More On IBM's Next-Gen Xbox Chipset Win

Pieroxy writes "EE Times reports further details on Microsoft's use of IBM chips in its next generation Xbox game and consumer electronics devices, dealing a blow to Intel and providing a much needed boost for IBM's lossmaking chip business." An analyst claims that "IBM is likely to modify its most advanced G5 PowerPC silicon, which is being used in Apple Computer's fastest Macintosh desktops, for the embedded market, reducing the cache and cutting power consumption", and further comments: "This is likely to heat things up at Intel, but it is competition that is healthy for the industry. It's ironic that IBM, with its roots in the computer industry, doesn't supply the processors for the main portion of the personal computer industry. Intel does." We covered IBM's initial announcement as a section-specific story earlier today.

Comments

Whose roots where?

[csnydermvpsoft]

It's ironic that IBM, with its roots in the computer industry, doesn't supply the processors for the main portion of the personal computer industry. Intel does.

Wouldn't it be more like the "computer industry, with its roots in IBM," not the other way around? Though that's not entirely accurate either - maybe if it was changed to be the personal computer industry.

Re:Whose roots where?

[Xeriar]

Wouldn't it be more like the "computer industry, with its roots in IBM," not the other way around? Though that's not entirely accurate either - maybe if it was changed to be the personal computer industry.

And the mainframe industry, and the software industry, and...

Other companies were involved, sure, but they invented FORTRAN and built at least half of the computing industry as we know it, both from their successes and their failures.

Re:Whose roots where?

[t0ny]

Considering Intel developed the first microprocessor, you could swing the statement around toward them as well. Perhaps its best if we just stick to the present.

Personally, this isnt too much of a surprise. Consoles generally use a chip specialized for their specific needs (much like the architecture of the XBox was specialized for it, rather than being a straight PC). Intel is most likely too big to do something that is both relatively (to Intel) small scale and essentially a single-purpose chip.

IBM, on the other hand, is hungrier than Intel, and is also the only company set up with the necessary chip fabrication resources to create a high volume of very specialized chips.

Honesly, the only other way this could work out is if the XBox2 were going to stick with the Pentium line; since the XBox2 is most likely going to aim much higher than the XBox (maybe with integrated PVR and expanded multimedia features), they seem to have outgrown their original product design.

Oh well, maybe Intel can nab that lucrative contract selling chips for Infinium Lab's "Phantom"

Re:Whose roots where?

[kfg]

"Considering Intel developed the first microprocessor, you could swing the statement around toward them as well."

Never trust a computer you can't lift. - Stan Mazor - Intel

And out of that simple philosophy. . .

KFG

Re:Whose roots where?

[homer_ca]

The overall architecture of consoles is specialized for gaming, but the CPUs themselves have definitely been standard commodity parts. Sony is the exception here, but for the rest they've used processors like the 6502, 68000, SH-4 (also used in some CE PDAs), and Powerpc. Still, the CPU is only a fraction of the processing power in a game console compared to the graphics and sound. You could say the same about a PC with a high end graphics card.

What about today's Xbox?

[O]

Are they not going to have backward compatibility? That seems like a big mistake in the game console market to me.

Re:What about today's Xbox?

[pi eater]

I doubt it.. Look at nintendo, none of their consoles are backwards compatible.. Although it is a point you could argue, I doubt that this had a big impact on their popularity.. I blame competition from sony and ms

funny geeky shirt

Re:What about today's Xbox?

[chill]

Are they not going to have backward compatibility? That seems like a big mistake in the game console market to me.

By the time the X2 comes out, Celeron 733s will probably be cheap enough to put on an add-on card and sell for $49.

That is the price of 1 game. Lot of people with existing X-Box titles would buy that.

Re:What about today's Xbox?

[no reason to be here]

The sales numbers for the Xbox and Game Cube are fairly comparable; however, the PS2, which was backwards compatible with the PS1, blows them both out of the water. Part of the reason that Nintendo has done so well in the portable arena is because the newer iterations of the Gameboy were backwards compatible with the older versions.

Re:What about today's Xbox?

[vanillacoke]

Well to be honest the gamecube is the first time Nintendo is using optical media, so i think they may finally have backwards compatibility.

But it's all speculation.

And suicide if they don't.

Re:What about today's Xbox?

[jo42]

Microsoft wants you to throw out all of your old stuff and buy new stuff. This is, after all, their business model. Don't use Windows 2000 with Office 2000 anymore! Throw it out and buy Windows XP and Office XP instead. Then throw all that out and buy Office 2003 and run it on, oh wait, we have no Windows 2003 for the desktop - doh! Just missed an opportunity to sell a new round of OS upgrades - cr*p!

Re:What about today's Xbox?

[yamla]

You have shown a correlation. Unfortunately, you have not shown causation. Apparently, the market research shows that backwards compatibility is only important at launch and then, only if you don't have a solid set of high-quality games available for that platform. Of course, the XBox didn't have this stable of games (and really only has KotOR now) upon release. I don't recall what the PS2 had. If it had GT3, that was almost enough to carry the platform there and then even without PS1 compatibility.

The question really comes down to this. How many people do you know who own a PS2 and play PS1 games on it? Not a single person I know does this. That's not to say that _you_ might not know people who do, only that this does not represent the average user. Most people who buy a PS2 only want to play PS2 games on it. They are unlikely to buy a PS1 game. They may already own some PS1 games but if so, they could just use their PS1 to play them on.

Re:What about today's Xbox?

[cfuse]

Are they not going to have backward compatibility? That seems like a big mistake in the game console market to me.

Well, since Microsoft owns Connectix (who make virtual PC), they can just run x86 via emulation. And according to the apple crowd, it runs faster on a G4/G5 than the real x86 hardware anyway.

I look forward to Windows 2005 running on a G5 emulating x86.

Re:What about today's Xbox?

[hotchai]

They can still maintain backwards compatibility if they think the market will demand it.

What prevents them from bolting on the current XBox processor (some P3 derivative IIRC) alongside the new IBM chip and having the hardware funnel the instructions to the appropriate processor depending on the version of the game being played? So an XBox1 game will run on the old processor whereas an XBox2 game can run on the new one - everything else can be shared (memory etc.) Again all this will depend on results from their "market surveys".

Re:What about today's Xbox?

[AuMatar]

I do. I own about 6 PS1 games to every PS2 game. Never owned a PS1, I was in college when it came out, had no money. So being able to do both was great, I could play all the games I missed.

In addition, if I had had a PS1, I'd still find it a great feature since it would let me get rid of a piece of hardware and remove clutter.

Re:What about today's Xbox?

[LilMikey]

oh wait, we have no Windows 2003 for the desktop - doh! Just missed an opportunity to sell a new round of OS upgrades - cr*p!

That's why Longhorn is initially coming out for the desktop. Server will follow in a couple years when 2003 starts to fall off of support.

Re:What about today's Xbox?

[yamla]

I'm sure, then, that backwards compatibility is very important to you. Take a look around other owners of PS2s, though, and you'll find this just doesn't hold for the vast majority of them.

What _is_ important is how many (and how many good) games are available at launch. Leave out backward compatibility _and_ ship without many good games and you are asking for a world of hurt.

Re:What about today's Xbox?

[startled]

"The sales numbers for the Xbox and Game Cube are fairly comparable; however, the PS2, which was backwards compatible with the PS1, blows them both out of the water."

I disagree-- I'm pretty sure it's the blue LED that's responsible for the PS2's popularity.

Re:What about today's Xbox?

[Osty]

Of course, the XBox didn't have this stable of games (and really only has KotOR now) upon release. I don't recall what the PS2 had. If it had GT3, that was almost enough to carry the platform there and then even without PS1 compatibility.

What? It's generally accepted that XBox has had the strongest launch lineup of any recent console (perhaps of any console?). With Halo, Project: Gotham Racing, Amped, NFL Fever, Oddworld, DoA3, and a bunch of other titles, the XBox did quite well for itself. The PS2 on the other hand had a pretty weak launch library, so weak in fact that I can't even recall what titles launched with the system (probably a Madden title, maybe the first SSX?). GT3 was released a year or more after the launch of the PS2, and was eventually bundled with the PS2 for a while, but it was definitely not a system seller for the PS2 launch except by anticipation, ie "Why should I buy a Dreamcast with a bunch of great games right now when I could buy a PS2 and get Gran Turismo 3 in a year even though there's really nothing worth playing yet?"

The question really comes down to this. How many people do you know who own a PS2 and play PS1 games on it? Not a single person I know does this. That's not to say that _you_ might not know people who do, only that this does not represent the average user. Most people who buy a PS2 only want to play PS2 games on it. They are unlikely to buy a PS1 game. They may already own some PS1 games but if so, they could just use their PS1 to play them on.

I know exactly two people who have a PS2 (and thus, I'm probably not typical, but oh well), and both of them use their PS2 for PS1 games. Mostly, they play Square's PS1 games like the Final Fantasy series and re-releases (Anthologies, Chronicles, Origins). Nobody is going to buy a PS2 so that they can play the PS1 version of John Madden Football, but I could see someone buying a PS2 because they want to play Gran Turismo 2 (they'll probably also get GT3, but the hardcore will dig up copies of 1 and 2 as well). And as for just playing the PS1 games on the old PS1, why bother? Do you really want to have another box connected to your TV/home theater system? More importantly, the PS2 does smooth the graphics somewhat of PS1 games, making them look a bit better, so except for the very few games that don't work with the PS2's PSOne-on-a-chip, there's no reason you wouldn't want to play your PS1 games on your PS2.

Re:What about today's Xbox?

[orthancstone]

Presuming IBM makes a new chip, I'd hope they could come up with some emulator that Intel would agree to (c'mon MS, you can pay Intel off...you've got the money).

Re:What about today's Xbox?

[nelsonal]

While I agree with you today, at launch I think backwards compatability and DVD player features, sold more consoles than the game stable at the time, and really accellerated this cycle. MS would be foolish not to adopt this since they aren't starting from a very strong competitive position in the next cycle.

Re:What about today's Xbox?

[vanillacoke]

Sorry, forgot about the GBA. DOH.

Thanks for being an asshat tho.

Re:What about today's Xbox?

[Tychoma]

It runs faster on a G5? Interesting.

Re:What about today's Xbox?

[blincoln]

Take a look around other owners of PS2s, though, and you'll find this just doesn't hold for the vast majority of them.

Like the parent's author, I never had a PS1, just a PS2. I have twelve PS2 games, and fifteen or sixteen PS1 games. That doesn't include the box full of demo discs for both platforms.

My friends with PS2s also use them for playing PS1 games. There are a lot of them out there that are still definitely worth playing (e.g. Soul Reaver), that look really good with the texture smoothing enabled.

Re:What about today's Xbox?

[Johnny Mnemonic]

I think you're mistaking perception with behavior. It's not really how people actually use the console; it's how they think they're going to use the console. The ability to play PS1 games meant something to me when I went to purchase a PS2; I recognized that that meant that a lot of games would be immediately available for it (cheaply, on eBay). However, I have only actually purchased one PS1 game to play.

If only one of PS3 and Xbox2 are backwards compatible, that will factor in on my purchasing decision; that's something that I can control. Whether or not I actually ever wind up using it that way is irrelevant. Although, that said, Sony would be the greater fool to sacrifice their customer base, inasumch as it is much larger; Sony has much more to lose from starting over at the starting line than does MSFT.

Re:What about today's Xbox?

[E_elven]

*cough* ngage *cough*

Re:What about today's Xbox?

[Dun Malg]

By the time the X2 comes out, Celeron 733s will probably be cheap enough to put on an add-on card and sell for $49.

Or perhaps they'll be able to emulate it even .

Re:What about today's Xbox?

[Daetrin]

I still play PSX games on my PS2, although not as much as i did when the PS2 first came out. I've got four friends with PS2s who i discuss games with on a regualr basis, and all four of them have used the PS2 to play PSX games on them within the last year at the very least (two of them i know have played PSX games on their PS2s within the last month)

Back when the PS2 first came out most of the PS2 games were crap, and everyone i knew was talking about either the backwards compatibility, the DVD player, or both. And just about everyone got SSX since it was the _only_ PS2 game everyone could agree was actually worth the money, but unlike the bitching you see after similar launches for other systems, most people were of the opinion "oh well, at least there are a lot of good PSX games i can play on it."

Re:What about today's Xbox?

[yamla]

This is a great point. I don't know if the market research was looking at perception or behavior. Slashdot, of course, is a bad place to look for either, we are all geeks. But yes, when Microsoft releases the XBox 2, the perception that it cannot play XBox 1 games may be far more important than the (possible) fact that most people wouldn't actually do so.

Marketing, of course, is all about perception.

Re:What about today's Xbox?

[yamla]

Halo is apparently an A1 game (I wouldn't know, I never played it). DoA3 was fine, I bought that when the price came down. I'm not convinced I'd list it as a must-get game, though, and I certainly wouldn't list any of the other games as must-buy. I'll likely get Amped 2 when it's released, never thought enough of Amped 1 to buy it though. That said, I'm a slashdot user and therefore not typical.

While I may disagree with the quality of the XBox games upon launch, I clearly wasn't remembering the launch of the PS2 correctly. GT3 was almost good enough to make me buy a PS2 but if it wasn't available at launch, I am sure you are right about their library problems at the beginning.

Now, what I thought the XBox did really badly was to not come with DVD playback. And their DVD adapter was _expensive_ (and mine wasn't very good). The PS2 could play DVDs out of the box, no? I'm not quite sure how Microsoft justified that, coming so long after the PS2 was released.

Re:What about today's Xbox?

[Osty]

While I may disagree with the quality of the XBox games upon launch, I clearly wasn't remembering the launch of the PS2 correctly. GT3 was almost good enough to make me buy a PS2 but if it wasn't available at launch, I am sure you are right about their library problems at the beginning.

Launch titles don't have to be "must have" titles a year or two after launch :) In fact, they don't even have to be "must have" titles at launch, they just need to be good. Whether or not DoA3 is a must-buy game (I feel it is, if only as a way to show off what the XBox can do, even two years into its life) doesn't matter, because it's still a good game (IMHO, of course). Compared to launch libraries of previous consoles, the XBox did damned well (also evidenced by the XBox's high attach rate from nearly day one).

Now, what I thought the XBox did really badly was to not come with DVD playback. And their DVD adapter was _expensive_ (and mine wasn't very good). The PS2 could play DVDs out of the box, no? I'm not quite sure how Microsoft justified that, coming so long after the PS2 was released.

I'm definitely not an authority here, but the story I've heard is that it was basically a monetary concern. Microsoft was already taking a loss on the XBox hardware, without factoring the cost of an MPEG2 license on every box. To at least salvage something from the situation, they chose to sell the DVD playback kit separately, and roll the MPEG2 license charge into that (I for one am glad they did this -- I don't give a crap about DVD playback on my game consoles, because my stand-alone DVD player is ten times better than any console could be). And if that doesn't do it for you, how about justifying it this way: the PS2 comes with DVD playback in the box, but no remote control. A good remote control costs around $30. The XBox doesn't come wiht DVD playback in the box, but if you spend $30 on the DVD playback kit you get a remote control. In the end, for the same price, you have either a PS2 with DVD playback and remote, or XBox with DVD playback and remote.

Re:What about today's Xbox?

[akuma(x86)]

By the time the X2 comes out, Celeron 733s will probably be cheap enough to put on an add-on card and sell for $49.

Except that Intel doesn't make Celeron 733s anymore.

Re:What about today's Xbox?

[Lars T.]

Then buy a 1 GHz one and underclock it.

Re:What about today's Xbox?

[WoTG]

Or failing that, almost any modern x86 will do. The beauty of a standard ISA, eh?

Re:What about today's Xbox?

[Lars T.]

Well, there could be problems with instruction timing. The beauty of coders depending on specific hardware. Who remembers the problems with >350 MHz K6-2s and Windows? - all because AMD made one instruction faster than Intel. I'm sure most Xbox games would act weird if run on a P4 core chip.

Re:What about today's Xbox?

[rifter]

"oh wait, we have no Windows 2003 for the desktop - doh! Just missed an opportunity to sell a new round of OS upgrades - cr*p!"

That's why Longhorn is initially coming out for the desktop. Server will follow in a couple years when 2003 starts to fall off of support.

Besides, Windows Server 2003 is the server OS of the same generation as XP. In other words, they already did the same thning as longhorn in the XP space.

Re:What about today's Xbox?

[PainKilleR-CE]

GT3 was almost good enough to make me buy a PS2 but if it wasn't available at launch, I am sure you are right about their library problems at the beginning.

If GT3 wasn't available at the US launch, it was available shortly afterwards (within about 6 months), which for many people was essentially the same thing, since the PS2s themselves weren't available at launch.

On the other hand, GT3 was the first DVD title for the PS2 (many PS2 titles were still on CDs), and caused major problems in Japan and on some US consoles (which couldn't read DVDs properly).

As for whether or not people actually use the backwards compatibility, I won't buy a system anywhere near launch without it any more, and I still play PS1 games nearly as much as PS2 games, and sold my PS1 when I got the PS2 (so I could buy more PS2 games). With Final Fantasy Origins having just come out in the last 6 months or so, I hardly see myself stopping with the PS1 games any time soon. The added bonus was that places like Wal-Mart really dropped the prices on PS1 games at the PS2's release, which meant there were plenty of good titles available for a few bucks each.

Re:What about today's Xbox?

[leifm]

A lot of what I've read about the PS3 seems to make it sound destined to be a high profile disaster (same goes for the PSP), I expect Microsoft to take the lead in the next generation, and Nintendo to either go software only or be assimilated by that other Redmond company. Sony is headed for a fall, they are getting to cocky and to ambitious.

Re:What about today's Xbox?

[big_a]

Seems like an odd choice considering Microsoft will have to port all their tools and API's (i.e. Direct X) to a different architecture. As I understand it, part of appeal of the Xbox to game developers was ease of portability from the desktop PC environment. This would seem to be a step in the other direction. (Not to mention the backward compatibility issue.)

Re:What about today's Xbox?

[GooberToo]

My son and his friends, as do many other people at his school, all seem to play PS1 and PS2 games on their PS2s. My son is a freshman in highschool.

Seems many people find a game or two that they just love that isn't replaced with a newer version. For example, some of the PS1 Amored Core and FF games, do not have a PS2 replacement. Function for function that is. As such, people that spend a lot of time with their beloved games, don't want to lose those games just because they upgraded their console.

In short, the problem is clearly worse than you think. People will miss backwards compatibility.

Re:What about today's Xbox?

[forlornhope]

I dont think backwards compatibility will be the problem everyone thinks it is. Doesnt apple or someone already have a windows emulator for the mac? Even if they dont, youve got a ppc970 core probably running at 2GHz, and a plenty beefy graphics card. So I doubt that microsoft will have a problem producing an emulator. Also what about PS2's PS1 on a chip. Microsoft could put a XBOX1 on a chip in the XBOX2. I seriously doubt that Microsoft will be stupid enough to not include backwards compatibility when it would be this simple to put it in.

