Kuturagi's 2001 Cell vision vs. current Cell chip

CELL will transform the fundamentals of the network topology. The old mechanism functioned by reading memory data into resistors and rewriting the arithmetic into memories. In short, it was just a repetition of loading and storing. Because each cash memory differs in time of access and capacity, it worked out in such hierarchical structure as primary cache, second cache, etc. On the other hand, CELL might completely transform the concept of cache as it would drastically accelerate the speed of networking. What comes into reality is that each of the astronomical number of computers around the globe could unite to form a CELL and operate by one operating system. Each CELL would be the broadband network itself. Just to give you a picture, it is like 1,000 units of computers at one company functioning as one server. In such a networking world, one would only see the overall strength of power decline when one computer drops out and vice versa.

Ken Kuturagi in 2001 basically said:
- Computers are connected via the internet but can't share computing resources
- This is not due to a lack of bandwidth; even if everyone had fibre, things still won't work due to topology
- Cell will somehow change this ("drastically accelerate the speed of networking")

It's 2005. We have all the info about the Cell chip. But in no way do I see how this chip can accelerate networking. And I can't see how it changes the network 'topology'.

Am I missing something?
 
I just had an apocalyptic vision of all computers on the internet melding into one cohesive entity that becomes self aware and enslaves the world. All thanks to CELL.....
 
JF_Aidan_Pryde said:
CELL will transform the fundamentals of the network topology. The old mechanism functioned by reading memory data into resistors and rewriting the arithmetic into memories. In short, it was just a repetition of loading and storing. Because each cash memory differs in time of access and capacity, it worked out in such hierarchical structure as primary cache, second cache, etc. On the other hand, CELL might completely transform the concept of cache as it would drastically accelerate the speed of networking. What comes into reality is that each of the astronomical number of computers around the globe could unite to form a CELL and operate by one operating system. Each CELL would be the broadband network itself. Just to give you a picture, it is like 1,000 units of computers at one company functioning as one server. In such a networking world, one would only see the overall strength of power decline when one computer drops out and vice versa.

Ken Kuturagi in 2001 basically said:
- Computers are connected via the internet but can't share computing resources
- This is not due to a lack of bandwidth; even if everyone had fibre, things still won't work due to topology
- Cell will somehow change this ("drastically accelerate the speed of networking")

It's 2005. We have all the info about the Cell chip. But in no way do I see how this chip can accelerate networking. And I can't see how it changes the network 'topology'.

Am I missing something?

50% Hype and 50% BS, it won't be the last from Ken. :D
 
and here is Kuturagi's ~ Sony's vision of Playstation3 in 1999

http://www.eetimes.com/story/OEG19991006S0040
The Phase 3 workstation, scheduled to appear around 2005-2006, will have the Emotion Engine 3 and Graphic Synthesizer 3, which will have drastically changed architectures. As a result, the workstation will have 1,000 times performance of the development tool workstation, and will handle 4,000 x 2,000-pixel pictures at 24-to-120p. At this stage, Playstation3 will be taking a shape, Sony said.


http://groups-beta.google.com/group/rec.games.video.sega/msg/551d3b67432c8445?dmode=source&hl=en
PlayStation3 Revealed?

Kutaragi breaks moore's law, claims 1000x chip performance for PS3

When playstation creator ken Kutaragi talks about his future plans,
listen.
in July 1996, he was already saying that his next product would be
capable of synthesizing emotions just as music chip synthesized sounds.
Now, with PS2 already in production, Kutaragi has announced his next
act.

First will be the commercialization of the PS2 tool development
station. The tool is a standalone graphic workstation that uses the sam
Emotion Engine and graphics synthesizer chips found in PS, coupled with
a Linux OS. Kutaragi was quoted in an EE times article as saying, "in
the past, workstations and PSx had more power than home game consoles,
so we could used them as development tools. But when the power {of
PS2}matches or exceeds their power, it becomes difficult to use them for
development."

the non-PS develperment-specific version of the tool, called the
creative Workstation, will be marketed to high end graphics
professionals and movie makers as an alternative to current
workstations. "the PC is losing it's position as the technology driver:
so are workstations," Kutaragi was also quoted as saying. Kutaragi also
said he doesn't care if sony actually loses money on the creative
workstations, because he feel the entire system series is an R&D
program. he also plans to open new markets with the workstations,
including digital projection of movies.

the first generation creative workstations, which sony plans to make
available in 2000 or 2002, will offer roughly 10 times the performance
of the tool, using after multiple version of faster-clock-speed emotion
engines and graphic synthesizer chops operation in parallel.For the
second-generation version, due in 2002, Kutaragi promises 100 times the
tool performance, using secondgeneration version of the EE and GS. By
2006, the third generation system will arrive, with EE3 and GS3,
featuring 1000 times the performance of the tool-just in time for the
PS2, which will also be delivered around that date, according to sony
sources mentioned in the EE times article.

HERE's the PlayStation3 part

So what kind of statistics would a game system capable of 1000 times
PS2's performance generate?
If you do the math, that's 66 billion polygon calculations per second.
On a super-high-definition 6400x4800 screen, that's more than enough
polygons to update pixel size triangles more than 120 times per
second.that's both bettor than movie resolution and a refresh rate
faster than the human eye can detect (in fact, those results can be doen
by a mere 3.6 billion polys per second).
so how can ken Kutaragi promise such results when intel and AMD
practically have to bathe their chips in liquid nitrogen just to eke out
another 100 MHz of clock speed without them melting? Simple. Unlike
intel and AMD, who are hamstrung by the need to maintain backwards
compatibility with the primitive, 20 year old 8080 architecture (which
is only two generations removed from the first micro processor, intel's
8008), Kutaragi and team were able to start from scratch, designing
chips from maximum efficiency without worrying about backwards
compatibility (or, apparently, the marking desire to pace changes so as
to introduce new, more powerful chops the next year).

