge accelerator, wasn't that designed to fit into the port used by the Video CD Card?

 

does the decoder card port have more system access then the cart slot port?

 

x

 

We have had this conversation before. The Real 3D/100 hardware would have been way too expensive to go into the machine in 1996. Part of Sega's problem was the price of the hardware already.

 

Yes we have had this conversation before, but I wanted to have it again.

While true, it would've been too expensive as it was originally designed by 1995, a 3 chip solution (geometry processor, texture processor, graphics processor) however if the 3 seperate processors had been combined into a single chip solution, put on a smaller nm process and mass produced for the console market, I believe it could've been cheap enough for use in a console in late 1996. Also don't forget that RAM prices dropped drastically that year, too. Now I believe it could be said that the Nintendo 64 chipset, the 3DO M2 chipset and the 3DFX Voodoo chipset would've been too expensive for 1996, but they were (or could've been in the case of 3DO M2) introduced at mass market prices for the reasons I listed. The Saturn was a mess. 8 chips, IIRC. A single chip Real3D/100 plus PowerPC CPU plus RAM could've cost less in 1996 than Saturn's huge list of chips did in 1994/1995.

The original PlayStation was a good example of chip integration and would've cost a fortune without it. Even moreso with the 3DO M2. The M2's BDA (Bulldog ASIC) had 10 processors in it, but with integration, it became a mass market machine--although it just never got released as a console. Also don't forget that Real3D/100 was not the super expensive Real3D/Pro-1000 used in pair in the Model 3 arcade board. If Lockheed had been more willing to integrate their multi-chip boards into single-chip solutions the Real3D/100 would've been perfect for Sega to use it to counter N64, 3DO M2 and all PC 3D accelerators. Sega would've had an awesome console to port their arcade games.

For the Dreamcast generation, Sega and Lockheed should've worked on a totally custom "Real3D/500", a GPU with dozens of times more performance than a single chip Real3D/100. Such a GPU would've far surpassed Nvidia's GeForce 256 (NV10) and been superior to even the original Xbox. With backing from Microsoft, Lockheed Martin, IBM and other hardware partners, the 3rd party devs would've flocked to this system instead of PlayStation2. Sega would've been in a vastly, vastly different situation, then and now.



IMO it'ssad though, even Lockheed Martin admitted Sega was going the right direction with the Black Belt using 3DFX.

http://web.archive.org/web/19970605161903/www.next-generation.com/news/042997b.chtml

 

Card slot has most/all signals also wired to the cartslot, so it would not matter so much which port you used. The cartslot is very versatile, and its also bootable :3

 

I was always under the impression that the 3D upgrade would've used the cartridge slot, not the MPEG video port.

 

Is this perhaps a reference to the cartridge being currently worked on that allows you to boot a copied disc as if it was an original, bypassing the Saturn's protection checks? I've heard this is being made in conjunction with a homebrew project, and that it will be modified so that you can only use that particular game - if they didn't include such a feature, my guess is that most people would simply buy it (providing this is ever completed, of course!) for the cart and possibly draw a lot of negative attention in the direction of its creators...

 

Here's the block diagram of the Real3D/100



This could've been used in a Saturn upgrade or replacement in 1996, and been better than 3DO M2 or 3Dfx Voodoo. It would've also allowed Sega to easily port their arcade games home. While Real3D/100 was not nearly as powerful as the Model 3 board, the similar architecture would've allowed decent Model 3 to home conversions.

 

I'm guessing the reason they didn't go the upgrade route was because of how miserably that strategy had failed in the case of the Mega Drive/CD/32X. I think they learned that when releasing new hardware, it's better for it to be seen as completely separate and different from the existing hardware - i.e. to be seen as "next generation". Otherwise, people are either confused or are simply unwilling to fork out more cash for a system they already own. This is still the case, I think. Look at Kinect/PlayStation Move/Wii Motion Plus. Very similar to the Sega CD/32X, and have all been commercial failures.

From a collector's standpoint, it's easy to say that it would've been awesome if we had a near-Model 3 Sega console to play around with. But what a collector wants is not at all what the market at large wants. People like things that are cheap and that blow their hats off. A slightly upgraded Saturn probably wouldn't have appealed to many.

Another possible reason is that the Saturn was much more popular in Japan than in the west, and at the time the Japanese market was more interested in 2D games. In other words, a 3D upgrade card probably wouldn't have appealed to the majority of people who owned Saturns at the time.

I think a more pertinent question to ask is why didn't Sega port more Model 3 games to the Dreamcast? Obviously the DC was not as powerful, but then again the Saturn was significantly less powerful than the Model 2 and there were tons of great Model 2 to Saturn ports. Sega could've easily made a great (though not arcade-perfect) port of Scud Race. Maybe it would've happened if the DC's life wasn't cut so unexpectedly short.

It's all just speculation, anyway. Interesting, but still speculation.

 

A Real3D/100 armed Saturn would not have looked or felt anything like the Saturn we got. It would've been much more than a slightly upgraded Saturn. It would've been inbetween Model 2 and Model 3

I agree we should've seen more Model 3 games on Dreamcast even if they weren't perfect ports. The Dreamcast was good enough to do decent Model 3 conversions, better than Saturn was able to do Model 2 conversions. Scud Race would've been great on Dreamcast, much better than the dev-kit demo.

 

x

 

Why? It's a dead system. I don't hear the Everdrive or similar devices getting much negative attention.

 

It might've seemed like half a Dreamcast but don't forget Real3D/100 could only push 750,000 polygons/sec while Dreamcast could push 3,000,000 to 5,000,000 polygons/sec. So in reality Real3D/100 would not have been half a Dreamcast although it would've been better than PS1, N64, Sega Model 2, 3DO M2, and every other PC 3D accelerator out there in the 1996/1997 timeframe. With a Real3D/100 powered Saturn, Sega could've delayed Dreamcast until 2001 and used much better technology (from IBM and Lockheed of course).