||2xPower PC602 @ 66MHz, each delivering over 70 MIPS.
||320х240 - 640х480,full color.
||16bit DSP @ 66MHz. Quad-speed CD-ROM.
|BDA (custom ASIC):
||Memory control, system control and video/graphic control. Includes setup engine, triangle engine, MPEG decoder, DSP for audio and different kinds of DMA control and port control. Random access of frame buffer and z-buffer possible at the same time.
|CDE (custom ASIC):
||Includes a power bus which is connected the BDA and the CPU. The 'bio-bus' is used as a low-speed bus for peripheral hardware.
||Unified memory system of 8MB SDRAM,(64bit bus) @ 500 Mb/sec, average access greater than 400MB/per second.
||Flat and Gouraud.
||Decal, modulation blending, tiling (16K/128K texture buffer built-in).
||Linear, bilinear, trilinear, mipmap, LOD, 3D perspective.
||(4 or 7-bit)
Matsushita M2: Hit or Myth?
M2 could very well change the way you game. If the machine ever sees the laight of day...
In the past two years M2 has undergone a metamorphosis from a scarify omnipotent force to something more akin to a sleeping enemy. Consider the evidence: Matsushita, the world's largest consumer electronics company, pays $ 100 million for the exclusive rights to The 3DO Company's M2 technology. With multiple worldwide branding, unrivaled financial clout, and its fingers in every new technological pie, it prepares an assault on the game industry armed with the most powerful videogame technology ever conceived. It looks unstoppable. And yet, two years down the line, there is still no sign of a working unit or any definable strategy from the company What's going on?
Next Generation has traced the development of Matsushita's console since it was first being touted as the natural successor to the 3DO platform. However, after getting its fingers burnt from the failed 3DO Multiplayer, Matsushita now faces a tall order: How to be taken seriously, and more disturbingly, how to release a machine that has already lost a considerable time lead over its forthcoming rivals.
Despite the M2 technology having received a revamp, and despite its undeniable position as the current standard bearer for home 3D technology, there is a lot of skepticism surrounding M2. Many doubt the console will ever be released.
The issue of software development is perhaps the most galling indictment of Matsushita's handling of M2 so far. While the company said in October that the console would have more titles than were available for Nintendo 64 at launch, it is apparent that few developers are working on titles, and those that do have development systems are reluctant to commit resources to the format until Matsushita makes clear its intentions. Rob Povey, technical director at Boss Game Studios, is one licensed M2 developer in the U.S. (others are Iguana and Studio 3DO) who concurs with this view. "To commit to it currently would obviously be a huge risk for any developer. Until Matsushita unveil its plans for developer support, M2 development, at least outside of Japan, will be minimal."
Only three developers officially make up the Japanese M2 roll call: tiny third-party developer Warp (previously a loyal 3DO disciple) and, conversely, the undeniable might of twin heavyweights Capcom and Konami.
So far, Warp has demonstrated its promising adventure game D2 (a sequel to its prerendered 3DO game, D) and the recent Tokyo AOU coin-op show revealed the first Konami coin-op powered by Matsushita's chipset, a cute take on its Pop 'n' Twinbee series using a full polygon environment. Capcom on the other hand has a 3D beat 'em up in development which was recently shown to Next Gen. Another third-party developer is Genki, which is rumored to be working on a shoot 'em up for the system.
In the U.K., Perception has broken its silence about its plans for the system. NG 29 exposed its Power Crystal title, an RPG, although this is being developed on an incomplete development board with only one PowerPC chip — hardly the kind of situation to inspire confidence, no matter how admirable the design ambitions behind it. Boss's Povey believes his team does have a final kit, though.
"Is it final? As far as I am aware it is, however since my units haven't come directly from Matsushita it's hard to say for sure. We had our first development system before Matsushita bought the technology, so we received it from 3DO. We have received updated boards since the sale." Has Matsushita been in touch regarding development for the machine? "No."
This state of affairs obviously bodes poorly for the future of the machine. However, what transpires most from companies that are working with the system, or considering developing pending Matsushita's position, is a unanimous optimism for the 64-bit technology destined to be — or not to be — included in the finished machine. According to Povey, it's more powerful than any other consumer level product he's seen, including PC accelerators.
And this view is echoed by those who have experimented with the console's hardware. However, until Matsushita reveals its own work on the system there exists a disparity between what Matsushita itself claims the console is capable of (Model 3 levels of performance) and the rather less spirited comparisons from third parties of two to three times Nintendo 64's polygon horsepower.
This relative aging of M2 technology could explain the delay. Does the company release the console in six months only to face stiff competition from hardware rivals? Or does it hold on and re-evaluate the technology for an upgrade? One reason for the delay could be the news of the development of M2's big brother, provisionally titled MX. This hardware is the natural successor to M2 and is being developed by the same hardware team that created M2 in 3DO's California offices. "I'm not sure the existence of MX means we will never see M2," adds Povey. "It's normal to start development of your next-generation technology before the current technology is complete."
At the heart of discussions about M2 is the notion that what is really holding Matsushita back from a full-scale assault on the console market is a level of ambition above and beyond the narrow constraints of traditional videogame hardware marketing. Yes, it may have shot itself in the foot with its handling of its 32-bit console but as much of this can be put down to the disappointing performance of the 3DO technology itself as to any negligence on Matsushita's part. Over the coming months it will hopefully become clear if the consumer electronics giant is serious about videogaming or, indeed, if its plans encompass a far wider interactive entertainment agenda. Time will tell.
The Technology Inside M2
1. With two CPUs each offering over 70MIPS there is no question that in terms of raw processing power M2 is substantially ahead of the game. 8MB of unified SDRAM are included with a peak bus bandwidth of 533Mb/sec and an average speed of greater than 400Mb/sec. Simultaneous random access of the frame buffer and z-buffer is possible, too. The BDA is a custom ASIC (application-specific integrated circuit) comprising the system controls for the memory, video and graphic modes of M2.The BDA also includes MPEG decoding and a 16-bit DSP for audio modulation. The CDE is another custom ASIC which handles a low-speed 'bio-bus' and connects the BDA and CPU.
2. The triangle engine is M2's equivalent of the PlayStation's GTE and handles all of the geometry calculations that drive the polygons. On top of this are several different layers of software abstraction that control it and provide easy access for programmers. The setup engine is included in the BDA and processes graphics before output to the screen.
3. The current M2 development environment includes an Apple Macintosh-based M2 card, a PowerPC-based ampiler, linker and assembler and an external quad-geed CD drive. There are PowerPC versions of 3DO chugging and sound tools as well as graphics tools lowing the importing of images from PC, Amiga and SGI packages.
/ NEXT GENERATION #30, June 1997 /