Re:What about today's Xbox?

[KirkH]

I'll likely get Amped 2 when it's released, never thought enough of Amped 1 to buy it though.

It's been released. Pretty sweet game too, as long as you don't mind a challange.

Re:What about today's Xbox?

[poot_rootbeer]

By the time the X2 comes out, Celeron 733s will probably be cheap enough to put on an add-on card and sell for $49.

Add-on processor cards for home/desktop computers have a tenuous reputation for hardware compatibility, though. Ask any Sun workstation user with a SunPCI card, or any G3 Mac user with the G4 upgrade card.

Re:What about today's Xbox?

[chill]

I didn't know that. My experience was with Amiga computers (A2000, A3000, A1200, A4000) and accelerators and they always worked fine.

Unless you went tried to integrate a different architecture altogether like adding PPC to an MC68K Amiga. But things like that add-in 80286 board (full PC on a card) worked great for me.

Re:What about today's Xbox?

[Lars T.]

Not Windows Visual Basic programmers, Windows device driver programmers from MS. They programmed a timing loop depending on a certain x86-instruction, which AMD had made several times faster on the K6-2 than on Pentiums. At 350+MHz, they got time of zero for the loop - bada-bing, division by zero.

Re:What about today's Xbox?

[wolrahnaes]

Whether or not DoA3 is a must-buy game...if only as a way to show off what the XBox can do

YES. This is exactly right! The Xbox is great for correct jiggle-physics simulation.

Seriously though, after I got done staring at the screen, I started using the game to demo the Xbox. Every one of my friends bought an Xbox after seeing DoA and Halo.

it's still a good game

Huh? There's a game in there too? ;)

About the DVD, here's my idea:

Integrate the reciever. Make it software-programmable so any universal remote with DVD functions can control it. Sell a MS branded remote to those who want it.
If you don't know, the Xbox can be controlled by any new DVD controlling RCA remote (you know, the ones that look like the Xbox remote). All three of my satellite remotes and my friend's universal (all RCA) can control my Xbox. This, by the way, makes for interesting battles over control during movies because all 5 people in the room have compatible remotes.
If MS just integrates the current reciever in to the Xbox2 (or even some redesigned X1), it would make alot of people with universal remotes happy.

A question

[FreeBSD Goddess]

Forgive me if I'm being moronic about this, but if it's based off the G5, it has a completely different instruction set. Does this mean that the modified Windows 2000 kernel used with the current XBox will be upgraded and ported to G5, or that we might see a completely different and new kernel?

Re:A question

[jonbryce]

NT has run on PPC chips in the past, so it won't be to much of a problem to port it.

Re:A question

[Ed Avis]

Perhaps Microsoft already has a Win2k kernel running on the G5. NT4 was available for PowerPC, that support was dropped, but it wouldn't surprise me if random PowerPC builds of bits of NT/2k were still being produced inside Microsoft.

Re:A question

[Anonymous Coward]

didn't some guy at microsoft get fired for posting a picture of lots of G5's bought by microsoft?

Re:A question

[nelsonal]

There was a test build of Win2k for Alpha, that I've seen (had a friend with an Alpha), but never heard of any PPC builds. NT was built with many architectures in mind, with the goal of being run as the workstation OS regardless of architecture, so I would think that what's needed to run an XBox would be pretty portable.

Re:A question

[NanoGator]

"Does this mean that the modified Windows 2000 kernel used with the current XBox will be upgraded and ported to G5, or that we might see a completely different and new kernel? "

Probably a little bit of both. First a recompile followed by some hand-tweaking in the slower areas of the code. Microsoft has operated with other processors before, so I doubt this'd catch their dev teams completely off-guard.

Re:A question

[His name cannot be s]

Apparently, MS is considering using it's recently--acquired Connectix stuff to do emulation of the X86 instruction set.

This should be quite interesting. Given that MS can change the specs of the box at will, I wonder how long it will take for the Xbox Linux hackers to keep up.

Re:A question

[AKAImBatman]

> didn't some guy at microsoft get fired for posting a
> picture of lots of G5's bought by microsoft?

Yes. However, you realize that Microsoft needs those machines to test the Mac software they MAKE. Right? You know, like Microsoft Office X.

Re:A question

[calica]

I also seem to remember a PowerPC port for NT. Either 3.5 or 5.

Re:A question

[spongman]

ms had a ppc version of nt back in the day. they never released it though, and i believe it was supposed to be a secret, but i played on an old ibm chirp box running nt when i was there...

Re:A question

[jonbryce]

http://www.google.com/search?q=nt%20power%20pc

The first link should do.

At one point it was available for i386, alpha, ppc and mips. They dropped mips, cause nobody bought it, then IBM/Motorola pulled out of the ppc project for the same reason.

Alpha lasted a bit longer but was withdrawn as of Windows 2000.

More recently it has been ported to ia64, and they are working on a port to AMD's 64bit chip.

Re:A question

[grotgrot]

In fact Microsoft developers on the OS side do a lot of their development on non-Intel platforms such as PowerPC. The reason is that it ensures the code is cleaner and architecture neutral. Later on it is "ported" to Intel.

Re:A question

[Anonymous Coward]

As of NT 4.0, there were still PPC-specific files on the NT install disk. Oddly, one of the primary design goals of NT was portability, and Alpha, PPC and something I forget were supported throughout its early development.

Re:A question

[sean23007]

Proof. But, like the link says, support was dropped in 1996, and it WAS only NT (obviously). But if it's been done before it can be done again.

Re:A question

[the_2nd_coming]

well, the Xbox 2 will have the DRM in the CPU, so it will be hard for Linux hackers to make a mod chip to get Linux on there.

Re:A question

[Rude Turnip]

It was no secret. My NT4 CDROM has binaries for X86, PPC, Mips & Alpha on it. I bought it at the college bookstore in 1996.

Re:A question

[MidnightBrewer]

The NT 4.0 system had a nice little bit called the hardware abstraction layer (HAL) that allowed it to be ported to different processors more-or-less easily.

Re:A question

[qa'lth]

I must be imagining the PPC arch directory on my NT4 Server disc, then...

Re:A question

[laird]

Actually, NT 3.x (and, possibly 4.0) ran on many CPU's including the PowerPC. If you remember back when the PPC was being promoted as the CPU of the Common Hardware Reference Platform, you could run MacOS, Linux, and NT on the same hardware. Anyway, I would hope that MS has retained that portability in the core of the OS (the same way Apple makes sure that MacOS X still runs on the x86) in order to keep its options open.

So it'd be easy for new "Xbox 2" games to run natively on PPC. That being said, they'd still need an x86 emulator to run the Xbox games. I bet a 2 GHz G5 could emulate an Xbox pretty well...

Re:A question

[Loadmaster]

It will be even harder when MS buys Infinium Labs.
In a press conference today CEO of Microsoft discussed their recent acquistion of Infinium Labs. Gates sighted Infinium's outstanding DRM technology stating that Microsoft will implement the Epoxy Encapsulation technology in the XBox2. About this new DRM feature Gates had this to say, "We plan to glue the shit out of everything inside XBox2. I don't know why we didn't think of it before. We don't believe it will cause any unwanted behavior in the console. We'll be sure to use Silicon caulk."

Swi
--
Load Clear

Re:A question

[ADRA]

If its in the die, then how do they 'upgrade' the system when someone invariably cracks the copy protection ala DeCSS?

Re:A question

[javiercero]

... and MIPS!!! Actually MIPS was the native platform of NT up to 3.5, the earlies NT kernels were actually targeted for the 80860, which was being designed by Intel in the mid/late 80s, and whose code name inside intel was N-10. Hence the name NT for the kernel microsoft was designing, those earlier NT systems were 100% research oriented and had nothing to do with the modern Win2K kernel (were more related to VMS).

Re:A question

[drinkypoo]

Most likely, most of the work for PowerPC compatibility must be done in the HAL. Obviously, recent versions of Windows NT are still 64 bit clean, as evinced by the itanic and x86-64 ports of Windows XP. I would further guess that nearly every part of Windows above the HAL (and portions of the kernel) is written in nicely portable C or C++ code (increasing portions of 2003 are written in dotnet, but what's dotnet written in?) so actually, porting the userspace tools would seem to be a nonissue if they can get a compiler working these days.

Re:A question

[the_2nd_coming]

how do you crack it if it refuses to run the software?

Re:A question

[bhtooefr]

Quick question: Was it possible to run NT4/PPC natively on a Mac?

Re:A question

[kill-1]

Can you provide any sources for that claim?

Re:A question

[ADRA]

If for instance the code was somehow embedded inside the CPU core, then the following holds:

- The Signature that allows the boot loader to run is stored inside the CPU, but it can only run a boot loader that has been signed by microsoft beforehand. Since the key can't be changed, all it guarantees is that the boot loader is secure. Now, you have the kernel, which one would assume to be upgradable. If the kernel is not upgradable, a major bug / exploit found in the kernel could be exploited and never get fixed until Xbox 3.

If the kernel isn't upgradable, you need to start hackimng the userspace apps to kill the lock-out.

If they make the kernel upgradable, they have to be able to sign the kernel from the boot loader. So, you either use the embedded CPU ID to sign the kernel, or they will have the key somewhere else that normal users won't be able to reach. Of course having the decode key only gives you half the prize. You need to find the encoding key in order to encrypt software that actually runs on the Xbox. Once software has been signed, there is no stopping anyone from running anything as long as that piece of software has kernel-level access to the hardware. If you rely on a userspace exploit, you can only perform a limited set of operations, like running arbitrary commands. Get the kernel, and you could burn in a new kernel in its place tricking everything on the higher levels as to who you are.

So, how do we hack the Xbox 2 if we can't run arbitrary code on the xbox to begin with?
Any external access to the system is exploitable through its IO subssytems. The kernel could have a buffer overrun, the game running could leak precious data, who knows. The more functionality left open, the more potential expoits there are to take advantage of.

I haven't bridged the topic of hardware mods which does get more sticky if the encryption is processes in the CPU. I'd have to see when and how the CPU decrypts a signed working set and how the decoded code is run after the decrypt.

Re:A question

[grotgrot]

Having a meeting with several of them at Redmond. Heck some of them even used vi as an editor. It is more than coincidence that the netstat command under Windows is very similar to the one on Unix.

Re:A question

[thoth]

Windows NT 4.0 and earlier (3.51 and 3.1) ran on four cpu's: intel x86, dec alpha, nec mips, power pc.

As a side note, the NT 4.0 CD's included all four architectures on the same CD. The OS was smaller back then ;)

Anyway, mips and power pc were dropped after 4.0. The alpha was supported during early Win2000 builds, but that ended somewhere around RC2 for Win2000 (?? this is from memory). Basically it was there as the 64 bit dev platform until the Itanium was ready to go, at which time you would use the Itanium. So alpha didn't make it out for the Win2000 release.

Re:A question

[jafac]

Prediction:
This will be the one PPC Platform where Microsoft will deliver a version of Virtual PC that works with the G5.

Re:A question

[the_2nd_coming]

right, so Hacking the system to use user land software is "easy", but as for modchips that allow the system to be wide open to anyone is probably going to be near impossible and will most likely only come around at the end of the life of the XB2

Re:A question

[Tychoma]

Question, is the HAL the reason that old games (that require the CD to be in the CDROM drive) why they fail to work on XP? The uber buget version of GTA for example?

Re:A question

[TelcusFreshbreeze]

I thought that the netstat command was similar to Unix because Microsoft used the BSD TCP/IP stack.

Re:A question

[grotgrot]

A lot of the people who worked on the networking were UNIX folks. Basically Microsoft recruited a bunch of UNIX weenies who brought UNIXy stuff with them (either code for BSD things, or following traditional UNIX command line arguments and tool names). That all happened in the early 90's.

Re:A question

[vought]

Was it possible to run NT4/PPC natively on a Mac

No. But it did run on an Apple machine - the Network Server 500 and 700 machines.

This unique and capable hardware wasn't a Macintosh, but did run AIX. They were initially available with 132 and 150MHz PowerPC 604 chips, and while the SMP 200MHz 604 model was killed before intro, many dual processor boards made their way through my group at Apple.

There were two other ROM sets for the Network Server 500 and 700. The Mac OS ROMs never worked very well, and the NT ROMs never worked that I knew of, but I heard whispers that it ran NT/PPC 3.5, same as Motorola's PowerStack series. Neither of these products were ever released or acknowledged by Apple that I know of.

Re:A question

[Slime-dogg]

With that, I'm not surprised that MS bought all them Apple machines.

Oh wait. They actually do Apple software.

Re:A question

[YE]

The Windows NT kernel is very portable - it was initially developed for a (now defunct) Intel RISC chip, used to run on PowerPCs and MIPS RISC CPUs, and has recently been ported to Intel's IA-64 (Itanium) architecture.

I strongly suspect they will mandate development of games using the .NET CLR to allow for easier porting between PC and the next Xbox.

Re:A question

[PainKilleR-CE]

But, like the link says, support was dropped in 1996, and it WAS only NT (obviously). But if it's been done before it can be done again.

The same methods are used for the IA64 port, and will be used for the AMD 64-bit port. The entire OS was originally built on MIPS to re-enforce the portability, since the designers of NT felt that if they started with x86 they may never have been permitted to do the remaining ports.

Re:A question

[Bert64]

Motorola always hoped the PPC would achieve the same widespread support that m68k had, unfortunately for them.. it never did, and support gradually faded away.

Re:A question

[Bert64]

*still* 64bit clean?
windows was never 64bit clean, the alpha ports of nt3/nt4 were hacked to use a 32bit compatibility mode of the alpha cpu, and none of the other supported cpu`s were 64bit at the time..
Around the time of 2000, i believe a 64bit version was being developed on the alpha, and development has since switched to the itanic

Re:A question

[PainKilleR-CE]

I would further guess that nearly every part of Windows above the HAL (and portions of the kernel) is written in nicely portable C or C++ code

A lot of the pre-.Net code from Microsoft is in C++, so I'd imagine that's what it's in. In fact, one person I know had an internship there and said that they were very big on making sure developers knew C++ as conceptually being a completely different language from C (something most schools don't seem to teach).

(increasing portions of 2003 are written in dotnet, but what's dotnet written in?) so actually, porting the userspace tools would seem to be a nonissue if they can get a compiler working these days.

The key with .Net is to get the CLR running on the target platform, then it doesn't matter what your code was written in. Of course, most of it is probably in Managed C++ and C#.

Re:A question

[drsmithy]

[...]those earlier NT systems were 100% research oriented and had nothing to do with the modern Win2K kernel (were more related to VMS).

Eh ? NT was never a "research" OS, it was always going to be a commercial product - and Win2k is as much like VMS as earlier (and later) versions of NT were.

Re:A question

[drsmithy]

No, because Apple never released a CHRP Mac.

Will this affect users?

[Ed Avis]

Time was, when the choice of CPU meant something. If two machines used the same CPU you had a good chance of getting a speedy emulation of one using the other - for example the Mac emulators for the Amiga which were close to 100% compatible. But even though this is a Power-derived processor it doesn't seem likely anyone will be running AIX or Mac OS X on the Xbox2, or the other way round.

Re:Will this affect users?

[nate1138]

No, but what about Linux or xBSD? They both already run on that architecture. The article mentioned that MS was pissed about people turning their heavily subsidized game platform into a PC without buying the games that make it a profitable product. Those folks aren't running windows on it anyhow. Now maybe if it was an "extended" power architecture, something not publicly documented, it could delay the hacks.

Re:Will this affect users?

[motorsabbath]

Which begs the question: with Apple running IBM CPUs *and* playing reasonably well with the open-source community, why isn't IBM pushing Apple desktops at all? Seems to me both could clearly benefit from some help from the IBM marketing machine, given it's size and scope. Apple wants to do it all itself (and is doing a pretty good job) but the benefits of such help are undeniable.

Won't MS have to rewrite everything?

[ajiva]

Uhhh, won't MS have to rewrite the whole OS, directX, etc for the XBox2? It seems that porting Windows to PowerPC may be harder than just getting a new Intel processor in there. Of course if MS does port XP to the G5, maybe the Apple guys can install something besides a real OS :)

Re:Won't MS have to rewrite everything?

[pi eater]

Maybe writing it all from scratch will teach them something about implementing a reliable operating system

funny geeky shirt

Re:Won't MS have to rewrite everything?

[stratjakt]

Yah, MS and IBM are stupid, you better call 'em and set them straight!

Some things to consider:

1) NT ran on PowerPC long ago, porting it wont be hard.

2) Wasn't IBM working on a new mega-chip that can run x86 and PowerPC codes side by side?

3) Perhaps making the Xbox so close to a PC blew up in MSFTs face, and they don't plan to do it again. You can turn your noses at "security through obscurity" all you want, but the Xbox, once initiall cracked, has been blown wide open - it's a no brainer to port PC emulators and apps to it. The GCN or PS2, however, have taken much longer, and are still an obfuscated mess to try and develop custom stuff for.

Re:Won't MS have to rewrite everything?

[titzandkunt]

Maybe, maybe not.

IBM could go the AMD route and surround the RISC core with an Intel personality, doing translation on the fly either to native instructions or some intermediate sorta-kinda p-code.

T&K.

Re:Won't MS have to rewrite everything?

[jtnishi]

There's a definite possibility of that, I think. Just the fact that the processor deals with memory in a big-endian way probably'll create havoc (think about the problems talked about moving VirtualPC to G5).

Since DirectX is supposed to be tied a bit to the hardware, one could predict a rewrite of that being necessary, though perhaps the OS might not need to be (after all, just because the XBox uses Win2K at its core doesn't mean that the XBox2 can't use a custom-written kernel).

Re:Won't MS have to rewrite everything?

[Anonymous Coward]

1) NT ran on PowerPC long ago, porting it wont be hard.


I keep seeing folks post this. Have y'all ever actually dealt with thechnology?

The PowerPC port of Windows NT was abandoned long ago. Since that time the nice microkernel architecture and the clean HAL that made the PPC and Alpha ports practical have both been largely compromised/abandoned because NT performance sucked.

Getting the modern codebase to run on the PPC again is going to be a bitch.

Re:Won't MS have to rewrite everything?

[ADRA]

Lets not even begin to talk about CE, which could run on literally a dozen different architectures!

It isn't that fcking hard. There IS a reason that Linux runs on so many different arch's. Its the effort of supporting each user space and chip maker politics that keeps Windows from being released to Architecture XYZ.

When .NET's CLR rules the earth, instantly every Chip maker will become comodity to microsoft. MS can then vice grip any chip manufacturer to their will since MS now has the keys to switch architectures without affecting existing code bases.

Re:Won't MS have to rewrite everything?

[jtnishi]

That's obviously only one example, and it's the only one that I knew for sure of. Obviously, there are other problems besides that one. It's unlikely that all of the software components that MS used for the XBox (and by extension, MS software in general) are written in any kind of hardware independant way. However, that's certainly something new that I wasn't aware of, so thanks.

