UntoldDreams said:
A lot of this comes down to Kuturagi's original vision.
(It was in a Japanese article I don't have the link anymore but this is true stuff)
His "Dream" (oh krazy ken you wabbit) was for Japan/Sony to create a new computer architecture.
He dreamed that the Cell Processor would be the first step towards seeding Earth with a new architecture which could rival Apple computer and the like.
These Cell computers would be capable of joining in a GRID and having ridiculous computational power much like Folding@Home does but more organized and controlled.
Imagine a company of 200 people where they all had these "Kuturagi computers" which if you wanted you could join them together. It would mean every small company out there would have some form of Super computing power readily available.
That, was his dream and it was not inherently wrong it actually could have worked except for the simple fact that Sony ran out of money and the economy took a nosedive.
*** It just wasn't the right timing for him and maybe Sony wasn't the right company
Too bad I think if the timing had been different Google would have jumped on the chance to seed Earth with their Linux based Google OS.
Ken was a visionary despite his crazy moments.
I also followed every piece he said back in the days. Sony had an idea (has?) of
a new computer based on a new architecture that should overcome the burden of
the von Neumann architecture to open up the door for massive multimedia
applications on your desktop. Ken Kutaragi nicknamed this computer Creative
Workstation. The Graphics Synthesizer Cube, the GSCube, was part of his vision
towards this goal;
"As Sony Computer Entertainment fused the technology of computers and world of
entertainment and evolve a new market called Computer Entertainment, we are now
pioneering the creation of realtime digital entertainment through an extension
of the technology of PlayStation2." - Ken Kutaragi, CEO of Sony Computer
Entertainment America
GSCube
gamasutra.com said:
... The GScube is a rendering device targeted at the content creation/broadband delivery markets. It consists of 16 cascaded processing units, each of them based upon an enhanced version of a Sony Playstation 2. Every processing unit has an Emotion Engine CPU (which was jointly developed by Sony and Toshiba), and an improved Graphics Synthesizer equipped with a 32MB frame buffer (eight times the memory of a regular PS2). This yields a total 512MB of VRAM, and it can theoretically reach a peak performance of 1.2 billion triangles per second -- a number that sounds like it comes from science fiction books. The device must be controlled by an external broadband server which feeds data to the GScube, and at Siggraph that device was the brand-new SGI Origin 3400. At the Sony booth, we enjoyed a real-time battle between characters from the movie Antz rendered in real time, as well as interactive sequences from the upcoming Final Fantasy movie shown at 1920x1080 pixels and a sustained rate of 60FPS.
In the Antz demo, I counted 140 ants, each comprising about 7,000 polygons, which were rendered using a ported version of Criterion's Renderware 3. All ants were texture mapped, and the results looked surprisingly close to the quality of the original movie. The Final Fantasy demo was just data from the now-in-development full-length CG movie based upon the game series, rendered in real time by the GScube. It showed a girl (with animated hair threads) in a zero-gravity spaceship, with a user-controllable camera viewpoint. The demo rendered about 314,000 polygons per frame, and included an impressive character with 161 joints, motion-blurring effects, and many other cinematic feats. According to Kazuyuki Hashimoto, senior vice president and CTO of Square USA, the GScube allowed them to show real-time quality, in "close to what is traditionally software rendered in about five hours." Sony believes that the GScube will deliver a tenfold improvement over a regular PS2, and future iterations of the architecture expect to reach a 100-fold improvement.
...
