'Sega Saturn M2: The Facts' - M2 rendered VF2 ??

Taken with a MASSIVE grain of salt.

http://www.gamespot.com/pages/unions/read_article.php?topic_id=25962473&union_id=12112

I don't remember reading this before. The supposed M2 rendered VF2 Pai looks realtime and about the quality we've seen from M2 graphics.

 

interesting.. I call bullshit!

 

it seems a pc rendered picture. too antialiased too.

beside that, all BS IMHO.

 

Yeah I guess this was complete BS (The M2 VF2). Although Sega working with M2 kits is pretty well known.

 

That image looks too much like Pai as she appears in the Virtua Fighter CG Portrait Collection to be a coincidence, so I have to call bullshit as well on the legitimacy of that being a render - whether in real time or not - from VF2 running on the M2. However, what I can assure you all of is that Sega was indeed preparing a VF conversion for the ill-fated console, though I've read it would have been the third game in this series and not its predecessor. Besides, wouldn't VF2 have been a little outdated by the time this platform was ready to launch, not to mention the fact Saturn owners already had a decent enough version of their own? I'm not exactly sure why the proposed deal fell through, but I definitely remember reading somewhere that Matsushita was involved by this point so you can't blame Sega not wanting to collaborate with non-Japanese companies for the collapse. Of course, it's well known that Sega was indeed starting to make decisions internally that led to a power struggle, often favouring its Japan divisions regardless of who actually had the better ideas. This ultimately led to the US wing of Sega being forced to take a back seat, which naturally limited any further mainstream success... and for a perfect example you only have to look at the MegaDrive era, when SOA called many of the best shots.

There were two SOJ programmers assigned to the task of creating in-house library tools for any support of the M2, though I've no idea who these were. Now, this might be a total guess, but one would probably be hardware R&D chief, Hideki Sato. As for the other, I'm sure AM2 lead programmer Keiji Okayasu would have been involved to some extent if a VF conversion was being planned. I'd sure love to know more for definite, as this isn't entirely a made-up story. The fact is that Sega briefly toyed with the idea of developing for the M2, but the idea this was being touted as a Saturn upgrade flies in the proverbial face of all we've since discovered about the actual Eclipse project. Anyway, that was being cooked up by US staff, though it wouldn't be the first time rumours from this period would totally confuse Sega enthusiasts - some people still ask SOJ members about the cartridge accelerator, despite it only being shown in working form to those in America who participated in its highly secretive creation over there! To this day I've read the comments of people who believe AM2 was utilising such technology for their unreleased Saturn versions of VF3 and what later became Shenmue. The only game officially confirmed to have any ties with Eclipse was Tomb Raider II, and I'm sure Core may have put together other tech demos during their acquaintance with the add-on that was never going to make it with SOJ taking back power. Can anyone say SCUD Race, perhaps? Well, Core did make the later Katana development kit rolling sample, though I'd rather not hijack this thread too much with my speculation on such off-topic points...

 

No please go on. It might be off-topic but this is the best post so far, and this thread is gonna die anyway. Might as well hear some more interesting info since you know a lot about these subjects, more than most people!

 

They worked with Nvidia, 3Dfx and Lockheed Martin.

Source?

 

NVIDIA collaborated with Sega on the Diamond Edge card, which previously formed the basis of an early design for the Saturn, so it's possible this former partner was contacted to help on the Eclipse project at some point. 3DFX was also definitely involved, though later pulled out because it preferred the idea of providing technology for a new Sega console, and this led to the US-based design rejected in favour of the NEC-backed Katana concept. As for Lockheed Martin, those in the know will confirm the working Eclipse prototypes featured a variant on the same Real3D processor found in the Model 2 board, though I'm not really sure this would have allowed the Saturn to reach Model 3 level quality of graphics... even with the added performance of the base platform taken into consideration. Finally, returning to the suggestion of VF3 being M2-bound, that came from an early update of the Wikipedia article on this game. I've done a bit of editing to this myself over the years, but for some reason it's received so many alterations (possibly from Sega insiders?) that it became a pointless exercise... I just don't have the spare time to keep changing things anymore. The fact remains that VF3 was briefly touted as Sega's key contribution to the M2, but this was a very short-lived idea. Even their neighbouring departments had no idea just how much work AM2 had put into a Saturn conversion of VF3, and closer to home for Yu Suzuki's team, to this day his replacement still doesn't believe such a thing existed despite a reliable and long serving - not to mention high ranking - member of Sega's European operations having in his possession a finished copy of the game. Once again, I don't want to stray too far off-topic for now!

 

The Saturn launched in Japan in November 1994, and it must have been completed at least sometime earlier that year.

The Diamond EDGE 3D card which contained Nvidia's NV1 chip, did not launch until late 1995, after the U.S. Saturn launch. I'm pretty sure the EDGE 3D/NV1 did not influence Saturn's design at all, but more like the other way around. The EDGE 3D card had Saturn controller ports, and it recieved several conversions of Saturn games (i.e. VF Remix, Panzer Dragoon).

Sega and Nvidia collaborated on the never-released NV2 chip during 1995-1996 after both the Saturn and EDGE3D/NV1 had been released. It was around the same time that Sega was also looking at 3DO/Matsushita M2, the first-gen 3DFX and PowerVR chips and (at least just looking) at Lockheed Martin Real3D.

The MODEL 2 board didn't technically use Real3D tech, it used Martin Marietta texture-mapping chip technology in 1993 (games released in 1994). So Model 2 was pre-Real3D, before Lockheed merged with MM to form LM and before LM formed the Real3D division/company from the combined graphics resources of General Electric Aerospace, Martin Marietta and Lockheed themselves. After Sega/MM MODEL 2, the new Real3D company came up with the midrange R3D/100 chipset for PCs and highend R3D/Pro-1000 chipset for image generators and MODEL 3 which all happened in 1995 (Model 3 games launched in 1996).