Re:Won't MS have to rewrite everything?

[John Whitley]

Recall the assumptions inherent to a game console software platform. Portability is explicitly traded away for performance and implementation simplicity/stability. The cut-down embedded NT kernel the XBox team uses (not "Windows") will certainly need some porting work for an entirely new processor. However, there's almost certainly an embedded NT kernel port to PPC already available. Whether or not this particular IBM PPC chip has a little-endian mode will probably no more than determine the NT group's starting branch & build number to be used as the seed cut for XBox2 development.

Likewise, DirectX for the XBox is not a port of the PC library -- it's a bare-metal optimized reimplementation. MS chose the DirectX API to simplify the lives of game developers coding to the XBox platform, not to reduce their own development work. To achieve meaningful performance and feature gains in XBox2, with entirely new CPU and graphics hardware, DirectX support has to be reimplemented pretty much by definition. At least at this point in time, this sort of thing is part and parcel of the console software platform world.

Re:Won't MS have to rewrite everything?

[bhtooefr]

That's not the AMD route, that's the P6/NetBurst route (Pentium Pro/2/3/M for P6, Pentium 4 for NetBurst). Those models use a RISC core with hardware x86 translation (as opposed to Transmeta's software translation). The P6 models all have similar RISC cores, and the NetBurst (family 15) models use a different, but similar between different NetBursts, RISC core.

Re:Won't MS have to rewrite everything?

[xswl0931]

You're the one talking out of your ass. NT still has the same portable architecture it's always had, hence ia64 and amd64 versions coming. Not to mention the various flavors of embedded NT.

Re:Won't MS have to rewrite everything?

[Excen]

Since you seem to know what you are talking about, what's the difference between x86 and PowerPC? I don't understand. . .

Re:Won't MS have to rewrite everything?

[akuma(x86)]

2) Wasn't IBM working on a new mega-chip that can run x86 and PowerPC codes side by side?

Yes, but this project (based in Vermont) was canned years ago because they couldn't figure out some pretty nasty technical problems.

Re:Won't MS have to rewrite everything?

[TheNetAvenger]

yeah - and how long did these ports take?

Um, lets see, WindowsXP 64bit Edition for Itanium was shipping in 2001, oh, just a bit behind the official release of WindowsXP for x86.

And considering the Itanium 64bit is a complete different architecture than the x86, if Windows(NT) was now SO dependant on the x86 platform as some people here mistakenly think, then moving to the Itanium would have been a long journey and not just quick port like it was.

The Windows AMD64bit Edition was delayed based on CPU availability (Intel had emulator and samples of the Itanium 64bit CPUs, and even Shipping CPUs long before AMD did.)

AMD 64bit support also got caught in the security restructuring that also delayed Windows 2003. Hence why the AMD 64bit version beta and the Windows 2003 SP1 Beta are in the same development beta process.

(BTW, this info is straight from the MS development team, horse's mouth.)

If Microsoft wanted a PPC version of Windows(NT), it wouldn't be as difficult as many people think. If you somehow think the basic NT design core has changed so much that it is dependant on x86 you are not only mistaken, but grossly overlooking several NT releases that debunk this theory.

When NT was restructured (as the parent post implied), it did not become more dependant on the x86 architecture, the major revision was moving the graphical processes into the kernel scheduling and adding an additional HAL layer for Video, giving Video in NT a more direct access to the hardware instead of handing all calls through the NT kernel.

This happened with NT 4.0, and if this somehow made NT dependant on the x86 platform then it would not have also been released on PPC, ALPHA, and RISC in the 4.0 version as it was.

Additionally, even though PPC and RISC support was dropped for the NT5(Windows 2000) development cycle, ALPHA CPU support was still very a part of the development process up until RC1 of Windows 2000 (around Aug of 1999) when Compaq acquired DEC, and decided to abandon NT support on the Alpha chip - which was NOT MS's decision, Compaq intentionally pulled the plug - they were NOT going to ship NT on the ALPHA CPU, virtually eliminating any market for Windows 2000 on ALPHA.

Additionally, people forget that the core of WindowsCE (Pocket PC) comes from NT design.

Even the name 'NT' comes from the MIPS based machines that it originally ran and was designed on before it ever breathed any life in the x86 world. (NT equaling 'New Technology' is a misconception.)

Do a search with NT and RISC, MIPS, PPC, or ALPHA, and you will find more information on the portability of NT and even the current portability NT (WindowsXP/Windows2003) still has.

Re:Won't MS have to rewrite everything?

[PainKilleR-CE]

One of the Windows CE.Net pages has this little piece of info:

Support for over 200 separate 32-bit CPUs from the ARM, MIPS, SH, and x86 architectures.

Over at http://www.microsoft.com/Windows/Embedded/devices/ mobile/default.asp

I think MS is doing a lot more cross-platform work than people would normally believe.

Re:Won't MS have to rewrite everything?

[TheNetAvenger]

Ah, if only CE didn't suck donkey balls

I guess you are the clear expert in this regard. :)

Bodes well for Apple too!

[MarcQuadra]

Who thinks that IBM is going to turn around and sell these things to Aplle as the CPU for the G5 laptop?

This bodes VERY well for IBM, Apple, Microsoft, and anyone else interested in low-power-draw PowerPC systems. It sucks for Motorola, but they lost my favor years ago, and they really charge ludicrous prices for their wares.

Also, Could IBM be developing their G3+AltiVec chip for this? It seems to me that if the G3 series was dead IBM would stop working on it, but there are 750GX CPUs due soon (just a 750FX with 1MB on-die cache), and rumors of a G3 with SIMD coming down the pipe. It seems to me that if IBM bastardized some of the SIMD logic from the 970 and strapped it to the 750 they'd have a pretty decent low-power SIMD chip that Apple could market as a 'G3', 'G4' or a 'G5.'

Maybe I'm just a hopeless romantic, but the G3 was the CPU best-suited for what I do, and I hope it doesn't disappear. I have little use for SIMD, and I really appreciate running a CPU without a fan strapped to it, it's just so... elegant.

Re:Bodes well for Apple too!

[JollyFinn]

G3 is dead for desktop, and in many other places. Enchanched G3 chips are aimed at in EMBEDDED market, thats right in printers in PDA:s and other such mediums where low power comsumption low price and sometimes ability to put custom logic on same chip with processor core may help. Thats where most of PPC:s go for not apple nor any IBM workstations/servers but for other products with different companies.

Re:Bodes well for Apple too!

[will592]

Would a G3 processor be any advantage when dealing with AltiVec code? I thought this was the big difference between the G3 and the G4 processor.

Re:Bodes well for Apple too!

[will592]

Oops I'm an idiot. I missed the previous post regarding IBM working on a G3 + Altivec chip. Color me stupid.

Re:Bodes well for Apple too!

[MarcQuadra]

I don't quite understand what you mean. The G3 series isn't capable of running native altivec floating-point code, but it's a great all-around CPU for desktop use, it's a kick-ass compiler and workhorse, but lacks with video editing and scientific modeling.

Apple's original G4 was just a G3 with a Motorola-brand AltiVec unit strapped to it. Over the years the G4 has integrated more and matured, and so has the G3, but IBM makes the G3 now, and they have their OWN breed of AltiVec. I would think IBM could take the most commonly used/important logic from their AltiVec implementation and add it to the G3 for a different 'genus' of G4.

Since the G3 was built from day-one to be alow-temperature, fanless CPU I think it would make a great laptop CPU and would also do well in video-game consoles.

Re:Bodes well for Apple too!

[haxor.dk]

"Also, Could IBM be developing their G3+AltiVec chip for this? It seems to me that if the G3 series was dead IBM would stop working on it, but there are 750GX CPUs due soon (just a 750FX with 1MB on-die cache), and rumors of a G3 with SIMD coming down the pipe. It seems to me that if IBM bastardized some of the SIMD logic from the 970 and strapped it to the 750 they'd have a pretty decent low-power SIMD chip that Apple could market as a 'G3', 'G4' or a 'G5.'
"

Just so you know, a G3 + Altivec is essentially a G4.

The G3+SIMD rumors are 5 years old, and the rumored CPU was called 770.

Re:Bodes well for Apple too!

[MarcQuadra]

Right, but since the G3 and G4 split several years ago the parity has been less and less. If IBM implemented a 750+SIMD it would be VERY different under a microscope (and to a compiler) than a motorola G4. IBM has their own homegrown SIMD logic, and their G3 chips have seen a lot more evolution.

G5 Powerbook

[sameerd]

This has got to be good news for anyone dying to get a g5 Powerbook. Please please apple can I have one for christmas.

Re:G5 Powerbook

[AKnightCowboy]

This has got to be good news for anyone dying to get a g5 Powerbook. Please please apple can I have one for christmas.

No, they just came out with new G4 Powerbooks. I wouldn't hold my breath about seeing G5 Powerbooks until at least next summer.

Re:G5 Powerbook

[AKAImBatman]

> No, they just came out with new G4 Powerbooks.

Those were iBooks. Powerbooks are the ones with the metal cases.

Re:G5 Powerbook

[bojolais]

New Powerbooks just came out, but it's been a month or two. Remember the new PB15?

Re:G5 Powerbook

[AKAImBatman]

> New Powerbooks just came out, but it's been a month or two

Thus they didn't *just* come out. And with iBooks now G4 equipped, they got to do something to widen the gap. ;-)

Re:G5 Powerbook

[PainKilleR-CE]

The iBooks top out at 1GHz, whereas the PowerBooks start at 1GHz.

At the most, at least for the near future, they'll push the high end of the G4s up a few hundred MHz in the PowerBooks and leave the iBooks alone.

You'll probably see the G5s in PowerBooks in 9-12 months, with a public showing at one of the MacWorld or similar events up to 3-6 months before they're available.

Backwards Compatible?

[realdpk]

It seems that it would be prudent for Microsoft to announce something about backwards compatibility if they're going to make such a dramatic platform change. That's one of the greatest things for early adopters of the PS2 - they could still play their PS1 games on 'em.

Re:Backwards Compatible?

[yamla]

There will be no backwards compatibility between the XBox 2 and the XBox. Trust me on this one.

Re:Backwards Compatible?

[Psykechan]

It would still be possible for backwards compatibilty in some form or another. Remember, MS owns Virtual PC.

What I want to know is if it will be easy for cross development between Xbox2 and the Gamecube... seeing as how they both use a PPC/ATi setup.

Re:Backwards Compatible?

[cK-Gunslinger]

Perhaps MS's recent purchase of emulation software (Virtual PC) plays into this?

Link

Re:Backwards Compatible?

[nelsonal]

Perhaps this was what drove the VirtualPC purchase, emulating a Pentium 700 Mhz shouldn't be tough for a G5 in a year or two, (when the console is released).

Re:Backwards Compatible?

[yamla]

Nope. With the current design specifications of the XBox 2, it would be impossible even with the use of Virtual PC for them to make it backwards compatible with the XBox 1. I suppose if they released the XBox 2 in, say, 2010 or with different specifications than they are currently using, they could do it. But it won't happen. The XBox 2 will not run games from the XBox 1.

Re:Backwards Compatible?

[yamla]

Nothing has been officially announced but I will bet you the purchase and shipping price of an XBox 2 that it will not run XBox 1 games.

There are several reasons why this is true, not least of which the use of the CPU which would make emulation troublesome at best. I am not sure what speed of G5 the XBox 2 will contain but it wouldn't be fast enough to provide a 733 Mhz x86 in emulation. IBM is already talking about stripping down the CPU, removing some of the cache, etc.

You can count this as FUD if you want. You can assume I know no more than you do about the XBox 2. But you'll see when they announce the full specs, the XBox 2 cannot run XBox 1 games.

Re:Backwards Compatible?

[Daniel_Staal]

Quick counterpoint:

1: The PowerPC architecture has proven fairly good at emulating Intel architecture. It's definitely not megahertz for megahertz, but a G5 chip that emulates a 733Mhz x86 is not out of the question for a game console. If you don't mind a performance hit you don't even need to keep the same preformance nessisarially... Just tell people that while they can run their old XBox games they aren't 'optimized' for the XBox2 and therefore won't run as fast.

2: Microsoft has emulation code for the PPC inhouse: they recently bought VirtualPC which did exactly that. Now, I'm sure it is not a direct port, but it would make a good starting code base.

I'm not saying it will happen. Just that it is not out of the question.

Re:Backwards Compatible?

[yamla]

I agree that the PowerPC architecture is fairly good at emulating Intel's. But it just isn't _that_ good. The stats I've seen show the G5 at 2 Ghz capable of emulating an Intel P3 at approximately 400 Mhz. Now, I could be out by an order of magnitude here but I don't think so. There's also some benefit potentially realisable from Microsoft dynamically recompiling ix86-code to PowerPC code in the XBox 2. But it just wouldn't help enough.

Based on what happened with the Xbox 1 and the fact that Microsoft needs to be getting the hardware to their developers fairly soon, we are likely looking at a modification of the G5 processor running between 1.5 Ghz and 2.0 Ghz. 1.5 Ghz is pretty much the minimum, I expect, but I'd be slightly surprised to see a 2.0 Ghz chip in the Xbox 2. The modifications, however, will slow things down compared to Apple's G5. Less cache, optimised for reduced heat, lower power, that sort of thing. Sure, the XBox 2 will have better dedicated video and probably dedicated audio processing than the current top-end Apple G5 systems but the CPU itself will be less powerful.

And you'd struggle to emulate ix86 at 733 Mhz on a current 2.0 Ghz G5. Cripple the chip (aka optimise-for-console) and maintain fairly strict requirements of being able to emulate everything the ix86 can do at 733 Mhz, and you just cannot pull it off.

Give it a 4 Ghz G5 with all the bells and whistles and this isn't a problem any more (most likely).

Re:Backwards Compatible?

[stud9920]

Except that when the xbox 2 is released (2006 ?), it will be five years since the original xbox was released. That's about four Moore's law generations. You can reimplement the functionality (basically the instruction set) on 2^-4 = 1/16th of the original surface.

And if you design the chip cleverly, you can even reuse the on-die caches and possibly more.

This all without a very significant surface loss. And I'm sure that +- 5% surface loss is fully justified by the myriad of Xbox 1 games xbox 2 owners will want to play. Both of them.

Re:Backwards Compatible?

[stud9920]

Even more, you can clock the reimplemented functionality even higher than the original one, and use this to make up for a slower, but better surface optimized design.

Re:Backwards Compatible?

[tgibbs]

I am not sure what speed of G5 the XBox 2 will contain but it wouldn't be fast enough to provide a 733 Mhz x86 in emulation. IBM is already talking about stripping down the CPU, removing some of the cache, etc.

Probably not. Of course, if Microsoft is serious about backwards compatibility, they'll go for hardware emulation, and include a 733 Mhz x86 in the box along with the G5.

Re:Backwards Compatible?

[Megane]

But how much will they really have to emulate? VirtualPC gives that "approximately 400 MHz" performance when emulating an entire PC, OS and all.

If Microsoft comes up with a PowerPC version of the NT kernel for XBox2, then it will only have to emulate the game code, not the operating system, not DirectX, not the drivers, not the networking stack. That should provide enough of a performance improvement for a 2GHz G5, or maybe even a 1.5GHz G5, to run XBox-1 games at a good speed.

Re:Backwards Compatible?

[yamla]

This is a good point... that they don't really need to emulate the rest of the operating system, nor do they need to emulate the drivers. I still doubt you'd have enough processing power to do it but I'll accept that it is at least a good possibility.

Re:Backwards Compatible?

[stud9920]

Actually I meant Halo and GTA when it will be out.

With hindsight, "Both of them" also refers to the xbox ownes in the grandparent post

Re:Backwards Compatible?

[Bakaneko]

Uh, console games are the ultimate in time sensitivity. A game that runs "slow" will probably be considered "unplayable"... Either the Xbox 2 will have near perfect compatibility for a specified set of games (remember that the PS2 backwards PSOne compatibility isn't for EVERY game, I have a few which a PS2 refuses to run) or they'll just forgo backwards compatibility.

Basically, my arguement is it'll either be "Runs Halo, MechAssault, Crimson Skies nearly exactly the same" or "sorry, no backwards compatiblity, but the new Xboxlite, at the low low price of 100 bucks, is half the size of the old Xbox and plays all Xbox 1 titles, oh, but BTW, its made out of dixie plates, so don't beat on it too hard."

Re:Backwards Compatible?

[be-fan]

Games make very few OS calls. The Win2k kernel in the XBox doesn't actually do all that much. It loads apps, and gets the hell out of the way. Most games are nearly completely CPU-bound, which indicates that performance might even be worse than the 400MHz figure.

Re:Backwards Compatible?

[Daniel_Staal]

'Running, but unplayable' is probably the best thing that could happen to the marketing department: They get to tell people that it can run their old games (since it can, technically), but anyone who tries will soon realize they want to upgrade to the new improved version...

Re:Backwards Compatible?

[Bakaneko]

Word spreads reasonably fast in well, any modern market. If Halo won't run well on the Xbox2, then most people will know that by day 2 of general release, and I think a fair amount would know well before. So any sellthrough capability providing by advertising backwards compatibilty would be pretty quickly negated. I know that Sony made an effort to make sure the Gran Turismos ran well on the PS2, for instance.

Also, a lot of console customers don't think of their consoles like computers, but appliances. If Halo looked and ran like crap on the Xbox2, but still ran, there would be an awful lot of people who'd just generalize that to "The Xbox2 looks and runs like crap."

Re:Backwards Compatible?

[Tychoma]

Thanks Bill.

Re:Backwards Compatible?

[_fuzz_]

The stats I've seen show the G5 at 2 Ghz capable of emulating an Intel P3 at approximately 400 Mhz.

I'm curious as to where these stats came from (see my sig). It wasn't VirtualPC, as it won't run on a G5. It requires the G4's ability to swap byte-ordering in the processor. Any other x86 emulator (e.g. Bochs) is just too dog slow to be used for anything performance intensive.

VirtualPC isn't all that snappy either. I have a G4 800 with VirtualPC 6 and it pretty much feels like I'm working on a 486 when I use it.

Theoretically, Microsoft could use the x86 emulation technology from VirtualPC to emulate a processor for xbox1 games, but I don't thing things will work out. It's more likely that if they were to offer backward compatibility it would be with an add-on card as someone else pointed out.

Re:Backwards Compatible?

[Johnny Mnemonic]

As other posts have mentioned, the G5 happens to suck at Intel emulation. It apparently lacks the endian compatibility mode that the G4 had; getting VPC to run on the G5 will be a major undertaking. I don't know that MSFT has this in their favor for making Xbox backwards compatibility work.

I think it remains an open question whether or not Connectix told MSFT that before they were purchased (they surely knew it themselves) or if it was a post-purchase discovery; I can only imagine the look on the faces. But I don't think we'll ever know.

Re:Backwards Compatible?

