• Hey Guest. Check out your NeoGAF Wrapped 2025 results here!

(Interview) Kutagari talks ALOT about CELL.

This interview is not new,interesting stuff anyway.
I wonder if next year we'll start hearing about Cell 2 and PS4 development :D
 
Talk, talk, talk, talk, talk, and more talk.

Show us some actual fucking games being controlled by someone using a controller! Not conceptual renderings on b-roll.
 
Nah, this was posted a week or two ago, and the "Cell could cause a worldwide system crash if we allow people to access its power" comments were appropriately derided and damage-controlled.
 
PS2, for example, made full use of the semiconductor technology available at the time, but only a software effort on the part of the game developers enabled high-performance titles like Gran Turismo—a realistic car racing game that has sold more than 43 million copies worldwide. The closer they accessed the hardware, the higher the performance they achieved.

Cell is not like that. Application programs can no longer directly access the hardware; instead, they will have to be written in high-level, object-oriented language. That was done for security reasons. If processors of high performance and wide bandwidth like the Cell were linked together without sufficient security, a worldwide system crash could occur with one attack.


OMG CELL=TEH SKYNET!

CinemaquetteTX1.jpg
 
Oh dear lord, same stupid responses.

We don't need another thread about this.

A virus could bring down a network of Cell computers. Just like it could a network of any computers. Cell could be particularly vulnerable given its closeness to the network, however. That's all he's saying.

:megaroll
 
yup reminds me of when my fellow CSCI classmates tried to be funny. you gotta just give them a fake laugh smile so they feel better...

haha good one guys, skynet... awesome you like mixed terminator with CELL for funny. :)
 
Razoric said:
yup reminds me of when my fellow CSCI classmates tried to be funny. you gotta just give them a fake laugh smile so they feel better...

haha good one guys, skynet... awesome you like mixed terminator with CELL for funny. :)

Thanks! :lol
 
gofreak said:
Cell could be particularly vulnerable given its closeness to the network, however. That's all he's saying.
Still doesn't make any sense. What does "closeness to the network" mean, and how is it any different from current gen consoles?
 
Rhindle said:
Still doesn't make any sense. What does "closeness to the network" mean, and how is it any different from current gen consoles?

Part of Cell's design is to share executable code across the network. There's hardware provision for that, apparently. Remember all that early talk about connecting Cell to a network, and getting it to share processing with other Cell devices? Sending out "cells" across the network to execute elsewhere? That's apparently still there in the PS3 OS according to recent Kutaragi/Chatani interviews (although they say you'd need a 100Mbps connection to make this feasible over the internet..but that might be reasonably available during Cell's lifetime, hence the need to factor this kind of thing in. Part of the reason for the gigabit port is to apparently let you do this locally, however). If that were the case, and a virus could exploit that, it could more easily accomodate the spread of malicious code. Of course, all systems have to be concerned about virii..the risk is too strong to ignore even without the above.
 
TheDiave said:
Talk, talk, talk, talk, talk, and more talk.

Show us some actual fucking games being controlled by someone using a controller! Not conceptual renderings on b-roll.
that goes for every developer showing "next-gen" vids...
 
EET: When did you first get the idea for the Cell processor?


Ken Kutaragi: It was immediately after we completed the development of the PS2 in 2000. After we presented the papers on the Emotion Engine processor at the International Solid-State Circuits Conference and the main work proceeded to the building of the PS2 console, I started thinking about what would be next.

to clarify, this actually happened in 1999, not 2000.
there must have been a mistake or typo along the way.
 
gofreak said:
Part of Cell's design is to share executable code across the network. There's hardware provision for that, apparently. Remember all that early talk about connecting Cell to a network, and getting it to share processing with other Cell devices? Sending out "cells" across the network to execute elsewhere? That's apparently still there in the PS3 OS according to recent Kutaragi/Chatani interviews (although they say you'd need a 100Mbps connection to make this feasible over the internet..but that might be reasonably available during Cell's lifetime, hence the need to factor this kind of thing in. Part of the reason for the gigabit port is to apparently let you do this locally, however). If that were the case, and a virus could exploit that, it could more easily accomodate the spread of malicious code. Of course, all systems have to be concerned about virii..the risk is too strong to ignore even without the above.
That's what I thought you were going to say. Given that the whole distributed computing supercomputing thing was a harebrained idea to begin with, if retaining those features (whatever they are) also introduces an additional security risk, it would seem that they should just be eliminated.