Kutaragi also has benefit of a billion dollar investment in two new chip
fabrication facilities, which enable him to predict the move to faster,
more powerful .15 and .13 micron process chip making facilites(the PS2
and Dolphin will use .18 micron process chips). How will he pay for the
facilities and the development of EE3 and GS3? Consumer sales of PS2.
Don't expect the traditional PS and workstation market (and the
companies that supplies the OSs for them, like Microsoft and sun) to
roll over and hand the keys to the future to sony, but clearly, Kutaragi
is very, very confident in the abilities of the chips his team has
designed to lay down suck a gauntlet. The response of the rest of the
graphics and PSX industry should be interesting, to say the least. "you
never know what will happen, " said Jack Lyon, director of Cnet's
computes.com, as expert on the computer industry, to whom we'll give the
last word in this story:
"This could be a wake-up call to the rest of the industry, the way
sputnik was to the aerospace industry in 1950's"

Source Nex-gen Dec. '99 issue.
 
Pozer said:
JF_Aidan_Pryde said:
CELL will transform the fundamentals of the network topology. The old mechanism functioned by reading memory data into resistors and rewriting the arithmetic into memories. In short, it was just a repetition of loading and storing. Because each cash memory differs in time of access and capacity, it worked out in such hierarchical structure as primary cache, second cache, etc. On the other hand, CELL might completely transform the concept of cache as it would drastically accelerate the speed of networking. What comes into reality is that each of the astronomical number of computers around the globe could unite to form a CELL and operate by one operating system. Each CELL would be the broadband network itself. Just to give you a picture, it is like 1,000 units of computers at one company functioning as one server. In such a networking world, one would only see the overall strength of power decline when one computer drops out and vice versa.

Ken Kuturagi in 2001 basically said:
- Computers are connected via the internet but can't share computing resources
- This is not due to a lack of bandwidth; even if everyone had fibre, things still won't work due to topology
- Cell will somehow change this ("drastically accelerate the speed of networking")

It's 2005. We have all the info about the Cell chip. But in no way do I see how this chip can accelerate networking. And I can't see how it changes the network 'topology'.

Am I missing something?

50% Hype and 50% BS, it won't be the last from Ken. :D


yeah. because now with PS4 going into the planning and early R&D phase, Ken is going to have to say something like, "we will achieving more processing and intelligence than the human mind"
 
Megadrive1988 said:
Pozer said:
JF_Aidan_Pryde said:
CELL will transform the fundamentals of the network topology. The old mechanism functioned by reading memory data into resistors and rewriting the arithmetic into memories. In short, it was just a repetition of loading and storing. Because each cash memory differs in time of access and capacity, it worked out in such hierarchical structure as primary cache, second cache, etc. On the other hand, CELL might completely transform the concept of cache as it would drastically accelerate the speed of networking. What comes into reality is that each of the astronomical number of computers around the globe could unite to form a CELL and operate by one operating system. Each CELL would be the broadband network itself. Just to give you a picture, it is like 1,000 units of computers at one company functioning as one server. In such a networking world, one would only see the overall strength of power decline when one computer drops out and vice versa.

Ken Kuturagi in 2001 basically said:
- Computers are connected via the internet but can't share computing resources
- This is not due to a lack of bandwidth; even if everyone had fibre, things still won't work due to topology
- Cell will somehow change this ("drastically accelerate the speed of networking")

It's 2005. We have all the info about the Cell chip. But in no way do I see how this chip can accelerate networking. And I can't see how it changes the network 'topology'.

Am I missing something?

50% Hype and 50% BS, it won't be the last from Ken. :D


yeah. because now with PS4 going into the planning and early R&D phase, Ken is going to have to say something like, "we will achieving more processing and intelligence than the human mind"

thanks for posting that article. I love the part about the "sythesizing human emotions" with ps2. I remember reading that in PSM mag back in 99 with screens of CG. I rememeber an article also talking about how the ps2 was gonna revolutionize the movie industry because it could render realtime CGi level gfx for so cheap. BAck then I actually believed it till I played the launch games which didn't hold a candle to my Geforce2 at the time.
 
Pozer said:
thanks for posting that article. I love the part about the "sythesizing human emotions" with ps2. I remember reading that in PSM mag back in 99 with screens of CG. I rememeber an article also talking about how the ps2 was gonna revolutionize the movie industry because it could render realtime CGi level gfx for so cheap. BAck then I actually believed it till I played the launch games which didn't hold a candle to my Geforce2 at the time.

LOL, I remember that either Newsweek or Time (both had PS2 cover stories) article mentioned that a single PS2 could render Phantom Menace special effects in real time. They even had photo of a render farm room full of mainframes and PS2 with caption along the lines of: "Old way: thousands of processors require up to 24 hours to render a single frame. Sony Revolution: New Playstations can perform the operations on the fly".

It's hard to believe how insane the PS2 hype got at one point.
 
Back
Top