[yamla]

The add-on cards were pretty much never good value for money. Sega did this, I think, with the Genesis. But the old Master Systems were so cheap even brand-new that you could buy one cheaper than you could buy the add-on card (which was roughly $100 Canuck at the time).

The add-on card for the XBox 2 would have a rather expensive component that makes the production of this card rather unlikely (but certainly not impossible).

Re:Backwards Compatible?

[truenoir]

I know the G5 is not good for emulation...but...aren't all XBox games written in DirectX? Wouldn't it be more possible to write a translator for whatever version of DirectX they wind up with on the IBM CPU for the old instructions? Not quite an emulator, more like what UltraHLE did in that it could understand the commands and run them rather than pretend to be the old hardware. That ran pretty well on a 350Mhz PII for me and the N64 hardware was near 100Mhz. Maybe I'm not too great at the whole coding thing (okay, I *know* I'm not great at it)...but it seems like they wouldn't have to do a full emulation deal a la Virtual PC to run the old games. Or they could just build the old CPU in too, like the PS2 did (it's the IO controller AFAIK).

Re:Backwards Compatible?

[yamla]

The 'standard' for CPU allocation is apparently roughly 5% for AI and closer to 30% for graphics (not quite sure why, though, as this should be more offloaded to the GPU, but then I'm not a graphics programmer). However, some games (including console games, specifically XBox games) are edging closer to 40% or more for AI, spending much less on graphics (or, rather, offloading almost all the processing to the GPU). These AI instructions, run on the CPU, would still need emulation. Creating DirectX for the PowerPC would, on these games, handle the 5 - 10% that the graphics were eating up. It wouldn't help with the AI, though, nor with some of the other elements that similarly must be emulated or translated. Graphics, fine. Sound, probably fine. Your input devices (which probably take up virtually no CPU already), you can provide high-speed reimplementations. But there's still a lot in modern games that you simply cannot reimplement in a general manner short of emulating the ix86 instruction set.

Basically, the GPU means that your CPU is more likely to spend its time doing general-purpose execution rather than graphics. So your suggestion, while good, isn't going to help as much as you may think.

Re:Backwards Compatible?

[Psychotext]

This would be economic suicide for the XBOX division of MS. One of the reasons MS are taking such losses on their current console is to build up a following for any future consoles that they release. Alienating their current supporters by not providing backwards compatibility would make a lot of people seriously reconsider upgrading when the new XBOX comes out.

Surely it would be simple enough for MS to follow the Sony route on this one and create an "XBOX on a chip". Lets face it, the XBox technology was hardly state of the art to start with (P3 / GeForce) and even if they would not go the addon chip route, I can't imagine that providing on board emulation supporting hardware would be beyond the designers of the XBox 2, in terms of complexity or cost.

Re:Backwards Compatible?

[mjm75]

One other thing that's going to make the x86 Xbox code easier to emulate is that the DirectX libraries will likely be native PowerPC code - and that's where all of the heavy computation happens. Even if it can't keep up 100% on emulated code, it can make up for lost time with native libraries.

Re:Backwards Compatible?

[JollyFinn]

Blah. They are probably far more than 6 months away from the business. So that means that atleast ONE process shrink that improves clock speed, and changing the endianess of a MMU isn't hard at all. MS could simply ask IBM to make em little endian, and thats it. But I think they integrate memory controller on chip also. The G5 could be the core for XBox2 but espect it heavily modified for MS needs. What 4 channel 3.2ghz XRDRAM controller ondie ... OR MCM with memory chip and 8 channel memory interface...
And new vector extensions spefically for XBOX2, doing 16FMAC:s /cycle, and ... All these things are possible, and wouldn't be overly expensive to do. 16(32bit)FMACS ~ 1/2M transistors... You know why special purpose chips kick ass for CPU:s and IBM probably offered them a special purpose chip for XBOX2. What you don't WANT memory as upgradable modules well that simply means they can do solutions that multiply their memory bandwith compared to PC:s.
What about the costs....
Well mass processed wafers costs 3000$ for And they have (150mm)*PI area... Expect yields relaticely close to 100% for 100mm chip. Lets round down and take that rounding as the yild loss lets say they get 500 peaces of 100mm die:s per wafer thats 6$ per chip, the packaging, testing and designer time is more expensive than pure wafer area. At 0.9u they could put 2MB of L2 cache and TWO G5 cores there from cost point of view. On the otherhand the power consumption will keep them from putting the 2nd core. So they probably put few special function accelerations on as extra instructions or separate cores, change endianess, add ondie memory controllers and put a Lot of cache, as cache consumes order of magnitude less power than logic. And special purpose logic is order of magnitude more energy efficient than doing same things in general purpose chip.
IBM does a lot custom chips while Intel and AMD is more of producers of standard chips in mass quantities.
To put it in perspective, they get few customizations and 2Ghz core there they probably beat G4 in x86 emulation by far.
2Ghz G5 with little endian mode by default 2MB of L2 cache with improved bandwith, and 24GB/s of memory bandwith available is not out of question for XBOX2. Think that trying to emulate 733 celeron, 3 times clockspeed 8 instructions/cycle /vs 3 ... a Lot more cache and a lots more memory bandwith , the difference should be enough to emulate it in real time.
Why they would get 24GB/s memory bandwith, its special purpose memorychips on board just like Nvidia and ATI and not just upgradable modules that cause interfaces pain in the ass signaling problems.

Stupid for MS

[afidel]

One of the few competitive advantages MS has with the XBox is that games created for it take little work to port over to the PC arena. By using a PPC chip much of that ease of porting is eliminated and along with it one of the few selling points for title owners.

Re:Stupid for MS

[ProtonMotiveForce]

How is this a Troll?! You people are crazy.

This post is completely on the mark, I wish I had moderator points. The reason the Xbox has been anywhere near successful is that it's easy to develop on Xbox _or_ PC and port to the other.

Also, developers understand the x86 architecture, and there are a shitload of tools for it. This is asinine on Microsoft's part.

Re:Stupid for MS

[terminal.dk]

Are you a developer without a clue ? Or have you been smoking the funny stuff ?

The CPU does mean shit when it comes to porting games. Games can only be written to the API since the hardware box is closed, and as long as the API is frozen, all it takes is a recompile. But there is a chance MS will use the longhorn API by then, and maybe even say C# - so they can be platform independent. Remember, C# is portable bytecode like Java.

There was a time when all it took was a recompile, and your Windows app would run on PowerPC or MIPS, Alpha or even Intel :)

A PowerPC in 2005 should also be able to do pretty well emulating the 733 MHz P3.

Re:Stupid for MS

[yamla]

The CPU may not mean much when it comes to porting (and I agree with you here) but if one of the parent articles is right and the XBox 2 ships without a hard drive, porting a PC game would be a royal pain in the behind.

Re:Stupid for MS

[dnhll]

ahh, but if the PS3 is PPC based, as it is widely speculated, then this will make it much easier for 3rd party game vendors to port their PS3 games to the XBox platform...(and vice versa) but give sony marketshare this will be a net win for M$

Re:Stupid for MS

[dadragon]

But there is a chance MS will use the longhorn API by then, and maybe even say C# - so they can be platform independent. Remember, C# is portable bytecode like Java.

Actually, all .Net code is compiled to MSIL bytecode (MSIL = MicroSoft Intermediate Language), not just C#. My C++ code can use the .Net API and be just as portable as C#, or Visual Basic. .Net can even be downloaded for Win2K and XP, as well as 98 and ME.

It compiles everything using the same format, so C++, C#, and VB can all instantiate the same objects, and use their methods. It's really quite a good system, even though it did come from MS.

Re:Stupid for MS

[tshak]

Well, C# doesn't have bytecode, but it's a similar concept.

The CPU does matter, but easily just as important is the GPU and the graphics API. Most of your graphics optimizations will happen through Microsoft's DirectX API, which will likely be very similar on the PC as well as the XBox (just like they are now). Then you just have to abstract your CPU optimizations which developers already do anyway (for SSE, SSE2, 3DNow!, and other instructions). Also, I'm sure that the next generation XBox will still be a Windows based system, so developing against the OS will be very similar to a PC. Finally, MS will probably provide tools for PC developers to make the switch virtually seemless. So, while a PPC based XBox may take a bit more development effort for PC ports, I wouldn't say it's a major setback at all.

Regarding backward compatibility with XBox titles, I agree with you completely. A 2Ghz G5 could very easily emulate a 733 Celron.

Re:Stupid for MS

[be-fan]

C#, in a console, are you insane! Game programmers are still rolling custom assembly loops for their engines, and there is no indication its changing anytime soon.

Re:Stupid for MS

[be-fan]

You've obviously not done game programming. Game programming is all about getting every last bit of performance from the hardware. That means pointer tricks and custom asm and whatnot.

Re:Stupid for MS

[ProtonMotiveForce]

Not really, the 3D engines will be _completely_ different. Porting between them won't be any easier than it is now.

Re:Stupid for MS

[antin]

It isn't a selling point for title owners, only developers. The end user couldn't care less how easy the development/porting was.

That said, the ease of porting was great for xbox 1 as it provided an incentive to develop for it, and therefore lots of games. However now that microsoft have a decent share of the console market, developers will continue support the system just as they do sony and nintendo, neither of whom have pc-compatable processors.

Re:Stupid for MS

[afidel]

by title owner I meant the developers/publishers. Eg those people who own the rights to a title. Also quite a few xbox games came about because of the compatibilty. For instance KoToR is based on the engine from NWN and was possible only because there was minimal issues in moving that codebase over to the xbox. If the porting hadn't been trivial then it's doubtfull that Bioware could have afforded to make a brand new console game.

Re:Stupid for MS

[afidel]

Your an idiot, the graphics chip is basically a limited GF4MX. The XBox uses a modified version of DX8 so porting back and forth from the PC is trivial so long as the PC minimum specs are made sufficiently high and the coders avoid doing stupid hardware specific things on the xbox side. For an example Bioware's KoToR is based on NWN's Aurora engine which they developed origionally for DX7 on the PC several years before the XBox was even a product.

Re:Stupid for MS

[BOFHelsinki]

You mean a "limited GF4". The XGPU (NV2A) in Xbox is in essence a GF3 (NV20) with an extra VS unit. GF4MX (NV17) is GF2 class (DX7).

Incidentally, I wonder if the "NV2A" codename comes not only from "somewhere between NV20 and NV25" but also in rememberance to the NV2 console chip, which Sega back then rejected because it was based on quadratic textures instead of textured triangles (like the non-DX-compliant NV1 which also almost killed Nvidia) and thus was a PITA to develop for.

Other links

[iJed]

This has also been mentioned on the following sites:

• The Register
• Mac Rumors
• Mac Central

It would be interesting to know exactly why they have picked an IBM chip rather than Intel or AMD. I wouldn't think the IBM (PPC?) chip would be more cost effective than the Intel/AMD but you never know...

Re:Other links

[Anonymous Coward]

So far it has not been stated that microsoft is using an IBM chip. Merely that they are licensing technology from IBM. This could mean a chip deal. But it might not. Any statement that the x-box 2 will contain an IBM chip is pure conjecture at this point.

It is wierd that the news sites are presenting conjecture as if it were a certainty. Overall I think both the slashsites and the register are guilty of very sloppy journalism here.

Re:Other links

[Otter]

And, after all this, there still are no specifics from either company about what is involved. Just a lot of uninformed speculation from "analysts" that it's about the core Xbox processor...

Re:Other links

[jtnishi]

It could and would, for the reason pointed out by GreatDrok on the other thread.

His post is here.

Basically, if the hardware is designed custom, you can make up for the costs in the long run by manufacturing lots of chips. Costs go down. While if MS went with Intel or AMD, the likely conclusion is that MS would've gone with stock chips. On the other hand, that logic fails to make sense considering that IBM would be using a G5 based core, supposedly. But there's probably more to this than these rumors.

Chips Ahoy!

[CptTripps]

So were going to have a loss-leader that runs off of a Microsoft machine that uses an IBM chip co-developed by Apple? talk about Clash of the titans....

Re:I have to ask this question:

[lostchicken]

Well, remember, WinNT 4 ran on Alpha, MIPS and PowerPC. Nobody bought it then, it's very unlikely that that would change.

And the Itanium is not a x86 chip, and we've got XP for that.

Erm...

[athakur999]

and then adding a pirated disk drive

This is just pure ignorance. Apparently going to Best Buy and buying a hard drive is now "piracy".

Re:Erm...

[cK-Gunslinger]

Apparently going to Best Buy and buying a hard drive is now "piracy".

Yes, but only because 80% of these people return home via ship and swing on a rope into their homes. Arrrrgh!

Re:Erm...

[Godeke]

I did have to ask myself what the heck they were talking about. I guess a lot of XBox owners are further reducing the cost of the conversion project by boarding trucks while in traffic and absconding with Maxtor drives.

Avast ye scurvy trucker, prepare to be boarded!

Re:Erm...

[TylerL82]

I don't know about you, but I regularly get my consumer technology products by hijacking Best Buy trucks.

Re:Erm...

[Jeff DeMaagd]

Unless copyright enforcement features are disabled, the only penalty I see that could stick is that the warranty could be voided.

Re:Erm...

[Nucleon500]

Also, such hard drives have many times the storage capacity of regular drives. The count double.

Re:Erm...

[gl4ss]

buying from pirates of the seven seas is cheaper than best buy, you gotta look out that they don't shangai you though.

.

IBM G5s?

[BlackBolt]

Yeah, but can it run Mac OS X?

NOT!

Re:IBM G5s?

[Jesrad]

Yeah, but will its games be able to run in a virtualized environment on my G5 ?

I CERTAINLY HOPE SO!

Re:IBM G5s?

[gl4ss]

yes, some 4 or 5 years after the launch.

not by official support though ever.

Mod Chip

[blackmonday]

I can't wait for the Morpheus II Mod-chip that lets you run OS X on the Xbox2.

Re:Mod Chip

[KojakBang]

I never understand why someone would mod an Xbox to put something else on it. People are so excited to use Linux on an Xbox. And for what?

ORN: So what have you been playing lately on your Xbox?

BH: Linux. I don't use my Xbox to play games.

I used to own a copy of Dead or Alive 3, but I gave that to a friend after I got bored of it. I also tried Halo once and bored of it pretty quickly. I tend to play the Nintendo GameCube the most; its games are the most fun. I am still working on beating the new Zelda.

Why would you buy one of these to put linux on? If your not playing games wiht it why not do one of the following:

1) Use a small older computer from ebay and install linux

2) Build a computer and install linux

3) Buy a PS2 with the Linux Kit

4) Buy a dreamcast and burn your linux boot cd

And even then WHAT IS THE POINT? It seems like people just do this because THEY CAN and are not asking whether they SHOULD be doing this?

And I still don;t see the use for it. What are you possibly going to do with Linux on an Xbox that you couldn't do with Linux on your computer?

The same argument can be said for using a Mod chip to put OS X on the Xbox2.

Re:Mod Chip

[Carnildo]

An Xbox is effectively a Microsoft-subsidized PC. What Linux geek wouldn't want MS to pay for their new server?

Re:Mod Chip

[wheany]

A Linux geek that doesn't want to give Microsoft money.

Re:Mod Chip

[BOFHelsinki]

>> What Linux geek wouldn't want MS to pay for their new server?

> A Linux geek that doesn't want to give Microsoft money.

Huh? The grandparent was correct. MS loses money on every Xbox machine they sell: in essence they subsidise the hardware.

So if you buy an Xbox, MS doesn't get your money. (Unless you buy games, which changes everything, as per the typical console business model. Whereas Sony seems to be making a small profit on the PS2 hardware too, at least nowadays with all the volume benefits, and up until the machine's price dives permanently into the gutter. Nintendo is rumored to be breaking even quite nicely with the hardware. -- Dunno about overall profiting, which pits the possibly bigger factors, per-game revenue and marketing costs against each other.)

Re:Mod Chip

[wheany]

Microsoft has already paid for the console when it sits on the shelf. Then you go and give Microsoft some (all?, I don't believe Microsoft sells them at a loss anymore) of the money back, and take the console home.

In addition to that, Microsoft Xbox now has a larger perceived market share.

Re:Mod Chip

[BOFHelsinki]

But then Microsoft has to pay for and put another console on the shelf, because now there's an empty slot ;)

(I believe they still less at a slight loss; I'd have to check this though. I don't think the arbitration with Nvidia helped much because retail prices have dropped much more. At least at the local Koneveljet...)

Agree about perceived market share, but it doesn't increase revenues.

PowerPC was *supposed* to become a commodity chip

[KojakBang]

IBM was going to come out with personal powerpc systems which would even have a common motherboard reference design with Macs. The volume of production would drive down costs dramatically. But IBM didn't and Apple basically got screwed on that deal.

Interestly enough, the reason IBM canned the personal powerpc systems was that OS2 for PPC completely blew its schedule several times over. IBM had a personal AIX edition for PPC ready but chose not to go with that. The reason. Unix would never make it as a mainstream operating system for PCs.

Like rain on your wedding day!

[joeytsai]

It's ironic that IBM, with its roots in the computer industry, doesn't supply the processors for the main portion of the personal computer industry. Intel does.

Disregarding the fact that the statement isn't ironic, I have been wondering when IBM is going to release a new line of Intellistations using Power5 chips. Does anybody know when we might be seeing these? Or when Linux will be running on them? I think I remember hearing that some people got Linux running on them, but not 64-bit...

For all the noise that Slashdot likes to make about Apple's G5s, IBM, and Linux, I think it's rather strange that we haven't heard of any news for Linux on G5.

I leave it to you to see if that's ironic. :o)

there will be backwards compatibility

[spir0]

as mentioned in Financial Times, Microsoft will likely be using their recently acquired Virtual PC software. This software is the way mac users run windows software on PPC chips. VPC technology will allow MS to provide backwards compatibility under Intel emulation.

Re:there will be backwards compatibility

[discstickers]

Ironic, VPC can't run on the G5.

Re:there will be backwards compatibility

[Goldfinger7400]

Holy Fuck! Does this mean that my G5 might soon be capable of playing x86 games? I mean, the hardware is pretty much the same as what they're gonna be putting into the X-Box...

Just as long as they aren't douches and drop mac support for VPC completely, this would kick ass. At least it means they're gonna try to fix the PPC970 compatibility issue with Virtual PC.

Re:there will be backwards compatibility

[spir0]

currently...

MS will fix that. eventually.

Re:there will be backwards compatibility

[spir0]

assuming that it's actually the G5 they will be using. The G4 would make financial sense, and is still grunty enough to run top end games with enough RAM and a decent frame buffer.

My pick will be that they'll announce a dual G4 or a low end G5 with shit on-chip cache.

Re:there will be backwards compatibility

[idlethought]

More likely, XBox2 will be incompatible, but there will be 'enhanced' ports of old games to the new platform. Using the old code base (mostly C++) this will create both a reasonable starting library of games and offer the chance for MS and games producers to milk a little more cash out of old titles cheaply.