Nothing that I am saying in this post is inside info. This is all publicly available information. Also nothing here (above) that is rumor.


Back to M2

 

x

 

3Dfx blew it all right. I dunno if they killed the Saturn (perhaps they did indirectly) but they sure killed any chance for the Black Belt to become Sega's new system

http://www.gamespot.com/pages/unions/forums/show_msgs.php?topic_id=26533319&union_id=17347

http://www.answers.com/topic/voodoo-graphics

 

I know it was alleged that Sega chose NEC to provide the chipset for its next console as a corporate favour on direct orders from Hayao Nakayama, but could the Black Belt project have really failed because 3DFX shot itself in the foot by sharing details of its own contract? Also, I'm not very familiar with the history of Lockheed Martin, so I always thought they were fully responsible for the technology in Model 2. If they did just come up with the final board's texturing abilities, it's hardly surprising this particular element of so many games had to be significantly reworked for Saturn conversions. While arcade programmers had no problems with being limited to monochrome, I distincly remember this being quoted in reference to certain AM ports. Finally, I thought NV1 dated from before the final Saturn design, and may have even originated from the early prototype. Of course, there was also the System 32-based version, something later recycled (in part) as the Nintendo 64 and also the original, single-chip idea that would be improved - or should that be over-complicated? - to enable more than just "modest" support for three dimensional graphics as the specification once demanded.

 

x

 

Brief history of Sega working with the defense industry to co-develop the ModelX boards.

Sega worked with General Electric Aerospace on the MODEL 1 board which launched in 1992 with Virtua Racing. GE Aerospace provided the polygon engine (made from a bunch of DSPs). When Sega's own efforts to upgrade the MODEL 1 with texture mapping failed, they turned back to GE Aerospace, who was now owned by Martin Marietta, in 1993. Martin Marietta (not Lockheed) provided a texture-mapping board with its own chipset and API/tools and whatnot. The additional hardware was integrated into the upgraded MODEL 1, which was now called MODEL 2. It was first seen running an early version of Daytona in mid to late 1993. The completed Daytona USA game was launched in early 1994. Martin Marietta made a game of their own called Desert Tank which came out in late 1994 just before Virtua Fighter 2. MODEL 2 was refined into newer revisions (CRX) for newer games. MODEL 2 became Sega's workhorse and the most successful of the ModelX boards.

In 1995, Martin Marietta merged with Lockheed to form Lockheed Martin Corp. LMC took the graphics aspects of their new company, the technology/IP/patents/the engineers, the buildings they worked in, etc) which LMC acquired from the former GE Aerospace, Martin Marietta, and perhaps whatever graphics tech Lockheed had of their own, if they did, naming it Real 3D. The purpose of the Real 3D division of Lockheed Martin in 1995-1996 was to take all of that graphics tech and decades of expertise those companies had from building an Apollo moon lander simulator in the 60s, many military flight & tank simulators in the 70s, 80s, early 90s, image generators for the industry, MODEL 1/MODEL 2 arcade boards for SEGA, etc, and apply that to both the low-cost PC market and a new generation of arcade technology for Sega (MODEL 3).

The Lockheed Real3D/Pro-1000 GPU technology in MODEL 3 is very different from the Martin Marietta texture-mapping tech in MODEL 2. Think of it like this, 3Dfx makes a Voodoo card for your PC in 1998. Nvidia buys 3Dfx in 2000. Nvidia makes a new generation GPU with its own technology combined with that of 3Dfx, and it's released in 2001. Now that old Voodoo card you have was not made by Nvidia, it was made by 3Dfx, even though the companies are now combined.

So you see, it took time for Lockheed Martin to be Sega's arcade graphics provider, LMC Real3D was the 3rd incarnation of the group that worked with Sega. It's important not to forget GE Aerospace (MODEL 1) and Martin Marietta (MODEL 2) before we get to LM Real3D (MODEL 3).


This article mentions some of the things I mentioned




Also, with the MODEL 1/2/3 boards, GE Aerospace / Martin Marietta / Lockheed Martin were only responsible for the graphics portion of those boards, not the host CPU side or the sound. Yu Suzuki wanted Lockheed to handle the entire MODEL3 board (he also wanted Martin Marietta to do the entire MODEL2 board) but Sega's top brass decided against this, they gave only the graphics portion of MODEL3 to Lockheed because of severe delays. MODEL3 was originally meant to launch in 1995 with 3 games. Instead, it launched in 1996 with only 1 game.

Nvidia was founded in 1993 but NV1 development was entirely seperate from Saturn. Saturn launched in Nov 1994, NV1/Diamond EDGE 3D launched in fall 1995. The NV1 is totally different than Saturn's VDP1 & VDP2 graphic processors. NV1 and Saturn were not developed together, did not have anything to do with each other, as far as I know. Maybe people tie NV1 and Saturn together because they both use quads instead of triangle polygons.

After the NV1 chip was completed when the Diamond EDGE 3D card came along, did they put Saturn controller ports in. NV1 got exellent ports of Saturn games.

Sega and Nvidia started working together on the chip level (NV2) after Saturn launched and after NV1's design was complete.

 

This goes to show that rumours and wishful thinking do not make thing fact, even if they appear to make sense and you want them to be true.

 

The NV1 did not influence the Saturn design, but both used quads instead of polygons. That's why the ports to the NV1 hardware were so solid.

-hl718

 

Exactly. That how I should've said it.

 

too bad...we will never know or see

 

That was originally made for the M2? I though the rumour was that it was a Saturn tech demo for VF3.

Yakumo