Xbox 2 CPU taped out

http://www.teamxbox.com/news.php?id=5327

It will be using a 65-nanometer process for the circuits. That’s some impresive tech IBM is using. Can’t wait to see what kind of video effects the next gen systems will be capable of.

But sometimes, it just could be as straightforward as Jodie Foster claims in the movie Contact, “The simplest hypothesis is most likely to be true.”

Yeah, or you know, it could be “Occam’s Razor”, first proposed by William of Occam sometime in the 14th century, by stating “Pluralitas non est ponenda sine neccesitate” or “plurality should not be posited without necessity.” I cannot believe dude quotes Jodie Foster. I know not everyone can be up to date on relatively obscure philosophy, but going to a movie quote for profundity is inane.

Sorry to go off topic. 65 nm processors. HOORAY! or something.

Ockham. Where the fuck is Occam?

I’ve seen it spelt both ways. I think either is acceptable, leastaways when it’s referring to the philosopher.

Spelt? Spelled?

What exactly does that mean for the Xbox 2? Does mean faster or cheaper to make or a combination of both?

It means you’re going to need to live in a fridge to be able to turn the Xbox 2 on.

Nah, just install it on an exterior wall and connect it to an air-conditioner. ;-)

It means you’re going to need to live in a fridge to be able to turn the Xbox 2 on.[/quote]

Huh? Since when do smaller traces mean massive heat?

It’s quite simple. Smaller traces mean that there’s less space to fit the elves. The remaining elves have to run that much faster in order to compensate for their now superfluous peers, generating vast quantities of heat. Thus, an entirely new paradigm in cooling solutions will be required. Traditional heatsink/fan assemblies will give way to arrays of tiny blizzard generators. The problem here, as I’m sure you’re aware, is that current technology limits the number of tiny blizzards that can be coordinated simultaneously, resulting in uneven coverage and the risk of elf attrition. Ultimately, this means that we’re unlikely to see Xbox 2 before 2014 at the earliest.

Hope that helps.

It’s quite simple. Smaller traces mean that there’s less space to fit the elves. The remaining elves have to run that much faster in order to compensate for their now superfluous peers, generating vast quantities of heat. Thus, an entirely new paradigm in cooling solutions will be required. Traditional heatsink/fan assemblies will give way to arrays of tiny blizzard generators. The problem here, as I’m sure you’re aware, is that current technology limits the number of tiny blizzards that can be coordinated simultaneously, resulting in uneven coverage and the risk of elf attrition. Ultimately, this means that we’re unlikely to see Xbox 2 before 2014 at the earliest.

Hope that helps.[/quote]

Using your analogy. I thought because the traces were smaller there was less resistance to elf movement. Less resistance = less heat.

Sorry, Sam, you’re wrong.

Smaller traces have less resistance, which means there would be less energy given off as heat. However, the smaller processes in use today allow for a lot more transistors to be packed onto the chip, which means that there’s more stuff there to generate heat.

Could I get a coffee? Latte, espresso, double decaf whatever, I’m not fussed.

Sorry, Sam, you’re wrong.

Smaller traces have less resistance, which means there would be less energy given off as heat. However, the smaller processes in use today allow for a lot more transistors to be packed onto the chip, which means that there’s more stuff there to generate heat.[/quote]

I guess it’s a trade-off. Usually the next step in trace size has meant lower power consumption.

Its all about tradeoffs. The other problem is that there have been quite a few problems going to smaller smaller die sizes and traces.

A 65-nm process is quite ambitious.

A story from the year 2000: http://www.ebnews.com/story/chipwire/OEG20001222S0013

As a result, the 2000 update lists three scenarios. Scenario 1, the least aggressive, envisions the 100-nm node occurring in 2005. Scenario 1.5 sees the 100-nm node arriving in 2004. The most aggressive scenario, 2.0, sees the 100-nm node’s arrival in 2003 and the 90-nm node in 2004.

Scenario 2.0 represents the thinking of some U.S. and European companies. If their forecasts prove accurate, the 65-nm, 45-nm, and 33-nm, and 23-nm nodes would be realized in 2007, 2010, 2013 and 2016, respectively.

A source is saying that we’re going to have them in 2005? I’ll chat with my brother, who works for Applied Materials, and ask him about the processes in use.

edit: And various recent articles about the current ITRS roadmap say the same thing, so I’m wondering whats going on. Is it a subtlety in the process or a bad source, or what?

edit2: Something’s not right here.

http://www.eetimes.com/story/OEG20040123S0041

ITRS said a week or so ago that 65nm process won’t be in major manufacturing (defined as more than 2 manufacturers producing more than 10,000 chips a month) until 2007, and they’ve been accurate so far.

edit3: Well, company PR from Intel and AMD/IBM all claim they’ll have 65nm processes up and going by 2005, but well… that’s PR. I know Intel has had production problems with 90nm.

So yeah, I’m sure they’d like to be able to do that, but don’t believe the hype.

Both Sony and Microsoft have said 65nm. I think we already discussed it briefly in another thread because this is old news.

I think that’s optimistic at best and suicidal at worst.

–Dave

IBM thinks they can get 65nm. They have a bunch of SOI related patents Intel wishes they had, so what Intel is struggling with at the moment is not necessarily an indication of what IBM will be able to do by the end of 2005.

Wouldn’t have been so bad if he had quoted from the book ‘Contact’ even, although it’s been a while and I can’t say hand on heart that that line is in there. It’s the kind of thing Sagan would make the character say though.

Whatever they’ve taped out is certainly not the CPU that will be used in the Xbox 2. And all the dumbass news sites that have repeated that headline are run by nincompoops. IBM has some experimental 65 nanometer silicon, but they don’t have an Xbox 2 CPU that far into production.

The past particple of “spelt” tends to be used only in Britain, whereas “spelled” is mostly American. This is the same for “learnt” and “learned”, “dreamt” and “dreamed”, etc. We are leaning ever more towards the American spelling, and most British newspapers seem to use the “ed” ending now, but it is by no means universal. I tend to find myself using the different endings depending on the situation, with burned describing recent events and burnt ones further in the past, e.g. “I just burned my toast” and “the toast was burnt”. We also differ in the use of the word while/whilst, and I tend to use the latter. An American once corrected how I spelt that word, whilst attempting to be erudite.

Sorry for the interruption, but I can’t resist being a spelling nerd.

Should be interesting. Sony is still struggling with 90nm:

OTTAWA, January 29, 2004 — Semiconductor Insights (SI), the leader in technical and patent analyses of integrated circuits and structures, today revealed that Sony’s widely publicized 90nm EmotionEngine + Graphics Synthesizer (EE+GS@90nm) found in the PlayStation X (PSX) is still shipping in 130nm technology. “This discovery is contrary to Sony’s announcement about this device”, said Derek Nuhn, Senior Vice President and Chief Operating Officer of Semiconductor Insights. “Manufacturers, including Sony, are under great market pressure to deliver at 90 nm, but the reality is many are not ready.”

http://www.semiconductor.com/resources/press_releases/2004_01_29_1.asp?c=29450