Re:there will be backwards compatibility

[spir0]

I don't know if you're serious and you actually think this is likely or if you're only kidding.

but for the record, hell would have to freeze over a second time before that happened.

Re:there will be backwards compatibility

[pi radians]

There is no way they are discussing the G4... since Motorola, not IBM are the manufacturers of it.

Re:there will be backwards compatibility

[troc]

But IBM do make a G3 with AltiVec which is essentially a G4.

troc

G5 ... firing

[maximus21]

wonder if this was the big deal with the Mac guy being fired by M$ for reporting that they purchased the gang of G5's..... for their xbox r&d department..

Re:G5 ... firing

[aristotle-dude]

Um... Maybe for the game gfx artists, game developers, and Virtual Xbox 1 developers in the Xbox R&D department?

Are you a product of the US educational system? Learn to read.

Okay, now this could be awesome.

[iamdrscience]

Assuming that this "next generation xbox" thing will be as hackable as the current Xbox (and the Dreamcast and the PS2, etc.) this could lead to a reasonably priced PowerPC machine to mess around with. This could be very cool. The current Xbox isn't especially competitive with a comparable low budget x86 system, but right now the only way you can mess with a G5 is to spend a couple grand on a new mac and likely once this new system comes out a cheap G5 system will still be the better part of a grand.

Re:Okay, now this could be awesome.

[cascino]

In two to three years I doubt this processor will matter much. I'd imagine the Xbox 2 of 2005-2006 will be equivalent to the Xbox 1 of 2002... i.e.: not terribly significant in terms of performance for the price.

interesting move for xbox..

[Thrakkerzog]

Wasn't one of the "Benefits" for people making games for xbox the ability to code just like they did for pc games?

perhaps they are trying to lure developers away from GCN by offering a similar cpu architecture?

Re:interesting move for xbox..

[NanoGator]

"perhaps they are trying to lure developers away from GCN by offering a similar cpu architecture?"

Away from GCN? I doubt it. But they could be after getting more ports from GCN games, that wouldn't be all that surprising.

I dunno if the processor matters as much as it used to anymore. I mean, you've got your time developing the engine etc, but a good deal of time spent making a game involves creating the assets. Building maps, creating 3D models and textures, sound effects, etc. Those formats are pretty much portable now. I don't think it's like it was in the olden 16-bit days where the machines were just so vastly different in how they behaved.

Re:interesting move for xbox..

[Thrakkerzog]

how much assembly do you think goes into the drawing routines?

You have to work with your byte order flipped.. there could still be a considerable amount of changes.

I don't think nintendo will ever port any of their 1st or 2nd party games to xbox.. other than that, there are not many gcn exclusives..

Re:interesting move for xbox..

[yamla]

They can still code just like they do for the PC games. Just make sure you port DirectX, etc. etc. Virtually nothing in modern games is programmed down to the assembly level. Yes, porting to a non-ix86 platform requires recoding and would likely require more optimisation but the coding itself wouldn't be any more difficult on the XBox 2 than on the XBox 1.

Re:interesting move for xbox..

[gl4ss]

they still code so(as they would code for x86 based system, if you mean by that they use language like c++ and mostly access the hardware only through some abstraction like directx).. the compiler/dev environment just compiles for the chosen system.

the cpu architechture itself is starting to be less and less meaningful as developers move away from very low level access(saving few opcodes every now and then isn't that precious anymore). it certainly isn't a ruse to get gcn developers to jump aboard(afaik barely any of gcn stuff is coded so low level that it would really matter anyways for them what architechture the cpu is)

-

So is it a fair assumption...

[NanoGator]

... that backwards compatibility isn't part of Microsoft's game here? I suppose they could do like the PS2 did and use the original Intel processor to act as a controller or something, but somehow I doubt that'd be cost effective enough.

Emulator? Eck I hope not. Well.. maybe that wouldn't be so bad. Maybe they could do a combination emulator and wrapper. The emulator would be for the processor instructions, and the wrapper would be to send the graphics commands to the new GPU. Presumably, the difference between the two GPUs wouldn't be big enough as to prevent that from working.

Eh I dunno. Personally, I'm hoping Microsoft does something a little more interesting than just throwing next-gen hardware into a box as an upgrade. Pushing polygons around is nice, but I really like how small and cheap my GameCube is. *Hint hint*

Re:So is it a fair assumption...

[user32.ExitWindowsEx]

2 words: Connectix VirtualPC.

They can customize and optimize the hell out of it since they know the *exact* HW platform it'll be on. All they have to do is have the boot ROM check to see whether it's an i386 game or a PPC game.

Re:So is it a fair assumption...

[NanoGator]

"They can customize and optimize the hell out of it since they know the *exact* HW platform it'll be on. All they have to do is have the boot ROM check to see whether it's an i386 game or a PPC game. "

Fair enough, but it seems like a daunting task. There are a lot of games for consoles that rely on tricks of the hardware to bost performance in problem areas. Seems forseeable that unless Microsoft does lots and lots of testing (ha!) there are going to be games that have less than ideal behaviour.

Think they'll be able to solve that?

Re:So is it a fair assumption...

[yamla]

Yes, Microsoft has no intentions of making the XBox 2 backward-compatible with the XBox 1. They decided it doesn't make enough business sense (how many people do you know who buy PS1 games to play on their PS2?). They could be wrong, you can debate that if you wish. If they are able to get a good selection of A1 games available at launch (they couldn't do this with the XBox 1 and even now, the selection is seriously lacking), it is unlikely to seriously impact sales.

Re:So is it a fair assumption...

[samdu]

Presumably, the difference between the two GPUs wouldn't be big enough as to prevent that from working

Actually, there's already been talk about the problems with backwards compatibility because of the change in graphic processors as well. nVidia had/has the contract for the X-Box 1 and ATI picked it up for XB2. Given that the XB2 will have both a new CPU architecture and a new GPU architecture (with incompatible optimizations), I think getting old XB1 games running on the new box may be a bit harder than some are anticipating. But, given MS's history, XB2 is not the machine we should be worrying about, it's the third generation X that should be done right.

Re:So is it a fair assumption...

[user32.ExitWindowsEx]

I'm sure they have a copy of every game legally made for the X-Box...and who knows what *limits* DirectX imposes on hardware tricks?

Re:So is it a fair assumption...

[NanoGator]

"and who knows what *limits* DirectX imposes on hardware tricks?"

Where'd that come from?

Let me get this straight...

[HoldmyCauls]

That's IBM... supplying Microsoft... with PowerPC processors... for a gaming console...?

If anyone needs me, I'll be conferring with my local pastor as to whether or not Hell has frozen over.

Re:Let me get this straight...

[Temporal]

Dude, hell froze over weeks ago. This is just an after-shock. Didn't you see the iTunes announcement?

Re:Let me get this straight...

[tabby]

And the Diablo2 1.10 patch was released as well. Maybe Bil^H^H^HSatan did install new air-con.
HellFrozenOver-text-640x.jpg

Re:Compatibility ?

[demonbug]

Does this mean that Xbox2 won't run Xbox 1 titles ?

Don't worry, when Xbox2 comes out your Xbox 1 won't run Xbox 1 titles either, so backwards compatability won't really be an issue.

I don't see anything about XBox2 using IBM chips..

[downix]

What I do see is Microsoft hedging its bets by licensing technology. Now, it can go to both Intel *and* AMD and go "if you two won't give us a better price, we'll cut you both off."

When businesses compete, the consumer wins.

Ironic

[Gregoyle]

It's ironic that IBM, with its roots in the computer industry, doesn't supply the processors for the main portion of the personal computer industry. Intel does.

That's not ironic. It would be ironic if IBM declared the PC industry dead, and said that the embedded industry was all that was viable, made this processor for the embedded industry, and someone used it to revitalize the PC industry and put IBM back on top there. The fact that they are not on top of an industry that they helped start is interesting, but it's a far cry from ironic.

Not to pick nits, but misuse of the word "irony" is one of my pet peeves.

Re:Ironic

[anomaly]

If the vast majority of english speakers use the word incorrectly, doesn't that mean that the effective meaning of the word has changed?

Re:Hope it's...

[Jesrad]

The Gamecube has a 486MHz PPC 750CX/e, which is 32bits, has limited front side bus bandwidth and 256K L2 on-die cache. The XBox2 will likely use a variant of the PPC 970 or 980, which are 64bits, have plentiful bandwidth and 512K on-die L2 cache, plus an altivec unit, and run around 2GHz.

Of course it'll be better.

intel/amd/ppc

[Sebastopol]

Intel went after Xbox originally so that AMD wouldn't get the win, but Intel still took it in the shorts, or so it is claimed, by offering a nutrageously low price to outbid AMD. Probably even at a loss given the timeframe.

Intel doesn't give a crap about PPC, as it isn't even a remote threat, what at 4% of the market. Intel could have EASILY played the same power-play and had another design win, but at the cost of lower ASPs for a niche market (compared to its $20b a year market, xbox isn't worth it).

just my $0.02.

Yes- article author is stupid.

[mekkab]

Yes, your clarification where you added "personal" changes everything. That statement says more about the ignorance of the one who uttered it than it does about IBM's "market" position (whatever that means).

Poor Man's Mac?

[FerretFrottage]

"The current Xbox is based on a Pentium 3 processor running at 800 MHz or less. One problem Microsoft has faced is the conversion of Xbox systems into personal computers. By buying a heavily subsidized $200 game machine from Microsoft, and then adding a pirated disk drive, the Xbox can be used as a "poor man's PC, turning a $200 game machine into a $600 personal computer, which Microsoft doesn't like at all," Doherty said. That may have led Microsoft to the PowerPC platform developed by IBM. " So people will be booting Yellow Dog instead on Mandrake (does that make a Yellow Dog Man-Drake's best friend?)

i'd buy an xbox if osx ran on it

[webdev]

microsoft are you listening?

Re:i'd buy an xbox if osx ran on it

[Jesrad]

You probably want to speak with Apple, instead.

Unfortunately Apple already tried to get in the videogame console market, with the Pippin, which was a monumental failure. Steve Jobs is not likely to send Apple this way again anytime soon.

Re:i'd buy an xbox if osx ran on it

[spir0]

why would microsoft listen to that?? you are telling them:

- that you will buy their heavily subsidised piece of hardware (which will cost them money)
- that you will most likely modify it illegally and place your own BIOS to run OSX
- that you won't be buying any of their games which, incidentally, is how the console makers actually turn a profit

if you want OSX, buy a Mac.

Apple will also have no interest in you, because you won't be buying any of their hardware. This is how they turn a profit.

Keep using windows. it was made for chumps like you.

Geez.

[Zenki]

MS isn't going to go PowerPC. Back in the day, the PowerPC port of NT was done by IBM for MS. More likely, IBM is going to be making the ATI designed video chip for Xbox. ATI probably worked out a deal where they sell MS the license to use the graphics chip design and leave the manufacturing problem to MS to work out. MS is probably just getting a deal with IBM so they can use IBM's fabs to churn out that chip for Xbox 2 and maybe other bits too.

Re:Geez.

[SEE]

Actually, the NT PPC port was done by Motorola. IBM did OS/2 PPC.

Re:Well, you know what this means...

[sy161e]

If they build the console from the ground up, they might as well use the most efficient means possible, as well as eliminating the case of installing alternative software on the machine. Right now they're selling the consoles *drastically* at cost, and I don't see them lowering the price to compete with the GC's $99, but if they had effective components and low-cost yet powerful hardware, they'd be a console manufacturer, and not simply a "We want to make a mini-PC so porting the W2k kernel and DX will be uber-easy" developer.

Megadrive was first

[DrSkwid]

With it's backwards compatibility with Master I & II

Re:Megadrive was first

[default luser]

Actually, the earliest instance of easy backward-compatibility I can recall was the Atari 7800. It could read 2600 and 7800 carts, all from the same cartridge slot. Of course, the 7800 was so heavily delayed that Nintendo stole their market...

Even though the z80 hardware was built in to the Megadrive / Genesis, you had to shell out for the Master System Converter.

This, plus the fact that the Master System didn't sell too well, made it a losing feature.

When you sell millions of PS1s, and you can get an optical drive cheap that reads your old CD media, and your new DVD media, then you've got backward-compatibility that's a selling point.

ATARI and Sega did it first. Nintendo and Sony were the first companies to do it right.

Re:Megadrive was first

[KernelHappy]

Maybe it doesn't count as backwards compatibility (either because it was an add on or because it added compatibility with someone elses product) but I think that Colecovision beat the 7800 to market.

Re:Megadrive was first

[KernelHappy]

Errrrr I meant to say that Colecovision and the 2600 -Adapter beat the 7800 to market.

X86 emulation under PowerPC

[ahchem]

This is actually a very interesting move on MS's part. Not too long ago they purchased the premier x86 on PPC emulator. Now they are going with IBM's PPC as there next Xbox. Maybe they are ready to have a win-tel divorce and declare their independence from Intel, or maybe just shake the relationship up a bit.

Re:X86 emulation under PowerPC

[yamla]

This would be great as far as I am concerned. However, ix86 is still the cheapest instruction set for the performance available for a general-purpose PC. I don't see any benefit, and considerable problems, with Microsoft divorcing itself from the ix86 instruction set. Don't forget, of course, that Microsoft already tried something similar back in the NT days and ended up abandoning all non-ix86 platforms.

Of course, ix86 isn't the be-all and end-all, not by a long shot. I'd rather run PowerPC, myself, if the price was right.

Re:X86 emulation under PowerPC

[yamla]

This should be marked as insightful. My experience is that the iBook and Powerbook, for things I am interested in, scale very closely Mhz for Mhz with my Athlon XP desktop. That means that the 866 Mhz Powerbook was only about 57% as fast as my 1.46 Ghz Athlon desktop. However, I primarily used it for programming and these are compiler benchmarks (gcc on both platforms). For other tasks, the G3 and G4 scale slightly better.

But, after my rather long-winded paragraph, the G3 and G4 seemed to have much better battery life than similar Intel-based laptops did. If they can do something similar with the G5 and get a 2.0 - 3.0 Ghz laptop out next year for a reasonable price, I'd probably buy that rather than a similar (mobile) Athlon64-based laptop. Battery life on a laptop is very important. That said, many people I know primarily plug their laptops in to AC and use them as ultraportable desktops. And the iBook and Powerbook systems I used had as significant problems with heat as did the much faster Intel-based laptops. The Powerbook, for example, actually burned my legs when I was wearing shorts.

Test?

[Dr. Evil]

If it compiles, ship it!

This chip will be used for the AI ....

[zymano]

and not graphics ? Xbox2 will have a dedicated graphics card which will do most of the heavy work . It kind of makes me wonder why they don't use a vector processor instead a general processor for non graphics and then have the vector and graphics card combine when doing difficult rendering. Could increase the polygon by alot. Just an idea.

Microsoft aren't total idiots...

[Thud457]

No internal standard HD == no internal standard BUS == less easy to hack.

I forsee some wacky "CD" format that it will be a DMCA violation to publish details about.

Re:Dear God ...

[AKAImBatman]

I guess they listened to my rant the other day. See? It's good to fly off the handle every once in awhile. ;-)

This could be a good move

[ryanw]

Could you imagine if Microsoft enjoys the PowerPC platform so much that they end up porting Windows to the MORE STABLE PowerPC platform? I mean they're going to have to port their APIs and such to make this sort of move easy for the developers anyway. This would enable consumers to buy any random intel box or buy a nice IBM Box. This would rekindle the love/hate relationship that DOS encouraged. IBM would sell the machines, microsoft would supply the OS. Most of the intel systems IBM sells are for Linux these days. The PowerPC systems run AIX. IBM has been phasing out Microsoft as much as possible. But if microsoft would embrace the PowerPC platform and promote sales of IBM's more expensive hardware, I'm sure IBM would be on that band wagon.

Re:This could be a good move

[Jesrad]

Could you imagine if Microsoft enjoys the PowerPC platform so much that they end up porting Windows to the [...] PowerPC platform?
Please read this slowly and carefully:
NOT.
HAPPENING.

Now a bit of history: in the 90s IBM convinced Microsoft to port Windows to the PowerPC, in order to make their PPC workstations more attractive to the corporate IS market. It was a failure because very few software vendors ported their apps over. What makes you think they would, now ? Besides, IBM has Linux now. It's only a matter of time before they stop "recommending" WindowsNT on their x86-based products.

Another thing to note: IBM would NOT build and sell the hardware for the XBox2. The most they'll do is provide a custom PPC, should it be a PPC 970/980 or a PPC 750GX variant.

Re:This could be a good move

[Xeger]

Microsoft has taken the first step toward removing the binary compatability burden from their vendors' shoulders by introducing .NET as their blessed-from-on-high development platform.

Inherent to .NET, of course, is the write-once-run-anywhere concept. IMHO this mentality comes just in time, as we are on the doorstep of 64-bit computing, at which time the latest architecture from a given vendor (Intel or AMD, take your pick) will no longer be compatible with the previous generation of chips. Ready or not, Windows is about to support multiple architectures.

Transmeta has proven that one can achieve non-native compatability (i.e. without building a compatability mode into the silicon); this, combined with the tremendous amount of research that's gone into virtualization and emulation since the mid 90s means that it's no longer a pipe dream to dynamically translate code between architectures. This is especially true if you're translating a binary that targets the same underlying API, and the API is designed to abstract away the BIOS and other architecture-dependent parts of a computer.

I'm not saying that we'll see a PPC port of Windows sold in stores -- I agree with you that it's about as likely as Bill Gates slapping a penguin bumper sticker on his car. I'm just saying that the technical obstacles that killed off Alpha/PPC Windows have since been surmounted.

Re:This could be a good move

[Unregistered]

Nope. IBM isn't in the hardware buisness anymore. They're in support and windows is much harder to support than linux or aix.

Re:This could be a good move

[Jesrad]

Very interesting, but somehow I don't think that Microsoft will try to compete with Apple on Macintoshes. Windows XP or Longhorn for Macs, even with .NET apps, would be too much a burden for them, and the Mac-community could react very strangely to this.

Re:This could be a good move

[BOFHelsinki]

You mean the "selling commodity hardware" business. They are very much going into the "commodity hardware subcontractor" business (next Nvidia chips, etc.). They'll remain in the "selling high-end hardware" business -- watch out for the Power5 Regattas. Maybe some laptops too.

Correct that they want to expand their Support and Services businesses. (Well, in Services they're global already, but they'll want to eat competitors or form new partnerships, and expand into total corporate solutions -- I won't be surprised when we see IBM Accounting, IBM Marketing, IBM Advertising, IBM Organising Entire Industries Into Worldwide Megacorporations, IBM Running An Efficient Government, and other business related services...) [Okay, I *was* joking with the last ones...]

Re:This could be a good move

[Xeger]

I think you've hit the nail on the head here. The real problem with a retail PPC port of Windows is that there's no target market!

Anyone with a PPC system is already a huge Mac fan, and anybody looking to buy a speedy Windows system is going to buy Intel hardware at 75% of the cost. Anyone who wants to run MacOS server applications on Intel hardware can simply run GNU Darwin.