It seems like eliminating the ability to do any assembly level coding could be a major constraint in the long term. Perhaps not in the first year or two, but certainly as the platform ages. If you are able to "code to the metal" on the 360 but not on the PS3, then it may be 360 games that pull away from PS3 in the long-term and not vice-versa.
 
"It may be regarded as a game application for the time being, but I want to realize the day when "computer entertainment" would mean all such entertainment applications, including games."

Hopefully the "it's not a gaming machine" fanboys will shut up now.
 
I said this before in a similar thread in the past, but damnit, Kutagari just can't stop talking, even when he has nothing new to say, he still talks.
 
Ark-AMN said:
I said this before in a similar thread in the past, but damnit, Kutagari just can't stop talking, even when he has nothing new to say, he still talks.

He stayed quite for 5 years when cell was in development, now that PS3 nears it's inevitable release Kuturagi will talk, talk and talk until he starts working on Cell 2 or whatever crazy processor design he might have in his mind.

If you don't like what he says, why not simply ignore the thread?
 
KK for president. :D

I honestly can't get enough of this guy, but this is an old interview. Don't hate, appreciate. BTW, the distributed computing concept is not "harebrained". It's just lacking the external bandwidth to make it really practical right now. But the idea is very Skynet-ish in concept, and you gotta admit that's very fucking cool. I mean, aggregate computing power of 100M Playstations would make for the ultimate scalable game systems. The last console you'd ever need to buy. :D But we are at least a decade away from that kind of bandwidth. OK, make it like 20 years away. But I like a big dreamer, and it seems STI has laid the groundwork for our future enslavement by the machines. Oooh, I can't wait. :lol PEACE.
 
I don't think he's referring to any constraints or functionality that we haven't known for some time now - I think he is mostly just referring to the protected memory space of each SPU in the Cell. In the PS2 you could write data around to the VUs pretty much as you please, without any regard for security. If there was malicious code running there it would be easy to interfere with the normal operation of any given part of the program. In the Cell, on the other hand, there are restrictions as who can write to and from the local storage of each SPU and when. That may be irrelevant for most console games, which don't have to worry much about unathorized code - but it doesn't impose alot of restrictions either, as long as you are sticking to the Cell's model of concurrency, instead of hacking around memory spaces like you could in the PS2.

For a hypothetical OS running on the Cell, though, it could be a pretty significant feature, especially when you consider that Apple's DRM used in iTunes was supposedly cracked by writing out the memory space of the decrypting application. The same techniques used today to crack DRM (which Apple or Sony probably care alot about but consumers don't) could be used by malware to ship personal passwords or movies across the Internet (and presumably in the future there will be users as comfortable storing movies and photos online as storing email messages on Yahoo or Gmail.) You rely on the operating system to keep the code running in a trusted and protected fashion, but that's a difficult thing to ensure.

In the Cell, each SPU can probably run as a black box without offering the ability for other code to interrupt it or spy on it - presumably even if there are buffer exploits or other security holes in the code running in the SPU. There could be code that decrypts content (say a Blu-Ray DVD, or a movie of your child's birth that you are storing encrypted online for your grandparents to watch) and it would run in a protected mode in a SPU, using whatever key is tied to your computer that authorizes you to run it, without writing back to system memory or leaving a trace other programs could use to disassemble it.

Basically I think when he refers to a "system crash" he's not referring to a DoS attack on the infrastructure itself - I think he's referring to a loss of trust in the privacy or security of internet services (and computers connected to the internet) by users or content providers. The Cell is nowhere near a be-all or end-all solution to that, but it has some intriguing features and new exploits that might need to be found.
 
Top Bottom