GSCube 16x specification:
- CPU 128Bit Emotion Engine x 16
- System Clock Frequency 294.912MHz
- Main Memory Direct RDRAM
- Memory Size 2GB (128MB x 16)
- Memory Bus Bandwidth 50.3GB/s (3.1GB/s x 16)
- Floating Point Performance 97.5GFLOPS (6.1GFLOPS x 16)
- 3D CG Geometric Transformation 1.04Gpolygons/s (65Mpolygons/s x 16)
- Graphics Graphics Synthesizer I-32 x 16
- Clock Frequency 147.456MHz
- VRAM Size 512MB (embedded 32MB x 16)
- VRAM Bandwidth 755GB/s (47.2GB/s x 16)
- Pixel Fill Rate 37.7GB/s (2.36GB/s x 16)
- Maximum Polygon Drawing Rate 1.2 Gpolygons/s (73.7Mpolygons/s x 16)
- Display Color Depth 32bit (RGBA: 8 bits each)
- Z depth 32bit
- Maximum Resolutions 1080/60p (1920x1080, 60fps, Progressive)
Sony was even eager to deliver a 64x version of the GSCube;
eetimes.com said:
Sony steps up Playstation-based graphics system plans
Yoshiko Hara Yoshiko Hara
EE Times
(09/13/2000 2:55 PM EDT)
TOKYO — Sony Computer Entertainment Inc. has accelerated its plans to roll out a high-end graphics computer based on its Playstation game console technology. The company announced plans this week to introduce a system called GScube next year that will use 64 processor boards with Playstation 2 technology.
The resulting parallel-processing computer will act as a graphics visualization machine with a 3-D processing capability of 4.16 gigapolygons per second and a resolution of 60 frames/second (progressive scan) at 1,080 x 1,980 pixels, the company said.
The 64-board GScube will put Sony ahead of the graphics system road map it announced a year ago, when it promised to develop a system with 10 times the processing clout of Playstation 2 in 2000, followed by a 100-times version in 2002 and a 1,000x version before the end of the decade.
...
But the 64x version never saw the light of the day.
If one looks closely at the Cell architecture, then one can see that the Cell
processor is a very flexible extension of the PS2's Emotion Engine (EE), i.e.
each vector unit (PS2:VU, PS3:SPE) now got his own DMA engine and a lot more
memory and other stuff. The busses were also improved a lot, leading to the
Elementary Interconnect Bus (EIB). One can say that the Cell processor houses
4 Emotion Engines just going by the vector units.
Whereas the Emotion Engine architecture was fixed by design, the Cell
architecture (the CBEA) is quite flexible and extendable -- ready to make Ken's
vision come true. But the situation turned out differently, unfortunately. The
CBEA is great by design, but building a great design is only one side of the
coin, having the right development tools (compiler, profiler, ...) in place
makes up for the other side. And as such the Cell processor never really took
off until recently. Only those who knew how to program such a processor without
the need of high-level tools could take advantage of it right from the start
(read; scientific computing).
Due to the lack of developer support and other issues, Ken's vision lost
momentum. In his vision the PS3 was just the beginning of a new computer. If
the PS3 would had been successful right from the start, I reckon Sony would had
offered a (kind-of) PC based on the PS3 utilizing a lot more RAM and other
things. And I do believe that
Sony's ZEGO BCU-100 board is what would have
become the motherboard of said Creative Workstation. By IBMs roadmap we should
had an PowerXCell 32iv processor by now, envisoned as follows;
- 4 PPEs + 32 eSPEs,
- higher frequency (~3.8GHz),
- 100% backward compatible,
- Performance on PPE significantly better,
- Performance per SPE equal or better,
- Better inter-SPE latency,
- More on-chip memory,
- Better main memory latency and bandwidth,
- ~ 1 Tflops,
But unfortunately, this vision was abandoned about a year ago. Imagine a ZEGO
BCU-200 board with a PowerXCell 32iv processor + 16GB RAM and stuff. That's the
computer I wanna have!
Anyway. The Cell architecture is quite beyond and it's pretty cool to be able
to access the Cell processor via OtherOS to tinker round with it. 3.21 D'oh!
Well, I hope Sony sticks with Cell for the PS4 such that they don't have to
build a new toolchain, dev-tools, etc., again. A slightly enhanced Cell is all
what is needed. Two of them, a little more RAM, and throwing out the weak RSX
for something better ... and there we go again!