The real market is to be found in the millions of Intel users who want to get their hands on MacOS, but don't want to shell out the greenbacks for premium-priced Apple PPC hardware, and/or who don't feel comfortable with Apple's "don't look inside the box" mentality.

This explains the G5s on the loading dock...

[klubar]

Just a thought... alternatively, it could trash the market for Macs if hackers modify the new Xbox to run OSX.

Re:Hope it's...

[nate1138]

Knowing Nintendo and their Kid-oriented market, A move like that would not be surprising. It makes it that much easier for little Johnny to convince his parent to get one for x-mas.

Windows for PPC

[Sri Lumpa]

I must say I absolutely didn't see that coming and if somebody had told be beforehand I probably would have chuckled.

This means a version of Windows for PowerPC and, if they use the G5, a 64 bit version of Windows.

I think the former used to exist a while ago (NT4?) and the latter is available in a more or less broken form but the fact that XBox2 would be a fixed hardware platform may allow them to make a better version.

I'm no MS fan but I can only see that as a good sign as long as they don't use the 32 bit subset of the PPC architecture, it could help them with a better 64 bit Windows which would help drive the adoption of such systems up and therefore drive the price down; which would help enlarging the 64 bit Linux userbase.

Ok, it's a lot of "if's" but at least it opens the possibility a bit wider.

Re:Windows for PPC

[licketyspit]

NT4 existed for the alpha, but I don't recall it on the ppc. If my memory serves me correctly, this will be the first windows for ppc, or in my case linux-ppc on xbox hopefully!!

Popping on the Conspiracy Theory Hat

[Rude Turnip]

I think MS may have figured out a way to leverage the fact that people like to use the Xbox as a cheap computer. Anyone care to bet that this new PPC-based Xbox will be intentionally easy to install OS X on? Thus, for the cost of the Xbox (say $300 when it first comes out) and a retail copy of OS X ($130), you could have a cheap Mac clone for less than $500, thus killing Apple's low-end ($800) hardware market.

Re:Popping on the Conspiracy Theory Hat

[Jesrad]

Well, you buy an XBox2 and try that. I'll probably have settled definitely on Mars before you succeed. Steve Jobs does not tolerate Mac clones. And Apple still has the ultimate say on OS X's source code, it's not Linux you're talking about.

Re:Popping on the Conspiracy Theory Hat

[Rude Turnip]

I'm not talking about tweaking OS X source code, I'm talking about buying a copy of OS X off the shelf at CompUSA and installing it on one of these PPC Xboxen. And realistically, Steve Jobs would be powerless against MS if this happened.

Um, so how is PPC more stable than Intel?!?!

[FatSean]

Please give examples of instability in the Intel hardware.

Re:Um, so how is PPC more stable than Intel?!?!

[aristotle-dude]

The powerPC architecture does not have the problems with legacy that X86 does: IRQs,DMA conflicts,paralell ports,serial ports segmented memory etc... None of those "features" belong in a gaming console.

History Repeats Itself

[ewhac]

Not long after the 3DO Interactive Multiplayer came out, we started designing the next-generation platform. The machine was to be PowerPC-based with true 3D rendering capabilities (triangle engine, MIP-mapping, perspective-correct textures, 32-bit rendering, etc.).

The CPU was supplied by IBM. What we ended up with was the PowerPC 602, which was essentially a 603 (?) with a smaller cache and single-precision floating point operations that executed in a single cycle, which were essential for 3D gaming. The part ran at 66MHz.

It was a really nice machine. Sadly, it essentially died on the vine, as Matsushita chose not to exploit its gaming potential, relegating it instead to "kiosk" activities.

IBM also manufactured the triangle engine. It was a five layer chip -- at that time, a rather sophisticated process -- occupying 144 square millimetres.

Schwab

Re:History Repeats Itself

[mekkab]

I always wondered what was between the 601 (the bridge chip, instruction set-wise, from the POWER family) and the 603!

Re:PowerPC was *supposed* to become a commodity ch

[Anonymous Coward]

If i remember correctly, IBM didn't cancel PPC CHRP (Common Hardware Reference Platform), Apple did when they killed the mac clones. There was already a few CHRP based mac's out (Power Computer and a Motorola Starmax) and OS8 CHRP edition was about to ship. Then they bought NeXT, Steve Jobs came back, got rid of clones and effectivly killed CHRP by canceling all related projects in Apple. BeOS, NT PPC, AIX, OS/2 etc weren't going to be able to carry the platform without Apple and MacOS. With all the virtual machine and the surge of Linux we might see a comeback in an open PowerPC platform based around the G5.

Definitely will be some rewriting...

[JMZero]

...but they're not going to be porting "Windows" in any meaningful way. DirectX will have to be fiddled with - but even that will be a much simpler enterprise in this case as they won't need to port the old interfaces.

Summary on FT

[donutello]

The Financial Times has the best summary of this that I've read so far.

It will give people second thoughts.

[mb12036]

Even if it's an add-on investment for your X-2. Some people (myself included) have some serious coin wrapped up in games and observing Sony's compatibility from one generation to the next will take a bite out of X-2 sales as current X-Box owners jump ship. I mean heck, if Sony's got a better product, my brand loyalty will go right out the window if I have to start from scratch purchasing all new games again.

I guess it's liberating from one perspective, since I won't be influenced by the money I spent on my existing games, but I might be biased against Microsoft for screwing me on backwards-compatibility too.

backwards compatibility at full speed?

[Zed2K]

So not only is the xbox 2 going to use a completely different graphics processor but also a totally different CPU? Isn't this going to cause some slight (but not impossible) big-endian/little-endian problems as far as the cpu is concerned. It doesn't even say anything about problems with xbox 1 games writing directly to the GPU hardware being able to run on xbox 2 graphics hardware.

I'm sure xbox 1 games will run on xbox 2 but at what speed hit and general compatibility?

The PS2 runs PS1 games so well because (if I remember things correctly) the cpu of the PS1 is basically in the PS2 as well to handle other things...

My Prediction

[Goyuix]

Just as with the original XBox, there were plans to use AMD chips, at least that is what most of the speculation on the internet said, up until they announced the specs and low and behold Intel had "won" the contract instead.

I could very easily see this deal with IBM as a backdoor sneaky tatic to get nVidia hardware under the hood again, especially with their cozy arrangements as of late. For that matter, AMD and IBM are also in bed so it wouldn't be surprising at all the see ATI booted and AMD/NV offering becoming the real guts of the console. IBM would be the fab for the chips and assembly for the mainboard... heck they might even just roll the whole thing and take a cool percentage.

Off the wall? Perhaps. But I think this xb0x0r war is far from over. There are no published specs and these talks of industry deals are by no means what the final product will necessarily be.

Re:My Prediction

[yamla]

Deciding between AMD and Intel, both using ix86, is a lot different between deciding between Intel (or AMD) (ix86) and IBM (PowerPC). These are completely different instruction sets. I just don't see Microsoft trying this tactic as a means to get NVidia as their graphics chip supplier, it just seems too much of a stretch.

Re:My Prediction

[JasonAsbahr]

That would be nice, then NVDA would go up again. : )

C'mon, there are lots of non-x86 chips out there

[Junks Jerzey]

Most embedded CPUs are not x86-based. They're not PowerPC or ARM based either. It's just that most people aren't familiar with what CPUs are out there, only what's available for PC boxes.

That said, consider that the PlayStation 1 and PlayStation 2 use MIPS processors. The Sega Saturn used a Hitachi SH-2. The Dreamcast used an SH-4. The 3DO console was ARM based. The Nintendo 64 uses a MIPS. The GameCube uses a PowerPC. The Game Boy Color is Z80 based. The Game Boy Advance uses an ARM. The Nokia N-Gage also uses an ARM.

In short, non-x86 based game consoles are the norm, not the exception. You simply can't put a super hot P4 in an embedded environment. Intel knows this. That's not the market they're after with the P4. This is basic embedded systems design.

One wrinkle using the VPC code from Connectix...

[LionMage]

as mentioned in Financial Times, Microsoft will likely be using their recently acquired Virtual PC software

Except that if Microsoft uses the G5 (PPC970) chip, as everyone is speculating, they'll have to tweak the Virtual PC code base to run on the G5. Why? Because the G5 silicon lacks the special "virtual little endian mode" that the Virtual PC code from Connectix relies upon for performance on the G3 and G4 chips.

Of course, a highly optimal bit of PPC assembly could be written to replace the missing mode and instructions on the G5.

Then again, Microsoft could twist IBM's arm and get them to make a custom variant of the G5 that includes this mode, and maybe chops some cache for cost conservation. I sincerely doubt that the chip IBM winds up fabbing for the next Xbox is going to be identical to the version currently shipping in Apple's G5 desktops.

Now I know why G5 loaded truck found at Microsoft

[shailu]

Even Microsoft wants G5s

Ironic?

[g00set]

"It's ironic that IBM, with its roots in the computer industry, doesn't supply the processors for the main portion of the personal computer industry. Intel does".

Oh come on. What Intel computer can you buy that has a friggen guy come with it? IBM deals with the big boys. Internation Business Machines. I don't find their lack of presense or desire in the home PC market at all ironic.

Now I know why G5 loaded truck found at Microsoft

[shailu]

Even Microsoft wants G5s

Mainly the kernel

[Sycraft-fu]

Windows NT and its successors were designed to be portable. NT ran on x86, PPC and Alpha. 2k had a beta Alpha version that got killed near the end for reasons I do not know. XP has already been ported to IA-64. Windows CE runs on tons of different chips.

While it will require some reworking, it won't be a whole hell of a lot. Just because Windows doesn't run on PPC doesn't mean it can't fairly easily be ported to it.

IBM could have offered more customizations.....

[JollyFinn]

IBM microelectronics make custom asics with PPC cores in them, and IBM:s chip designing is in higher level stuff than AMD so they can modify that cheaper at expense of clock speed that they get... But now at 0.9u the PPC970 is supposed to be quite tiny so what ELSE they will put in there besides the CPU core and cache? Instead of using altivec they might go something more excess like putting 16 FMACS. Which would give microsoft both superiour numbers and performance but also guarantee that other chips wouldn't be compatible with it, as they would have instructions that no one else has, and in other way their developement package might be really only way to port software for it, and the customizations might even make reverse engineering the thing without full developement package from microsoft impossible. They could offer packaging with low latency mainmemory in the package, and something like 4-8 channels to the memory chip, inside the package. And only put outside interface to graphic chip outside the package and put all the other supporting logic in the same chip with CPU. Hey IBM has LOTS of options and modifications and stuff that they could have offered for microsoft besides price point. IBM could have made point hey we offer you 4 times as much memory bandwith and 4 times as many flops as our competitors in same price if you take the reduction of other chips in the system in account. And AMD and INTEL in their highly tuned hand optimized design methologies where not able to offer something even resembling the beast that IBM could customize for microsoft, at reasonable price. IBM makes great business selling G3:s with lots of custom stuff attached to it on single chip. They might even maker HARDWARE decryption on the processor chip for instructions stream, that could mean a LOT harder modifications for it than for original xbox.

MOD PARENT UP

[Jesrad]

Since the XBox1 was (also) test grounds for future Trusted architectures, and more or less failed, parent is most probably right on target here.

Re:MOD PARENT UP

[Thud457]

Good point. But I don't think leaving out local storage is going to work very well for a "trusted" general-purpose computer.
Or are trusted architectures by definition not general purpose machines?

The OS

[DrCode]

Will it be Linux? OS X? How about OS/2?

Re:One wrinkle using the VPC code from Connectix..

[spir0]

Why? Because the G5 silicon lacks the special "virtual little endian mode" that the Virtual PC code from Connectix relies upon for performance on the G3 and G4 chips.

interesting. I didn't know that. how does this pseudo little endian mode work? bit swapping in the registers? or is it done with different opcodes? surely the 64bit chip has enough registers to shift a truckload of data around.

Then again, Microsoft could twist IBM's arm and get them to make a custom variant of the G5 that includes this mode, and maybe chops some cache for cost conservation. I sincerely doubt that the chip IBM winds up fabbing for the next Xbox is going to be identical to the version currently shipping in Apple's G5 desktops.

I agree. However a dual proc G4 based Xbox would also have me in need of a towel.

This doesn't make sense

[Aqua OS X]

This doesn't make any sense. I thought the whole point of the xbox was that it eased console development by being nothing more then a stripped down Wintel box.

That press release was kind of vague. perhaps MS is going to use an IBM semiconductor for something besides the CPU.

So how does this relate to ATI?

[BlackjackGuy]

I thought ATI was doing graphics chips for the Xbox 2? Is that no longer the case, or is ATI designing the chips and IBM manufacturing them? Not sure about the relationship here. Can anyone explain?

Thanks.

Re:So how does this relate to ATI?

[jared_hanson]

What makes you think this announcement has anything to do with graphics chips? IBM's forte is in processor design, and supporting chipsets. Just as Apple uses nVidia and ATI for graphics cards, and IBM for processors (in the G5 and the newest iBook).

This announcement is more important for its affect on the main CPU in the Xbox 2, not the graphics capabilities.

Re:So how does this relate to ATI?

[BOFHelsinki]

First off, I'm inclined to agree with you about the announcement.

However, IBM is manufacturing (soon at least) graphics chips for Nvidia, and has designed and manufactured their own in the past (the GT1000 and RC1000 chips in the Diamond-era Fire GL cards, before ATI took over the Fire GL team and brand), so graphics chips is at least familiar ground if not the forte to IBM... And ATI likely won't be manufacturing the XGPU2, they'll be just designing it for MS, who then decides on the fabber...

Regardless, I do think this announcement is CPU related. And I think IBM's Cell/Grid biz with Sony and Toshiba doesn't really clash here -- the projects and the concepts involved are just too different.

It's an interesting scene! Just when the prev gen had everybody so neatly camped ;-)

Emulation-Virtual PC

[Enrique1218]

For all who question backward compatibility. Microsoft does have an emulator for PowerPC (Apple) system. It is VirtualPC. Virtual PC doesn't currently support 3d hardware but we are not talking about a system that will be out by this Christmas. Microsoft could in time optimized the VirtualPC technology for the 64-bit G5 and add 3D support. This way they can support older games at current speeds and possibly (although not likely) use it to run the xbox operating system + directx. As an added bonus, not only can we spectulate Apple will get a lower power processor for Powerbooks but they can get a better pc emulator.

no, it would be crud

[DrSkwid]

like it is everywhere else

Re:Hope it's...

[Paladine97]

I wouldn't say there is a Gamecube "performance issue." It runs beautifully fast. 485 Mhz of PowerPPC goodness. It is right on par with the XBox CPU since the PowerPPC design is so much more efficient.

Power5 instead of G5???

[Bodysurf]

Why don't they use IBM's Power5 CPU instead of the G5?

Wonder Why

[illumina us]

I wonder why they chose IBM chips. Is it for the better FPU and processing power/cycle or are they thinking it's better as a detriment to hacking being that it's not running on x86 architecture and therefore not as widely known. Personally, I think IBM chips are a good step seeing as that I am still an Amiga fan and still have my A500 and A1200 laying around :)

Re:Wonder Why

[MGS Hartman]

oh cool. death of Intel. film at 11.

Re:Hope it's...

[AtaruMoroboshi]

Frankly, what are you talking about? Performance issues?

The Gamecube is way more powerful than the PS2, most importantly with none of the jaggie aliasing problems nor the atrocious load times..

Yes, most of the time those 2 issues are no longer a big deal on the PS2 but it took a lot of developer cleverness to do it.

Re:Hope it's...

[CrypticSpawn]

They just might make it 1GHz and knock off L2 cache, or reduce it by half

Read between the lines

[psyconaut]

The Xbox using Power architecture is so that Microsoft can side-swipe Apple by bringing WIndows to Apple desktop hardware ;-)

-psy

IBM and the personal computer market

[AchilleTalon]

"It's ironic that IBM, with its roots in the computer industry, doesn't supply the processors for the main portion of the personal computer industry. Intel does."

I don't know the details of the agreement IBM is having with Apple. But, it seems IBM relies only on Apple to offer a personal computer product to the mass market (not so massive considering the pricing). I am still amazed IBM has done very few to enable third party vendors to produce hardware (motherboards, etc) for the PC market using their G serie processors.

Maybe this played a role in the decision of MS to use the chip. Since it is not easily available and hacked by the community, it will prevent their next generation X-Boxes to be as easily hacked as the first generation.

Re:IBM and the personal computer market

[BOFHelsinki]

Amazed?

IBM wants to be just a contract fabber in the low margin beige box PC market. What kind of third party vendors are you thinking of? Mac clones? PowerPC Linux machines? How could IBM "enable" those kind of vendors to even surface, let alone succeed (well enough to warrant any effort from IBM) in a market with Windows, x86 Linux and Apple machines already -- what's the magic recipe? :)

I do agree with your idea of Microsoft's reason for going PPC. The small increase in developer learning curve is probably more than worth the decrease in hacking (and piracy).

Ironic? Piffle

[the eric conspiracy]

How is it ironic that the company that invented the microprocessor be the market leader in desktop computers that are based on microprocessors?

Re:Hope it's...

[The_dev0]

It does accept DVD media, in fact the proprietary disk used is a one-of-a-kind DVD. If anything, using a non-standard DVD would be more expensive when you factor in production, etc. The proprietary format was simply to make piracy harder, if not impossible.

Microsoft wants to kill that market

[rsilvergun]

Last I checked, there's no licensing fees involvoved beyond buying the development tools. With Xbox they probably get $15 or so a game for each new release. It's vastly more profitable. They'd love to see the PC market die in favor of the XBox. What they don't want (and the reason they're being so nice to their developers) is for the Market to die in favor of the Playstation.

Even if MS is willing to let to PC game market live on (or is forced to), I think you'll find more and more developers using cross-platform API calls instead of hand written assembly. Between advances in computing power and in compilers, there's not much point to resorting to assembly anymore. The fact is, hardware has progressed faster than artists' ability to make use of it. Just check out any review of a recent high end video card for proof. They're all having trouble testing the cards because current games don't tax them at all.

First they steal our Halo...

[ProfessionalCookie]

Then they want to use our G5? I think I'll be headed to church this sunday as well.

I wonder if this brings up price/performance issues at IBM- I mean making chips for all three major console suppliers. I wonder how much research will be bread across product lines?

Yes, quite out of character...

[rsilvergun]

for the paragon of Jounalistic Virtue that is /. :).

Re:XBOX-n == GameCube-n

[spir0]

I'd like them to move to an enclosed disc like a minidisc, but with the capacity of a DVD.

I hate it when people touch the surface of my media...

Re:Hope it's...

[Anonymous Coward]

Don't be so sure. Don't be fooled by the clock speeds. The GCN is as powerful as the PS2, if not moreso, and is the superior of the two machines if actual output is a layman's comparison. Check out a comparison.

Why would they bother?

[RatBastard]

How many consoles can run programs written for previous generations of their "line"? The only one that springs to mind is the PlayStation 2, and that was done to hide the fact that there weren't all that many PS2 titles at launch. While it would be nice, and it would save a few thousand cubic feet of living room space, it's probably not in their interest to do so.

Re:Why would they bother?

[realdpk]

Gameboy Advance can play Gameboy games. SP can play 'em without an adapter - I think some games required an adapter on the older model.

I seem to remember an adapter you could get to play Gameboy games on the SNES(?).

Gameboy games can be played on a Gamecube, admittedly with a relatively expensive adapter.

Atari 5200 could play 2600 games. Ok so I'm stretching here. Point is, out of the relatively modern popular gaming platforms, many of them do offer backwards compatibility.

Re:Why would they bother?

[SEE]

Atari 7800 played Atarti 2600 games.

Can it really help IBM's lossmaking chip business?

[Warlock48]

providing a much needed boost for IBM's lossmaking chip business

I thought the XBox was selling below cost... In this case, unless Microsoft foots the total bill itself, I don't see how IBM will stop losing money!
They'll get volume, no problems, but at what margin?

Re:Hope it's...

[Logicdisorder]

The questions is will it be better than the CPU IBM is building to the PS3. I love IBM they are great

Re:C'mon, there are lots of non-x86 chips out ther

[Vaystrem]

"The GameCube uses a PowerPC"

And we've seen Nintendo port gameboy games to the Gamecube....

So what if Microsoft's new console could run Gamecube software natively and gameboy software via emulation or somesuch? That would be quite the coup in the software wars vs. Sony.

It might also explain the RARE sale to Microsoft from Nintendo to a degree, perhaps that was simply the first step in establishing a business relationship....

IBM made the chips for a most awesome console

[maudite]

Atari Jaguar

maudite

Nice, BUT, i have some comments:

[ruiner13]

"By buying a heavily subsidized $200 game machine from Microsoft, and then adding a pirated disk drive, the Xbox can be used as a "poor man's PC, turning a $200 game machine into a $600 personal computer, which Microsoft doesn't like at all," Doherty said.

That may have led Microsoft to the PowerPC platform developed by IBM.

Also, Microsoft has been bothered by the relative ease with which hackers have copied games that run on Xbox consoles."

Apparently this writer doesn't realize that the problem will still exist with PPC, as there are no shortage of open source OS out there. Unless, of course, M$ decides to use a custom board and bridge, making it more console like than the current Xbox, but making it more expensive. I'm not surprised they are going with IBM, the G5 is compared to the high-end P4s and opterons, even the G4, a very inexpensive chip. It also draws far less power than the intel and amd offerings, and scales very nicely allowing 2, 4 or 8 way setups. Supporting 1GHz busses is nice too for a game machine. That console could fly... might be the first bit of M$ taintware i'd actually buy.

Longhorn is being built using C#

[zeoslap]

Which like java is essentially write once run anywhere. If XBoX2 runs on a version of Longhorn then the only change needed to run on a PowerPC as opposed to x86 is a new CLR.

Pentium bug!!!!

[caveat]

Touche!

Now we know why they wanted Virtual PC!

[mabhatter654]

Now we know why they wanted Virtual PC! That way they can simply run their current code under emmulation and burn the results to ROM...seems pretty simple to me.

MS porting Windows to G5?

[Sean Clifford]

Okay, call me a conspiracy theorist with this one.

Essentially the XBox runs a win2k derivative kernel with DirectX slapped on top of it. Right? Everyone's making the transition to 64 bit, and Intel's hobbled a bit in that department (dot)

Since Redmond had decided to go with the PPC for XBox2 (dot) Maybe they're gonna pull something interesting. Mayhap the big spaz over that guy's G5 snaps, mebbe? (dot)

Who knows whether the dots connect?

I dunno if Redmond would really bother to get Apple folks to run Windows. But maybe this goes hand in hand with their fight in the server market. Big time Windows clustering.

Sony + IBM?

[Negativeions101]

I thought SONY was working with IBM on the Playstation 3?? So what's going on?

Re:Sony + IBM?

[BOFHelsinki]

They still are. IBM may end up designing or at least fabbing CPUs for all three (PS3, XB2, GC2), no problem.

I can't help but...

[Soong]

...thumb my nose at all the nay-sayers who kept kibitzing that Apple should move to x86. Microsoft is switching to PPC! Nyah!

Re:XBOX-n == GameCube-n

[spir0]

Nintendo makes games that don't suck?

Only if you're still in diapers.

Nintendo, excluding what they release on the GBA, make complete shit aimed at the lowest age group possible. The reason for this may well be because they believe in the value of getting their consumers young and keeping them loyal.

only problem is, when those gamers grow up, they're going to either change their console of choice, or get frustrated with their loyalty because the games don't have enough variety.

The GBA is the complete opposite of Nintendo's non-handheld console strategy. There are a huge variety of games for the GBA and they rival the PS1 in quality and complexity. If the big N put as much effort in the NGC, then maybe their sales might be better.

Hello? Connectix?

[bobobobo]

From an article

"Compatibility with older Xbox software could be provided by emulation technology Microsoft acquired in February from Connectix, which sells Virtual PC, allowing Intel-based software to run on PowerPC chips."

$200 Mac?

[jamesmrankinjr]

By buying a heavily subsidized $200 game machine from Microsoft, and then adding a pirated disk drive, the Xbox can be used as a "poor man's PC, turning a $200 game machine into a $600 personal computer, which Microsoft doesn't like at all," Doherty said.

That may have led Microsoft to the PowerPC platform developed by IBM.

Now Apple will be upset when the new XBox is hacked into a $200 iMac.

Peace be with you,
-jimbo

Re:One wrinkle using the VPC code from Connectix..

[Johnny Mnemonic]

Microsoft could twist IBM's arm

That's something that I'd like to see.

windows ppc

[simpl3x]

there was a version of windows for ppc. perhaps microsoft in a shrewd move to ensure its continued relevance should intel fail, has maintain the os in parallel. apple is thought to have an intel version maintained along side ppc. this makes business sense, not to mention that most of the world will not be intel ia32 or 64. remember that arm based processors for paltops and cell phones, as well as other risc processors run on most portable devices, where microsoft has strong interest. the future is not the pc!

How many cpu cycles are spend in DirectX?

[3770]

In a regular game, how much time is spent in DirectX? Lets say, for the sake of argument that it is 50% and that the DirectX code is compiled natively. Furthermore, lets assume that it runs about 4 times faster then the 800Mhz PIII.

This means that the rest of the game code, the emulated part, would have to run just a little faster than half (about 60% I believe) the speed of a PIII for the game to have the same speed on average.

It isn't perfect, but it goes to show that not all code has to be emulated. And DirectX isn't the only code that can run natively. Every call to an API that the game makes will be running native code. Also, recall that much of the DirectX code will be executed on the grafx card which never was and never (knock on wood) will be Intel.

Re:C'mon, there are lots of non-x86 chips out ther

[benzapp]

The Game Boy Color is Z80 based.

Your knowledge of processors is impressive, but that is an X86 processor if I ever saw one. Maybe its a very OLD X86 processor... but it still is one none the less...

oh oh

[presearch]

Not that I -always- suspect M$ of dirty dealing, but perhaps
M$ is just promising some biz to IBM so they disclose some info
and then M$ can indirectly lay the smack down on Apple's chip supply.
Almost everything M$ does is always to leverage Windows in some way.

Mod parent up

[SysKoll]

Yep. Don't forget that IBM is one of the very few licensed makers of x86 processors. So the bets are not on a PowerPC Xbox2. The bets are on an IBM-supplied x86 processor integrating the graphic chip and the CPU, or maybe the x86 with an integrated high-bandwidth bridge.

Why does this make sense? Because IBM has a very 90-nm advanced process on 300-mm wafers that can spit out small -- hence cheap and fast -- dice with a very reasonable yield. And BTW, their new Fishkill factory floor in entirely under Linux!

Re:Hope it's...

[Nazmun]

" Yes, most of the time those 2 issues are no longer a big deal on the PS2 but it took a lot of developer cleverness to do it."

The worst thing about the ps2's three cpu's is that they are a bitch to program for and everything requires cleverness.

Re:C'mon, there are lots of non-x86 chips out ther

[mandolin]

Nope. The Zilog z80 was a riced-out Intel 8080. The original 8086 (and 8088) was source-, but not binary-, compatible with the 8080. Google for "z80 8080" and "8086 8080" to verify.

It will impact *Windows* users...

[leonbrooks]

...at least until they bring out a PPC64 version of retail MS-Windows (IIRC, the x86 XboX runs a severely mangled Win2k natively), and that'll probably have its wings clipped so it's-a-no-run on a PPC64 XboX until somebody effectively cracks it.

Why do I say that? If you RTFA it says...

By buying a heavily subsidized $200 game machine from Microsoft, and then adding a pirated disk drive [intermission: er... what? How can you pirate* a disk drive?], the Xbox can be used as a "poor man's PC, turning a $200 game machine into a $600 personal computer, which Microsoft doesn't like at all," Doherty said.

Here in Oz, an XboX is AUD$330. Add in the cost of a keyboard ($10) and mouse ($30), bigger RAM ($60) and it comes to $420, which is in the same range as new PCs clocking at five times the speed, starting with five times the hard disk space and built with plenty of room for expansion. I think the "turning it into a PC" effect is being overblown by Microsoft both for "poor little me" abuse/sympathy points and to subtly assert that an XboX is as powerful as a PC when it ain't (weeeelll... possibly modulo the graphic card unless you shell out for a good one).

I predict that one effect of this CPU change will be to give Linux an even wider road into the XboX market. One has to wonder whether XboX will get zero, one or two subarchitectures in the kernel tree. (-:

.

* "Piracy" is a major misnomer. So you publish a program, and it gets "pirated". Well? Nobody sails up, shoots you and your crew, pushes the bodies over the side and makes off with a multimillion dollar vessel and its cargo, do they? So it's not piracy - although it is criminally illegal copying. But wait 'till somebody publishes SLPWA's open letter to Vietnam's Science and Technology ministry.

OS X on XBox?!?

[illumin8]

Wow... I can see the potential implications now:

Microsoft releases XBox 2 with a PPC 970 running at 2 ghz.

XBox hackers break the encryption and are able to run unsigned code on the PPC 970 processor.

XBox hackers use MacOnLinux to run a full-fledged Mac OS X on cheap Xbox hardware.

???

Profit!!!

Just kidding about the profit part... But who wouldn't want a 2 ghz. G5 running Mac OS X for about $300. This would be a killer workstation and would run circles around the existing Xbox 1 running Linux. Can you imagine?

Re:Hello? Connectix?

[yamla]

This assumes the only reason the XBox 2 would have problems running XBox 1 games would be the different CPU. This is not the case.

IBM root of the computer industry?

[lecca]

It's ironic that IBM, with its roots in the computer industry, doesn't supply the processors for the main portion of the personal computer industry. Intel does.

Huh? Since at least the 8088 (the essental birth of the industry) IBM's PC's have always had Intel processors. The only irony is that IBM is now the "other" chipmaker competing with intel.

Re:MOD PARENT DOWN

[antin]

Of course it is off-topic you dolt - where is the news post on stupid mods? I fully agree with the parent, half the mods these days are modding things in mad directions.

Sensible posts become trolls, trollish posts become insightful, off-topic posts become interesting, and depending on the topic (ie whether 'we' like the company or dislike it) fanboy responses get modded accordingly.

The quality of stories, and of modding on slashdot has fallen methinks.

Re:C'mon, there are lots of non-x86 chips out ther

[benzapp]

my bad. 8080, 8086...

Re:PowerPC was *supposed* to become a commodity ch

[mcrbids]

"There are two major products that come out of Berkeley: LSD and UNIX.
We don't believe this to be a coincidence."

Please, if you are going to quote people, REFERENCE THE QUOTE!

That quote came from Jeremy S. Anderson... (ref.)

Deja Vu: SGI + Nintendo

[ayeco]

SGI kind of did the same thing w/ the N64. Unfortunately it didn't pull SGI out of their downward spiral. Maybe IBM will faire better.

Re:Hope it's...

[0x0d0a]

The problem is lack of primary memory.

This is the primary lack on all current consoles (*especially* the PS2). It's incredibly irritating that console manufacturers were so short-sighted here.

xbox1 on xbox2?

[sl0ppy]

while some user called yamia keeps making claims that the xbox2 will not be backward compatible, i have to ask why not?

it's certainly possible, and i'm not talking emulation. last i heard, ibm had some broad cross licensing agreements with intel, and only got out of the x86 hardware business because it wasn't profitable anymore.

what happens if ibm is asked to use their dual core fabrication techniques to embed some sort of x86 processor next to that g5? it starts to sound like a ps2, doesn't it?

besides, the 700mhz celeron equivalent has dropped so much in price, it could probably be included like the ps1 processor in the ps2, at very little cost, even if it wasn't on the same dye as the chip. it would make one heck of a secondary processor, same as how the ps1 chip is used in the ps2.

microsoft isn't stupid, they've seen the headstart that sony got with the ps2, and they'd be insane to not cover every base to get ahead in the market.

they're ruthless, they're aggressive, they lie, they cheat, they steal, but they're not stupid, and it's best that nobody forget that.

i86 == shitty architecture

[bobobobo]

Why would you want to develop a gaming console using the i86 instruction set. It's main redeeming value is it's backwards compatibility, but since the xbox isn't pc compatible who cares?

Re:Hope it's...

[The_dev0]

The drive is standard, yes. The disks are not.

Bad for Linux?

[G3ckoG33k]

Will this be bad for Linux? Will MS now be able to say - Hey that open source thang you're doin', couldn't you just pace down a bit? I wonder: how much is open source worth, for an already a big pocket?

Re:PowerPC was *supposed* to become a commodity ch

[Type-R]

Err, except that AIX has been running on PowerPC for about 7 years...

Brilliant Move by Microsoft

[vwjeff]

I hate to speculate but....... Going with IBM would make Intel and AMD quite unhappy. I predict Intel or AMD will make an offer Microsoft cannot refuse. Their business model looks something like this... 1. Create FUD by announcing partnership with IBM. CPU prices drop for Microsoft naturally. 2. ? 3. Yes. Profit!!!!!

Re:One wrinkle using the VPC code from Connectix..

[Lars T.]

I have yet to see anything substantiating that claim. Heck, there never was a "virtual little endian mode". When somebody (esp. Microsoft) invents new names for something to prove they don't exist (anymore), I feel uneasy.

3.1.4 PowerPC Byte Ordering

The PowerPC architecture supports both big and little-endian byte ordering. The default byte ordering is bigendian. However, the code sequence used to switch from big to little-endian mode may differ among processors.

The PowerPC architecture defines two bits in the MSR for specifying byte ordering--LE (little-endian mode) and ILE (exception little-endian mode). The LE bit specifies the endian mode in which the processor is currently operating and ILE specifies the mode to be used when an exception handler is invoked. That is, when an exception occurs, the ILE bit (as set for the interrupted process) is copied into MSR[LE] to select the endian mode for the context established by the exception. For both bits, a value of 0 specifies big-endian mode and a value of 1 specifies little-endian mode.

The PowerPC architecture also provides load and store instructions that reverse byte ordering. These instructions have the effect of loading and storing data in the endian mode opposite from that which the processor is operating. See Section 4.2.3.4 Integer Load and Store with Byte-Reverse Instructions for more information on these instructions.

I have yet to see anything by IBM saying that there were changes in that behaviour in the 970.

Irony? "You keep using that word...

[BiOFH]

"You keep using that word. I do not think it means what you think it means." - Inigo, The Princess Bride

IBM has never supplied the processors for the "main portion of the personal computer industry" and Intel always has. How is this ironic? Did you just discover the PC industry? Do you need a history lesson?

Congratulations - Its a GameCube?

[justin_saunders]

What do you get when add an ATI graphics chip and a PowerPC processor? - thats right, the XBox 2 is a GameCube!.

But seriously folks, what does this mean for GameCube 2's "air supply". With ATI and IBM tied up (legally,financially) with exclusive tech deals with M$, how is the GameCube 2 going to retain backwards compatibility without using "second best" technology.

Re:Congratulations - Its a GameCube?

[Junks Jerzey]

But seriously folks, what does this mean for GameCube 2's "air supply". With ATI and IBM tied up (legally,financially) with exclusive tech deals with M$

IBM is not "tied up" with Microsoft. Remember, IBM is also supplying processors for Sony (the "CELL"). The PowerPC is a commodity processor used in lots of devices. Maybe the modified PPC970 is just for Microsoft, but that doesn't apply to the PPC970 or PPC line in general.

Re:Congratulations - Its a GameCube?

[samdu]

Actually, as a happy GameCube owner (in addition to PS2, Dreamcast, PS1, Jaguar, CD32, N64, and Saturn), The 'Cube is the one box that I hope doesn't carry backwards compatibility on to its next incarnation. Not that the games aren't great, there are some fantastic ones, its more that I hope they veer far away from the tiny little disc format and move on to some DVD-based standard format, both for the consumer and for developers. It's time for the big N to embrace some format standards.

Scary

[Anonymous Coward]

Now it would be scary if both the ATI (GameCube gpu) and the IBM (GameCube cpu) deal were done only to block Nintendo's source for quality chips, and that they really just put the next bigger Intel and NVidia chips into Xbox 2...

Except that...

[Channard]

Master System compatibility required you to plug in an adapter into the MD get play Master System cards. It wasn't backwards compatible from the word go.

Hmmm...

[Solokron]

I have not seen someone commenting on this but I can't stop suspecting that this may also be a political move by MS. With IBM having its new strength in the Linux world could MS be trying to gain better ties with IBM for IBM's Win Servers?

Re:Hope it's...

[zeno_2]

Doesn't the Gamecube also read discs opposite of a normal dvd player? (from outside to in, or spinning counter-clockwise?)

Re:Hope it's...

[rifter]

" Yes, most of the time those 2 issues are no longer a big deal on the PS2 but it took a lot of developer cleverness to do it."

The worst thing about the ps2's three cpu's is that they are a bitch to program for and everything requires cleverness.

Actually, I consider this a good thing. Do you really want non-clever programmers writing the software for your platform? That's the way Microsoft went and look where it landed them! The whole OS and all the applications are written by script kiddies!

IBM looks set to own the console processor market

[scottme]

IBM and Sony announced quite a time ago that IBM would be providing the CPU for the Playstation 3. See this BBC News article.

Re:Hope it's...

[msh104]

exactly. making the linux on the xbox thingy even less attractive not to mention pc emulating xbox games. (a smart stratagy from the MS side if you ask me)

Porting vs. Piracy

[BOFHelsinki]

Just an idea that popped into mind: maybe MS is *willing* to trade development ease for piracy difficulty.

Maybe now that MS is an established player in the console scene they'll have easier time getting developers (they already have secured exclusive houses) even if the learning involves more than the previous small x86-PC to x86-Xbox step. And PPC is a well established architecture (mature, documented, lots of libraries, etc.) after all -- compare to PS2 at launch.

(And maybe even more of the game code is offloaded into the DX9/DX10 pipeline; geometry-related physics like deformation and collision detection may move completely to vertex shaders from CPU code. But this is probably a minor factor here.)

So, assuming that the move to PPC doesn't make the learning curve too steep, and assuming that MS currently loses billions or at least millions to games piracy, maybe it's worth their while to make the platform more alien to all the x86 hackers out there. (I hope I didn't alienate the hordes of PPC hackers now...) I mean, maybe the x86 compatibility has turned from a benefit into a burden for MS?

[Pardon my burst of unorganised thoughts written in a hurry; hope this is at least legible if not terribly thought-provoking. Off to lunch...]

Interesting presage...

[headGasket]

With this processor switch the XBOX 2 has the exact same component manufacturers the GameCube had (ATI for gfx & IBM for cpu). Could this be a presage about a Nintendo takeover by M$?

Emulation this, Emulation that

[Zetta Matrix]

I hear a lot of talk about emulating x86 on this PPC in order to have backwards compatibility with XBox 1 games.

Remember, this is not the only viable option. Back in the days of NT4, DEC came up with FX!32 which would emulate, profile, and then selectively translate x86 code into the Alpha equivalent. I think that this approach may be quite viable given that the XBox has a HDD which could store this profiling data and translated code snippets.

For the first few runs, you notice some lag and missed frames, and then after 20 minutes of playing the game (after some profiling and translation kicks in) you're back up to full speed!

That would rock, and IBM and MS both have the brain power to pull such a thing off, no problem.

Developer Learning Curve...

[BOFHelsinki]

Developers care. They don't necessarily care about backward compatibility, but they care about an architecture that's already intimately familiar to them.

Then again, it won't be too difficult to get going in the PPC world. Lots of stuff and learning help already available.

Overall, for them, it's ultimately not about how good the architecture is, it's how good *they* are in it! (Or how soon they'll get good enough.)

Learning curve is what delayed PS2 titles. Luckily there was ample time before Xbox and GC.

As an aside, maybe you dismissed the "common x86 arch" benefit too easily. Xbox games aren't PC compatible, but they are the easiest to port to PC, and that saves time and money for a dev house (wanting the PC slice of the pie too). And both are crucial resources, of course.

Re:Hope it's...

[nate1138]

Look at the number of polygons that each deck is capable of rendering and you will think differently. I realize that alot of this has to do with memory/graphics/bus/etc but that is what counts. Compared to the other systems, the Gamecube is seriously lacking in HP.

Re:Hope it's...

[nate1138]

Way more powerful than the PS2? Can I have some of your crack?

• Gamecube - 33M Polygons/sec
• PS2 - 66M Polygons/sec
• XBox - 60M Polygons/sec

Now how is that more powerful?

Check here for complete tech specs on these three (and a bunch more) consoles

Re:Hope it's...

[Junks Jerzey]

Look at the number of polygons that each deck is capable of rendering and you will think differently. I realize that alot of this has to do with memory/graphics/bus/etc but that is what counts. Compared to the other systems, the Gamecube is seriously lacking in HP.

Roughly in order of power, from highest to lowest:

Xbox
GameCube
PS2

And guess what? The PS2 is outselling the other two consoles COMBINED, and by a large margin. Tech is not everything. In the next round of consoles, tech will mean even less.

Business Consideration

[Klanglor]

Unwanted result from cpu change,

1) Better and faster ports of games for MAC not PCs, which may shiff the sales of personal computers to apple.

2) BSD:XB2 instead of Linux:XB1

3) Death of microsoft. :'(

You can't embed red-hot components

[Morgaine]

You simply can't put a super hot P4 in an embedded environment.

I agree, but it goes further than that.

Actually, you can't sensibly put a super hot P4 into a normal home user PC either, but Intel and Microsoft haven't felt the heat because the hot potato has been dumped onto the laps of end users, and they've more or less coped with the problem.

However, with Xbox, the problem is in Microsoft's lap, and I bet that they have felt the heat. For a start, a hot CPU requires a fan, and fans are noisy in a lounge hifi environment. Hot CPUs also place a lower limit on size and therefore on cost of the console, so it's inevitable that x86 has had a detrimental effect on the Xbox balance sheet. You can tell just by lifting it up that the Xbox has had much more engineering put into it than is normal for loss-leading hardware in the games industry, and part of the reason for that will inevitably be heat management.

On that basis, it wouldn't surprise me in the slightest if X2 replaces x86 by G5 technology. Intel (and AMD) have simply lost the thread in their current direction. Ultimately, everything wants to be "embedded", and you can't embed things that run red hot.

Re:One wrinkle using the VPC code from Connectix..

[ivan256]

Because the G5 silicon lacks the special "virtual little endian mode" that the Virtual PC code from Connectix relies upon for performance on the G3 and G4 chips.


PPC chips can't run in big and little endian mode at the same time. There's no way VirtualPC uses the little endian mode of the PPC processor under MacOS. There are byte-swapping load and store PPC instructions that are no more expensive than non-swapping loads and stores, so there is no real performance hit due to byte swapping when emulating.

G3+AltiVec chip = G4 iBook

[MacDork]

Take a look at the new iBooks, it's already here and being marketed as a G4. I can't find PPCXXXX listed anywhere on Apple's site right now, but I've read reports from people who've purchased one. PPC 750FX I think?

Re:G3+AltiVec chip = G4 iBook

[MarcQuadra]

Nope, sorry. The G4 in the new iBooks is actually a Motorola 74xx G4. Apple's not being specific about it because when they DO move to something from IBM they don't want to blow a hole in their branding by having to ditch the G4 name.

IBM hasn't even begun production of an AltiVec-enabled 750-series CPU (codename 'Mojave'). The only thing IBM is admiting to right now is the 750GX, which is really nothing special, just a beefier 750FX.

The new iBooks have 'real' G4s in them, nothing else exists. People HAVE opened them up and gawked the CPU, they're Motorola 7455s or 7457s, IIRC.

Re:Hope it's...

[Radius9]

Lack of foresight has less to do with it than cost. Putting more memory into it brings up the cost. The PS2 has enough memory, only its segmented in such a way to make much of it practically useless. Who needs 8 megs of sound RAM when you only have 4 megs of video RAM?

Re: i don't get it

[mb12036]

i don't get it. why would you buy an xbox2 if you weren't planning on buying any new xbox2 games? and if you already have an xbox1, what difference does it make if xbox2 is backwards compatible?

Probably just because I'd rather only have one console. If I could sell my X-Box to help pay for part of the X2 I'd do it in a minute - if it was backwards compatibile. Plus I could buy original XBOX games cheap to play on my new X2. But if they take that functionality away, I'll be just like a new consumer entering the market and I can consider the Sony console and MS console equally on its own merits. I won't be swayed by the fact that I have a thousand bucks in games sitting at home that I could play on the new console.

Re:Hope it's...

[AtaruMoroboshi]

If that was truly the case, why do gamecube games, on the whole, look significantly better than PS2 games?

Any figures you cite that show the PS2 to be more powerful than the Xbox are to be considered extremely suspect.

Consider this: Splinter Cell is on all three platforms. The Xbox version is the most graphically intensive. The Gamecube version is slightly pared down from the Xbox version. The PS2 version is radically pared down. I've seen comparison screenshots and the PS2 version is significantly more simple, because the hardware couldn't handle it.

.

Re:Hope it's...

[nate1138]

why do gamecube games, on the whole, look significantly better than PS2 games

That's a bit subjective, isn't it?

And this argument is wholly irrelevant anyhow. After the very disappointing sales of the cube, I'd be surprised if Nintendo was still in the console biz in another 5 years [cough]SEGA![cough]

Re:One wrinkle using the VPC code from Connectix..

[poot_rootbeer]

if Microsoft uses the G5 (PPC970) chip, as everyone is speculating, they'll have to tweak the Virtual PC code base to run on the G5.

They're going to have to do this anyway, to get VirtualPC to run on the Apple G5s.

Re:Hope it's...

[AtaruMoroboshi]

it's not subjective when it's widely acknowledged that it's:

Xbox > Gamecube > PS2

and when the polygon counts in games like Soul Calibur 2 and Splinter Cell bear this out.

I mean, you seriously thing the PS2 is more powerful than the Xbox? That's what your figures said, and that's simply ridiculous.

as for the rest of your comment, you're simply trolling.

.

Re:Hope it's...

[nate1138]

Ok, first off, those aren't "my" figures.

Second, widely acknowleged by who? All the info I can find puts the gamecube at the bottom of the heap.

I'm sorry you invested in a shitty console, but don't take it out on me.

Re:Hope it's...

[AtaruMoroboshi]

well, you're ignoring the plain facts then, go read ign.com's multi-platform reviews of soul calibur 2 or splinter cell.

the figures you've cited are from a site called "psxfreaks", i'm sure they are unbiased. What Sony claims the PS2 can do and what it's real world performance is are radically different beasts.

Please admit that the Xbox's hardware is far better than the PS2's... and then explain how your "facts" can be right when it's clear that the Xbox should be stronger.

btw, I have no gripe with the PS2's library of games, it's the clear leader. I'm buying one soon, and I've bought PS2 games to play at friends' houses, including the (possibly-best-game-of-all-time) Virtua Fighter 4: Evolution. But it's blatant fanboy-ism to claim that the PS2 more powerful than the GC.

There's only one area where the PS2 is more "powerful" and that's capacity of the individual data discs.

.

Re:PowerPC was *supposed* to become a commodity ch

[KojakBang]

I CAN'T REFERENCE THE QUOTE unless I can put more than 120 characters on my signature, and I'm NOT going to add the reference for every post I make. You must be very proud of yourself nonetheless. ;P

Re:Hope it's...

[BOFHelsinki]

Then again, Splinter Cell is very PCish game, relying heavily on pixel shaders and large textures. Arguably it's not the kind of game that PS2 was designed for, which I guess would be something with lots of small untextured polygons, funky particles, light effects and just weird unorthodox hacks ;-)

Undeniably Xbox has immense geometry and light power; PS2 is showing its age. But possibly Splinter Cell is one of the least favourable comparos for PS2. (But I don't know what would offer a "neutral" comparison. And I don't own any of these consoles. Real men play with a mouse.)

YAY!! a blow to intel's ego!!

[XdarkstarX]

it's about time there's some more competition. AMD is greater than Intel in every way, but Intel is not as threatened by them as they are when IBM joins the opposition. heh heh heh. INTEL MUST BE BURNED!! freedom and quality to the people!! now if someone will put up some heavy competition against M$...

Re:Hope it's...

[nate1138]

I have never said that the PS2 is more powerful. I merely quoted perfomance figures from a third party. I don't even own a PS2. I have an XBox, and love it. Live rocks, the games are cool, and the HD games look awesome. Hell, the XBox had better be faster, it came out over a year later than the PS2.

But that wasn't the focus of my post. I said the Gamecube was underpowered in comparison to its competition, and I stand by that opinion. The Gamecube came out around the same time as the XBox, and was nowhere near as powerful a console.

Re:Hope it's...

[YellowElectricRat]

Raw polygon pushing is a vey misleading figure, but it's not surprising when coming from a PSX fan site. Even if I was sure about those figures (which still don't look quite right to me), they would only be applicable for games which used straight, unshaded polygons.

The main figures come in how many render passes you need to do to get the effects you're after. The GameCube for example, can apply 8 render effects to a polygon per pass (i.e. texturing, bump mapping, specular highlighting, stenciling, alpha blending etc etc), whereas the PS2 only has hardware support for a couple (i.e. texturing and bump mapping, at a guess).

Taking that into account, with all effects turned on, the PS2 would have to make 2-4 render passes to render the same scene that the GCN could do in a single pass. If a game uses fewer effects, the PS2 could compete, but the game would probably look pretty shite.

Re:One wrinkle using the VPC code from Connectix..

[WinDoze]

I sincerely doubt that the chip IBM winds up fabbing for the next Xbox is going to be identical to the version currently shipping in Apple's G5 desktops.

Think about that statement for a minute...

Suppose Microsoft's goal is to do exactly that: Include the Apple-version G5 in their gaming console. What better way to poke fun at Apple than to have, as a marketing strategy, "Yeah, their machines are so fast, we use their CPU's as toys".

Now THAT is a truly nifty (and evil) idea!

Random Conspiracy Theories

[SEE]

1) Anyone remember this rumor? So, the Xbox2 is going to use ATI and PPC technology, just like the Game Cube. Maybe backwards compatibility is going to be in the box, just not with the Xbox . . .

2) Now if you hack an Xbox2 to run Linux, do you get a cheap, MOL-capable G5-based machine? Wouldn't that be nice and Apple-infuriating . . .

Re:Hope it's...

[AtaruMoroboshi]

way to ignore everything and just repeat your opinion. my point is that the performance figures you quoted are OBVIOUSLY wrong, since the Xbox is way better than the PS2.

.

Re:Hope it's...

[nate1138]

Very true. Those figures don't look quite right to me either after a second look. However, for a more raw-perfomance oriented picture, look at the MIPS rating for each proc (from whatever source you choose). The GC still lags behind both of the other consoles (in terms of raw CPU, not GPU) power. I understand that this number is not the end-all of performance, I am simply stating that there is significant room for improvement.

Re:Hope it's...

[nate1138]

Are you even literate? Did you read the last post? I concede that those figures may be skewed. So find some of your own, don't just expect me to take your word for it.

Anyhow, this thread isn't about XBox vs. PS2. It's about the shitty proc in the GC.

Idiot.

Re:Hope it's...

[AtaruMoroboshi]

I didn't ask you to take my word on it, I asked you to read some reviews and find out for yourself that almost every cross platform game looks best on xbox, then on the cube, and then on the ps2.

at this point, i'm just curious to see how many more times you are going to respond...

Re:One wrinkle using the VPC code from Connectix..

[LionMage]

I have yet to see anything by IBM saying that there were changes in that behaviour in the 970.

I've seen precious little out of IBM regarding the PPC 970, period.

Regardless of what you choose to believe, the issue with Virtual PC is very real. Perhaps you take umbrage at the nomenclature for the CPU mode in question. I don't disagree that it's an idiotic PR-speak name. Regardless, there is something missing from the PowerPC 970 silicon that was in the PPC 750 (G3) and the G4 series from Motorola.

This has been discussed months ago on Slashdot. The PPC 970 isn't 100% Book-E compliant, if my memory serves correctly. It's still a PowerPC chip, but it contains some things that the other family members don't (e.g., hardware support for square root in the FPU), and omits some minor things that the other family members have.

The bottom line is, Virtual PC does not work on the G5. This problem existed before Microsoft acquired Connectix, and is trivially provable by attempting to run a recent pre-Microsoft version of Virtual PC on a G5. As I understand it, Virtual PC is the only software on the Mac that uses the little-endian hackery present in the G3 and G4 chips. That the IBM engineers responsible for the G5 chose to spend their transistor budget on other things is unsurprising.

Re:One wrinkle using the VPC code from Connectix..

[spir0]

hahah... trademark that....

Re:One wrinkle using the VPC code from Connectix..

[LionMage]

Dude, you're arguing with the wrong person. Notice I said "virtual little endian mode" and not "little endian mode." This is the marketing speak that I am quoting verbatim from the press release that announced Virtual PC 6.x will not work with the G5. If you want to argue the point, argue with the people who wrote the press release, or the Connectix engineers that Microsoft Borg-ishly assimilated when they bought the VPC product.

The bottom line is, there is either a missing mode, or missing instructions, from the G5 processor. Something that recent versions of Virtual PC rely upon to accelerate the emulation task on PowerPC hardware. If you want to argue the specifics, argue with the engineers who actually are writing the code. Don't yell at me because I repeated something that was in a Microsoft press release.

At least one former Connectix engineer confirmed that the problem was real, and had to do with some architectural decisions that were made for the PowerPC 970 (G5). Make of that what you will.

Re:Hope it's...

[nate1138]

at this point, i'm just curious to see how many more times you are going to respond...

One, bitch.

Re:One wrinkle using the VPC code from Connectix..

[LionMage]

Suppose Microsoft's goal is to do exactly that: Include the Apple-version G5 in their gaming console. What better way to poke fun at Apple than to have, as a marketing strategy, "Yeah, their machines are so fast, we use their CPU's as toys".

Maybe. I don't think Microsoft would make such a left-handed compliment, however, especially since it would cast their own product in a less-than-positive light.

Especially considering that Microsoft has lofty goals for the Xbox that extend far, far beyond mere gaming applications. The Xbox is a Trojan Horse.

Re:One wrinkle using the VPC code from Connectix..

[LionMage]

I know little more than what was in the press release announcing that Virtual PC would not work on Apple's new G5 computers. One Connectix engineer confirmed in a semi-public forum that there was a real problem, that it wasn't Microsoft spewing bullshit.

Others in this thread have mentioned that there are byte-swapping load/store instructions in the PowerPC architecture which reduce the cost of doing little-endian loads and stores. According to what I've read, Virtual PC uses a special mode available on the G3 and G4, and not available on the G5. Supposedly, it's the only Mac software to ever use this mode. Without disassembling the Virtual PC code myself and examining it carefully, I couldn't tell you what specifically got "lost" from the G5 silicon.

It's amazing the amount of skepticism that's greeted my single comment, even though this whole issue was hashed out months ago in Slashdot and on various Mac forums, right after the Virtual PC press release.

Re:One wrinkle using the VPC code from Connectix..

[spir0]

I've read the other posts... my POV is simply from lack of knowledge - not trying to be smart or anything.

I've since read a post on MS's knowledge base, but have come to the conclusion that MS has the money to get around or fix the problem :)

Re:C'mon, there are lots of non-x86 chips out ther

[damiam]

I highly doubt the XB2 could run GC games natively, if for no other reason than that the discs wouldn't fit (plus, the GC has some pretty specialized hardware). Gameboy games might be a different story. The current XBox can already run GB emulators under Linux, it's only legal issues that prevent some company from releasing a GB player for XBox. The same legal issues would presumably persist with the XBox 2, but the technology is there and has been for a long time.

Re:One wrinkle using the VPC code from Connectix..

[Lars T.]

All this doesn't prove that the 970 can't effectively emulate a x86 processor, nor that it doesn't have the little-endian access described in the PPC docs. Just that Connectix used it in a way they maybe shouldn't have (namely the "virtual" mode that isn't part of the PPC spec), depending on it always working that . Which just happens to be my other point in this thread.

Re:Hope it's...

[n0wak]

PS2's numbers are theoretical. Gamecube's numbers are a little more down to Earth. If you can find me a PS2 game that pushes 66M Polygons/sec, I'll BUY YOU A PS2! Seriously.

And, c'mon, do you really take numbers at face value at a site called PSXfanatics when it's comparing the GC? Or do you really fall for the Sony hype machine (and their outlandish pre-PS2 launch claims that they've yet to deliver on)?